. .5 , :
.. I .2. A...
. Ln. . in. .
I... Samsrzunx 1. .13.:
1 «w. x ‘2: 1312.1:
3.714 v.
I: . ..
i. . 4
.53...» . llaﬁgaaﬁt
:2??? he”. . .iu
; . ,z
vvl '03.?!
”.1. CV9."

s3,x....:l.rit
.1. x 83.33;). .I’.
a?! . ,

, 7:
. , :31 .

3mm:

.43.... gum. ‘

_.. 333533.?!
. $2. 1..
. await}.
Hum: ..
h". ah“. V
... ﬁgmrsmﬁw. v.
1.3.5.9.. Hzrwwﬁun.
’Vuitri aha
h, :3
a .

I 2. .,
5‘1 .81“
3.23.21?

{hi

.2
II.
.8.)

«1:1...
.

a. 1.8. gar...“
.. i .3
gammy,

 

 

72-33318

M

Mb

This is to certify that the
dissertation entitled

ECOSYSTEM CONSEQUENCES OF AGGREGATION
FOLLOWING SOIL DISTURBANCE

presented by
ANDREW STUART GRANDY

has been accepted towards fulﬁllment
of the requirements for the

PhD. degree in Crop and Soil Sciences

(3% 26.2%.

MajoéF’rofessor’s Signature

914.2% }ﬁ(

Date

 

MSU is an Afﬁrmative Action/Equal Opportunity Institution

 

UBRARYVIL
Michigan State
Universitxr

 

.‘n-o
--

 

PLACE IN RETURN BOX to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE | DATE DUE DATE DUE

 

{QB 559%“). 2007

 

 

 

Ds§¥121.‘§r%9

 

 

 

 

 

 

 

 

 

 

 

 

2/05 p:/ClRC/Date0ue.indd-p.1

ECOSYSTEM CONSEQUENCES OF AGGREGATION FOLLOWING SOIL
DISTURBANCE

BY

ANDREW STUART GRANDY

A DISSERTATION

Submitted to
Michigan State University
in partial fulﬁllment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Crop and Soil Sciences
Ecology, Evolutionary Biology, and Behavior Program

2005

ABSTRACT

ECOSYSTEM CONSEQUENCES OF AGGREGATION FOLLOWING SOIL
DISTURBANCE

BY

ANDREW STUART GRANDY

Carbon sequestration and greenhouse gas abatement in soils are two of a limited
number of rapidly-deployable, high impact C02 stabilization options now available to
policy makers. The long-term persistence of stored C, however, remains a major
uncertainty in C sequestration forecasts. In this dissertation, I examine the ecological
processes underlying soil organic matter permanence and the ecosystem and agronomic
consequences of long-term no-till.

Our experimental site is a series of ecosystems that differ in management intensity
located at the Kellogg Biological Station (KBS) Long-Term Ecological Research (LTER)
site. KBS is located in SW Michigan and receives ca. 90 cm of precipitation annually,
about half as snow, and has a mean annual temperature of 9 °C. All ecosystems are in
close proximity to each other on the same or very similar soil series, mainly Kalamazoo
(Fine-loamy) and Oshtemo (Coarse-loamy) mixed, mesic, Typic Hapludalfs developed
on glacial outwash.

In the ﬁrst series of experiments, reported in Chapter 2, I determined aggregate-
associated soil C pools in ten ecosystems on the same soil series along a management
intensity gradient. I also quantiﬁed the degree to which C is protected by aggregates
using size and density fractionation techniques coupled with long-term mineralization

assays of crushed and intact aggregates. Enhanced C storage relative to conventional

agriculture principally occurred in macroaggregate (>250 um) size classes. Increases in
active pool C when 2000-8000 um aggregates were broken into microaggregates (<250
um) ranged from 18% in conventional agriculture to 59% in alfalfa.

In the second set of experiments, reported in Chapters 3-5, I cultivated a never-
previously cultivated ﬁeld and minimized plant community changes to look at soil
disturbance free from the inﬂuence of other agricultural management practices. I infer
soil C permanence from responses of aggregate-protected soil organic matter, enzyme
activities that reﬂect soil microbial activity, and trace gas ﬂuxes. Cultivation
immediately reduced 2000-8000 um aggregates to levels commonly found in agricultural
soils tilled for > than 50 years. The destruction of aggregates released particulate C from
protected microsites and limited the incorporation of aboveground C into new aggregates.
This lead to a series of biogeochemical transformations due to greater soil organic matter
availability I term an aggregate cascade: microbial activity and N cycling increase,
substantially increasing ﬂuxes of both nitrous oxide and carbon dioxide. Growing plants
can modify the effects of tillage on N availability and N20 ﬂux but have little effect on
aggregation or CO; emissions.

In chapter 6, I report on an analysis of data from the till and no-till com-soybean-
wheat treatments of the KBS LTER and conclude that over 12 years. there were no yield
declines or increases in N20 emissions associated with no-till. This work was performed
in collaboration with classmates in a graduate seminar. Together, my results demonstrate

the importance and feasibility of protecting no-till soils from even occasional cultivation.

To my parents, Jeffrey and Mary Grandy, who courageously engaged challenges and
inspired critical thinking, curiosity, and open mindedness.

ACKNOWLEDGMENTS

I would like to thank Phil Robertson, my advisor, for his constant support and
shared enthusiasm for ecological science. His insight into identifying key research
questions and designing ecological experiments has been invaluable. Equally important
has been the generosity with which he has supported my overall development as a
scientist, particularly manuscript preparation, grant writing, and travel to meetings. I am
also grateful to my other advisory committee members, Steve Hamilton, Mike Klug, and
Alvin Smucker for their guidance and support. I have beneﬁted from many spirited
discussions with each.

Tim Bergsma and Claire McSwiney patiently shared information on trace gas
sampling and analysis; Terry Loecke, Sara Parr and Pongthep SuWanwaree asked critical
questions throughout the study and Terry and Sara contributed signiﬁcantly to Chapter 6
and are coauthors of this work. Andrew Corbin helped considerably with the C and N
analysis and Stacey Andres and Starr Shelton with the soil inorganic N analysis. Barb
Fox was always available to help troubleshoot graphing and soﬂrware problems. Joe
Simmons and Greg Parker supported my ﬁeld work and Laura Faber, Justin Rensch,
Carly Szekely, and Sara Wamers provided assistance in the ﬁeld and lab. Nina
Consolatti often provided logistical help, frequently on short notice, and John Gorentz
freely gave his time to help solve computer problems. Chad Brassil, Sasha Kravchenko
and Alan Tessier provided statistical advice. Kristin Huizinga and Stephanie Eichorst
explained many aspects of studying microbial communities. Sherri Morris, Elizabeth
Brewer, and Anne-Marie Fortuna provided suggestions for modeling active, slow and

passive pool C. I am particularly grateful to Sven Bohm for his many suggestions

regarding the long-term mineralization assays and modeling in chapter two and assistance
with analyzing samples for trace gases.

This work was supported ﬁnancially by the National Science Foundation through
a Doctoral Dissertation Improvement Grant, a Research Training Grant to the WK.
Kellogg Biological Station, and the WK. Kellogg Biological Station Long-Term
Ecological Research Project. Additional ﬁnancial support was provided from two USDA
Sustainable Agriculture grants, the Consortium for Agricultural Soils Mitigation of
Greenhouse Gases (CASMGS) and a Dissertation Completion Fellowship from the
College of Natural Resources at Michigan State University.

My time at KBS has been greatly enriched by interactions with the other students
and postdocs. Tara Darcy, Meg Duffy and Spencer Hall have been particularly great
friends and colleagues. Tara and Spencer have supported my work and my family in
innumerable ways.

This degree has been a shared journey with my remarkable wife Sarah. Her
humor, wisdom, creativity, and love have made pursuing this degree a joyful experience.
She has been unbelievably patient and supportive of my science although it often took me
away from home or made me mentally inaccessible. We are both lucky to have very
supportive families that have made balancing the demands of academia and family
manageable. I am also grateful to my children Marlon and Lyra who amaze me daily
and, along with Sarah, see to it that my life remains balanced. My entire family extends
our deepest thanks to Dr. and Mrs. Richard Light and family who donated the Lux Arbor
Reserve to Michigan State University. Our home for ﬁve years, this land and its wild

inhabitants will always be part the love that binds and sustains us.

vi

TABLE OF CONTENTS

LIST OF TABLES ................................................................................... ix
LIST OF FIGURES .................................................................................. xi
CHAPTER 1

Thesis Overview ...................................................................................... 1
CHAPTER 2

Carbon Storage in Soil Aggregates Along a Management Intensity

Gradient: Implications for Soil Carbon Persistence .............................................. 9
Introduction .......................................................................................... 10
Materials and Methods .............................................................................. 12
Results ................................................................................................ 19
Discussion ............................................................................................. 23
Conclusion ........................................................................................... 30
References ............................................................................................. 42
CHAPTER 3

Initial Cultivation of a Temperate-Region Soil Immediately Accelerates

Aggregate Turnover and C02 and N20 Fluxes ................................................. 46
Introduction .......................................................................................... 47
Materials and Methods ............................................................................. 49
Results ................................................................................................ 56
Discussion ............................................................................................. 60
Conclusion ........................................................................................... 66
References ............................................................................................. 76
CHAPTER 4

Changes in Aggregate-Protected Carbon, Inorganic N and Enzymes

Following Initial Tillage of an Undisturbed Soil Proﬁle ....................................... 82
Introduction .......................................................................................... 83
Materials and Methods .............................................................................. 85
Results ................................................................................................ 90
Discussion ............................................................................................. 94
Conclusion .......................................................................................... 100
References ........................................................................................... l 1 3
CHAPTER 5

Plant-Mediated Effects of Initial Cultivation on Carbon Stabilization,

Carbon Dioxide, Nitrous Oxide, and Methane ................................................ l 19

vii

Introduction .......................................................................................... 1 20

Materials and Methods ............................................................................ 123
Results ................................................................................................ 130
Discussion .......................................................................................... 1 35
Conclusion .......................................................................................... 141
References .......................................................................................... 1 6 1
CHAPTER 6

Long-Tenn Trends in Nitrous Oxide Emissions, Soil Nitrogen, and Crop Yields of Till
and No-Till Cropping Systems .................................................................. 166
Introduction .......................................................................................... 1 67
Materials and Methods ............................................................................ 170
Results ................................................................................................ 175
Discussion .......................................................................................... 177
Conclusion .......................................................................................... 183
References .......................................................................................... 1 90

viii

 

 

Table 2.1.

Table 2.2.

LIST OF TABLES

Cropping system and successional vegetation effects on soil C and N to 5 cm

soil depth at the Kellogg Biological Station Long-Term Ecological Research
Project in 2001 ....................................................................... 32

Active (Ca), Slow (Cs) and Passive (Cr) Pool C associated with ecosystems
and intact aggregate size classes ................................................... 33

Table 3.1. Cultivation effects on the distribution of soil organic matter in 2002 (DOY

Table 4.1.

Table 4.2.

Table 4.3.

Table 5.1.

Table 5.2.

Table 5.3.

Table 5.4.

236) and 2003 (DOY 294) to 20 cm. Values are means (n = 4) with
standard errors in parentheses ...................................................... 67

Tillage effects on soil C and N and bulk density at 0-7 and 7-20 cm depths.
Initial cultivation occurred on 25 June 2002 and again in the spring of 2003
and 2004 on the same plots ....................................................... 102

Tillage and depth effects on the concentration of particulate organic matter
(POM) in four density fractions and clay-associated C (<53 pm) on three
sampling dates. Initial cultivation occurred on 25 June 2002 and again in
the spring of 2003 and 2004 on the same plots ................................. 103

Tillage, depth and organic matter fraction effects on C/N ratios of different
soil fractions in 2002, 2003 and 2004. Initial cultivation occurred on 25
June 2002 and again in the spring of 2003 and 2004 on the same plots. 104

Tillage effects on top six dominant plant species by biomass and proportion of
total biomass. Samples were collected on 26 May 2004, approximately one
year after initial cultivation and before the second cultivation. Means with
standard errors in parentheses .................................................... 143

Aboveground C and N in plant and litter pools after initial cultivation in 2003
and prior to the second cultivation in spring 2004. Means with standard
errors in parentheses ............................................................... 144

Initial cultivation on soil C and N concentrations and bulk density at 0-7 and
7-20 cm sampling depths. Initial cultivation was on 15 June 2003 and again
on the same plots in the spring of 2004 .......................................... 145

Mean (interpolated) C02, N20 and CH4 ﬂuxes following initial cultivation of
sites with different vegetation management in 2003 and 2004. Individual
trace gas contributions to global warming potential (GWP) were calculated
for a period of 6 mo (corresponding approximately with the lengths of our
sampling campaign and growing season) using 100-yr time horizon factors
of310 for N20 and 21 for CH4 ................................................... 146

ix

Table 6.1. No-till soil management effects on soil aggregation, bulk density, and C
sequestration in 2001 after 12 y at the KBS LTER Site ...................... 184

Table 6.2. Global warming potential (GWP) contributions in CO2 equivalents from soil C
storage and N20 emissions in conventional and no-till systems between
1991 and 2002 at the KBS LTER Site .......................................... 184

 

 

 

LIST OF FIGURES

Figure 1.1. Description of the potential effects of tillage in soils conceptualized as an
aggregate cascade. In the model, tillage immediately reduces soil
aggregation. Reductions in the physical protection of organic matter
coupled with incorporation of aboveground C increases unprotected soil
organic matter (SOM). This increases CO2 emissions and N mineralization.
Increased NH4+ concentrations increase nitriﬁer enzyme activity, N03'
production and N20 emissions via nitriﬁcation and denitriﬁcation. Plant
grth can modify N03" ﬂuxes and N20 emissions. This cascade is
accelerated by increased soil temperature and other environmental changes
following decomposition that promote decomposition ........................... 6

Figure 2.1. Outline of the soil organic matter pools (SOM) produced by density
separation and long-term mineralization of different aggregate size classes.
SOM pools that we quantify in this study are shown within ovals ........... 34

Figure 2.2. Ecosystem effects on total sand-free C concentrations in different aggregate
size fractions at the KBSLTER. Within an aggregate size class, ecosystems
with different lowercase letters are signiﬁcantly different (p<0. 05). Within
an ecosystem, size classes with different uppercase letters are signiﬁcantly
different. Bars are means i: S. E ................................................... 35

Figure 2.3. Ecosystem effects on total sand-free C concentrations in different aggregate
size fractions at the KBS LTER. Within an aggregate size class, ecosystems
with different lowercase letters are signiﬁcantly different (p<0.05). Within
an ecosystem, size classes with different uppercase letters are signiﬁcantly
different. Bars are means :1: SE ................................................... 36

Figure 2.4. Ecosystem effects on the distribution of sand-free C in different aggregate
size classes. Results are a presented on a per unit soil basis and are thus a
function of C concentration and the proportion of soil within an aggregate
size class. Within an aggregate size class, ecosystems with different
lowercase letters are signiﬁcantly different (p<0.05). Within an ecosystem,
size classes with different uppercase letters are signiﬁcantly different.
Statistical results and stacked bars for size classes are shown from top to
bottom in the same order ........ . .................................................... 3 7

Figure 2.5. Ecosystem effects on the concentration of heavy-fraction C in different
aggregate size classes at the KBS LTER. Within an aggregate size class,
ecosystems with different lowercase letters are signiﬁcantly different
(p<0.05). Within an ecosystem, size classes with different uppercase letters
are signiﬁcantly different. Bars are means i: S.E ................................ 38

Figure 2.6. Ecosystem effects on the concentration of inter- and intra-aggregate light
fraction (LF) C in aggregate size classes. Within a column, the results on

xi

 

 

 

top are from intra-aggregate LF and those on the bottom are for inter-
aggregate LF. Within an aggregate size class, ecosystems with different
lowercase letters are signiﬁcantly different (p<0.05). Within an ecosystem,
size classes with different uppercase letters are signiﬁcantly different.
Statistical results and stacked bars for light fraction location are shown from
top to bottom in the same order ................................................... 39

Figure 2.7. Ecosystem effects on the potential release of labile C from physical
protection following aggregate destruction. Results are presented as the
difference between crushed and intact aggregates. The larger the difference
the greater the transition of C from slow to active pools. Within an
aggregate size class, ecosystems with different lowercase letters are
signiﬁcantly different (p<0.05). Within an ecosystem, size classes with
different uppercase letters are signiﬁcantly different.

Bars are means :l: S.E ............................................................... 40

Figure 2.8. Ecosystem effects on the total potential release of labile C from physical
protection following destruction of aggregates in different size classes.
Results are a presented on a per unit soil basis and are thus a function of the
increase in active pool C due to aggregate destruction and the proportion of
soil within an aggregate size class. Within an aggregate size class,
ecosystems with different lowercase letters are signiﬁcantly different
(p<0.05). Within an ecosystem, size classes with different uppercase letters
are signiﬁcantly different. Statistical results and stacked bars for size
classes are shown from top to bottom in the same order ....................... 41

Figure 3.1. Changes in soil CO2 ﬂux, moisture, temperature, and N20 ﬂux following
cultivation of a previously never tilled soil in 2002. Measurements were
made in 2002 (a; this page), 2003 (b; following page), and 2004 (c; two
pages ahead). Arrows indicate cultivation dates. Note different x-axis
ranges in each year, reﬂecting different durations of sampling. Soil
temperature and moisture samples were determined for the 0-7 cm depth.
Treatment means are shown i standard error (n = 4); * indicates statistically
signiﬁcant (P < 0.05) differences between control and cultivated treatments
within a single day of year (DOY) where there was a signiﬁcant treatment
by DOY interaction .................................................................. 68

Figure 3.2. Mean weight diameter (WMD) of soil aggregates following cultivation of a
previously uncultivated soil. Tillage occurred on day of year (DOY) 176 in
2002 and DOY 166 in 2003. Aggregation was determined for the 0-20 cm
depth. Treatment means are shown i standard error (n = 4); * indicates
statistically signiﬁcant (P < 0.05) differences between control and cultivated
treatments within a single DOY ................................................... 71

Figure 3.3. Soil aggregate distribution in four size classes. Black bars represent control
plots; patterned bars, cultivated plots. Tillage occurred on day of year

xii

(DOY) 176 in 2002 and DOY 166 in 2003. Aggregation was determined for
the 0-20 cm depth. Treatment means are shown i standard error (n = 4); *
indicates statistically signiﬁcant (P < 0.05) differences between control and
cultivated treatments within a size class and DOY .............................. 72

Figure 3.4. Distribution of inter- and intra-aggregate light fraction organic matter (LP)
in control and cultivated plots in 2002 and 2003. Samples were collected to
a depth of 20 cm on 24 August 2002, day of year (DOY) 236 and 15 June
2003 (DOY 166) and wet-sieved into aggregate size classes. 1‘, I, £5 and 11
indicate signiﬁcant differences (P<0.05) within an aggregate size class
between cultivation treatments for inter-aggregate LF, intra-aggregate LF,
total LF and the proportion of total LF within aggregates, respectively. In
2003 there was trend towards a decreased proportion of intra-aggregate LF
(p<0.054) in 2000 - 8000 um aggregates. ....................................... 73

Figure 3.5. Changes in inorganic N following initial cultivation. Arrows indicate
cultivation dates for 2002 (left panel), 2003 (middle panel), and 2004 (right
panel). Inorganic N was determined for the 0-7 cm depth. Treatment means
are shown i standard error (n = 4); * indicates statistically signiﬁcant (P <
0.05) differences between control and cultivated treatments within a day of
year (DOY) where there was a signiﬁcant treatment by DOY interaction... 74

Figure 3.6. Changes in nitriﬁer enzyme activity following cultivation on 13 October
2003 (DOY 286) and 17 August 2004 (DOY 230). Treatment means are
shown i standard error (n = 4); * indicates statistically signiﬁcant (P < 0.05)
differences between control and cultivated treatments within a day of year
(DOY) where there was a signiﬁcant treatment by DOY interactions ....... 75

Figure 4.1. Mean weight diameter (WMD) of soil aggregates. The June 2002 sampling
was done prior to cultivation. The May 2003 samples were taken prior to
the second cultivation. Treatment means are shown i standard error (11 = 4);
* indicates statistically signiﬁcant (P < 0.05) differences between control
and tilled treatments within a single sampling date. llndicates signiﬁcant
differences between depths within a treatment and sampling date .......... 105

Figure 4.2. Tillage effects on the distribution of soil in four aggregate size classes.
Numbers along the x-axis correspond with aggregate size: 2000 = 2000-
8000 pm; 250 = 250 - 2000 um; 53 = 53-250 um; and <53 = 53-250 pm.
The May 2003 samples were taken prior to the second cultivation.
Treatment means are shown i standard error (n = 4); * indicates statistically
signiﬁcant (P < 0.05) differences between control and tilled treatments
within a single sampling date. IIndicates signiﬁcant differences between
depths within a treatment and sampling date. Size classes within a sampling
date and treatment with different letters are signiﬁcantly different ......... 106

xiii

Figure 4.3. Tillage effects on total sand-free aggregate C in 2002, 2003 and 2004. Initial

cultivation was in June 2002. Treatment means are shown 2h standard error
(11 = 4); * indicates statistically signiﬁcant (P < 0.05) differences between
control and tilled treatments within a single sampling date. l'Indicates
signiﬁcant differences between depths within a treatment and sampling date.
Size classes within a sampling date and treatment with different letters are
signiﬁcantly different .............................................................. 107

Figure 4.4. Tillage effects on the distribution of sand-free inter-aggregate light fraction

organic matter in 2002, 2003 and 2004. Initial cultivation was in June 2002.
Treatment means are shown i standard error (n = 4); * indicates statistically
signiﬁcant (P < 0.05) differences between control and tilled treatments
within a single sampling date. Tlndicates signiﬁcant differences between
depths within a treatment and sampling date. Size classes within a sampling
date and treatment with different letters are signiﬁcantly different ......... 108

Figure 4.5. Tillage effects on the distribution of sand-free intra-aggregate light fraction

Figure 4.6.

Figure 4.7.

Figure 4.8.

Figure 5.1

organic matter in 2002, 2003 and 2004. Initial cultivation was in June 2002.
Treatment means are shown :1: standard error (n = 4); * indicates statistically
signiﬁcant (P < 0.05) differences between control and tilled treatments
within a single sampling date. 'lIndicates signiﬁcant differences between
depths within a treatment and sampling date. Size classes within a sampling
date and treatment with different letters are signiﬁcantly different ......... 109

Changes in inorganic N following initial cultivation. NH4+ (circles) and
N03‘ (triangles) in control plots (open symbol, solid line) and cultivated
plots (closed symbol, dashed line). Arrows indicate cultivation dates for
2002 (left panel), 2003 (middle panel), and 2004 (right panel). Inorganic N
was determined for the 0-7 cm depth. Treatment means are shown :1:
standard error (n = 4); * indicates statistically signiﬁcant (P < 0.05)
differences between control and cultivated treatments within a day of year
(DOY) and sampling depth ....................................................... 110

Changes in nitriﬁer enzyme activity following cultivation on 13 October
2003 (DOY 286) and 17 August 2004 (DOY 230). Black bars represent
control plots; patterned bars, cultivated plots. Treatment means are shown :l:
standard error (11 = 4); Bars with different letters are statistically different (P
< 0.05) ............................................................................... 111

Changes in denitriﬁer enzyme activity in August, 2004 after three tillage
events. Treatment means are shown i standard error (n = 4); Bars with
different letters are statistically different (P < 0.05) ........................... 112

Cultivation effects in a previously uncultivated ﬁeld on soil aggregation in

2003 and 2004 at 0-7 (a, this page), 7-20 (b, next page), and 0-20 (c, two
pages ahead) cm depth. Bars are arranged within sampling date by size

xiv

 

 

class going left to right: 2000-8000 urn, 250-2000 um, 53-250 um and <53
pm. No litter plots had all biomass removed prior to cultivation.

Bareground plots were weeded and contained root exclusion barriers.
Bareground plots were measured in October 2004 only. Treatment means
within a sampling date, depth, and size class followed by different uppercase
letters are statistically different (p<0.05). Within a sampling date, depth and
treatment, size classes followed by different lowercase letters are

statistically different (p<0.05). Within a sampling date, treatment and size
class, means at 0-7 cm followed by an asterisk are different from those at 7-
20 cm ................................................................................. 147

Figure 5.2. Cultivation effects in a previously uncultivated ﬁeld on soil organic matter
distribution in 2004 at 0-7 and 7-20 cm depth. Values going leﬁ to right on
the x-axis represent aggregate size classes: 250-2000 um, 53-250 um and
<53 pm, respectively. No litter plots had all biomass removed prior to
cultivation. Bareground plots were weeded and contained root exclusion
barriers. Treatment means within a depth and size class followed by
different uppercase letters are statistically different (p<0.05). Within a depth
and treatment, size classes followed by different lowercase letters are
statistically different (p<0.05). Within a treatment and size class, means at
0-7 cm followed by an asterisk are different from those at 7-20 cm. . . . .....150

Figure 5.3. Cultivation effects in a previously uncultivated ﬁeld on soil surface CO2
ﬂuxes in 2003 and 2004. No litter plots had all biomass removed prior to
cultivation. Litter plots had all aboveground biomass incorporated with
tillage. Bareground plots were weeded and contained root exclusion
barriers. Arrows indicate cultivation dates. Bars are
standard errors (n=4) ............................................................... 151

Figure 5.4. Cultivation effects in a previously uncultivated ﬁeld on soil surface N20
ﬂuxes in 2003 and 2004. No litter plots had all biomass removed prior to
cultivation. Litter plots had all aboveground biomass incorporated with
tillage. Bareground plots were weeded and contained root exclusion
barriers. Arrows indicate cultivation dates. Bars are
standard errors (n=4) ............................................................... 152

Figure 5.5. Cultivation effects in a previously uncultivated ﬁeld on CH4 oxidation in
2003. Measurements are not available for 2004. No litter plots had all
biomass removed prior to cultivation. Litter plots had all aboveground
biomass incorporated with tillage. Bareground plots were weeded and
contained root exclusion barriers. The arrow indicates the cultivation date.
Bars are standard errors (n=4) .................................................... 153

Figure 5.6. Cultivation effects in a previously uncultivated ﬁeld on soil surface N20

ﬂuxes in 2003 and 2004. No litter plots had all biomass removed prior to
cultivation. Litter plots had all aboveground biomass incorporated with

XV

tillage. Bareground plots were weeded and contained root exclusion
barriers. Arrows indicate cultivation dates. Bars are
standard errors (n=4) ............................................................... 154

Figure 5.7. Relationship between soil temperature and CO2 emissions in 2003 and 2004.
Analysis of covariance (ANCOVA) results showed a signiﬁcant temperature
effect in 2003 (p<0.003): C02 ﬂux = 14.9 + 0.026 (temperature), r2 = 0.22
(p<0.000). The line in 2003 thus represents all treatments. In 2004 there
was a temperature by treatment interaction (p<0.001). Control plots: CO2
ﬂux = 13.3 + 0.020 (temperature), r2 = 0.08 (p<0.000); no litter plots: C02
ﬂux = 12.1 + 0.033 (temperature), r2 = 0.40 (p<0.000); litter plots: CO2 ﬂux
= 13.8 + 0.018 (temperature), r2 = 0.22 (p<0.000); bareground plots: CO2
ﬂux = 15.5 + 0.026 (temperature), r2 = 0.20 (p<0.000) ....................... 155

Figure 5.8. Cultivation effects in a previously uncultivated ﬁeld on soil N0;;' in 2003 and
2004 at 0-7, 7-20, and 0-20 cm depth. No litter plots had all biomass
removed prior to cultivation. Litter plots had all aboveground biomass
incorporated with tillage. Bareground plots were weeded and contained root
exclusion barriers. Arrows indicate cultivation dates. Letters within a day
of year are stacked in order: control, no litter, litter, bareground. Different
letters indicate treatment differences within a day of year and soil depth.
Asterisks indicate differences between soil depths Within a treatment and
day of year. Bars are standard errors ............................................ 156

Figure 5.9. Cultivation effects in a previously uncultivated ﬁeld on NH4+ in 2003 and
2004 at 0-7, 7-20, and 7-20 cm depth. No litter plots had all biomass
removed prior to cultivation. Litter plots had all aboveground biomass
incorporated with tillage. Bareground plots were weeded and contained root
exclusion barriers. Letters within a day of year are stacked in order: control,
no litter, litter, bareground. Different letters indicate treatment differences
within a day of year and soil depth. Asterisks indicate differences between
soil depths within a treatment and day of year. Bars are
standard errors ...................................................................... 157

Figure 5.10. Cultivation effects in a previously uncultivated ﬁeld on mineralizable C in
2003 at 0-7, 7-20, and 0-20 cm depth. No litter plots had all biomass
removed prior to cultivation. Bareground plots were weeded and contained
root exclusion barriers ............................................................. 158

Figure 5.11. Cultivation effects in a previously uncultivated ﬁeld on denitriﬁcation
enzyme activity (DEA) in 2003 and 2004 at 0-7, 7-20, and 0-20 cm depth.
No litter plots had all biomass removed prior to cultivation. Bareground
plots were weeded and contained root exclusion barriers ..................... 159

Figure 5.12 Cultivation effects in a previously uncultivated ﬁeld on nitriﬁer enzyme
activity in 2003 and 2004 at 0-7, 7-20, and 0-20 cm depth. No litter plots

xvi

had all biomass removed prior to cultivation. Bareground plots were
weeded and contained root exclusion barriers .................................. 160

Figure 6.1. (a) Gravimetric soil moisture content in till and no-till treatments between

1989 and 2002. Letters indicate the crop harvested in that year: c = com; S =
soybean; w = wheat. * indicates statistically different responses within a
year (p<0.05) determined by slicing of the treatment by year interaction. (b)
Difference between conventional till (Ct) and no-till (Nt) normalized by the
overall mean soil moisture content for that year. Points greater than zero
indicate a till response greater than a no-till response. Year means were
signiﬁcantly different (P<0.05) and separated using Tukey’s test. Linear
regression analysis showed no trends in soil moisture concentration (p<0.05)
across all crops and years or for speciﬁc crops over time ..................... 185

Figure 6.2. (a) Soil N03‘ concentration in till and no-till treatments between 1989 and

2002. Letters indicate the crop harvested in that year: c = com; S = soybean;
w = wheat. * indicates statistically different responses within a year
(p<0.05) determined by slicing of the treatment by year interaction. (b)
Difference between conventional till (Ct) and no-till (Nt) normalized by the
overall mean N03' concentration for that year. Points greater than zero
indicate a till response greater than a no-till response. Year means were
signiﬁcantly different (P<0.05) and separated using Tukey’s test. Linear
regression analysis showed no trend in N03' concentration (p<0.05) over
across all crops and years; N03' concentrations in wheat signiﬁcantly
increased in till relative to no-till over time [normalized N03’ difference = -
15650 + 7.84 (year); r2 = 0.44; n = 3 years]. Mean 3: SE ................... 186

Figure 6.3. (a) Soil NH4+ concentrations in till and no-till treatments between 1989 and

Figure 6.4.

2002. Letters indicate the crop harvested in that year: c = com; S = soybean;
w = wheat. * indicates statistically different responses within a year
(p<0.05) determined by slicing of the treatment by year interaction. (b)
Difference between conventional till (Ct) and no-till (Nt) normalized by the
overall mean NH4+ concentration for that year. Points greater than zero
indicate a till response greater than a no-till response. Year means were
signiﬁcantly different (P<0.05) and separated using Tukey’s test. Linear
regression analysis showed no trends in soil NH.+ concentration (p<0.05)
across all crops and years or for speciﬁc crops over time ..................... 187

Figure 4. (a) Soil N20 emissions in till and no-till treatments between 1991
and 2002 (no data available for 1995). Letters indicate the crop harvested in
that year: c = com; S = soybean; w = wheat. * indicates statistically
different responses within a year (p<0.05) determined by slicing of the
treatment by year interaction. (b) Difference between conventional till (Ct)
and no-till (Nt) normalized by the overall mean N20 emissions for that year.
Points greater than zero indicate 3 till response greater than a no-till
response. There was no detectable difference between treatments based on a

xvii

 

Tukey’s test. Linear regression analysis showed no trend in N20 emissions
(p<0.05) across all crops and years; emissions from soybean signiﬁcantly
increased in till relative to no-till over time [normalized N20 difference =
31210 - 15.7 (year); r2 = 0.48; n = 4 years]. Mean i S.E ................... 188

Figure 6.5 (a) Crop yields in till and no-till treatments between 1989 and 2002. Letters
indicate the crop harvested in that year: c = corn; 5 = soybean; w = wheat. *
indicates statistically different responses within a year (p<0.05) determined
by slicing of the treatment by year interaction. (b) Difference between
conventional till (Ct) and no-till (Nt) normalized by the overall mean crop
yield for that year. Points greater than zero indicate a till response greater
than a no-till response. Year means were signiﬁcantly different (P<0.05)
and separated using Tukey’s test. Linear regression analysis showed no
trend in crop yields (p<0.05) across all crops and years; wheat yields
signiﬁcantly increased in till relative to no-till over time [normalized yield
difference = -5980 + 3.00 (year); r2 = 0.40; n = 3 years].

Mean i S.E .......................................................................... 189

xviii

 

CHAPTER 1
Dissertation Overview

Radiative forcing of the Earth’s atmosphere is increasing at unprecedented rates
due to human activities. Since 1750, atmospheric concentrations of C02, N20, and CH4
have increased by more than 30, 150, and 17%, respectively (IPCC, 2001). Agriculture
has contributed substantially to global ﬂuxes of these gases and now represents an
important mitigation option as carbon sequestration and greenhouse gas abatement in
soils are two of a limited number of rapidly-deployable, high impact C02 stabilization
options now available to policy makers (Kauppi et al., 2001; Caldeira et al., 2004).
Although management strategies such as cover-cropping and improved N fertilizer
management potentially reduce the global warming potential of agricultural ecosystems,
eliminating tillage may have the broadest impact because of its potential to alter
emissions of all three biogenic greenhouse gases.

The area of cropland under no-till management in the US. between 1994 and
2004 rose from 14 to 23%, but only a fraction of this is in continuous no-till (CTIC,
2004). This raises questions about the permanence of stored C, which Marland et al.
(2001) identiﬁed as the fundamental challenge to terrestrial C sequestration. Further, no-
till deployment has been limited in part because it may accelerate N leaching (Martens,
2001) and N20 emissions (MacKenzie et al., 1997; Ball et al., 1999; Baggs et al., 2003)
but decrease N availability and yields (Rice et al., 1986; Niehues et al., 2004). In this
dissertation I address the ecological processes underlying soil organic matter permanence

and the ecosystem and agronomic consequences of long-term no-till.

Short-term total soil C changes are notoriously difﬁcult to detect because of high
background C concentrations in most arable soils, spatial variability, and the
redistribution of above-ground pools upon cultivation (Sollins et al., 1999). The few
studies in temperate ecosystems on SOM permanence have nonetheless focussed on
short-term total soil C change (e. g. Tiessen and Stewart, 1983; Bowman et al., 1990;
Pierce et al., 1994; VandenBygaart and Kay, 2004). Because of this, I believe they do
not adequately predict the long-term effects of converting a long-term no-till ﬁeld or
successional community to conventional tillage; these studies also do not predict the
effects of periodic tillage of no-till soils. The consequences of both these strategies on C
storage and trace gas ﬂuxes may take decades to fully understand. Here I take a different
approach: I infer permanence from short-term responses to tillage of aggregate-protected
soil organic matter, enzyme activities that reﬂect soil microbial activity, and trace gas
ﬂuxes. I, therefore, attempt to advance our ability to predict tillage effects on soils by
understanding the basic ecological processes controlling soil’s response to disturbance.
In Chapter 2, I determine aggregate-associated soil C pools in ten ecosystems on

the same two co-occurring soil series along a management intensity gradient. I also
assess the degree to which C is protected within aggregates using size and density
fractionation techniques coupled with long-term mineralization assays of crushed and
intact aggregates. I found that relative to conventional agriculture, enhanced C storage
Primarily occurred in 2000-8000 um aggregate size classes. Increases in active pool C
when 2000-8000 um aggregates were broken into microaggregates (<250 um) ranged

from 18% in conventional agriculture to 59% in alfalfa. Potential release of whole-soil

labile C from physical protection following macroaggregate destruction was seven to
nine-fold greater in successional systems than conventional ag1iculture.

These C losses following aggregate destruction in successional systems led me to
question the persistence of soil organic matter in long-term no-till soils. How fast do
aggregates break down following tillage and what are the biogeochemical consequences
of aggregate destruction following tillage? Many studies have shown that long-term
tillage depletes aggregation (e. g. Six et al. 2000) but the immediate effects of soil
disturbance are not as well known. Further, authors have speculated that a substantial
portion of structural breakdown following cultivation is due to changes in plant
communities, organic inputs and long-term C losses (Paustian et al. 1997; 2000). I
hypothesize that immediate and persistent reductions in soil aggregation occur after
cultivation in conjunction with changes in soil temperature and other environmental
controls over decomposition and that rapid changes in SOM turnover, microbial activity
and trace gas ﬂuxes will occur after a single plowing. These changes can be
conceptualized as an “aggregate cascade” (Figure 1) whereby changes in aggregation
occur immediately and have far-reaching effects that include changes in microbial
communities and nitrous oxide ﬂuxes — as well as changes in organic C protection.

In Chapters 3, 4, and 5 I examine the effects of tillage (moldboard plowing and
discing) in a previously undisturbed soil. Tillage in temperate ecosystems has been
examined almost exclusively in soils with long histories of agricultural management.
The legacies of prior tillage, fertilization and annual cropping effects on soil structure,
microbial communities, aggregation, and other factors that control trace gas ﬂuxes may

last for more than a century, confounding the inﬂuence of recent tillage. Because of this,

our basic understanding of tillage effects on soils and, in particular, our ability to predict
ﬁiture effects of tillage in long-term no-till soils is limited.

In Chapter 3, I report that cultivation reduced aggregate mean size by 35% within
60 d and that these effects were persistent. Mean additional daily C02-C losses due to
cultivation ranged between 1 and 2 g C m2 d’1 and N20 ﬂuxes increased from 2 to 6-fold
over the three years following initial tillage and N20 ﬂuxes were associated with
increased soil N03' concentrations, which approached concentrations found in fertilized
ﬁelds. In Chapter 4 I focus on the changes in aggregation following cultivation and its
relationship to SOM availability and microbial processes. In Chapter 5, I look at the
ability of plant litter and growing plants to modify the effects of tillage on aggregation,
microbial activity and ﬂuxes of C02, N20 and CH4.

The work supports the idea of an aggregate cascade (Figure 1). The destruction
of aggregates releases particulate C from protected microsites and also limits the
incorporation of aboveground C into new aggregates. Instead, C increases in unprotected
inter-aggregate pools where it is quickly oxidized. This stimulates microbial activity and
N cycling, substantially increasing ﬂuxes of both nitrous oxide and carbon dioxide.
Growing plants can modify the effects of tillage on N availability and N20 ﬂux but have
little effect on aggregation or CO2 emissions. This cascade is accelerated by increased
soil temperature and other environmental changes following decomposition that promote
decomposition.

In the ﬁnal, sixth chapter I perform an analysis of data from the till and no-till
treatments of the KBS LTER and conclude that over 12 years there were no yield

declmes or increases in N20 emissions associated with no-till. This work was performed

in collaboration with classmates in a graduate seminar class using data from the KBS
LTER data catalogue.

Results show that cultivation immediately destabilizes physical and microbial
processes related to C and N retention in soils. We also demonstrate that no-till cropping
can be practiced with no yield or environmental trade-offs. Together, our results
demonstrate the importance and feasibility of protecting no-till soils from periodic
cultivation. This information will become increasingly valuable as greenhouse gas

mitigation policies are further developed and deployed.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Above-ground litter

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Plant
growth

 

 

 

 

 

 

 

 

Cultivation
Soil 1 2
aggregation
Aggregate
formation
L
Inter-
aggregate,
unprotected
LF
SOM oxidation _ C02 T
emissions
1
NH,+ T
Nitriﬁer enzyme 1‘
activity
~ . 2 ‘ nitrification
‘ denitriﬁca‘tieﬁ‘“
N03'T “3 N20 T
7 emissions

 

 

 

 

 

 

Figure 1.1. Description of the potential effects of tillage in soils conceptualized as an aggregate
Case-ade. In the model, tillage immediately reduces soil aggregation and prevents the long-term
Stabdilation of newly formed aggregates. Reductions in the physical protection of organic matter
Coupled with incorporation of aboveground C increases unprotected light fraction (LF) soil organic

In

.mattel‘ (SOM). This increases CO2 emissions and N mineralization. Increased NH4+ concentrations
(”lease nitriﬁer enzyme activity, N03' production and N20 emissions via nitriﬁcation and

e . . . - . . . .
n_'tl‘1f1cat1on. Plant growth can mod1fy N03 ﬂuxes and N20 emissrons. This cascade rs accelerated
dec'ncreased soil temperature and other environmental changes following tillage that promote
or“position and can be reversed by regeneration of aggregates following conversion to no-till.

REFERENCES

Baggs, E. M., M. Stevenson, M. Pihlatie, A. Regar, H. Cook, and G. Cadisch. 2003.
Nitrous oxide emissions following application of residues and fertiliser under zero and
conventional tillage. Plant and Soil 254:361-370.

Ball, B. C., A. Scott, and J. P. Parker. 1999. Field N20, C02 and CH4 ﬂuxes in relation to
tillage, compaction and soil quality in Scotland. Soil & Tillage Research 53:29-

39.

Bowman, R. A., J. D. Reeder, and R. W. Lober. 1990. Changes in soil properties after in
a central plains rangeland soil after 3, 20, and 60 years of cultivation. Soil Science

150:851-857.

Caldeira, K., M. G. Morgan, D. Baldocchi, P. G. Brewer, C. T. A. Chen, G.-J. Nabuurs,
N. Nakicenovic, and G. P. Robertson. 2004. A portfolio of carbon management
options. Pages 103-130 in C. Field and M. Raupach, editors. The Global Carbon
Cycle. Island Press, Washington, DC, USA.

IPCC. 2001. Climate Change 2001; Synthesis Report. Cambridge University Press,
Cambrdge, UK.

Kauppi, P., and R. D. Sedjo. 2001. Technological and economic potential of options to
enhance, maintain, and manage biological carbon reservoirs and geo-engineering.
Pages 301-344 in B. Metz, 0. Davidson, R. Swart, and J. Pan, editors. Climate
Change 2001, Mitigation. Cambridge University Press, Cambridge, UK.

MacKenzie, A. F., M. X. Fan, and F. Cadrin. 1997. Nitrous oxide emission as affected by
tillage, com—soybean-alfalfa rotations and nitrogen fertilization. Canadian Journal

of Soil Science 77:145-152.

Marland, G., K. Fruit, and R. Sedjo. 2001. Accounting for sequestered carbon: the
question of permanence. Environmental Science & Policy 3:259-268.

Martens, D. A. 2001. Nitrogen cycling under different soil management systems.
Advances in Agronomy 70: 143-192.

Ni ehues, B. J ., R. E. Lamond, C. B. Godsey, and C. J. Olsen. 2004. Starter nitrogen
fertilizer management for continuous no-till corn production. Agronomy Journal

96:1412-1418.

Paustian, K., H. P. Collins, and E. A. Paul. 1997. Management controls on soil carbon.
Pages 15-50 in E. A. Paul, K. Paustian, E. T. Elliott, and C. V. Cole, editors. Soil
organic matter in temperate agroecosystems. CRC Press, New York.

Paustian, K., J. Six, E. T. Elliott, and H. W. Hunt. 2000. Management options for
reducing CO2 emissions from agricultural soils. Biogeochemistry 48:147-163.

 

Rice, C. W., M. S. Smith, and R. L. Blevins. 1986. Soil nitrogen availability after long-
tenn continuous no-tillage and conventional tillage corn production. Soil Science
Society of America Journal 50:1206-1210.

Six, J ., K. Paustian, E. T. Elliott, and C. Combrink. 2000. Soil structure and organic
matter: 1. Distribution of aggregate-size classes and aggregate—associated carbon.
Soil Science Society of America Journal 64:681-689.

Sollins, P., C. Glassman, E. A. Paul, C. Swanston, K. Lajtha, J. W. Heil, and E. T. Elliott.
1999. Soil carbon and nitrogen: pools and fractions. Pages 89-105 in G. P.
Robertson, D. C. Coleman, C. S. Bledsoe, and P. Sollins, editors. Standard Soil
Methods for Long-Term Ecological Research. Oxford University Press, New
York.

Tiessen, H., and J. W. B. Stewart. 1983. Particle size fractions and their use in studies of
soil organic matter: II. cultivation effects on organic matter composition in size
fractions. Soil Science Society of America Journal 47 :509-514.

VandenBygaart, A. J ., and B. Kay. 2004. Persistence of soil organic carbon aﬁer plowing
a long-term no-till ﬁeld in southern Ontario, Canada. Soil Science Society of
America Journal 68: 13941402. '

CHAPTER 2

Carbon Storage in Soil Aggregates along a Management Intensity Gradient:

Implications for Soil Carbon Persistence

ABSTRACT

The recovery of aggregate-associated organic C pools is an important mechanism
for increased soil C storage following reductions in land use disturbance. The persistence
of this stored C, however, remains a major uncertainty in C sequestration forecasts. We
determined aggregate-associated soil C pools in ten ecosystems on the same soil series
along a management intensity gradient in southern Michigan (USA). We also assessed
the degree to which C is protected by aggregation using size and, density fractionation
techniques coupled with long-term mineralization assays of crushed and intact
aggregates. Ecosystems included four annual row-crop systems (conventional, no-till,
low input, and organic), two perennial cropping systems (alfalfa and poplar), and four
native communities (early successional, midsuccessional historically tilled,
midsuccessional never-tilled, and late successional forest). Enhanced C storage relative to
conventional agriculture ranged from 150 in low input row crops to 1890 g C rn’2 in late
successional forest and principally occurred in heavy-fraction C pools (SOM >1.6 g cm'3
Plus SOM <53 um) and macroaggregate (>250 um) size classes. In the 2000-8000 urn
size class, no-till cropping and early successional communities increased C from 4.1 to
8.5 and 15.7 g C kg'l sand-free whole soil, respectively. Active pool C increases when
2000-8000 um aggregates were broken into microaggregates (<250 um) ranged from

18% in conventional agriculture to 59% in alfalfa. Potential release of whole-soil labile

C from physical protection following macroaggregate destruction was seven to nine-fold

greater in successional systems than conventional agriculture. Our results suggest that
conventional agriculture releases microaggregates and light fraction organic matter from
within macroaggregates and C in these pools is then rapidly oxidized. Legume cover
crops, no-till, perennial crops and unmanaged successional ecosystems can enhance soil
C storage but the potentially rapid destruction of macroaggregates following tillage raises

concerns about the long-term persistence of these C pools.

INTRODUCTION

Restoring some fraction of terrestrial soil C pools through changes in agricultural
management is a high impact, rapidly deployable strategy for partially mitigating
increases in atmospheric CO2 (Caldeira et al. 2004, CAST 2004, Pacala and Socolow
2004). Management strategies that increase soil C storage include reducing tillage
intensity and increasing residue inputs with cover crops, green manures, or perennial
crops (West and Post 2002, Lal et al. 2004). These practices modify decomposition rates
by changing soil aeration, water dynamics, and aggregation as well as the biochemistry
and quantity of crop residues (Angers and Caron 1998, Martens 2000).

The quantity and turnover rates of soil aggregates may be a particularly important
control over soil organic matter (SOM) protection following changes in cultivation or
cropping intensity (Cambardella and Elliott 1994, J astrow et al. 1996, Paustian et al.
2000). Evidence for protection of SOM by aggregates includes low interior-aggregate
Oxygen concentrations that limit respiration rates, small intra-aggregate pore spaces that

llmrt SOM access to decomposers, and increases in soil respiration rates when SOM is

10

released from within aggregates (Sexstone et al. 1985; Mikha and Rice 2004). Field

studies have demonstrated that C sequestration in afforested soils and restored grassland
ecosystems principally occurs within soil aggregates (Six et a1. 1998; J astrow et a1. 1996;
DeGryze et a1. 2004).

The mechanical impact of tillage can immediately destroy a substantial proportion
of macroaggregates while simultaneously exposing soil surfaces to the impacts of rain
and freeze-thaw and wet-dry cycles. In a recent study, Grandy and Robertson (2005, in
review) found that soil in the 2000-8000 urn aggregate size class declined from 0.34 to
0.19 g g'1 after plowing once a previously uncultivated ﬁeld, levels identical to those in
adjacent agricultural ﬁelds on the same soil type continuously tilled for >50 y. This
immediate effect of cultivation on soil structure raises concerns about the persistence of
stored soil C following periodic cultivation of no-till soils, a common practice in no-till
management. The area of cropland under no-till management in the US. between 1994
and 2004 rose from 14 to 23%, for example, but only a fraction of this is in continuous
no-till (CTIC 2004). Stored C might also be lost rapidly from sites in the USDA
Conservation Reserve Program and other set-aside programs upon cultivation.

Robertson et al. (2000) demonstrated the potential for cropping system
management and succession to modify total soil C storage in 10 ecosystems on the same
soil type between 1989 and 1999 at a site in the northern U.S. corn belt. Here, we
investigate the microsite distribution of C in these ten ecosystems and the susceptibility
of physically-protected C to loss following aggregate destruction. We combine physical
SOM separation techniques (e. g. Elliott 1986, Conant et a1. 2004, DeGryze et a1. 2004)

with direct estimates of the effects of soil structure on long-term C mineralization rates

11

 

(e.g. Plante and McGill 2002, Mikha and Rice 2004) to provide additional insights into

the nature of C accumulation in different size fractions and the vulnerability of C in
these size fractions to aggregate destruction. Our speciﬁc objectives were to determine
the effects of cropping system management (conventional, low input, organic, no-till, and
perennial) and ecological succession (early, mid, and late old-ﬁeld succession) on: 1)
stabilizing water-stable soil aggregates in different size classes; 2) enhancing C storage in
physically protected aggregate-associated C pools; and 3) controlling changes to the size
and decay rates of physically-protected C pools following soil structural disturbance. We
hypothesize that soil aggregation and total soil C storage will increase in agricultural
systems where legume cover-crops, no-till, or perennial crops are used and that late
successional ecosystems will have the greatest structural stability and C storage. Further,
we hypothesize that C preferentially accumulates in intra-aggregate C pools, particularly

in macroaggregate size classes that are especially susceptible to loss upon tillage.

MATERIALS AND METHODS
Experimental Site and Approach
Our experimental site is a series of ecosystems that differ in management intensity

located at the Kellogg Biological Station (KBS) Long-Tenn Ecological Research (LTER)
site (Table 2.1). KBS is located in SW Michigan and receives ca. 90 cm of precipitation
annually, about half as snow and a mean annual temperature of 9 °C. All ecosystems are
in close proximity to each other on the same or very similar soil series; mainly
Kalamazoo (Fine-loamy) and Oshtemo (Coarse-loamy) mixed, mesic, Typic Hapludalfs

developed on glacial outwash. These two series co-occur in all ecosystems and differ

12

mainly in their Ap horizon texture, although variation within a series can be as great as
variation between series (Crum and Collins 1995, Robertson et al. 1997).

The experimental ecosystems include four annual cropping systems, two
perennial cropping systems, and four successional plant communities (Table 2.1). The
annual cropping systems are com-soybean-wheat rotations and include four management
regimes: 1) conventional chemical management with tillage; 2) tilled, low chemical
input; 3) tilled organic; and 4) no-till. Both low input and organic management systems
have a leguminous winter cover crop (T rzfolium pretense L.) to provide nitrogen in two
out of every three years. No systems receive compost or manure and the standard input
systems receive inorganic N fertilizers according to regional best management practices.
The perennial crops include poplar trees (Populus sp.) on a 10-year rotation cycle and
alfalfa (Medicago sativa) on a 6-8 year rotation. The successional communities include
recently abandoned, 12 yr-old early successional old-ﬁelds, historically tilled 50-year-old
midsuccessional communities, a never-tilled 50-year old midsuccessional community,
and a set of late successional oak-hickory forests that were never cleared or plowed. The
annual and perennial cropping systems as well as the early successional community are
replicated in six one ha plots within the 60 ha KBS LTER main experimental site. These
treatments were all established in 1989 in a previously conventionally managed row-crop
ﬁeld. Midsuccessional historically tilled, midsuccessional never-tilled, and late
successional ecosystems are replicated at three different locations within a 2 km radius of
the main site on the same soil series. These ecosystems have been organized along a
management intensity gradient based on tillage, external inputs, and above-ground net

primary productivity (Table 2.1).

13

Our overall approach was to sieve soils from each site into 4 aggregate size
classes and characterize the total organic C associated with each aggregate size fraction
using size and density fractionation techniques, as well as to assess the potential for
aggregate structure to control the distribution of C in biologically-deﬁned pools by

comparing long-term mineralization dynamics of crushed and intact aggregates.

Soil Sampling and Storage

Previous research at KBS (De Gryze et al. 2004) and other studies (e. g. West and
Post 2002) have shown that C accumulates primarily near the soil surface following
cessation of tillage. To better understand the mechanisms underlying this accumulation
and the persistence of accumulated C we sampled to 5 cm. Soil samples were collected
from ﬁve locations within each plot in June and July, 2001. At each of the ﬁve sample
locations, two subsamples with a diameter of 7.6 cm to a soil depth of 5 cm were taken
by gently hammering a PVC core into the ground to minimize compression and slicing of
aggregates. In row-crop ecosystems, one of the subsamples at each location was taken in
the row and the other between rows. All ten subsamples from each plot were combined
to produce one representative sample for each of the 52 plots. Four separate samples for
bulk density analysis were taken at the same time as those for aggregate analysis, using
an 8 cm diameter root corer.

Field-moist soil samples were put into a cooler (4 °C) prior to being broken along
natural fracture planes and passed through an 8 mm sieve within 72 h of the sampling

time. Rock fragments > 8 mm were discarded. After sieving, soils were dried at room

14

temperature in paper bags prior to storage in plastic bags. Care was taken throughout
the study to minimize disturbance of the samples that might inﬂuence aggregate

SII'UCILII'C.

Water-Stable Aggregate Distribution

Aggregate distribution was determined on four replicate 100 g air-dried soil
samples by wet-sieving in water through a series of 2000 um, 250 um, and 53 um sieves
(Cambardella and Elliott 1993, Six et al. 1998). Soil was submerged for 5 min on the
surface of the 2000 um sieve which was then moved up and down for two minutes with a
stroke length of 3 cm for 50 strokes. Sieving was repeated on the 250 um (50 strokes)
and 53 um (30 strokes) sieves using the soil plus water that passed through the next
larger sieve. Aggregates remaining on each sieve were dried at 60 °C. Sand content was
determined on an aggregate subsarnple after dispersing soil in sodium

hexarnetaphosphate (0.5%) for 48 h on a rotary shaker at 190 rpm.

Aggregate-Associated Light Fraction Organic Matter
The method we used to separate inter- and intra-aggregate light fraction (LF;
organic matter of relatively low density) is based on previously published protocols (Six
et al. 1998, Gale et al. 2000). Aggregate subsamples were pre-wet prior to LP analysis in
order to minimize aggregate destruction during LF separation. An 8 g subsample of
aggregates was divided in half and placed on two membrane ﬁlters (47 mm diameter; Pall
Supor-450) overlaying two paper ﬁlters (70 mm diameter; Whatman 42) in a 10 cm Petri

dish. Four mL of deionized (DI) water were trickled onto the paper ﬁlters in order to

15

slowly wet all of the aggregates by capillarity. Aggregates were transferred from the

membrane ﬁlters to 100 mL beakers after 16 h with 5 mL aliquots of sodium
polytungstate (NAPT) at a density of 1.62 g cm'3. A total of 55 mL NAPT was used for
each sample. A preliminary test showed that the ﬁnal density of NAPT was about 1.60 g
cm'3 following equilibration with the water contained in aggregates.

After 24 h, LF was aspirated from the surface of the NAPT and then rinsed on a
hardened, ashless ﬁlter paper with at least 600 mL DI H2O. We refer to this pool as inter-
aggregate LF. After removal of this pool, we aspirated the remaining NAPT. Aggregates
were then dispersed to release the intra-aggregate LF using sodium hexametaphosphate
as described previously and resuspended in NAPT (d=1.62 g cm'3). The intra-aggregate
LF was collected from the surface. The mineral-associated aggregate C plus POM with a
density >1.6 was determined by difference and is referred to as heavy-fraction C (HF).
Organic C and total N concentrations of organic matter and whole soil samples were
determined by dry combustion and gas chromatography in a CHNS analyzer (Costech
ECS 4010, Costech Analytical Technologies, Valencia CA.). Inorganic carbonates were
not a signiﬁcant interference in these soils to organic C measurement (S.K. Hamilton,

pers. comm.)

Laboratory Incubations
A subsample of aggregates (15-20 g) was transferred into 60 mL glass serum vials
with a 13 mm diameter opening. We estimated the bulk density of each of our samples
after tamping down the serum vials 10 times on a laboratory bench and from this

determined the amount of water needed to bring them to 55% water-ﬁlled pore space

16

 

(WFPS). It has been shown that microbial activity is highest at this moisture content

across a range of soil types (Linn and Doran, 1984; A.J. Franzluebbers, personal
communication). Water applications were made slowly via 5 mL pipettes to minimize
breakdown of aggregate structure following rewetting.

After wetting the samples, the serum vials were placed in a '/2 pint jar with ca. 60
mL of water in the bottom. These jars were covered with polyﬁlm that permits relatively
ﬁ'ee O2 and CO2 exchange but retains water. Jars were put into boxes and then into dark
incubation chambers maintained at 25 °C. Samples were periodically checked for water
loss by weighing them and rewetted, as necessary.

We added additional sand to the <53 urn size class to minimize 02 depletion. The
2000-8000 um size class contained an average of 45% sand and, the 250-2000 um size
class an average of 60% sand while the <53 um size class contained an average of 33%.
Sand additions consisted of particles with a diameter between 250-1000 um and brought
the average sand content in this size class up to 48%.

Respiration in all size classes was measured a minimum of 12 times over 205 d
with greater sampling intensity early in the incubation. At each sampling date serum
vials were ﬂushed for 45 s with a humidified air stream. After ﬂushing, bottles were
sequentially capped with a rubber septum. A 0.5 mL sample of headspace was
immediately drawn with a syringe and then two additional samples were taken over a 90
min sampling interval. CO2 content of each gas sample was analyzed using an infrared
gas absorption (IRGA) analyzer, followed by calculation of the respiration potential for

the time interval (Robertson et al. 1999).

17

Active, slow and passive pool C associated with different aggregate size
fractions were determined by modelling the long-term respiration data (Paul et al. 2001).
We used a differentiated version of a standard three-pool ﬁrst order model to

accommodate discontinuous sampling:

dC/dt = C.*k.e""“""”s’ + (Csoc — Cr — C.) * lee-km”) + C. * k.e<‘kr*days>

where Ca and ka are the active C pool size and decay rate constant, CS and k5 are the slow
pool size and decay constant, and Cr and kr are the resistant pool C size and decay rate
constant. Csoc is total soil C. Ca, ka and k5 were determined by modelling; slow pool C
(C,) was determined by difference and Cr was determined by acid hydrolysis. All
detectable plant residues and POM were removed with tweezers prior to hydrolysis
because of the potential for relatively young lignin to resist hydrolysis and inﬂate Cr
estimates (Paul et a1. 2001). After removal of these materials soil samples were ground
with a mortar and pestle. Hydrolysis was carried out for 16 h at 110 °C in 110 mL test
tubes containing 2 g soil and 20 mL 6N HCl. The kr was assumed to be 8.3-104’ (1'1 (Paul
et al. 2001). Carbon dating of Cr in previous studies at our site has demonstrated that that
this pool ranges from hundreds to thousands of years old (Paul et al. 1997), resulting in a
mean residence time so large and a kr so small that deviations of the assumed value from
the actual decay rate constant have little affect on the other parameters (Paul et al. 2001;
Fortuna et al. 2003).

To determine the potential for aggregate structure to control the distribution of C,

an additional 15 - 20 g subsample of aggregates was crushed prior to carrying out the

18

long-term mineralization assays. Aggregates ranging in size from 2000-8000 and 250-

2000 um were fractured to create microaggregates by passing them through a 250 um
sieve. The structure of aggregates in the 53-250 um size class was destroyed by crushing
aggregates in a mortar and pestle. Total potential physical protection of C by aggregates
was estimated from increases in Ca associated with aggregate destruction where the
difference in Ca between crushed and intact aggregates was positive. Samples where

differences were negative were analyzed separately.

Statistical Analysis

Statistical analysis was carried out using a completely random-design analysis of
variance (ANOVA) with the Proc Mixed procedure in SAS (SAS Version 8.2, SAS
Institute 1999). Data were analyzed by considering ecosystem and aggregate size class as
ﬁxed effects after log transformation where necessary to improve homogeneity of
variance. Single degree of freedom comparisons were made using the LSD statistic to
calculate a 95% conﬁdence interval around the differences between means generated
using the diff option in Proc Mixed. The LSD was carried out using the PDMIX800
algorithm (Saxton 1998). Distribution of C in different size fractions as well as potential
increases in Ca due to aggregate breakdown were analyzed per gram aggregate and also
on a whole soil basis after making adjustments for the percentage of soil in a particular
size class. Aggregates and aggregate-associated SOM pools were corrected for sand >53
urn.

RESULTS

Total Soil C and N

19

Relative to conventional agriculture, increases in soil C concentration from 0 to

5 cm occurred with no-till (48%), low input (24%) and organic (32%) treatments (Table
2.1). Perennial crops increased soil C concentrations relative to conventional by 63% in
alfalfa and 68% in poplar stands. The early successional community increased soil C
concentrations by >100% relative to conventional agriculture and contained C
concentrations similar to the historically tilled midsuccessional community, although soil
C concentrations were substantially less than those in the never-tilled midsuccessional
soil and deciduous forest. Organic N concentrations also increased across the
management intensity gradient. C/N ratios were similar among agricultural systems but

increased in perennial and successional communities (Table 2.1).

Aggregate C Distribution

All ecosystems except for low input had an increased mass of soil in the 2000-
8000 um size class aggregates relative to conventional (Figure 2.2). No-till increased the
mass of soil in the 250-2000 um aggregate class and decreased the proportion of soil in
smaller size fractions relative to conventional (Figure 2.2). Alfalfa and poplar
ecosystems and the successional communities increased the mass of soil in the 250-2000
um class relative to conventional.

All ecosystems except for zero-input increased total C concentrations in the 2000-
8000 um aggregate size class (Figure 2.3). In the 250-2000 um size class, low input,
perennial and successional ecosystems increased total C relative to conventional
agriculture. Poplar, early successional communities, and the mid and late successional

communities increased C in the 53-250 um size class relative to the annual agricultural

20

treatments. In the <53 um size class the successional treatments increased total C
relative to the annual agricultural treatments (Figure 2.3). C concentrations in macro-
and micro-aggregates were similar in mid and late successional communities but in other
ecosystems C concentrations were lower in microaggregates. Expressed on a whole soil
basis after correcting for the amount of soil in a particular size class, increased C storage
relative to conventional agriculture primarily occurred in the two macroaggregate (>250
um) size fractions (Figure 2.4). Heavy fraction organic matter accounted for most of the

C difference between ecosystems (Figure 2.5)

In the 2000-8000 um aggregate size class, the inter-aggregate LF concentrations
were indistinguishable among annual cropping systems and early successional
communities (Figure 2.6). Midsuccessional historically tilled, midsuccessional never-
tilled, and late successional ecosystems increased inter-aggregate LP by 301 , 247, and
503%, respectively, compared to conventional agriculture (Figure 2.6). In the 250—2000
um aggregate size class, inter-aggregate LF was generally similar among ecosystems
(Figure 2.6). In the 53-250 um size class, inter-aggregate LF was similar among annual
cropping systems. All four successional communities had higher inter-aggregate LF than
the agricultural treatments.

Intra-aggregate LP in the 2000-8000 um size class was similar among agricultural
treatments; alfalfa, poplar, and early successional ecosystems all increased LF relative to
conventional ag1iculture. Poplar increased intra-aggregate LF relative to conventional,
organic, and no-till agriculture. The mid and late successional communities had similar
intra-aggregate LF but early successional ecosystems had lower intra-aggregate LF than

did midsuccessional historically tilled and late successional systems. Intra-aggregate LP

21

was higher in organic than in no-till or conventional ecosystems. Intra-aggregate LF
concentrations in poplar and early successional ecosystems were greater than those in
conventional and alfalfa systems. Early and late successional systems had greater intra-

aggregate LF than midsuccessional historically tilled ones.

Active, Slow and Passive C

Annual and perennial cropping systems and early successional communities
generally had similar active C pool sizes while the mid and late successional
communities had greateractive pool C than the other ecosystems (Table 2.2). Low input,
organic and no-till organic management increased slow pool C relative to conventional
agriculture. Alfalfa and early successional systems had greater slow pool carbon (Cs)
than annual cropping systems other than no-till. The mid and late successional
communities had greater C, than other treatments. Annual crop management had no
effect on resistant pool C (Cr). Alfalfa increased Cr relative to conventional and organic
agriculture; poplar and early successional systems increased Cr relative to annual row
crop systems. Alfalfa had the lowest k,; the other perennial systems and successional
communities had higher ks than the annual systems. Total C, Ca and Cs were greater in
large and medium sized aggregates than small ones while ka and kS were greater in small

and medium aggregate sizes (Table 2.2).

Physical Protection of C

In the 2000-8000 um size class, no-till, perennial crops, and successional

treatments had greater transfer of C from C5 to Ca following aggregate breakdown than

22

did conventional agriculture. Increases in Ca ranged from 98 mg C kg'1 in conventional

agriculture to 626 mg C kg"l aggregated soil in midsuccessional historically tilled
communities (Figure 2.7). The potential transfer of C from slow into active pools
following aggregate destruction represents a large proportion of the active pool C of
intact aggregates (Table 2.2; Figure 2.7): conventional (18%); low input (50%); organic
(26%); no-till (29%); alfalfa (59%); poplar (46%); early successional (52%);
midsuccessional historically tilled (29%); midsuccessional never tilled (43%); and late
successional (32%). Crushing aggregates had no effect on ka and effects on ks were
generally small or nonsigrriﬁcant (data not shown).

Differences in aggregate distribution and potential transfer of C5 to Ca led to large
differences in the effects of aggregate breakdown expressed on a whole soil basis (Figure
2.8). All treatments except low input increased potential C losses from 2000-8000 um
aggregate size classes. The increase in Ca following breakdown of 2000-8000 um
aggregates was 4-fold greater in no-till than conventional cropping systems (86.1 vs. 20.4
mg C kg”l whole soil). Differences were particularly great in successional communities
where increases in Ca on a whole soil basis were 7-fold greater in the two
midsuccessional communities and 9 times greater in late successional systems.
Breakdown of 250-2000 um aggregates increased C transfer from C3 to Ca in the
successional communities while destruction of microaggregates generally had less effect
(Figure 2.8).

In each aggregate size class there was a total of 55 comparisons between crushed
and intact aggregates. In the 2000-8000 um size class three of these comparisons were

negative, indicating a transfer of C from C, to Cs following aggregate crushing (data not

23

 

shown); in the 250-2000 urn size class ﬁve comparisons were negative (data not

shown); and in the 53-250 um size class 29 comparisons were negative.

DISCUSSION
Soil C Storage

Our results demonstrate the potential for no-till soil management, cropping
intensity and successional age to enhance total soil C storage and that this is related to
changes in aggregation and the distribution of C in different size fractions (Table 2.1).
Increases in soil C over 12 yr in no-till are equivalent to an annual C increase in the top 5
cm of 26 g C rn'2 y". These rates are similar to those reported by Robertson et al. 2000
(30 g C m‘2 y") for this no-till system between 1989 and 1999. Although these rates are
lower than those reported in some studies (e. g. West and Post 2002) they are consistent
with results presented in recent review papers (Davidson and Ackerman 1993; Six et al.
2004) and are similar to reported average C accumulation rates for the Midwest US.
(Franzluebbers and Steiner 2002).

Increases in C with organic and low input systems demonstrate the potential for
legume cover crops to increase soil C pools, although at lower rates than no-till cropping.
Some studies have reported that C additions from legume cover crops are relatively small
and therefore insufﬁcient to increase total soil C (MacRae and Mehuys 1985). However,
long-term use of legume cover crops can increase SOM relative to rotations that consist
exclusively of residues with higher C :N ratios, perhaps because of effects on aggregation
and/or microbial communities (Drinkwater et al. 1998, Grandy et al. 2002). The potential

effects of plant communities on soil C were further demonstrated by the increased C

24

 

measured in perennial cropping systems and early successional communities relative to

no-till since these communities were converted from conventional agriculture in 1989;
SOM differences between these treatments are likely a function primarily of plant
community structure and productivity and their effects on residue quality and quantity,

aggregation, and microbial communities.

Aggregate Stability

Our results show potential for tillage and plant communities to inﬂuence soil

A-.r- -‘L ~

aggregate size distributions. It is widely recognized that tillage decreases
macroaggregation (e.g. Six et al. 2000, Mikha and Rice 2004, Wright and Hons 2004).
The destructive effects of tillage on soil structure begin to occur immediately after
cultivation and are persistent (Grandy and Robertson 2005a, in review). Opinions
regarding the effects of plant communities on soil structure, however, remain varied as
residue quality and quantity, as well as root growth, may inﬂuence aggregation (Angers
and Caron 1998, Kavdir and Smucker 2005). Aboveground biomass inputs from the
organic and low input systems are similar to those from conventional crops (KBS LTER,
2005) suggesting that root dynamics or the quality of residues may be increasing
aggregation.

Legume cover crop decomposition may rapidly stimulate production of
polysaccharides and fungal hyphae capable of stabilizing soil aggregates (Haynes and
Beare 1996). Legumes grown in crop rotations can increase aggregation but in
monoculture legumes may decrease aggregation because their stabilizing effects are

short-lived relative to residues with higher C/N ratios (Haynes and Beare 1996, Wright

25

 

and Hons, 2004). Greater macroaggregation in poplar and early successional

communities than in alfalfa highlights the potential for soil structural declines in legume
monocultures. Additional evidence that residue quality differences and/or belowground
dynamics are controlling aggregate stabilization is that poplar produces similar
aboveground biomass to the agricultural row-crop systems while early successional

communities produce less (DeGryze et al. 2004).

Aggregate Associated C Pools

Heavy-fraction (HF) organic matter, rather than light fraction (LF), accounted for
the largest proportion of C increases across ecosystems. For example, in the no-till
system differences in HF accounted for 82% of the C increase in 2000-8000 um
aggregates and in midsuccessional never-tilled systems HF accounted for 80% of the
difference in total C. Six et a1. (1999) found that between 50 and 66% of new C
associated with macroaggregates was in the form of mineral-associated C, and Jastrow et
al. (1996) similarly found that C deposition in macroaggregates occurs principally in
mineral-associated pools and that particulate soil organic matter accounted for only 20%
of C gain.

Differences in HF among size classes within ecosystems were proportionally
smaller than LF differences, and HF differences explained a smaller proportion of the
variation in total C among size classes. In the no-till system the difference in C
concentration between the 2000-8000 um and 53-250 urn size classes was 8.1 g kg'1 and
47% of this was accounted for by differences in HF. In early successional systems, HF

was similar in the 2000-8000 (24.7 g kg'l) and 53-250 (24.4 g kg") um size classes so the

26

 

differences between size classes in total C concentration could be accounted for entirely
by changes in LF. In midsuccessional historically-tilled, midsuccessional never-tilled,
and late successional systems there were no differences in HF between size classes.
There were differences in LF, although they were small relative to total pool sizes and did
not result in a change in total C across aggregate size classes. These results suggest that
LF and HF C pools are an important component of C differences between size classes but
that between ecosystems HF C accounts for most of the total C gain. Further, succession

reduces differences in total C and heavy C between size classes.

‘-.1 J“

The lack of agricultural management effects on inter-aggregate LP is similar to
other reports that this pool is highly sensitive to any kind of agricultural management and
that differences primarily occur between managed and unmanaged ecosystems (Arrouays
and Pelissier 1994, Six et al. 1999); its accumulation may occur slowly, driven primarily
by plant inputs, root dynamics, soil disturbance, and other activities that inﬂuence the

decomposition environment (DeGryze et al. 2004).

SOM Mineralization
When aggregate structure was left intact there were no differences in active pool
C (C...) among the annual or perennial agricultural treatments or early successional
communities (Table 2.2). There were differences, however, in the concentration of C in
the slow (Cs) and resistant pools (C,) of intact aggregates, demonstrating that C
accumulation occurred principally in SOM pools that are chemically or physically
protected. Those ecosystems with the greatest inter-aggregate POM were also the ones

with the greatest Ca, suggesting that some portion of the free, unprotected LF pool

27

 

contributes to Ca.

Increases in Ca following the breakdown of large aggregates in no-till, alfalfa,
poplar and early successional systems suggest that physical mechanisms make important
contributions to C accumulation in the 2000-8000 urn size class and that the
incorporation of LF and mineral-associated C into aggregates will decrease its
decomposition rate. Fewer differences among treatments in the 250-2000 um size class
and no differences when microaggregates were crushed are likely due to greater chemical
protection of SOM in these size classes. Microaggregates contain higher concentrations
of humic materials and other degraded and chemically complex compounds than do
macroaggregates (Tisdall and Oades, 1982; Haynes and Beare, 1996; .

The potential transfer of C from slow into active pools following aggregate
destruction is within range of those from other comparisons of C mineralization rates
from crushed and intact aggregates (Balesdent et a1 2000). Elliott (1986) found in a
cultivated soil that breaking macroaggregates down to microaggregates increased C
mineralization 19% in a cultivated soil and 4% in a previously uncultivated ﬁeld. In a
review of experimental comparisons of crushed and intact aggregates, Balesdent et al.
(2000) concluded that the protective capacity of aggregates increases with SOM
concentration, clay content, and the absence of tillage. Our results support the hypothesis
that protective capacity is related to SOM concentration and tillage intensity.

Potential whole soil additional C losses with soil disturbance (Figure 2.8) are a
function of aggregate C content, tillage intensity, and also aggregate size distribution.
Successional plant communities should thus be highly susceptible to C losses following

cultivation. They have high SOM concentrations within macroaggregates and also

28

 

contain >30% soil in macroaggregate size classes that break down immediately

following cultivation. In a recent study, Grandy and Robertson (2005a, in review)
demonstrated that initial cultivation of the midsuccessional never-tilled communities
reduced 2000-8000 um aggregates within 60 d to levels indistinguishable from those in
conventional ag1iculture. Concomitantly, there were signiﬁcant decreases in the
proportion of LF protected within aggregates and CO2 ﬂuxes increased an average of 1.0
to 1.9 g C In2 (1'1 over three years, likely as a result of decreased aggregate protection of
LP.

The potential for oxidation rates to decrease following aggregate destruction was
demonstrated when Ca in crushed aggregates was smaller than that in intact aggregates.
This principally occurred in 53-250 mm aggregates. Other researchers have also reported
decreases in respiration rates when soil structure is destroyed (Balesdent et al. 2000).
Although we added supplemental sand to this size class it is possible that anaerobic sites
reduced respiration rates or that crushing facilitated organo-mineral interaction that

protected C.

Management Implications
SOM losses following a decade or more of cultivation range from 30-60% of original
surface soil carbon (Davidson and Ackerman 1993). Restoring a portion of this oxidized
C is one of a limited number of rapidly-deployable, high impact CO2 stabilization options
now available to policy makers (Caldeira et al. 2004). Widespread adoption of no-till
cropping and other soil C conservation strategies could re-sequester up to 0.5 - 1.0 Pg soil

C y'1 and contribute signiﬁcantly to stabilizing global CO2 levels (Caldeira et al. 2004;

 

29

CAST 2004). Our results demonstrate that physical protection of C by aggregates is an

important mechanism controlling SOM turnover in managed ecosystems and that 2000-
8000 um aggregates may reduce active pool C turnover by as much as 50%. Ultimately,
however, however, net greenhouse gas mitigation by due to soil C storage depends on the
long-term persistence of stored soil organic matter (Paustian et al. 2000; Pacala and
Socolow 2004). No-till soil management in the US, although increasing, continues to be
rotated with tillage because of perceptions that no-till limits nitrogen availability and
decreases yields (Martens 2001). This practice increases aggregate turnover rates,
oxidizes C, and reduces the potential for agricultural land to sequester C. In a recent
analysis of long-term data from the WK. Kellogg Biological Station, Grandy et al.
(2005b, in review) demonstrated no negative effects of continuous no-till on yields,
inorganic N availability, or N20 ﬂuxes over 12 yr. Further, Six et a1. (2004) demonstrate
that long-term no-till may decrease N20 ﬂuxes and that this may be due to long-term soil
structural improvements in no-till. Practices such as no-till and legume cropping and set-
aside programs that increase soil C storage within aggregates should thus be continuously

maintained long-term to maximize potential ecosystem beneﬁts.

CONCLUSIONS
Our results support theories that agricultural soil C losses near the soil surface can
be partially reversed by using less intensive cultivation and enhancing plant community
complexity. We found that the highest C accumulation rates occur in perennial cropping
systems and early successional communities. Across ecosystems, C accumulation

relative to conventional agriculture principally occurred in heavy-fraction ((1 >16 g cm'3

30

 

 

or particle sizes <53 urn), slow, and acid-resistant pools of macroaggregates >250 pm.

The 2000-8000 um size range demonstrated the greatest increases in active pool C
following aggregate destruction. This suggests that macroaggregates (and their
constituent microaggregates) have the greatest protective capacity across a range of
ecosystems with contrasting histories of plant and soil management but raises concerns
about the persistence of sequestered C. The vulnerability of macroaggregates to
destruction following tillage intensiﬁcation and substantial shifts of C from physically
protected slow pools into active pools following aggregate destruction demonstrates the
need to protect stabilized SOM from increases in tillage intensity. Greatly increased
macroaggregation and enhanced protective capacity of macroaggregates in perennial
crops and successional ecosystems underscores the need to protect these systems, in

particular, from even occasional tillage and other increases in management intensity.

31

 

.momoﬁcosa E 38.5 “ENE—Sm 5S» meeovf

 

 

 

 

 

Gee mam ﬁve 3; 3m: m: S3 82 :98 :._ 82: NE 63: we... seem 25:2er
2 I: an 92.8 v.2 5d 8. A28 28 95.8 85 898 mg 95.8 NS. $20 22
5.8 9% ﬁve as 3.8 m: 9H: 8: see e: :98 owe 82.8 new E: 22
5:: one 3.2: n: 3:; _: 2&3 SS 32: a; :3: w; 3:: 3N ram
moEanEoU 356883
8:: Se 3.2: w: 3.3 See 8.3 82 A38 2: :98 Se 8:: of 5%..
Gee 3e A58 2: see o: 3.3 2 : :3: m: :98 25 A58 a: are?
320 Eccoeom
no.8 see 5.8 N? 83V 2: A23 82 33: e2 :3: m; 28.8 E... =: oz
:58 w _ .e 2&8 23 5.3 m. a 5.8 So :58 e2 :2: m3 :3: ME .28 3&2 E agate
5:: were 3.8 was 893 new :33 5 A38 42 898 :5 A38 n: a$8 25?. E gas 33
3.8 8e 8:: a: sad 2: 93 Se 2:8 as :3: Se :2: :2 aeeeeeaeu
€03.88 «3:3-cmonxdm-€obv macho 3:52
N.E w N.E w m.Ee w .x. ﬁx.
:e 23m 25 z 38 o 33 base 0:5 z 38 o Eek

 

.Loom E 8085 :oemomom .eo_wo_00m Suﬁ—twee:
=o_§m _8_wo_o_m memo—EM 2: we save :8 Eu m 9 Z 25 O :8 :o acute 223%? 1283888 use 889m wEnaBO ._.m 2ng

32

Table 2.2. Acive (Ca), Slow (C,)and Passive (C,) Pool C associated with ecosystems and intact
aggregate size classes”

 

 

 

 

 

 

CPool
Active Slow Reslstant
Ca ka Lmrt Fmrt C, ks Lmrt Fmrt C,
mg 8' dl ('10'2) d" d" mg g" d" (-10“‘) y" y" mg g"
- '2

Conventional 0.545Ac 4.91 20.7 62.8 11.4Ac 3.06Bbc 10.3 31.3 7.366
Low input 0.593Ac 5.43 18.6 56.3 15.8Abc 2.56bc 11.7 35.4 9.66de
Zero input 0.568Ac 5.77 17.4 52.8 13.4Abc 3.60ABab 8.29 25.1 9.67de
No till 0.585Ac 5.22 19.5 59.2 14.1Abc 3.78ab 8.38 25.4 11.2cd
Alfalfa 0.577Ac 4.343 23.2 70.4 17.5Ab 1.73c 16.8 50.911.4cd
Poplar 0.447c 5.01 20.2 61.2 14.8bc 4.7OBa 6.12 18.5 11.4cd
Early 0.576Ac 5.05 20.6 62.3 15.9Abc 4.00ab 7.05 21.4 14.8c
Mid-HT 2.17ABa 3.193 32.1 97.2 47.0Aa 3.003 9.42 28.6 28.8b
Mid-NT 1.39ABb 3.368 32.9 99.6 30.6ABa 3.95 7.72 23.4 28.6ABb
Late 1.32ABb 3.07 33.0 100 28.1ABa 3.20 9.27 28.1 4.94a
250-2000 um size class
Conventional 0.353Bbc 5.24 23.1 70.0 8.16Bc 4.63Ab 9.49 28.8 9.67d
Low input 0.372Abc 6.21 17.1 51.8 16.9Ab 3.98b 7.33 22.2 11.1cd
Zero input 0.330Abc 6.46 16.6 50.5 13.9Ab 4.60b , 6.11 18.5 11.9cd
No till 0.386Abc 6.22 16.5 49.9 14.6Ab 3.82b 7.51 22.8 12.9cd
Alfalfa 0.352Abc 5.298 20.9 63.3 17.5Ab 1.50Ac 19.2 58.1 11.9cd
Poplar 0.251c 7.60 15.7 47.7 13.1b 6.78Aa 4.20 12.7 15.3bc
Early 0.356Abc 6.14 17.2 52.2 18.1Ab 5.00ab 5.73 17.4 15.7b
Mid-HT 1.26Aa 4.36AB 32.8 99.5 34.1ABa 6.77a 4.54 13.7 24.6ab
Mid-NT 0.405Abc 4.94AB 22.8 69.0 44.2Aa 5.12ab 6.58 20.0 19.3Ccd
Late 0.686Ab 4.37 23.5 71.2 40.4Aa 3.90b 7.53 22.8 39.6a
_53-250 um size clgss
Conventioanl 0.177c 5.14bc 20.9 63.5 5.31Cc 3.93AB 7.20 21.8 9.64d
Low input 0.23OBbc 6.63ab 17,5 53.1 7.94Bb 3,27 917 27.8 10.2d
Zeroinput 0.19ch 5.01bc 20.4 61.7 7.53Bb 3.08 10.7 32.6 9.46d
No till 0.24SBbc 4.26c 40,7 123 7.528b 3.22 12.6 33,1 10.1d
Alfalfa 0.142bc 8.29Aa 15.1 45.7 954% 2.8213 12.1 36.8 12.4cd
POplar 0.189bc 6.60abc mg 51.0 10.5b 4.668 6.51 197 16.5bc
Early 0.243Bbc 6.1821bc 17,3 539 10.1Bb 4,30 6.49 197 15.6c
Mid-HT 0.588Bab 6.51A 154 46.6 24.7Aa 4.47AB 6.18 13,7 27.1ab
Mid-NT 0.476Babc 6.26A 132 55,] 2].SBa 6.33 4,73 14,3 38.6Aa
13¢ 0.867Ba 5.57abc 20,0 60.7 24.7821 577 4_79 145 42.5a

 

i - . . . . . . .
Wrthln an aggregate Slze class, ecosystems Wlth dlfferent lowercase letters are srgnlﬁcantly dlfferent
(p<0.05). Within an ecosystem, size classes followed by different uppercase letters are signiﬁcantly

different.

Ca and k,, represent active pool C and kinetics, k, represents slow pool kinetics, and C, and k, represent
resistant pool C and kinetics. Ca, ka and k5 were determined by modeling; slow pool C (Cs) was
determined by difference. k, was assumed to be 83-104)6 d".
Laboratory mean residence time (Lmrt) was calculated as UK Field mean residence time (Fmrt) was
determined after doing a Qlo correction for the difference in lab temperature (25°C) and ﬁeld mean
temperature at KBS (90°C).

33

GEE» 56% 2a beam £5 5 €ch 25 85 28m 20m .8320 8mm commocwmm “subway mo gonna—«SEE
€8-98— ?a 5:883 bane—o 3 326on G209 Econ “BEE £5qu :8 05 mo 0:280 .mm onE

 

 

 

60¢
96me
a
SUI 29
£9.92; 68% $3
Eom cozm~=SwEE
€22-32
mmummmammm 853.6 .0
SEE maEmmnzm m mm - me

 

 

 

 

9E0 m 009
8:26: 26ch

‘ 0 8860339652

38 6 8e
c0390: 36ch

_ maEmmnsw m 9w _

 

ago 6% 28202 ~

 

.20.... mummwammmézE

S_On_ + Ucmw

_ooa 60¢ at: 08 $06
265 w>uo< new $635

 

      

ow. _.A 20m

  

 

ll.Y 016916.338 220 + :_w M.

 

   

>
llllY 20a 663.3395

   

 

 

      

 

 

 

 

E 0.60“ 2000—8000 pm a
U) a a
“"7
go 045“ be b
01% cd be
935, 030—
CHI)
0):—
<8 015$
(U
°’ 0.007
§ 0.60-250—20000m a a
U)!-
9'6:
“503
050
0')!—
<8
(U
0)
§ 0.60753—250pm
(I)!-
9'0:
was
8‘3.
030)
mi...
<8
“I
O)
E
(D
(D‘—
22b)
‘00)
8‘5
Lou
ma)
(3);.
<8
(13
O)
0'00 —~5.92aa_>'f§2
S .E g g “—- ‘5 ﬂ 9 32 -'
(D O
> _1
I:
O
0

Figure 2.2. Ecosystem effects on total sand-free C concentrations in different aggregate size
fractions at the KBS LTER. Within an aggregate size class, ecosystems with different
lowercase letters are signiﬁcantly different (p<0.05). Within an ecosystem. size classes with
different uppercase letters are signiﬁcantly different. Bars are means i 5.13.

35

Total aggregate C Total aggregate C
9 kg'1 9 kg'1

9 k9"

Total aggregate C Total aggregate C
9 kg"

 
     
 
 

 

 
 
    
 

 

   

 

    

 

  

Aa Aa
100 - 2000—8000 11m
75-«
50.1
Abc Acd Abe Ab Abc Ab
25— Ad
0..
100 7 250—2000 pm 3,, A3
75"
2 A
50 A8 Acd Ade Ade Acd Acd °
25‘
0—1
100- 53—250um A8
75_ Ba A3
50‘ ADC 80
25 _ Be Bde Bde Bde 3°“
0‘
100- (53%
75“
50“
Ca Ba Ba
25‘ CchdeBbc Cb
0" _ c.
g *5 .2 E i! g a I '2 .3
-f-3 .3 5 ° g 6 u“) ‘9 T9 4
c 9 Z < o. E 2
a) 3 O
> O
C _l
O
0

Figure 2.3. Ecosystem effects on total sand-free C concentrations in different
aggregate size fractions at the KBS LTER. Within an aggregate size class,
ecosystems with different lowercase letters are signiﬁcantly different (p<0.05).

Within an ecosystem, size classes with different uppercase letters are signiﬁcantly
different. Bars are means i SE.

36

ﬁ

.590 083 05 E Seton 9 m8 80¢ 8527. Be momma? 0N6 8m was @385 93 83.6.2 Eoumcﬁm

.Eeoﬁme Nassau—mama. 08 ESE 03825: Eoeobmc £3, momma—o 86 .806mean SW 555/ Amodvﬁ “cobbmw
bocmommemmm 2m 3032 33.532 Sesame 53> mEBmkmOoo .220 mu? owawouwwm :a 23:3 .820 on? 8&8wa
cm 55:5 :8 mo comtomoa 05 98 :otmbcoocoo 0 mo :03er a 2:: 03 can memes :8 ES be a so @8536
m 0.8 338% .mommﬂo 0N6 8&8wa Eobbﬁ E 0 02.3653 mo cousngﬂe 2: so 38me Eowmzmoom in oeswE

_ 9181

— J.H le
— A1193

 

 

om<
m<
m0
mm

 

 

— Je|dod

_ l‘v’lleil‘v’

— 1111 0N
——1ndug 0192
—— 1ndu1 M01

 

 

 

 

 

 

 

 

 

 

 

 

ono

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

o< no< on< on<
6?. B< 68 68 Bo
So no Em E E
8m 8m
E: mmv
E: 833
E: 88103

E; 88 1 88.

 

[if |euonuaAu03

nos e|ouM eeii-pues ,-6)1 6
3 912691669 19101

I
_H_
D

 

[I

 

mm

om

mu

2:

37

Heavy Fraction C Heavy Fraction C

Heavy Fraction C

9 kg"

9 kg."

100‘“

2000 - 8000 pm a

 

100—

 

250 - 2000 pm a

 

100‘“

 

53 — 250 pm

ab a

    

Bde Bde

 

 

No tIII
Alfalfa
Polplar
Early
Mid NT
Late

1—
I
E
2

Low input
Organic

Conventional

Figure 2.5. Ecosystem effects on the concentration of heavy-fraction C in different
aggregate size classes at the KBS LTER. Within an aggregate size class, ecosystems
with different lowercase letters are signiﬁcantly different (p<0.05). Within an
ecosystem, size classes with different uppercase letters are signiﬁcantly different. Bars
are means 3: S13.

38

  
    

  

 

 

 

 

      

 

 

(U
C
.9
+4
C
0)
>
C
O
0

Figure 2.6. Ecosystem effects on the concentration of inter— and intra-aggregate light fraction
(LF) C in aggregate size classes. Within a column, the results on top are from intra-aggregate

LF and those on the bottom are for inter-aggregate LF.

  

   

73 20— 2000—8000 pm
(I)
c 'o 15 lntra-aggregate LFC
o co -
*3 *5 1:1 Inter-aggregate LFC
E 8 Ade Abc
LL 5) 10 “ e Af Acde Abcd
"‘ Acd
.6) 8 5 _ ACd Acd ACd Ad .
x
a) 0 I
E 207 250—2000 pm
(I)
'0
5 2 15 -
.8 to Aa
(U U)
L: a 10‘ Af Adf Abd AbC Ab Babcd Ba
e ce C
E 8 A3 Bf Aab Aa Ba, Bdef
.33.? 5— Aab Baby Aab Bab Bb . ,3 Bab
°’ N H
:0 ;Q__ l
l l l I l
8 207 53—250pm
13
.52 15—
*5 cu
m c»
*- 9 _
it 8 10
fail“ B B C 13 B B C B
‘4 to
x m
m m... - _
_ L 1—
‘5 ‘5’ E % .02 g E z ,3
Q (0 ° ~_— ‘—1 .12 .12
3 o
O
_1

Within an aggregate size class,

ecosystems with different lowercase letters are signiﬁcantly different (p<0.05). Within an
ecosystem, size classes with different uppercase letters are signiﬁcantly different. Statistical
results and stacked bars for light fraction location are shown from top to bottom in the same

order.

39

    

     
    
 

 

 

£81200—
'0
236 2000-8000 m a
eta 9004 “ Aa
CU)
TE
AU) _
93(0) 600
'0‘?
23 300- Aab Aab
mo
3c»
$581200—
68 250—2000um
9‘5 900-
CO)
T92
”EU! __1 a
8‘? 600
E9 300- ab
«3 o ABb Aab Aab 8,,
52’ 0.
A=1200—
g8
58 900—
£9.13
58
flag 600—
94:
E2? 300—
w M
3
0% 0—

Conventional
Low input
Zero input
No till
Alfalfa
Polplar

Early

M1d HT

Mid NT

Late

Figure 2.7. Ecosystem effects on the potential release of labile C from physical protection
following aggregate destruction. Results are presented as the difference between crushed and
intact aggregates. The larger the difference the greater the transition of C from slow to active
pools. Within an aggregate size class, ecosystems with different lowercase letters are
signiﬁcantly different (p<0.05). Within an ecosystem, size classes with different uppercase
letters are signiﬁcantly different. Bars are means :t S.E.

40

.820 08mm 2: E Seton 9 a8 Got 5527.

2m 3320 3mm how was woo—03m can 3.3.2 Eozmzﬁm .thomtv banEcw_m 8a 8032 838%: Engage 53» 8320 36
5.29308 5 £55» Amodvav EBQEE banEcw_m 98 £33— 38332 Engage £3» mEomxmooo awn—o oﬁm oawwohwwm
cm 55;? .330 on? Samohwwm am 5:23 :8 mo 5an8.“ 05 use .8525me 8&8me 9 26 D Eon 038m 5 8835

2: mo 532.3 a 35 8a was £93 :8 20:3 oo¢6§m a co 3289a 8a $.33— .mommﬂo 0N5 Evacuate E moamwohwwm

.6 cotuzbmov wage—.8 5:8on 323% 80¢ Aaov U 038“ no 9.328 33:82. :38 2: :o acute Ecumxmoom .wd 8sz

 

 

 

 

 

 

m

2 .I U

m. o m

m a d v m. m. u
D. B m. H... N U U w...
H U M. WI: 0 d d U
l_ .A m B m m. m pk

_ _ _ _ p _ o

 

 

   

 

 

 

 

 

 

 

 

 

 

l l 2:
o< o
E o l 8N
onm< onm<
onm<
22 I 8”

 

 

:5 omwém U l oov

E: ooom I omN D

l. com
ca 88 .. 88

"08 elouM t-ﬁx 0 W
(90) 1091Ul - (Bo) pausmo

 

 

l 000

41

REFERENCES

Angers DA, Caron J. 1998. Plant-induced changes in soil structure: Processes and
feedbacks. Biogeochemistry 42: 55-72.

Arrouays D, Pelissier P. 1994. Changes in carbon storage in temperate humic loamy soils
after forest clearing and continuous corn cropping in France. Plant Soil 160: 215—
223. '

Balesdent J, Chenu C, Balabane M. 2000. Relationship of soil organic matter dynamics to
physical protection and tillage. Soil & Tillage Research 53: 215-230.

Caldeira K, Morgan MG, Baldocchi D, Brewer PG, Chen CTA, Nabuurs G-J,
Nakicenovic N, Robertson GP. 2004. A portfolio of carbon management options.
Field C, Raupach M, editors. The global carbon cycle. Washington, DC: Island
Press. p103-130.

Cambardella CA, Elliott ET. 1993. Methods for physical separation and characterization
of soil organic matter fractions. Geoderma 56: 449-457.

Cambardella CA, Elliott E. 1994. Carbon and nitrogen dynamics of soil organic matter
fractions from cultivated grassland soils. Soil Science Society of American
Journal 58: 123-130.

Conant RT, Six J, Paustian K. 2004. Land use effects on soil carbon fractions in the
southeastern United States. 11. Changes in soil carbon fractions along a forest to
pasture chronosequence. Biology and Fertility of Soils 40: 194-200.

CAST. 2004. Climate change and greenhouse gas mitigation: challenges and
opportunities for agriculture. Ames, Iowa, USA: Council for Agricultural Science
and Technology (CAST).

Crum, J .R., and HP. Collins. 1995. KBS Soils [Online]. Available at
www.lter.kbs.msu.edu/soil/characterization. W. K. Kellogg Biological Station
Long-Term Ecological Research Project, Michigan State University, Hickory
Corners, MI.

CTIC (Conservation Technology Information Center). 2004. 2004 National Crop Residue

Management Survey [Online]. Available at www.ctic.purdue.edu/ctic/crm.htm1.
CTIC, West Lafayette, IN.

Davidson EA, Ackerman IL. 1993. Changes in soil carbon inventories following
cultivation of previously untilled soil. Biogeochemistry 20: 161-193.

DeGryze S, Six J, Paustian K, Morris SJ, Paul EA, Merckx R. 2004. Soil organic carbon
pool changes following land-use conversions. Global Change Biology 10: 1120-
1 132.

42

Drinkwater LE, Wagoner P, Sarrantonio M. 1998. Legume-based cropping systems have
reduced carbon and nitrogen losses. Nature 396: 262-265.

Elliott ET. 1986. Aggregate structure and carbon, nitrogen, and phosphorous in native
and cultivated soils. Soil Science Society of American Journal 50: 627-633.

Fortuna A, Harwood RR, Kizilkaya K, and Paul EA. 2003. Optimizing nutrient
availability and potential carbon sequestration. Soil Biology & Biochemistry
35:1005-1013.

Franzluebbers AJ, Steiner J L. 2002. Climatic inﬂuences on C storage with no tillage. .
Kimble JM, Lal R, Follett RF, editors. Agriculture practices and policies for
carbon sequestration in soil. Boca Raton (FL): CRC Press. p71-86.

Gale W, Cambardella C, Bailey T. 2000. Surface residue- and root-derived carbon in

stable and unstable aggregates. Soil Science Society of American Journal 64: 196-
201.

Grandy AS, Porter GA, Erich M. 2002. Organic amendment and rotation crop effects on
the recovery of soil organic matter and aggregation in potato cropping systems.
Soil Science Society of American Journal 66: 131 1-1319.

Haynes RJ, Beare MH. 1996. Aggregation and organic matter storage in meso-therrnal,
humid agricultural soils. In: Structure and organic matter storage in agricultural
soils. Carter MR, Stuart BA, editors. New York: CRC Press. p213-262.

J astrow JD, Boutton TW, Miller RM. 1996. Carbon dynamics of aggregate-associated
organic matter estimated by carbon-13 natural abundance. Soil Science Society of
American Journal 60:801-807.

Kavdir Y, Smucker AJM. 2005. Soil aggregate sequestration of cover crop root and
shoot-derived nitrogen. Plant and Soil 272:263-276.

KBS LTER. WK. Kellogg Biological Station Long-Term Ecological Research Project.
2005. Available at www.lter.kbs.msu.edulData/DataCataloghtrnl. KBS LTER,
Hickory Corners, MI.

Lal R, Grifﬁn M, Apt J, Lave L, Morgan MG. 2004. Managing soil carbon. Science 304:
393-393.

Linn, D. M., and J. W. Doran. 1984. Effect of water-ﬁlled pore space on carbon dioxide
and nitrous oxide production in tilled and nontilled soils. Soil Science Society of
America Journal 48:1267-1272.

MacRae RJ, Mehuys GR. 1985. The effects of green manuring on the physical properties
of temperate-area soils. Advances in Soil Science 3: 71-94.

43

Martens DA. 2000. Plant residue biochemistry regulates soil carbon cycling and carbon
sequestration. Soil Biology and Biochemistry 32: 361 -369.

Mikha MM, Rice CW. 2004. Tillage and manure effects on soil and aggregate-

associated carbon and nitrogen. Soil Science Society of American Journal 68:
809-816.

Pacala S, Socolow R. 2004. Stabilization wedges: Solving the climate problem for the
next 50 years with current technologies. Science 305: 968-972.

Paul EA, Follett RF, Leavitt SW, Halvorson A, Peterson GA, Lyon DJ. 1997.
Radiocarbon dating for determination of soil organic matter pool sizes and
dynamics. Soil Science Society of American Journal 61: 1058-1067.

Paul EA, Morris SJ, Bohm S. 2001. The determination of soil C pool sizes and turnover
rates: biophysical fractionation and tracers. Lal R, editor. Assessment methods for
soil carbon. Boca Raton (FL): Lewis Publishers. p193-206.

Paustian K, Six J, Elliott ET, Hunt HW. 2000. Management options for reducing C02
emissions from agricultural soils. Biogeochemistry 48: 147-163.

Plante AF, McGill WB. 2002. Soil aggregate dynamics and the retention of organic
matter in laboratory-incubated soil with differing simulated tillage frequencies.
Soil Tillage and Research 66: 79-92.

Robertson GP, Klingensmith KM, Klug MJ, Paul EA, Crum JR, Ellis BG. 1997. Soil
resources, microbial activity, and primary production across an agricultural
ecosystem. Ecological Applications 7: 158-170.

Robertson GP, Paul EA, Harwood RR. 2000. Greenhouse gases in intensive agriculture:

contributions of individual gases to the radiative forcing of the atmosphere.
Science 289: 1922-1925.

Robertson GP, Wedin D, Groffrnan PM, Blair JM, Holland EA, Nadelhoffer KJ, Ham's
D. 1999. Soil carbon and nitrogen availability: nitrogen mineralization,
nitrification, and soil respiration potentials. Robertson GP, Coleman DC, Bledsoe

CS, Sollins P, editors. Standard soil methods for long-term ecological research.
New York (NY): Oxford University Press. p25 8-271.

Saxton AM. 1998. A macro for converting mean separation output to letter groupings in
Proc Mixed. In: Proc. 23rd SAS Users Group Intl. Nashville (TN): Sas Institute.
p1243-1246.

Sexstone AJ, Revsbech NP, Parkin TB, Tiedje JM. 1985. Direct measurement of oxygen

profiles and denitriﬁcation rates in soil aggregates. Soil Science Society of
American Journal 49: 645-651.

44

Six J, Elliott ET, Paustian K. 1999. Aggregate and soil organic matter dynamics under
conventional and no-tillage systems. Soil Science Society of American Journal 63:
1350-1358.

Six J, Elliott ET, Paustian K, Doran JW. 1998. Aggregation and soil organic matter
accumulation in cultivated and native grassland soils. Soil Science Society of
American Journal 62: 1367-1377.

Six J, Paustian K, Elliott ET, Combrink C. 2000. Soil structure and organic matter: 1.
distribution of aggregate-size classes and aggregate-associated carbon. Soil
Science Society of American Journal 64: 681-689.

Six J, Ogle SM, Breidt FJ, Conant RT, Mosier AR, Paustian K. 2004. The potential to
mitigate global warming with no-tillage management is only realized when
practised in the long term. Global Change Biology 10: 155-160.

Tisdall, J. M., and J. M. Oades. 1982. Organic matter and water-stable aggregates in
soils. Journal of Soil Science 33:141-163.

West TO, Post WM. 2002. Soil organic carbon sequestration rates by tillage and crop
rotation: a global data analysis. Soil Science Society of American Journal 66:
1930-1946.

Wright AL, Hons FM. 2004. Soil Aggregation and carbon and nitrogen storage under

soybean cropping sequences. Soil Science Society of American Journal 68: 507-
513.

45

CHAPTER 3
Initial Cultivation of a Temperate-Region Soil Immediately Accelerates Aggregate
Turnover and C02 and N20 Fluxes
ABSTRACT
The immediate effects of tillage on protected soil C and N pools and ontrace gas
emissions from soils at pre-cultivation levels of native C remain largely unknown. We
measured the response of C02 and N20 emissions and associated environmental controls
to cultivation in a previously uncultivated US. Midwest Alﬁsol with organic C
concentrations indistinguishable from those in adjacent old-growth forests on the same
soil type (3.2%). Within 2 days of initial cultivation, tillage signiﬁcantly (P < 0.001 , n =
4) increased C02 ﬂuxes from 91 to 196 mg COz-C m'2 hr'1 and within the ﬁrst 30 (1
higher ﬂuxes due to cultivation were responsible for losses of an additional 85 g COz-C
m‘z. Additional daily C losses were sustained during a second and third year of
cultivation at rates of 1.9 and 1.0 g C m'2 d], respectively. Associated with the C02
responses were increased soil temperature, substantially reduced soil aggregate size
(mean weight diameter decreased 35% within 60 d), and a reduction in the proportion of
intra-aggregate, physically protected light fraction organic matter. N20 ﬂuxes increased
6-fold in 2002, 2-fold in 2003 and 6-fold in 2004 and were associated with increased soil
NO3’ concentrations, which approached 15 u g N g]. Decreased plant N uptake
immediately after tillage, plus increased mineralization and nitriﬁcation rates (nitriﬁer
enzyme activity increased 5-fold), contributed to increased N03' concentrations. Our
results demonstrate that initial cultivation of a soil at pre-cultivation levels of native soil

C immediately destabilizes physical and microbial processes related to C and N retention

46

in soils and accelerates trace gas ﬂuxes. Policies designed to promote long-term C
sequestration may thus need to protect soils from even periodic cultivation in order to

preserve sequestered C.

INTRODUCTION

Carbon sequestration and greenhouse gas abatement in soils are two of a limited number
of rapidly-deployable, high impact CO2 stabilization options now available to policy
makers (Kauppi et al., 2001; Caldeira et al., 2004). Converting agricultural land to
perennial crops or to successional communities has the potential to sequester ca. 60 g soil
C m'2 y'1 (CAST, 2004) and conversion to no-till annual crops has the potential to
sequester C at about half this rate (West and Post, 2002; La], 2003). Ultimately,
however, net greenhouse gas mitigation by soil C storage depends on the persistence of
stored soil organic matter (SOM) and an important challenge to persistent C and N is the
potential reversibility of C sequestration (Paustian et al., 2000; La], 2004; Pacala and
Socolow, 2004). Some models predict that soils with sequestered C may rapidly lose C
and N following cultivation (Baisden and Amundson, 2003) but predictions remain
contentious because data are not currently available to generalize for temperate
ecosystems on a time scale of less than 10 years following cultivation (Davidson and
Ackerman, 1993; West and Post, 2002; Miller et al., 2004).

Models predict that N20 emissions in long-term no-till soils may be lower than
those in tilled soils (Six et al., 2004). Over time, increased aggregation, decreased bulk
density, and lower mineralization rates in no-till soils may limit nitriﬁer and denitriﬁer

N20 production. These changes, in addition to cessation of annual N fertilizer

47

amendments, reduce N20 emissions when agricultural soils are converted to unfertilized

successional communities (Robertson et al., 2000). Cultivation of no-till soils or other
disturbances that increase soil N availability may accelerate N20 emissions. Pinto et al.
(2004) found that cultivation of a l7-year old, unfertilized perennial pasture increased
N20 emissions from 2.7 to 4.6 mg N20-N In2 over 5 days of sampling. Keller et al.
(1993) found that converting forest to pasture increased N20 ﬂuxes in the ﬁrst ten years.
In both of these studies authors attributed increases to accelerated SOM turnover
producing excess inorganic N that was subsequently used by N20-producing nitriﬁers
and denitriﬁers.

In temperate regions, the processes controlling C and N cycling responses to
cultivation have been studied almost exclusively in soils with a recent history of
disturbance (Martens, 2001). Studies have demonstrated that long-term tillage aerates
soils, exposes C in protected microsites (Del Gado et al., 2003; DeGryze et al., 2004),
increases soil temperature, modiﬁes trace gas ﬂuxes (Smith et al., 2001; Smith and
Conen, 2004), and alters microbial community structure and function (Cavigelli and
Robertson, 2000; Buckley and Schmidt, 2001). Over time, these changes lead to 30-60%
declines in SOM (Davidson and Ackerman, 1993; Buyanovsky et al., 1997; West and
Post, 2002; Lal, 2003).

The quantity and turnover rates of soil aggregates may be particularly important
controls over SOM protection following changes in cultivation intensity (Paustian et al.,
2000). Evidence for protection of SOM by aggregates includes low interior-aggregate
oxygen concentrations that limit respiration rates, small intra-aggregate pore spaces that

limit SOM access to decomposers, and increases in soil respiration rates when SOM is

48

released from within aggregates (Sexstone et al., 1985; Mikha and Rice, 2004). Field

studies have demonstrated that C sequestration in afforested soils and restored grassland
ecosystems principally occurs within soil aggregates (J astrow et al., 1996; DeGryze et al.,
2004)

We hypothesize here that if immediate and persistent reductions in soil aggregation
occur after cultivation in conjunction with changes in soil temperature and other
environmental controls over decomposition, rapid changes in SOM turnover, microbial
activity and trace gas ﬂuxes will occur after a single plowing. Our ﬁrst objective is to
measure changes in CO2 and N20 emissions for three years following cultivation of a
previously uncultivated ﬁeld at pre-cultivation levels of native soil C to evaluate the
immediacy, magnitude, and persistence of change. Our second objective is to determine
the persistence of soil aggregation and changes in the distribution of physically protected
light-fraction SOM after initial cultivation. We additionally followed changes in soil
inorganic N, nitriﬁer enzyme activity and soil moisture to better understand factors

driving changes in gas ﬂuxes.

MATERIALS AND METHODS
Site Description and Experimental Design
Our experimental site was a previously never-tilled ﬁeld with little topography and
no known soil gradients located at the WK. Kellogg Biological Station (KBS) in
southwest Michigan, USA (42° 24' latitude, 85° 24' longitude). In 1956 the site was
cleared of trees and thereby converted from a northern hardwood forest to a

midsuccessional grassland community that has since been mowed each fall. Soils at the

49

site are Kalamazoo (ﬁne-loamy) and Oshtemo (coarse-loamy) mixed, mesic, Typic

Hapludalfs developed on glacial outwash (Crum and Collins, 1995). The two series co-
occur at KBS with variation within a series often as great as variation between series.
Prior to initial cultivation, surface soils (0-7 cm) contained 31.8 (i 1.4) g C kg", 2.59 (i
0.11) g N kg", and soil pH averaged 5.93 (:t 0.11); the same soil C levels had been
measured 3 years and 12 years prior to this study and are statistically indistinguishable
from C levels in nearby undisturbed forests on the same soil types (Robertson et al.,
2000)

Dominant plant species at the site include Bromus inermis,‘Rubus allegheniensis,
Monardaﬁstulosa, Juncus sp., Solidago canadensis, Arrhenatherum elatius, Poa
pratensis, and Elytrigia repens (Robertson et al., 2000). In 2002 we established in this
ﬁeld eight 3 x 6 m plots to which we assigned 4 replicates of two tillage treatments
(cultivated and uncultivated control) in a randomized complete block design. Plots were
laid out in an east — west direction and arranged in two rows running north — south with 4
plots and two blocks in each row. There were 5 m between plots in north — south and
east - west directions. Cultivated sites were mowed with biomass left in place prior to
cultivation to 19 cm. We used a moldboard plow and a disc for primary and secondary
cultivation, respectively, methods commonly used for preparing fallow land for
agricultural production. Cultivation occurred in the same plots on 25 June 2002 (DOY
176), 15 June 2003 (DOY 166), and 20 June 2004 (DOY 172). To study soil disturbance
effects separately from other factors associated with agricultural conversion (e. g. the use
Of annual plant monocultures and fertilizers), we left the site fallow following tillage to

allow a diversity of annual and perennial plants to recolonize.

50

Soil and Plant Sampling

Samples for aggregate and total soil C and N analysis were collected prior to initial
cultivation on 18 June 2002, and post-initial-cultivation on 24 August 2002, 27 May
2003, and 21 October 2003. An additional sample for total C analysis was taken 24
September 2004. Five 3.8 cm soil cores were taken from each plot to a depth of 20 cm,
placed in plastic bags, and refrigerated (< 7 (1) prior to sieving through an 8 mm sieve and
air-drying at 20 °C.

At each of two locations within a plot two 2.5 cm soil cores were collected to a
depth of 7 cm at each time of trace gas sampling. Cores were passed through an 8 mm
sieve, homogenized, and used for gravimetric soil moisture determinations.

Plant and litter C estimates prior to initial cultivation were estimated by drying and
analyzing litter and plant samples collected on 21 June 2002 (DOY 172) ﬁ'om two 625
cm'2 quadrats in each plot. Changes in the litter layer after cultivation were estimated on
15 September 2003 from two 625 cm'2 quadrats in each of four adjacent plots cultivated
for the ﬁrst time in 2003. Organic C and total N concentrations of plant and soil samples
were determined by dry combustion and gas chromatography in a CHN S analyzer

(Costech ECS 4010, Costech Analytical Technologies, Valencia CA.).

Trace Gas Fluxes
Gas ﬂuxes were determined using a single 25 cm diameter static PVC chamber
(5500 cm'3) located within each plot (Livingston and Hutchinson, 1995). Prior to

sampling, gas-tight lids with sampling ports were placed on chamber bases permanently

51

installed to a soil depth of 2.5 cm and accumulated headspace was then sampled four

times over 90 min by removing 20 mL of headspace gas to 12 mL vials (Labco
Unlimited, Buckinghamshire, UK). Gas sampling was generally performed between the
hours of 0900 and 1400 and all plots were sampled on each sampling day. Within 48 h of
collection CO2 was analyzed using an infrared gas absorption analyzer (IRGA) and N20
analyzed using a gas chromatograph outﬁtted with a 63Ni electron capture detector (350
°C). Flux for each chamber was calculated as the linear portion of the gas accumulation
curve for that chamber. Trace gas measurements were made 13 times in 2002 between
24 June and 22 August, 10 times in 2003 between 29 May and 13 October, and 19 times
in 2004 between 15 April and 29 October. We interpolated daily gas ﬂuxes from our
periodic measurements to estimate total CO2 and N20 ﬂuxes over the measurement

period.

Particulate SOM Density Distribution

Cultivation effects on the distribution of particulate SOM in four density fractions
was determined using a sequential fractionation technique in sodium polytungstate
(NAPT). Whole soil samples (15 g) were dispersed in 0.5% sodium hexametaphosphate
by shaking for 48 h on a rotary shaker set at 190 rpm. Dispersed samples were poured
through a 53 um sieve and rinsed thoroughly with deionized (DI) water. Sand and
particulate organic matter (POM) remaining on the sieve were transferred to ﬁlter paper
and then backwashed into a 100 mL beaker with 40mL NAPT (density = 1.9 g cm‘3).
Aﬁer equilibrating overnight we aspirated the ﬂoating material off the surface. POM

remaining in the beaker was classiﬁed as having a density > 1.9 g cm’3. POM with a

52

density < 1.9 g cm'3 was then sequentially suspended in NAPT with densities of 1.6 and
1.3 g cm'3, resulting in POM in four density fractions (> 1.9, 1.6 — 1.9, 1.3 - 1.6, < 1.3 g

cm'3). Clay plus silt associated SOM was determined by difference.

Aggregate and Light Fraction Separation

Aggregate distribution was determined by hand on triplicate 35 g air-dried soil
samples (0-20 cm soil depth) by wet-sieving in water through a series of 2000 pm, 250
um, and 53 um sieves. Soil was submerged for 5 min on the surface of the 2000 um
sieve which was then moved up and down for two minutes with a stroke length of 3 cm
for 50 strokes. Sieving was repeated on the 250 um (50 strokes) and 53 um (30 strokes)
sieves using the soil plus water that passed through the next larger sieve. Aggregates
remaining on each sieve were dried at 60 °C. Sand content was determined on an
aggregate subsarnple after dispersing soil in sodium hexametaphosphate (0.5%) for 48 h
on a rotary shaker at 190 rpm.

Mean weight diameter (MWD) of sand-free aggregates was determined by
calculating the sum of the products of the mean diameter of each size fraction and the
proportion of the total sample weight in that fraction (Kemper and Rosenau, 1986).

The method we used to separate inter- and intra-aggregate light fraction (LF) is
based on previously published protocols (Six et al., 1998; Gale et al., 2000). Aggregate
subsamples were pre-wetted prior to LP analysis to minimize aggregate slaking during
LF separation. An 8 g subsample of aggregates was divided in half and placed on two
membrane ﬁlters (47 mm diameter; Pall Supor-450) overlaying two paper ﬁlters (70 mm

diameter; Whatman 42) in a 10 cm petri dish. The paper ﬁlters conducted water to the

53

membrane ﬁlters, which, in turn, facilitated the smooth transfer of soil into beakers.

Four mL of DI water were trickled onto the paper ﬁlters in order to slowly wet all of the
aggregates by capillarity. Aggregates were transferred from the membrane ﬁlters to 100
mL beakers after 16 h with 5 mL aliquots of NAPT at a density of 1.62 g cm'3. A total of
55 mL NAPT was used for each sample. A preliminary test showed that the ﬁnal density
of the sodium polytungstate was 1.60 g cm"3 following equilibration with the water
contained in aggregates.

After 24 h, LF was aspirated from the surface of the sodium polytungstate and
then rinsed on a hardened, ashless ﬁlter paper with at least 600 mL DI H2O. We refer to
this pool as inter-aggregate LF. Aﬁer removal of this pool, we aspirated the remaining
sodium polytungstate. Aggregates were then dispersed to release the intra-aggregate LF
using sodium hexametaphosphate as described previously and resuspended in NAPT (d =

1.62 g cm'3). The intra-aggregate LF was collected form the surface.

Soil Temperature and Inorganic Nitrogen Dynamics

Soil temperature was determined at the time of each trace gas sampling to a depth
of 7 cm at a distance of 25-35 cm from the sampling chamber. NH4+ and NOg' were
extracted with l M KCl from duplicate ﬁeld-moist 10 g soil samples using a 1:5
soil/extractant ratio. Soil extracts were ﬁltered with a syringe ﬁlter using a type A/E
glass ﬁber ﬁlter (Pall Corporation, East Hills, NY). Filtrates were stored in 7 mL
scintillation vials and ﬁozen until analysis for NH4+ and NO3'. Both analyses were
performed on an Alpkem 3550 Flow Injector Analyzer (01 Analytical, College Station,

TX). Inorganic N was determined on samples collected to a depth of 7 cm in 2002 on

54

DOY 175, 178, 190, 200, 207, 228, and 234; in 2003 on DOY 149, 170, 178, 189, 216,

226, 245, 266, and 286; and in 2004 on DOY 131, 156, 180, 194, 208, 230, and 268.
Additionally, inorganic N was also determined to a depth of 7-20 cm on three dates in
2003 and all dates in 2004. Trends at this depth were similar to those at 0-7 cm but

generally smaller in magnitude and are not presented here.

We used a shaken slurry method (Hart et al., 1994) to test for changes in nitriﬁer
enzyme activity to 0-7 and 7-20 cm soil depths following cultivation on 13 October 2003
(DOY 286) and 17 August 2004 (DOY 230). Brieﬂy, 30 g ﬁeld-moist soil was combined
with 100 mL solution containing non-limiting quantities of NIHr in a 160 mL jar capped
with polyﬁlm. The polyﬁlm was perforated with a needle to allow rapid gas exchange
while minimizing water loss. Jars were placed on a rotary shaker and rotated at 200 rpm.
Each ﬂask was sampled 4 times during a 28 h incubation period. Soil extracts were

centrifuged, ﬁltered, and frozen until analysis for NO3‘.

Statistical Analysis

Soil aggregate size distributions were corrected for sand content of the same size as
the aggregates since this is usually not part of the aggregate structure. Carbon content of
soil organic matter fractions was calculated on an area basis by correcting for bulk
density and soil sampling depth. Tillage effects on trace gas ﬂuxes, soil moisture,
temperature and inorganic N were analyzed by Proc Mixed (Version 8.2, SAS Institute,
1999) using a randomized complete block design analysis of variance (AN OVA) with
repeated measures (SAS Version 8.2, SAS Institute, 1999). Treatment and sampling date

were considered ﬁxed effects and block a random effect. Where there were signiﬁcant

55

day of year by treatment interactions, differences between treatments on separate days

were determined using the slicing command in SAS. Cultivation effects on aggregation
and SOM pools were similarly analyzed but without repeated measures. Analysis of
covariance (Goldberg and Scheiner, 2001) was used to examine the relationship between

soil temperature, moisture and trace gas ﬂuxes in 2002, 2003, and 2004.

RESULTS
Trace Gas Fluxes

Within 2 days of initial cultivation in 2002 tillage had signiﬁcantly (P < 0.001; n =
4) increased co2 ﬂuxes > 100%, from 91 to 196 mg CO2-C m'2 hr" (Figure 3.1). We
estimate that within the ﬁrst 33 d following cultivation 85 g C In2 was lost due to tillage.
During the next 30 d, tillage effects on CO2 ﬂux were somewhat lower, resulting in a

daily average loss of 1.4 g C m'2 (1'1 over the 60 (1 measurement period.

Prior to the second cultivation in 2003, C02 ﬂuxes were again higher in the
cultivated plots (p<0.001). In 2003 the tillage-induced CO2 response was measurable for
80-120 days and average daily CO2-C loss due to cultivation was 1.9 g In2 (1’1 for the 138

d sampling period.

Prior to cultivation in 2004 there were no signiﬁcant differences (p<0.05) between
treatments. Following cultivation (DOY 172), cultivated plots had signiﬁcantly greater
CO2 ﬂuxes on DOY 203 (104%), 204 (73.4%), and 211 (152%). Additional daily CO2-C

losses due to cultivation were 1.00 g C m'2 d'I over the 198 sampling period in 2004.

56

Analysis of covariance showed that soil moisture and temperature (Fig. 1) were
related to CO2 ﬂux differently in different years. In 2002 there was no temperature

effect but a signiﬁcant soil moisture by treatment interaction (P < 0.05), indicating that
the relationship between CO2 ﬂux and moisture differed between treatments [control:
logic C02 ﬂux = 0.28 + 0.77 (logto moisture), r2 = 0.74; cultivated: logio CO2 ﬂux = 0.98
+ 0.33 (logm moisture), r2 = 0.21; units in Figure 3.1]. Moisture effects were not
signiﬁcant in 2003 or 2004. In 2003 there was a signiﬁcant (p<0.05) soil temperature
effect [log1o CO2 ﬂux = 061 + 1.40 (logm temperature), 1'2 = 0.42]. In 2004 there was
again a signiﬁcant (p<0.05) soil temperature effect, although a smaller proportion of the

ﬂux was explained [loglo CO2 ﬂux = 0.05 + 0.89 (logm temperature), 1'2 = 0.16].

N20 ﬂuxes in 2002 increased in response to cultivation but not until 30 d after
cultivation when ﬂuxes in cultivated plots were as high as 87 ug N20-N rn'2 h"l (Figure
3.1). In 2003, mean ﬂuxes were higher in cultivated plots on all days but signiﬁcantly
different only on DOY 170 and 202. Overall, N20 ﬂuxes were considerably higher in
2004 than in 2002 or 2003. In 2002, average N20 ﬂuxes were 2.87 in control plots and
22.17 1.1g N20-N m'2 h'1 in cultivated plots; in 2003 the average N20 ﬂuxes were 3.72 in
control plots and 11.54 11g N20-N m'2 h'1 in cultivated plots; in 2004 control plots emitted
11.37 11g N20-N m‘2 h’] compared to 76.14 11g N20-N rn‘2 h'1 in cultivated sites. In 2004,
there were extremely high ﬂuxes due to cultivation on DOY 163 (217.4 :1: 153.3 pg N20-

N m'2 h") and DOY 189 (1292 a 573.2 ug N20-N m'2 h").

Soil Aggregation

57

The mean weight diameter (MWD) of aggregates measured 63 d after tillage was
35% lower in cultivated than in control plots (P < 0.01 , n = 4; Figure 3.2). These
differences were still evident prior to tillage in 2003 (P<0.01, n = 4). In October 2003
(DOY 294) the reduction in MWD associated with tillage was 37% (P < 0.05; Figure
3.2), similar to the difference found in 2002. The size class of 2000 — 8000 um
aggregates declined to 19% from 34% in control plots by DOY 236, 2002 (Figure 3.3).
In 2003, 2000 — 8000 um aggregates remained lower in cultivated plots while aggregates

in the 53 — 250 and <53 um size classes increased.

Soil Organic Matter Distribution

There were no differences in total soil C to 20 cm depth between control and
cultivated plots following cultivation in 2002 (4.36 :1: 0.05 vs. 4.70 a: 0.15 kg m'z), 2003
(4.38 a 0.28 vs. 4.81 a 0.26 kg m'Z), or 2004 (4.32 a 0.09 vs. 4.74 a: 0.32 kg m'z).
Cultivation reduced litter C on the soil surface from 195 to 18 g C m'2 (p<0.01) after a
single cultivation. In 2002 and 2003 POM with a density > 1.9 g C cm'3 was similar in
control and cultivated plots (Table 3.1). In 2002, there was an increase of POM in the
lower density fractions (< 1.3, 1.3 — 1.6, and 1.6 — 1.9). Cultivation did not signiﬁcantly
change POM distribution in 2003 (p<0.05), although there was a trend (p < 0.1) toward

more POM with a density of 1.3 — 1.6 g C m"2 in cultivated plots (Table 3.1).

The proportion of intra-aggregate to total LP in 2000 — 8000 11m aggregates
declined from 28 to 16% within 60 d of the ﬁrst cultivation in 2002 (p<0.05; n = 4;
Figure 3.4). In 2003 there was a trend towards a reduced proportion of intra-aggregate

LF associated with 2000 — 8000 um aggregates in cultivated (21%) compared to control

58

(27%) plots (p<0.l; n = 4) . The intra-aggregate LP in the 250 - 2000 pm size class
increased 73% (P<0.05; n = 4) and there was a trend towards an increase in inter-
aggregate LF from 56.7 to 118.0 g LF C m'2 (p<0.1). In the 53 — 250 pm size class there

was an increase in the inter- and intra-aggregate LF and total LF.

Soil Inorganic Nitrogen

Cultivation in 2002 increased extractable soil NO3’ concentrations to 3.18 pg N03'-
N g‘l (compared to 0.41 pg NO3'-N g'1 in control plots) after 24 (1 (Figure 3.5). NO;'
concentrations in cultivated plots remained higher on DOY 207, 228 and 234 and peaked
on DOY 228 (14.6 vs. 0.44 pg NO3'-N g"). NHa+ concentrations were also increased by

cultivation on DOY 190 (87%), 200 (312%), 207 (176%), 228 (193%), and 234 (82%).

In 2003 cultivation signiﬁcantly increased NO3’ concentrations on DOY 170 and
216 but there were no differences on the other sampling dates (Figure 3.5). NHa+
concentrations were signiﬁcantly greater in cultivated plots on DOY 170 and 216 but
lower in these plots relative to control plots on the following three sampling dates. In
2004 N03' concentrations on DOY 180 were 11.8 pg NOj-N g'1 in cultivated plots and
0.55 pg NO3'-N g'1 in control plots (Figure 3.5). NO;' concentrations remained higher in
cultivated plots on DOY 194 and 208. NH4+ concentrations were reduced by cultivation

on four of seven sampling dates in 2004.

Nitriﬁer enzyme activity (Figure 3.6) in 2003 was 5-fold higher in cultivated than
uncultivated treatments (25.4 vs. 127 pg N kg'l soil h") to 7 cm soil depth and 3-fold

higher at 7 — 20 cm (14.9 vs. 45.2 pg N kg" soil h"). Similar trends were observed in

59

2004 when control plots had lower nitriﬁer enzyme activity at 0 - 7 cm (35.5 vs. 109 pg

N kg" soil h") and 7 - 20 cm (10.5 vs. 45.7 pg N kg" soil h“).

DISCUSSION

We documented substantial, immediate losses of CO2-C following cultivation. On
average, cultivated plots lost an additional 1.4, 1.9 and 1.0 g C m'2 d'1 in 2002, 2003 and
2004, respectively, equivalent to 84 g C m‘2 over the 60 (1 following cultivation in 2002,
260 g C over the 138 d sampling period in 2003, and 198 g C over 2004’s 198 d sampling
period. N20 emissions also markedly increased during our measurement period
following cultivation: 672% in 2002, 210% in 2003, and 570% in 2004. Based on IPCC
calculations using a 100-y time horizon for N20 these emissiondifferences are equivalent
to 22.5, 9.1, and 75.4 g CO2-C equivalents In2 for 2002, 2003, and 2004, respectively.
The 3-year average of 35.7 g CO2-C m'2 is similar to the average annual C gain (ca. 30 g
C m'2 y“) under no-till systems in the US. Midwest (Robertson et al., 2000;
Franzluebbers and Steiner, 2002). Below we discuss how changes in the soil
environment and biological processes likely contributed to these ﬂuxes and, more

generally, to destabilizing soil C and N.

Soil Structure and Organic Matter

Many studies have demonstrated that long-term repeated cultivation reduces soil
structural stability and changes the distribution of SOM (e. g. Six et al., 1998; Grandy et
al., 2002; DeGryze et al., 2004). Our results indicate that a signiﬁcant amount of the

structural degradation and change in C distribution, particularly in large soil fractions,

60

occurs after plowing only once. There was no measurable additional decline in soil

structure following cultivation in 2003, further demonstrating that cultivation effects
occur immediately and are persistent. Mean weight diameter (MWD) differences
measured 63 d after tillage were largely attributable to declines in 2000 - 8000 pm
aggregates (Figure 3.4). In cultivated sites, 19% of the soil was in the 2000 - 8000 pm
size class, compared to 34% in the control plots, identical to adjacent agricultural ﬁelds
on the same soil type that have been tilled for > 50 y (data not shown). The persistence
of these effects was evident prior to tillage in 2003 when aggregation in cultivated sites
remained substantially lower, despite the potential for freeze-thaw and wetting-drying

cycles to have affected aggregation during winter months.

Soil CO2 emission responses to cultivation in previously cultivated soils may occur
for only hours or days, suggesting that physical phenomenon such as diffusion rates are
driving increases (Kessavalou et al., 1998; Calderon et al., 2001; Jackson et al., 2003). In
our previously uncultivated soils, sustained CO2 ﬂuxes suggest that microbial respiration
increased following cultivation due to increased substrate availability. These substrates
included LF released from large aggregates in addition to aboveground C that entered
low-density, unprotected, inter-aggregate POM pools (Table 3.1; Figure 3.4) that are
rapidly oxidized following disturbance (Arrouays and Pelissier, 1994; Six et al., 1999;

DeGryze et al., 2004).

In 2002 this aboveground C consisted of litter (142 i: 30 g C m’z) and plant
biomass (228 :1: 11 g C m'z). In agricultural systems, Lupwayi et al. (2004) found that
incorporation of wheat litter with conventional tillage increased its decomposition rate by

48%; Burgess et al. (2002) found that litter decomposition rates at 20 cm were 52 - 105%

61

greater than those at the soil surface. Light fraction pools have been shown to be
correlated with soil surface respiration rates (J anzen et al., 1992; Alvarez and Alvarez,
2000) and its depletion represents a major portion of C loss in cultivated soils

(Cambardella and Elliott, 1992, 1994).

Slow recovery of the plant community following cultivation suggests that soil C
turnover rather than increased root respiration accounts for the additional CO2-C
emissions. It generally took about four weeks for signiﬁcant plant recovery to occur.
During this time, heterotrophic respiration will have accounted for most of the CO2 ﬂux
in the tilled plots while in the control plots autotrophic respiration may have produced
50% or more of the measured CO2 (Hanson et al., 2000). Our inability to detect changes
in total soil C is a common ﬁnding in short-term C loss studies due to the need to detect
relatively small changes in soil C against large and spatially heterogeneous background

pools (Sollins et al., 1999; Brye et al., 2002).

Soil Moisture and Temperature

Analysis of covariance showed that soil moisture and temperature were related to
C02 ﬂux differently in different years. In 2002, the signiﬁcant soil moisture by treatment
interaction resulted primarily from differential responses to low and high gravimetric soil
water contents; speciﬁcally, at soil moisture contents less than 20 %, CO2-C emissions
were an average of 84% higher in cultivated sites (2.64 vs. 4.88 g C m'2 d") whereas at
higher soil moisture contents the CO2-ﬂux difference was only 21% (5.84 vs. 7.08 g C m'
2 d'l). These differences may be due to a greater inﬂuence of soil moisture on C

availability in uncultivated sites. Soil moisture can stimulate C and N mineralization by

62

 

enhancing aggregate turnover (Denef et al., 2001), by accelerating diffusion of active C

compounds (Borken et al., 2003), and by increasing lysis of microbial cells and the
release of intra-cellular solutes (F ierer et al., 2002). These processes may have been
important sources of C following wetting of undisturbed soils; in cultivated sites,
however, increased C availability associated with aggregate destruction and changing
SOM pool sizes following tillage may have elevated emissions in drier soils. Kessavalou
et al. (1998) similarly found that wetting effects varied with disturbance intensity. They
experimentally increased soil moisture content with 5.1 cm of water in a wheat-fallow
cropping system and found that between 24 and 72 h after wetting increases in CO2

emissions averaged 109% for subtill, 82% for no-till, and 24% for plowing treatments.

A signiﬁcant temperature effect on CO2 emissions suggests that measured increases
in the average temperature of cultivated plots in 2003 (16.9 vs. 19.7 °C) and 2004 (16.2
vs. 18.4) contributed to increased CO2 ﬂuxes. Other studies have also shown that
decomposition and CO2 ﬂux are related to soil temperature (e. g. Wagai et al., 1998; Inoue
et al., 2004; Knorr et al., 2005). Davidson et al. (2000) argue that changes in soil
temperature should not be viewed in isolation but rather in conjunction with changes in
other controls over SOM. In our experiment, it is likely that SOM pools became more
susceptible to the effects of warming after cultivation because of aggregate destruction
and decreases in the physical protection of labile C pools. Although some studies
addressing the effects of soil warming on CO2 emissions indicate that soil respiration
responses to increased temperature may decrease over time (Kirschbaum , 2000; Luo et
al., 2001) or that only certain pools of C are susceptible to mineralization after soil

warming (Davidson et al., 2000), the long-term effects of soil warming on resistant C

63

pools that represent the majority of SOM are difﬁcult to infer from ﬁeld experiments

lasting only a few years (Knorr et al., 2005; Powlson, 2005).

N20 ﬂuxes and N Availability

N20 ﬂuxes did not increase in 2002 until 34 d after cultivation, likely due to the
lack of adequate NO3' and moisture: not until then did cultivated soils exhibit both a high
soil moisture content (>20%) and substantially elevated soil N03' concentrations (Figures
3.1, 3.5). In 2003 and 2004 N20 ﬂuxes also were highest on sampling days with high
soil NO3' concentrations. Pinto et al. (2004) similarly found that N20 ﬂuxes following
cultivation of a perennial pasture increased synchronously with increases in soil N03'

following an initial lag period.

In these unfertilized soils, changes in nitrogen availability may be a particularly
important control over N20 emissions via denitriﬁcation. Low NO3' concentrations in
our control soils suggest a tight coupling of plant N uptake and the microbial processes of
N-mineralization and nitriﬁcation. High NO3‘ concentrations in the cultivated sites
indicate that soil disturbance disrupted the synchrony between inorganic N production
and consumption. Accelerated SOM mineralization and increased nitriﬁer enzyme
activity likely enhanced NO3' production and in all three years soil NO3' concentrations
and N20 ﬂuxes declined as vegetation re-established, likely due to plant N uptake.
Additional changes in the synchrony between nitrogen availability and plant demand,
including the use of supplemental N fertilizer or conversion to annual crops, will likely

lead to additional N20 emissions.

64

The effects of tillage on soil surface N20 ﬂuxes have been primarily studied

following conversion of long-term, conventionally tilled cropping systems to no-till
(Grant et al., 2004; Six et al., 2004; Mackenzie et al., 1997). These studies demonstrate
the potential for changes in soil water content, pore space structure, nitrogen cycling, and
plant productivity following adoption of no-till to modify denitriﬁcation rates and N20
emissions (MacKenzie et al., 1997, 1998; Baggs et al., 2003). The results we present
here demonstrate the potential for sizeable N20-N losses following cultivation of no-till
ecosystems and corroborate measurements by Pinto et al. (2004) showing increases in
N20 emissions over 5 d of between 1.8 and 23 mg N20-N rn'2 following the plowing of a

perennial pasture.

Low N20 emissions in the spring and fall of 2003 and 2004 suggest that we
captured those seasonal periods with the highest ﬂux, however, some studies have
reported that winter and early spring N20 emissions can be important components of the
total annual budget and that spring thaw emissions, in particular, may be among the
highest of the year (Flessa et al., 1995; Kammann et al., 1998). Previous winter and early
spring N20 sampling campaigns near our study site on the same soil type, however, have
found low or undetectable emissions (Robertson, unpublished data). Laboratory
experiments have demonstrated that unfrozen, super-cooled water ﬁlms around clay
particles can support denitriﬁcation at temperatures as low as -2 to -4°C (Dorland and
Beauchamp, 1991; Kopenen et al., 2004). In our soils, with clay contents generally <
20% (Crum and Collins, 1995), the availability of unfrozen water to support

denitriﬁcation in frozen soils may limit winter denitriﬁcation.

65

CONCLUSIONS

Overall our results illustrate the rapid and destabilizing effect of cultivation on C and
N cycling in a soil at pre-cultivation levels of native C. Following a single tillage event,
19% of the soil was present as aggregates in the 2000-8000 pm size class, identical to
adjacent agricultural ﬁelds on the same soil type tilled for > 50 y, compared to 34% in
control plots. Aggregate destruction directly released light-fraction organic matter from
within intra-aggregate microsites and limited the incorporation of particulate organic
matter originating from aboveground C pools into aggregates. These processes increased
substrate availability that, along with changes in temperature, contributed to higher
mineralization rates. As a result, cultivated plots lost an additional 1.4, 1.9 and 1.0 g C
m’2 (1" in 2002, 2003 and 2004, respectively, equivalent to 84 g C rn'2 over the 60 (1
following cultivation in 2002, 260 g C m'2 over the 138 d sampling in 2003, and 198 g C
In2 over 2004’s 198 d sampling period. Increased mineralization rates also increased
nitriﬁer enzyme activity and soil N03'-N, which approached 15 pg g’I in cultivated plots.
Cultivation increased N20 emissions by 672% in 2002, 210% in 2003, and 570% in
2004. Based on IPCC calculations using a 100-y time horizon for N20, these emission
differences are equivalent to a 3-year average of 35.7 g CO2-C m’z, which is similar to the
average annual C gain under no-till systems in the US. Midwest. Our results
demonstrate that successional communities at pre-cultivation levels of native C
experience rapid increases in ecosystem C and N cycling immediately following
cultivation. This acceleration translates into substantive destabilization of soil C and N
stocks and dramatic increases in trace gas ﬂuxes, suggesting that policies designed to

promote soil C sequestration need to protect soils from even periodic plowing.

66

 

be. we. 93 a a: .9 86 a 33 38 5 as... we. 8.0 u :3. .9 ”we a ”we
moom Ade wx m _ .o a one .m> 36 a 00$ 88 5 90—0 08933 new .0580 50.502 Eu cm 9 O :8 :38 E moocopogn 0: 203 22:. ..
data—E 20m 98 So» 0 5.23 80.0 08?an 0:0 65:00 0003000 $055k? ESEEB 83065 _..

See 32

 

 

A .3: SE 838 £03 8930 $8 6. _ a 8mm 333:5
62: 2: 0.3: 32 83: 32 A _ .26 32 Same ”an .8250 88
Ga. _ c we: 5 .o: aim 2.4.8 4.8. 833 33W 833 e _ mm 822:5
.330 N03 .23: 2: save :2 :33 $3 82% of m .8200 88
“.8 0 w
0 00508.8
2 v e._ -2 3 -3 S A -3052

 

Arlene 3 .3800 .552: 0:890 22:20.23

+

2805530 5 Echo 08053 :23 C. u 5

88:. as 82e> .5 cm 9 0% >09 88 use 68 >08 88 s asses seems :8 .6 assesses 2: 8 asses 8:325 .2 2.3

67

Treatment

. ‘ OControl (solid line)
400'“ f 1. - Cultivated (dashed line)

2.
“
~

   

 

 

40- j
30—
20—

10"

 

Soil Moisture g g'1

 

40—

30"

 

20"

10“

 

Temperature °C

 

20007 i
1504
90%‘
60-

30‘

 

pg N20-N m'2 h'1

 

171 183 195 207 . 219 - 231 . 243
Day of Year 2002

Figure 3.1. Changes in soil CO2 ﬂux, moisture, temperature, and N20 ﬂux following
cultivation of a previously never tilled soil in 2002. Measurements were made in 2002 (a;
this page), 2003 (b; following page), and 2004 (c; two pages ahead). Arrows indicate
cultivation dates. Note different x-axis ranges in each year, reﬂecting different durations of
sampling. Soil temperature and moisture samples were determined for the 0-7 cm depth.
Treatment means are shown 3: standard error (n = 4); * indicates statistically signiﬁcant (P <
0.05) differences between control and cultivated treatments within a single day of year
(DOY) where there was a signiﬁcant treatment by DOY interaction.

68

500-1 i Treatment

0 Control (solid line)
' Cultivated (dashed line)

 

 

 

407 l
30-

20‘

 

10"

 

Soil Moisture g g'1

 

40? i

30—J

20—

 

10—

Temperature °C

 

 

 

 

 

 

 

o .
125 155 185 215 245 275 305
Day of Year 2003

Figure 3.1. Cnt’d

69

i Treatment

0 Control (solid line)
° Cultivated (dashed line)

or
o
o

g]

4:.

o

O
I

0.)
o o
o o
l J

100—

 

mg COz-C m2 h'1
N

 

 

O
.1

Soil Moisture g g'1

 

 

 

 

 

 

 

 

9
9
3
E
(D
Q.
E
G.)
'— o

1290
‘_ 2000 217i .
'5 l ' 1
N 150— i I\
E 90% ’/i I" i

\
Q“ 60— I, “ I} y’
z _ 7
c, 30 I L__it \ —’-’- ““
:- O_W ' r l I
100 135 170 205 240 275 310
Day of Year 2004

Figure 3.]. Cnt’d

7O

.>OD 233 a 5:23 35.58: 33323 Ea .888 50252.
moocoeotﬁ God v a: EmoEcwﬁ 335:“; 3:36.: .1 xv n 5 coho weaves: a :32? 2a £82: 2253; .ﬁaov

:5 o2 2: .3 85.52% a; Seamed? .88 5 of >8 Ba 88 s a: :68 3% do be 8 eoesooo $2.:
.28 3:32:05. 3325a a mo gags—so wEBozom moumwohwwm :8 mo AQEBV 33:86 £303 :32 .N.m 23w:—

amm> 06 >8

  
      

       

 

 

 

 

 

 

 

V
vmm 21 0mm %

. .. ., 1 l 0.0
1.... 1 . . nJu
1 . M .. m o e
W w. M ._ W1 9

M... a 1.. .2. . 1..“ l.

1 . 1 ~. m o P w u
l m P 5
862:5 I N
9E0 . Mu.
_ O I a. a. l. O N e
#5559... a w
1-
moom l m N m

71

Aggregation
%

Aggregation
°/o

Aggregation
%

Aggregation
%

40‘
30“
20‘

 

 

 

 

   
 
 
 
   

 

2000— 250-

8000 2000

2003: DOY 294

Treatment

I Control
I Cultivated

2003: DOY 147

2002: DOY 236

 

2002: DOY 169 (pre-till)

   

53- < 53
250

Aggregate Size Class (um)

Figure 3.3. Soil aggregate distribution in four size classes. Black bars represent control plots;
patterned bars, cultivated plots. Tillage occurred on day of year (DOY) 176 in 2002 and DOY
166 in 2003. Aggregation was determined for the 0-20 cm depth. Treatment means are
shown :1: standard error (11 = 4); * indicates statistically signiﬁcant (P < 0.05) differences
between control and cultivated treatments within a size class and DOY

72

2003
250 a 2000-8000 um 250-2000 pm 53250 um
7 § T. I . §

   
 
  
 

 

0
g 200-
? “(E 150- Soil C Pools
E a, 100 - ' Inter-aggregate LF
:3 ‘ U lntra-aggregate LF
:1 50 -
o _
2002
b
250—
200- ‘Il T. § T. §

Light Fraction C
9 m“2

 

 

Cont. Cult. Cont. Cult. Cont. Cult.

Cultivation Treatment

Figure 3.4. Distribution of inter— and intra-aggregate light fraction organic matter (LF) in
control and cultivated plots in 2002 and 2003. Samples were collected to a depth of 20
cm on 24 August 2002, day of year (DOY) 236 and 15 June 2003 (DOY 166) and wet-
sieved into aggregate size classes. 1', I, § and 1] indicate signiﬁcant differences (P<0.05)
within an aggregate size class between cultivation treatments for inter-aggregate LF, intra-
aggregate LF, total LF and the proportion of total LF within aggregates, respectively. In
2003 there was trend towards a decreased proportion of intra-aggregate LF (p<0.054) in
2000 - 8000 um aggregates.

73

628885 >OQ .3 Ems—“mob “:85ch a $3 205 Bop—3 909v

So» mo than a 55; Savannah cow—«>220 28 35:8 5953 moococobﬁ God v A: Emoﬁcwi £85355 83305 ... Xv u 5 hobo
$8288 a :32? 0.8 888 E0832 1r Snow 80 To 2: 8» 355.2% 33 Z 25w85 .2058 Ewtv voom can .283 23:5
moon 50:8 to: moon .8 mouse gags—so 2865 m30t< docm>£=o REE wage—.8 Z oEmwBE E mowcmzu .m.m Ezwi

 

 

 

 

 

 

38 55> 6 .60 88 55> a >8 88 55> 5 .60
En EN oew wow 02 m2 2: mom mew mew m5 m9 mmr mm? mmm 5m 9a SN m2 2: So
_ . _ i . _ r _ . . . _ . _ p b . .L
111111 ”in c 1717::

1 .. 1 18 m.
1 $ ....... * 1 18 m
. v i ’* 6
.228 1 :2 5 . e
1 ; 1 .. 18 m.
.8800 1 :2 o 1 .. O
1 umumzzso 4.02 4 1 ,,+. .. rev .W
a _§coo.-1oz 4 a 1 .. M
1 1 H 18 v

 

74

$85885 >OQ 3 E0830:

“53$:me a was» 0.85 80:3 909 So» we then m GEES 35888“ 32323 new 35:8 5253 mooceotmv

God v a Became racemes; asses ... Re 1 a 5.5 useasm « 53% ea ages access: .82. >09 38
unmade S 98 Sam >09 moom $9900 2 co gags—so wc§o=8 53:8 2558 Samba E mowemnu 6M oSwE

meow meow
Eoomue 53.10 Eooml eoeio

l
o
In

 
    

852:5 l
.9200 I
2958;

 

low—1

l
O
52
1M nos 6» N--€ON 5“
Kimnov ew/izua JeuunN

 

l CON

75

REFERENCES

Alvarez, R., and C. R. Alvarez. 2000. Soil organic matter pools and their associations
with carbon mineralization kinetics. Soil Science Society of America Journal 64:184-
189.

Arrouays, D., and P. Pelissier. 1994. Changes in carbon storage in temperate humic
loamy soils after forest clearing and continuous corn cropping in France. Plant
and Soil 160:215-223.

Baggs, E. M., M. Stevenson, M. Pihlatie, A. Regar, H. Cook, and G. Cadisch. 2003.
Nitrous oxide emissions following application of residues and fertiliser under zero
and conventional tillage. Plant and Soil 254:361-370.

Baisden, W. T., and R. Amundson. 2003. An analytical approach to ecosystem
biogeochemistry modeling. Ecological Applications 13:649-663.

Borken, W., E. A. Davidson, K. Savage, J. Gaudinski, and S. E. Trumbore. 2003. Drying
and wetting effects on carbon dioxide release from organic horizons. Soil Science
Society of America Journal 67: 1 888-1 896.

Brye, K. R., S. T. Gower, J. M. Norman, and L. G. Bundy. 2002. Carbon budgets for a
prairie and agroecosystems: Effects of land use and interannual variability.
Ecological Applications 12:962-979.

Buckley, D. H., and T. M. Schmidt. 2001. The structure of microbial communities in soil
and the lasting impact of cultivation. Microbial Ecology 42:11-21.

Burgess, M. S., G. R. Mehuys, and C. A. Madramootoo. 2002. Decomposition of grain-
com residues (Zea mays L.): A litterbag study under three tillage systems.
Canadian Journal of Soil Science 82:127-138.

Buyanovsky, G. A., J. R. Brown, and G. H. Wagner. 1997. Sanbom ﬁeld: effect of 100
years of cropping on soil parameters. Pages 205-226 in E. A. Paul, K. Paustian, E.
T. Elliott, and C. V. Cole, editors. Soil organic matter in temperate
agroecosystems. CRC Press, New York.

Caldeira, K., M. G. Morgan, D. Baldocchi, P. G. Brewer, C. T. A. Chen, G.-J. Nabuurs,
N. Nakicenovic, and G. P. Robertson. 2004. A portfolio of carbon management
options. Pages 103-130 in C. Field and M. Raupach, editors. The Global Carbon
Cycle. Island Press, Washington, DC, USA.

Calderon, F., L. Jackson, K. Scow, and D. Rolston. 2001. Short-term dynamics of
nitrogen, microbial activity, and phospholipid fatty acids after tillage. Soil
Science Society of America Journal 65:118-126.

76

Cambardella, C. A., and E. T. Elliott. 1992. Particulate soil organic-matter changes across
a grassland cultivation sequence. Soil Science Society of America Joumalz777-783.

Cambardella, C. A., and E. T. Elliott. 1994. Carbon and nitrogen dynamics of soil

organic matter fractions from cultivated grassland soils. Soil Science Society of
America Journal 58:123-130.

CAST. 2004. Emissions and mitigation of agricultural greenhouse gases. in Climate
Change and Greenhouse Gas Mitigation: Challenges and Opportunities for
Agriculture. Council for Agricultural Science and Technology (CAST), Ames,
Iowa, USA.

Cavigelli, M., and G. Robertson. 2000. The functional signiﬁcance of denitriﬁer
community composition in a terrestrial ecosystem. Ecology 81 : 1402-1414.

Crum, J .R., and HP. Collins. 1995. KBS Soils [Online]. Available at
www.lter.kbs.msu.edu/soil/characterization. W. K. Kellogg Biological Station

Long-Term Ecological Research Project, Michigan State University, Hickory
Corners, MI.

Davidson, E. A., and I. L. Ackerman. 1993. Changes in soil carbon inventories following
cultivation of previously untilled soil. Biogeochemistry 20:161-193.

Davidson, B. A., S. E. Trumbore, and R. Amundson. 2000. Biogeochemistry - Soil
warming and organic carbon content. Nature 408:789-790.

DeGryze, S., J. Six, K. Paustian, S. J. Morris, E. A. Paul, and R. Merckx. 2004. Soil
organic carbon pool changes following land-use conversions. Global Change
Biology 10:1120-1132.

Del Gado, I., J. Six, A. Peressotti, and M. F. Cotrufo. 2003. Assessing the impact of land-
use change on soil C sequestration in agricultural soils by means of organic matter
fractionation and stable isotopes. Global Change Biology 911204-1213.

Denef, K., J. Six, H. Bossuyt, S. D. Frey, E. T. Elliott, R. Merckx, and K. Paustian. 2001.
Inﬂuence of dry-wet cycles on the interrelationship between aggregate, particulate

organic matter, and microbial community dynamics. Soil Biology & Biochemistry
33:1599-1611.

Dorland, S., and E. G. Beauchamp. 1991. Denitriﬁcation and ammoniﬁcation at low soil
temperatures. Canadian Journal of Soil Science 71:293-303.

Fierer, N., and J. P. Schimel. 2002. Effects of drying-rewetting frequency on soil carbon
and nitrogen transformations. Soil Biology & Biochemistry 34:777-787.

F lessa, H., P. Dorsch, and F. Beese. 1995. Seasonal-variation of N20 and CH4 ﬂuxes in
differently managed arable soils in southern Germany. Journal of geophysical
research-atmopheres 100123 1 15-23124.

77

Franzluebbers, A. J ., and J. L. Steiner. 2002. Climatic inﬂuences on C storage with no
tillage. Pages 71—86
in J. M. Kimble, R. Lal, and R. F. Follett, editors. Agriculture Practices and Policies for
Carbon Sequestration in Soil. CRC Press, Boca Raton.

Gale, W., C. Cambardella, and T. Bailey. 2000. Surface residue- and root-derived carbon
in stable and unstable aggregates. Soil Science Society of America Journal
64:196-201.

Goldberg, D. E., and S. M. Scheiner. 2001. ANOVA and ANCOVA: Field competition
experiments. Pages 77-98 in S. M. Scheiner and J. Gurevitch, editors. Design and
analysis of ecological experiments. Oxford University Press, New York.

Grandy, A. S., G. A. Porter, and M. Erich. 2002. Organic amendment and rotation crop
effects on the recovery of soil organic matter and aggregation in potato cropping
systems. Soil Science Society of America Journal 66:131 1-1319.

Grant, B., W. N. Smith, R. Desjardins, R. Lemke, and C. Li. 2004. Estimated N20 and
C02 emissions as inﬂuenced by agricultural practices in Canada. Climatic
Change 65:315-332.

Hanson, P. J ., N. T. Edwards, C. T. Garten, and J. A. Andrews. 2000. Separating root and
soil microbial contributions to soil respiration: A review of methods and
observations. Biogeochemistry 48:1 15-146.

Hart, S. C., J. M. Stark, E. A. Davidson, and M. K. Firestone. 1994. Nitrogen
mineralization, immobilization, and nitriﬁcation. Pages 985-1018 in R. W.
Weaver, J. S. Angle, P. J. Bottomley, D. F. Bezdicek, M. S. Smith, M. A.
Tabatabai, and A. G. Wollum, editors. Methods of Soil Analysis, Part 2-
Microbiological and Biochemical Properties. Soil Science Society of America,
Madison, Wisconsin, USA.

Inoue, Y., A. Olioso, and W. Choi. 2004. Dynamic change of C02 ﬂux over bare soil
ﬁeld and its relationship with remotely sensed surface temperature. International
Journal of Remote Sensing 25:1881-1892.

Jackson, L. E., F. J. Calderon, K. L. Steenwerth, K. M. Scow, and D. E. Rolston. 2003.
Responses of soil microbial processes and community structure to tillage events
and implications for soil quality. Geoderma 114:305-317.

J anzen, H. H., C. A. Campbell, S. A. Brandt, G. P. Lafond, and L. Townley-Smith. 1992.
Light fraction organic matter in soils from long term crop rotations. Soil Science
Society of America Journal 56:1799-1806.

Jastrow, J. D., T. W. Boutton, and R. M. Miller. 1996. Carbon dynamics of aggregate—

associated organic matter estimated by carbon-13 natural abundance. Soil Science
society of America Journal 60:801-807.

78

Kammann, C., L. Grunhage, C. Muller, S. Jacobi, and H. J. Jager. 1998. Seasonal
variability and mitigation options for N20 emissions from differently managed
grasslands. Environmental Pollution 102:179-186.

Kauppi, P., and R. Sedjo. 2001. Tecnological and economic potential of options to
enhance, maintain, and manage biological carbon reservoirs and geo-engineering.
Pages 301-344 in B. Metz, 0. Davidson, R. Swart, and J. Pan, editors. Climate
Change 2001, Mitigation. Cambridge University Press, Cambridge, UK.

Keller, M., E. Veldkamp, A. M. Weitz, and W. A. Reiners. 1993. Effect of pasture age on
soil trace-gas emissions from a deforested area of Costa-Rica. Nature 365:244-
246.

Kemper, W. D., and R. C. Rosenau. 1986. Aggregate stability and size distribution. Pages
377—382 in A. Klute, editor. Methods of Soil Analysis 1. Physical and
Mineralogical Methods Second Edition. American Society of Agronomy,
Madison, Wisconsin, USA.

Kessavalou, A., J. W. Doran, A. R. Mosier, and R. A. Drijber. 1998. Greenhouse gas
ﬂuxes following tillage and wetting in a wheat-fallow cropping system. Journal of
Environmental Quality 27 :1 105-1 1 16.

Kirschbaum, M. U. F. 2000. Will changes in soil organic carbon act as a positive or
negative feedback on global warming. Biogeochemistry 48:21-51.

Knorr, W., I. C. Prentice, J. I. House, and E. A. Holland. 2005. Long-term sensitivity of
soil carbon turnover to warming. Nature 433:298-301.

Koponen, H. T., and P. J. Martikainen. 2004. Soil water content and freezing temperature
affect freeze-thaw related N20 production in organic soil. Nutrient Cycling in
Agroecosystems 69:213-219.

Lal, R. 2003. Global potential of soil carbon sequestration to mitigate the greenhouse
effect. Critical Reviews in Plant Sciences 22:151-184.

Lal, R. 2004. Soil carbon sequestration impacts on global climate change and food
security. Science 304:1623-1627.

Livingston, G. P., and G. L. Hutchinson. 1995. Enclosure-based measurement of trace
gas exchange: applications and sources of error. Pages 14-51 in P. Matson and R.
Harriss, editors. Biogenic Trace Gases: Measuring Emissions from Soil and
Water. Blackwell Science, Osney Mead, Oxford, UK.

Luo, Y. Q., R. E. White, P. R. Ball, and R. W. Tillman. 1996. Measuring denitriﬁcation
activity in soils under pasture: optimizing conditions for the short-term
denitriﬁcation enzyme assay and effects of soil storage on denitriﬁcation activity.
Soil Biology & Biochemistryz409-417.

79

Lupwayi, N. Z., G. W. Clayton, J. T. O'Donovan, K. N. Harker, T. K. Turkington, and W.
A. Rice. 2004. Decomposition of crop residues under conventional and zero tillage.
Canadian Journal of Soil Science 84:403-410.

MacKenzie, A. F.,.M. X. Fan, and F. Cadrin. 1997. Nitrous oxide emission as affected by
tillage, com-soybean-alfalfa rotations and nitrogen fertilization. Canadian Journal
of Soil Science 77 :145-152.

MacKenzie, A. F., M. X. Fan, and F. Cadrin. 1998. Nitrous oxide emission in three years
as affected by tillage, com-soybean-alfalfa rotations, and nitrogen fertilization.
Journal of Environmental Quality 27:698-703.

Martens, D. A. 2001. Nitrogen cycling under different soil management systems.
Advances in Agronomy 70:143-192.

Mikha, M. M., and C. W. Rice. 2004. Tillage and manure effects on soil and aggregate-
associated carbon and nitrogen. Soil Science Society of America Journal 68:809-
816.

Miller, A. J ., R. Amundson, I. C. Burke, and C. Yonker. 2004. The effect of climate and
cultivation on soil organic C and N. Biogeochemistry 67 :57-72.

Pacala, S., and R. Socolow. 2004. Stabilization wedges: Solving the climate problem for
the next 50 years with current technologies. Science 305:968-972.

Paustian, K., J. Six, E. T. Elliott, and H. W. Hunt. 2000. Management options for
reducing C02 emissions from agricultural soils. Biogochemistry 482147-163.

Pinto, M., P. Merino, A. del Prado, J. M. Estavillo, S. Yamulki, G. Gebauer, S. Piertzak,
J. Lauf, and O. Oenema. 2004. Increased emissions of nitric oxide and nitrous
oxide following tillage of a perennial pasture. Nutrient cycling in agroecosystems
70:13-22.

Powlson, D. 2005. Will soil amplify climate change? Nature 433:204-205.

Robertson, G. P., E. A. Paul, and R. R. Harwood. 2000. Greenhouse gases in intensive
agriculture: Contributions of individual gases to the radiative forcing of the
atmosphere. Science 289:1922-1925.

Sexstone, A. J ., N. P. Revsbech, T. B. Parkin, and J. M. Tiedje. 1985. Direct
measurement of oxygen proﬁles and denitriﬁcation rates in soil aggregates. Soil
Science Society of America Journal 49:645-651.

Six, J ., E. T. Elliott, and K. Paustian. 1999. Aggregate and soil organic matter dynamics
under conventional and no-tillage systems. Soil Science Society of America
Journal 63:1350-1358.

80

Six, J ., E. T. Elliott, K. Paustian, and J. W. Doran. 1998. Aggregation and soil organic
matter accumulation in cultivated and native grassland soils. Soil Science Society of
America Journal 62:1367-1377.

Six, J ., S. M. Ogle, F. J. Breidt, R. T. Conant, A. R. Mosier, and K. Paustian. 2004. The
potential to mitigate global warming with no-tillage management is only realized
when practised in the long term. Global Change Biology 10:155-160.

Smith, K. A., and F. Conen. 2004. Impacts of land management on ﬂuxes of trace
greenhouse gases. Soil Use And Management 20:25 5-263.

Smith, P., K. W. Goulding, K. A. Smith, D. S. Powlson, J. U. Smith, P. Falloon, and K.
Coleman. 2001. Enhancing the carbon sink in European agricultural soils:
including trace gas ﬂuxes in estimates of carbon mitigation potential. Nutrient
cycling in agroecosystems 60:237-252.

Sollins, P., C. Glassman, E. A. Paul, C. Swanston, K. Lajtha, J. W. Heil, and E. T. Elliott.
1999. Soil carbon and nitrogen: pools and fractions. Pages 89-105 in G. P.
Robertson, D. C. Coleman, C. S. Bledsoe, and P. Sollins, editors. Standard Soil
Methods for Long-Tenn Ecological Research. Oxford University Press, New
York.

Wagai, R., K. Brye, S. Gower, J. Norman, and L. Bundy. 1998. Land use and
environmental factors inﬂuencing soil surface C02 ﬂux and microbial biomass in

natural and managed ecosystems in southern Wisconsin. Soil Biology &
Biochemistry 30:1501-1509.

West, T. O., and W. M. Post. 2002. Soil organic carbon sequestration rates by tillage and

crop rotation: a global data analysis. Soil Science Society of America Journal
66:1930-1946.

81

CHAPTER 4
Changes in Aggregate-Protected C, Inorganic Nitrogen and Enzymes

Following Initial Tillage of an Undisturbed Soil Profile

ABSTRACT

Understanding the short-term effects of tillage on the persistence of soil organic
matter following years or decades of no-till is critical to developing C conservation
strategies. We annually plowed replicated plots in a previously uncultivated
midsuccessional soil between 2002 and 2004 and investigated changes in aggregation,
soil organic matter dynamics, inorganic N, and N enzyme activities. Within 60 d of
initial cultivation, soil aggregates in the 2000-8000 um size class at 0-7 cm depth
declined to levels indistinguishable from those in an adjacent agricultural soil cultivated
for >50 y. Inter-aggregate, unprotected light fraction (LF) increased following
cultivation, as did particulate C in soil fractions with densities < 1.9 g cm'3. Nitriﬁer
enzyme activity in 2003 was 5-fold higher in cultivated than in uncultivated treatments
(127 vs. 25.4 ug N kg’l soil h'l) to 7 cm soil depth and 3-fold higher at 7 — 20 cm depth
(45 .2 vs. 14.9 vs. ug N kg'l soil h"). Both nitriﬁcation and denitriﬁcation were
stimulated by tillage in 2004. Changes in the mass of total soil C were not detectable over
the 3 years of this study but there was signiﬁcant vertical homogenization of C within the
proﬁle across all soil C pools. Our study demonstrates that plowing once immediately
and substantially alters aggregation, the vertical distribution of C in the soil proﬁle, light-

fraction and particulate C dynamics, and N-cycle enzyme activities. The effects of even

82

occasional cultivation in no-till soils may thus be detrimental to soil structure and C

stocks. Results highlight the value of permanent no-till.

INTRODUCTION

Changes in soil organic matter (SOM) and other soil properties following the
conversion of native ecosystems to agriculture are well known. SOM declines, C is
redistributed between surface and subsurface horizons and is released from protected
microsites (e.g. Del Gado et al., 2003; DeGryze et al., 2004). Accompanying changes
include increased soil temperature, modiﬁed trace gas ﬂuxes (e. g. Smith et al., 2001;
Smith and Conen, 2004), and altered microbial community structure and function (e. g.
Cavigelli and Robertson, 2000; Buckley and Schmidt, 2001), among others. Typically
SOM declines over time to 30-60% of original values (e. g. Davidson and Ackerman,
1993; Buyanovsky et al., 1997; West and Post, 2002; Lal, 2003). In tropical ecosystems
change occurs rapidly, within months of initial cultivation (Houghton et al., 1985;
Detwiler, 1986; Brown and Lugo, 1990). In temperate ecosystems it is more difﬁcult to
generalize because data for the ﬁrst 10 y of cultivation are not currently available
(Davidson and Ackerman, 1993; West and Post, 2002; Miller et al., 2004).

Ultimately, net greenhouse gas mitigation by soil C storage depends on the
persistence of stored SOM. But an important challenge to sequester C in soils is the
vulnerability of stored carbon to oxidation following intermittent tillage, a common
practice in US. soils under no-till management (Paustian et al., 2000; Lal, 2004; Pacala
and Socolow, 2004). Theory and a handful of relevant studies suggest that immediate C

losses from temperate region soils may be less dramatic than from tropical ecosystems

83

(c.f. Bowman et a1. 1990). Tiessen and Stewart (1983) found no difference four years

after cultivating a silt loam. Other authors (e. g. Pierce et al., 1994; VandenBygaart and
Kay, 2004) have also failed to ﬁnd an effect within the ﬁrst few years of initial
cultivation, perhaps because of detection difﬁculties: short-term total soil C changes are
notoriously difﬁcult to detect because of high background C concentrations in most
arable soils, spatial variability, and the redistribution of above-ground pools upon
cultivation (Sollins et al., 1999). Consequently, soil C permanence may be better
predicted by changes in more readily detected factors known to affect C storage such as
soil aggregation, the distribution of C in the soil proﬁle, the dynamics of speciﬁc SOM

fractions in soil, and enzyme activities that reﬂect soil microbial activity.

Macroaggregates (>250 um soil particles) may be espeCially good predictors
because of their importance for protecting recently deposited, labile organic matter
(Angers and Giroux 1996; J astrow 1996). Isotope studies and '3 C CP/MAS NMR
spectroscopy have demonstrated that newly incorporated crop residues, root derived
carbon, and young SOM are found in macroaggregates (Golchin et al. 1994; Six et al.
1998; Gale et al. 2000a; 2000b). In ecosystems with frequent soil disturbance,
accelerated turnover rates of macroaggregates limit the physical stabilization of labile
SOM compounds such as particulate C (SOM > 53 pm). In a recent study Grandy and
Robertson (2005, in review) demonstrated C protection within 2000-8000 um aggregates

reduces its turnover rate by as much as 50%.

In this paper we investigate the short-term effects of tillage on soil physical and
microbial processes that control C and N turnover and persistence. We observed

aggregation, inter and intra-aggregate LF pools, particulate organic matter, and nitriﬁer

84

to
soil
me

pri

If]

C0

and denitriﬁer enzyme activities in a previously uncultivated mid-successional

community subjected to three years of cultivation. By studying a previously undisturbed
soil proﬁle and leaving the site fallow following cultivation we are able to isolate and
measure cultivation effects independent of changes in plant community composition and

prior cultivation.

MATERIALS AND METHODS

Site Description and Experimental Design

Our experimental site was a previously never-tilled ﬁeld at the WK. Kellogg
Biological Station (KBS) Long-Tenn Ecological Research site in southwest Michigan,
USA (42° 24' latitude, 85° 24' longitude). In 1956 the site was Cleared of trees and
thereby converted from a northern hardwood forest to a midsuccessional grassland
community that has since been mowed each fall, with mown biomass left in place. Soils
at the site are Kalamazoo (ﬁne-loamy) and Oshtemo (coarse-loamy) mixed, mesic, Typic
Hapludalfs developed on glacial outwash (Crum and Collins, 1995). The two series co-
occur at KBS with variation within a series often as great as variation between series.
Prior to initial cultivation, soils contained 31.8 (i 1.4) g C kg'1 and 2.59 (:t 0.11) g N kg'l,
and soil pH averaged 5.93 (:t 0.11); the same soil C levels had been measured 3 years and
12 years prior to this study and are statistically indistinguishable from C levels in nearby
undisturbed forests on the same soil types (Robertson et al., 2000).

In 2002 we established in this ﬁeld eight 3 x 6 m plots to which we assigned 4
replicates of two tillage treatments (cultivated and uncultivated control) in a randomized

complete block design. Cultivated sites were mowed with biomass left in place prior to

85

cultivation to 19 cm. We used a moldboard plow and a disc for primary and secondary
cultivation, respectively. Cultivation occurred on 25 June 2002 (DOY 176), 15 June

2003 (DOY 166), and 20 June 2004 (DOY 172).

Soil and Plant Sampling

Samples for aggregate and total C and N analysis were collected prior to initial
cultivation on 18 June 2002, and post-initial-cultivation on 24 August 2002, 21 October
2003, and 24 September 2004. An additional sample was taken on 27 May 2003 for
aggregate analysis. On each sample date, ﬁve 3.8 cm diameter soil cores were taken
from each plot to a depth of 20 cm, placed in plastic bags, and refrigerated (< 7 (1) prior to
sieving through an 8 mm sieve and air-drying at 20 °C.

Plant and litter C estimates prior to initial cultivation were estimated by drying and
analyzing litter and plant samples collected on 21 June 2002 from two 625 cm'2 quadrats
in each plot. Changes in the litter layer after cultivation were estimated on 15 September
2003 from two 625 cm'2 quadrats in each of four adjacent plots cultivated for the ﬁrst
time in 2003. Organic C and total N concentrations of plant and soil samples were
determined by dry combustion and gas chromatography in a CHNS analyzer (Costech

ECS 4010, Costech Analytical Technologies, Valencia CA.).

Particulate SOM Density Distribution
Cultivation effects on the distribution of particulate SOM in four density fractions
were determined using a sequential fractionation technique in sodium polytungstate

(NAPT). Whole soil samples (15 g) were dispersed in 0.5% sodium hexametaphosphate

86

and poured through a 53 um sieve. Particulate materials were sequentially fractionated
in sodium polytungstate at four densities to separate POM into four density fractions: (>
1.9, 1.6 — 1.9, 1.3 — 1.6, < 1.3 g cm‘3). Clay plus silt associated SOM (< 53 pm) was

determined by the difference between whole-soil C and total particulate C.

Aggregate and Light Fraction Separation
Aggregate distribution was determined by hand on triplicate 35 g air-dried soil
samples by wet-sieving in water through a series of 2000 pm, 250 um, and 53 um sieves.
Mean weight diameters (MWD) of sand-free aggregates were determined by calculating
the sum of the products of the mean diameter of each size fraction and the proportion of
the total sample weight in that fraction (Kemper and Rosenau, 1986).

The method we used to separate inter- and intra-aggregate light ﬁaction (LF) is
based on previously published protocols (Six et al., 1998; Gale et al., 2000). Aggregate
subsamples (8 g) were pre-wetted with 4 mL H20 prior to LP analysis to minimize
aggregate slaking during LF separation. Aggregates were transferred from the membrane
ﬁlters to 100 mL beakers after 16 h with 5 mL aliquots of NAPT at a density of 1.62 g
cm’3. A total of 55 mL NAPT was used for each sample. A preliminary test showed that
the ﬁnal density of the sodium polytungstate was 1.60 g cm'3 following equilibration with
the water contained in aggregates.

After 24 h, inter-aggregate LF was aspirated from the surface of the sodium
polytungstate and then rinsed with at least 600 mL DI H20. Aggregates were then

dispersed to release the intra-aggregate LF using sodium hexametaphosphate and

87

resuspended in NAPT (d = 1.62 g cm'3). The intra-aggregate LF was collected form the

surface.

Inorganic Nitrogen Dynamics

NH; and N03‘ were extracted with l M KC1 from duplicate ﬁeld-moist 10 g soil
samples using a 1:5 soil/extractant ratio. Soil extracts were ﬁltered with a syringe ﬁlter
using a type A/E glass ﬁber ﬁlter (Pall Corporation, East Hills, NY). F iltrates were
stored in 7 mL scintillation vials and frozen until analysis for NIL;+ and NO3'. Both
analyses were performed on an Alpkem 3550 Flow Injector Analyzer (01 Analytical,
College Station, TX). Additionally, inorganic N was determined to a depth of 7-20 cm

on three dates in 2003 and on all dates in 2004.

Nitriﬁer and Denitriﬁer Enzymes

We used a shaken slurry method (Hart et al., 1994) to test for changes in nitriﬁer
enzyme activity to 0-7 and 7-20 cm soil depths following cultivation on 13 October 2003
(DOY 286) and 17 August 2004 (DOY 230). Brieﬂy, 30 g ﬁeld-moist soil was combined
with 100 mL solution containing non-limiting quantities of NHa+ in a 160 mL jar capped
with paraﬁlm. The paraﬁlm was perforated with a needle to allow rapid gas exchange
while minimizing water loss. Jars were placed on a rotary shaker and rotated at 200 rpm.
Each ﬂask was sampled 4 times during a 28 h incubation period. Soil extracts were

centrifuged, ﬁltered, and frozen until analysis for NO3’.

Denitriﬁer enzyme activity was measured to 0-7 and 7-20 cm soil depths on soil

samples collected on 17 August 2004 using methods described in Groffrnan et a1. (1999).

88

We combined 10 g ﬁeld-moist soil with a 10 mL solution consisting of NaC4H404 (2

mM) and KNO3' (1mM) in 160 mL serum vials. We capped these vials with rubber
septa and ﬂushed them with N2 ﬁve times over 10 min to create anaerobic conditions
prior to adding acetylene (15%). Acetylene blocks the conversion of N20 to N2 so that
N20 measurements alone can be used to estimate total denitriﬁcation potential. High
microsite substrate availability was maintained in the jars by keeping them on a rotary
shaker (190 rpm) during the incubation. N20 concentrations were determined on four 0.5
mL headspace samples collected every 12 min. We completed the entire analysis within
60 min of the substrate additions in order to reduce the potential for cell growth or
reproduction to alter NzO production during incubation (Tiedje, 1987; Luo et al., 1996).
N20 was analyzed using a gas chromatograph outﬁtted with a 63 Ni electron capture

detector (350 °C).

Statistical Analysis

Soil aggregate size distributions and aggregate-associated C pools were corrected
for sand content > 53 um. Carbon content of soil organic matter fractions was calculated
on an area basis by correcting for bulk density. Tillage effects on SOM pools and enzyme
activities were analyzed by Proc Mixed (Version 8.2, SAS Institute, 1999) using a
randomized complete block design analysis of variance (ANOVA). Treatment, soil depth
and, in the case of C pools, SOM fraction were considered ﬁxed effects and block a
random effect.

Single degree of freedom comparisons were made to compare differences among

soil fractions and treatments at different depths using the LSD statistic to calculate a 95%

89

conﬁdence interval around the differences between means generated using the diff

option in Proc Mixed. The LSD was canied out using the PDMIX800 algorithm
(Saxton, 1998). Soil inorganic N concentrations were analyzed by Proc Mixed using a
randomized complete block design analysis of variance (ANOVA) with repeated
measures (SAS Version 8.2, SAS Institute, 1999). Treatment and sampling date were
considered ﬁxed effects and block a random effect. Where there were signiﬁcant day of
year by treatment interactions, differences between treatments on separate days were

determined using the slicing command in SAS.

RESULTS
Total Soil C and N
On 24 August 2002, 60 d after initial tillage, there were signiﬁcant changes in the
depth distribution of soil C and N (Table 4.1). C and N concentrations in tilled plots
were signiﬁcantly lower than those in conventional systems at 0-7 em but greater at 7-20
cm. Expressed on an areal basis, soil C and N showed the same trend, despite tilled plots
having a slightly higher bulk density at 0-7 cm and lower bulk density at 7-20 cm. These
changes in soil C and N generally persisted in 2003 and 2004 although differences
between C concentrations at 7-20 cm were not signiﬁcantly different (Table 4.1).
Expressed on an areal basis, total soil C at 0-20 cm was not different between treatments

in any year.

Aggregation

90

Prior to tillage in June 2002 there were no differences in soil aggregation between

treatments (Figures 4.1, 4.2). Within 60 d, tillage decreased 2000-8000 um aggregates
at both soil depths and increased the amount of soil in the 53-250 um and <53 urn size
classes at 0-7 cm. 2000-8000 um aggregates declined from 0.47 to 0.15 g g". These
changes persisted over the winter and into the following spring when control plots had
0.40 g g" in the 2000-8000 um size class and tilled plots had 0.24 g g". Differences in
2000-8000 um aggregates at both depths were also evident in October 2003 but not in
September 2004. Tillage increased soil in size classes < 250 um at 0-7 cm throughout the
experiment although there were generally no differences in these classes at 7-20 cm.
Shifts in soil from 2000-8000 um size classes into smaller sizes following cultivation are
also evident from statistical comparisons among size classes within treatments. Control
plots generally contained greater soil in macroaggregate than microaggregate classes,
particularly at 0-7 cm, whereas in tilled plots the 53-250 um class contained amounts of
soil frequently equal to that in the 250-2000 um class and higher than that in the 2000-

8000 um class.

Particulate and Aggregate-associated C
Initial cultivation had signiﬁcant effects on the distribution of SOM in different
particulate organic matter POM density fractions and soil depths (Table 4.2). On 24
August 2002 cultivation reduced SOM in clay fractions and POM > 1.9 g at 0-7 cm and
increased all POM classes and clay-associated SOM at 7-20 cm. Cultivation effects on
POM distribution were similar in 2003 and 2004 although they were not as consistent

across density fractions. Treatment differences were related to a homogenization of C

91

availability at 0-20 cm due to cultivation. In control plots, there were signiﬁcant

differences in all POM fractions and clay associated C at all sampling dates after tillage
(except the 1.3 - 1.6 g cm'3 in 2004); in contrast, the tilled plots had few differences in C
concentrations between depths. At 0-20 cm POM increased in densities < 1.9 g cm'3
while there were no increases in the clay fractions or POM pools > 1.9 g cm'3 (Table 4.2).

Tillage changed the depth distribution of C associated with aggregate size fractions

(Figure 4.3) and increased inter-aggregate light fraction C in the 2000-8000 um class at
0-7 and 7-20 cm in 2002 and also in the 250-2000 um class at 7-20 cm in 2002 (Figure
4.4). In 2003, tillage increased inter-aggregate LF C in the 250-2000 um class at 0-7 cm
and LF C in the 2000-8000 um class at 7-20 cm. Similar to whole soil C, there was a
redistribution of inter-aggregate and intra-aggregate LP in the soil proﬁle following
cultivation. Inna-aggregate LF concentrations were generally indistinguishable among
treatments; however, tillage increased intra-aggregate LP in the 250-2000 um class in

2002 and the 2000-8000 um class in 2003 (Figure 4.5). Tillage, soil fraction and depth

effects on C/N ratios are presented in Table 4.3.

Inorganic Soil N and Enzymes

Cultivation in 2002 increased extractable soil NO3' concentrations at 0-7 cm to 3.18
ug NO3'-N g'1 from 0.41 pg NO3'-N g'1 in control plots after 24 (I (Figure 4.6). NO3‘
concentrations in cultivated plots remained higher on DOY 207, 228 and 234 and peaked
on DOY 228 (14.6 vs. 0.44 p g NO3'-N g"). NH4+ concentrations were also increased by
cultivation on DOY 190 (87%), 200 (312%), 207 (176%), 228 (193%), and 234 (82%).

No data are available for 7-20 cm in 2002.

92

In 2003 cultivation signiﬁcantly increased NO3' concentrations on DOY 170 and

216 but there were no differences on the other sampling dates (Figure 4.6). NH4+
concentrations were signiﬁcantly greater in cultivated plots on DOY 170 and 216 but
lower in these plots relative to control plots on the following three sampling dates. At 7-
20 cm there were no differences in N03' or NHa+ between treatments on the three dates
for which data are available. In 2004 N03' concentrations on DOY 180 were 11.8 pg
NO3'-N g”1 in cultivated plots and 0.55 pg NOg’-N g’1 in control plots. N03”
concentrations remained higher in cultivated plots on DOY 194 and 208. NH4+
concentrations were reduced by cultivation on four of seven sampling dates in 2004. At
7-20 cm, results were similar but treatment differences were smaller in magnitude.
Tillage increased NO3' concentrations on DOY 180, 194, 208, 230, and 268. The greatest
differences occurred on DOY 180 when tilled plots had 6.09 and control plots 1.40 pg

NOg‘-N g". Tillage increased NH4+ concentrations on DOY 194 (Figure 4.6).

N Enzyme Activity

Nitriﬁer enzyme activity in 2003 (Figure 4.7) was 5-fold higher in cultivated than in
uncultivated treatments (127 vs. 25.4 pg N kg'l soil h") to 7 cm soil depth and 3-fold
higher at 7 — 20 cm depth (45.2 vs. 14.9 pg N kg'1 soil h"). Similar trends were observed
in 2004 when cultivated plots had higher nitriﬁer enzyme activity at 0 - 7 cm (109 vs.
35.5 pg N kg'I soil h'l) and 7 - 20 cm depths (45.7 vs. 10.5 pg N kg'I soil h"). Denitriﬁer
enzyme activity (Figure 4.8) was similar at 0-7 cm depth; at 7-20 cm depth tillage

increased DEA from 15.6 to 153 pg N20-N kg'1 h].

93

DISCUSSION
Soil Aggregates

Changes in soil aggregates were immediate, signiﬁcant, and persistent following
initial cultivation (Figure 4.1). Within 60 d of cultivation, soil in the 2000-8000 pm size
class at 0-7 cm soil (Figure 4.2) declined to levels indistinguishable from those in an
adjacent agricultural soil cultivated for >50 y (Grandy and Robertson, 2005 in review).
Cultivation also led to fewer aggregates in this size class at 7-20 cm depth, demonstrating
that declines at the soil surface were not solely due to a redistribution of aggregates to

increased soil depths.

Many prior studies have demonstrated that long-term cultivation reduces soil
aggregation (e. g. Gupta and Germida, 1988; J astrow et al., 1996; Mikha and Rice, 2004).
DeGryze et al. (2004) reported that at 0-7 cm soil depth, soil aggregate mean weight
diameter (MWD) in conventionally managed agricultural soils was ca. 3-fold lower than
in early successional, afforested, and native forest ecosystems; at 7-25 cm depth
reductions in MWD were also signiﬁcant and due largely to declines in aggregates >
2000 pm. Six et al. (2000) showed that cultivation reduced aggregation in four soils
ranging in texture from sandy loam to silty clay loam. Our results show that tillage
immediately destroys macroaggregates and that these declines in soil structure persist

throughout the growing season and following winter.

Increased turnover rates of macroaggregates following cultivation are due to the
direct effects of tillage and also to changes in annual plant cover and long-term changes
in soil C and microbial activity. Fungal hyphae and ﬁne-root networks contribute to

macroaggregate stability (Bethlenfalvay et al., 1994). These networks are particularly

94

susceptible to disruption following disturbance (J ansa et al., 2003). Bare soil surfaces

following tillage expose aggregates to the direct impacts of rainfall and make them more
susceptible to slaking and dispersion following rapid wetting (Kooistra and van
Noordwijk, 1996). A combination of direct and indirect effects of tillage likely

contributed to the rapid declines in soil structure at our site.

The persistence of cultivation effects on soil structure was evident prior to tillage in
2003 when aggregation in cultivated sites remained substantially lower (Figures 4.1, 4.2),
despite the potential for root growth and freeze-thaw and wet-dry cycles to have affected
aggregation in both tillage systems. Plant roots can increase aggregation by enmeshing
small particles into stable macroaggregates; by supplying organic substrates such as root
hairs, sloughed cells and mucilage; and by inﬂuencing soil moisture content (Reid and
Gross, 1982; Perfect et al., 1990). Freeze-thaw and wetting drying cycles can directly
break down aggregates but also inﬂuence structure by changing the proximity and
alignment of mineral and organic particles to facilitate aggregation (Kemper and
Rosenau, 1986; Rasiah et al., 1992; Denef et al., 2001). The direct and indirect
destructive effects of tillage were apparently greater than these other forces that serve to

equalize aggregation in the two treatments.

Soil Organic Matter

The vertical homogenization of soil C and N in the proﬁle following cultivation is
consistent with results from other studies (Tables 4.1, 4.2). VandenBygaart and Kay
(2004) studied the effects of moldboard plowing soils in three different textural classes

(sandy clay loam, silty clay loam, and sandy loam) after 22 yr of no-till management. In

95

each of these textural classes they found that soil C declined at 0-5 cm and increased at

lower depths with the net result being no change in total soil C. Kettler et al. (2000)
studied the effects of moldboard plowing a soil that had been in a no-till wheat-fallow
system. Five yr after plowing there remained differences in SOM stratiﬁcation to 30 cm
but no differences in total soil C. Pierce et al. (1994) found in Michigan that plowing
ﬁelds in no-till for 6 to 7 yr redistributed C to 150 mm depth and that this effect persisted
for 4 to 5 yr aﬁer plowing although there were no changes in total soil C. We similarly
found changes in SOM stratiﬁcation and that this homogenization occurred across SOM

pools of different densities.

After three years of annual tillage we were still unable to detect differences in total
soil C. Rates of SOM loss from temperate ecosystems immediately following cultivation
have rarely been studied and the available data are inconclusive. Bowman et al. (1990),
for example, found 40% organic C losses within three years of cultivation (15.7 vs. 9.4
kg 1113) and that between three and twenty years after cultivation there was little
additional C loss. These rapid rates of soil C change are unusual and may have been due
to their sandy soil texture (Davidson and Ackerman, 1993; West and Post, 2002) as well
as changes in vegetation that accompanied tillage. Substantial decreases in litter inputs
associated with agricultural conversion may also have contributed to the C losses as
vegetation changed from short-grass steppe to an unfertilized wheat-fallow rotation.
Tiessen and Stewart (1983) found no change in soil C concentration aﬁer four years of
cultivating a previously no-till silt loam soil. The high silt (49%) and clay (20 %)
contents of these soils may have increased their protective capacity and made them more

resistant to the destructive effects of cultivation (Wander and Bidart, 2000).

96

Despite the lack of change in total soil C we measured signiﬁcant changes in POM
pools to 20 cm and also in inter-aggregate SOM (Table 4.2; Figure 4.2). Collectively,

these changes suggest increases in organic matter availability. POM concentrations and
quantity to 20 cm in 2002 increased in fractions < 1.9 g cm’3 while in 2004 changes
occurred in fractions <1.6 g cm'3. Light fraction pools have been shown to be correlated
with soil surface respiration rates (J anzen et al., 1992; Alvarez et al., 2000) and its
depletion represents a major portion of C loss in cultivated soils (Cambardella and Elliott,
1992, 1994). We measured no changes in clay fractions or in POM fractions with a
density > 1.9 g cm’3. Both these fractions tend to be more stable than LF POM (Golchin
et al., 1994). Heavy fraction SOM is considered both more chemically resistant to
decomposition (recalcitrant) and more physically protected than LF (Swanston et al.,
2000). Decreased physical protection of SOM is also suggested by the increases in inter-
aggregate LF at both 0-7 at 7-20 cm. The biomass incorporated with tillage thus moved
primarily into POM pools with low densities and inter-aggregate light fraction pools. In
2002 this aboveground C consisted of litter (142 i 30 g C m'z) and plant biomass (228 :1:

11ng'2).

Soil N Pools and Enzymatic Activity

Long-terrn cultivation changes microbial community structure and function
(Cavigelli and Robertson, 2000; Steenwerth et al., 2002). Buckley and Schmidt (2001)
quantiﬁed the effects of plant community composition, fertilization, tillage and historical
cultivation on microbial community structure at the KBS LTER. They found that

changes in crop management and recent variations (7 yr) in tillage intensity had no

97

inﬂuence on microbial community structure although agricultural systems differed from

the never-tilled midsuccessional community examined in our study. Further, a 50-yr-old
historically—tilled midsuccessional community had similar microbial community structure
to annual and perennial cropping systems, demonstrating the persistent effects of
agricultural management on microbial communities. The persistent effects of agricultural
management on microbes was also demonstrated by Compton and Boone (2000) who
found that nitriﬁcation rates remained higher in agricultural than in undisturbed soils
more than a century after agricultural abandonment. Our study demonstrates that tillage
effects on microbial processes begin to occur immediately after cultivation (Figures 4.7,
4.8). We found changes in nitriﬁer enzyme activity the year following initial tillage
(2003) and continued to detect differences due to tillage in 2004, when we were also able
to detect differences in denitriﬁer enzyme activity. Increased nitriﬁer enzyme activities
likely contributed to the greatly increased soil NO3' concentrations as did the slow
recovery of the plant community following tillage. It is also likely that increased
denitriﬁer enzyme activities accelerated N20 ﬂuxes, which increased 6-fold in 2002, 2-

fold in 2003 and 6-fold in 2004 (Grandy and Robertson, 2005, in review).

Other studies have also demonstrated immediate changes in microbial processes
following soil disturbance. Calderon and Jackson (2000) demonstrated that sieving can
alter microbial biomass C and phospholipid fatty acid abundance and distribution within
2 d of cultivation. In a follow-up study, Calderon et al. (2001) showed in the ﬁeld that
tillage of a soil used for vegetable production changed denitriﬁcation rates and
phospholipid fatty acid proﬁles, indicating that microbial communities may have

changed.

98

Tillage thus has both immediate and persistent effects on microorganisms. One
explanation for this is that soil properties inﬂuencing microbes respond rapidly to
cultivation and also do not recover quickly. Changes in the spatial distribution and
protection of resources may be critical. Our results show that vertical homogenization of
soil C occurs after one plowing. Other studies have demonstrated that the spatial
distribution of resources responds rapidly to cultivation and may take decades to recover
(Robertson etal., 1988; 1997). Further the relocation of resources to deeper in the soil
proﬁle may inﬂuence long-term decomposition dynamics and microbial processes. In
agricultural systems, Lupwayi et a1. (2004) found that incorporation of wheat litter with
conventional tillage increased its decomposition rate by 48%; Burgess et a1. (2002) found
that litter decomposition rates at 20 cm were 52 - 105% greater than those at the soil
surface.
Implications for Soil management

Marland et al. (2001) identiﬁed soil C permanence as the fundamental challenge
to C sequestration. Although tillage is the major threat to stored soil C, most farmers for
agronomic reasons periodically cultivate otherwise no-till ﬁelds. Rather than focussing
on total soil C responses, which can be extremely hard to detect in short-term studies, we
measured changes in the mechanisms underlying C and N cycling to infer potential long-
term changes. We show immediate reductions in the physical protection of C,
homogenization of C in the soil proﬁle and changes in microbial processes following
cultivation of a previously uncultivated soil. These changes drastically increased C02
ﬂuxes (1.0-1.9 g C02-C m-2 (1'1 over three growing seasons; Grandy and Robertson, 2005,

in review) and over a longer time this enhanced C02 efﬂux would likely lead to

99

detectable C losses. These results and recent research demonstrating that long-term no-

till cropping can reduce soil NzO emissions (Six et al., 2004) with no signiﬁcant
ecological or yield tradeoffs (Grandy and Robertson, 2005, in review) highlight the need
to maintain no-till management. Efforts should be made in speciﬁc regions to identify

and overcome the agronomic limitations to no-till associated with different soil types.

CONCLUSIONS
We tilled a previously uncultivated ﬁeld to investigate the immediate effects of
tillage on aggregation, SOM dynamics, and nitrifer and denitriﬁer enzyme activities. By
choosing a previously undisturbed soil proﬁle and leaving the site fallow following
cultivation we were able to isolate and measure the effects of cultivation without the

confounding effects of changes in plant communities or prior cultivation. Results include:

1) Within 60 d of cultivation, soil in the 2000-8000 pm size class at 0-7 cm declined
to levels indistinguishable from those in an adjacent agricultural soil cultivated for
>50 y (Grandy and Robertson, 2005, in review). These changes were persistent.

2) Inter-aggregate, unprotected LF increased as a result of cultivation.

3) Nitriﬁer enzyme activity one year after initial cultivation was 5-fold higher in
cultivated than uncultivated treatments (25.4 vs. 127 pg N kg'l soil h!) to 7 cm
soil depth and 3-fold higher at 7 — 20 cm depth (14.9 vs. 45.2 pg N kg’l soil h").
In the following year there were signiﬁcant tillage effects on both nitriﬁer and

denitriﬁer activity.

100

4) Total soil C changes were not detectable by the end of year 3 but there was

signiﬁcant vertical homogenization of C within the proﬁle.

Our results demonstrate that soil C and N are immediately inﬂuenced by cultivation and

that these changes are persistent through the growing season and subsequent winter.

101

:00» E0 5 32: 3:02:00 0:0 25:00 50.50: Eu om 9 U :8 .08: :_ 80:83:: 0: 0:03 0.0.05.
:00: :3 :0 0.500 Amodmo 380$? b.0330; 00 00:0. 20.0%:6 .3 330:8 :528 0 55:5 0:002:
500

ovod mmod _oo.o mood _oo.o mmod wood wood onod Cod wood _oo.o Sod _ood _oo.o *0w0zﬁ.
_oo.o mood .ood _oo.o _ood :od wood _oo.o Sod mood mood omod _ood _ood _oo.o c300

momd mmmd mood Bod mood VmNd :Nd Nvmd wwod wwnd Nﬁd 02d owod did mo_.o 0w0=C1

mo=_m> m . bosom

 

 

3001—1111. <>OZ<

08d 0mo.m 0:: “Em—d £10.— 0v~d 03M 2mg Ed 09:; 0mmd 0ow.m Dom; “om—d coo; “00::

fwd oomN 0N0; 02d oom._ ofd nmo.m 091 odd ooo._ 32d and; 091 oo_.o too; 35:8 Eoomd

02d £0.— 0mm._ nod—d new; dad nod; 0mm— o_d nmvN 02d com; um: fwd 23nd “00::

ootd fod :mod 08d 02.». Ltd ommd ooo._ mmd 0:...“ oomd owﬁm woo; 08d 0_m.m 3::8 Eond

 

N.E we. N.E mo. n.Eo m X. ..\e NE mo. ~-E wx 0.80 w e\e e\.. we: mo. N.E wx M.Ee w 5\5 e\e
€200 35:09 .9659
Z O V=:m Z D Z 0 ism Z D Z 0 53m 2 U
voom :0oE0E0m 0N moom 09800 _N Noom Hmsw=< 0N

 

v.92: 0500 0:: :o voom 0:0 moomoo
max: 05 E 5000 0:0 moon 0:2. mm :0 02.500: 5:03:00 3::— 0530 So om-» 0:0 01o :0 56:0: 0:3 o:0 2 0:0 0 :8 :o $00to 0w0=§ 4.0 030:.

102

.0000 w::a.:00 0:0 .5000 4:08.000. 0 55.3 05:00.0 .200. :003000. AmodVgo 0005-50-00 0:02-2:30 0.00.05 0.0000. 20.0.0.0 .3 003028 0:002H
.0000 w:=o:.00 0:0 £2.00... .200. 0:08.00: 0 5505 0.0000 50.500 0005.00.00 00.00.05.
.0000 m:=q:.00 0:0 £20000... 200. .5000 0 :20...» 00:00:00: 00000:: :003000. 08:20.0...0 .:00....:w_0 00.00.05...

 

 

 

 

 

 

 

 

 

00.0 000.0 00.0 000.0 000.0 00.0 000.0 00.0 000.0 000.0 00.0 000.0 00.0 000.0 000.0 00.00
0.0.0 0.0.0 00.0 000.0 000.0 00.0 000.0 00.0 000.. 000.0 0.00.0 0.0..0 00....0 000.0 00.0 .0080 00-0
0.0: mg
000.0 000.. 000.0 00... 000.0. 000.0 00... 000.0 000.0 000.0. 000.0 000.0 000.0 0.0.0 0.0.0. 00:...
0.00.0 0..0.0 0000.0 000.. 000.0. 000.0 000.0 000.0 000.0 000... 0.00.0 0.00.0 0.000 000.0 000.0. .0080 00-0
00.0.0 000.0 000.0 0000.0 000.0. 000.0 000.0 000.0 0.0.0 000.: 000.0 000.0 0000.0 000.0 000... 00.00
0000.0 0.00.0 000.0 000.0 00.... 000.0 000.0 00.0 0.0.0 0.00.0 0.00.0 0.00.0 0.00.0 0.00.. 0.00.0 .0050 00-0
0000.0 000.. 0.0.0 0.00.. 000.0. 000.0 0.0.. 000.0 0.0.0 0.00.0. 000.0 000.0 000.0 0.00.0 0.00.0. 00::
0.00.0 000.. 0..00.. 0.00.0 0...0.00 0.00.0 0.00.. 00.00.. 0..00.0 0.0.00 0.00.0 0.00.. 00.30 ..000.0 0..00..0 .0080 0-0
=8..wa8
0..v 0..-0.. 00-0.. 02A 00.0 0..v 0.0.. 00-0.. 0..A 00.0 0..v 0..-0.. 0.0.. 0..A 00.0
0000 0000.00000- 00 0000 0000.00 .0 0000 000000. 00

 

0.0003 0800 0:. :o voom 0:0 moom .00 w::% 0.... :_ :_0w0 0:0 moom 0:...- mN :0 000.5000 :2003030 BE:— .00.00 3:90.00 00....
:o 0:... mmvo U 000060000003 0:0 0:00.000.“ 36:00 0.00.. 5 0200: 00.00:. 0_§w.o 0.00.0000: .00 :o_.0b:00:00 0... :0 0.00000 5000 0:0 0w0=:- .m... 030:-

103

.0000 w:.....:00
0:0 .5000 05:500.. 0 25.3 0:260... :00 :003000. Anodqu 03:25.5 .:00....:w.0 0.8.0:. ::.=.00 0 55.3 0.0.00. 0:20....0 .00. 0032.5 0:00.20
.0000 w:..0.:.00 0:0 £280.. :00 0:05.005 0 55.3 050.00 :003000. 000:0.0t.0 00.00.05.
.0000 w:..0..:00 0:0 62.00.. :00 .5000 0 55...» 00:20:00.. 0w0E0 50.5.00. 000:0.0...0 0:00....:w.0 00.00.05...

 

 

 

 

00.00.: nmmdm 00.0.0.3 0mmdm 0Nodm 0.00060 0.20.3 0...”de 00.3.5 000.02.? 009...; 000.352. m..v
0003.3. nomsm 009N5NN 00N.mm 009mm 00.0om.mm undo..m 0.0.o..mm oncodm 00.9.3 00.02.?” 0.0.2.09; 0..-»...
who-w. 0.Ndm 0003.0. 00.90.». 00nd. 00.0.6. 00056. 000.6. ododm 0.0.0.0. 0.9mm 0.3.8 0.6..
33.2 0.3.0. .036. 00%.? own-mm 09*mw.0m 2.02.3 n0mo..m 0mm.m. 0dm.m. mom... 0a.... o..A
:de. 00.Nd. wvod 000m... 03d .2”... wowm... 0030... 0.Nd. .0006 .mvd. 0$d. 00.0 .200. m3
0.06m 00....»- 0..mdm 00.8.3 0.8.: 00.0.0.3 0.0.9.3 0.0.0....N 00m.wm 0.03.2 008.3 Sham omm-mm
00m...mm 0.....de 009mm.mm 0%.0N £08m 059mm 0.2.5 90.093 0.o.wm 090%.? 0om..0 00.0mo.vm doom-omm
000.0N0N DEER 0..m..0m 036m 0wm.~m 0030.3 00.0.3.3 0.0.3.2.. nmmdm 00omdm 00No.mm 090modm ooowooom 3.00.0.2.
.02.... 0modm 00....w. 00000.2 0oo..~ 08.3 00de. 00.0Nodm 0.93 02.8.3 0N0.NN 03.0mm omm-mm
005de now-6m oncodm 030.3 0.34% 00.03...” 00.0....N 0.0.wm.N~ 00.00....wm 000.08.3- 00Nodm 0.2...N ooomomm
0.x..om 0.0mm...” 00.0.8.3 0.0..NwNN 0N..NN 00.00% 0.0mm.mm 0.0.0.9? 0.0.0-om 000..o.om 00N.dm 00.0.3.8 ooowooom $.00.-.0.:.
5%.» 0m. .w wwmd 0. md. ovod Moms .2 . .o. 08d. 0.5... Km... ﬁnd 030 mmv
:wwd ovhd wood. 000m... 0mod. .ood. .02.... 00w..m. 03d. .03d .50.... 0om... omm-mm
:w..~.~. 00o... wmmc... 00m..m. 000.0... 3.0... .0om.m. 00o..~. 030... .03... .3... 0mm... ooomomm 000.0 00.0
0.0003... 003... mummN. 0000.2 008... .00... mommd. 00oo.m. 0.0... 00...... 000.2 0.00.... ooowooom 0.0w0.ww<
0.3:... 00owd. wvm... 00o..~. 00de. .mm... .0m...m. 000w.m. 0am... -.0ood. 00m... 02.... =00 0.00.3
E... .50 E... 0:00 E... 0:00 E... 0:00 E... 0:00 E... 800 5.80.... :om
own no on-.. no own To
voom .00....080m 0N moom .3900 .N Noom 3.0.0.3.. 0N

 

0.0.0.0. 0500 05 :o voom 0:0 moom .o w:..0.0 05 :_ :.0w0 0:0 moom 0:3. mm :o
00....000 5.03.30 .0...:. .voom 0:0 moom .Noom :. 050.000.. :00 30.0-00.0 .0 00.00. 200 :o 0.00..-.0 :0500... 8.00:. 0.:0w.o 0:0 50.00 6.0.0.2.... .020 0.0.0...

104

 

 

Mean Weight Diameter

 

 

an.) 4— 7-20 Treatment

2 I control
3‘ .

.92 I tilled

D * *

E E 2-

.9 E

a:

E 1:

C

8 _

2 0

June Aug. May October Sept.
2002 2002 2003 2003 2004

Figure 4.1. Mean weight diameter (MWD) of soil aggregates. The June 2002
sampling was done prior to cultivation. The May 2003 samples were taken
prior to the second cultivation. Treatment means are shown i standard error (n
= 4); * indicates statistically signiﬁcant (P < 0.05) differences between control
and tilled treatments within a single sampling date. 'l'Indicates signiﬁcant
differences between depths within a treatment and sampling date.

105

4:26.06 bEmoE:w_m 2m 95:2 E2006 .23 9:253: 28 8.8 3:068 m 5.23 momma—o 05m .229 9.2088 98 0:258: m ESE»
23% 5233 30:28am. Enocva. 3839:: 9% mEEEnm ﬁwim a 553» 35.585 3:: 98 .9550 5253 38.2006 $0.0
v 5 «52.:sz bﬁocmssm moanzuﬁ * 5» n 5 6:0 22:88 a :38? 2a mcmoE 0:258; £28323 vacuum 2: 8 SE :81:
$3 $353 88 92 2F .5: 88.? 1 av 95 we: 88-8 u m we: 88 I 88 u 88 Ha 88-88 H 88 "as 282mm“
59» 0:908:00 38-x 2: mac? 88:52 .8320 cam oumwoumwm 500 E :8 mo cossnﬁﬂu 2: co acute own—E1 .N.v 83w;

voom .Eww moom .60 meow 3.2 «com .m:< Noom 0:2.
8v 8 08 88 8v mm 08 88 8v mm 08. 88 8v mm 08 88 8v 8 o8 88

 

 

09:: I
63:8 I So omﬁ

“:95me

 

NP E0 RIO

I

I

l

 

00.0
3.0
00.0
mvd
00.0
00.0
3.0
00.0
0T0

00.0

I!Os t.6 919691669 6

gos L6 919691669 6

se1969166v

se1969166v

106

.Eouutmw bagoEcwmm 2w

€892 E80006 £3» 32585 was 8% @5083. m £55 momma? 03m 2% mam—08mm was E9889: 8 55:3 2390
52,302 80:20th 2.805%? 8830:: .829 mam—088 oﬁmﬁm a 55:3 £22585 3:: can 35:8 59253
889508 80.0 v A: 230$:me 3328sz 38065 * xv n 8 hobo @8888 H 850% 2a 8889588801 .Noom
95:. E 33 8:83:30 REE .voom Ea moon "meow E O oumwohwwm 80.058 38 co $80.3 own—=01 .mé oSwE

 

 

38 .86 88 .60 88 .8<
5 m.
W 9|.
.1 v
B 6
m m
8:: E m 1 oo m m
Sacco I I 00 W a
E0869 1.. Eu om; 009 9 O
o 6 1
0
ON 2W4 9"
a R
6 6
om 6. m
m w
8 m m
e 0

 

009

107

8:20am: 80:85:09 08 8030— 0:20.06 :33 20.500:

0:0 0:00 wEEEwm 0 55:5 momma—0 0&0 .0000 wEEEmm 0:0 30:500.: 0 £53, 0500: 50250: 03:20am: 0:005:06
0886:: .80: w:=o:8m 03:9 0 5:23 3:258: 00:: :5 39:90 :0050: 000:0:0.E: Goo v a: EmoﬁEwmm
£85080 080065 .. xv u 5 :85 0:30:80 0” 559% 0:0 0:008 0:05:00; .Noom 0:3. E 003 gags—:0 BE:— .voom
0:0 moon doom 5 6:08 0_:0w._o :2800 Em: 8000:302095 000-050 .00 5:359: 05 :o $00.00 own—E1 .39 0.53“—

  

 
 

 

voom moom Noom
mm omm ooom om omN oooN mm omm oooN 0 l
. I 1. ,1 0.1:. in u
a 1m 0 n u . . 5 a
. . l0 “4 d
a. 0
10 9 6.
90.9 I a 10 m m
69:00 I a m
EmEﬁmﬁ T NF w 9
Eu 8-: .. 2 a 13
O

 

919691669 L6» 6
0 fl 919691669-191u1

 

108

20600:: b::00..0_:m_m 00 £050. 30:00::

:23 0:09:00: ::0 0:0: w::aE0m 0 55:5 000003 005 .000: 05:53 ::0 0:05:00: 0 55:5 05:0: :0030: 0080:0050
0:005:me 02006:: .000: 0:58:00 0.050 0 55:5 35:58: 00:: 0:0 35:00 50300: 00050000? Goo v a: 0:0o_0_:m_m
0:005:08 0000065 .., Xv u 5 00:00 0:00:08 ”m 56:0 00 0:00_:E0E000:01 .Noom 0:3 :_ 003 8:02:30 .025 .voom
0:0 moom .Noom E 000:. 2:006 8:00: Em: 000M0:mw0-0::_ 00:00:00 00 50330:: 0.: :0 0000000 0w0:_01 .mé 05$:

VOON moom NOON
mm 00w 000m 00 00m 000N mm 00w 000m

    

 

 

1 o lo M
6 I.
n L mm
V6
IN 961
6 9
11m” 6 6
m m 8:: a. a w
_o=:oo I 1v w 9
E09509... E0 om; «.4 _..l._

Im
O

 

 

E0 N10

919691669 L6» 6
O :11 919691669-911m

109

.5000 0:03:00 0:0 908 :00» 0o :00 0 5:00,» 00:08:00: 00:03:30 0:0 75:00

:0050: 0005000000 Amoo V AC 0:00—00:wmm b_00:m:08 08020:: _. xv u 5 8:0 0:00:80 H 53:0 0.0 0:008 0:08:00;
.580 :5 2 09 9: 850509 93 z 2509: .920: :05 88 95 .220: 28:5 88 :88 000

Noom 80 00:00 5:03:50 8030:: 038.3 .A0:: 000000 60800 000200 32: 00:03:50 0:0 3:: 0:00 488% :003
32: 38:90 :0 30%:005 -00 Z 0:0 303:3 +52 .:o:0>:_:0 00:0: 3:59:00 2 00:09:95 :_ 00w:0:U .06 05000

 

 

 

 

  
  

 

 

 

 

88 .0010 >0: 88 50> 9 >00 88 .0010 >00
80 08 :8 08 80 02 :8 08 08 08 08 mm: 000 80 08 8 08 :8 08 80 E
_ F —L _ p L . _ p — h _ . _ L
4 O 41 Q 4 < \\.1 l<< O
7%
to? X a [or
low . low
I 8 $ I 8
n P b _ p o It!!! ss\
(1* 18
o .. :
[om

 

 

l.6 6r! N ogu961ou1 1103

110

 

Amoo v AC 50000000 bﬁoumuﬁm 000 00032 E06000 505 000m Av u 5 00:00 0:00:80
0 850:0 00 0:008 0:09:00; .0020 00:03:00 .000: 00:00:00 M00000 000:8 0:000:00: 0:00 00000 .83 >09 voom
Hmsws< 2 0:0 Gwm >09 moom 000800 2 :0 8:03:30 0:50:00 53:00 088:0 00:00.00: :0 00w:0:o .50 00:30

 
  

moom Noom
9081: 50010 908...: 90:10
10
n m
low 5 m...
N 0
g cm 0
108 Nb.
»
5 m
S 9
0 00:: n 80 m w
.2900 u .. M M
0:08.099 0 A
:80

 

111

Treatment

   

 

 

 

 

I Control
g 2507 I Tilled
> a
E ‘7 2000 a a
<1: .:
w ‘2»
Es! 150~
c 2.
w e
5 z 100-
SE c»
:0: 1
g 50-
D
O I
0-7 7-20
Depth (cm)

Figure 4.8. Changes in denitriﬁer enzyme activity in August, 2004
after three tillage events. Treatment means are shown :h standard error
(n = 4); Bars with different letters are statistically different (P < 0.05).

112

 

REFERENCES

CAST, 2004. Emissions and mitigation of agricultural greenhouse gases. in Climate
Change and Greenhouse Gas Mitigation: Challenges and Opportunities for Agriculture.
Council for Agricultural Science and Technology (CAST), Ames, Iowa, USA.

CTIC, 2004. 2004 National Crop Residue Management Survey. CTIC, West Lafayette,
IN.

Alvarez, R., and C. R. Alvarez. 2000. Soil organic matter pools and their associations

with carbon mineralization kinetics. Soil Science Society of America Journal
64:1 84-189.

Angers, D. A., and M. Giroux. 1996. Recently deposited organic matter in soil water-
stable aggregates. Soil Science Society of America Journal 60: 1647-1655 1.

Bethlenfalvay, G. J ., and J. M. Barea. 1994. Mycorrhizae in sustainable agriculture. 1.
effects on seed yield and soil aggregation. American Journal of Alternative
Agriculture 92157-161.

Bowman, R. A., J. D. Reeder, and R. W. Lober. 1990. Changes in soil properties after in
a central plains rangeland soil after 3, 20, and 60 years of cultivation. Soil Science
150:851-857.

Brown, S., and A. E. Lugo. 1990. Effects of forest clearing and succession on the carbon
and nitrogen content of soils in Puerto Rico and US Virgin Islands. Plant and Soil
124:53-64.

Buckley, D. H., and T. M. Schmidt. 2001. The structure of microbial communities in soil
and the lasting impact of cultivation. Microbial Ecology 42:11-21.

Burgess, M. S., G. R. Mehuys, and C. A. Madramootoo. 2002. Decomposition of grain-
corn residues (Zea mays L.): A litterbag study under three tillage systems.
Canadian Journal of Soil Science 82: 127-138.

Buyanovsky, G. A., J. R. Brown, and G. H. Wagner. 1997. Sanbom ﬁeld: effect of 100
years of cropping on soil parameters. Pages 205-226 in E. A. Paul, K. Paustian, E.
T. Elliott, and C. V. Cole, editors. Soil organic matter in temperate
agroecosystems. CRC Press, New York.

Calderon, F., L. Jackson, K. Scow, and D. Rolston. 2000. Microbial responses to
simulated tillage in cultivated and uncultivated soils. Soil Biology &
Biochemistry: 32:1547-1559.

Calderon, F., L. Jackson, K. Scow, and D. Rolston. 2001. Short-term dynamics of

nitrogen, microbial activity, and phospholipid fatty acids after tillage. Soil
Science Society of America Journal 65:118-126.

113

Cambardella, C. A., and E. T. Elliott. 1992. Particulate soil organic-matter changes across
a grassland cultivation sequence. Soil Science Society of America Journal: 777-783.

Cambardella, C. A., and E. T. Elliott. 1994. Carbon and nitrogen dynamics of soil

organic matter fractions from cultivated grassland soils. Soil Science Society of
America Journal 58:123-130.

Cavigelli, M., and G. Robertson. 2000. The functional signiﬁcance of denitriﬁer
community composition in a terrestrial ecosystem. Ecology 81 :1402-1414.

Compton, J. E., and R. D. Boone. 2000. Long-term impacts of agriculture on soil carbon
and nitrogen in New England forests. Ecology 81:2314-2330.

Davidson, E. A., and I. L. Ackerman. 1993. Changes in soil carbon inventories following
cultivation of previously untilled soil. Biogeochemistry 20:161-193.

DeGryze, S., J. Six, K. Paustian, S. J. Morris, E. A. Paul, and R. Merckx. 2004. Soil
organic carbon pool changes following land-use conversions. Global Change
Biology 10:1120-1132.

Del Gado, I., J. Six, A. Peressotti, and M. F. Cotrufo. 2003. Assessing the impact of land-
use change on soil C sequestration in agricultural soils by means of organic matter
fractionation and stable isotopes. Global Change Biology 9:1204-1213.

Denef, K., J. Six, H. Bossuyt, S. D. Frey, E. T. Elliott, R. Merckx, and K. Paustian. 2001.
Inﬂuence of dry-wet cycles on the interrelationship between aggregate, particulate

organic matter, and microbial community dynamics. Soil Biology & Biochemistry
33:1599-1611.

Detwiler, R. P. 1986. Land use change and the global carbon cycle: the role of tropical
soils. Biogeochemistry 2:67-93.

Gale, W., C. Cambardella, and T. Bailey. 2000. Surface residue- and root-derived carbon

in stable and unstable aggregates. Soil Science Society of America Journal
64:196-201.

Gale, w. J., C. A. Cambardella, and T. B. Bailey. 2000. Root-derived carbon and the
formation and stabilization of aggregates. Soil Science Society of America
Journal 64:201-207.

Golchin, A., J. M. Oades, J. O. Skjemstad, and P. Clarke. 1994. Study of free and
occluded particulate organic-matter in soils by solid-state C-1 3 CP/MAS NMR-

spectroscopy and seaming electron-microscopy. Australian Journal of Soil
Research 32:285-309.

Groffman, P. M., E. A. Holland, D. D. Myrold, G. P. Robertson, and X. Zou. 1999.
Denitriﬁcation. Pages 272-290 in G. P. Robertson, D. C. Coleman, C. S. Bledsoe,

114

and P. Sollins, editors. Standard Soil Methods for Long-Tenn Ecological
Research. Oxford University Press, New York.

Gupta, V. V. S. R., and J. J. Germida. 1988. Distribution of microbial biomass and its
activity in different soil aggregate size classes as affected by cultivation. Soil
Biology & Biochemistry 20:777-786.

Hart, S. C., J. M. Stark, E. A. Davidson, and M. K. Firestone. 1994. Nitrogen
mineralization, immobilization, and nitriﬁcation. Pages 985-1018 in R. W.
Weaver, J. S. Angle, P. J. Bottomley, D. F. Bezdicek, M. S. Smith, M. A.
Tabatabai, and A. G. Wollum, editors. Methods of Soil Analysis, Part 2-
Microbiological and Biochemical Properties. Soil Science Society of America,
Madison, Wisconsin, USA.

Houghton, R. A., R. D. Boone, J. M. Melillo, C. A. Palm, G. M. Woodwell, N. Myers, B.
Moore, and D. L. Skole. 1985. Net flux of carbon dioxide from tropical forests in
1980. Nature 316:617-620.

Jansa, J ., A. Mozafar, G. Kuhn, T. Anken, R. Ruh, I. R. Sanders, and E. Frossard. 2003.
Soil tillage affects the community structure of mycorrhizal fungi in maize roots.
Ecological Applications 13:1 164-1 176.

Janzen, H. H., C. A. Campbell, S. A. Brandt, G. P. Lafond, and L. Townley-Smith. 1992.
Light fraction organic matter in soils from long term crop rotations. Soil Science
Society of America Journal 56:1799-1806.

Jastrow, J. D., T. W. Boutton, and R. M. Miller. 1996. Carbon dynamics of aggregate-

associated organic matter estimated by carbon-13 natural abundance. Soil Science
Society of America Journal 602801-807.

Kemper, W. D., and R. C. Rosenau. 1986. Aggregate stability and size distribution. Pages
377-382 in A. Klute, editor. Methods of Soil Analysis 1. Physical and
Mineralogical Methods Second Edition. American Society of Agronomy,
Madison, Wisconsin, USA.

Kettler, T. A., D. J. Lyon, J. W. Doran, W. L. Powers, and W. W. Stroup. 2000. Soil
quality assessment after weed-control tillage in a no-till wheat-fallow cropping
system. Soil Science Society of America Journal 64:339-346.

Kooistra, M. J ., and M. Van Noordwijk. 1996. Soil architecture and the distribution of
organic matter. in M. R. Carter and B. A. Stuart, editors. Structure and organic
matter storage in agricultural soils. CRC Press, New York.

Lal, R. 2003. Global potential of soil carbon sequestration to mitigate the greenhouse
effect. Critical Reviews in Plant Sciences 22:151-184.

115

Lal, R. 2004. Soil carbon sequestration impacts on global climate change and food
security. Science 304:1623-1627.

Luo, Y. Q., R. E. White, P. R. Ball, and R. W. Tillman. 1996. Measuring denitriﬁcation
activity in soils under pasture: optimizing conditions for the short-term

denitriﬁcation enzyme assay and effects of soil storage on denitriﬁcation activity.
Soil Biology & Biochemistry: 409-417.

Lupwayi, N. Z., G. W. Clayton, J. T. O'Donovan, K. N. Harker, T. K. Turkington, and W.
A. Rice. 2004. Decomposition of crop residues under conventional and zero
tillage. Canadian Journal of Soil Science 84:403-410.

Martens, D. A. 2001. Nitrogen cycling under different soil management systems.
Advances in Agronomy 70: 143-192.

Mikha, M. M., and C. W. Rice. 2004. Tillage and manure effects on soil and aggregate-
associated carbon and nitrogen. Soil Science Society of America Journal 68:809-
816.

Miller, A., R. Amundson, I. Burke, and C. Yonker. 2004. The effect of climate and
cultivation on soil organic C and N. Biogeochemistry 67:57-72.

Niehues, B. J ., R. E. Lamond, C. B. Godsey, and C. J. Olsen. 2004. Starter nitrogen
fertilizer management for continuous no-till corn production. Agronomy Journal
96:1412-1418.

Pacala, S., and R. Socolow. 2004. Stabilization wedges: solving the climate problem for
the next 50 years with current technologies. Science 305:968-972.

Paustian, K., J. Six, E. T. Elliott, and H. W. Hunt. 2000. Management options for
reducing COz emissions from agricultural soils. Biogeochemistry 48:147-163.

Perfect, E., B. D. Kay, W. P. K. van Loon, R. W. Sheard, and T. Pojasok. 1990. Rates of
change in soil structural stability under forages and corn. Soil Science Society of
America Journal 54:179-186.

Pierce, F. J ., M.-C. Fortin, and M. J. Staton. 1994. Periodic plowing effects on soil

properties in a no-till farming system. Soil Science Society of America Journal
58:1782-1787.

Reid, J. B., and M. J. Gross. 1981. Effect of living roots of different plant species on the
aggregate stability of two arable soils. Journal of Soil Science 32:521-541.

Rice, C. W., M. S. Smith, and R. L. Blevins. 1986. Soil nitrogen availability after long-
terrn continuous no-tillage and conventional tillage corn production. Soil Science
Society of America Journal 50: 1206-1210.

116

 

Robertson, G. P., M. A. Huston, F. C. Evans, and J. M. Tiedje. 1988. Spatial variability in
a successional plant community: patterns of nitrogen availability. Ecology 69: 1 517-
1524.

Robertson, G. P., K. M. Klingensmith, M. J. Klug, E. A. Paul, J. R. Crum, and B. G.
Ellis. 1997. Soil resources, microbial activity, and primary production across an
agricultural ecosystem. Ecological Applications 7 : 158-170.

Robertson, G. P., E. A. Paul, and R. R. Harwood. 2000. Greenhouse gases in intensive
agriculture: Contributions of individual gases to the radiative forcing of the
atmosphere. Science 289:1922-1925.

Saxton, A. M. 1998. A macro for converting mean separation output to letter groupings in
Proc Mixed. Pages 1243-1246 in Proc. 23rd SAS Users Group Int]. Sas Institute,
Nashville, TN.

Six, J ., E. T. Elliott, and K. Paustian. 1999. Aggregate and soil organic matter dynamics
under conventional and no-tillage systems. Soil Science Society of America
Journal 63:1350-1358.

Six, J ., E. T. Elliott, K. Paustian, and J. W. Doran. 1998. Aggregation and soil organic
matter accumulation in cultivated and native grassland soils. Soil Science Society
of America Journal 62:1367-1377.

Six, J ., S. M. Ogle, F. J. Breidt, R. T. Conant, A. R. Mosier, and K. Paustian. 2004. The
potential to mitigate global warming with no-tillage management is only realized
when practised in the long term. Global Change Biology 10:155-160.

Six, J ., K. Paustian, E. T. Elliott, and C. Combrink. 2000. Soil structure and organic
matter: 1. Distribution of aggregate-size classes and aggregate-associated carbon.
Soil Science Society of America Journal 64:681-689.

Smith, K. A., and F. Conen. 2004. Impacts of land management on ﬂuxes of trace
greenhouse gases. Soil Use And Management 20:255-263.

Smith, P., K. W. Goulding, K. A. Smith, D. S. Powlson, J. U. Smith, P. F alloon, and K.
Coleman. 2001. Enhancing the carbon sink in European agricultural soils:
including trace gas ﬂuxes in estimates of carbon mitigation potential. Nutrient
Cycling in Agroecosystems 60:237-252.

Steenwerth, K. L., L. E. Jackson, F. J. Calderon, M. R. Stromberg, and K. M. Scow.
2002. Soil microbial community composition and land use history in cultivated

and grassland ecosystems of coastal California. Soil Biology & Biochemistry
34:1599-1611.

Swanston, C. W., B. A. Caldwell, P. S. Homann, L. Ganio, and P. Sollins. 2002. Carbon

dynamics during a long-term incubation of separate and recombined density
fractions from seven forest soils. Soil Biology & Biochemistry 34:1121-1130.

117

Tiessen, H., and J. W. B. Stewart. 1983. Particle size fractions and their use in studies of
soil organic matter: 11. cultivation effects on organic matter composition in size
fractions. Soil Science Society of America Journal 47:509-514.

VandenBygaart, A. J ., and B. Kay. 2004. Persistence of soil organic carbon aﬁer plowing
a long-term no-till ﬁeld in southern Ontario, Canada. Soil Science Society of
America Journal 68: 13941402.

Wander, M. M., and M. G. Bidart. 2000. Tillage practice inﬂuences on the physical

protection, bioavailability and composition of particulate organic matter. Biology
and Fertility of Soils 32:360-367.

West, T. O., and W. M. Post. 2002. Soil organic carbon sequestration rates by tillage and I
crop rotation: a global data analysis. Soil Science Society of America Journal ij
66:1930-1946.

‘E' f. -l~ _

118

CHAPTER 5
Plant-Mediated Effects of Initial Cultivation on Carbon Stabilization, Carbon

Dioxide, Nitrous Oxide, and Methane

ABSTRACT
Tillage can signiﬁcantly alter trace gas ﬂuxes and the stability of soil C and N. The
effects of soil disturbance, however, are usually confounded by fertilizer inputs, annual
cropping or prior tillage. We cultivated a previously uncultivated midsuccessional
community and left the site fallow and unfertilized to study cultivation effects separate
from other agricultural disturbances. We hypothesized that rapid changes in aggregation,
microbial activity and trace gas ﬂuxes would occur after a single plowing and that both
plant litter and growing plants will modify responses to tillage. On one set of cultivated
plots we removed aboveground biomass prior to cultivation and in another set of
cultivated plots we removed vegetation throughout the growing season to examine plant
litter and growing plant effects on trace gas ﬂuxes, aggregation and microbial processes
following tillage. Tillage with litter plowed under increased C02 ﬂuxes by 72% in year 2
and 61% in year 3 and 40 and 54% of these increases were due to moving aboveground C
belowground. N20 ﬂuxes doubled the ﬁrst year and increased seven-fold the second year
due to cultivation in plots with litter. These changes were related to increased nitrate
concentrations in the soil which reached levels commonly found in fertilized ﬁelds (10-
20 ug N g") and increased nitriﬁer and denitriﬁer enzyme activities. Plant litter had little
effect on N cycling but growing plants strongly modiﬁed soil NO3' concentrations and

N20 ﬂuxes. Global warming potential increases in cultivated plots over six month

119

periods in 2003 and 2004 represent 12 and 13 times more C, respectively, than is
annually gained (30 g C m'2 y") under no-till cropping in the Midwest US. Our results
demonstrate that tillage alone can immediately change aggregation, trace gas ﬂuxes and

microbial processes and thereby destabilize soil C and N stocks.

INTRODUCTION

Radiative forcing of the Earth’s atmosphere is increasing at unprecedented rates
due to human activities. Since 1750, atmospheric concentrations of C02, N20, and CH4
have increased by more than 30, 150, and 17%, respectively (IPCC, 2001). Agriculture
has contributed substantially to global ﬂuxes of these gases and now represents an
important mitigation option as carbon sequestration and greenhouse gas abatement in
soils are two of a limited number of rapidly-deployable, high impact CO; stabilization
options now available to policy makers (Kauppi et al., 2001; Caldeira et al., 2004).
Although management strategies such as cover-cropping and improved N fertilizer
management potentially reduce the global warming potential (GWP) of agricultural
ecosystems, eliminating tillage may have the broadest impact because of its potential to
alter soil-atmosphere exchanges of all three biogenic greenhouse gases.

Eliminating tillage by converting agricultural land to perennial crops or to
successional communities has the potential to sequester ca. 60 g soil C rn’2 y'l (CAST,
2004) and conversion to no-till annual crops has the potential to sequester C at about half
this rate (West and Post, 2002; Lal, 2003). Ultimately, however, net greenhouse gas
mitigation by soil C storage depends on the persistence of stored soil organic matter

(SOM) and an important challenge to persistent C and N is the potential reversibility of C

120

sequestration (Paustian et al., 2000; Lal, 2004; Pacala and Socolow, 2004). No-till soil

management in the US, although increasing, continues to be rotated with tillage because
of perceptions that no-till limits N availability and decreases yields (Martens, 2001).
Further, there is currently little incentive to maintain the C stocks in long-term no-till
soils or successional communities. Some models predict that soils with sequestered C
may rapidly lose C and N following cultivation (Baisden and Amundson, 2003) but
predictions remain contentious because data are not currently available to generalize for
temperate ecosystems on a time scale of less than 10 years following cultivation
(Davidson and Ackerman, 1993; West and Post, 2002; Miller et al., 2004).

The effects of tillage on N20 emissions are related to changes in soil structure,
moisture, porosity and other factors that inﬂuence C, N and Oz availability. Higher NzO
emissions in no-till than in conventionally tilled cropping systems have been frequently
reported (MacKenzie et al., 1997; Ball et al., 1999; Baggs et al., 2003) although some
studies have found lower emissions in no-till soils or no difference between tillage
systems (Robertson et al., 2000; Elmi et al., 2003), demonstrating the potential for
responses to vary across cropping systems, soil types, and the time since no-till
implementation. In an assessment of potential European land-use changes on non-C02
greenhouse gases, Smith et a1. (2001) calculated that converting all possible arable land
to no-till could result in an increase in N20 emissions equivalent to 20.5 Tg C02-C y".
In a recent review Six et al. (2004) concluded that differences in N20 ﬂuxes between till
and no-till change over time. Grandy et al (2005, in review) reported that no-till had no

effects on total N20 emissions from com-soybean-wheat cropping systems over 12 yr.

121

Tillage has been reported to decrease methane oxidation in some agricultural and
natural soils (Hiitch et al., 1994; Cochran et al., 1997; Mosier et al., 1997) but in others
increase it (Kruse and Iversen, 1995) or have no effect (Mosier et al., 1998; Burke et al.,
1999). Suwanwaree and Robertson (2005) found that tillage per se — separate from
fertilization effects, a confounding factor in earlier studies —— had no effect on CH4
oxidation in agricultural, forest, or successional ecosystems.

These studies indicate that despite recent efforts to quantify tillage effects on trace
gases, signiﬁcant uncertainties remain. Tillage may inﬂuence C02, CH4 and N20 ﬂuxes
but its net effect on greenhouse gas exchange is largely unknown. Further, tillage in
temperate ecosystems has been examined almost exclusively in soils with long histories
of agricultural management. The legacies of prior tillage, fertilization and annual
cropping effects on soil structure, microbial communities, aggregation, and other factors
that control trace gas ﬂuxes may last for more than a century, confounding the inﬂuence
of recent tillage. Because of this, our basic understanding of tillage effects on soils and,
in particular, our ability to predict future effects of tillage in long-term no-till soils is
limited.

In this study our overall objective was to determine the effects of tillage on trace
gas ﬂuxes and their underlying physical and microbial controls, and the extent to which
plants modify tillage effects. Our speciﬁc objectives were to: 1) determine tillage effects
on C02, N20 and CH4 emissions in a previously uncultivated mid-successional soil to
determine the effects of cultivation on GWP independently of other factors; 2) quantify
the persistence of aggregation and aggregate-associated C immediately following tillage;

and 3) examine disturbance effects on soil N cycling and microbial processes that control

122

C and N turnover. By cultivating a never-previously cultivated ﬁeld and minimizing

plant community changes we were able to look at soil disturbance free from the
inﬂuence of other agricultural management practices, something that, to best of our
knowledge, has never been done before. We hypothesized that rapid changes in
aggregation, microbial activity and trace gas ﬂuxes will occur after a single plowing and

that both plant litter and growing plants will modify responses to tillage.

MATERIALS AND METHODS
Site Description and Experimental Design

Our experimental site was a previously never-tilled ﬁeld at the WK. Kellogg
Biological Station (KBS) in southwest Michigan, USA (42° 24' latitude, 85° 24'
longitude). The soils, plant communities, and management at this site have been
described previously (Grandy, 2005) and the six dominant plant species are listed in
Table 5.1. Brieﬂy, soils at the site are Kalamazoo (ﬁne-loamy) and Oshtemo (coarse-
loamy) mixed, mesic, Typic Hapludalfs developed on glacial outwash (Crum and Collins,
1995). Prior to initial cultivation, soils contained 31.8 (d: 1.4) g C kg'1 and 2.59 (i 0.11)
g N kg", and soil pH averaged 5.93 (d: 0.11); the same soil C levels had been measured 3
years and 12 years prior to this study and are statistically indistinguishable from C levels
in nearby undisturbed forests on the same soil types (Robertson et al., 2000).

In 2002, four uncultivated 3 x 6 m control sites were established in this ﬁeld as part
of a related study (Grandy, in review). In 2003 we established in this ﬁeld eight 3 x 6 m
plots to which we assigned 4 replicates of two tillage treatments with differing

aboveground biomass inputs. In one treatment we removed all aboveground plant and

123

litter biomass prior to cultivation and in the other treatment all aboveground biomass

was mowed and left in place prior to cultivation. Vegetation was clipped at ground-
level. After raking as much biomass from the site as possible, remaining plant material
was blown off of the soil surface with a push-blower. Within each of the four plots in
which aboveground biomass was tilled into the soil, a single 1m x 1m bare-ground
mieroplot was established. These micr0plots were separated from the surrounding plots
by root-exclusion barriers inserted to 30 cm. Bare-ground micr0plots were established on
29 June 2003 and maintained by removing aboveground plant biomass weekly.
Microplots were removed prior to cultivation in 2004 and then re-established in the same
location.

Cultivation occurred in the same plots on 15 June 2003 and 20 June 2004. To study
soil disturbance effects separately from other factors associated with agricultural
conversion (e. g. the use of annual plant monocultures and fertilizers), we left the site
fallow following tillage to allow a diversity of annual and perennial plants to recolonize.
We anticipated that cultivation would alter dominant plant species to a degree but that
this change would be relatively small compared to the establishment of annual or

perennial monocultures.

Trace Gas Fluxes
Field protocols for determining trace gas emissions and laboratory analysis of C02
and N20 ﬂuxes have been described previously (Grandy, in review ). Gas ﬂuxes were
determined using a single 25 cm diameter static PVC chamber (5500 cm'3) located within

each plot (Livingston and Hutchinson, 1995). Prior to sampling, gas-tight lids with

124

_ If ghﬁ“

 

sampling ports were placed on chamber bases permanently installed to a soil depth of

2.5 cm and accumulated headspace was then sampled four times over 90 min by
removing 20 mL of headspace gas to 12 mL vials (Labco Unlimited, Buckinghamshire,
UK). Gas sampling was generally performed between the hours of 0900 and 1400 and all
plots were sampled on each sampling day. Within 48 h of collection, CO; was analyzed
using an infrared gas absorption analyzer (IRGA); N20 analyzed using a gas
chromatograph outﬁtted with a 63 Ni electron capture detector (350 °C); and CH4 analyzed
using a gas chromatograph equipped with a ﬂame ionization detector (FID). Flux for
each chamber was calculated as the linear portion of the gas accumulation curve for that
chamber. Trace gas measurements were made 1 times in 2002 between 24 June and 22
August, ltimes in 2003 between 29 May and 13 October, and 19 times in 2004 between
15 April and 29 October. Gas ﬂuxes were measured 19 times in 2003 between 25 May
2003 and 13 October 2003 and 19 times in 2004 between 15 April 2004 and 29 October
2004. CH; measurements are not available for 2004.

Soil Inorganic N
NH4+ and NO3' were extracted with 1 M KCl from duplicate ﬁeld-moist 10 g soil

samples using a 1:5 soil/extractant ratio. Soil extracts were ﬁltered with a syringe ﬁlter
using a type A/E glass ﬁber ﬁlter (Pall Corporation, East Hills, NY). Filtrates were
stored in 7 mL scintillation vials and frozen until analysis for NFL,+ and N031 Both
analyses were performed on an Alpkem 3550 Flow Injection Analyzer (01 Analytical,
College Station, TX).

Enzyme Dynamics

125

 

We used a shaken slurry method (Hart et al., 1994) to test for nitriﬁer enzyme
activity to 0-7 and 7-20 cm soil depths on 27 May 2003 (pre-cultivation), 02 September
2003, 13 October 2003, 10 May 2004 and 17 August 2004. Brieﬂy, 30 g ﬁeld-moist soil
was combined with 100 mL solution containing non-limiting quantities of NH4+ in a 160
mL jar capped with paraﬁlm. The paraﬁlm was perforated with a needle to allow rapid
gas exchange while minimizing water loss. Jars were placed on a rotary shaker and
rotated at 200 rpm. Each ﬂask was sampled 4 times during a 28 h incubation period.

Soil extracts were centrifuged, ﬁltered, and frozen until analysis for NO3'.

Denitriﬁer enzyme activity was measured at 0-7 and 7-20 cm soil depths on soil
samples collected on 16 July 2003, 18 August 2003, 21 October 2003 17 and 17 August
2004 using methods described in Groffrnan et al. (1999). We. combined 10 g ﬁeld-moist
soil with 10 mL solution consisting of NaC4H4O4 (2 mM) and KNO3‘ (1 mM) in 160 mL
serum vials. We capped these vials with rubber septa and ﬂushed them with N2 ﬁve
times over 10 min to create anaerobic conditions prior to adding acetylene (15%).
Acetylene blocks the conversion of N20 to N2 so that NzO measurements alone can be
used to estimate total denitriﬁcation potential. High microsite substrate availability was
maintained in the jars by keeping them on a rotary shaker (190 rpm) during the
incubation. N20 concentrations were determined on four 0.5 mL headspace samples
collected every 12 min. We completed the entire analysis within 60 min of the substrate
additions in order to reduce the potential for cell growth or reproduction during the
incubation to alter N20 production (Luo et al., 1996). N20 was analyzed using a gas

chromato graph outﬁtted with a 63 Ni electron capture detector (350 °C).

126

Readily mineralizable C was measured in 2003 on May 27 (pre-till), July 16,
September 2 and October 13. A subsample of soil (15-20 g) was transferred into 60 mL
glass serum vials with a 13 mm diameter opening. We estimated the bulk density of each

of our samples after tamping down the serum vials 10 times on a laboratory bench and
from this determined the amount of water needed to bring them to 55% WFPS. After
wetting the samples, the serum vials were placed in a ‘/2 pint jar with ca. 60 mL of water
in the bottom. These jars were covered with polyﬁlm that permits relatively free 02 and
C02 exchange but retains water. Jars were put into boxes and then into dark incubation
chambers maintained at 25 °C for 7d. After 7 (1, serum vials were flushed for 45 s with a
humidiﬁed air stream. After ﬂushing, bottles were sequentially capped with a rubber
septum. A 0.5 mL sample of headspace was immediately drawn with a syringe and then
two additional samples were taken over a 90 min sampling interval. CO; content of each
gas sample was analyzed using an infrared gas absorption (IRGA) analyzer, followed by

calculation of the respiration potential for the time interval (Robertson et a1. 1999).

Aggregation
Soil aggregation was determined on samples collected 27 May 2003, 16 July 2003,
and 21 October 2003. Aggregate distribution was determined by hand on triplicate 35 g
air-dried soil samples by wet-sieving in water through a series of 2000 pm, 250 pm, and
53 um sieves (Six et al. 1998). The method we used to separate inter- and intra-
aggregate light fraction (LF) is based on previously published protocols (Six et al., 1998;
Gale et al., 2000). Aggregate subsamples (8 g) were pre-wetted with 4 mL H20 prior to

LF analysis to minimize aggregate slaking during LF separation. Aggregates were

127

transferred from the membrane ﬁlters to 100 mL beakers after 16 h with 5 mL aliquots

of NAPT at a density of 1.62 g cm'3. A total of 55 mL NAPT was used for each sample.

A preliminary test showed that the ﬁnal density of the sodium polytungstate was 1.60 g
cm'3 following equilibration with the water contained in aggregates.

After 24 h, inter-aggregate LF was aspirated from the surface of the sodium
polytungstate and then rinsed with at least 600 mL DI H20. Aggregates were then
dispersed to release the intra-aggregate LF using sodium hexametaphosphate and
resuspended in NAPT (d = 1.62 g cm'3). The intra-aggregate LF was collected form the

surface.

Data Analysis

Soil aggregate size distributions and aggregate-associated C pools were corrected
for sand content > 53 um. Carbon content of soil organic matter fractions was calculated
on an area basis by correcting for bulk density. Tillage effects on aggregation and
aggregate-associated SOM pools were analyzed by Proc Mixed (Version 8.2, SAS
Institute, 1999) using a randomized complete block design analysis of variance
(ANOVA). Treatment, soil depth and aggregate size fraction were considered ﬁxed
effects and block a random effect. Treatment effects on bulk density, total soil C,
denitriﬁer and nitriﬁer enzyme activities and mineralizeable C were analyzed similarly
but without the size class effect. Soil inorganic N concentrations were analyzed by Proc
Mixed using a randomized complete block design analysis of variance (ANOVA) with
repeated measures (SAS Version 8.2, SAS Institute, 1999). Treatment and sampling date

were considered ﬁxed effects and block a random effect.

128

_-.....‘

 

We interpolated daily gas ﬂuxes from our periodic measurements to estimate total

C02 and N20 ﬂuxes over the measurement period. Individual trace gas contributions to
global warming potential were made using 100-yr time horizon factors of 310 for N20
and 21 for CH4 and calculated, conservatively, for a six month period, which corresponds
approximately with the length of our growing season and sampling campaign. The
difference in global warming potential was calculated as the sum of the difference in
N20, C02 and CH4 between tilled and control plots with CH4 oxidation considered a
negative offset of CO; and N20 emissions. Gas ﬂuxes and differences in GWP were
analyzed similarly as other variables except treatment was the only ﬁxed effect. Analysis
of covariance (Goldberg and Scheiner, 2001) was used to examine the relationship
between soil temperature, moisture and C02 ﬂuxes in 2003 and 2004.

To determine gas ﬂuxes in bare-ground plots in 2003, prior to the regeneration of
vegetation and establishment of the micr0plots, we used emission rates corresponding to
the same replicate of the tilled plots with litter to create a data set that covered the entire
sampling period. These emissions should have been identical to those from the bare-
ground microplots. This data set was used to determine mean ﬂuxes for the sampling
period. In 2004, measurements from the bare- ground micr0plots spanned the entire
sampling period and these were used for interpolation.

Single degree of freedom comparisons were made to using the LSD statistic to
calculate a 95% conﬁdence interval around the differences between means generated
using the diff option in Proc Mixed. The LSD was carried out using the PDMIX800
algorithm (Saxton, 1998). Data were log transformed where necessary to improve

homogeneity of variance

129

.0 _'~ Anna
.4

 

Plant and Soil Analysis
Cultivation effects on plant litter were determined on 15 September 2003 from two
625 cm'2 quadrats in each plot. Cultivation effects on plant species, biomass and litter
were determined on 26 May 2004 prior to the second cultivation in two 1250 cm"2
micr0plots. Organic C and total N concentrations of plant and soil samples were
determined by combustion and gas chromatography in a CHNS analyzer (Costech ECS

4010, Costech Analytical Technologies, Valencia CA.).

RESULTS

Plant and Litter Dynamics

Changes in the plant communities following cultivation were mainly due to the
reduction in Rubus accidentalis (Table 5.1). Although some regenerated following
tillage, it was no longer one of the top six species. Even in the control plots, however,
Rubus accidentalis was patchily distributed as indicated by the high standard error.
There was none in one of the replicates and only 8 g m"2 in another while there was 830 g
m'2 in another replicate. Bromus inermis, Agropyron repens, and Poa Pratensis L. were
among the top six biomass producers in all of the treatments.

Prior to the second cultivation in Spring 2004, we estimated that there was 192 g C
In2 in aboveground plant pools and 58 g C rn‘2 in litter pools of plots where vegetation
was removed; in plots where aboveground biomass is incorporated with tillage, we
estimated that C in the plant and litter pools was 260 and 43 g C m'z, respectively (Table

5.2). These estimates are similar to estimates made in 2002 prior to the initiation of this

130

. 5::

;L

 

experiment when there was 228 d: 11 g C m'2 in plant pools and 142 :1: 30 g C m'2 in

litter pools.

Total Soil C and N

Cultivation redistributed C and N from the soil surface to 7-20 cm (Table 5.3). C
concentrations in cultivated plots were reduced by 23-34% at 0-7 cm and increased by
52-54% at 7-20 cm. Control plots had 23-30% higher N concentration at the 0-7 cm
depth than any of the cultivated treatments; at 7-20 cm cultivated plots had a 33-40%
higher N concentrations. There were no differences in overall soil C and N
concentration at 0-20 cm depth. Expressed on an areal basis, the biomass removal
treatment and bare- ground micr0plots had more C than control plots to 20 em but this

was at least partially due to higher bulk densities in these plots than control plots.

Aggregation and Aggregate-Associated C

Cultivation immediately changed soil aggregate distributions (Figure 5.1). On 16
May 2003, 31 days after tillage, the proportion of soil in the 2000-8000 um aggregate
size class at 0-7 cm soil depth was reduced from 0.35 g g’1 in control plots to 0.20 in no
litter plots and 0.16 g g'1 in litter plots. These trends were still evident in September 2003
when cultivation reduced soil in the 2000-8000 um aggregate size from 0.48 in control to
0.21 in no litter and 0.18 g g‘1 in litter plots. Cultivation increased soil in the
microaggregate size classes (<250 pm). There were also reductions at the 7-20 cm soil
depth, although they were proportionally less than those near the soil surface (Figure 5.1).

At the 0—20 cm soil depth of cultivated plots there was a lower proportion of soil in the

131

 

2000-8000 um size class on 16 July 2003 and 21 October 2003. On 24 September 2004

cultivated plots with litter and bare- ground micr0plots had lower soil in the 2000-8000

 

um size class than control plots. There were generally no differences among no litter,
litter and bare-ground plots except at 0-7 cm where bare—ground plots had a lower
proportion of soil in the 250-2000 um size class than vegetated plots with litter
incorporated at tillage.

Cultivation generally homogenized the distribution of aggregate-associated C at 0-7 hr
and 7-20 cm soil depth (Figure 5.2). There were no differences among depths for any of
the cultivated treatments while control sites generally had more inter, intra and heavy
fraction C at 0-7 cm. Cultivation reduced inter-aggregate C associated with 2000-8000
um aggregates at 0-7 cm in no litter and bare-ground plots. Bare-ground plots also
reduced inter-aggregate C associated with the 53-250 um size class relative to control
plots. At 7-20 cm, cultivated plots with litter increased C in the inter- and intra-aggregate

pools of the 250-2000 urn size class (Figure 5.2).

Trace Gases and GWP
No litter and litter plots had higher C02 ﬂuxes after mowing but before cultivation
on 20 May 2003 (Figure 5.3). Cultivation immediately increased C02 emissions in litter
and no litter plots and these emissions remained higher throughout 2003 (Table 5.4;
Figure 5.3). Bare-ground micr0plots established after the recovery of vegetation
generally had similar CO; ﬂuxes to the plots with biomass incorporated at tillage. Tillage
effects were still evident prior to cultivation in 2004 when cultivation again accelerated

C02 ﬂuxes from all plots. Mean C02 ﬂuxes in 2003 were 33% greater in plots with no

132

 

litter, 72% greater in plots with litter and 61% greater in bare-ground micr0plots (Table

5.4). In 2004, no litter increased CO; emissions by 36% and litter by 61% relative to
control plots (Table 5.4). N;O ﬂuxes were also accelerated by tillage and removal of
plants throughout the growing season (Figure 5.4). In 2003, no-litter and litter doubled
N;O emissions while bare-ground increased N;O emissions four-fold (Table 5.4). In
2004 cultivation of litter and no-litter treatments increased N;O emissions seven-fold and
in bare-ground plots 12-fold. CH4 oxidation was reduced by tillage and plant
manipulations (Figure 5.5; Table 5.4), although reductions in methane oxidation in bare-
ground and no-litter plots were relatively small compared to cultivation effects on C0;
and N20 emissions. CO; contributions to GWP were far greater than those from CH4 or

N;O.

Soil Temperature and Moisture

Soil moisture and temperature (Figure 5.6) were inﬂuenced by tillage. In 2003,
mean soil temperature in cultivated treatments ranged from 18.2 in bare- ground plots to
19.3 °C in litter plots. All cultivated plots had a higher (p<0.05) temperature than control
plots (16.6 °C). In 2004, bare-ground plots had the highest temperature (19.5) but all of
the tilled treatments had a higher temperature than control (16.2). Analysis of covariance
results showed a signiﬁcant temperature effect in 2003 (p<0.003): CO; ﬂux = 14.9 +
0.026 (temperature), r2 = 0.22 (p<0.000) (Figure 5.7). In 2004 (Figure 5.7) there was a
temperature by treatment interaction (p<0.001) indicating the slope of the line describing
the relationship between temperature and CO; emissions differed between treatments.

Control plots: CO; ﬂux = 13.3 + 0.020 (temperature), r2 = 0.08 (p<0.000); no litter plots:

133

CO; ﬂux = 12.1 + 0.033 (temperature), r2 = 0.40 (p<0.000); litter plots: CO; ﬂux = 13.8
+ 0.018 (temperature), r2 = 0.22 (p<0.000); bare-ground plots: CO; ﬂux = 15.5 + 0.026
(temperature), r2 = 0.20 (p<0.000). In 2004, there was additionally a moisture effect

(p<0.02), although the relationship with CO; was weak (r2=0.02; data not shown).

Soil Inorganic N

Cultivation increased soil NO3’ concentrations at 0-7 and 7-20 cm soil depth (Figure
5.8). In litter plots, NO3' peaked four days after tillage (DOY 170), when it exceeded 20
pg g'1 and exceeded 10 pg g'1 on DOY 215. On three sampling dates in 2003, N03'
concentrations were greater in litter plots than no-litter plots. The highest NO3'
concentrations were in bare-ground micr0plots where concentrations exceeded 30 pg g'1
at 0-7 cm and 20 pg g'1 at 0-20 cm in 2003. In 2004 bare-ground micr0plots had the
highest NO3' at both soil depths on DOY 208 and 230. Ammonium concentrations
(Figure 5.9) responded strongly to tillage in 2003 at 0-7 em but tillage effects were

smaller at 7-20 cm and at both depth strata in 2004.

Microbial Processes
Potentially mineralizeable C was generally greater at 0-7 than 7-20 cm depth across
treatments, except for litter plots on 13 October 2003 (Figure 5.10). Control plots had
greater mineralizeable C at 0-7 cm on 16 July 2003 and again on 13 October 2003. Litter
plots increased mineralizeable C at 7-20 cm on 02 September 2003 and again on 13
October 2003 and increased mineralizeable C at 7-20 cm on 02 September 2003.

Cultivation decreased denitriﬁer enzyme activity at 0-7 cm and increased it at 7-20 cm.

134

 

 

 

Overall, the only signiﬁcant cultivation effect at 0-20 cm was between no litter and
control plots on 20 August 2004 (Figure 5.11). In contrast to denitriﬁer enzyme
activity, cultivation increased nitriﬁer enzyme activity at 0-7, 7-20 and overall at 0-20 cm

depths (Figure 5.12).

DISCUSSION

Soil Aggregation

Within 60 d of cultivation, soil in the 2000-8000 pm size class at 0-7 cm depth
(Figure 5.1) declined to levels indistinguishable from those in an adjacent agricultural
soil cultivated for >50 y (Grandy and Robertson, 2000, in review). There were also
declines at 7-20 cm in the 2000-8000 pm size class demonstrating that declines at the soil
surface were not solely due to a redistribution of aggregates to increased soil depths.
Multiple studies have demonstrated that long-term cultivation reduces soil aggregation
(e. g. J astrow etal., 1996; Six et al., 2000; Mikha and Rice, 2004); our results demonstrate
that tillage immediately destroys macroaggregates and that these declines in soil structure
persist. Plant residues and growing plants are known to modify soil aggregation in
multiple ways but we found few differences between litter, no litter and bare-ground

plots. The effects of vegetation on aggregation were small compared to tillage effects.

Global Warming Potential (GWP) Impact
We documented (Table 5.4) immediate changes in trace gas ﬂuxes after cultivation
that dramatically altered global warming potentials (GWP). Global warming potential

increases in cultivated plots with litter over a 6 month period in 2003 and 2004 represent

135

12 and 13 times more C, respectively, than is annually gained (30 g C In2 y’l) under no-

till cropping in the midwest (Robertson et al. 2000; Franzluebbers and Steiner, 2002).
These differences were primarily attributable to variation in CO; ﬂux but in 2004
differences in N;O ﬂux between tilled plots with litter and control plots were equivalent
to 35 g C In2 over 6 months (Table 5.4). Methane oxidation offset little of the GWP
contributions from C0; and N;O. Similar differences in GWP were found in bare-ground
micr0plots, although no litter plots had reduced GWP (though not signiﬁcant) relative to

litter plots. This difference was primarily due to reduced CO; ﬂux in control plots.

Carbon Dioxide

Based on differences between no litter and control plots relative to differences
between litter and control plots, between 40 and 54% of the soil surface CO; emissions in
litter plots were from the incorporation of aboveground biomass (Table 5.4). We
estimate that cultivation incorporated 302 g C m'2 of aboveground C in 2004 (Table 5.2).
Although we do not have aboveground C estimates from pre-tillage in 2003, sampling of
the control sites in spring 2002 provided an estimate of 370 g C m'z, which is our best
estimate of the 2003 C contribution with tillage. In 2003, the difference in CO;-C loss
between litter and no litter treatments was 188 g C m'z, or 51% of the aboveground
residue C. In 2004, the difference in CO;-C loss between litter and no litter treatments
was 147 g C m'z, or 49% of the aboveground residue C (Table 5.4). Rochette et a1 (1999)
found that 40% of maize residue C was oxidized in a single season. Burgess et al. (2002)
found somewhat higher residue decomposition rates so that after two years 89-98% of

buried corn residue was lost. Lupwayi et al. (2004) estimated that within 12 months

136

decomposition losses from canola residue were 50% in conventional tillage. Our
estimates of aboveground C contributions to CO; ﬂux are thus in line with prior

estimates.

Striking declines in soil aggregation (Figure 5.1), increases in inter-aggregate LF
(Figure 5.2), and greater mineralizeable C at 7-20 cm (Figure 5.10) indicate aboveground
biomass C principally entered low-density, inter-aggregate POM pools that are rapidly
oxidized following disturbance (Arrouays and Pelissier, 1994; Six et al., 1999; DeGryze
et al., 2004). Lupwayi et al. (2004) found that incorporation of wheat litter with
conventional tillage increased its decomposition rate by 48%; Burgess et al. (2002) found
that litter decomposition rates at 20 cm were 52 - 105% greater than those at the soil
surface. Increases in soil temperature may have also contributed to increased CO;

emissions.

Bare- ground micr0plots in 2003 give us an estimate of the plant contributions to
CO; ﬂux after tillage. Similar ﬂuxes in these plots and the tilled plots with litter indicate
that plant contributions to CO; ﬂux were very low during the season (Table 5.4). This
was expected since recovery of the plant community generally took 3-4 weeks after
cultivation, during which time heterotrophic respiration (mainly bacteria and ﬁmgi)
would have accounted for most of the difference in CO; ﬂux. In contrast, autotrophic
(vascular plant roots) respiration may have accounted for 50% or more of the measured
CO; in control plots (Hanson et al., 2000). Low mean CO; ﬂuxes from bare-ground plots
in 2004 may have been partially due to the absence of autotrophic respiration but may
have also been due to decreases in C inputs. By establishing the plots in the same

location in 2004 as in 2003 we minimized the above and below-ground C inputs.

137

Nitrous Oxide

The effects of tillage on soil surface N;O ﬂuxes have been primarily studied
following conversion of long-term, conventionally tilled cropping systems to no-till
(Grant et al., 2004; Six et al., 2004; Mackenzie et al., 1997). We found that tillage
without fertilizer inputs or annual cropping doubled N;O ﬂuxes in 2003 and increased
emissions seven-fold in 2004 (Table 5.4). Sizeable N;O-N losses following cultivation of
no-till ecosystems were also demonstrated by Pinto et al. (2004) who reported increases
over 5 d of between 1.8 and 23 mg N;O-N m'2 following the plowing of a perennial
pasture. Keller et al. (1993) found that converting forest to pasture increased N;O ﬂuxes
in the ﬁrst ten years. In both of these studies, authors attributed increased N;O emissions
to accelerated SOM turnover producing excess inorganic N that was subsequently used
by N;O -producing nitriﬁers and denitriﬁers.

Our results also point to the importance of N availability because the highest N;O
emissions (Figure 5.4) generally coincided with elevated NO3' concentrations (Figure
5.8). Low NO3' concentrations in the control soils suggest a tight coupling of plant N
uptake and the microbial processes of N-mineralization and nitriﬁcation. Accelerated
SOM turnover, increased nitriﬁer enzyme activity (Figure 5.12) and reduced plant uptake
following cultivation disrupted the synchrony between N availability and plant uptake.
As a result, NO3' concentrations reached levels equal to or exceeding those typically
found in fertilized agricultural soils (10-20 pg N g'l soil).

If this site were converted to an annual cropping system, N;O would contribute

proportionally more to GWP over time due to soil C loss and increases in N;O ﬂux. For

138

 

1: SF»; _. _ {q

example, Robertson et al. (2000) demonstrated that once an agricultural soil has reached

equilibrium N;O dominates contributions to GWP from soil surface gas emissions.
Even in no-till cropping systems that are accruing C, annual C gains can be nearly offset
by N;O emissions (Grandy et al. 2005). Also, additional changes in the synchrony
between nitrogen availability and plant demand, including the use of supplemental N
fertilizer or conversion to annual crops, would lead to additional N;O emissions (Mosier
et al., 1998; Mosier, 2001).

Removing aboveground biomass prior to cultivation had little effect on N;O
emissions or on soil NO3' or NH4+ concentrations. Bare-ground micr0plots, however,
greatly increased N;O ﬂuxes and soil NO3' concentrations in both years. Both of these
results demonstrate the importance of growing plants in buffering nitrogen pulses alter
tillage; i.e., greater C turnover in litter than no litter plots, as evidenced by increased CO;
ﬂuxes, and its potential inﬂuence on N mineralization-immobilization dynamics had little

effect on N cycling compared to the presence or absence of plants.

Methane
We detected small decreases in CH; oxidation with cultivation that were
statistically signiﬁcant in the no litter plots (Table 5.4). Other studies have also reported
decreases in methane oxidation following tillage. In a grassland soil, Mosier et al. (1997)
found that tillage reduced immediately CH4 consumption by 34% and that this effect
persisted for three years. Kessavalou (1998b) found that over three years CH4 uptake in
no-till systems averaged 7.4-8.1 g C ha"l (II and in plowed soils averaged 5.980 g C ha’l

d". In contrast, other studies have found tillage had no effects on CH4 uptake (Robertson

139

et al., 2000; Suwanwaree and Robertson, 2005) or even increased oxidation (Kessavalou

et al., 1998a). Differences in climate and soil texture may modify the direct effects of
tillage but the indirect effects on pH and N turnover and availability may also be
important (Bédard and Knowles, 1989; Hﬁtsch et al. 1994; Suwanwaree and Robertson,
2005). In our study, tillage, per se, may not be decreasing CH4 oxidation — despite
declines in soil aggregation -- but having an indirect effect by increasing soil N cycling.
Increases in N availability and turnover frequently decrease CH4 oxidation rates (Hutsch

et al. 1993; Hilger et al. 2000).

Tillage Management

Cultivating this previously uncultivated soil immediately changed physical and
microbial processes regulating trace gas ﬂuxes. Release of C from physical protection
within macroaggregates > 250 pm, coupled with movement of aboveground C pools into
unprotected belowground pools, greatly increased C substrate availability. Associated
with this was a series of changes in microbial activity, soil inorganic N concentration, and
trace gas ﬂuxes. Although we detected no difference in soil C concentrations after two
years and other studies have suggested that periodic cultivation may have little effect on
otherwise no-till soils (Pierce et al., 1994; VandenBygaart and Kay, 2004), our results
suggest that plowing once destabilizes C and N stocks. These same processes are likely
occurring in agricultural soils rotated between till and no-till but are undetectable because
SOM pools and aggregation are relatively low. Detecting C losses due to periodic

cultivating may thus require observing these systems over many years.

140

The effects of cultivation on N;O ﬂuxes and soil NO3' concentrations can be
strongly mediated by plants. Reducing the period of time following cultivation that

soils are bare by using catch crops, cover crops or other strategies will reduce N losses.

CONCLUSIONS

We found that cultivation reduced the proportion of soil in the 2000-8000 pm
aggregate size classes from 36% to 16% at 0-7 cm within 30 days. Declines in aggregate
stability were also detected at 7-20 cm. These declines persisted throughout the study.
Plant litter and growing plants had little effect on aggregation. Tillage with litter plowed
under increased CO; ﬂuxes by 72% in 2003 and 61% in 2004 and between 40 and 54%
of these increases were due to moving aboveground C beloWground. N;O ﬂuxes doubled
in 2002 and increased seven-fold in 2003 due to cultivation in plots with litter. These
changes were related to increased nitrate concentrations in the soil which achieved levels
commonly found in fertilized ﬁelds (10-20 pg N g'l) and increased nitriﬁer and
denitriﬁer enzyme activities. Plant litter had little effect on N cycling but growing plants
strongly modiﬁed soil NOg' concentrations and N;O ﬂuxes. Global warming potential
increases in cultivated plots with litter over a 6 month period in 2003 and 2004 represent
12 and 13 times more C, respectively, than is annually gained (30 g C m'2 y'l) under no-
till cropping in the US. Midwest. These differences were primarily attributable to
variation in CO; ﬂux but in 2004 differences in N;O ﬂux between tilled plots with litter
and control plots were equivalent to 35 g C rn'2 over 6 months. Methane oxidation offset

little of the GWP contributions from C0; and N;O. Our results demonstrate the

141

importance of maintaining long-term no-till soils and protecting them from future

disturbance.

 

 

142

 

Table 5.1. Tillage effects on top six dominant plant species by biomass and
proportion of total biomass. Samples were collected on 26 May 2004,
approximately one year after initial cultivation and before the second cultivation.
Means with standard errors in parentheses.

 

Treatment Plant Species Plant Biomass Proportion Total
Biomass
g m’2
Control Rubus accidentalis 246 (198) 0.19 (0.12)
Dactylis glomerata L. 83.3 (38.9) 0.10 (0.05)
Bromus inermis 71.4 (31.5) 0.11 (0.05)
Poa pratensis L. 63.2 (23.9) 0.09 (0.04)
Arrhenatherum elatius L. 48.8 (21.5) 0.06 (0.04)
Agropyron repens 25.8 (15.0) 0.02 (0.01)
No Litterl Agropyron repens 125 (58.3) 0.25 (0.09)
Dactylis glomerata L. 92.1 (57.3) 0.20 (0.14)
Bromus inermis 26.4 (8.30) 0.06 (0.02)
Aster sp? 18.8 (8.05) 0.05 (0.03)
Daucus carrota L. 11.4 (6.66) 0.03 (0.02)
Poa pratensis L. 9.6 (6.47) 0.02 (0.01)
Litter Agropyron repens 136 (53.7) 0.22 (0.07)
Bromus inermis 90.7 (44.2) 0.15 (0.07)
Dactylis glomerata L. 85.4 (34.4) 0.15 (0.06)
Poa pratensis L. 37.0 (21.6) 0.06 (0.04)
Barbarea vulgaria 23.6 (18.6) 0.04 (0.03)
Daucus carrota L. 21.1 (7.22) 0.04 (0.01)

 

fNo litter plots had all aboveground biomass removed prior to cultivation; litter
plots had all aboveground biomass left in place.

143

Table 5.2. Aboveground C and N in plant and litter pools alter initial
cultivation in 2003 and prior to the second cultivation in spring 2004. Means
with standard errors in parentheses:r

 

Date C Source Carbon Nitrogen
g m'2 g m'2
15 Sept. 2003 Litter Control 195 (40.1)A 5.79 (1.32)A
No Litter 20.84 (14.5)B 0.33 (O.23)B
Litter 17.8 (4.42)B 0.39 (0.10)B
26 May 2004 Plant Control 423 (110)A 13.7 (3.57)A
No Litter 192 (25.0)B 6.23 (0.81)B
Litter 259 (24.1)AB 8.39 (0.78)AB
Litter Control 192 (27.6)A 5.09 (0.73)A
No Litter 58.3 (7.14)B 1.55 (0.19)B
Litter 43.3 (2.63)B 1.15 (0.07)B

 

TMeans within a column, date and C source followed by different letters are
signiﬁcantly different (p<0.05).

144

0:0:0000 0:0 50:30 0 5:03 0:008 :00 20000 w:_3::00 :003000 0005:0000 0000005...
.8.on 00 5:00 80 omé 0:0 :0.: 30:00.00 $003005 0:0 :8030 0 5505 0:000: 0000:0000 30:00.00 .3 00323.0 00003: 0
.mo.oV0 00 320000 3000000000 0:0 00:00 :00 0:0 5:38 0 :00? 0:002 0008030— 300000 .3 00323.0 8002.,

 

£S.8 20
0.2008 000
£S.8 02
0S8 02.

£008 S0
£38 2.0
£S.8 00.0
£008 00.0

LS8 S:
L28 00.:
L028 80
58 S0
0-6 mo—

0

$00.8 0.0

0<S8 3.0
0.188 8.0

mCodv wmd

£S.8 SS
£88 000
£88 :00
£ S8 :00

LS8 :0
LS8 0:0
LS8 03
S8 :0
0-8 wx

Z

£S.8 a:
003.8 00.:
m<S8 00.:
9000.8 00.:

S8 81
5.8 00.:
S8 3.:
S8 00.:

£008 00.:
£S.8 2.:
£88 00.:
..£8.8 00.0

0.50 w

005: 0:30

5.8 3.:
$08 80
8:8 00.0
S8 00.:

£28 S:
£38 3.:
£008 3.:
£: :8 00.:

£: :8 3.0
£28 80
£28 0.0

.058 2.0

o\o

0

S8 :0
S8 03
S8 03
S8 2.0

£S.8 £0
£S.8 :0
£008 :0
£S.8 0:0

£S.8 0:0

£S.8 00.0

£S.8 8.0

.258 00.0
as

Z

0::0:w-0:0m
:80..—

:003 02
30:00

0::0:w-0:0m
:005

:00: 02
30:00

0::0:w-0:0m
:005

:005 02
30:00

::0 omé

::0 onus

EU \uro

 

réoom .00 m:.::% 000 E 003 0800 05 :0 :_0w0 0:0 meow 0:3 2 :0 003 500.030
3.55 . 00000 $09000 E0 3-0 0:0 Do 00 @050 0:00 0:0 0:0000:00:00 Z 0:0 0 :00 :0 0000.00 8.002000 000:5 .m.m 030,—.

145

.Amcdvnc 30:00.00 0:800:00 0:0 :000: 30:00.00 0 :3 003230 00% 0:0 :::030 0 0:00:03 0:000):F
00:03:03: 0:: :: b00000 0:00: 3:333 0:0 0:3: 0?: 00: 00 0:030:03: 0:00:w-0:00 :: 0::3: w::_:::00 :0300 0:03 0:00: voom :00

0:2: 30:00 80:0 300 0:0 002 .000 00 8000 :30 :03: :0 00:0:0005:
0000:0320: :: 0:000 0:00:00 0:03 :3000 0:0 0:00:20

 

0:008 :00 - £0.00: 000 $08 0000 - £008 00: 00:00:: :0: 00:20.20:
<38 00:: - :00: :0: L000: 0000 - £S.8 0.: £00: 000 :8:
02:00.0: 000 - £008 :0: L008 0000 - £0.08 0.00 m<S8 00: :30: oz
- - 00.8 0.00 L000: 00:0 - 0:08 00: 080:: 00: 30:80 SSO
:0: 8 00:: :88 00.? S8 0.00 :00: 0000 0:0,: 0.000808 0.00 $0.08 00: 05:20.20:
:80: 000: L008 000. 0:08 :00 0:00: 0000 0:00.: 0.00-3.5. :: 0.0: <00: 00: 00::
L008 000 L: 0.8 00.0- 00.0: 000 0.0:: 0000 0230.0: 0.00-060. :: 0.: <3: :0 8.: 00:: oz
- :38 000- 0.0.8 00. :: S0: 000: £3.00 0.00-0800: 00.0 0:00: 0:: 3080 0000
:3 0-0: 000 w :3 0-8 000 w 0% 0-0: 000 w 7: 0-:: w: 70 0:: w: :0: 0-:: wE
: a 3 00 .00 002 N00 0-..:0 2-00.2 0-000

 

030003000 0000 :30 0: 030000030 000 000: 0.

mDKDTm mac Emu—2

 

:.. 0.400 :00 :0 0:0 007: :00

o: m 00 0:0:000 300:0: 0:0: 3.2:: mi? 300000 0:030:w 0:0 030908 0:03:00 :00 00 00:30: 0:: 0:03 3800:0853 w::0:0:00::000

0:: 0 00 000:0: 0 :00 3:030:00 0:03 3.30: 30:03: m:::::03 30300 0: 030000030 00» 0000 00000005 .0oom 0:0 moom

:: ::0E0w0:0:: 3003000, 30:00.00 0:3 00:00 00 30:02:30 3.00:: 303030 00x00 £0 0:0 002 .000 08038030 :002 06 030.0.

146

.::0 om-n :0 0000: 30:0 30:0003 0:0 0:000:00 :0 000 0030000 30 0-0 :0 0:003 .0003 000 0:0

33:00:: 0:00 w:=:::00 0 3003 Amodvﬁ 3200000 0000000080 000 0:000: 0008030: 3800000 00 003230 000003
0000 33000:: 0:0 0::00 6:00 w:0:::00 0 3003 Amodvﬁ 3200000 000000080 0:0 0:000: 0000:0::0 30:00.00 :00
0030:00 000:0 0000 0:0 .0300 6:00 30:00:00 0 3033 0:003 300500; .0030 voom :000:0O 0: 0030003 0:03 03::
0:00:0m:0m 0:00:00 000000800 :00: 0030300 0:0 000003 0:03 03:: 0:00:m-0:0m 00003000 0: :00:: 00>0::0:
00002.: :0 :02: 0:20 :00:: oz .81 00v 05 a: 000-00 .81 0000-000 .51 0000-0000 000:: 0: 00: 00:00 0.00:0

00:0 :00 0:00 w:0:::00 :00::3 00w:0:.:0 0:0 000m .0::00 30 000000 00m0: 03: .0: om-o 0:0 .0090: :08: .0: 3-0 A000:
000: .0: 0.0 :0 voom 0:0 moom :: :000w0:mm0 :00 :0 0:20 00:03:33: 300038: 0 :0 0:00000 3002000 0m 03?:

00-30 modmm mo-_00 05-000

 

 

7: > 7: 0
g 0 0 0 m .
p... - , I oo o
.0 m. I 0:0 6
m am... 6
Di J: J
. e
00.. 1-
e
a”... . S
- - I mvo 5
000050006 0 0000- 0 Rm.
000: 0 [ 00.0
:00: 0z 0-0

.0080 l

147

I Control

[:1 No Litter

 

'D
C
3
O
L

'58
E5
-—|CD
am
<m
<m
O
“H
[x
.Q
l l
0 L0
“3. V.
O O

L.5 6 391953166»!

148

 

Oct-04

Sep-O3

Jul-O3

Jun-03

Figure 5.1. cont’d (7-20 cm soil depth)

 

' 098
0002

7, ~. “h -‘ 3 ‘_ ,1 ‘ v '
,'*--‘~-- x 12‘31“ ‘ -‘_ , ‘v :4 ~
. I Li" I“ ”I. ‘1‘ A',

2“

(1360 0
can (DD

2
< oczoawﬁm

5E4

5:3 02
6:50

     

 

.25 8-8 3:8 .3 2:5

I 00.0

l
Lo
".
o

I
o
(‘9.
o

I
no
‘1'.
o

 

o

f
o
‘9.
o

L.6 6 391969166»!

149

A 0;; cm Treatment

  

41 ' Control
a D No Litter
0 AB : - Litter
% 3‘ ’ ‘3 Bareground
a)
92 2-
c»
ow
<9 1a
5
E o-

 

 

 

 

lntra-aggregate C
9 kg'1 sand-free aggregate g kg'1 sand-free aggregate

 

 

 

OJ b O)
O 01 O
l I J

15‘

Heavy Fraction C
9 kg'1 sand-free aggregate

 

 

 

O
l

 

2000
250

Figure 5.2. Cultivation effects in a previously uncultivated ﬁeld on soil organic matter
distribution in 2004 at 0-7 and 7-20 cm depth. Values going left to right on the x-axis
represent aggregate size classes: 250-2000 um, 53-250 mm and <53 pm, respectively.
No-litter plots had all biomass removed prior to cultivation. Bare-ground plots were
weeded and contained root exclusion barriers. Treatment means within a depth and
size class followed by different uppercase letters are statistically different (p<0.05).
Within a depth and treatment, size classes followed by different lowercase letters are
statistically different (p<0.05). Within a treatment and size class, means at 0-7 cm
followed by an asterisk are different from those at 7-20 cm.

150

.Avncv £85

28.8% 08 mam .moﬁw aoumzzso 8865 $58.2 asthma :29:on 88 3595c was 3303 803 92a vasohwbbwm
own—E 53> 365885 $9.85 258w??? =m was 92m 835 eggs—so 9 SE 3382 $285 .3 we: Boa
8:: o Z .voom can moom 5 355 Nov 083m :8 no 22m BEE—3:: 392.65 a 5 38m? :ozgmxsu .m.m 2sz

 

 

 

 

:53 50> .6 >3 88$ 59 .6 >8
mum ovm new 02. mm? 09 mom mum mvm 9N mm? mm? mNP
_ _ _ _ _ L Ll . _ _ _ _ H _ _ p _ . _ L _ L o
m
a . -_.._\m ..n... ., M
7...... H _ w-.. _.\.....,// . ../../ / 53. ....» \ _ m 30
.../ .. .. .1 . .. .r - ..r / / .. A... . ... .. nu .
.2... a .../A. HH... m . / a. u x -1... H. MHMHK , , loom O
, WI .... W..— W / ..\..u.... w . 4». .... m
..H . .. H loom um
..H H $29 ucsemmamm ..
I 85. 3:03 .3: < ..oov
as: 85335:: 02 x
35. 26$ .228 .

E9599...

151

.QHE macho cavemen.

8m mam .moﬁv eowgmzso 8865 $58.3 .2053 56288 88 358:8 can @383 80>» 30E nanokwoam
.quE .23 368985 32:05 958w??? =m was $03 .554 dosgﬁso 9 Sta 3388 @285 EN was 82m
5:: oz .voom 98 Bow 5 $56 ONZ ouﬁbﬁ :8 co Bow 33323:: bmsogoa a E Somme :oumzzso 42m 035

 

 

 
  

 

 

«com 50> .6 >8 _ meow 59 .5 >8
mum ovw mom 0: mm? 02. mom mum EN EN mm? m2. mm?
—l . ﬁl _ p .... _ l.LIlll_II lo
. Ill 3...,/./. ....... . .
I I cm
I ... .- I oo 5N
0

I ..8 N
999 uczoamocmm ... w?
65. 8:03 at: < u.

I Ace: 3:33 5:: 02 x I om? ...
as. 2.93 .928 . H

I EmEﬁmc. I om?

 

 

152

.Awu5 28.8

58:88 08 5.3 .08: 883:8 2: 88065 388 on... 85:82 :59:on 88 85888 8:8 8803 895 $2: :::o:w-o8m
88:: :53 888985 @8505 882888 :8 8: 32: 8:5 558358 9 :55 8852 88505 :8 8: £2: 8:: 0 Z
.voom :8 03288 8: 08 38583802 .83 5 :58:me £0 :0 Bow 88558:: £95305 8 5 Beebe 885550 .m.m 2:me

88 86> 6 8a

mom mum mvm mFN mm? mm? mm?
_ _ _ _ — _ H _ H _ H r O

 

 

III Mm
............... I . ..l. M . K M .. W ON
.m . . IkW / I -II:
x / H... Li. ' am. 3
F Ial ... :0
....m. wm‘ .twg. H
.. d.— a. .. W v
' . x ‘ _ . II 0v 0
...... I z w. X
4,...“ I _ Z mu.
LI I. m.
I oo u

 

low

153

dens 88:8 5:85:88 88 m8m .88: 858558 88585 830...}: 88:85 88288 88 85888 5:8 883»
883 «.88 8:88-855 88:: 553 888885 888555 55888858 :8 85 888 .855 .:o_8>_58 o: .88 885...: 88:83 :8
85 888 85: oZ .voom 5:8 moon 5 2:885 8:8 8.58898: :8 :o .52.: 88558:: 582588 8 5 38.5.8 8585550 .86 28E

 

 

 

 

38 86> 6 :8 88 86> 6 :8

8:8 8:8 888 888 8:: 88: 88: 888 888 888 8:8 88: 88: mm:
_ . _ . _ . _ L H . _ . _ _ P _ Ll _ o S
m.
I rm mm.
Tor w
I8: m
B
Iom m.
I88 N
58 b

_ _ o

 

 

 

Io_.

 

c'.
N
(%)91n13!0W IIos

 

 

999 88898 I on
H 85.886368: < H
I 65828358: 62 x row
85.263688 c

58889»

154

 

 

 

8
A
I
0 I l I l
Treatment
0 Control (soilid line)
X No litter (dashed line)
500‘ A Litter (dotted line)
A -

Bareground (grey line)

 

 

 

Soil Temperature (°C)

Figure 5.7. Relationship between soil temperature and C02 emissions in 2003 and 2004.
Analysis of covariance (ANCOVA) results showed a signiﬁcant temperature effect in
2003 (p<0.003): C02 ﬂux = 14.9 + 0.026 (temperature), r2 = 0.22 (p<0.000). The line in
2003 thus represents all treatments. In 2004 there was a temperature by treatment
interaction (p<0.001). Control plots: C02 ﬂux = 13.3 + 0.020 (tem rature), r2 = 0.08
(p<0.000); no litter plots: C02 ﬂux = 12.1 + 0.033 (temperature), r = 0.40 (p<0.000);
litter plots: C02 ﬂux = 13.8 + 0.018 (tem erature), r2 = 0.22 (p<0.000); bare-ground plots:
C02 ﬂux = 15.5 + 0.026 (temperature), r = 0.20 (p<0.000).

155

Treatment
0 Control (soilid line)

 

0-7 cm 0-7 cm X No litter (dashed line)
50‘ b
lc b 1 l A Litter (dotted line)
b. I Bareground (grey line)
40‘ a
a

Soil Nitrate (pg 9")

 

 

Soil Nitrate (pg 94)

 

 

 

 

Soil Nitrate (pg 9")

 

125 155 185 215 245 275 305 100 135 170 205 240 275 310

Figure 5.8. Cultivation effects in a previously uncultivated ﬁeld on soil NO3' in
2003 and 2004 at 0-7, 7-20, and 0-20 cm depth. No litter plots had all biomass
removed prior to cultivation. Litter plots had all aboveground bimass incorporated
with tillage. Bare-ground plots were weeded and contained root exclusion barriers.
Arrows indicate cultivation dates. Letters within a day of year are stacked in order:
control, no litter, litter, bare-ground. Different letters indicate treatment differences
within a da)I of year and soil depth. Asterisks indicate diffences between soil depths
within a treatment and day of year. Bars are standard errors.

156

Treatment

0-7 cm Control (soilid line)

    

 

 

 

 

 

 

 

 

 

b b:
.1“ J; g.
b) 50 a 2 av- 1 l X No litter (dashed line)
a) a a A Litter (dotted line)
3 40‘ b " % Bareground (grey line)
.2 3O _ i. ’i a — a“ t t
C H a w * i n- *
_- bC i a
O _ .- r‘: _ a“ * , * a“
E 20 b. . b be
E .- b‘ *7 _ ~ b
< 10‘ — b. “'8" ' 5“ C
o O ﬁﬁ fl ' l 1 l " I ' 5: ’ i ' i ' l Til ' l
(D
6‘ - 7-
ca 5077 20 cml 20 cm i
O)
3 40- B .
E
.2 30 — g -
C ' b
O ‘ _ b b
E 20‘ is) ab a ”x. B a b b a
E a a ab / a ab .3 g a
< i" g a byla,
(’0) ~' I l

 

 

 

l ' l ' l ' I ' I ' I

 

' I ' I ' ' I ' I ' I
125 155 185 215 245 275 305 125 155 185 215 245 275305

Soil Ammonium (pg 9")

Day of Year 2003 Day of Year 2004

Figure 5.9. Cultivation effects in a previously uncultivated ﬁeld on NH4+ in 2003 and
2004 at 0-7, 7—20, and 7-20 cm depth. No litter plots had all biomass removed prior to
cultivation. Litter plots had all aboveground bimass incorporated with tillage. Bare-
ground plots were weeded and contained root exclusion barriers. Letters within a day of
year are stacked in order: control, no litter, litter, bare-ground. Different letters indicate
treatment differences within a day of year and soil depth. Asterisks indicate diffences
between soil depths within a treatment and day of year. Bars are standard errors.

157

Mineralizeable C
we COz'C 91'1 d‘1

 

 

 

 

 
       

  

75 Treatment
3 '30 1 7-20 cm ' Control
3 To) 60 D No litter
(U 32:23" '
g L? 45_ LItter
E 6'
.“g’ g,
E 1
75 ‘1
o T 60 A 0-20 cm
2 'C
a To: 45 ~
E 9
E 6‘ 3° ‘
G) l
.5 ‘3, 15—
E a. . I
O_ :31. xi: . ~ 15;."

 

Sap-03 Oct-03

Figure 5.10. Cultivation effects in a previously uncultivated ﬁeld on mineralizable C
in 2003 at 0-7, 7-20, and 0-20 cm depth. No-litter plots had all biomass removed
prior to cultivation. Bare-ground plots were weeded and contained root exclusion
barriers.

May-03 Jul-O3

158

 

 

I Control
400 “ a No litter
'7 7-20 cm -- -
< f 300 _ a Litter A
UJ on D Bareground A
D x

 

 

 

0-20 cm

 

 

Jul-03 Aug-O3 Oct-O3 Aug-O4

Figure 5.11. Cultivation effects in a previously uncultivated ﬁeld on denitriﬁcation
enzyme activity (DEA) in 2003 and 2004 at 0-7, 7-20, and 0-20 cm depth. No-litter
plots had all biomass removed prior to cultivation. Bare-ground plots were weeded
and contained root exclusion barriers.

159

Treatments

 

I Control
250‘" A D No litter
:7: 200_ 0-7 cm a Litter
g) D Bare-ground
Z. A A
'co
0
Z
a)
1

 

 

 

pg NO3--N kg h-1

 

Nitriﬁer Enzyme Activity Nitriﬁer Enzyme Activity

 

 

 

 

>5

Ev- 250-
21; 200—
“5’; 1504
E I” 100-
mg

3?ng 50-
’2' 0

 

 

Jun-O3 Sep-03 Oct-03 May-04 Aug-O4

Figure 5.12 Cultivation effects in a previously uncultivated field on nitriﬁer enzyme
activity in 2003 and 2004 at 0-7, 7-20, and 0-20 cm depth. No-litter plots had all
biomass removed prior to cultivation. Bare-ground plots were weeded and
contained root exclusion barriers.

160

REFERENCES

(CAST). 2004. Emissions and mitigation of agricultural greenhouse gases. in Climate
Change and Greenhouse Gas Mitigation: Challenges and Opportunities for Agriculture.
Council for Agricultural Science and Technology (CAST), Ames, Iowa, USA.

Crum, J .R., and HP. Collins. 1995. KBS Soils [Online]. Available at
www.lter.kbs.msu.edu/soil/characterization. W. K. Kellogg Biological Station

Long-Term Ecological Research Project, Michigan State University, Hickory
Comers, MI.

Arrouays, D., and P. Pelissier. 1994. Changes in carbon storage in temperate humic

loamy soils after forest clearing and continuous corn cropping in France. Plant
and Soil 1602215-223.

Baggs, E. M., M. Stevenson, M. Pihlatie, A. Regar, H. Cook, and G. Cadisch. 2003.
Nitrous oxide emissions following application of residues and fertiliser under zero
and conventional tillage. Plant and Soil 254:361-370.

Baisden, W. T., and R. Amundson. 2003. An analytical approach to ecosystem
biogeochemistry modeling. Ecological Applications 13:649-663.

Ball, B. C., A. Scott, and J. P. Parker. 1999. Field N20, C02 and CH4 ﬂuxes in relation to
tillage, compaction and soil quality in Scotland. Soil & Tillage Research 53:29-
39.

Bédard, C., and R. Knowles. 1989. Physiology, biochemistry, and speciﬁc inhibitors of
CH4, NH4+, and CO oxidation by methanotrophs and nitriﬁers. Microbiological
Reviews 53:68-84.

Burgess, M. S., G. R. Mehuys, and C. A. Madramootoo. 2002. Decomposition of grain-
com residues (Zea mays L.): A litterbag study under three tillage systems.
Canadian Journal of Soil Science 82:127-138.

Burke, R. A., J. L. Meyer, J. M. Cruse, K. M. Birkhead, and M. J. Paul. 1999. Soil-
atmosphere exchange of methane in adjacent cultivated and ﬂoodplain forest
soils. Journal of Geophysical Research-Atmospheres 104:8161-8171.

Caldeira, K., M. G. Morgan, D. Baldocchi, P. G. Brewer, C. T. A. Chen, G.-J. Nabuurs,
N. Nakicenovic, and G. P. Robertson. 2004. A portfolio of carbon management
options. Pages 103-130 in C. Field and M. Raupach, editors. The Global Carbon
Cycle. Island Press, Washington, DC, USA.

Cochran, V. L., E. B. Sparrow, S. F. Schlentner, and C. W. Knight. 1997. Long-term
tillage and crop residue management in the subarctic: ﬂuxes of methane and
nitrous oxide. Canadian Journal of Soil Science 77:565-570.

161

Davidson, B. A., and I. L. Ackerman. 1993. Changes in soil carbon inventories following
cultivation of previously untilled soil. Biogeochemistry 20:161-193.

DeGryze, S., J. Six, K. Paustian, S. J. Morris, E. A. Paul, and R. Merckx. 2004. Soil
organic carbon pool changes following land-use conversions. Global Change
Biology 10:1120-1132.

Elmi, A. A., C. Madramootoo, C. Hamel, and A. Liu. 2003. Denitriﬁcation and nitrous
oxide to nitrous oxide plus dinitrogen ratios in the soil proﬁle under three tillage
systems. Biology and Fertility of Soils 38:340-348.

Franzluebbers, A. J ., and J. L. Steiner. 2002. Climatic inﬂuences on C storage with no
tillage. Pages 71-86 in J. M. Kimble, R. Lal, and R. F. Follett, editors. Agriculture
Practices and Policies for Carbon Sequestration in Soil. CRC Press, Boca Raton.

Gale, W., C. Cambardella, and T. Bailey. 2000. Surface residue— and root-derived carbon

in stable and unstable aggregates. Soil Science Society of America Journal
64:196-201.

Goldberg, D. E., and S. M. Scheiner. 2001. ANOVA and ANCOVA: Field competition
experiments. Pages 77-98 in S. M. Scheiner and J. Gurevitch, editors. Design and
analysis of ecological experiments. Oxford University Press, New York.

Grant, B., W. N. Smith, R. Desjardins, R. Lemke, and C. Li. 2004. Estimated N20 and
C02 emissions as inﬂuenced by agricultural practices in Canada. Climatic Change
65:315-332.

Groffman, P. M., E. A. Holland, D. D. Myrold, G. P. Robertson, and X. Zou. 1999.
Denitriﬁcation. Pages 272-290 in G. P. Robertson, D. C. Coleman, C. S. Bledsoe,
and P. Sollins, editors. Standard Soil Methods for Long-Term Ecological
Research. Oxford University Press, New York.

Hanson, P. J ., N. T. Edwards, C. T. Garten, and J. A. Andrews. 2000. Separating root and
soil microbial contributions to soil respiration: A review of methods and
observations. Bio geochemistry 48:1 15-146.

Hart, S. C., J. M. Stark, E. A. Davidson, and M. K. Firestone. 1994. Nitrogen
mineralization, immobilization, and nitriﬁcation. Pages 985-1018 in R. W.
Weaver, J. S. Angle, P. J. Bottomley, D. F. Bezdicek, M. S. Smith, M. A.
Tabatabai, and A. G. Wollum, editors. Methods of Soil Analysis, Part 2-
Microbiological and Biochemical Properties. Soil Science Society of America,
Madison, Wisconsin, USA.

Hilger, H. A., A. G. Wollum, and M. A. Barlaz. 2000. Landﬁll methane oxidation
response to vegetation, fertilization, and liming. Journal of Environmental Quality
29:324-333.

162

Hiitsch, B. W., C. P. Webster, and D. S. Powlson. 1993. Long—term effects of nitrogen
fertilization on methane oxidation in soil of the broadbalk wheat experiment. Soil
Biology & Biochemistry 25: 1307-1315.

Hiitsch, B. W., C. P. Webster, and D. S. Powlson. 1994. Methane oxidation in soil as
affected by land use, soil pH and N fertilization. Soil Biology & Biochemistry
26:1613-1622.

IPCC. 2001. Climate Change 2001; Synthesis Report. Cambridge University Press,
Cambrdge, UK.

J astrow, J. D., T. W. Boutton, and R. M. Miller. 1996. Carbon dynamics of aggregate-
associated organic matter estimated by carbon-13 natural abundance. Soil Science
Society of America Journal 60:801-807.

Kauppi, P., and R. DSedjo. 2001. Tecnological and economic potential of options to
enhance, maintain, and manage biological carbon reservoirs and geo-engineering.
Pages 301-344 in B. Metz, 0. Davidson, R. Swart, and J. Pan, editors. Climate
Change 2001, Mitigation. Cambridge University Press, Cambridge, UK.

Keller, M., E. Veldkamp, A. M. Weitz, and W. A. Reiners. 1993. Effect of pasture age on
soil trace—gas emissions from a deforested area of Costa-Rica. Nature 365:244-
246.

Kessavalou, A., J. W. Doran, A. R. Mosier, and R. A. Drijber. 1998. Greenhouse gas
ﬂuxes following tillage and wetting in a wheat-fallow cropping system. Journal of
Environmental Quality 27 :1 105-1 1 16.

Kessavalou, A., A. R. Mosier, J. W. Doran, R. A. Drijber, D. J. Lyon, and O.
Heinemeyer. 1998. Fluxes of carbon dioxide, nitrous oxide, and methane in grass

sod and winter wheat-fallow tillage management. Journal of Environmental
Quality 27: 1 094-1 104.

Kruse, C. W., and N. Iversen. 1995. Effect of plant succession, plowing, and fertilization
on the microbiological oxidation of atmospheric methane in a heathland soil.
FEMS Microbiology Ecology 18:121-128.

Lal, R. 2003. Global potential of soil carbon sequestration to mitigate the greenhouse
effect. Critical reviews in Plant Sciences 22:151-184.

Lal, R. 2004. Soil carbon sequestration impacts on global climate change and food
security. Science 304:1623-1627.

Livingston, G. P., and G. L. Hutchinson. 1995. Enclosure-based measurement of trace
gas exchange: applications and sources of error. Pages 14-51 in P. Matson and R.
Harriss, editors. Biogenic Trace Gases: Measuring Emissions from Soil and
Water. Blackwell Science, Osney Mead, Oxford, UK.

163

Luo, Y. Q., R. E. White, P. R. Ball, and R. W. Tillman. 1996. Measuring denitrifrcation
activity in soils under pasture: optimizing conditions for the short-term denitrifrcation

enzyme assay and effects of soil storage on denitriﬁcation activity. Soil Biology &
Biochemistry:409-417.

Lupwayi, N. Z., G. W. Clayton, J. T. O'Donovan, K. N. Harker, T. K. Turkington, and W.
A. Rice. 2004. Decomposition of crop residues under conventional and zero
tillage. Canadian Journal of Soil Science 84:403-410.

MacKenzie, A. F ., M. X. Fan, and F. Cadrin. 1997. Nitrous oxide emission as affected by

tillage, com-soybean-alfalfa rotations and nitrogen fertilization. Canadian Journal
of Soil Science 77:145-152.

Martens, D. A. 2001. Nitrogen cycling under different soil management systems.
Advances in Agronomy 70: 143-192.

Mikha, M. M., and C. W. Rice. 2004. Tillage and manure effects on soil and aggregate-
associated carbon and nitrogen. Soil Science Society of America Journal 68:809-
816.

Miller, A. J ., R. Amundson, I. C. Burke, and C. Yonker. 2004. The effect of climate and
cultivation on soil organic C and N. Biogeochemistry 67 :57-72.

Mosier, A., J. A. Delgado, and M. Keller. 1998. Methane and nitrous oxide ﬂuxes in an
acid oxisol in western Puerto Rico; effects of tillage, liming, and fertilization. Soil
Biology & Biochemistry 30:2087-2098.

Mosier, A. R. 2001. Exchange of gaseous nitrogen compounds between agricultural
systems and the atmosphere. Plant and Soil 228217-27.

Mosier, A. R., J. M. Duxbury, J. R. Freney, O. Heinemeyer, and K. Minami. 1998.
Assessing and mitigating N;O emissions from agricultural soils. CLIMATIC
CHANGE 40:7-38.

Mosier, A. R., W. J. Parton, D. W. Valentine, D. S. Ojima, D. S. Schimel, and O.
Heinemeyer. 1997. CH; and N20 ﬂuxes in the Colorado shortgrass steppe: 2.
Long-term impact of land use change. Global Biogeochemical Cycles 11:29—42.

Pacala, S., and R. Socolow. 2004. Stabilization wedges: Solving the climate problem for
the next 50 years with current technologies. Science 305:968-972.

Paustian, K., J. Six, E. T. Elliott, and H. W. Hunt. 2000. Management Options for
reducing C02 emissions from agricultural soils. Biogeochemistry 482147-163.

Pierce, F. J ., M.-C. Fortin, and M. J. Staton. 1994. Periodic plowing effects on soil

properties in a no-till farming system. Soil Science Society of America Journal
58:1782-1787.

164

Pinto, M., P. Merino, A. del Prado, J. M. Estavillo, S. Yamulki, G. Gebauer, S. Piertzak,
J. Lauf, and O. Oenema. 2004. Increased emissions of nitric oxide and nitrous oxide
following tillage of a perennial pasture. Nutrient Cycling in Agroecosystems 70213-22.

Robertson, G. P., E. A. Paul, and R. R. Harwood. 2000. Greenhouse gases in intensive
agriculture: Contributions of individual gases to the radiative forcing of the
atmosphere. Science 289:1922-1925.

Rochette, P., D. A. Angers, and L. B. Flanagan. 1999. Maize residue decomposition and
measurement using soil surface carbon dioxide ﬂuxes and natural abundance of
carbon-13. Soil Science Society of America Journal 63:1385-1396.

Six, J ., E. T. Elliott, and K. Paustian. 1999. Aggregate and soil organic matter dynamics
under conventional and no-tillage systems. Soil Science Society of America
Journal 63:1350-1358.

Six, J ., S. M. Ogle, F. J. Breidt, R. T. Conant, A. R. Mosier, and K. Paustian. 2004. The
potential to mitigate global warming with no-tillage management is only realized
when practised in the long term. Global Change Biology 10:155-160.

Six, J ., K. Paustian, E. T. Elliott, and C. Combrink. 2000. Soil structure and organic
matter: 1. Distribution of aggregate-size classes and aggregate-associated carbon.
Soil Science Society of America Journal 64:681-689.

Smith, P., K. W. Goulding, K. A. Smith, D. S. Powlson, J. U. Smith, P. Falloon, and K.
Coleman. 2001. Enhancing the carbon sink in European agricultural soils:
including trace gas ﬂuxes in estimates of carbon mitigation potential. Nutrient
Cycling in Agroecosystems 60:237-252.

Suwanwaree, P., and G. P. Robertson. 2005. Methane oxidation in forest, successional,
and no-till agricultural ecosystems: effects of nitrogen on soil disturbance. Soil
Science Society of America Journal In press.

VandenBygaart, A. J ., and B. Kay. 2004. Persistence of soil organic carbon after plowing
a long-term no-till ﬁeld in southern Ontario, Canada. Soil Science Society of
America Journal 68:13941402.

West, T. O., and W. M. Post. 2002. Soil organic carbon sequestration rates by tillage and

crop rotation: a global data analysis. Soil Science Society of America Journal
66:1930-1946.

165

CHAPTER 6

Long-Term Trends in Nitrous Oxide Emissions, Soil Nitrogen, and Crop Yields of

Till and No-Till Cropping Systems

ABSTRACT
No-till cropping can increase soil C stocks and aggregation but potential long-term
changes in N20 emissions, soil N availability, and crop yields still need to be resolved.
We measured soil C accumulation, aggregation, soil moisture, N;O emissions, soil
inorganic N, and crop yields in till and no-till com-soybean-wheat rotations between
1989 and 2002 in southwest Michigan and investigated whether tillage effects varied over
time or by crop. Mean annual NO3' concentrations in no-till were signiﬁcantly less in
three of six corn years and during one year of wheat production. Yields were similar in
each system for all but three of the 14 years during which yields were higher in no-till,
indicating that soil NO3' reductions did not translate into yield losses. N;O emissions
were similar with till (3.27 :t 0.52 g N ha (1") and no-till (3.63 :i: 0.53 g N ha (1") soil
management. C accumulated in no-till soils at a rate of 26 g C m'2 y'I over 12 yr;
between 56 and 61 % of the reduction in CO; equivalents associated with this C
sequestration was offset by N20 emissions, however. After controlling for rotation and
environmental effects by normalizing treatment differences between till and no-till
systems we found no signiﬁcant trends in soil N, N;O emissions or yields through time.
In these sandy-loam soils, no-till cropping enhances C storage, aggregation and

associated environmental processes with no significant ecological or yield tradeoffs.

166

INTRODUCTION

Soil organic matter (SOM) losses are rapid following initial tillage of
undisturbed soils (Martel and Paul, 1974; Balesdent et al., 1988; Davidson and
Ackerman, 1993) and commonly approach or exceed 20-40% of the original soil C after
twenty years (Davidson and Ackerman, 1993). Restoration of some portion of the 55 Pg
C lost from arable soils globally is one of several short-term, high impact options for
stabilizing global CO; levels (Caldeira et al., 2004; Cast, 2004; Lal, 2004). No-till
cropping generally increases soil C stocks at an average annual rate of 30 - 60 g C m'2 y'I
(Davidson and Ackerman, 1993; West and Post, 2002) and is thus a prominent
agricultural CO; mitigation strategy. Use of no-till soil management has been steadily
increasing the past decade and between 1994 and 2004 the percentage of total cropland in
no-till rose from 13.7 to 22.6%, or 9.5 million hectares (CTIC, 2004). Many more
hectares are suitable for no-till but its deployment has been limited in part by the
perception that accelerated rates of N leaching and immobilization may reduce plant N
availability and decrease yields (Martens, 2001). Further, studies have suggested that
increases in N;O emissions following conversion to no-till might offset some portion of
the CO; mitigation, making the long-term global warming potential (GWP) of the two
systems similar (MacKenzie et al., 1997; Six et al., 2004).

Higher N;O emissions in no-till than conventionally tilled cropping systems have
been frequently reported (MacKenzie et al., 1997; Ball et al., 1999; Baggs et al., 2003)
although some studies have found lower emissions in no-till soils or no difference
between tillage systems (Robertson et al., 2000; Elmi et al., 2003), demonstrating the

potential for responses to vary across cropping systems and soil types. In a recent review

167

Six et al. (2004) analyzed N;O emissions data in a linear mixed-effect model and

concluded that differences in N;O ﬂuxes between till and no-till change over time. In
humid ecosystems N20 ﬂuxes are higher in no-till during the ﬁrst decade but after 20
years no-till emissions are lower. In an assessment of potential European land-use
changes on non-C02 greenhouse gases, Smith et al. (2001) calculated that eliminating
tillage on all potential no-till lands could result in an increase in N20 emissions
equivalent to 20.5 Tg COz-C y'l. The potential mechanisms explaining this include
decreased 02 availability associated with increased bulk density and water-ﬁlled pore
space and increased soil organic matter decomposition in no-till soils (Linn and Doran,
1984; Guzha, 2004); alterations to the synchrony between plant N uptake and N
availability (Six et al., 2004); and litter C accumulation at the soil surface (Arshad et al.,
1999; Baggs et al., 2003).

Along with soil surface N20 emissions, soil inorganic N cycling frequently
changes following conversion to no-till and these changes may decrease yields (Rice et
al., 1986; Niehues et al., 2004). Application of urea ammonium nitrate (UAN) fertilizer
to the soil surface results in greater ammonia volatilization in no-till than in conventional
till systems where fertilizer is incorporated into the soil (Keller and Mengel, 1986; Fox
and Piekielek, 1993). Along with volatilization, immobilization of N in surface residues
and decreased N mineralization rates can limit N availability and potentially decrease
yields in no-till (Kaspar et al., 1987; Ismail et al., 1994). Reduced mineralization rates
and other problems (e. g. delayed and uneven germination) due to cool and water-

saturated soils may be particularly detrimental to yields in the northern corn belt (Kaspar

168

et al., 1987; Vetsch and Randall, 2000) and on ﬁne-textured soils (Vyn and Raimbault,
1993; Beyaert et al., 2002).

As with N;O emissions, soil N cycling and crop productivity responses to no-till
may change over time. Vyn and Raimbault (1993) found that com yields increased in no-
till during the ﬁrst eight years and then declined. Other studies have found that no-till
has no effect on yields (Mehdi et al., 1999; Beyaert et al., 2002), decreases yields initially
before a yield recovery (Rice et al., 1986), or is related to annual environmental
conditions or other factors independent of time since initiation (Ismail et al., 1994).
Additional data are needed describing long-term annual trends in soil N cycling, yields
and, particularly, N20 emissions to develop predictive models for long-term no-till
cropping systems.

Our objectives in this study are to: 1) determine the effects of long-term no-till
soil management in a com-soybean wheat rotation on C sequestration and the extent to
which increases in N;O ﬂux can offset the GWP impact of enhanced C storage; 2)
determine whether changes in soil moisture content, N availability, and yields occur that
could limit the deployment of no-till in our region (southwest Michigan); and 3)
investigate whether responses to no-till vary over 14 y of no-till management. In a
previous study Robertson et al. (2000) reported no difference in the mean annual NzO
ﬂux from these treatments between 1991 and 1999. Here, we investigate soil moisture
content, N;O emissions, soil N concentrations, and crop yield by year to better
understand the inter-annual patterns associated with no-till management between 1989

and 2002. We predict that N;O emissions will initially be higher in no-till soils and then

169

decrease over time and that soil inorganic N concentrations will be lower in no-till soils,

potentially lowering yields of corn and wheat in these northern corn belt soils.

MATERIALS AND METHODS

Experimental Site

Experimental plots were located at the WK. Kellogg Biological Station Long-
Term Ecological Research (LTER) Site (http://lter.kbs.msu.edu) in southwest Michigan
(85° 24'W, 42° 24'N; 920 mm y‘l precipitation). Tillage treatments were established in
1989 in six replicated one ha plots organized in a randomized complete block design.
Soils are Kalamazoo (ﬁne-loamy) and Oshtemo (coarse-loamy) mixed, mesic, Typic
Hapludalfs developed on glacial outwash (Crum and Collins, 1995). The two series co-
occur in all ecosystems and differ mainly in their Ap horizon texture, though variation

within a series can be as great as variation between a series (Robertson et al., 1997).

Agronomic Protocols

The crop rotation prior to 1995 consisted of corn followed by soybeans. In 1995,
wheat was planted after soybean, initiating a com-soybean-wheat rotation. Primary
tillage from 1989 to 1998 consisted of spring moldboard plowing and spring chisel
plowing from 1999 and 2002. Secondary tillage consisted of disking before wheat
planting, ﬁeld conditioner prior to soybean and corn planting, and inter-row cultivation
for soy and corn. Fertilizer applications were identical in both systems. Detailed
information about annual applications can be found on the KBS LTER website

(http://lter.kbs.msu.edu). Side-dress NH4NO3 fertilizer was broadcast over corn in 1989

170

(112 kg N ha"), 1991 (112 kg N ha") and 1993 (84 kg N ha"). In 1996 and 1999, corn

received 28 kg N ha"1 UAN (28%) at planting and an additional 135 kg side-dress N ha'1
as NH4+NO3I In 2002, corn received 29 kg N ha'1 starter fertilizer and an additional 124
kg side-dress N ha]. In April 1995 and 1998 wheat received 56 kg side-dress N ha’1
broadcast as NH4+NO3I The 2001 wheat crop received 71 kg side-dress N as UAN. No
fertilizer was applied to soybeans. Lime, P, and K were applied as needed according to

Michigan State University best management recommendations.

Soil Sampling and Storage

Soil samples for aggregate size distribution and total C and N analysis were
collected from ﬁve sites within each plot in June and July, 2001. At each of the ﬁve
sample locations, two subsamples with a diameter of 7.6 cm were taken to a depth of 5
cm by gently hammering a PVC core into the ground to minimize compression and
slicing of aggregates. One of the subsamples at each station was taken in the row and the
other between the rows. All ten subsamples from each plot were combined to produce
one representative sample. Four separate samples for bulk density analysis were taken at

the same time as those for aggregate analysis, using a 8.0 cm diameter root corer.

F ield-moist soil samples were put into a cooler (4°C) prior to being broken along
natural fracture planes and passed through an 8 mm sieve within 72 h of sampling. Aﬂer
sieving, soils were air-dried in paper bags at 20°C prior to storage in plastic bags. Care
was taken throughout the study to minimize disturbance of the samples that might

inﬂuence aggregate structure. Total soil C and N concentrations were determined by dry

171

combustion methods in a CHNS analyzer (Costech ECS 4010, Costech Analytical

Technologies, Valencia CA.).

Soil Aggregation
Aggregate distribution was determined on triplicate 35 g air-dried soil samples by
hand-sieving in water through a series of sieves (2000 um, 250 um, and 53 um). Mean
weight diameter (WMD) of sand-free aggregates was determined by calculating the sum
of the products of the mean diameter of each size fraction and the proportion of the total

sample weight in that fraction (Kemper and Rosenau, 1986).

. N;O Gas Emissions

N20 measurements were made using static chambers (Livingston and Hutchinson,
1995) at weekly to monthly intervals when soils were not frozen (Robertson et al., 2000).
Single chambers were located in four of the six replicates of each treatment. Chamber
lids were placed on semi-permanent aluminum bases removed only for cropping
activities and accumulated headspace sampled four times over 120 min. All chambers
were sampled on the same dates between 1991 and 2002, although no data are available
for 1995. Samples were stored in 3 ml crimp-top vials and analyzed in the laboratory for
N;O with the ﬂux for each chamber calculated as the linear portion of the gas
accumulation curve for that chamber. N;O was analyzed by gas chromatography using a

63 Ni electron capture detector (ECD).

Inorganic Soil N

172

NH4+ and NO; were extracted monthly from 1991-1995 and bimonthly from
1996-2002 from triplicate, ﬁeld-moist, 10 g soil samples using a 1:10 soil/extractant
ratio. Soil slurries were shaken for 1 minute, left to equilibrate overnight, and re-shaken
>1 h prior to ﬁltering. Extracts were ﬁltered with a syringe ﬁlter using a one um, glass
ﬁber ﬁlter. Filtrates were stored in 7 ml scintillation vials and ﬁozen until analysis for
NH; + and N03' using an automated ﬂow injection analyzer (NH4 + via diffusion

colorimetry and N03' via cadmium reduction and colorimetry).

Global Warming Potential
Global Warming Potential (GWP) calculations were made for soil C accumulation
in no-till relative to till between 1989 and 2001 and N;O emissions from both tillage
systems. GWP for soil C accumulation in no-till was calculated according to Robertson

et al. (2000) as follows:

 

_, _(xl —x,)kgC x 44kgCO2 x103 gCO2
mZ-x3y 12kgC 1kgCO2

X g CO2 ~m‘2 - y
where x1 = soil C in no-till (kg C m'z); x2 = soil C in conventional till (kg C ml); and x2 =
years (y) of accumulation. GWP calculations for N20 used an IPCC 20-year time

horizon factor of 280 for N20:

_._x,gN,o-Nx 44gN,o x365dx lha x280gco2
ha-d 28gNZO-N 1y 104m2 lgNzO

 

XgCO2 ~m’2-y

where x1 = average daily N20-N emission rate (g N ha'1 d'l).

Statistical Analysis

173

Tillage effects were analyzed by Proc Mixed (Version 8.2, SAS Institute, 1999)
using a randomized complete block design analysis of variance (ANOVA) with repeated
measures. Year and tillage treatments were considered ﬁxed effects. Where signiﬁcant
treatment by year interactions occurred, results were sliced to determine whether
treatment means signiﬁcantly differed within individual years. Additionally, to
determine trends in N;O ﬂux, soil inorganic N and crop yields over time, differences
between treatments within each block and year were normalized to the overall mean for

that year:
Normalized percent difference = [((Yc - Yn) / Yannua1)* 100]

where Yc and Y., are the annual mean for each block within a year for conventional and
no-till, respectively, and Yannual is the overall mean for that year, averaged across
treatments. Regression analysis was used to determine whether there were trends in the
response of till relative to no-till sites across crops and years and also whether trends
differed by crop. By normalizing the data prior to the regression analysis rather than
using absolute differences we were able to isolate the relative effect of no-till over time
and control for factors, such as the use of fertilizer or climatic differences, that might
cause inter-annual variation in the magnitude of response in both treatments. The
normalized differences were also analyzed by ANOVA using Proc Mixed with year as
the ﬁxed effect. Where there were signiﬁcant year effects, indicating that the normalized
difference varied with year, mean separation and ranking was carried out using a Tukey’s

test with the PDMix8OO Algorithm (Saxton, 1998).

174

RESULTS
Soil Properties

In 2001, 12 years after the adoption of no-till, soil aggregate MWD was 55%
higher in no-till soils and there was an additional 310 g C m’2 to a depth of 5 cm (Table
6.1). This change over 12 years represents an annual C increase to 5 cm of 26 g C m‘2 y'
I. No bulk density differences were detected between the two treatments (Table 6.1).

There was a signiﬁcant treatment by year interaction (p<0.05) for gravimetric soil
moisture content (Figure 6.1). Soil moisture content was greater in no-till than till in
2000 but in all other years soil moisture content was similar in the two treatments.
Regression analysis of the normalized difference between till and no-till treatments found
no trend across all years or for particular crops.

There was a signiﬁcant treatment by year interaction (p<0.05) for soil NO3'
concentrations (Figure 6.2) indicating that the effects of tillage differed by year. No till
reduced NO3‘ concentrations in 1991 (com, 80%), 1996 (com, 29%), 1999 (com, 41%)
and 2001 (wheat, 28%). There was not a signiﬁcant linear trend in NO3' concentrations
over all years, although there was a signiﬁcant positive relationship between time and the
normalized N03“ difference during wheat years (normalized NOg' difference = -15650 +
7.84(year); r2 = 0.44; n = 3). This suggests that over 1995, 1998 and 2001 soil NO3'
concentrations declined in no-till wheat plots relative to tilled plots. Mean separation
showed that these three points were statistically equal and that 1991 was different from
the other years because of higher NO3' concentrations in tilled soils.

Response of soil NI-L;+ concentration to no-till also varied by year (p<0.05; Figure

6.3). NH4+ concentrations were higher in tilled sites in 1991 (corn) and 2002 (corn) and

175

no-till sites in 2000 (soybean). Linear regression showed that there was not a

signiﬁcant trend in NH4+ concentrations across all the crops and years and there were no
signiﬁcant trends for individual crops. Mean separation of the normalized difference by
year showed the potential for inter-annual variability in the response to tillage but also

showed no consistent trends by crop.

N20 Emissions and GWP

There was no detectable difference in N;O emissions between tillage treatments
averaged across years (conventional till = 3.27 :h 0.53; no-ti11= 3.63 i 0.53 g ha‘l d'l),
but there was a signiﬁcant year by treatment interaction (Figure 6.4) with higher
emissions occurring in no-till plots in 1991 (com, 300%) and 2000 (soybean, 300%).
Linear regression showed that there was no signiﬁcant relationship between time and the
normalized difference in N20 ﬂux across years and crops but indicated that for the
soybean crop N;O emissions increased in no-till plots relative to till plots over 4 growing
seasons between 1992 and 2000 (normalized N20 difference = 31210 — 15.7 (year); r2 =
0.48; n = 4 years). Tukey’s test (p<0.05) showed that normalized differences in N20
emissions were similar across years. In the no-till system C storage between 1989 and
2001 reduced soil C02 ﬂuxes by 95 g CO; rn'2 y'1 but 56-61% of that mitigation was

offset by NzO emissions of 53-58 g C02 equivalents m'2 y'1 (Table 6.2).
Crop Yields

There was a signiﬁcant (p<0.05) treatment by year interaction for crop yields

(Figure 6.5). Crop yields were signiﬁcantly greater in no-till sites in 1992 (soybean),

176

 

 

1996 (corn) and 1997 (soybean). There was not a signiﬁcant linear relationship between
crop yield and time across years. There was a signiﬁcant positive relationship between
time and wheat yields in no-till relative to conventional till (normalized yield difference =
-5980 + 3.00(year); p < 0.05; r2 = 0.40; n = 3 years) for three growing years between
1995 and 2001. Mean separation indicated that the normalized mean difference was

statistically equal for these three wheat years.

DISCUSSION
N20 and GWP
Although we measured signiﬁcantly greater NzO ﬂuxes from no-till plots in 1991
and 2000 such differences were not sustained throughout the study, and averaged across
years there was not a signiﬁcant effect of tillage (Figure 6.4). Grant et al. (2004)
predicted that across all of Canada adoption of no-till would reduce N;O emissions by an
average of 17% but in Eastern Canada greater precipitation and soil water-ﬁlled pore
space would increase N;O emissions. Grant et al.’s predictions are consistent with Six et
a1. (2004) that in humid climates no-till systems will have higher NzO emissions for a
decade or more after conversion. MacKenzie et al. (1997) found in eastern Canada that
no-till increased N;O emissions from 17 to 60%. In Saskatchewan, Canada, Aulakh et al.
(1984) found that no-till doubled N;O ﬂuxes.
The mechanisms proposed to explain higher N;O ﬂuxes in no-till soils are related
mainly to C accumulation at the soil surface, increased bulk density, and decreased crop
yields altering the availability of C, N and 02 (Ball et al., 1999; Smith et al., 2001).

While some no-till soils may be more susceptible to Oz depletion following precipitation

177

(Doran, 1980; Martens, 2001), the 50% greater aggregate MWD and statistically equal

bulk density that we measured in no-till suggest that drainage and hence Oz availability
are similar or higher than those in tilled soils. Although these measurements were made
12 yr after no-till conversion and may not represent soil structural dynamics during the
ﬁrst few years after adoption of no till, they are conﬁrmed by measurements made on
samples taken in 1995 (Six et al., 2000) showing statistically greater aggregation in no-
till. Other studies have shown that no-till increases aggregation (Liebig et al., 2004), and
that these changes can occur within the ﬁrst few years after adoption (Rhoton, 2000), and
have a positive impact on water dynamics and crop yields (Diaz-Zorita etal., 2004).

The association between decreased crop yields and increased N;O ﬂux in no-till
has been attributed partly to an asynchrony between crop N uptake rates and soil N
availability (Six et al., 2004). Resulting changes in temporal patterns of soil inorganic N
availability coupled with greater water-ﬁlled pore space and C availability stimulate N;O
production. In our system, no-till did not signiﬁcantly affect crop yields averaged over
the entire experiment although in 1992 (soybean), 1996 (corn), and 1997 (soybean) there
were higher yields in no-till, suggesting that this asynchrony did not differ between
treatments.

Our results demonstrate that N;O emissions are a critical component of the GWP
of agricultural ecosystems (Table 6.2). We found that 56-61% of the reduction in GWP
associated with C sequestration in the top 5 cm was offset by N;O emissions from no-till.
These results are consistent with Robertson et al. (2000) who found that 51% of the soil
C storage to 7 cm in no-till was offset by NzO emissions between 1991 and 1999 in these

plots. Over time, C sequestration rates will slow down in the no-till soils as they

178

approach a new C equilibrium (West and Post, 2002) and, concomitantly, the proportion
of annual C sequestration that is offset by N20 emissions will increase.

Robertson et al. (2000) found that conventional, no-till, low-input, organic, and
continuous alfalfa cropping systems at the KBS LTER produced statistically similar N20
emissions between 1991 and 1999 and that only successional communities, where soil N
concentrations were substantially lower, had lower N;O emission rates. In our soils,
increases in N20 emissions from no-till are nil or a small proportion of the total N;O ﬂux
from agricultural soils receiving supplemental anthropogenic N (Table 6.2). Asynchrony
between N availability and crop uptake in systems receiving supplemental N results in
excess soil inorganic N and enhanced N20 production (MacKenzie et al., 1998; Grant et
al., 2004). Efforts to mitigate NzO emissions from agricultural cropping systems should
thus focus on improving N use efﬁciency in cropping systems (Mosier et al., 1998;
Mosier, 2001) rather than on tillage differences. Management practices designed to
better synchronize plant uptake with N availability have been described in detail
elsewhere (Mosier et al., 1998; CAST, 2004) and include using plant and soil sampling to

determine crop N requirements, split N fertilizer applications, and slow-release fertilizers.

Changes in Yields and Soil Properties
In the northern corn belt of the US, cold soils, compaction, and decreased N
availability can be particularly problematic in no-till soils, resulting in yield declines
(Mehdi et al., 1999; Vetsch and Randall, 2004). We found that that there was no yield

cost for no-till corn, soybean or wheat. In fact, yields were similar in each system for all

179

but three of the 14 years during which yields were higher in no-till, indicating that soil
NO3' reductions did not translate into yield losses.

The use of starter N fertilizer, side-dress N, and a two to three year rotation in our
experiment likely contributed to the persistence of equal or greater yields in no-till.
These management strategies reduce N limitation due to low mineralization rates or
fertilizer N immobilization in no-till. Vetsch and Randall (2000) reported an increase in
continuous corn yields of 0.5 Mg ha'1 when starter fertilizer was used in no-till systems
and Vetsch and Randall (2002) concluded that optimizing no-till com-yields following
soybean also required the use of starter fertilizer. Starter fertilizer enhances early season
crop growth and increases corn yields in no-till soils that may be susceptible to spring N
deﬁciencies (Jokela, 1992; Bullock et al., 1993). Side-dressing N after corn emergence
further optimizes crop N use efﬁciency by targeting the crop when it is actively removing
soil N (Fox et al., 1986; Mosier, et al., 1998; Vetsch and Randall, 2004). In a review of
N cycling in no-till, Martens (2001) suggested that N limitation may not be as great when
com is grown in rotation with other crops with differing N requirements.

Increased aggregation and soil C content indicates that, over time, no-till may
have become a better habitat for crop growth. Increased aggregate MWD will improve
aeration and drainage and make the system more resilient to compaction. Increased soil
C concentrations will improve soil structure and erosion resistance and potentially
increase the N mineralization potential during the growing season when soils are warm

enough to support active decomposition.

180

Trends over Time

Predicting N20 emissions and crop yields in no-till systems depends on
understanding short and long-term dynamics following transition to no-till. The potential
for N cycling, crop yield and N;O emissions in no-till to be dependent on the time since
conversion has been demonstrated by other studies (Rice et al., 1986; Vyn and
Raimbault, 1993; Six et al., 2004). We did not detect a trend through time in the
difference between till and no-till for soil moisture, N;O emissions, soil nitrate and
ammonium, or crop yields. Analysis of variance and mean separation of the normalized
differences by year further showed that while there was some inter-annual variability,
different years were generally very similar. Corn yields in both treatments were lower in
1996, 1999 and 2002 than in previous years due to localized, seasonal droughts during
critical stages of corn development.

There was some evidence of changes in N cycling shortly after the adoption of
no-till. Relatively high rates of N;O ﬂux were measured in 1991 and although these
differences were not sustained, they do corroborate other results showing the potential for
increased N;O production after no-till conversion (Ball et al., 1999; Yamulki and Jarvis,
2002; Baggs et al., 2003). Additionally, there seemed to be a trend towards decreased
soil N concentrations in no-till corn over time. In 1989, the N03" concentrations were the
same in both treatments but in 1991, 1996, 1999, and 2001 there was lower NO3’ in no-
till sites. However, linear regression, even with the anomalous 1991 results removed

(data not shown), did not show a signiﬁcant relationship between time and normalized

NO3' difference.

181

Long-term responses to no-till may vary considerably with soil type, climate and

management practices and their effects on litter accumulation and soil bulk density.
Following conversion to no-till, increased N immobilization is likely where litter C
accumulates rapidly at the soil surface (Baggs et al., 2003), and N20 production should
increase where increased bulk density alters soil pore-space dynamics (Linn and Doran,
1984; MacKenzie et al., 1998). McConkey et al. (2002) reported that N cycling and grain
yield responses to no-till varied considerably with soil type. On a sandy-loam soil, the
apparent N balance (applied N minus N removed in grain) was 26% greater in no-till than
minimum till sites over 13 years of continuous wheat; on a clay soil the apparent N
balance was 78% greater in no-till than continuous till sites suggesting that there was
greater immobilization or gaseous losses of N on ﬁne-textured soils. Several studies
reporting increased N;O emissions following no-till conversion have also been conducted
on ﬁne-textured soils including Ste. Rosalie clay and Ormstown silty clay loam
(MacKenzie et al., 1997; 1998); imperfectly drained clay loam (Ball et al., 1999); clay-
loam (Aulakh et al., 1984); and silt loam (Baggs et al., 2003). In contrast, Elmi et al.
(2003) found that no-till did not have higher N20 emissions than conventional till on a
sandy-loam soil in Quebec, Canada.

Over time, increases in soil aggregation and associated soil physical properties
and increased yields might decrease NzO production in ﬁne-textured, no-till soils. Rapid
increases in N;O emissions following cultivation of long-term no-till soils provides
evidence for the beneﬁts of long-term no-till (Estavillo et al., 2002; Pinto etal., 2004). In

our system with relatively low clay contents and low rates of litter accumulation due to

182

modest productivity and high decomposition rates, the negative effects of no-till are nil

or short-lived over the ﬁrst 12 years.

CONCLUSIONS

At the KBS LTER Site no-till increased soil C in the top 5 cm from 0.69 to 1.00
kg C In2 and increased aggregate MWD from 1.32 to 2.04 mm. N;O emissions were
higher in no-till in two out 10 years but there was no signiﬁcant effect of tillage averaged
across years. N;O emissions offset between 56 and 61% of the C02 reduction associated
with C sequestration in no-till. In the three years where signiﬁcant tillage effects on yield
occurred, no-till had higher yields than conventional till, indicating that decreases in N
availability with no-till did not reduce yields. There were no trends in soil N cycling,
yields, or N;O emissions across all years. No-till effects on ﬁne-textured soils are likely
stronger than those on sandy loam and other coarse-textured soils with high sand
concentrations. Our results demonstrate that the adoption of no-till can increase soil C

storage and physical structure without increased N20 emissions or yield trade-offs.

183

Table 6.1. No-till soil management effects on soil aggregation, bulk density, and C
sequestration in 2001 after 12 y at the KBS LTER Site. I

 

Aggregate MWD Bulk Density Organic C ACt
mm g cm'3 kg rn’2 g m'2 y'1
Conventional
till 1.32 (0.14)* 1.37 (0.01) 0.69 (0.04)* O
No till 2.04 (0.11) 1.36 (0.03) 1.00 (0.04) 26

 

* Signiﬁcant at the p<0.05 probability level
l 0-5 cm sampling depth
1 MWD: mean weight diameter.

‘5 Robertson et al. (2000) reported a C accumulation rate of 30 g C m'2 to 7 cm.

Table 6.2. Global warming potential (GWP) contributions in C02 equivalents
from soil C storage and N;O emissions in conventional and no-till systems
between 1991 and 2002 at the KBS LTER Site.

Conventional till No till

CO; Equivalents (g m"2 y")

 

Soil C Storagel 0 -95
N20 Emissions 53 58
GWP (N20 + C02) 53 -37

 

1 0-5 cm sampling depth.

* N;O emissions were statistically equal in till (3.27 i 0.52 g N ha (1") and no-
till (3.63 i 0.53 g N ha (1"). N;O production thus offset between 56 and 61%
of the CO; stabilization associated with soil C increases in no-till.

184

a
CS CS CSWCSWC SWC

 

25—
Treaﬂnent
_ 20_ - Till
'05) \o No-Till *
o
5:3 915— I ‘
iii 3
E 2
-- 010—
52
(D 5_
0—
b
50— Crop
. COITI

_ ct > nt A soybean
25 X wheat

ababab abab abab aibabab babab

 

-25 —

Soil Moisture (Ct — Nt)
NormaIIzed %
O

 

-50

l l l I l
1988 1991 1994 1997 2000 2003
Year

 

Figure 6.1. (a) Gravimetric soil moisture content in till and no-till treatments
between 1989 and 2002. Letters indicate the crop harvested in that year: c =
com; S = soybean; w = wheat. * indicates statistically different responses
within a year (p<0.05) determined by slicing of the treatment by year
interaction. (b) Difference between conventional till (Ct) and no-till (Nt)
normalized by the overall mean soil moisture content for that year. Points
greater than zero indicate a till response greater than a no-till response. Year
means were signiﬁcantly different (P<0.05) and separated using Tukey’s test.
Linear regression analysis showed no trends in soil moisture concentration
(p<0.05) across all crops and years or for speciﬁc crops over time.

185

 

 

 

 

 

 

 

20 a
C S C S C S W C S W C S W C
2' 15 _ * Treatment -
C's-,7 l Till
5’ l No-Till
Z _
__= 310
O
(D
zoo—b a .Cmp
com
A A soybean
2 °\° 100_ Ct>nt X Wheat
... m b b b b i b
o .N. b b b i I
v— _-- -- ..... 5-- ........... LI ..... l ..... __
2% ° i5 ‘1 i
In 8
(Z) 2 400a
nt>ct
-200 I I I I I
i 988 1991 1994 1997 2000 2003

Year

Figure 6.2. (a) Soil NO; concentration in till and no-till treatments between 1989 and
2002. Letters indicate the crop harvested in that year: c = corn; 5 = soybean; w = wheat.
* indicates statistically different responses within a year (p<0.05) determined by slicing
of the treatment by year interaction. (b) Difference between conventional till (Ct) and
no-till (Nt) normalized by the overall mean NO; concentration for that year. Points
greater than zero indicate a till response greater than a no-till response. Year means
were signiﬁcantly different (P<0.05) and separated using Tukey’s test. Linear
regression analysis showed no trend in N05 concentration (p<0.05) over across all
crops and years; NO; concentrations in wheat signiﬁcantly increased in till relative to
no-till over time [normalized NO3' difference = -15650 + 7.84 (year); r2 = 0.44; n = 3
years]. Mean i S.E.

186

15oa
cscscswcswcswc

—

Z' 12 Treatment
+ .

‘T I TI"
:1? C» 9— .
2 U) I NO-Tlll
_ :3. *

'5
U) 6"

 

   

 

 

 

 

3 _
O _
_ b
1 20 Crop
ct > nt 0 com a
ab A soybean E
60‘ abc E x wheat
bcd

 

NH4+ -N (Ct — Nt)
Normalized %
O
4.

 

 

 

 

 

'6Od nt > ct
-120 I I I I I
1 988 1991 1 994 1 997 2000 2003

Year

Figure 6.3. (a) Soil NI-I4+ concentrations in till and no-till treatments between 1989
and 2002. Letters indicate the crop harvested in that year: c = corn; 5 = soybean; w
= wheat. * indicates statistically different responses within a year (p<0.05)
determined by slicing of the treatment by year interaction. (b) Difference between
conventional till (Ct) and no-till (Nt) normalized by the overall mean NH4+
concentration for that year. Points greater than zero indicate a till response greater
than a no-till response. Year means were signiﬁcantly different (P<0.05) and
separated using Tukey’s test. Linear regression analysis showed no trends in soil
NH4+ concentration (p<0.05) across all crops and years or for speciﬁc cr0ps over
time.

187

so—a

z 70 20 _ Treatment

| 1- .

O '(D I Tlll

ZN; * I No-Till
or

 

 

 

 

 

 

N20-N (Ct - Nt)

 

 

 

 

 

 

— b
400 Crop
I com
o 200.. Ct > nt A soybean
o\ X wheat '
8 5 I
as
TE OT"""'"""""""’£ """ E "I" "E """ Z"§"'
E l i
O
z ~200d
nt>ct
—400 I I I I I
1 988 1 991 1‘ 994 1997 2000 2003

Year

Figure 6.4. (a) Soil N20 emissions in till and no-till treatments between 1991 and
2002 (no data available for 1995). Letters indicate the crop harvested in that year: c
= corn; 3 = soybean; w = wheat. * indicates statistically different responses within a
year (p<0.05) determined by slicing of the treatment by year interaction. (b)
Difference between conventional till (Ct) and no-till (Nt) normalized by the overall
mean N20 emissions for that year. Points greater than zero indicate a till response
greater than a no-till response. There was no detectable difference between
treatments based on a Tukey’s test. Linear regression analysis showed no trend in
N;O emissions (p<0.05) across all crops and years; emissions from soybean
signiﬁcantly increased in till relative to no-till over time [normalized N;O
difference = 31210 — 15.7 (year); r2 = 0.48; n = 4 years]. Mean i S.E.

188

a
10000 1

 

 
    
   
    

 

 

 

cscscswcswcswc
'o 8000 ~ Treatment
7, .7 - Till
; 2 6000 _ E NO-Tl”
°- 0)
2 x *
O 4000 _
2000*
0—
(D
8 60 b Crop
93 0 corn
0.) A soybean
:3: _t > nt
'5 304 ab >< wheat a ab
2 ababT ab: abab igabE-I—
.9 °\° 0_ I E
>' _IL I ab E b I
U z:
(D
% -3o— 1
nt > ct
g
2 -60 I I I I I
1988 1991 1994 1997 2000 2003
Year

Figure 6.5. (a) Crop yields in till and no-till treatments between 1989 and 2002.

Letters indicate the crop harvested in that year: c = com; s = soybean; w = wheat.
* indicates statistically different responses within a year (p<0.05) determined by

slicing of the treatment by year interaction. (b) Difference between conventional

till (Ct) and no-till (Nt) normalized by the overall mean crop yield for that year.
Points greater than zero indicate a till response greater than a no-till response.
Year means were signiﬁcantly different (P<0.05) and separated using Tukey’s test.
Linear regression analysis showed no trend in crop yields (p<0.05) across all crops

and years; wheat yields signiﬁcantly increased in till relative to no-till over time

[normalized yield difference = -5980 + 3.00 (year); r2 = 0.40; n = 3 years]. Mean d:

S.E.

189

REFERENCES

Arshad, M. A., A. J. Franzluebbers, and R. H. Azooz. 1999. Components of surface soil
structure under conventional and no-tillage in northwestern Canada. Soil & Tillage
Research 53:41 -47.

Aulakh, M. S., D. A. Rennie, and E. A. Paul. 1984. Gaseous nitrogen losses from soils
under zero-till as compared with conventional-till management systems. Journal
of Environmental Quality 13:130-136.

Baggs, E. M., M. Stevenson, M. Pihlatie, A. Regar, H. Cook, and G. Cadisch. 2003.
Nitrous oxide emissions following application of residues and fertiliser under zero
and conventional tillage. Plant and Soil 254:361-370.

Balesdent, J ., G. H. Wagner, and A. Mariotti. 1988. Soil organic matter turnover in long-
term ﬁeld experiments as revealed by carbon-13 natural abundance. Soil Science
Society of America Journal 52:118-124.

Ball, B. C., A. Scott, and J. P. Parker. 1999. Field N20, C02 and CH4 ﬂuxes in relation to
tillage, compaction and soil quality in Scotland. Soil & Tillage Research 53:29-
39.

Beyaert, R. P., J. W. Schott, and P. H. White. 2002. Tillage effects on corn production in
a coarse-textured soil in southern Ontario. Agronomy Journal 94:767-774.

Bullock, D. G., F. W. Simmons, 1. M. Chung, and G. 1. Johnson. 1993. Growth analysis
of corn grown with or without starter fertilizer. Crop Science 33:112-117.

Caldeira, K., M. G. Morgan, D. Baldocchi, P. G. Brewer, C. T. A. Chen, G.-J. Nabuurs,
N. Nakicenovic, and G. P. Robertson. 2004. A portfolio of carbon management
options. Pages 103-130 in C. Field and M. Raupach, editors. The Global Carbon
Cycle. Island Press, Washington, DC, USA.

CAST. 2004. Emissions and mitigation of agricultural greenhouse gases. in Climate
Change and Greenhouse Gas Mitigation: Challenges and Opportunities for
Agriculture. Council for Agricultural Science and Technology (CAST), Ames,
Iowa, USA.

CT I C. 2004. 2004 National Crop Residue Management Survey. CTIC, West Lafayette,
IN.

Crum, J .R., and HP. Collins. 1995. KBS Soils [Online]. Available at
www.lter.kbs.msu.edu/soil/characterization. W. K. Kellogg Biological Station
Long-Term Ecological Research Project, Michigan State University, Hickory
Corners, MI.

Davidson, E. A., and I. L. Ackerman. 1993. Changes in soil carbon inventories following
cultivation of previously untilled soil. Biogeochemistry 20:161-193.

190

Diaz-Zorita, M., J. H. Grove, L. Murdock, J. Herbeck, and E. Perfect. 2004. Soil
structural disturbance effects on crop yields and soil properties in a no-till production
system. Agronomy Journal 96: 1651-1659.

Doran, J. W. 1980. Soil microbial and biochemical changes associated with reduced
tillage. Soil Science Society of America Journal 44:765-771.

Elmi, A. A., C. Madramootoo, C. Hamel, and A. Liu. 2003. Denitriﬁcation and nitrous
oxide to nitrous oxide plus dinitrogen ratios in the soil proﬁle under three tillage
systems. Biology and Fertility of Soils 38:340-348.

Estavillo, J. M., P. Merino, M. Pinto, S. Yamulki, G. Gebauer, A. Sapek, and W. Corre.
2002. Short term effect of ploughing a permanent pasture on N20 production
from nitriﬁcation and denitriﬁcation. Plant and Soil 239:253-265.

Fox, R. H., J. M. Kern, and W. P. Piekielek. 1986. Nitrogen fertilizer source, and method
and time of application effects on no-till corn yields and nitrogen uptake.
Agronomy Journal 78:741-746.

Fox, R. H., and W. P. Piekielek. 1993. Management and urease inhibitor effects on
nitrogen use efﬁciency in no-till corn. Journal of Production Agriculture 6:195-
200. '

Grant, B., W. N. Smith, R. Desjardins, R. Lemke, and C. Li. 2004. Estimated N20 and
C02 emissions as inﬂuenced by agricultural practices in Canada. Climatic
Change 65:315-332.

Guzha, A. C. 2004. Effects of tillage on soil microrelief, surface depression storage and
soil water storage. Soil & Tillage Research 76:105-114.

Ismail, I., R. L. Blevins, and W. W. Frye. 1994. Long-term no-tillage effects on soil

properties and continuous corn yields. Soil Science Society of America Journal
53:193-198.

Jokela, W. E. 1992. Effect of starter fertilizer on corn silage yields on medium and high
fertility soils. Journal of Production Agriculture 51233-237.

Kaspar, T. C., T. M. Crosbie, R. M. Cruse, D. C. Erbach, D. R. Timmons, and K. N.
Potter. 1987. Growth and productivity of four corn hybrids as affected by tillage.
Agronomy Journal 79:477-481.

Keller, G. D., and D. B. Mengel. 1986. Ammonia volatilization from nitrogen fertilizers
surface applied to no-till corn. Soil Science Society of America Journal 50:1060-
1063.

Kemper, W. D., and R. C. Rosenau. 1986. Aggregate stability and size distribution. Pages
377-382 in A. Klute, editor. Methods of Soil Analysis 1. Physical and

191

Mineralogical Methods Second Edition. American Society of Agronomy,
Madison, Wisconsin, USA.

Lal, R. 2004. Agricultural activities and the global carbon cycle. Nutrient cycling in
Agroecosystems 70: 103-1 16.

Liebig, M. A., D. L. Tanaka, and B. J. Wienhold. 2004. Tillage and cropping effects on
soil quality indicators in the northern Great Plains. Soil & Tillage Research
78:131-141 .

Linn, D. M., and J. W. Doran. 1984. Effect of water-ﬁlled pore space on carbon dioxide
and nitrous oxide production in tilled and nontilled soils. Soil Science Society of
America Journal 48:1267-1272.

Livingston, G. P., and G. L. Hutchinson. 1995. Enclosure-based measurement of trace
gas exchange: applications and sources of error. Pages 14-51 in P. Matson and R.
Harriss, editors. Biogenic Trace Gases: Measuring Emissions from Soil and
Water. Blackwell Science, Osney Mead, Oxford, UK.

MacKenzie, A. F., M. X. Fan, and F. Cadrin. 1997. Nitrous oxide emission as affected by
tillage, com-soybean-alfalfa rotations and nitrogen fertilization. Canadian Journal
of Soil Science 77:145-152. '

MacKenzie, A. F., M. X. Fan, and F. Cadrin. 1998. Nitrous oxide emission in three years
as affected by tillage, com-soybean-alfalfa rotations, and nitrogen fertilization.
Journal of Environmental Quality 27 :698-703.

Martel, Y. A., and E. A. Paul. 1974. Effects of cultivation on organic matter of grassland
soils as determined by fractionation and radiocarbon dating. Canadian Journal of
Soil Science 54:419-426.

Martens, D. A. 2001. Nitrogen cycling under different soil management systems.
Advances in Agronomy 70: 143-192.

McConkey, B. G., D. Curtin, C. A. Campbell, S. A. Brandt, and F. Selles. 2002. Crop and
soil nitrogen status of tilled and no-tillage systems in semiarid regions of
Saskatchewan. Canadian Journal of Soil Science 82:489-498.

Mehdi, B. B., C. A. Madramootoo, and G. R. Mehuys. 1999. Yield and nitrogen content
of corn under different tillage practices. Agronomy Journal 91:631-636.

Mosier, A., J. A. Delgado, and M. Keller. 1998. Methane and nitrous oxide ﬂuxes in an
acid oxisol in western Puerto Rico; effects of tillage, liming, and fertilization. Soil
Biology & Biochemistry 30:2087-2098.

Mosier, A. R. 2001. Exchange of gaseous nitrogen compounds between agricultural
systems and the atmosphere. Plant and Soil 228:17-27.

192

Niehues, B. J ., R. E. Lamond, C. B. Godsey, and C. J. Olsen. 2004. Starter nitrogen
fertilizer management for continuous no-till corn production. Agronomy Journal
96:1412-1418.

Perfect, E., B. D. Kay, W. P. K. van Loon, R. W. Sheard, and T. Pojasok. 1990. Rates of
change in soil structural stability under forages and corn. Soil Science Society of
America Journal 54:179-186.

Pinto, M., P. Merino, A. del Prado, J. M. Estavillo, S. Yamulki, G. Gebauer, S. Piertzak,
J. Lauf, and O. Oenema. 2004. Increased emissions of nitric oxide and nitrous
oxide following tillage of a perennial pasture. Nutrient Cycling in
Agroecosystems 70: 13-22.

Rhoton, F. E. 2000. Inﬂuence of time on soil response to no-till practices. Soil Science
Society of America Journal 64:700-709.

Rice, C. W., M. S. Smith, and R. L. Blevins. 1986. Soil nitrogen availability after long-
terrn continuous no-tillage and conventional tillage corn production. Soil Science
Society of America Journal 50:1206-1210.

Robertson, G. P., K. M. Klingensmith, M. J. Klug, E. A. Paul, J. R. Crum, and B. G.
Ellis. 1997. Soil resources, microbial activity, and primary production across an
agricultural ecosystem. Ecological Applications 7:15 8-170.

Robertson, G. P., E. A. Paul, and R. R. Harwood. 2000. Greenhouse gases in intensive
agriculture: Contributions of individual gases to the radiative forcing of the
atmosphere. Science 289:1922-1925.

Saxton, A. M. 1998. A macro for converting mean separation output to letter groupings in
Proc Mixed. Pages 1243-1246 in Proc. 23rd SAS Users Group Intl. Sas Institute,
Nashville, TN.

Six, J ., E. T. Elliott, and K. Paustian. 2000. Soil structure and soil organic matter: 11. A

normalized stability index and the effect of mineralogy. Soil Science Society of
America Journal 64:1042-1049.

Six, J ., S. M. Ogle, F. J. Breidt, R. T. Conant, A. R. Mosier, and K. Paustian. 2004. The
potential to mitigate global warming with no-tillage management is only realized
when practised in the long term. Global Change Biology 10:155-160.

Smith, P., K. W. Goulding, K. A. Smith, D. S. Powlson, J. U. Smith, P. Falloon, and K.
Coleman. 2001. Enhancing the carbon sink in European agricultural soils:
including trace gas ﬂuxes in estimates of carbon mitigation potential. Nutrient
Cycling in Agroecosystems 601237-252.

Vetsch, J. A., and G. W. Randall. 2000. Enhancing no-tillage systems for corn with

starter fertilizers, row cleaners, and nitrogen placement methods. Agronomy
Journal 92:309-315.

193

Vetsch, J. A., and G. W. Randall. 2002. Corn production as affected by tillage system and
starter fertilizer. Agronomy Journal 94:532-540.

Vetsch, J. A., and G. W. Randall. 2004. Corn production as affected by nitrogen
application timing and tillage. Agronomy Journal 96:502-509.

Vyn, T. J ., and B. A. Raimbault. 1993. Long-term effect of ﬁve tillage systems on corn
response and soil structure. Agronomy Journal 85:1074-1079.

West, T. O., and W. M. Post. 2002. Soil organic carbon sequestration rates by tillage and
crop rotation: a global data analysis. Soil Science Society of America Journal
66:1930-1946.

Yamulki, S., and S. C. Jarvis. 2002'. Short-term effects of tillage and compaction on

nitrous oxide, nitric oxide, nitrogen dioxide, methane and carbon dioxide ﬂuxes
from grassland. Biology and Fertility of Soils 36:224-231.

194

   

IIIIIIIIIIIIIIIIIIIIIIIIIIII