at
.1394!
5’4...‘

.19.?
. .w’rgz

Va
a
x

r...

35.1512...
’1!

.5-
. O:

x a
.l...).3:
. ‘ .533:
, 532.45..
.‘ .‘ u
I...

.93:
. n 2 a... a»: ‘
g :3... N31...

1..
.! 55. .\
$1119.59: .J.
rltavcavi

.11}. 1:21.?)
3-!

I.
.1} ,li.2.
:luc. 31.1))»

5:.
.VAAxliwu...

l‘ .79... .a... I
< $332.31.:
A r:

 

 

>l

 

LIBRARY
Michigan State
University

 

 

 

This is to certify that the

dissertation entitled

Effects of Low—Intensity Prescribed Fire
on Fine Roots of Red Pine

presented by

Joseph Daniel Zeleznik

has been accepted towards fulfillment
of the requirements for

#11.— degree in M

 

Major professor

Date. Auggst 6, 2001

MSUI: an Affirmative Action/Equal opportunity limitation 0.127."

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

 

 

‘_i

 

 

 

 

 

 

 

 

 

 

 

 

”1’14

 

 

 

 

 

6/01 c:/ClRClDateDue.p65-p.15

 

 

EFFECTS OF
LOW—INTENSITY PRESCRIBED FIRE
ON FINE ROOTS
OF RED PINE
By

Joseph Daniel Zeleznik

A DISSERTATION
Submitted to ’
Michigan Sate University
in partial fulfillment of the requirements
for the degree of
DOCTOR OF PHILOSOPHY
Department of Forestry

2001

Abstract
EFFECTS OF
LOW-INTENSITY PRESCRIBED FIRE
ON FINE ROOTS
OF RED PINE
By

Joseph Daniel Zeleznik

The above ground ecological effeCtS 0f bOIh wildfire and Prescribed fire are fairly
well characterized while belowground responses 0f the ecosystem are just beginning to
be elucidated. The goal of this study was to determine the CffCCtS Oflow-intensity

prescribed fire on soil nutrient dynamics and fine root production and turnover in the top
25 cm of mineral soil in a stand of red pine (Pinus resinosa Ait.) in southern Lower
Michigan. The experiment was set Up as a randomized complete block, with plots split
based on soil depth. Soil and root dynamics were followed over time using a repeated
measures analysis. Root characteristics were determined using both coring and
minirhizotron techniques and turnover was calculated using several different methods. A
new method of calculating fine roor turnover, using both production and mortality values
as a proportion of average annual Standing crop, is proposed. The effects of fire on
diameter growth were determined. An additional experiment on the thermal death point
of fine roots of red pine was performed to determine if these roots are killed at 60°C, the
temperature that is often quoted as the ultimate thermal death point. Contrary to
expectations, soil cation concentrations did not increase after fire but rather decreased

substantially in both burned and unburned (control) plots; some factor other than fire was

controlling nutrient dynamics. Burning also had no effect on fine root production and

 

turnover. However, there were large fluctuations in standing crop of fine roots at

different depths. It appeared that soil water availability played a larger role than did fire

in controlling root dynamics. The often-observed spring and autumn peaks in 1‘00t

initiation and growth were not seen in this study. Death of fine roots of red pine began at

approximately 52.50C and followed the generally observed patterns of increased mortality
roots

with increasing time-of-exposure or with increasing temperature. However some
3

remained alive after being exposed to 60°C, indicating that red pine fine roots may be

more thermotolerant than other species and/or tissues.

Dedication

Dedicated to my parents, Donald and Shelli Zeleznik,

and to the memory 0f Carl Ramm, a good friend and mentor.

TABLE OF CONTENTS

LIST OF TABLES. .......................................................................................................... v11

LIST OF FIGURES .................................................................................................

CHAPTER 1

LITERATURE REVIEW .......................................................................
Introduction ...............................................................................

Nutrients...

Nutrient losses from the site ..................................................

Nutrient gains......
pH changes...
Repeated burning .
Mycorrhizae and other soil microbes
Foliar nutrients... ...
Fine root turnover and carbon allocation...
‘ ‘Fine’ ro.ots..
Fire and roots .,
Fire and overstory growth
Thermal death point...
Literature Cited

CHAPTER 2

EFFECTS OF FIRE ON AVAILABLE NUTRIENTS...
Introduction...
Site description.mm.....................................
Methods.......
Results.......

Literature Clted

CHAPTER 3

EFFECTS OF FIRE ON FINE ROOTS AND STEM GROWTH...
Introductionw ....

Methods...

ReSults... .
DiScussion , .

Literature cited...

......... ix

........ l
........ 2
........ 3
........15
......16
......18
......22
.....25
.....25
......32
.....34
......38
...43

.....62
......63
.....63
Discussion................................... .. .74

....82

.....86
.....87

......88
..95
.....112
...128

CHAPTER 4

ROOT THERMAL DEATH POINT .......................................................................... 135
Introduction .................................................................................................... l 36
Methods, . . . . ...................................................................................................... I 36
Results ............................................................................................................ .141
Discussion........................ ................................................................................ 147
Literature Cited ................................................................................................ 154

APPENDIX

Data for lO-25crn depth, not statistically analyzed ............................................ 159

vi

Table 1'1 .
Table 1‘2'
Table 1'3-

Table 2-1-
Table 2—2.

Table 2-3.

Table 2-4-
Table 2-5.

Table 2-6.

Table 2-7,

Table 2-3.

Table 2-9.
Table 2-10.

Table 2-1 1.
Table 3-1.

Table 3-2.

LIST OF TABLES

Net primary production (N PP) values of fine roots of various
coniferous forests ....................................................................................... 29

Various calculations used in determining annual fine root
turnover .................................................................................................... .30

Direct estimates of median root longevity as obtained by various

rhizotron techniques ................................................................................... 31
Dates on which soils and roots were sampled ........................................... 65
Weather, fuel and fire characteristics, June 1997 ...................................... 66
Prefire soil cation p-values ......................................................................... 68
Postfire soil cation p-values ....................................................................... 71
Postfire FF p-values ................................................................................... 73
Red pine FF weights and cation concentrations from the literature .......... 73

Potential amounts of cations released over the course of two postfire
growing seasons, from burned litter and decomposed duff ....................... 74

Total amount (standard error) of available nutrients in the top 10cm
of mineral soil (average of all prefire samples), and the potential
amount of cations released by the fire and subsequent breakdown

of the forest floor, as a percentage of that total .......................................... 74
Coefficients of variation (CVs) for various nutrients in the soil ............... 79
Variability of foliar nutrient concentrations .............................................. 80

Changes in nutrient concentrations following individual prescribed
fires in a variety of ecosystems .................................................................. 8l

Equations used in comparison of methods of calculating fine root
turnover ...................................................................................................... 93

Weather, fuel and fire characteristics, June 1997 .................................... .94

vii

Table 3-3.

Table 3-4-

Table 3‘5-

Table 3'6'

Table 3‘7-

Table 3-8-

Table 3-9-

Table 3—10-

Table 3-11-

Table 4-1.

Table 4-2,
Table 4-3.
Table 4-4.

Table 4.5-

General timetable of samples ..................................................................... 95

AN OVA results for root lifespan, as directly estimated from

MR images .............................................................................................. .98
ANOVA results for root length production and mortality calculated
fiom MR data .......................................................................................... 100

AN OVA results for live and dead fine roots, core data ......................... .107

Mean (standard error) May 1997 prefire fine root mass (Mg/ha)
and root mass density (g/m3) among the different depths ........................ 108

Significant changes in live and dead fine roots during 1997 and

1 998 used to calculate NPP ...................................................................... 111
ANOVA results of ring width data .......................................................... 1 11
Comparison of turnover values in a variety of ecosystems, as
calculated by a variety of methods ........................................................... 1 14
Comparison of root mass (Mg/ha) in the forest floor and the mineral
soil .......................................................................................................... .119
Time-temperature combinations that were tested (X) ............................. 138

Results of the logistic regression analysis of survival of fine roots
one week following a hot water treatment ............................................... 143

Results of the logistic regression analysis of survival of fine roots
two weeks following a hot water treatment ............................................. 145

Results of the logistic regression analysis of survival of fine roots
three weeks following a hot water treatment ........................................... 146

Actual time and temperature information at the surface of the
mineral soil during a backfire in a red pine stand .................................... 152

viii

r

7—7—1

Figure 2‘1-

Figure 2‘2-

Figure 2'3-
Figure 2—4
Figure 2‘5-
Figure 2-6-

Figure 3-1-

Figure 352--

Figure 3—3.

Figure 3-4.

Figure 3-5,
Figure 3-6.
Figure 3_7.
Figure 3-8.
Figure 3-9.

Figure 3-10.

LIST OF FIGURES

Residuals of the prefire calcium data ..................................................... .67

Prefire soil calcium concentration, across depths in two

experimental treatments ......................................................................... .68
Prefire H+ by depth ................................................................................... 69
Prefire potassium by depth, time .............................................................. 70
P re- and postfire soil calcium by depth and sample date .......................... 71
Mass of the Forest Floor during the second postfire growing

season ....................................................................................................... .72

Diameter distribution of “new” roots observed in minirhozotron

images through the course of two growing seasons ................................... 96
Soil temperatures during a prescribed backfire in red pine ....................... 97
Soil temperatures during a prescribed backfire in red pine ....................... 97

Median root lifespan, as determined by direct observation of
minirhizotron images ............................................................................... .99

Root production (m of root length/m3 soil volume) between
depths, over time .................................................................................... .100

Root loss (m of root length/m3 soil volume) between depths, over
time.................. ........................................................................................ 10]

Root production (m of root length/m3 of soil volume) compared

to the rainfall of the previous 30 days .................................................... .102
Root loss (m of root length/m3 of soil volume) compared to the
rainfall of the previous 30 days ................................................................ 102
Total number of “New” roots in MR images by depth and month,
all plots and tubes combined .................................................................... 103 ,
Production vs. loss, 0-10 cm .................................................................... 104

ix

Figure 3-l l .

Figure 3-12.

Figure 3-13.
Figure 3-14-
Figure 3-15'

Figure 3-16-

Figure 44 '

Figure 4-2.

Figure 4-3.

Figure 4-4.

Figure 4_5.

Figure 4‘6.

Figure 4-7,

Figul'e 4‘8

Turnover (times per year) at the 0-10 cm depth, as estimated
from direct observations of root lifespans and as calculated from
the methods of Hendrick and Pregitzer (H&P) (1992) and Cheng
et al. (1991) ............................................................................................. 105
Turnover (times per year) at the 10-25 cm depth, as estimated
fiom direct observations of root lifespans and as calculated from
I—Iendrick and Pregitzer’s (1992) method and Cheng et al.’s (1991)
method ..................................................................................................... 106
Mass of live fine roots in the forest floor over time ................................ 107
Mass of live fine roots over time ............................................................. 109
Mass of dead fine roots over time ............................................................ 110
Mass of the forest floor (FF) in various red pine stands .......................... 117
Temperatures at the surface of the mineral soil, 2 cm and 6 cm
depths at one location during low-intensity prescribed burning
of a red pine stand, June 1997 .................................................................. 138
Diagrammatic representation of the cell in which roots were
bathed in hot water ................................................................................... 139
Survival of fine roots 1 week following hot-water treatment .................. 142
Idealized model of root survival after one week, based on the
results of logistic regression ..................................................................... 143
Survival of fine roots 2 weeks following hot-water treatment ................ 144
Idealized model of root survival after two weeks, based on the
results of the logistic regression ............................................................... 145
Survival of fine roots 3 weeks following hot-water treatment ................ 146

Idealized model of root survival after three weeks, based on the
results of the logistic regression ............................................................... 147

Chapter 1
Literature Review

In troduction
The overall goal of this project was to determine the effects of low-intensity

prescribed fire on fine roots of red pine. These effects could be direct, due to the heat of
the fire, or indirect, due to fire’s effects on nutrient status of the soil. Both the direct and
indirect effects may cause a shift in within-tree carbon allocation. The direct effect of
tempermure is a] so related to the amount of time a given tissue is exposed to that
temperature- This literature review covers (1) the effects of fire on nutrient availability,

(2) root distribution, fine root turnover and within-tree carbon allocation, and (3) the

effects of high temperatures on plant tissues.

Nutrients
The ecological effects of both wildfire and prescribed fire have been studied
intensively through the years. While many questions still remain, much has been learned.
This liter ature review will discuss the effects of low-intensity prescribed fire on the
mineral nUtt‘ient content of soils. Slash fires and wildfires will not be discussed here
unless only these types of fires can illustrate specific points.

Results of different studies have varied because of many factors, including: (1)
species, (2) site, (3) amount of accumulated fuel, (4) proportion Of that fix] actually
consumed (mcl therefore fire intensity), (5) season 0f burn, and (6) type 0f fir e. Other
factors include variation in study design, sampling techniques and analytical procedures
(Metz et a1. 1 9 61). Nonetheless, some general conclusions can be reached, and they will
be discussed here. Examples will be given to illustrate the generalities, and a few

exec ' - . . . .
ptlons W111 be discussed. General rev1ews about the effects of fire on $011 nutrients,

k)

and reviews about fire effects in certain ecosystems include Viro (1974), Alban (1977),

Wells et al. ( l 979), Raison (1980), Boerner (1982), Covington and Sackett (1990) and

DeBaIlO (199 1 )-

NUTRIENT LOSSES FROM THE SITE

Specific: nutrients are often grouped by their potential fate: those that are most
easily lost durit‘l g the fire due to volatilization (nitrogen [N] and sulfur [8]), and those that
are not so eaSily lost, including basic cations (calcium [Ca], magnesium [Mg] and
potassium [1(1) and phosphorous [P]. The fates of the non-volatile elements may include

loss via surface runoff, or leaching into and/or through rooting zone (Metz 1954).

Volatilization and Convection
Nutrients can be lost from a site during or following burning by a variety of
mechanisms . Certain elements, especially N and S, can be volatilized at relatively low
temperatures, and they are most easily lost during fires (Wells et al. 1979). Because N is
often the mOst-limiting nutrient on a given site, its loss can have large consequences. In
gener 31’ the amount of loss is related to the intensity of the fire (e.g., Gillon and Rapp
1989); 1033 Of nitrogen is directly related to loss of fuel dry weight at a 1:1 ratio (Grier
1975; Raison et al. 1985a, b).
Nutrie tits may also be lost by convection of the ash during the fire (Boerner
1982)“ F91” eD-Itz».11nple, Gillon and Rapp (1989) estimated that 31% of P and 54% of K was
lost to the annosphere in a winter burn in a French Mediterranean Pinus halepensis

fore
St' The percentage loss to the atmosphere often follows the order N, K, P, Ca. Loss

b)

til is general}

airman/at;

"1131.1. hm

r.‘ ‘ \
111:: rum:

.1

H~

“4N 1011116. ll

31’?!» ,'
. Ll; 410 1

(3pm)
..tz'xznéwis '

not ecessarilx‘ \O

of N is generally due to volatilization, while loss of K and P results from particulate and
non-particulate mechanisms. Calcium is mainly lost due to transport of ash (Raison et al.
1985b). Raison et al. (1985a) found that ash transport constituted 093.9% of the mass
of the combusted fuel in a Eucalyptus stand, values which they stated were comparable to

those found in the literature.

Nutrient Capture and Return

Nutrients that are volatilized or removed in the particulate matter from a fire are
not necessarily lost from a site. They might be intercepted by the canopy and returned to
the soil via stemflow and throughfall during precipitation. Results have been variable.

DeBell and Ralston (1970) concluded that only minor amounts of N released from
organic matter by burning are in forms that could be returned to the soil in precipitation.
Kodama and Van Lear (1980) found that in a young unthinned plantation of loblolly pine,
there was some capture of volatilized nutrients in the canopy, which were then released
during rainfall. However, they concluded that the quantities of nutrients intercepted and
released by the canopy were small when compared to nutrient transfer by leaf fall and
precipitation. In a red pine stand that was not burned, Bockheim et al. (1983) also found
that litterfall returned greater amounts of nutrients to the soil surface than did throughfall
and stemflow.

However, Lewis (1974) found that cation concentrations in rainfall plus fallout
were about twice as high on burned plots than on unburned plots in mixed slash and
loblolly pine. There were no significant differences between treatments for N and P,

although the burn plots had consistently higher concentrations of these nutrients.

y.

,‘fi .
111.111 mga so)

11mm lb.

.fmat/‘Izndfrt

Abfné/Izs
axe-cling this im‘

swam Q
resented b.

.@z/‘//‘”’§/‘?{;

-»-~ra®*\§\\'§

A

\\ \\ .
\‘ \\\

Following a series of old-field burns in southern Ontario, Smith and Bowes (1974)
determined that approximately 30% of the nutrients lost from biomass during burning
was recovered in downwind deposits of fly-ash adjacent to the burned areas. When burn
temperatures are relatively low, as in their study, the loss via fly ash is more important

than the loss from volatilization.

Runoff and Erosion

Nutrients can be. lost from a site following fire via surface runoff. Factors
affecting this include topography, amount of cover left after a burn, the intensity of a
given rainfall event and the time elapsed since the fire. However, low-intensity
prescribed burning often results in little or no change in the amount of nutrients found in
runoff. For-example, during the first winter following a prescribed burn in Chaparral,
DeBano and Conrad (1978) found that only trace amounts of N and P were lost in runoff
water. In loblolly pine, two low-intensity prescribed fires did not increase the amount of
runoff to, or the nutrient concentrations of, streams draining the watersheds (Van Lear et
al. 1985). Sediment export, a measure of water quality, also did not increase following
buming (Douglass and Van Lear 1983). Wright (1976) measured minimal nutrient
increases to two lakes during the first growing season after a wildfire in Minnesota, even
though runoff increased by 30 to 60%. The fire occurred in the spring, and he postulated
that results would have been different had it taken place in the fall. Specifically,
following a fall fire, revegetation of the site does not begin until the following spring.
Even when nutrient losses do occur from runoff and erosion, they often do not represent a

significant proportion of total site nutrients (Tiedemann et al. 1979).

Timing and intensity of a given precipitation event is important in determining the
specific amounts and pathways of nutrient movement. For example, following-a wildfire
in a mixed conifer stand in California in August 1960, Johnson and Needham (1966)
followed ionic composition of a stream running through the watershed. They expected to
see large increases in ionic concentrations during the following spring, but instead found
no specific effects of the fire. They concluded that ash constituents were dissolved by
light rainfall and leached into the soil before the first snow. In the acidic soils, the
cations were adsorbed on the exchange complex rather than washed directly into the
stream.

Nutrient loss following fires can also occur via surface erosion. The factors
affecting this are essentially the same as those that affect runoff. However, as long as no
surface soil is exposed, erosion losses will be negligible. In the Georgia Piedmont, soil
movement was negligible after a single burn and afier one repeat burn in a loblolly pine
stand (Brender and Cooper 1968). Biswell and Schultz (1957) observed the results of
exceptionally high rainfalls on sites of varying slopes and different ages since being
burned. They concluded that there was no indication of runoff or erosion that could be
related to the burning itself. Annual spring burns, applied for 1 to 3 years in an oak-
hickory forest in Wisconsin, did not result in increased overland flow or sedimentation
(Knighton 1977). However, Sampson (1944) found increased soil erosion following
burning in Chaparral, especially when slopes exceeded 40%.

Erosion following fire can also indirectly affect nutrient availability to plants by
causing losses of mycorrhizal inoculum. For example, Amaranthus and Trappe (1993)

found that topsoil that had eroded from the soil surface (which was captured in sediment

Elinollou

,. T1,”. .
talc-431501

I \« l '
nilfhmg a
\u
' V:
{pr

“1“ ml:

' n/
M” ”I’d/1
Al‘é’géfifl 501/5.

no 03W elemQt

\\\\R\\ \\ fir Q T

I-ll .

.. w,
1L‘l.”'.{1nn r
I ‘ ‘

A]: d

. «“‘K§\\\XV

} 7‘
“Y:

traps) following a wildfire had relatively high VAM inoculum potential, while the

residual soil had low potential.

Leaching (below rooting zone)

Nutrient losses below the rooting zone depend on several factors, including which
specific nutrient is being measured, soil type, and amount of postfire vegetation available
to take up any increases in available nutrients. Stark (1977) found net losses of Ca and
Mg when soil surface temperatures exceeded 300 °C during burns in Douglas-fir stands;
no other elements were lost from the soil due to burning. Over 90% of nutrient cations
(which were released from the ash) were retained in the top 19cm of mineral soil
following a wildfire in a mixed conifer ecosystem (Grier 1975). Similarly, Knighton
(1977) found that three annual spring burns in an oak-hickory forest only slightly
accelerated nutrient loss below 150m. Except for Ca and Mg, precipitation inputs easily
compensated for these losses. These results and those of Lewis (1974) indicate that fire
apparently increases the solubility of the major nutrient cations in the following order:
Ca2+ > Mg2+ > K+. However, in the New Jersey Pine Barrens, Boerner and Forman
(1982) found that as a percentage of inputs from rainfall, potassium losses below the
rooting zone were greater than losses of Mg or Ca. They also found that rates of mineral
output were inversely proportional to biomass and forest floor mass, which in turn
depended on fire intensity and time of recovery. In a pine savannah in Belize, Kellman et
al. (1985) found large increases of most elements in the rooting zone immediately
following fire, but these effects disappeared after one week. Despite much percolation,

there were no comparably large increases deeper in the soil.

NUTRIENT GAINS
Nutrient gains following fire are observed in the mineral soil, with the bulk of the
nutrients coming from what were the organic soil horizons. These gains can also be

viewed as on-site nutrient re-distribution (c.f. Wienhold and Klemmedson 1992).

Organic matter changes

Changes in soil nutrient status are initiated in the forest floor (FF), as this is the
fuel consumed by fire. Low-intensity fires usually consume the litter (L) layer and
sometimes part of the fermentation (F) layer (duff). The humus (H) layer is usually not
consumed in low-intensity fires, but extremely hot burns may affect this layer.
Experience with red pine in the Lake States has shown that typically only the surface
litter is removed during a single low-intensity prescribed burn (Dickmann 1993). Annual
or biennial burns may destroy both the L and F horizons in red pine (Alban 1977). The
fuel arrangement characteristics of certain species (e.g., ponderosa pine) may result in a
fire that smolders for a long time, which can burn well into the deep duff (Covington and
Sackett 1984, 1986; White 1986).

Organic matter distillation starts at ZOO-315°C, but substantial organic matter
(OM) loss can occur at lower temperatures (DeBano et al. 1998). Organic matter content
of the mineral soil may be decreased by fire, but usually it is not affected or is slightly
increased. Moehring et al. (1966) found no changes in OM concentrations in the top 4
cm of mineral soil following 9 years of annual prescribed burning in loblolly pine.

However, increases in soil OM have been found in a few instances. For example, Wells

[[3, f‘ .
114,523”.

N~

4152224 £092 4 ‘

“Kym

§;\E

(1971) found that a series of burns on the South Carolina Coastal Plain over 20 years
caused a shift in the distribution of OM from the forest floor to the top 5cm of the mineral
soil. Similar results have been found in other southern pine stands (Metz et al. 1961,
McKee 1982, McKee and Lewis 1983) and in red pine (Alban 1977). However, a severe
fire in a jack pine ecosystem resulted in lower soil OM in the top 2cm of mineral soil
during the first 3 months following burning (Smith 1970). Smith also found that organic
matter concentrations increased above pre-fire levels 10 to 15 months following bunting,
at several depths, indicating evidence of the percolation of organic colloids through the
soil.

This increase in soil OM can result in increased nutrient availability indirectly by
increasing the cation exchange capacity (CBC) of the soil. This increased CEC helps to
reduce leaching losses following fire. However, Wells and Davey (1966) suggest that the
forest floor may have a higher total CEC than several inches of underlying mineral soil,

especially in sandy forest soils.

Decomposition and nutrient release

The rate of decomposition of the residual forest floor may increase following
burning (e.g., loblolly pine, Schoch and Binkley 1986; ponderosa pine, Covington and
Sackett 1984, Monleon and Cromack 1996). However, this is not always the case. For
example, in a litter-bag study in northeastern Minnesota, Grigal and McColl (1977) found
no differences in litter decomposition rates of quaking aspen and aster between burned
and unburned areas. They conducted this study for three years following the fire and

attributed their results to compensating factors. That is, higher surface temperatures

1111111111.; in
ruins :1
11111111111. 1‘
1:1-1:111:11
Ra
pracnbcd 1

31.4731? [Min
4, 12115513 6";

1mm am

QM <1

following fire would be expected to increase decomposition rates, but lower surface
moisture would retard it. Stark (1977) tracked litter decomposition in Douglas-fir forests
following burning, and concluded that although burning slightly increased
decomposition, the difference was not usually statistically significant.

Raison (1980) recommended monitoring changes in forest metabolism following
prescribed burning as an indicator of nutrient turnover rates. Following this, White
(1986) found a decrease in soil C02 evolution (a proxy for rate of decomposition) for up
to 10 weeks after a prescribed fire in ponderosa pine, even though N-mineralization
increased during this time.

OM decomposition rates should be inversely related to the C/N ratio of the
organic substrate. After 30 years of annual or periodic burning in loblolly/longleaf pine,
Binkley et al. (1992) found an increasing C/N ratio as burning interval decreased. These
were the same sites on which Bell and Binkley (1989) found decreasing N mineralization
with decreasing fire-return interval. Eight to 65 years of annual or periodic burning in the
Gulf and Atlantic Coastal plains resulted in a higher C/N ratio than that found in control
plots (McKee 1982). Moehring et al. (1966) found that 9 years of annual burning on
loess soils slightly increased the C/N ratio in the top 10cm of mineral soil, but biennial
bunting slightly decreased the ratio.

In addition to the C/N ratio, the “quality” of the organic N substrate will have an
effect on N mineralization rates. It does appear that repeated burning affects OM quality
and subsequent N mineralization, as suggested by Vance and Henderson (1984). They
found that 30 years of annual and periodic (4-yr) prescribed burning in an oak-hickory

forest reduced the amount of mineralizable N. More recent studies are beginning to

10

1:1578‘2‘04

11111.1
1111311111

t.“
“it.

‘ t
'mlcul.
., ‘
:14 I.)
41.11511)“

411 “/t' 5 0,.

(mm
Qittr
’33:. 5:. J.

t , )
y%’(7/é ,

‘ am

is,

“Vt

determine the mechanisms behind these changes. Guinto et al. (1999) found lower N
mineralization in plots that had been repeatedly burned in eucalyptus stands in Australia.
Using NMR spectroscopy, they found that this change was paralleled by a shift in the
OM structure from carbohydrates (easier to break down) to more waxes and cutins
(difficult to break down). Eivazi and Bayan (1996) studied the same site used by Vance
and Henderson (1984). They found that the soil in burned plots had significantly reduced

activities of certain soil enzymes that are important in the cycling of N, P, and S.

Cations

Often, the availability of basic cations in the mineral soil increases following fire
(e.g., St. John and Rundel 1976). The nutrients that were locked up in the slowly
decaying organic material of the forest floor are released by the fire and available in the
ash (Wells et al. 1979). The nutrients then dissolve in rain water and percolate down
through the soil profile. In some cases, the released nutrients result in a “pulse” that
works its way down through time (e.g., Richter et al. 1982). This pulse of available
nutrients may be short-lived, however, as cations may be adsorbed on the soil exchange

complex, taken up by microbes, or rapidly utilized by vigorously-growing plants.

Nitrogen

Nitrogen in particular has drawn much attention, as it is often the limiting nutrient
on a given site and is most easily lost by burning. The total amount of N lost is
proportional to the amount of fuel burned/fire intensity (e.g. Gillon and Rapp 1989, Grier

1975). However, losses in the upper organic horizons (where the litter is fresh and has

l
\

5.10:]
\‘A\" \’ {h
u \ L

\

111,110] (7‘.

111113301. ‘

t“ l

u: '
11111121
1

““11

‘ r
'9

all. 1
r.. I“ .
hint“.
\

F0]

We». 1'0. _

Mable \' in
i Q

mixing a“

J."
LII. '
‘71-’31 z)
5
“
7" .
‘* t1» -
9
@5161]
1;" FJ‘
\< ’1

barely begun to decompose) can be offset by gains in either lower organic horizons
(Mroz et al. 1980), or by gains in the mineral soil (Klemmedson et al. 1962). That is,
although the total amount of N may decrease, there is an increased concentration in the
residual material (Knight 1966), and the amount of available N usually increases
following fire (St. John and Rundel 1976, Wells et al. 1979).

For example, Ryan and Covington (1986) found that soil NI-If-N significantly
increased during the summer after a fall prescribed burn in a ponderosa pine stand,
although NO3'-N did not increase. Other mechanisms that contribute to the increase in
available N include increased N-mineralization, nitrification and N-fixation (both

symbiotic and non-symbiotic).

Nitrogen Mineralization

Schoch and Binkley (1986) measured the amount of N in the forest floor of a
mature loblolly pine stand. They found that the total amount of N in the forest floor was
not significantly reduced by low-intensity prescribed burning, but the decomposition rate
of the forest floor more than doubled for the first growing season after burning, releasing
60 kg N ha'1 more than was released in the unburned portion of the same stand. White
(1986) also found increased N-mineralization and nitrification following prescribed
burning in ponderosa pine.

The timing of sampling following burning can affect the results of prescribed-
burning experiments. For example, following a slash fire in a clearcut Douglas-
fir/westem larch stand, there was an immediate increase in the concentration of NH4+, but

nitrification showed a 3-week lag following the fire (Jurgensen et al. 1981). Similar

12

results have been observed following surface fires in a ponderosa pine pole stand in New
Mexico (Kovacic et a1. 1986), in Texas rangelands (Shanow and Wright 1977), and in
chapparal in Arizona (Wienhold and Klemmedson 1992). Lab experiments involving red
pine litter (Mroz et al. 1980) or soils from Eucalyptus stands (Jones and Richards 1977)
gave comparable results. Although spring burning increased the amount of soil NHX,
Sharrow and Wright (1977) found decreased soil nitrate levels following bunting. They
attributed this to rapid uptake of nitrate by the vigorously-growing plants on these plots.
That is, they believed that they missed detecting a nitrate pulse because of the timing of

their sampling (3 weeks following burning).

SYmbiotic N-fixation
Whether N-fixation rates are affected by fire depends on the response of
leguminous plants to fire. In some cases there has been in increase in the numbers of
legUDIes following fire (Chen et a1. 1975). These species are adapted to invade and
Survive on sites that may be low in nutrients, e. g. sites that have experienced severe
Wfldfires. The amount of N fixed may not be enough to replace that which was lost by
fir e. For example, Hamilton et al. (1993) found that the amount of N fixed by Acacia
Species over 27 months following a prescribed fire in Eucalyptus was small. More N was
103‘ t0 the atmosphere during the burn than was gained by fixation. However, they

attributed these findings to the low density of Acacia plants in their study area, not to low

fixation rates per plant.

w

mmsmhiott'c 1w

    
  

Me ro/e o/ ‘

£335, “arms 61 L

 

 

Non-symbiotic N-fixation
The role of non-symbiotic nitrogen fixers following fire has been studied

extensively. Most of these studies have measured acetylene reduction (an indication of
nitrogenase activity) in soil or forest-floor samples. Maggs and Hewett (1986) found that
a single prescribed burn in a slash pine plantation in Australia increased nitrogenase
activity in the forest floor by about 3 times at 18-30 months post-fire. They attributed
this increase to the increase in soil pH and calcium concentration. Studies of repeated
burning have given varied results. For example, Vance et al. (1983) found that annual
and periodic (4-year) burning in an oak-hickory forest had no influence on nitrogen
flXation ratesor proportion of samples displaying activity. However, J orgensen and
Wells (1971) found that annual burning (for over 20 years) in loblolly pine resulted in a

ten-fold increase in N-fixing ability compared to samples from unbunted areas.

Gr ee11120 use assays
While the increase in available nutrients following fire is usually measured
directly via soil analysis, it can be measured indirectly using greenhouse plant-growth

aSSaYS. Vlamis et al. (1955) grew barley and lettuce in soils that had been collected from
pr escribed-burned ponderosa pine stands. They found that burning increased the N- and
P‘supply of these soils. The effect was more pronounced on areas that had been burned
one year before the test, compared to those that had been burned two years prior. Vlamis
and (30me ( 1 961) also used a greenhouse pot test to study the effects of burning on
several 1"nations of crops. They found that burning increased the yields from the first

plantings 0f lettuce and barley compared to controls. However, the differences observed

14

l“

 

in the first planting had largely disappeared in the second planting. Wagle and Kitchen
(1972) also used a greenhouse bio-assay to determine the effects of burning on soil
nutrient availability. They used lettuce plants and ponderosa pine seedlings in their
study, with similar results to the earlier studies. However, because of the different
nutrient requirements of lettuce plants and tree seedlings, they cautioned about
extrapolating results from the lettuce bio-assay to forest growth.

Caution must be used when interpreting results from these greenhouse
experiments. Some long-tenn studies of the effects of prescribed burning indicate
reduced growth of residual trees (e. g., Boyer 1993), and that has been attributed to a
lOWer water-holding capacity of the soil (Boyer and Miller 1994). Thus, lower moisture
availability may be more important than the increased nutrient availability following
burning. Greenhouse assays avoid the water-holding capacity problem by using optimal

Watering regimes.

PH CHANGES

Soil pH has been found to increase following fires in a variety of ecosystems
(e.g,, Grier 1 975 — single wildfire) although many investigators have reported no
Significant change in pH following a single fire (Christensen 1977, Kovacic et al. 1986,
Ryan and Covington 1986, Masters et al. 1993). An increase in the concentration of the
basic cations in the ash is usually associated with the increased pH (Alban 1977). This is
important in many forest types because of the acid soils on which they grow. Besides the
increased availability of the basic cations, the increased pH affects the availability of

other nutrients, specifically phosphorous, which is most available at a pH of 6.5 (Brady

15

1990). However, most of the reported increases have been slight and have not resulted in
pH values this high.

Several studies have measured pH changes in a variety of ecosystems that have
received repeated burns (annual or periodic) over the course of 8 to 65 years (Lunt 1950,
Metz et al. 1961, Wells 1971, Alban 1977, McKee 1982, Binkley et al. 1992, Eivazi and
Bayan 1996). Most of these report either no change in pH, or only a slight increase in the

most superficial soil layers (0-5cm). The pH of deeper soil layers is usually not affected

by bunting.

REPEATED BURNING:
The effects of one fire are different from the effects of repeated burning over a
longtime. For example, some studies indicate an increase in available nutrients
f0“OWing an initial burn, but a decrease or no change following a series of burns (e.g.,
Hunt and Simpson 1985). An initial study by Covington and Sackett (1986) was
condttcted in ponderosa pine stands that had many years of fire exclusion, which resulted
in e)tttremely high fuel loads. Consequently, there were initial large increases in available
N- HOWever, a series of repeated burns resulted in a decrease in total N (Wright and Hart
1 997), which could have long-tenn impacts on the site.

Many forest stands, especially in the South, are burned at regular intervals to
cOntrol competition and reduce buildup of hazardous fuel loads. Several studies have
ShOWn an increase in available nutrients following a program of repeated burns (68-,
McKee and I..ewis 1983, McKevlin and McKee 1986), or at least no significant changes

in nutrient C3<>Iicentrations (e.g., Boyer and Miller 1994). Binkley et al. (1992) found that

30 years of prescribed burning in loblolly pine reduced the amount of C and N in the
forest floor, but nutrient content of the mineral soil was little affected by burning. There
were no differences for Mg or K between burning and control treatments in the top 20cm
of the soil; however, exchangeable Ca was greater in the 1- and 2-year burn interval
plots. Following a series of annual or periodic summer or winter burns at four sites in the
Atlantic and Gulf Coastal Plains, McKee (1982) concluded that burning consistently

resulted in increased P availability and an increase or no change in soil N.

Nitrogen
In ponderosa pine, Covington and Sackett (1986) found that soil NH: and N03'
Were higher on plots that had been burned annually or biennially. However, plots that
had not been rebumed for 4 to 5 years had concentrations similar to controls. A later
Study on the same site (Wright and Hart 1997) found that NH4+ and NO3' were similar on
bienniaIIy-bumed plots and the controls after 20 years of burning. However, there was a
r eduCtion in total N in the upper 15cm of mineral soil.
Lon g-tenn burning in other ecosystems had similar outcomes. Total N in the
Surface mineral soil has been found to be unchanged or slightly increased by annual or
PefiOdic hurtling in red pine (Alban 1977), loblolly pine (Waldrop et a1. 1987), or in oak-
hickory stanlds (Vance and Henderson 1984). However, Vance and Henderson found that
blurring reduced the amount of available NHf-N, and the amount of mineralizable N,
indicating a change in the quality of the organic N substrate. Similar results were found

for a single burn in ponderosa pine in Oregon (Monleon et al. 1997) where, 12 years after

burning, net N-mineralization was lower than in controls, even though total N pools were
similar.

Season of burn may also have an effect on soil N. McKee (1982) found that
annual or periodic summer burning in southern pines had a detrimental effect on the
amount of N in the surface mineral soil, while winter burning increased N. While McKee
didn’t mention the specific reason for this increase, an increased population of legumes
after winter burns may have been involved. For example, periodic (4- to 5-year)
dormant-season burning in loblolly pine — hardwood stands resulted in increased
frequency, density, and diversity of legumes, compared to sites that had no prior burn
history (Hendricks and Boring 1999). Chen et a1. (1975) also found increased production
0f legumes in loblolly — shortleaf pine stands that were repeatedly burned during the

winter.

MYCORRHIZAE AND OTHER SOIL MICROBES
Fire may have indirect effects on nutrient cycling via effects on soil

microorganisms (Chambers et al. 1986). Soil microbes are important for a number of
I.easons. They decompose organic matter and thus increase the rate of nutrient cycling;
they help to improve soil structure; they can also help plant nutrition directly via
symbiotic associations. Intense fires such as wildfires or slash burns can sterilize the soil,
heating it to temperatures above the lethal threshold for microbes. Indeed, much of the
research that has been done in this area has involved the effects of intense fires on soil

microbes (e - g -, Widden and Parkinson 1975).

Generally, burning has been found to be beneficial or at least neutral to soil
microbes. For example, Fuller et al. (1955) found that light bunting resulted in higher
numbers of bacteria, but lower numbers of fungi. They also found that microbial activity
of soils from the burned areas (as determined by C02 evolution) was lower than that of
unburned controls. Bissettand Parkinson (1980) also found an increase in the number of
bacteria, and reduced fungal flora, following a wildfire in a spruce-fir subalpine forest.
Jorgensen and Hodges (1970) found that prescribed burning in loblolly pine stands had
no effect on the number of fungi per gram of soil, although annual bunting for 20 years
reduced total numbers of fungi through a decrease in the weight of the forest floor.

Earlier studies focused on the general effects of fire on soil microbes, while more
recent studies have divided results based on functional groups or genera. There is a
general trend for heterotrophic microbes to decline following a moderate- or high

intensity fire, although certain autotrophic microbes may increase dramatically above
pr e‘fire levels (Neary et al. 1999). As with studies of nitrogen dynamics, timing of
sampling has an effect on the results of microbial studies. For example, Ahlgren and
AhIgren ( 1 965) found decreased numbers and activity of microorganisms immediately
after Slash burning in a jack pine stand. However, both numbers and activity increased
abruptly to a very high level after the first rainfall following burning. Following a slash
fire in a Douglas-fir/western larch stand, nitrification showed a three-week lag following
the fire (Jurgensen et al. 1981). At six weeks post-fire, the population of nitrifying

bacteria (Ni trosomonas) was significantly higher in both the humus and mineral soil.

l9

,\l}'corrhl-7«3C

As with :
depend on seaso
burning. Most c
three areas: myc
succession of m
leg. Visser 199
titerse fires (\V
literature exists
rttcorrhizae. H
batiiomycetes
Mountains. Th,
ectomycorrhizat
results following
219911 found dc
Northtt‘est Spair

compared to cor

Long-let

tutlotting forest

1].
mitoueh th
\ e Sp:
percent
COVEI‘
of
19831.

. n prairie

o
k prescn

Myconhizae
As with studies of N fixers, results of studies of fire effects on myconhizae may
depend on season of burn, depth sampled, and the amount of time that has elapsed since
burning. Most of the literature dealing with the effects of fire on myconhizae is from
three areas: mycorrhizal associations following grassland (prairie) burning, the
succession of myconhizal species following intense fires such as wildfires or slash burns
(e.g., Visser 1995), and the myconhizal colonization of tree seedlings regenerating after
intense fires (Wright and Tanant 1958, Miller et al. 1998, Horton et al. 1998). Little
literature exists on how low-intensity underburning in pine stands will affect
myconhizae. However, Palmer et al. (1994) found increased fruiting of ectomyconhizal
basidiomycetes after prescribed burning in pine/hardwood stands in the Allegheny
Mountains. Their results suggest that there was minimal damage to living
CCtOmyconhizae during a February prescribed burn. Horton et al. (1998) found similar
r eSults following a wildfire in bishop pine in Califontia. However, Vilarifio and Arines
(1991) found decreased densities of VA myconhizal propagules following wildfire in
NOFthwest Spain. This was paralleled by lower VA colonization of post-fire grasses
compared to control plots.

Long-tenn, successional changes in the myconhizal community are often found
follOWing forest wildfires (e.g., Visser 1995), paralleling changes in the plant community.
Although the species composition changed following a severe fire in rangeland, the
percent COVer of myconhizal species was approximately equal (Pendleton and Smith
1983). In prairie ecosystems, the myconhizal community usually doesn’t change

following Prescribed burning (e.g., Gibson and Hetrick 1988).

20

Dhillion
Earle blueslem d
22: year hotWVC'
{11111051 [31211115
Bentu'enga and
an increase in 3'
:ionth. MedW
intensity of VA

Myconl
at eerease leg
The recovery tin
.fie.g..11'right ar.
Miller et al. (19
wildfire depend
were infected St
111:1 white pine
tuned site did
TWO months aft:
mmlcorrhiza
himmg Clean

Dt‘t‘m'
“elawfir see:

i

‘ .ti‘firs after a

litej ‘
ensm
. Of n
t

Dhillion et al. (1988) found that burning resulted in lower VAM colonization in
little bluestem during the first post-fire growing season. These results didn’t last into the
2"d year however. Their results suggest that with increased nutrient availability following
fire, host plants may not need to form the symbiotic relationship with the fungi.
Bentivenga and Hettick (1991) reported that a spring burn in tallgrass prairie resulted in
an increase in active myconhizal colonization, but the effect didn’t last more than one
month. Medve (1985), however, found that one or two spring burns had no effect on the
intensity of VAM infection on the prairie plant blazing star.

Mycorrhizal colonization or hyphal density may increase (e.g., Hen et al. 1994)
0r decrease (e.g., Buchholz and Gallagher 1982) following a fire. In any case, the longer
the recovery time following fire, the higher the rate of myconhizal infection of plants
(e.g., Wright and Tanant 1958; Malajczuk and Hingston 1981; Wicklow-Howard 1989).

Miller et al. (1998) found that first-year survival of ponderosa pine seedlings following a
Wildlire depended on their infection by ectomyconhizae; that is, only those seedlings that
Were infected survived the first growing season. On the other hand, colonization of red-
and White pine seedlings by ectomyconhizae following outplanting on a harvested and
burned site did not affect survival (Hen et al. 1994). However, they took samples only
two months after outplanting. Fire intensity positively conelated with percent
eCtomYcorrhizal roots for white pine but not for red pine. Similarly, slash burning
fOllowing clearcutting had no effect on the number of infected root-tips of first-year
Douglas-fir Seedlings (Pilz and Peny 1934). Fuller et al. (1955) qualitatively stated that
4‘5 years after a forest fire (presumably a wildfire), there appeared to be no difference in

the density of myconhizae on pine seedlings growing on burned and unburned areas.

21

Greenhouse 3553.1
Many stut
sites that have be
plants grown in
1111116 soils from
also found a red
temperatures dt
HOWCVt
Sehoenberger 2
eetomycorrhiz
53136 than by b
"‘13) a fungus
<greatest affini
and Malajcm
mlCOlThlZae ‘
liner. the ma?

Greenhouse assays

Many studies have involved greenhouse assays, growing seedlings in soils from
sites that have been burned. For example, Wicklow-Howard (1989) found that big sage
plants grown in soils from high-intensity burned areas had an 18% VAM infection rate,
while soils from a low-intensity burn gave a 67% infection rate. Klopatek et al. (1988)
also found a reduction in VAM colonization which was significantly conelated with soil
temperatures during simulated burning of pinyon-juniper litter.

However, following a greenhouse assay with western hemlock and Douglas-fir,
Schoenberger and Peny (1982) found that bunting only minimally affected number of
ectomyconhizal root tips and root weight. These variables were more affected by soil
type than by burning. They concluded that the late seral species hemlock was associated
with a fungus that was relatively sensitive to fire, whereas the pioneer Douglas-fir had the
greatest affinity for ectomyconhizal types that were unaffected by disturbance. Redell
and Malajczuk (1984) concluded that the 90% reduction in brown and white types of
myconhizae following low-intensity burning in Australia was simply due to the loss of
litter, the main site of their formation. Black-type myconhizae were usually found in the

mineral soil and were not affected by bunting.

FOLIAR NUTRIENTS
While an increase in available soil nutrients is often measured following fire,
changes in foliar nutrient concentrations offer evidence of the direct effects of fire on

plant nutrition. Much of the literature on this subject is from studies focused on wildlife

22

mggement. C017
rationing prescri
:‘ullotting prescri
1.33 1957). Ho“
nature. non-spit

Prescribe
ot‘species and et
nutrient content
03121110. Bumir
Ca and P for ch.
consistent differ
fining resultet
Prescribed bUrr
tnliar nutrient i

SChOch
?0\11ng-53a501
351001115 after
lRC‘bfinS and .\
tum ie"€15. \Vh
flamed three I

Foliar r.

management, concerning the effects of fire on nutrient status of browse and forage
following prescribed fire. Most of these studies have found an increase in foliar N
following prescribed burning, present as increased protein (e.g., DeWitt and Derby 1955,
Lay 1957). However, these results may or may not be comparable to those found in
mature, non-sprouting pines.

Prescribed fire has resulted in increased foliar nutrient concentrations in a variety
of species and ecosystems. For example, James and Smith (1977) found increased foliar
nutrient concentrations of 24-42% following burning in a quaking aspen stand in southern
Ontario. Burning in the New Jersey Pine Barrens resulted in increased first-year foliar
Ca and P for chestnut oak, but not for white oak (Boerner et al. 1988). There were no
consistent differences in foliar N between treatments. In a prairie ecosystem, spring
burning resulted in increased concentrations of N and P in May foliage (Old 1969).
Prescribed burning resulted in increased concentrations of all soil nutrients, followed by
foliar nutrient increases in a longleaf pine — wiregrass ecosystem (Christensen 1977).

Schoch and Binkley (1986) also found increased foliar N in loblolly pine one
growing-season after a spring prescribed fire. Foliar concentrations of P and K increased
3 months after a wildfire, whereas foliar N did not, in a red pine stand in Newfoundland
(Roberts and Mallik 1994). Four years later, N levels were significantly lower than pre-
bum levels, while K was significantly higher. Cunent-year foliar P was still only slightly
elevated three years later.

Foliar nutrient concentrations do not always increase following fire, however.

For example, bunting in a prairie ecosystem resulted in lower tissue concentrations of K,

Ca and Mg in little bluestem (Dhillion et al. 1988). However, these results were due to

the dilution eflee‘
interns of total :
iiterertces bet“
effect on foliar 1
1131.1. Landsber
needle mass in
and Miller 199'
and P coneentr
pine. there we
8’. al. 1992).

F ew 5'
increased set
after present
Shortqiyed. I
num‘em mm
3‘~ailable nu

the dilution effect of increased growth on the burned site. When nutrients were expressed
in terms of total amounts of nutrients per unit area of soil surface, there were no
differences between the burned and unburned areas. A single prescribed burn had no
effect on foliar N concentrations (percent) in ponderosa pine (Landsberg and Cochran
1980, Landsberg et al. 1984), but there was a loss of total foliar N (kg/ha) via a loss of
needle’mass in a high-fuel-consumption fire. Periodic burning of longleaf pine (Boyer
and Miller 1994) or slash pine (Hunt and Simpson 1985) also had no effect on foliar N
and P concentrations. After 30 years of annual or periodic bunting in loblolly/longleaf
pine, there were no significant differences for any foliar nutrient concentration (Binkley
et al. 1992).

Few studies have related changes in foliar nutrients to changes in soil nutrients.
Increased soil N availability was reflected in increased foliar N concentrations one year
after prescribed burning in loblolly pine (Schoch and Binkley 1986). This effect may be
short-lived, though, as Hunt and Simpson (1985) found no differences in soil or foliar
nutrient concentrations three years after bunting, even though there were increases in
available nutrients immediately after bunting. Similar results were found by Binkley et
al. (1992) following 30 years of interval bunting in loblolly/longleaf pine. They found no
differences in foliar nutrient concentrations, and the only change in soil nutrient
concentration was increased Ca in the top 10cm of the annual- and biennial-burn plots.
Boyer and Miller (1994) found the same — no significant differences in soil N and P or
foliar N and P following 16 years of biennial burning in longleaf pine. At one of the red
pine wildfire sites studied by Roberts and Mallik (1994), foliar nutrient concentrations

did not necessarily follow changes in soil nutrients. While available N appeared to

24

 

increase slt
Atalable l

otslightly

fnerefore.

tees.

line root
F i

nutrient u

ecoloetez

increase slightly, foliar N concentrations decreased significantly following fire.
Available P also was slightly increased following fire, but foliar levels were unchanged
or slightly lower. Only changes in soil K were paralleled in foliar K concentrations.
Recovering vegetation, usually understory plants, can act as a sink for nutrients
released by a fire (e.g., Stark 1977; Boerner 1983; Harris and Covington, 1983).
Therefore, increased nutrient concentrations may not be found in the foliage of overstory

trees.

Fine root turnover and carbon allocation

Fine roots serve a variety of functions. For trees, fine roots function in water and
nutrient uptake and are the location of production of certain hormones. From an
ecological standpoint, fine roots are important because they are the primary means by
which carbon enters the soil system. Roots had been ignored in studies of tree growth
and carbon allocation until relatively recently. Several studies over the last 20 (or so)
years have begun to illuminate their importance as a sink in within-tree carbon-allocation

studies.

“FINE” ROOTS

Roots are usually divided into size classes by diameters. However, there is no
agreed-upon definition in the literature of what diameter constitutes a “fine” root. Fine
roots are those that have no secondary growth (Marshall and Waring 1985) and which are
shed easily. Diameters most-commonly used as cutoff points between fine- and coarse-

roots include 1mm (e.g., Santantonio and Hermann 1985, Santantonio and Grace 1987,

25

 

and GM
iimeter
roots by

role at it

Sword 1988), 2mm (Persson 1983, Vogt et al. 1983, Marshall 1986, Espeleta and
Eissenstat 1998, Rytter and Rytter 1998), and 5mm (Joslin and Henderson 1987, Haynes
and Gower 1995, Steele et a1. 1997). In most cases, there is no functional reason for the
diameter size class chosen (Vogt and Persson 1991), and studies which do not distinguish
roots by their structural and functional categories will result in a misinterpretation of the
role of roots in ecosystem processes (Vogt et al. 1991).

The 1mm diameter limit may be most appropriate for defining “fine” versus
“coarse” roots. Roots >lmm tend to be secondary roots in which the epidermis and
cortex have sloughed off and which increase in diameter by carnbial growth (Vogt and
Persson 1991). Morphologically, the exposed peridenn of secondary roots of both
balsam fir and white pine is light in color, in contrast with the dead cortex of smaller
roots (Tippett 1982). Using the minirhizotron (MR) technique, Majdi and Persson (1995)
found the average diameter of living white and brown roots of Norway spruce to be
<1mm. In a radiata pine stand in New Zealand, Santantonio and Santantonio (1987)
observed roots of diameters <1, 1-2, and 2-5mm. They found that only roots <lmm
diameter exhibited seasonal cycles of growth, death and replacement. Nambiar (1990)

also reported negligible mortality of radiata pine roots >1mm diameter.

Root distribution

Maximum rooting depth of pines varies from as little as 1m to as great as 24m,
with most species rooting between 2 and 5m (Stone and Kalisz 1991). Most “fine” roots
are found in the top several centimeters of mineral soil, with rooting density decreasing

quickly with depth. For example, Harris et al. (1977) found that 80-90% of the root

26

biomass of yellow-poplar and loblolly pine forests were within 30cm of the surface.
Similar values have been reported elsewhere (e.g., Coile 1937), but values can change
dramatically in soils that have a sandy texture at the surface, with a finer texture at depth
(e.g., Day 1941, Fayle 1975, Van Rees and Comerford 1986). Within the mineral soil,
fine roots are ofien most dense at the interface between the forest floor and the mineral
soil, with density decreasing with depth (e.g., Braekke and Kozlowski 1977, Kimmins
and Hawkes 1978, Ares and Peinemann 1992). Furthermore, some species’ fine roots
extend upward from the mineral soil into the organic material of the forest floor.
Presumably, these roots would be the ones most affected by the heat of a fire because
only 8-10% of the heat from burning is transmitted downward to the soil or litter
(DeBano et al. 1977, Packham 1969, Raison et al. 1986, Hungerford et al. 1991).

The amount of fine root mass located in the forest floor ranges from <1 % to
>50% of total root mass (e.g., Kimmins and Hawkes 1978, Gholz et al. 1986, Vogt et al.
1987). This value is dependent on many things, including species, site (V ogt et al. 1987),
stand age (V ogt et al. 1987), or previous management (McQueen 1973), including fire-
free interval. One report for red pine (McClaugherty et al. 1982) indicated that the forest
floor held approximately 30% of the root mass (<3mm, live + dead) to a depth of 60- to

120cm.

Net Primary Production
Fine root net primary production (N PP) and subsequent turnover is a major
carbon input to forest soils (Steele et al. 1997). Although fine roots might account for

<1% of the biomass of mature forest stands, they can constitute up to two-thirds of

27

.tmuil pr
proluctlo

talculatin

mesh 11

txv
0111518111

1
'- .

{1.1013 [n
a 1.510“
t‘ttn \\

e131,]-

annual production (Grier et al. 1981). Fine root NPP is the difference between total root
production and root mortality (Sutton and Tinus 1983). Several methods exist for
calculating fine root NPP and are reviewed in Vogt et al. (1998).

Estimates of absolute values of annual NPP of fine roots of coniferous trees range
nearly 10-fold (Table 1'—l). It is difficult to compare values directly because of: (1)
different species studied, (2) different definitions of “fine” roots, (3) different depths of
sampling, (4) different site indices (productivities), (5) different stand ages, and (6)
different methods of calculating NPP. However, a few general conclusions are apparent
from the literature. Fine root NPP tends to be higher in coniferous forests of the western
US. than in those found in the Midwest. High productivity sites have higher NPP of fine
roots than low productivity sites (Keyes and Grier 1981). Water availability may serve as
a proxy for these productivity differences (Santantonio and Hermann 1985). However,

even within a site, NPP can vary by as much as two times, between different years (Steele

et a1. 1997).

28

 

Afiofieoeg +v boEo>O

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2% use: so. : on.
Anna: 53.8; 28% wE5> Sm. 3 Ed 350 Eoom EEN oEQ £on
Sacco BEE
Goo S .E 8 $300 wed 8.50 Eomm EEN tcéflmsoo
Boss: 2: 2 _ E
2M3: ._m 6 8:0 Eoéaom mm 0.3 880 Each EEN 63% 3:8;
53 as :9: E
:3: 5:0 98 moxox 52: new 33 9m 850 E0? EEN Ens—ween
63 we
£3: casters: 2
SEES: Ea 2:9:ng in ad 350 Eomm EE~ $-33on
awe: .E E steamsfluuS— eofioE EEaaE m.m 380 E02 EEm 02E com
eofioE EEeaE
No.2 m _ .o
Goo: 5266 can 8:me 33 mad 880 «Eu? EEm oca com
Ammo: ._~N 8 Sn< eofioE EE¢§2 :V I $6 880 E02 EEm oEq com
mozm 03% dofiaa— and .36 880
$3: ._—N 8 203m 8% 93. .3132 omd .vmd EBEwE Eoom EEm 8E v.03.
33m .88 Z flab—“£36 @0522 Emoa BuoEfle SE 36on
A:2 wEEEmm new EEEEwE

 

 

.338.“ mzoeomEoo mzotg mo 92: can we mos—g EAZV :ozozeoa >8th 82 ._-~ 2an

29

lumoter
Rt
roots tf

tees. Ho

pertain.

trained
131mm
CT‘OI‘l and

3:2. not

Tumover/Root longevity

Root turnover is the simultaneous addition to, and loss from, a given population
of roots (Sutton and Tinus 1983). That is, turnover is the concurrent birth and death of
roots. However, there is no agreed-upon standard for calculation of turnover rates (times
per year). Methods for calculating turnover use somemeasure of either production or
mortality as a percentage of “standing crop” (Table 1-2). Some investigators have used
initial standing crop of live roots, mean annual standing crop, or the average of initial and
final standing crop in the denominator. These calculations can be used with biomass data
obtained from coring techniques or estimated from length data from rhizotrons.
Important factors in these methods include starting date (and therefore initial standing
crop) and ending date (temporal extent of sampling). Ideally, data will span an entire

year, not just the growing season.

Table 1-2. Various calculations used in determining annual fine root turnover.

 

 

 

Reference

Hendrick and Pregitzer Tl (Turnover Index) = 2 Mortality

(1992) Initial Standing Crop
Cheng et al. (1991) T1 = E Mortalig

(Initial + Final Standing Crop)/2

 

Rytter and Rytter (1998) Calculated from a third-order polynomial, describing
growth and decay phases of numbers of roots

 

 

 

 

Burke and Raynal (1994) Turnover = Production/Mean biomass

 

Where turnover has been determined using these methods, investigators have

attempted to determine root lifespan, as the inverse of turnover:

turnover = 1/ (median) lifespan. (1)

30

ntzntrh
1.25 it
1- .

[3,.
111 u :p a.
rem \‘

Table

tee 1111‘.

Soil coring techniques generally underestimate production and mortality rates because of
the temporal overlap in these processes (Kurz and Kimmins 1987). Estimates of root
longevity in pine forests, obtained by the sequential coring techniques, are summarized in
Schoettle and F ahey (1994), and range from 73 days to 6.4 years. Using length data fiom
minirhizotrons, Hendrick and Pregitzer (1992) estimated root lifespan to be 168 to 237

days in a sugar maple-dominated forest. However, direct estimates of median root

lifespan, based on MR observations (Table 1-3), are much shorter than those calculated

from weight or length data.

Table 1 -3. Direct estimates of median root longevity as obtained by various rhizotron

 

 

techniques'
species/ Median Root Notes Study
Ecosystem Lifespan (days)
Sitka sprucc >63 l-year-old seedlings Black et al. (1998)
Populus 36 Nine different cohorts
Sycamore 30 combined
Cherry 1 8

Ponderosa pine

 

139 (123-185)

Seedlings; myconhizal
root tips only; not affected

Rygiewicz et al.
(1997)

 

 

 

 

 

 

 

by C02 or N additions
Kiwifruit 328 Season of root initiation Reid et al. (1993)
had no effect
Populus <21 Myconhizal Hooker et al.
>49 Non-myconhizal (1 995)
Citrus 16-51 Shallow, two rootstocks Kosola et al.
23-57 Deep, two rootstocks (1995)
3 cohorts combined
Holrn Oak 35-471 Several depths to 60cm Lopez et al. (1998)
(NE Spain) 85 Unthinned plots
76 Thinned plots

 

Many factors affect the longevity of a given root, including species or cultivar

(Harris et al. 1995, Black et al. 1998, Kosola et al. 1995), myconhizal status (Hooker et

31

 

al. 1995. Hook
plant age (Espt
and Van Rees

1‘fe'ct estimate

that roots that t

FIRE AND RC

Theefl‘
much ofthe m
395 looked at t
othat those Sp
can re-sprout a

The stu
within-plant ca
Plants followir
Some authors ]
tem’een bums
follotn'ng bum
Anderson 199:
mtreased 3110c

gU‘Mh- Rathe

A t3 _
“ >72.)
flue

al. 1995, Hooker and Atkinson 1996), water availability or drought (Marshall 1986),
plant age (Espeleta and Eissenstat 1998), and fertilization (Gower et al. 1992, Eissenstat
and Van Rees 1994, Haynes and Gower 1995). Sampling intensity (over time) may also
affect estimates of root lifespan. The shorter the sampling interval, the more likely it is

that roots that die and decay very quickly will be accounted for.

FIRE AND ROOTS

The effects of fire on roots, especially fine roots, has been little studied, with
much of the work being done in prairie and grassland ecosystems. Much of this research
has looked at the rooting habits of species that sprout after fire. Generally, the conclusion
is that those species and individuals that have deeper roots are better able to survive and
can re-sprout after a fire (e. g., Shearer 1975).

The studies of grasslands and prairies have generally looked at productivity and
within-plant carbon allocation patterns with or without fire. Enhanced growth of prairie
plants following burning has been found repeatedly (see Dhillion and Anderson 1993).
Some authors have found no differences in root production (Melgoza and Nowak 1991)
between burned and unburned plots, while others have found increased root mass
following burning (e.g., Hadley and Kieckhefer 1963, Medve 1987, Dhillion and
Anderson 1993). Medve’s (1987) results suggest that increased root mass is not due to

increased allocation of carbon to belowground structures at the expense of aboveground
gr 0Wth. Rather, plants on burned plots were larger overall than those on control plots. In

a greenhouse experiment that used soils from unburned (control) and from burned plots,

32

tt‘lregr:

from b

t<imIT
from it
author:

tire.

wiregrass plants grown in control soils had roots that were stunted, while those in soils
from burned plots had root growth that “appeared to be normal” (Christensen 1977).
Results from standard soil cores indicated that fire had no effect on fine root
(<lmm) density in red shank chaparral (Kummerow and Lantz 1983). However, results
from ingrowth cores indicated that bunting significantly increased fine-root density; the
authors suggested that this was the result of enhanced nutrient availability following the

fire.

Trees
Most of the references from the literature on effects of fire on the fine roots of
trees are anecdotal and not quantitative. For example, Landsberg and Cochran (1980)
report that soil profile observations after a high-fuel-consumption burn in ponderosa pine
indicated that most of the fine feeding roots located near the surface of the soil were
destroyed, but the authors report no data. Heyward (1934) concluded that the intense
heat of a wildfire killed all fine feeding roots to a depth of 2.5cm, although he observed
that “both longleaf and slash [pine] quickly develop new feeding roots after a fire.”
Other studies merely speculate that the cause of mortality for certain species after fire is
root injury (e.g., Connaughton 1936, Lutz 1960).
Where quantitative data are reported, results have indicated that fine root
mortality occurred during burning, and corresponded to increased soil temperatures (e.g.,
Shearer 1975, Swezy and Agee 1991 ). Total (live + dead) fine root mass did not change
following light burning in ponderosa pine (Grier 1989), while there was a significant

increase in unburned plots. Grier also reported crown scorch from the fires. Therefore, it

33

is not clear Wht
induced mortal
ro.ots or groxttl
wage and on
of root inj ury v

Swezy J
ponderosa pine
the forest fl oor
oflive fine roo
Citameter class t
‘5an] PaSSage l

5 hours.

FlRE AND O\

The que
he years. Som
lbuatl958.lc

reduced Emmy
Gmsrh

35m at all t A 1.

lottt E1 a] 196{

l“ .1.
Wimp 6t al. 1

l
1‘».

gas

is not clear whether the lack-of-increase in fine—root biomass was caused directly by heat-
induced mortality, or if there simply was not enough foliage left to support the remaining
roots or growth of new roots. McConkey and Gedney (1951) quantified both root
damage and crown injury to white pine following wildfire. They concluded that degree
of root injury was more important than crown injury in determining post-fire survival.
Swezy and Agee (1991) looked at the effects of reintroducing fire into old-growth
ponderosa pine ecosystems. Before burning, there were more live fine (<1mm) roots in
the forest floor than in the mineral soil (0-10 and 10-20cm). After the fire, the reduction
of live fine root mass was most pronounced in the forest floor and in the smallest
diameter class of roots. In their study, the thick forest floor smoldered for hours after the
initial passage of the fire, and lethal temperatures (>60 °C) were sustained for more than

5 hours.

FIRE AND OVERSTORY GROWTH
The question of prescribed fire’s effects on tree growth has been debated through
the years. Some studies have indicated increased growth following bunting (Morris and
Mowat 1958, Johansen 1975, Somes and Moorhead 1954) while others have found
reduced growth (Boyer 1983, 1987, 1993, Cochran and Hopkins 1991, Grier 1989,
Gruschow 1951, Haywood and Grelen 2000, Landsberg et al. 1984, Zahner 1989) or no
effect at all (Alban, 1977, Burrows et a1. 1989, Grelen 1976, Hunt and Simpson 1985,
Lotti et al. 1960, Mann and Whitaker 1955, Ross et al. 1995, Van Lear et al. 1977,
Waldmp et al. 1987, Wyant et al. 1983). Other studies (e.g., Brockway and Lewis 1997)

have found mixed results, depending on which measure of growth — height, diameter, or

34

\‘0illme ‘ was
is diameter grt‘
ire rotation or
{1904) suggest
reducing long-
intervals result
While 1
$0th (\l'aldr
Ltponderosa p
but the burning
a“? data for th.
Eminded into ,
30' efffi‘l 0n nu
flush Was redu‘.

concluded that

volume — was reported. It appears that height growth may be more sensitive to fire than
is diameter growth (Waldrop and Lloyd 1988). The effects of fire on growth depend on
fire rotation or fire-retum interval (Grelen 1983, Peterson et al. 1994). Peterson et a1.
(1994) suggest that an interval of 4 to 6 years provides adequate fuel reduction without
reducing long-term growth in southwestern ponderosa pine. Shorter or longer return
intervals resulted in growth loss.

While height growth does appear to be more sensitive to fire than is diameter
growth (Waldrop and Lloyd 1988), Wyant et al. (1983) found that a fall prescribed burn
in ponderosa pine had no effect on shoot growth during the first post-fire growing season,
but the burning did result in significantly larger buds. Unfortunately, they don’t report
any data for the second growing season, the one in which those larger buds would have
expanded into new shoots. Waldrop and Lloyd’s (1988) results showed that burning had
no effect on number of growth flushes in loblolly pine, but rather the length of the each
flush was reduced. In a literature review of the effects of fire on pines, Landsberg (1993)
concluded that tree grth usually decreases after prescribed underburning.

The mechanisms that account for these varying results are becoming clearer.
Most researchers (e.g., Cain 1985, 1996, Lilieholm and Hu 1987, Mann and Whitaker
1955) agree that medium-to-heavy crown scorch results in a loss of growth, although

light crown scorch usually has no deleterious effects on growth. Amount of crown
scorch can be related to type of fire. For example, Gruschow (1951) found that plots
burned With headfires had significantly more crown scorch than plots that had backfires.

Consequently, scorched trees had significantly less height and diameter growth.

35

However. both l
atire even thOUt
Other grt
this area has not
found decrease:
burned longleaf
increased moist
Increase
leen reduced ir
This reduction
{Morris and M

Chambers (197

However, both height and diameter growth have been shown to decline significantly after
a fire even though little or no visible damage occurred (Chambers et al. 1986).

Other growth-loss mechanisms include injury to roots (Landsberg 1993), although
this area has not received intensive study (see above). Also, Boyer and Miller (1994)
found decreased moisture-holding capacity of surface and subsurface soils in repeatedly-
bumed longleaf pine plots. They suggested that growth losses are due, at least in part, to
increased moisture stress associated with changes in soil physical properties.

Increased growth following burning has often been found where competition has
been reduced in overcrowded stands (e.g., Waldrop and Lloyd 1988, Lloyd et al. 1995).
This reduction in competition may compensate for growth losses due to crown scorch
(Morris and Mowat 1958, Johansen 1975, Burrows et al. 1989). Villarrubia and
Chambers (1978) actually found increased growth following low levels of crown scorch.
The authors attributed these results to elimination of the non-productive “free-loading”
lower portions of the crown.

Many researchers that have found an increase in available nutrients following
fires suggest that this increase will improve the growth rate of the overstory. However,
most of these studies didn’t measure overstory growth responses. For those that did,
some have found no changes in growth (Alban 1977, Hunt and Simpson 1985, Lotti et al.
1960) and occasionally decreases in growth (Boyer and Miller 1994). Ross et al. (1995)

suggested that burning enhances nutrient cycling, creating an improved environment for
tree gr OW’th. However, their results also indicated no effects of fire on growth of either

longleaf or loblolly pine. Data from Lunt (1950) suggest increased nutrient availability

36

miti-

“A. “ .
h

i0 i'lt

( J-
.tj.

coupled with increased growth following fire in red and white pine stands. However, his
study was not properly replicated and his data must be interpreted cautiously.

The results of greenhouse experiments are usually different from field studies.
For example, McKevlin and McKee (1986) took soils from a site that had 33 years of
annual burning, or from control plots, and brought them into the greenhouse. Following
sieving, they germinated loblolly pine seeds on these soils and found improved nutrition
of seedlings grown on the previously-bumed soils. Others have found similar results
with a variety of species (cf. Vlamis et al. 1955, Wagle and Kitchen 1972). These results
might be explained by the lack of competition and the ideal watering conditions often
found in greenhouse experiments. That is, in the field studies competition for water may
be more important than the potential increase in available nutrients.

There does appear to be a growing consensus that even low-intensity prescribed
bunting does indeed reduce productivity. This has even prompted Tiedemann et al.
(2000) to harshly criticize the widespread use of prescribed fire as a solution to forest
health problems in the Blue Mountains because of the potential (probable) loss of growth

following burning.

Red pine
The effects of fire on the growth of red pine are fairly consistent. None of the
studies (that I’m aware of) have shown increased growth following bunting. Alban
( 1 9 77) showed that annual and periodic burning in red pine had no effect on overall
gr owth, A single, low-intensity prescribed fire also had no effect on radial growth of

trees (Methven and Murray 1974). However, Roberts and Mallik (1994) concluded that

37

tailoring more-

to 4 Yea”

Thermal death

Extreme
reached at less-r
.1961 ) states tit.
plants is betwec-
att'hole. “her
induced tissue <
1943. Levitt 19
1993). Visible
exposure. or tht
Weeks (Sachs 1
Secondary caus

The ten
the time of fit)

following more-intense wildfires, growth of the annual ring is generally depressed for up

to 4 years.

Thermal death point

Extreme temperatures will cause death to living cells. However, death can be
reached at less-extreme temperatures, with longer exposures to those temperatures. Hare
(1961) states that the thermal death point at the cellular level for average mesophytic
plants is between 50°C and 55°C, while 60°C is frequently given as lethal for the plant as
a whole. Where death does not occur, a cell or tissue may be injured. The onset of heat-
induced tissue damage is between 50°C and 55°C for most plant species (Daubenmire
1943, Levitt 1980, Kappen 1981, Larcher 1983, Seidel 1986, Colombo and Timmer
1992). Visible symptoms of heat injury may be direct and acute and appear soon after
exposure, or they may be indirect and chronic, with symptoms not appearing for days or
weeks (Sachs 1864, in Baker 1929; Colombo and Timmer 1992). Death may be due to
secondary causes, such as fungi attacking at the point of initial wound (Baker 1929).

The temperature that is considered lethal has little meaning without mention of
the time of exposure. As temperature approaches 60°C, the tirne-of-exposure resulting in
death or injury decreases exponentially (Blodgett 1923; Lorenz 1939; Byram 195 8;
Colombo and Timmer 1992; Hare 1961; Seidel 1986; Ursic 1961; Wright 1970; Wright

and Bailey 1982). Exposure times at 60°C that result in plant death, vary from
inStantaneous (e.g., Vines 1968) to 1 minute (e.g., Seidel 1986).
Thermal death point has been estimated for seedlings of many tree species,

usually to determine how well they will survive on harsh, exposed sites (reviewed by

38

Hagen
agener
Neethli

Kolb at

mater.
:entper
uuei
{Sniper
mecha:
can be

Kolb a

"filer I;
ff‘is‘fise.
mahot

Lila.- ,
“056 I

.Ih;ij“

Helgerson 1990). Maximum temperatures at the soil surface in nursery beds or
regenerating sites can range from 55 to 75°C (Korstian and Fetherolf 1921; Tourney and
Neethling 1923; Isaac 1938; Vaartaja 1949; Smith 1951; Harrington and Kelsey 1979;
Kolb and Robberecht 1996).

It is difficult to determine the actual temperature of the tissue under study. It has
usually been estimated to be the same temperature as that of the surrounding medium
(water, air, or sand). Thermocouples have sometimes been used to measure the
temperature of these tissues (cf. Baker 1929); however, inserting the thermocouples may
cause injury to the tissue and it is therefore difficult to separate the effects of elevated
temperatures from the effects of wounding. Plants may also use transpiration and other
mechanisms to maintain lower temperatures than the air or soil. Temperature of tissues
can be lower than that of surrounding air or soil by as much as 12 - 20°C (Baker 1929;
Kolb and Robberecht 1996).

Roots seem to be more sensitive to high temperatures than are above-ground
tissues (Shirley 1936). Leitch (1916, in Baker 1929) found that pea roots died when
exposed to temperatures of only 45°C. Gentile and Johansen (1956) exposed roots of
dormant sand pine and slash pine seedlings to temperatures from 47 to 54°C in a hot
water bath. They found almost complete mortality of outplanted seedlings that were

eXposed to 52°C and higher. Ursic (1961) immersed roots of 1-0 loblolly pine seedlings
in a hot water bath. He concluded that the mortality rates were acceptable for seedlings
Whose roots were immersed at 48°C for 5 minutes, or 46°C for up to 2 hours.
Connaughton (193 6) attributed delayed mortality of Douglas-fir, following a

wfldfif e, to damage of roots. McConkey and Gedney (1951) found that root injury was

39

  
 

more seriouS “14

surface roots we

Ether factors
There is
Harrington and
Ofponderosa p
seedlings sun:

'lan Iemperatt

1- SPEI
moms follc
fillings had

\e'.
“5% (193a.

more serious than crown injury in determining the survival of white pine following a
wildfire. They considered root injury to be “light” if 25% or less of the visible roots were
killed or severely injured, while damage was “heavy” if 75% or more of the major

Surface roots were killed or severely injured.

Other factors ,
There is variability in the susceptibility of tissue to damage. For example,
Harrington and Kelsey (1979) found that surface temperatures of 55°C caused mortality
of portderosa pine seedlings on some sites in western Montana, but on other sites
seedlings survived temperatures up to 66°C. Factors involved in tissue damage, other

than temperature and time-of—exposure, are listed below.

1. Species. Seidel (1986) found that Englemann spruce seedlings had 100%
mortality following a 1 minute exposure to 60°C, but ponderosa pine and Douglas-fir
seedlings had up to 50% survival at the same time-temperature combination. Byram and
Nelson (1952) found that needles of pitch pine withstood higher temperature-time
combinations than did loblolly pine. Slash pine was most sensitive, while longleaf pine
needles gave variable results, depending on the time-temperature combination. Critical
temperature, Tc, differed for two species of Ilex whose roots had been exposed to high
temperatures (Ingram 1986).

2. Type of tissue. Roots seem to be more sensitive to high temperatures than
aboveground tissues (e.g., Shirley 1936). While Kayll (1966) and Vines (1968) defined

the lethal threshold for cambial cells to be instantaneous exposure to 60°C, Bradstock et

40

211.1199M W“
exposures Of U:
Ptmr's life CFC
3. AS"
toleranl than -V
1936; 1(0an

4.Sea5
temperature of
that conifers at
the spring and

5. Free:

:Alexandrov 1‘

says prior to trt
there was no Si
Colombo et a1.
more thermotol
increased level:
phenomenon c

6. \l'ate

pelalures‘ A!

al. (1994) found that seeds of Hakea species in Australia may be able to withstand
exposures of up to 5 minutes at 60°C. Seeds may be the most heat-resistant stage in a
plant’s life cycle (Hare 1961).

3. Age of the tissue or plant. Older seedlings or tissues tend to be more thermo-
tolerant than younger seedlings or tissues (Bates and Roeser 1924; Baker 1929; Shirley
1936; Koppenal and Colombo 1988).

4. Season of the year (which affects the succulence, water content, and initial
tenmerature of tissues). Koppenaal and Colombo (1988) cite several studies that indicate

that conifers are most heat-tolerant during winter dormancy, and most susceptible during
the spring and early summer, when shoots are actively growing.

5. Preconditioning the tissue to high temperatures (a.k.a. heat hardening
(Alexandrov 1977)). Koppenaal et al. (1991) determined that acquired thermo-tolerance
persisted after 14 days for jack pine seedlings that were pre-conditioned (hardened) for 6
days prior to treatment. However, for seedlings that were hardened for only 1 or 3 days,
there was no significant difference between their thermotolerance and that of controls.
Colombo et a1. (1995) found that both roots and shoots of black spruce seedlings were
more thermotolerant in afternoons than in mornings. This pattern was accompanied by
increased levels of heat shock proteins (HSPs) in the afternoon. Whether this
phenomenon constitutes true pre-conditioning or not is debatable.

6. Water status of the tissue. Drier seeds are more heat-tolerant than moister
seeds (Smith 1951). Assuming that denaturation of proteins is the primary result of high

temperatures, Alexandrov (1977) states that heat treatments of 120°C to 150°C are

41

needed to dena
denatured at 6"

7. lniti
lines 1968). '
needed to rear

8. Car
ceiis from the
knell] Olilimt

mm“ diffu.

The rr
treatments ha
eater bath C0
cause of deatl
seedlings of s
No. injUI‘y “a
shoots in roor
says of expo

1933
’ or by u

Seedlings in 8
ha

:1) thrOth

needed to denature dehydrated proteins, while the same proteins, in a water solution, are

denatured at 60°C to 70°C.

7. Initial temperature of the tissue (Byram 1948; Nelson 1952; Byram 1958;
V fines 1968). This doesn’t affect thermotolerance directly, but rather affects the time
needed to reach critical temperatures.
8. Capacity of other tissue such as leaves or bark to insulate buds or other living
cells from the heat source (Hare 1961). Similar to variable 7, these factors affect the
length of time to reach critical temperatures. Reifsnyder et al. (1967) found that the

thermal diffusivity of bark was lower for red pine than either shortleaf or longleaf pine.

The methods of exposing roots to high temperatures vary. Various hot air
treatments have been used, as have constant-temperature water baths. Submersion in a
water bath could possibly result in anaerobiosis, thus complicating determination of the
cause of death. To eliminate anaerobiosis as a cause of death, Shirley (1936) exposed
seedlings of several species to water at room temperature, for either 2 hours or 5 hours.
No injury was observed. Similarly, Colombo and Timmer (1992) immersed black spruce
shoots in room-temperature water for up to 3 hours, which also caused no damage. Other
ways of exposing roots have included sprinkling hot sand over entire seedlings (Roeser
1932) or by using a “dry-water” bath (Seidel 1986). This technique involves planting

seedlings in a sand mixture, and then circulating hot water (from a constant-temperature

bath) through tubes that ran through the sand.

42

Literature Cited

Aber, J. D., J. M. Melillo, K. J. Nadelhoffer, C. A. McClaugherty, and J. Pastor. 1985.
Fine root turnover in forest ecosystems in relation to quantity and form of nitrogen
availability: a comparison of two methods. Oecologia (Bed) 66: 317-321.

Ahlgren, I . F ., and C. E. Ahlgren. 1965. Effects of prescribed burning on soil
mjcroorganisms in a Minnesota jack pine forest. Ecol. 46: 304-310.

Alban, D. H- 1977. Influence on soil properties of prescribed burning under mature red
pine- USDA For. Serv. Res. Pap. NC-139. 8p.

lexandrov, V. Ya. 1977. Cells, molecules and temperature. Conformational flexibility
of macromolecules and ecological adaptation. Springer-Verlag, New York. 330p.

Amaanthus, M. P., and J. M. Trappe. 1993. Effects of erosion on ecto- and VA-
mycon' 'zal inoculum potential of soil following forest fire in southwest Oregon. Pl. Soil.
1 50'. 41‘49.

Ares, A., and N. Peinemann. 1992. Fine-root distribution of coniferous plantations in
re1ation to site in southern Buenos Aires, Argentina. Can. J. For. Res. 22: 1575-1582.

Baker, F. S. 1929. Effect of excessively high temperatures on coniferous reproduction. J.
For. 27: 949-975.

Bates, C. G., and J. Roeser, Jr. 1924. Relative resistance of tree seedlings to excessive
heat. USDA Bull. 1263. 16p.

Bell, R., and D. Binkley. 1989. Sol nitrogen mineralization and immobilization in

response to periodic prescribed fire in a loblolly pine plantation. Can. J. For. Res. 19:
816-820.

Bentivenga, S. P., and B. A. D. Hetrick. 1991. Relationship between mycorrhizal
activity, burning, and plant productivity in tallgrass prairie. Can. J. Bot. 69: 2597-2602.

Binkley, D., D. Richter, M. B. David, and B. Caldwell. 1992. Soil chemistry in a
loblolly/longleaf pine forest with interval burning. Eco. Apps. 2: 157-164.

Bissett, J ., and D. Parkinson. 1980. Long-tenn effects of fire on the composition and

activity of the soil microflora of a subalpine, coniferous forest. Can. J. Bot. 58: 1704-
1721.

Biswell, H. H., and A. M. Schultz. 1957. Surface runoff and erosion as related to
prescribed burning. J. For. 55: 372-374.

43

Bit:

M
111'

Black, K. E., C. G. Harbron, M. Franklin, D. Atkinson and J. E. Hooker. 1998.
Differences in rootlongevity of some tree species. Tree Phys. 18: 259-264.

Blodgett, F. M. 1923. Time-temperature curves for killing potato tubers by heat
treatments. Phytopath. 13: 465-475.

Bockheim, J. G., S. W. Lee, and J. E. Leide. 1983. Distribution and .cycling of elements
in 3 Firms resinosa plantation ecosystem, Wisconsin. Can. J. For. Res. 13: 609-619.

Boerner, R- E. J. 1982. Fire and nutrient cycling temperate ecosystems. BioSci. 32: 187-
l 92.

oemer, R- E. J. 1983. Nutrient dynamics of vegetation and detritus following two
inmnsities of fire in the New Jersey pine barrens. Oecologia. 59: 129-134.

Boerner, R. E. J ., and R. T. T. Forman. 1982. Hydrologic and mineral budgets of New
1 61.56): Pine Barrens upland forests following two intensities of fire. Can. J. For. Res. 12:

503-510.

Boerner, R. E. J ., T. R. Lord, and J. C. Peterson. 1988. Prescribed burning in the oak-
ine forest of the New Jersey Pine Barrens: Effects on growth and nutrient dynamics of
two Quercus species. Am. Mid. Nat. 120: 108-119.

Boyer, W. D. 1983. Growth of young longleaf pine as affected by biennial burns plus
chemical or mechanical treatments for competition control. pp. 62-65 in Jones, E. P., Jr.

ed. Proceedings of the second biennial southern silvicultural research conference. USDA
For. Serv. Gen. Tech. Rep. SE-24.

Boyer, W. D. 1987. Volume growth loss: A hidden cost of periodic prescribed burning in
longleaf pine? South. J. Appl. For. 11: 154-157.

Boyer, W. D. 1993. Eighteen years of seasonal burning in longleaf pine: Effects on
overstory growth. pp. 602-610 in Proceedings of the 12Lh Conf. on Fire and Forest
Meteorology. Oct. 26-28, 1993. Jekyll Island, GA.

Boyer, W. D., and J. H. Miller. 1994. Effect of burning and brush treatments on nutrient

and soil physical properties in young longleaf pine stands. For. Ecol. and Manage. 70:
3 1 1-3 18.

Bradstock, R. A., A. M. Gill, S. M. Hastings, and P. H. R. Moore. 1994. Survival of
serotinous seedbanks during bushfires: Comparative studies of Hakea species from
southeastern Australia. Aust. J. Ecol. 19: 276-282.

Brady, N. C. 1990. The nature and properties of soils. 10th ed. Macmillan Pub. Co., New
York. 621p.

44

Beet
resins
Nont

Brae
prest

one;

‘EJ' I I U
' !
(D CA-

n.
.L‘
(5*

C?

:7 I
('1!
H

{/1 "Te!
:3
(tr

Braekke, F. H., and T. T. Kozlowski. 1977. Distribution and growth of roots in Pinus
resinosa and Betula papyrifera stands. Med. fra Norsk Instittutt for Skogforskning (Rep.
Norwegian For. Res. Inst.) 33(10): 437-451.

Brender, E. V., and R. W. Cooper. 1968. Prescribed burning in Georgia’s Piedmont
loblolly pine stands. J. For. 66: 31-36.

Brockway, D. G., and C. E. Lewis. 1997. Long-term effects of dormant-season
rescribed fire on plant community diversity, structure and productivity in a longleaf pine
Wil—egrass ecosystem. For. Ecol. Manage. 96: 167-183.

Buchholz, K., and M. Gallagher. 1982. Initial ectomyconhizal density response to
Wildfire in the New Jersey Pine Barren Plains. Bull. Torrey Bot. Club 109: 396-400.

Burke, M. K., and D. J. Raynal. 1994. Fine root growth phenology, production, and
turnover in a northern hardwood forest ecosystem. Pl. Soil 162: 135-146.

Burrows, N. D., Y. C. Woods, B. G. Ward, and A. D. Robinson. 1989. Prescribing low
.1 mensity fire to kill wildings in Pinus radiata plantations in Western Australia. Aust. For.

Byram, G. M. 1948. Vegetation temperature and fire damage in the southern pines. Fire
Cont. Notes 9(4): 34-36.

Byram, G. M. 1958. Some basic thermal processes controlling the effects of fire on
living vegetation. USDA For. Serv. Southeast. For. Exp. Sta. Res. Notes 114. 2p.

Byram, G. M., and R. M. Nelson. 1952. Lethal temperatures and fire injury. USDA For.
Serv. Southeast. For. Exp. Sta. Res. Notes 1. 2p.

Cain, M. D. 1985. Prescribed winter burns can reduce the growth of nine-year-old
loblolly pines. USDA For. Serv. Res. Note SO-312. 4p.

Cain, M. D. 1996. Growth expectations from alternative thinning regimes and prescribed

burning in naturally regenerated loblolly-shortleaf pine stands through age 20. For. Ecol.
Manage. 81: 227-241.

Chambers, J. L., P. M. Dougherty, and T. C. Hennessey. 1986. Fire: Its effects on growth
and physiological processes in conifer forests. pp. 171-189 in Stress physiology and

forest productivity. Henessey, T. C., P. M. Dougherty, S. V. Kossuth, and J. D. Johnson
eds.

Chen, M., E. J. Hodgkins, and W. J. Watson. 1975. Prescribed burning for improving

pine production and wildlife habitat in the hilly Coastal Plain of Alabama. Ala. Ag. Exp.
Sta. Bull. 473. 19p.

45

Cheng. \V 00' h
minirhizotron to

 

Christensen. .\'.
coastal plain of

Cochran. P. 11..
ponderosa pine
islontane iores‘

Coile. T. S. 19

For. 35: 247-2

Colombo. S. J
causing direct

Colombo. S. .1
Variation in he
WWW). Ca

Connaughton.
416-51,

Cm'ington. tt'
mamags‘vment i
Fatum] resour.

C0“hasten. \V
Silulhtt" ~

a_ Estem
.toor. For. Sci

CUHCenu-aIiOn

Rjubenmue.
“L'I‘Udlnal

D

lirr

3F. M. “1 1

Cheng, W., D. C. Coleman, and J. E. Box, Jr. 1991. Measuring root turnover using the
minirhizotron technique. Ag. Ecosystems Env. 34: 261-267.

Christensen, N. L. 1977. Fire and soil-plant relations in a pine-wiregrass savanna on the
coastal plain of North Carolina. Oecologia 31: 27-44.

Cochran, P. H., and W. E. Hopkins. 1991. Does fire exclusion increase productivity of
ponderosa pine? pp. 224-228 in Proceedings — Management and productivity of Westem-
Montane forest soils. USDA For. Serv. Gen. Tech. Rep. INT-280.

Coile, T. S - 1937. Distribution of forest tree roots in North Carolina Piedmont soils. J.
For- 35: 247-257.

Colombo’ S . J ., and V. R. Timmer. 1992. Limits of tolerance to high temperatures
causing direct and indirect damage to black spruce. Tree Phys. 11: 95-104.

Colombo, S. J., V. R. Timmer, M. L. Colclough, and E. Blumald. 1995. Diurnal
variation in heat tolerance and heat shock protein expression in black spruce (Picea

mariana). Can. J. For. Res. 25: 369-375.

Connaughton, C. A. 1936. Fire damage in the ponderosa pine type in Idaho. J. For. 34:
46-5 1 .

Covington, W. W., and S. S. Sackett. 1990. Fire effects on ponderosa pine soils and their
management implications. pp. 105-111 in Effects of fire management of Southwestern
natural resources. USDA For. Serv. Gen. Tech. Rep. RM-191.

Covington, W. W., and S. S. Sackett. 1984. The effect of a prescribed burn in
southwestern ponderosa pine on organic matter and nutrients in woody debris and forest

floor. For. Sci. 30: 183-192.

Covington, W. W., and S. S. Sackett. 1986. Effect of periodic burning on soil nitrogen
concentrations in ponderosa pine. Soil Sci. Soc. Am. J. 50: 452-457.

Daubenmire, R. F. 1943. Soil temperature versus drought as a factor determining lower
altitudinal limits of trees in the Rocky Mountains. Bot. Gaz. 105: 1-13.

Day, M. W. 1941. The root system of red pine saplings. J. For. 39: 468-472.

DeBano, L. F. 1991. The effects of fire on soil properties. pp. 151-156 in Proceedings —

Management and productivity of Westem-Montane Forest Soils. USDA For. Serv. Gen.
Tech. Rep. INT-280.

DeBano, L. F ., and C. E. Conrad. 1978. The effect of fire on nutrients in a chaparral
ecosystem. Ecology 59: 489-497.

46

DeBarlo, L. F ., P. H. Dunn, and C. E. Conrad. 1977. Fire’s effects of on physical and
chemical properties of chaparral soils. pp. 65-74 in Proceedings of the symposium on the
environmental consequences of fire and fuel management in the Mediterranean
ecosystems. Mooney, H. A., and C. E. Conrad, Tech. Coords. USDA For. Serv. Gen.
Tech. Rep. WO-3.

DeBano, L. F°9 Neary, D. G., and P. F. F folliott. 1998. Fire effects on ecosystems. John
Wiley & Sons, New York. 333p.

DeBell, D- S., and C. W. Ralston. 1970. Release of nitrogen by bunting light forest fuels.
Soil Sci. Soc. Am. Proc. 34: 936-938.

DeWitt, J. B., and J. V. Derby, Jr. 1955. Changes in nutritive value of browse plants
following forest fires. J. Wildl. Manage. 19: 65-70,

Dhillion, S. S., and R. C. Anderson. 1993. Growth dynamics and associated mycorrhizal
fungi of little bluestem grass [Schizachyrium scoparium (Michx.) Nash] on burned and
unburned sand prairies. New Phytol. 123: 77-91.

Dhillion, S. S., R. C. Anderson, and A. E. Liberta. 1988. Effect of fire on the
mycorrhizal ecology of little bluestem (Schizachyrium scoparium). Can. J. Bot. 66: 706-

713.

Dickmann, D. I. 1993. Management of red pine for multiple benefits using prescribed
fire. North. J. Appl. For. 10: 53-62.

Douglass, J. E., and D. H. Van Lear. 1983. Prescribed bunting and water quality of
ephemeral streams in the Piedmont of South Carolina. For. Sci. 29: 181-189.

Eissenstat, D. M., and K. C. J. Van Rees. 1994. The growth and function of pine roots.
Ecol. Bull. 43: 76-91.

Eivazi, F ., and M. R. Bayan. 1996. Effects of long-term prescribed burning on the
activity of select soil enzymes in an oak-hickory forest. Can. J. For. Res. 26: 1799-1804.

Espeleta, J. F., and D. M. Eissenstat. 1998. Responses of citrus fine roots to localized
soil drying: a comparison of seedlings with adult fruiting trees. Tree Phys. 18: 113-119.

F ayle, D. C. F. 1975. Extension and longitudinal growth during the development of red
pine root systems. Can. J. For. Res. 5: 109-121.

Fuller, W. H., S. Shannon, and P. S. Burgess. 1955. Effect of burning on certain forest
soils of northern Arizona. For. Sci. 1: 44-50.

Gentile, A. C., and R. W. Johansen. 1956. Heat tolerance of slash and sand pine
seedlings. USDA For. Serv. Southeast. For. Exp. Sta. Res. Notes 95. 1p.

47

Gholz
the r

Gholz, H. L., L. C. Hendry, and W. P. Cropper, Jr. 1986. Organic matter dynamics of
fine roots in plantations of slash pine (Pinus elliottii) in north Florida. Can. J. For. Res.
1 6: 52 9-538.

Gibson, D. J., and B. A. D. Hetrick. 1988. Topographic and fire effects on the
composition and abundance of VA-mycorrhizal fungi in tallgrass prairie. Mycologia 30:
43 3-441.

Gillan, D., and M. Rapp. 1989. Nutrient losses during a winter low-intensity prescribed
fire in a Mediterranean forest. Plant and Soil. 120: 69-77.

Gower, 5- T., K. A. Vogt, and C. C. Grier. 1992. Carbon dynamics of Rocky Mountain
DouglaS-fir: Influence of water and nutrient availability. Ecol. Monog. 62: 43-65.

Grelen, H. E. 1976. Response of herbage, pines, and hardwoods to early and delayed
bun-{mg in a young slash pine plantation. J. Range Manage. 29: 301-303,

Grelen, H. E. 1983. Comparison of seasons and frequencies of burning in a young slash
pine plantation. USDA For. Serv. Res. Pap. SO-185.

Grief, C. C. 1975. Wildfire effects on nutrient distribution and leaching in a coniferous
ecosystem. Can. J. For. Res. 5: 599-607.

Grier, C. C. 1989. Effects of prescribed springtime underbunting on production and
nutrient status of a young ponderosa pine stand. pp. 71-76 in Multiresource management
of ponderosa pine forests. USDA For. Serv. Gen. Tech. Rep. RM-185.

Grier, C. C., K. A. Vogt, M. R. Keyes, and R. L. Edmonds. 1981. Biomass distribution
and above- and below-ground production in young and mature Abies amabilis zone
ecosystems of the Washington Cascades. Can. J. For. Res. 11: 155-167.

Grigal, D. F ., and J. G. McColl. 1977. Litter decomposition following forest fire in
Northeastern Minnesota. J. Appl. Ecol. 14: 531-538.

Gruschow, G. F. 1951. Effect of winter bunting on slash pine growth. Southern
Lumbennan 183(2297): 260-264.

Guinto, D. F., P. G. Saffigna, Z. H. Xu, A. P. N. House, and M. C. S. Perera. 1999. Soil
nitrogen mineralisation and organic matter composition revealed by '3 C NMR

spectroscopy under repeated prescribed burning in eucalypt forests of south-east
Queensland. Aust. J. Soil Res. 37: 123-135.

Hadley, E. B., and B. J. Kieckhefer. 1963. Productivity of two prairie grasses in relation
to fire frequency. Ecology 44: 389-395.

48

Hamilton, S. D., P. Hopmans, P. M. Chalk and C. J. Smith. 1993. Field estimation of N2
fixation by Acacia Spp. using l5N isotope dilution and labelling with 35 S. For. Ecol. and
Manage. 56: 297-313.

Hare, R. C. 1961. Heat effects on living plants. USDA For. Serv. South. For. Exp. Sta..
Occas. Pap 183. 321).

Harrington, M. G., and R. G. Kelsey. 1979. Influence of some environmental factors on
initial establishment and growth of ponderosa pine seedlings. USDA For. Serv. Res. Pap.
INT-230. 26p.

Harris, G. R., and W. W. Covington. 1983. The effect of a prescribed fire on nutrient
concentration and standing crop of understory vegetation in ponderosa pine. Can. J. F 01',
Res. 13: 501-507.

Harris, J. R., N. L. Bassuk, R. W. Zobel, and T. H. Whitlow. 1995. Root and shoot
growth periodicity of green ash, scarlet oak, Turkish hazelnut, and tree lilac. J. Am. Soc.
Hort. Sci. 120: 211-216.

Harris, W. F ., R. S. Kinerson, Jr., and N. T. Edwards. 1977. Comparison of belowground
biomass of natural deciduous forest and loblolly pine plantations. Pedobiologia 17: 369-
381.

Haynes, B. B., and S. T. Gower. 1995. Belowground carbon allocation in unfertilized
and fertilized red pine plantations in northern Wisconsin. Tree Phys. 15: 317-325.

Haywood, J. D., and H. E. Grelen. 2000. Twenty years of prescribed bunting influence
the development of direct-seeded longleaf pine on a wet pine site in Louisiana. South. J.
Appl. For. 24: 86-92.

Helgerson, O. T. 1990. Heat damage in tree seedlings and its prevention. New For. 3:
333-358.

Hendrick, R. L., and K. S. Pregitzer. I992. The demography of fine roots in a northern
hardwood forest. Ecol. 73: 1094-1104.

Hendricks, J. J ., and L. R. Boring. 1999. Nz-fixation by native herbaceous legumes in

bggned pine ecosystems of the southeastern United States. For. Ecol. Manage. 113: 167-

Herr, D- (3., L. C. Duchesne, R. Tellier, R. S. McAlpine, and R. L. Peterson. 1994. Effect

of prescribed burning on the ectomyconhizal infectivity of a forest soil. Int. J. Wildland
Fire 4: 95-102.

Heyward, F. 1934. Comments on the effect of fire on feeding roots of pine. Savannah
weekly naval stores review and journal of trade. 44(19): 4.

49

Hooker, J. E., and D. Atkinson. 1996. Arbuscular mycorrhizal fungi-induced alteration
to tree-root architecture and longevity. Z. Pflanzenemahr. Bodenk. 159: 229-234.

Hooker, J. E., K. E. Black, R. L. Perry, and D. Atkinson. 1995. Arbuscular myconhizal
fungi induced alteration to root longevity of poplar. Plant and Soil 172: 327-329.

Horton, T. R., E. Cazares, and T. D. Bruns. 1998. Ectomycorrhizal, vesicular-arbuscular
and dark septate fungal colonization of bishop pine (Pinus muricata) seedlings in the first
5 months of growth after wildfire. Myconhiza 8: 11-18.

Hungerford, R. D., M. G. Harrington, W. H- Frandsen, K. C. Ryan, and G. J. Niehoff.
1991. Influence of fire on factors that affect site productivity. pp. 32-50 in Proceedings —
Management and productivity of Westem-Montane forest soils. USDA For. Serv. Gert.
Tech. Rep. INT-280.

Hunt, S. M., and J. A. Simpson. 1985. Effects of low intensity prescribed fire on the
growth and nutrition of a slash pine plantation. Aust. For. Res. 15: 67-77.

Ingram, D. L. 1986. Root cell membrane heat tolerance of two dwarf hollies. J. Am. Soc.
Hort. Sci. 111: 270-272.

Isaac, L. A. 1938. Factors affecting establishment of Douglas-fir seedlings. USDA Circ.
No. 486.

James, T. D. W., and D. W. Smith. 1977. Short-tenrt effects of surface fire on the

biomass and nutrient standing crop of Populus Iremuloides in southern Ontario. Can. J.
For. Res. 7: 666-679.

Johansen, R. W. 1975. Prescribed burning may enhance growth of young slash pine. J.
For. 73: 148-149.

Johnson, C. M., and P. R. Needham. 1966. Ionic composition of Sagehen Creek,
California, following an adjacent fire. Ecol. 47: 636-639.

Jones, J. M., and B. N. Richards. 1977. Effect of reforestation on turnover of 15N-

labelled nitrate and ammonium in relation to changes in soil microflora. Soil Biol.
Biochem. 9: 383-392.

Jorgensen, J. R., and C. S. Hodges, Jr. 1970. Microbial characteristics of a forest soil
after twenty years of prescribed burning. Mycologia 62: 721-726.

Jorgensen, J. R., and C. G. Wells. 1971. Apparent nitrogen fixation in soil influenced by
prescribed burning. Soil Sci. Soc. Am. Proc. 35: 806-810.

50

Joslin, J. D., and G. S. Henderson. 1987. Organic matter and nutrients associated with
fine root turnover in a white oak stand. For. Sci. 33: 330-346.

Jurgensen, M. F ., A. E. Harvey, and M. J. Larsen. 1981. Effects of prescribed fire on soil
nitrogen levels in a cutover Douglas-fir/westem larch forest. USDA For. Serv. Res. Pap.

INT-275. 6p.

Kappen, L, 1981 . Ecological significance of resistance to high temperature. pp. 439-474
in Physiological plant ecology 1: Responses to the physical environment. Encyclopedia
of plant physiology, New series, Vol. 12A. Eds. O. L. Lange, P. S. Nobel, C. B. Osmond
and H Ziegler. Springer-Verlag, New York.

Kayll, A. J. 1966. Some characteristics of heath fires in NE Scotland. J. Appl. Ecol, 3;
29-40.

Kellman, M., K. Miyanishi, and P. Hiebert. 1985. Nutrient retention by savanna
ecosystems: 11. Retention after fire. J. Ecol. 73: 953-962.

Keyes, M. R., and C. C. Grier. 1981. Above- and below-ground net production in 40-
year-old Douglas-fir stands on low and high productivity sites. Can. J. For. Res. 11: 599—
605.

Kimmins, 1R, and B. C. Hawkes. 1978. Distribution and chemistry of fine roots in a
white spruce-subalpine fir stand in British Columbia: implications for management. Can.
J. For. Res. 8: 265-279.

Klemmedson, J. 0., A. M. Schultz, H. Jenny, and H. H. Biswell. 1962. Effect of

prescribed burning of forest litter on total soil nitrogen. Soil Sci. Soc. Am. Proc. 26: 200—
202.

Klopatek, C. C., L. F. DeBano, and J. M. Klopatek. 1988. Effects of simulated fire on

vesicular-arbuscular myconhizae in pinyon-juniper woodland soil. Plant and Soil. 109:
245-249.

Il<5nzight’ H. 1966. Loss of nitrogen from the forest floor by burning. For. Chron. 42: 149-

Knighton, M. D. 197 7. Hydrologic response and nutrient concentrations following spring
burns in an oak-hickory forest. Soil Sci. Soc. Am. J. 41: 627-632.

Kodama, H. E., and D. H. Van Lear. 1980. Prescribed burning and nutrient cycling
relationships in young loblolly pine plantations. South. J. Appl. For. 4: 118-121.

K0118 P. F., and R. Robberecht. 1996. High temperature and drought stress effects on
surv1va1 of Pmus ponderosa seedlings. Tree Phys 16: 665-672.

51

Koppenaal, R. S., and S. J. Colombo. 1988. Heat tolerance of actively growing, bud-
initiated, and (10th black spruce seedlings. Can. J. For. Res. 18: 1103-1105.

Koppenaal, R. S., S. J. Colombo, and E. Blumwald. 1991. Acquired thermotolerance of
jack pine, white spruce and black spruce seedlings. Tree Phys. 8: 83-91.

Korstian, C. F ., and N. J. F etherolf. 1921. Control of stem girdle of Spruce transplants
caused by excessive heat. Phytopath. 11: 48 5-490.

Kosola, K. R., D. M. Eissenstat, and J. H. Graham. 1995. Root demography of mature
citrus trees: the influence of Phytophthora nicotianae. Plant and Soil 171: 283-288.

Kovacic, D. A., D. M. Swift, J. E. Ellis, and T. E. Hakonson. 1986. Immediate effects of
prescribed burning on mineral soil nitrogen in ponderosa pine of New Mexico. Soil Sci.
141: 71-76.

Kurnmerow, J ., and R. K. Lantz. 1983. Effect of fire on fine root density in red shank
(Adenostoma sparsifolium Torr.) chaparral. Pl. Soil. 70: 347-352.

Kurz, W. A., and J. P. Kimmins. 1987. Analysis of some sources of error in methods

used to determine fine root production in forest ecosystems: a simulation approach. Can.
J. For. Res. 17: 909-912.

Landsberg, J. D. 1 993. A review of prescribed fire and tree growth response in the genus
Pinus. pp. 326-346 in Proceedings of the 12th Conf. on Fire and Forest Meteorology. 001-
26—28, 1993. Jekyll Island, GA.

Landsberg, J. D., and P. H. Cochran. 1980. Prescribed burning effects on foliar nitrogen

content in ponderosa pine. pp. 209-213 in Proc. 6th Conf. on Fire and Forest Meteorology.
April 22-24, 1980, Seattle, WA.

Landsberg, J. D., P. H. Cochran, M. M. Finck, and R. E. Martin. 1984. Foliar nitrogen

content and tree growth after prescribed fire in ponderosa pine. USDA For. Serv. Res.
Note PNW-412. 15p.

Largeher, W. 1983. Physiological plant ecology. 2"d ed. Springer-Verlag, New York.
p.

Lay, D. W. 1957. Browse quality and the effects of prescribed burning in southern pine
forests. J. For. 55: 342-347.

Levitt, J. 1980. Responses of plants to environmental stresses, 2nd ed. Academic Press,
New York. 497p,

LCWiS, W. M., Jr. 1974. Effects of fire on nutrient movement in a South Carolina pine
forest. Ecol. 55: 1120-1127.

Lilieholm, R. J ., and S. C. Hu. 1987. Effect of crown scorch on mortality and diameter
growth of 19-year-old loblolly pine. South. J. Appl. For. 11: 209-211.

Lloyd, F. T., T. A. Waldrop, and D. L. White. 1995. Fire and fertilizer as alternatives to
hand thinning in a natural stand of precommercial-sized loblolly pine. South. J. Appl.
For. 19: 5-9.

Lopez, B., S. Sabaté, and C. Gracia. 1998. Fine roots dynamics in a Mediterranean
forest: effects of drought and stem density. Tree Phys. 18: 601-606.

Lorenz, R. W. 1939. High temperature tolerance of forest trees. Minn. Ag. Exp. Sta,
Bull. 141. 25p.

Lotti, T., R. A. Klawitter, and W. P. Legrande. 1960. Prescribed bunting for understory
control in loblolly pine stands of the coastal plain. USDA For. Serv. Sta. Pap. SE-116

19p.

Lunt, H. A. 1950. Liming and twenty years of litter raking and bunting under red (and
white) pine. Soil Sci. Soc. Am. Proc. 15: 381-390.

Lutz, H. J. 1960. Fire as an ecological factor in the boreal forest of Alaska. J. For. 58:
454-460.

Maggs, J ., and R- K. Hewett. 1986. Nitrogenase activity (Csz reduction) in the forest
floor of a Pinus elliottii plantation following superphosphate addition and prescribed
bunting. For. Ecol. and Manage. 14: 91-101.

Majdi, H., and H. Persson. 1995. A study on fine-root dynamics in response to nutrient
applications in a Norway spruce stand using the minirhizotron technique. Z.
Pflanzenernahr. Bodenk. 158: 429-433.

Malajczuk, N., and F. J. Hingston. 1981. Ectomycorrhizae associated with jarrah. Aust.
J. Bot. 29: 453-462.

Mann, W. F., Jr., and L. B. Whitaker. 1955. Effects of prescribe-burning 4-year-old
planted slash pine. USDA For. Serv. Fire Cont. Notes 16(3): 3-5.

Marshall, J. D. 1986. Drought and shade interact to cause fine-root mortality in Douglas-
fir seedlings. Plant and Soil 91: 51-60.

Marshall, J. D., and R. H. Waring. 1985. Predicting fine root production and turnover by
momtonng root starch and soil temperature. Can. J. For. Res. 15: 791 -800.

53

Masters, R. E., D. M. Engle, and R. Robinson. 1993. Effects of timber harvest and
periodic fire on soil chemical properties in the Ouachita Mountains. South. J. Appl. For.

17: 139-145.

McClaugherty, C. A., J. D. Aber, and J. M. Melillo. 1982. The role of fine roots in the
organic matter and nitrogen budgets of two forested ecosystems. Ecol. 63: 1481-1490.

McConkey, T. W., and D. R. Gedney. 1951. A guide for salvaging white pine injured by
forest fires. USDA For. Serv. NEFES Northeast. Res. Notes No. 11. 4p.

McKee, W. H., Jr. 1982. Changes in soil fertility following prescribed burning on
Coastal Plain pine sites. USDA For. Serv. Res. Pap. SE-234. 23p.

McKee, W. H., Jr., and C. E. Lewis. 1983. Influence of bunting and grazing on soil
nutrient properties and tree growth on a Georgia Coastal Plain site after 40 years. pp, 79-
86 in Proceedings of the second biennial southern silvicultural research conference.
USDA For. Serv. Gen. Tech. Rep. SE-24.

McKevlin, M. R., and W. H. McKee, Jr. 1986. Long-tenn prescribed burning increases
nutrient uptake and growth of loblolly pine seedlings. For. Ecol. Manage. 17 : 245-252.

McQueen, D. R. l 973. Changes in understorey vegetation and fine root quantity
following thinning of 30-year Pinus radiata in central North Island, New Zealand. J.
Appl. Ecol. 10: 13-21.

Medve, R. J. 1985. The effect of fire on the root hairs and myconhizae of Liam's spicata.
Ohio J. Sci. 85: 1 51-154.

Medve, R. J. 1987. The effects of fire on resource allocation and growth of Liatris
spicata. Am. Mid. Nat. 117: 199-203.

Melgoza, G., and R. S. Nowak. 1991. Competition between cheatgrass and two native

species after fire: Implications from observations and measurements of root distribution.
J. Range Manage. 44: 27-33.

Methven, I. R., and W. G. Murray. 1974. Using fire to eliminate understory balsam fir in
pine management. For. Chron. 50: 77-79.

Metz, L. J. 1954. Forest floor in the Piedmont region of South Carolina. Soil Sci. Soc.
Am. Proc. 18: 335-338.

Metz, L. J ”Lotti, T., and R. A. Klawitter. 1961. Some effects of prescribed burning on
Coastal Plain forest soil. USDA For. Serv. Sta. Pap. SE-133. 10p.

54

Miller, S. L., T. M. McClean, N. L. Stanton, and S. E. Williams. 1998. Myconhization,
physiognomy, and first-year survivability of conifer seedlings following natural fire in
Grand Teton National Park. Can. J. For. Res. 28: 115-122.

Moehring, D. M ., C. X Grano, and J. R. Bassett. 1966. Properties of forested loess soils
after repeated prescribed burns. USDA For. Serv. Res. Note SO-40. 4p.

Monleon, V. J ., and K. Cromack, Jr. 1996. Long-term effects of prescribed underburning
on litter decomposition and nutrient release in ponderosa pine stands in central Oregon.
For. Ecol. and Manage. 81: 143-152.

Monleon, V. J ., K. Cromack, Jr., and J. D. Landsberg. 1997. Short- and long-tenn effects
of prescribed underburning on nitrogen availability in ponderosa pine stands in central
Oregon. Can. J. For. Res. 27: 369-378.

Morris, W. G., and E. L. Mowat. 1958. Some effects of thinning a ponderosa pine thicket
with a prescribed fire. J. For. 56: 203-209.

Mroz, G. D., M. F. Jurgensen, A. E. Harvey, and M. J. Larsen. 1980. Effects of fire on
nitrogen in forest floor horizons. Soil Sci. Soc. Am. J. 44: 395-400.

Nambiar, E. K. S- 1990. Interplay between nutrients, water, root growth and productivity
in young plantations. For. Ecol. Manage. 30: 213-232.

Neary, D. G., C. C. Klopatek, L. F. DeBano, and P. F. Ffolliott. 1999. Fire effects on
belowground sustainability: a review and synthesis. For. Ecol. Manage. 122: 51-71.

Nelson, R. M. l 952. Observations on heat tolerance of southern pine needles. USDA
For. Serv. Southeast. For. Exp. Sta. Pap. 14. 6p.

Old, S. M. 1969. Microclimate, fire, and plant production in an Illinois prairie. Ecol.
Monogr. 39: 355-384.

Packham, D. R. 1969. Heat transfer above a small ground fire. Aust. For. Res. 5: 19-24.

Palmer, J. G., O. K. Miller, Jr., and C. Gruhn. 1994. Fruiting of ecotrnycorrhizal
basrcromycetes on unburned and prescribed burned hard-pine/hardwood plots after

drought-breaking rainfalls on the Allegheny Mountains of southwestern Virginia.
Myconhiza 4: 93-104.

Pendleton, R. L., and B. N. Smith. 1983. Vesicular-arbuscular myconhizae of weedy and
colonizer plant species at disturbed sites in Utah. Oecologia 59: 296-301.

Persson, H. A. 1983. The distribution and productivity of fine roots in boreal forests.
Plant and Soil 71: 87-101.

55

    
  

 

-.‘?)?3:‘l‘3’,;“‘o‘}:f"k‘i‘I9‘, ”$961.“. "h

 

   

Peterson, D. L., S. S. Sackett, L. J. Robinson, and S. M. Haase. 1994. The effects of
repeated prescribed burning on Pinus ponderosa growth. Int. J. Wildland Fire 4: 239-247.

Pilz, D. P., and D. A. Perry. 1984. Impact of clearcutting and slash burning on
ectomyconhizal associations of Douglas-fir seedlings. Can. J. For. Res. 14: 94-100.

Raison, R. J. 1980. A review of the role of fire in nutrient cycling in Australian native
forests, and of methodology for studying the fire-nutrient interaction. Aust. J. Ecol. 5: 15-
21.

Raison, R. J., P. K. Khanna, and P. V. Woods. 1985a. Mechanisms of element transfer to
the atmosphere during vegetation fires. Can- J. For. Res. 15: 132-140.

Raison, R. J ., P. K. Khanna, and P. V. Woods. 1985b. Transfer of elements to the
atmosphere during low-intensity prescribed fires in three Australian subalpine eucalypt
forests. Can. J. For. Res. 15: 657-664.

Raison, R. J., P. V. Woods, B. F. Jakobsen and G. A. V. Bary. 1986. Soil temperatures

during and following low-intensity prescribed burning in a Eucalyptus pauciflora forest.
Aust. J. Soil Res. 24: 33-47.

Reddell, P., and N. Malajczuk. 1984. Formation of myconhizae by jarrah (Eucalyptus
marginata Donn ex Smith) in litter and soil. Aust. J. Bot. 32: 511-520.

Reid, 1. B., I. Sorensen, and R. A. Petrie. 1983. Root demography in kiwifruit (Actinidia
deliciosa). Plant, Cell Envir. 16: 949-957.

Reifsnyder, W. B., L. P. Herrington, and K. W. Spalt. 1967. Thennophysical properties
of bark of shortleaf, longleaf, and red pine. Yale Univ. School of For. Bull. No. 70. 41p.

Richter, D. D., C. W. Ralston, and W. R. Harms. 1982. Prescribed fire: Effects on water
quality and forest nutrient cycling. Science 215: 661-663.

Roberts, B. A., and A. U. Mallik. 1994. Responses of Pinus resinosa in Newfoundland
to Wildfire. J. Veg. Sci. 5: 187-196.

Roeser, J ., Jr. 1932. Transpiration capacity of coniferous seedlings and the problem of
heat injury. J. For. 30: 381-395.

ROSS, S. M., W. H. McKee, Jr., and M. Mirns. 1995. Loblolly and longleaf pine
responses to litter raking, prescribe burning and nitrogen fertilization. pp. 220-224 in

Edwards, M. B., comp. Proceedings of the eighth biennial southern silvicultural research
conference. USDA For. Serv. Gen. Tech. Rep. SRS-l.

56

 
    

    

e!)?5‘:‘ui';_§i".tt .2:;,,,,.;,...-_,.,. . ....

 

   

Ryan, M. G., and W. W. Covington. 1986. Effect of a prescribed burn in ponderosa pine
on inorganic nitrogen concentrations of mineral soil. USDA For. Serv. Res. Note RM-

464. 5p.

Rygiewicz, P. T., M. G. Johnson, L. M. Gartio, D. T. Tingey, and M. J. Storm. 1997.
Lifetime and temporal occurrence of ectomyconhizae on ponderosa pine (Pinus
ponderosa Laws.) seedlings grown under varied atmospheric C02 and nitrogen levels,
Plant and Soil 189: 275-287.

Rytter, R.-M,, and L. Rytter. 1998. Growth, decay, and turnover rates of fine roots of
basket willows. Can. J. For. Res. 28: 893-902.

Sampson, A. W. 1944. Effect of chaparral burning on soil erosion and on soil-moisture
relations. Ecol. 25: 171-191.

Santantonio, D., and J. C. Grace. 1987. Estimating fine-root production and turnover
from biomass and decomposition data: a compartment-flow model. Can. J. For. Res. 17:
900-908.

Santantonio, D., and R. K. Hermann. 1985. Standing crop, production, and tumover of

fine roots on dry, moderate, and wet sites of mature Douglas-fir in western Oregon. Ann.
Sci. For. 42: 113-142.

Santantonio, D., and E. Santantonio. 1987. Effect of thinning on production and

mortality of fine roots in a Pinus radiata plantation on a fertile site in New Zealand. Can.
J. For. Res. 17: 919-928.

Schoch, P., and D. Binkley. 1986. Prescribed burning increased nitrogen availability in a
mature loblolly pine stand. For. Ecol. Manage. 14: 13-22.

Schoenberger, M. M., and D. A. Perry. 1982. The effect of soil disturbance on growth

and ectomyconhizae of Douglas-fir and western hemlock seedlings: a greenhouse
bioassay. Can. J. For. Res. 12: 343—353.

Schoettle, A. W., and T. J. F ahey. 1994. Foliage and fine root longevity of pines. Ecol.
Bull. 43: 136-153.

Seidel, K. W. 1986. Tolerance of seedlings of ponderosa pine, Douglas-fir, grand fir, and
Engelmarm Spruce for high temperatures. Northwest Sci. 60: 1-7.

Shanow, S. H., and H. A. Wright. 1977. Effects of fire, ash, and litter on soil nitrate,

temperature, moisture and tobosagrass production in the Rolling Plains. J. Range.
Manage. 30: 266-270.

Shearer, R. C. 1975. Seedbed characteristics in Western larch forests after prescribed
bunting. USDA For. Serv. Res. Pap. INT-167. 26p.

57

«VJ;

 

- t of
Shirley, H. L. 1936. Lethal high temperatures for conifers, and the cooling eff“
transpiration. J. Agr. Res. 53: 239-258.

' of Eastern
Smith, D. M. 1951. The influence of seedbed conditions on the regeneration
white pine. Conn. Ag. Exp. Sta. Bull. 545. 61p. _ .
II . Can. J. Soil Sc1.
Smith, D. W. 1 970. Concentrations of soil nutrients before and after fire
50: 17-29.

. _ h during old-field
Smith, D. W., and G. C. Bowes. 1974. Loss of some elements 1n fly as
burns in southern Ontario. Can. J. Soil Sci. 54:215-224.

cfibed burning affect the

Somes, H. A., and G. R. Moorhead. 1954. D0 thinning and pres te No. 34. 2p.

gI‘OWth of shortleaf pine? USDA For. Serv. NEFES For. Res. N0
- h forest. Ecol. 58: 16-30.
Stark, N. M. 197 7. Fire and nutrlent cycling in a Douglas—fir/We

. Root mass, net primary
Steele, S. J" S. T. Gower, J. G. Vogel, and J’ M' Norman- 199fZ>rests in Saskatchewan and
production and turnover in aspen, jack pine and black spruce
Manitoba, Canada. Tree Phys. 17: 577-537.

0 . . 9 a
St. John, T. V., and P. W. Rundel. 1976. The role of fire as a mmerahzmg agent in
Sierran coniferous forest. Oecologia 25: 35-45.

Stone, E. L, and P. J. Kalisz. 1991. On the maximum: extent of tree roots. For. Ecol-
Manage. 46: 59-102.

Sutton, R. F., and R. W. Tinus. 1983. Root and root System ‘emfindogy- For. Sci -
Monograph No. 24. 137p.

Swezy, D. M., and J. K. Agee. 1991. Prescribed-fire effects on fine‘FOOt and tree
mortality in old-growth ponderosa pine_ Can. J, For. Res. 21: 626-634,

Sword, M. A. 1998. Seasonal deveIOpmem ofloblolly pine lateral roots

. in respons
stand densrty and fertilization. Plant and Soil 200: 21-25. e

to

Tiedemann, A. R., C. E. Conrad, J. H. Dieterich, J- W. Hombeck, W. F. Megahan, L
Viereck, and D. D. Wade. 1979, Effects of fire on water: A state-of-knowledge revl'e: '
USDA F or- Serv. Gen. Tech. Rep, wo- 1 o. 28p. W-

Tiedemann, A. R., J. 0. Klemmedson, and E. L. Bull. 2000. Solution of forest health

problems With prescribed fire: are forest productivity and wildlife at risk? For. Ecol.
Manage. 127: 1 _13.

58

Tippett, J. T. 1 982. Shedding of ephemeral roots in gymnosperms. Can. J. Bot. 60: 2295-
2302. -

Tourney, J, W - , and E. J. Neethling. 1923. Some effects of cover over coniferous
seedbeds in southern New England. Yale Univ. School of Forestry Bull. 9. 39p.

Ursic, S. J. 1961. Lethal root temperatures of 1-0 loblolly pine seedlings. Tree P\.N01es
47: 25-29

Vaartaja, O. 1 949. High surface soil temperatures: On methods of investigation, and
thermocouple observations on a wooded heath in the south of Finland. Oikos 1: 6‘28

Van Lear, D. H., J. E. Douglass, S. K. Cox, and M. K. Augspurger. 1985. Sediment
nutrient export in runoff from burned and harvested pine watersheds in the South and
Carolina Piedmont. J. Env. Qual. 14: 169-174.

Van Lear, D. H., J. R. Saucier, and J. G. Williams, Jr. 1977. Growth and wood pr _
of longleaf pine following silvicultural treatments. Soil Sci. Soc. Am. J. 41: 989_9(9)§emes

Vance, E. D., and G. S. Henderson. 1984. Soil nitrogen availability following long-re
burning in an oak-hickory forest. Soil Sci. Soc. Am. J. 48: 184-190. rm
Vance, E. D., G. S. Henderson, and D. G. Blevins. 1983. Nonsymbiotic nitrogen fixation
in an oak-hickory forest following long-term prescribed burning. Soil Sci. Soc. Am. J.
47: 134-137.

Van Rees, K. C. J., and N. B. Comerford. 1986. Vertical root distribution and strontium
uptake of a slash pine stand on a Florida Spodosol. Soil Sci. Soc. Am. J. 40: 1042- 1046-

Vilarifio, A., and J. Arines. 1 991. Numbers and viability of vesicular-arbuscular filngal
propagules in field 801‘ samples after wildfire. Soil Biol. Biochem. 23: 1083-1087.

Villarubia, C~ Ru and 3. L. Chambers. 1978. Fire: Its effects on growth and survival of
loblolly pine. Pinus taeda L. La. Acad. Sci. 41: 85-93.

Vines, R. G. 1968. Heat transfer through bark, and the resistance of trees to fire. Aust. J.
Bot. 16: 499-514.

Viro, PJ. 1974. Effects of forest fire on soil. in Fire and Ecosystems, Kozlowski, T. T.,
and C. E. Ahlgren eds. pp 7-46.

Visser, S. 1995 . Ectomycorrhizal fungal succession in jack pine stands following
Wlldfil'e- New Phytol. 129: 389-401.

Vlamis,.J., H. H. Biswell, and A. M. Schultz. 1955, Effects of prescribed burning on soil
fertility in second growth ponderosa pine. J. For. 53: 905-909.

59

., ‘—

\lmnn.l
afierbrus

Vog.K.

introot
dfiEmnt

ngtK.
477-50
\hfrt

\\:

PFOductj v

\V W‘s-\e - R.
Wndero s;

"N 83.6.: Qg
\Qb\<>\\§

‘94 a‘ldr
0‘: pre
Se“ -1

"V. 1"
“CL;

1‘; a;

53::
i1 55:115.

3'30. /l

i".- .. . - ,_
Mini/5L

ii 6k . C.
in?
//

l
«1.1“!
/ ‘a
t‘.

‘I‘ ‘ In. I
r /r"d[’
I

. far
if

   

Vlamis, J., and K. D. Gowans. 1961. Availability of nitrogen, phosphorus, and sulfur
afier brush burning. J. Range Manage. 14: 38-40.

Vogt, K. A., E- E. Moore, D- 1- Vogt, M. J. Redlin, and R. L. Edmonds. 1983. CmfifeI
fine root and mycorrhizal root biomass within the forest floors of Douglas-fir stands Oi
different ages and site production. Can. J. For. Res. 13: 429-437.

Vogt, K. A., and H. Persson. 1991. Measuring growth and development of roots. pp,

477-501 in Techniques and approaches in forest tree ecophysiology. Lassoie, J . P., and T
M. Hinckley eds. CRC Press, Boca Raton, FL. '

Vogt, K. A., D. J. Vogt, and J. Bloomfield. 1991. Input of organic matter to th

tree roots. pp. 1 71-190 in Plant roots and their environment. Eds. B. L. McMi c
H. Persson. Elsevier Publishers.

e Soil b
had and

Vogt, K. A., D. J- Vogt, and J. Bloomfield. 1998. Analysis of some direct and indi
methods for estimating root biomass and production of forests at an ecosystem levefect
Plant and Soil 200: 71 -89.

Vogt, K. A., D. J- Vogt, E. E. Moore, B. A. Fatuga, M. R. Redlin, and R. L. Edmonds.

1987. Conifer and angiosperm fine-root biomass in relation to stand age and site
productivity in Douglas-fir forests. J. Ecol. 75: 857-870.

Wagle, R. F., and J. H. Kitchen, Jr. 1972. Influence of fire on soil nutrients in a
ponderosa pine type. Ecol. 53: 118-125.

Waldrop, T- A., and F. T. Lloyd. 1988. Precommercial thinning of a sapling-sized
loblolly pine stand with fire. South. J. Appl. For. 12: 203-207.

Waldrop, T. A., D. H. V311 Lear, F. T. Lloyd, and W. R, Harms. 1987. Long-term studies
of prescribed burning in loblolly pine forest of the Southeastern coastal plain. USDA For.
Serv. Gen. Tech. Rep. SE-45. 23p,

Wells, C. G. 197 1. Effects of prescribed burning on soil chemical properties and nutrient

availability. pp. 86-99 in Prescribed burning symposium proceedings. USDA For. Serv.
Southeast. For. Exp. Sta., Asheville, NC.

Wells, 0 G» R. E. Campbell, L. F. DeBanO. C. E. Lewis, R. L Fredriksen, E- C-
Franklin, R. C. Froelich, and P. H. Dunn. 1979. Effects of fire on soil. A state-of-
knowledge mView. USDA For. Serv. Gen. Tech. Rep. WO-7. 34p.

Wells, (3' Grand C. B. Davey. 1966. Cation-exchange characteristics of forest floor
matenals- 5011 Sci. Soc. Am. Proc. 30: 399-402.

White, C S. 19.86- Effects of prescribed fire on rates of decomposition and nitrogen
madman” 1“ a Ponderosa pine ecosystem. Biol. Fertil. Soils 2: 87-95.

60

«Ht 1 t. .b 1 fit. \ . . .... u t . u... . 1 la‘ ri\ 1 .l. 7 . . . .. v .
HM Wm ”W U . at..- \h xhc . v t Jr . i. .s . it A 44.. ...h nu ch.
1....» -.mu. ..nmm mm Un \M... “9. w\ ,0 v . Dr.

 

Wicklow-Howal'd, M. 1989. The occurrence of vesicular-arbuscular myconhizae it;
burned areas of the Snake River Birds of Prey Area. Idaho. Mycotaxon, 34; 253—25 -

Widden, P., and D. Parkinson. 1975. The effects of a forest fire on soil microfufigl- SO“
Biol. Biochem- 7: 125-138.

Wienhold, B . J ., and J. O. Klemmedson. 1992. Effect of prescribed fire on nitrogen and
phosphorus in Arizona chaparral soil-plant systems. Arid Soil Res. and Rehab. 6: 285-

296.

Wright, E., and R. F. Tarrant. 1958. Occurrence of myconhizae after logging and

s
burning in the Douglas-fir forest type. USDA For. Serv. Res. Note PNW-16O - lash

Wright, H. A. 1970. A method to determine heat-caused mortality in bunchgraSSe
51.- 582-587. 8' ECO]-

Wright, H. A., and A. W. Bailey. 1982. Fire ecology: United States and southern
Canada. John Wi 1ey and Sons, New York, 501p.

Wright, R. F. 1976. The impact of forest fire on the nutrient influxes to small lakes in
northeastern Minnesota. Ecol. 5 7: 649-663.

Wright. R- J., and S. C. Hart. 1 997. Nitrogen and phosphorus status in a ponderosa pine
forest after 20 years of interval burning. Ecosci. 4: 526-533.

Wyant, J. G., R. D. Laven, and P. N. Omi. 1983. Fire effects on shoot growth
characteristics of ponderosa pine in Colorado. Can. J. For. Res. 13: 620-625.

Zahner, R. 1989. Tree-ring series related to stand and environmental factors in south
Alabama longleaf pine stands. pp. 193-197 in Miller, J. H., Jr. comp. Proceedings of the
fifth biennial SOUthem 811Vicultural research conference. USDA For. Serv. Gen. Tech.
Rep. SO-74.

6l

 

 

Chapter 2
Effects of fire on available nutrients

62

r»; rt I K?
\MN \‘1

mm
1.3 {a i“

If {1)

0E if

Introduction

in ' The benefits of prescribed fire include reduction of hazardous accumulations 0f

fuels, reduced competition from less fire-tolerant plants, an increase in sprouting 1“

certain hardwood species — which is beneficial for wildlife — and an increase in available
nutrients. Low-intensity burning often has a “fertilizer effect” on the community. The
level of this nutrient increase is usually proportional to the amount of fuel burned. Based
on this, I attempted to evaluate the nutrient dynamics of the soil in a red pine (Pinu

s

resinosa) stand before and after a low-intensity prescribed fire. This study Was in

conjunction with a larger project which studied the effects of prescribed fire On Certa‘
in

physiological aspects of red pine, in particular, fine-root turnover.

Hypotheses

The hypotheses for this part of the experiment were:
1. A low-intensity prescribed fire in red pine will result in an increase in available baSic
cations (Ca, Mg, K).
2. This increase in basic cations will result in a decrease in H ion concentration

(increased pH).

3. Cation concentrations will return to prefire levels within two growing seasons.

Site description

The study site is located at the WK. Kellogg Experimental Forest of Michigan
State University near Augusta, MI. The site was planted in 1950 with 2-1 red pine

seedlings on degraded farmland at a 3 x 3m spacing, Over the next two years, seedlings

63

 

ll ti.” )1
$231.0
\‘ ““1

MN

50.?)

55:31

that dled were replaced. The Stand was originally part of a thinning study FIVC p10ts 0f
0 06 ha each were established in 1994 for both the Burn and Control treatments. 1310‘s
were selected based on amount of understory (little to none) and blocked based on
prevrous {tnnnlng treatment and overstory density. Directed application of glyphosate
occurred where understory vegetation had established (mostly herbaceous plants and
some woody Vines) ThlS ensured that nearly every root that was sampled was fI‘Om re d
pine. As of 1990, basal area ranged from 32. 2— 55. 2 mz/ha (140 to 240 ftZ/A) With qu

23. 6— 28. 4 cm (9. 3 — l 1.2 in). Site index is approximately 19. 8 m (65 feet) at 50

ears.
Soils are a Kalamazoo loam and an Oshtemo sandy loam (Typic Hapludalfs) (3011
Conservation Service 1979).

Methods

Statistical design

The experiment was set up as a Randomized Complete Block with two
treatments burn and control and 5 replications per treatment. The blocks were split by
depth The data were analyzed over time using a repeated-measures analysrs (Littell et
3‘ 1996111“? prefire data Were analyzed to confirm that there were no a priori differences
between the treatments Where interactions were found, unadjusted p-values were
calculated usrng the ‘SLICE’ option of SAS’s Proc Mixed. Adjusted p-values were
calculated 118mg TUkeY’S HSD test. An alpha level of 0. 1 0 was used because of the

extremely hlgh variability found in studies of soil and roots (c f Steele et a1 1997)

64

 

r ll
“it‘ll“

l
7; m1
,, ml
41
r .‘,
(5:qu
new

'5“. VA“

5m
Llll. .

Field
Beginrling in the fall of 1996, soils were sampled at regular intervals to a depth 05

25 cm. Soils vvere sampled on four dates before burning and nine dates postfiro- Six
randomly-located cores of 4.76 cm (1-7/8 in) diameter were extracted from each
treatment plot. Each core was separated into the following depths: forest floor (FF), 0_2
em, 2-6 cm, 6— 1 0 cm, and 10-25 cm. At each depth, the six cores were composited within
each plot to reduce variability. Half of the composite (at each depth) was reset-Ved for
analysis of fine roots while the other half was used for nutrient analysis. Each COmPOsite
was air dried and then sieved to pass a 2-mm screen. Because the 10-25cm depth Was not
sampled after 1997, none of the data from this depth was used in statistical analyses,

Soils were sampled on four prefire dates and nine postfire dates (Table 2-1).

Table 2-1. Dates on which soils and roots were sampled.

 

 

[Prefire Sample DateL Pmre Sample Dates

 

 

 

   
 
   

September 1996 July 1997
October 1996 August 1997
April 1997 September 1997
May 1997 OCtober 1997

November 1997
May 1998
June 1998
August 1998

‘ October 1998

 

Forest floor (FF) samples were taken during the second post-burn growing season
(1993) to dotermine the amount of litter that burned and the amount of duff that

decomposed following the fires. Three samples, 232 cm2 each (36 inz), were taken from

65

 

1 -‘ 5'
(sin

(I.

i
l
mu)

'4‘ n
is.)

l“ \

(

-. I:
".1

,2
.r b

ere

each plot. Each sample was divided into litter (L) and duff (F /H layers). Samples W

dried at 70 °C for 2-3 days, then weighed.

The plots were burned with low-intensity backfires on June 9 and 10, 1997-
Because the plots were not contiguous within the stand, each plot was burned Separately.
Flamelengths were low as was the rate-of-spread (Table 2-2). On occasion, the Wind
shifted, turning the fire to a headfire. However, these episodes usually didn’ t last more

than about one minute before returning to backfire conditions.

Table 2-2. Weather, fuel and fire characteristics, June 1997.

 

    
  

 

 

 

 

 

 

 

 

weather Condi tions Fire Characteristics

Sunny ' Fuel (litter) loadings 4700-6100 kg/ha
Temperature 24-2 8°C Rate of spread 0.1 m/min
Windspeed 0.7-1. 8m/s F lamelengths <0.3m backfire
Relative Humidity l7-25% 0.6-0.9m headfire
Days since last rain 7-8

Amount of rain 097cm (0.38in)

Laboratory

80“ PH was measured using a 1:1 solution with distilled, deionized water. Basic

cations (Ca, Mg, and K) were extracted using an ammonium acetate solution (1.0 N, pH
7.0). The extract was then analyzed using Direct-Coupled Plasma — Atomic Emission
Spectroscopy .
A Plot of residuals indicated that the variability increased with increasing
concentration of the basic cations (e.g., Figure 2-1). The data were then log-transformed.
In all cases, the transformation eliminated this pattern. Because pH is measured on a log

scale, the data used for the original analysis was the concentration of the hydrogen ion

[W] in the soil,

66

~flr1

300

 

 

 

 

 

 

 

zoo — —
100 —
a
3
.2 o e
m
m 0
n‘ -100 a...”...---”_._--—e—~—¢!oe—_—..w-—94—+-m— - —— e -
. ~ .
-200 “L‘s”--- *. ...... -_ -.-. _ ___- u ,_ - M __ - __- n ._
O
-300 . ~ . , . . J
o 100 200 300 400 500 600 700
Predicted [eazt]

Figure 2-1. Residuals of the prefire calcium data. The increase in the spread of residue]
values, as calcium concentration increased, indicated the need to transform the data,
Similar results were found for Mg and K, both pre- and postfire.

Results

Prefire

Among the main effects, there were no prefire differences between the treatments
(Table 2-3), but there were significant differences across depths for all cations. Cation
concentration decreased with depth for all cations (Figure 2-2). There was no effect of
either time or treatment on cation concentration.

Although there is a treatment x depth interaction for each basic cation, this
interaction is probably due to the main effect of depth. Within each depth, there was no
significant difference between treatments, but within each treatment, there were
significant differences among the depths. This interaction was not eliminated by use of

the log transformation, but the main effect of depth was still the cause of the significant

67

Ti

”PM. 7”.

Ti

. ° C ' ' ° - + .
interaction. The prefire ngnlé ant depth x mm interaction for the hydrogen ion (H ) 18

mainly a result of the large (no 1‘ ease in H concentration in October 1996 (Figure 2-3).

Table 2-3. Prefire soil cation p'Values. Bold values are significant at or = 0.1 O.

 

 

 

 

 

 

 

 

 

ffect Ca Log (C3) [Mg L08 (Mg) L08 (K)
Treatment 0.720 0.762 .948 0.91 7 0.846 0.942
Depth <.001 <.001 .001 <.001 <.001 <.001
Trt x Depth 0.001 0.006 .005 0.005 0.058 0.029
Time 0.972 0.837 .936 0.947 0.162 0.245
Trt x Time 0.883 0.914 .832 0.943 0.832 0.595
Depth x Time 0.137 0.010 .110 0.068 0.055 0.019
Trt x Depth
x Time [0.904 0.595 .885 0.555 0.937 0.682

600

 

 

 

 

 

 

Calcium. [Ca2+] (ppm)

 

 

 

O-2cm

 

2-6cm

6-10cm

Figure 2-2. Prefire soil calcium concentration, across depths in two experimental
treatments. Because there was no significant difference for the main effect of time, all
four prefire sample dates are combined. Similar results were found for magnesrum and

potassium. Error bars indicate standard error; 11 = 20.

68

 

":3 7.026}:
, , + 2—6cm
+ 6- 1 Gem

 

 

1.5E-05.. —————- flew—re H___ we h__hs_... “.7- __.

 

 

 

1 .OE-OS .
Sept96 Oct96 Apr97 May97

Figure 2-3. Prefire H+ by depth. Data from both treatments are combined. Error bars
indicate standard error; 11 = 10.

Use of the log transformation revealed two additional significant differences (Ca
and Mg, depth x time) (Table 2-3). For Ca and Mg, there were no Significant differences
between treatments within any depth. For K, statistical output indicates a significant
difference among times for the shallowest depth (0-2 cm) (Figure 2'4). The pattern 0f
changing cation concentrations displayed in Figure 2-4 was similar for all three cations.
Concentration increased at the 0-2 cm depth from September 1996 to 0010b“ 1996, and

peaked the following spring.

69

3“
Few
You .rbu-a-‘ifl

l‘u F

\

pron

 

H .- . _ _ -- :e_0-2cmi
.+2-60m
,,_ ,- ._- +6-10cm

 

 

 

 

 

 

 

“1.6 2—4 - Prefire potassium by depth, time. The significant depth x time interaction is

E igstly due to the main effect of depth. However, within the 0-2 cm depth, there is a
@ignificant difference among the sampling dates. Error bars indicate standard error. 11 =
’30. 7
Postfire

There were no postfire differences in cation concentrations between treatments
(Table 2-4), but cation concentrations still decreased with depth (Figure 2-5). There was
an especial 131 large drop in cation concentrations at 0-2 cm from May to July 1997 (e.g.
Ca, Figure 2 -5) followed by a slow increase through the rest of the year. Cation
concentrations increased from November 1997 to May 1998. This increase was more
pronounced in the Burn plots than in the Control plots. After the fire, time was a
significant factor for the basic cations. Depth was the only significant main effect for H.
However, there was a depth x time interaction for all cations that was mostly due
to the main effect of depth. Cation concentrations were more variable across time at 0-2

c m than at 10 Wer depths (Figure 2-5). The log transformation eliminated this source of

70

Ti

:1 :H.

TL

Variation for Ca, K, and H, and reduced it for Mg (Table 2-4). The treatment x depth

interaction was not significant for the basic cations after the fire.

Table 2-4. Postfire soil cation p-values. Bold values are significant at a = 0.10.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

fi'ect a ~ Log (Ca) Mg Log (Mg)lK Log (K) [H pH
reatment 0.665 0.764 0.598 0.668 0.326 0.306 0.116 0.156
epth <.001 <.001 <.001 <.001 <.001 <.001 <.001 0.081
rt x Depth 0.796 0.488 0.615 0.139 0.195 0.724 0.674 0.348
Time 0.016 0.025 <.001 0.001 <.001 <.001 0.331 0.260
Trt x Time 0.312 0.111 0.305 0.104 0.687 0.560 0.222 0.443
Depth X Ti I-ne 0.001 0.221 <.001 0.040 0.016 0.446 0.014 0.251
Trt xDePth
Fiji/me/ 0.172 0.725 0.012 0.089 0.885 0.850 0.902 0.965
600
5oo __._.. ... _ £222
E 400 4%.i’ ii {\i ._ /£\fi{ Q}; “I—
S \f,E:i "f I “:3 3.763%
9'3, 30° ‘rx—e—we ---—-—- «e — m .— — £+2-5cm
é ‘—‘—6-10cm
55,200. ... _ ‘. ii
3 in”.
0:6

 

 

’\ '\ '\ $92 $9.31:
@9088, O".9 .359 ‘35? 090% 089%
Sample date

'\ «

F 1g age 2‘3. Pre- and postfire soil calcium by depth and sample date. Both treatments
cor; lned because of no significant difference between treatments. Arrow indicates date
of 1re treatments. Error bars indicate standard error. n— = 10. Similar results were found

for magnesium and potassium.

71

The three-way interaction for Mg is slightly more complicated. This interaction is
Jnainl y due to the effects of the two-way depth x time interaction. As stated above, the
depth x time interaction is mostly due to the main effect of depth. Through all
combinatiom of depth and sample date, treatment was significant only at the 0-2 cm
depth during September 1997. With the log transformation, treatment was significant at

2-6 cm and 6-10 cm during September 1997.

F crest F100 r mass
Approximately 1800-1980 kg/ha of litter was burned in the fires, and an
additional 2 600 kg/ha of duff was decomposed following the first postfire winter (Figure

2-6} There were significant differences in forest floor weights by treatment, by depth,

1 6000

14000 - i~———-~~———- ~——»- —~—~r~ — - -—~—-——— ~ H_nm
12000 ,,_ , .. , . _ . _ . ...“ --.» ,

 

 

 

E r—"“—“* 2....“ — e —l
3-
£ 10m0 .1? .. .._._ ____ _ __ __ _i T __ r'——-.‘-—-—r-- - ~--——
3:3 ______ g—Q—COMIOI L
9 "' T ’ _ +Control FIH i
3 800° - - -- a -» ~ ’ 2 v .
é -— Q — Bum L
c _ :- . - Burn F/H” __
‘6':
2
O
LI—

  

 

 

June98 August98 October98
Sample date

1,:- igure 2‘6- Mass of the Forest Floor during the second postfire growing season. The
11... tter lay

998 er is labeled “L” and the duff is “F/H”. Samples for the Control plots from June
1 were 2“'l'fillyzed improperly and are not included.

72

and over time (Table 2-5). Using values from the literature for red pine forest floor

cation concentrations (Table 2-6), relatively small amounts of nutrients were made

available (Table 2-7). However, these represent 25-3 5% of the nutrient capital of the top

2 cm of mineral soil (Table 2-8).

Postfire FF p-values. Bold values are significant at or = 0.10. Because the

Control data from June 1998 were suspect, they were not included in this analysis.

 

 

 

Table 2-5.

W FF weight
Treatment 0.005
Depth <.001
Treatment 3* Depth 0.068

0.008

' C
ilgatment >< Time 0.358
Drepth 7‘ Time 0.038
Treatment >< Depth x Time 0.443

 

 

 

 

.{ able 2—6- Red pine FF weights and cation concentrations from the literature.

 

W

 

(1982)

 

 

 

 

 

 

 

  

(1975)

 

 
 
 

 

 

This Study

   

 

 

 

 

 

FF weight % Ca % Mg % K Notes
(kg/ha)
Wd Alban 29390 1.0 0.15 0.13 Sandy soils
Win et al. (1986) 20670 0.77 0.068 0.08 cent. Wisc.
M0982) 46300 1.4 0.14 0.13 n. Minn.
34800 0.8 0.11 0.08 n. Minn.
We Cote (1994) 15600 0.82 0.11 0.048 40-year-old
__ in s. Quebec
Tappemer and Alm 15000 0.51 0.06 0.23 n. Minn.
Bockheim et al. (1989) 1440021000 0.51-0.59 0.065- 0070 Wisconsin—
0.078 0.091 5 sites
0.73 0.09 0.10
15532 May 1998
estimate

 

73

 

Table 2-7,

Potential amounts of cations released over the course of two postfire growing

seasons, from burned litter and decomposed duff. Values in parentheses are those
calculated using the assumption that all the litter that burned was freshly-fallen, and had
not Weathered at all.

 

 

Amount of cation released (kg/ha)

 

 

 

    

 

 

 

  

 

 

 

Oct. 97, Litter May 98, Duff Total
Calcium 14.5 (10.3) 19.1 33.6 (29.4)
Magnesium 1.78 (2.16) 2.36 4.14 (4.52)
Potassium 1.98 (4.66) 2.61 4.60 (7.27)

 

 

 

Table 2’8-
mineral

':l"otal amount (standard error) of available nutrients in the top 10cm of
so i 1 (average of all prefire samples), and the potential amount of cations released

by the fire amd subsequent breakdown of the forest floor, as a percentage of that total.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Nutrient Depth Average (kg/ha) Amount released Percentage of pre-
from FF fire quantities
Ca 0-2 cm 97.4 (6.0) 33.6 34.5%
2-6 cm 109.9 (8.8) 30.6
6-10 cm 75.0 (7.3) 44.8
Total 282.3 (20.6) 11.9
"Tori", 0-2 cm 11.8 (0.7) 4.14 35.1%
2-6 cm 14.8 (1.1) 28.0
6-10cm ll.0(1.0) 37.6
/_ Total 37.6 (2.7) 11.0
7:” 0-2 cm 18.0 (0.5) 4.60 25.6%
2-6 cm 26.2 (0.7) 17.6
6-10 cm 19.3 (0.4) 23.8
L/" Total 63.5 (1.3) 7.2

 

 

 

 

 

 

DichSSion

The absence of a significant prefire difference in cations and H+ between

treatments indicates that the blocking used in this experiment was appropriate. Although

there was a S ignificant treatment x depth interaction for each of the basic cations, it was

it’lOStly due to the effect of depth. Although there appeared to be a difference between

74

 

treatments at the 0-2 cm depth (e.g., Ca — Figure 2-2), results of Tukey’s HSD test
indicated no such difference (data not shown).
The lack of a “spike” is difficult to explain. While nutrients can be lost via
volatilizati on and convection of ash during a fire, cations are usually not removed in
these ways - Vaporization temperatures for inorganic forms of these elements range from
759°C (K) to 1484°C (Ca) (see Lide 1997). If it is assumed that negligible amounts of
s we: re removed from the site by these mechanisms, then those that were released

cation

during burn ing and through increased breakdown of the forest floor (Table 2-7) represent

2 53 50/0 of the nutrient capital of the top 2 cm of mineral soil (Table 2-8). One might
expecl‘ an increase of this size to be detected; however, less than half of the potential
increase W0 111d have been “instantaneous” because of the fire. The rest would have been
¢eleased slowly through increased breakdown of the forest floor.

When compared to the entire 10 cm depth, the total amount released from the
forest floor was only 7-12% of that which was available. Following prescribed fires in
slash pine stands in Florida and Georgia, the amount of cations remaining in the residual
ash represented 7.5-13.6% of that available in the top 10cm of mineral soil (Hough
198 1 ), KDdama and Van Lear (1980) also concluded that nutrient quantities lost from the

forest- floor through burning are small compared to amounts in the residual forest floor
and 5017 (although they don’t report any soil nutrient quantities).

7716 arnount of litter and duff that had accumulated in this stand was near the low

and 0f the rEl-I‘lge reported in the literature (Table 2-6). Using fuel accumulation equations

for red pine (LaMois 195 8, Dieterich 1963), this stand should have had up to twice as

[filUCh 11tter as was actually On-site (data not shown). Low litter amounts will result in

75

relatively low nutrient accumulation in the fuels, and therefore only a small amount
released following burning. Incomplete combustion will also result in the release of only
a small portion of nutrients contained in the fuels. However, backfires (used in this
experiment) usually result in more complete combustion of surface fuels than do
headfires (DeBano et al. 1998).

Nutrients that are released by fire may also be lost from runoff or deep percolation
following rains. An increase in available nutrients might not be detected if the rain
occurs before the first post-fire sample is taken (cf. Sharrow and Wright 1977). A
rainstorm dropped 1.6cm (0.62in) of precipitation on this site approximately 1 week after
the fires. A larger storm dropped 6.0cm (2.37in) of rain four days later. This may help to
explain the large drop in cation concentrations between May and July 1997 (Figure 2-5).
Runoff probably did not occur on this site, as it relatively flat. Although the results have
only been shown for the top 10cm of mineral soil, samples were taken to 25cm.
However, there were no differences in cation concentrations between treatments at the
10-25cm depth (results not shown). Water from the second (larger) storm probably
percolated below this depth. In a separate study on this site Gerlach (2001) found that the
soil contains 81-91% sand to a depth of 1.5m in two of the control plots (loamy sand to
sand). The Soil Survey (Soil Conservation Service 1979) classifies these soils as well-
drained with moderate to moderately-rapid permeability. If leaching did occur, K is the
cation most-susceptible to loss as it is highly mobile (Grier 1975, Nissley et al. 1980,
Bockheim and Leide 1986).

There was no increase in soil pH (decrease in soil H) following burning.

76

 

i.
\:
fit
a
'9.
p0
,
C
i?
1
l
i
l'
i
C

ll
r:
Kr
V‘

 
 

While soil pH usually increases as a result of burning (DeBano et al. 1998), several
researchers have found no increase in soil pH following burning (Christensen 1977,
Kovacic et al. 1986, Ryan and Covington 1986, Masters et al. 1993). Increased pH is
often associated with an increase of basic cations following burning (Alban 1977, Wells
et al. 1979). Because no such increase was observed in this study, the lack of change in
pH is not surprising.

Foliar nutrients were not measured in this study. Given the amount of nutrients
potentially released by these fires (Table 2-7), foliar nutrient concentrations would
probably not have been affected. In fertilization studies of red pine, the amounts of
cations applied are much larger than the potential amount released by fire. For example,
Bockheim et al. (1986) applied lOOkg/ha of K and 100 kg/ha of N to a red pine plantation
in Wisconsin. After 4 years they did not find any differences in foliar K concentrations
between control and fertilized plots. A long-term experiment in New York tested K
fertilization in red pine plantations on a sandy outwash plain. Amounts of added K
ranged from 56 to 560 kg/ha (Nowak et al. 1991). Based on earlier results of that
experiment, Heiberg et al. (1964) recommended 120 kg K/ha as the optimal rate for
efficiently maximizing growth of red pine on these sites.

F oliar nutrient concentrations of herbaceous vegetation have often increased
following prescribed burning (e.g., DeWitt and Derby 1955). Anderson and Menges
(1997) also found increases in foliar nutrient concentrations but no detectable increase in
soil nutrient availability. They concluded that foliar nutrient concentrations were higher
only because the plants on burned plots were growing quickly and recovering from fire;

differences dissipated with time since burning. Foliar nutrients have increased after

77

 

 

 

3131‘

burning in quaking aspen stands (James and Smith 1977, Johnston and Elliott 1998).
However, following a wildfire in red pine stands in Newfoundland, foliar nutrients
increased, decreased, or were unchanged, depending on the site (Roberts and Mallik
1994).

An additional source of variation in this study was the fire itself. Fire
characteristics differ among fires and even within a fire (Rouse 1988). Although I
attempted to treat all plots exactly the same, there were some differences in temperature
and relative humidity while burning the different plots. Also, there were a few instances
where the wind shifted, and the backfires turned to headfires, even though these episodes
only lasted for about a minute. The fires were low-intensity which could have resulted in
incomplete combustion, which would also result in low instantaneous nutrient release.

In general, soils are highly variable (Table 2-9), and those sampled in this study
are no exception. The relative variabilities of cation concentrations found in this study,
as measured by coefficients of variation (CVs), tend to be higher than those typically
found in the literature for Ca and Mg. For K, the CVs tended to be lower in this study but
were within the range reported. Although absolute variability as measured by standard
error (e.g., Figures 2-4 and 2-5) was higher at the shallower depths, relative variability as
measured by CV usually increased with depth. The only exception was that prefire K had
lower CVs at the deeper soil depths. The variabilities of foliar nutrient concentrations
(Table 2-10) are comparable to those found in soils (Table 2-9).

Other researchers have found increases, decreases, or no change in available
nutrients following single burns (Table 2-11). Possible reasons for the lack of increase in

soil cations following prescribed burning include highly buffered soils (Smith and Bowes

78

 

 

t. . . ..--....,~.~ ..~:.~...~m\.\.:». V ..\.In,.. .v\.\s.~\.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Aoév >0 dough; mo 620580

 

 

v.8 9% So 2:28 .283

v.2 2% gm 2.35 .58;

.Eae SN 3:. 2m ”Ema as:

8% :ocfifiooag Q: mdv fl; 2:05 .834”

was .85 ones v.2 92 ..mm 258a .53
=m mmouom mcwfiu>< 8E v“: Q: m.wm mam Suva .Eomd 32% m2...

2% v.5. 3m 3n n: 635%:

Q: 92 SN 98 2 Saga:

Nam tam SM 2:. EN owes;
«£3 a 9% a Na 3N EM ”.3 an $8:
5.5 a «a; Ea we :3 v.5. ”.3 as. at .83 82 as Eaafi

saga v.3 cam SN :3 5 803-2
2 353% 2i we «.2 2: :a 3m o. a as; £3 a a a :22

N. a EN o. a Sm m

EN 2m 3m 3m 502.:
Ba 3332 a. a N. a m on” 0.8 we song 68: a a so:

82:: . 2a a 3 8 mm :
cascoeem . 2a 2 2 a 8 E”

amass: I 2% 8a we 2 mm a ; om 532$ 2.8: 532
852 860% m w: ‘No a 2 snow :8 €sz

 

 

.=8 2: 5 35:5: 32.5» “8 Am>Uv sonata? we 3565000 .o-~ 053.

79

Table 3-10.

__-_________,_.

Study
Nowak et
al.1991
Bockheim
eta].1986
lletzetal. -~
(1966) 1
Alban e1
£41078)

 

1974). the t
rapid immo
possible lea
Conrad 197
decreased. t
1986). If it

Chfipter 3')_

The
mm? in 3V“-

L“13H. Begl

 

Table 2-10. Variability of foliar nutrient concentrations.

 

 

 

 

CV (%)
Study N P Ca Mg K Species Notes
Nowak et 6.7 - red Fertilized and unfertilized
al. 1991 34.3 pine over several decades
Bockheim 6.5 6.0 10. 7 4.9 13 .5 red Control

etal. 1986 3.8 4.7 8.0 4.9 8.8 pine FertilizedwithNandK
Metzetal. 10.1 10.6 26.9 23.0 18.4 loblolly

 

 

 

 

 

 

 

 

 

 

 

(1966) pine
Alban et 21.8 22.0 30.1 15.4 26.8 red
al. (1978) pine

 

1974), fire effects were secondary to seasonal changes (Anderson and Menges 1997),
rapid immobilization of nutrients in plant or microbial mass (Masters et al. 1993),
possible leaching (Masters et al. 1993), and losses from erosion and runoff (DeBano and
Conrad 1978). Where root mortality has occurred during a fire, nutrient uptake may be
decreased, thus increasing the amount of available nutrients (Covington and Sackett
1986). If it is assumed that little or no root mortality occurred because of this fire (see

Chapter 3), then no change in available nutrients would be expected by this mechanism.

Conclusions

The results of this study did not confirm the original hypotheses. There was no
spike in available basic cations following burning, and there was no concomitant increase
in pH. Because there was no increase the third hypothesis was moot. While there was no
increase in available nutrients, soils were highly variable and leaching of nutrients below
the depth of sampling was a definite possibility. Although soil N and P were not
measured, significant differences between treatments would probably not have been

found as variabilities of soil N and P are similar to those found for cations (Table 2-9).

80

‘1.

 

 

» Ill'li.l- ‘ ‘ ~ P .‘\F‘.N.‘

 

 

 

. Z Eco «.2382: 85 “”3me :28ng 1:38 03 205 ”3oz

95888 8 @6385 823a countenance 5?» .338. 03min.» n 90% .8 22:
:ozabcoocoo E053: E 835% u out
aosgcoocoo E052: E 93205 n 2:

cop—Boga o: n o

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

52:6 588%
2.0 o o o as: .502 3523 \wsam 8a Ex 3% m5
oEQ
o o EEofi :8:
mod Bow 3% o o 28996 .853 -xmo 852 .3 no E232
25.68; 35: fig C 3322
Ed o o o obi o Eom To 35 5:55 :Encam Ea c8523
258m
EA; .0 ..oE-N mEchSE uofiEzm $83 $25».—
.EBEBEEH o o o o Eco E .Eow To Buenos \wctmm mEoc-EO can 556
54% $8 3 RES
mod o o o C out Sam I 8:5 .2556 3:329 98 ocmmoo
20:3
2: 2: 2: Eomé SE55 05m xofl. 8.3: .EEm
o o o o o E + m NE 2550: macaw Bag 8%: 33 cm>
mod 0% out out 8c 0% 4-: 35 23 .23qu was «635%
Samoa .084. C out out out 0% mm 2:93: 059 A33 3 28.0mm
258m .08; o o o o o ”E 9ch =£ «8595; can coqu>oU
Z out: Eowmxmoom
882 d v. m: «U A Each Sana 2E :Omaom x8625 33m

 

 

.mEBmzmooo mo but? a E 35 near—82a .szZuE wEBo=£ 2238:0200 325:: E mom—6:0 ._ TN 035,—.

81

 

Literature

Alban. D. 1
Minnesota.

Alban. D. l I
pine. ['SD.“

Alban. D. 1
properties.

 

   
  
 
 
  
  
  

1le. D. l
in aspen. pi
390-299.

Anderson.
mFConhi

Bockheim.
CO-Ittpositio
\\ 15COIISln.

Bockheim
maCIOnmri
'1' F01 Res.

CIfiSIensex
W plai

CO‘JHQIOn
6"x

CC"Ingram

 

 

 

 

 

Literature Cited

Alban, D. H. 1974. Soil variation and sampling intensity under red pine and aspen in
Minnesota. USDA For. Serv. Res. Pap. NC-106. 10p.

Alban, D. H. 1977. Influence on soil properties of prescribed burning under mature red
pine. USDA For. Serv. Res. Pap. NC-139. 8p.

Alban, D. H. 1982. Effects of nutrient accumulation by aspen, spruce, and pine on soil
properties. Soil Sci. Soc. Am. J. 46: 853-861.

Alban, D. H., D. A. Perala, and B. E. Schlaegel. 1978. Biomass and nutrient distribution
in aspen, pine, and spruce stands on the same soil type in Minnesota. Can. J. For. Res. 8:
290-299.

Anderson, R. C., and E. S. Menges. 1997. Effects of fire on sandhill herbs: Nutrients,
mycorrhizae, and biomass allocation. Am. J. Bot. 84: 93 8-948.

Bockheim, J. G., J. E. Leide, and L. E. Frelich. 1989. Red pine growth and chemical
composition of foliage and forest floors across a precipitation-chemistry gradient in
Wisconsin. Can. J. For. Res. 19: 1543-1549.

Bockheim, J. G., J. E. Leide, and D. S. Tavella. 1986. Distribution and cycling of
macronutrients in a Pinus resinosa plantation fertilized with nitrogen and potassium. Can.
J. For. Res. 16: 778-785.

Christensen, N. L. 1977. Fire and soil-plant relations in a pine-wiregrass savanna on the
coastal plain of North Carolina. Oecologia 31: 27-44.

Covington, W. W., and S. S. Sackett. 1984. The effect of a prescribed burn in
southwestern ponderosa pine on organic matter and nutrients in woody debris and forest
floor. For. Sci. 30: 183-192.

Covington, W. W., and S. S. Sackett. 1986. Effect of periodic burning on soil nitrogen
concentrations in ponderosa pine. Soil Sci. Soc. Am. J. 50: 452-457.

DeBano, L. F., and C. E. Conrad. 1978. The effect of fire on nutrients in a chaparral
ecosystem. Ecology 59: 489-497.

DeBano, L. F., Neary, D. G., and P. F. F folliott. 1998. Fire effects on ecosystems. John
Wiley & Sons, New York. 333p.

DeWitt, J. B., and J. V. Derby, Jr. 1955. Changes in nutritive value of browse plants
following forest fires. J. Wildl. Manage. 19: 65-70.

82

 

Dieterich, J. H. 1963. Litter fuels in red pine plantations. USDA For. Serv. Res. Note
LS-14. 3p.

Fyles, J. W., and B. Cote. 1994. Forest floor and soil nutrient status under Norway
spruce and red pine in a plantation in southern Quebec. Can. J. Soil Sci. 74: 387-392.

Gerlach, J. P. 2001. A comparison of productivity and related traits for European larch
(Larix decidua Miller) and red pine (Pinus resinosa Ait.) across a site quality gradient in
the Great Lakes region. Unpublished Masters thesis. Michigan State University, East
Lansing, MI.

Grier, C. C. 1975. Wildfire effects on nutrient distribution and leaching in a coniferous
ecosystem. Can. J. For. Res. 5: 599-607.

Hough, W. A. 1981. Impact of prescribed fire on understory and forest floor nutrients.
USDA For. Serv. Res. Note SE-303. 4p.

James, T. D. W., and D. W. Smith. 1977. Short-terrn effects of surface fire on the
biomass and nutrient standing crop of Populus tremuloides in southern Ontario. Can. J.
For. Res. 7: 666-679.

Johnston, M., and J. Elliott. 1998. The effect of fire severity on ash, and plant and soil
nutrient levels following experimental burning in a boreal mixedwood stand. Can. J. Soil
Sci. 78: 35-44.

Kodama, H. B., and D. H. Van Lear. 1980. Prescribed burning and nutrient cycling
relationships in young loblolly pine plantations. South. J. Appl. For. 4: 118-121.

Kovacic, D. A., D. M. Swift, J. E. Ellis, and T. E. Hakonson. 1986. Immediate effects of
prescribed burning on mineral soil nitrogen in ponderosa pine of New Mexico. Soil Sci.
141: 71-76.

LaMois, L. 1958. Fire fuels in red pine plantations. USDA For. Serv. Sta. Pap. LS-68.
19p.

Lide, D. R. (Ed) 1997. Handbook of chemistry and physics. 78th edn. CRC Press, Inc.
Boca Raton, Florida.

Littell, R. C., G. A. Milliken, W. W. Stroup and R. D. Wolfinger. 1996. SAS system for
mixed models. SAS Institute, Cary, NC. 656p.

Masters, R. B., D. M. Engle, and R. Robinson. 1993. Effects of timber harvest and
periodic fire on soil chemical properties in the Ouachita Mountains. South. J. Appl. For.
17: 139-145.

83

 

 

Metz, L. J ., C. G. Wells, and B. F. Swindel. 1966. Sampling soil and foliage in a pine
plantation. Soil Sci. Soc. Am. Proc. 30: 397-399.

Nissley, S. D. R. J. Zasoski, and R. E. Martin. 1980. Nutrient changes after prescribed
surface burning of Oregon ponderosa pine stands. pp. 214- 219 in Proc. 6th Conf. on Fire
and Forest Meteorology. April 22- 24, 1980, Seattle, WA.

Nowak, C. A., R. B. Downard, Jr., and E. H. White. 1991. Potassium trends in red pine
plantations at Pack Forest, New York. Soil Sci. Soc. Am. J. 55: 847-850.

Perala, D. A., and D. H. Alban. 1982. Rates of forest floor decomposition and nutrient
turnover in aspen, pine, and spruce stands on two different soils. USDA For. Serv. Res.
Pap. NC-227. 5p.

Roberts, B. A., and A. U. Mallik. 1994. Responses of Pinus resinosa in Newfoundland
to wildfire. J. Veg. Sci. 5: 187-196.

Rouse, C. 1988. Fire effects in Northeastern forests: Red pine. USDA For. Serv. Gen.
Tech. Rep. NC-129. 9p.

Ryan, M. G., and W. W. Covington. 1986. Effect of a prescribed burn in ponderosa pine
on inorganic nitrogen concentrations of mineral soil. USDA For. Serv. Res. Note RM-
464.5p.

Sharrow, S. H., and H. A. Wright. 1977. Effects of fire, ash, and litter on soil nitrate,
temperature, moisture and tobosagrass production in the Rolling Plains. J. Range.
Manage. 30: 266-270.

Smith, D. W. 1970. Concentrations of soil nutrients before and after fire. Can. J. Soil Sci.
50: 17-29.

Smith, D. W., and G. C. Bowes. 1974. Loss of some elements in fly-ash during old-field
burns in southern Ontario. Can. J. Soil Sci. 54: 215-224.

Soil Conservation Service. 1979. Soil survey of Kalamazoo County, Michigan. F. R.
Austin, ed. 102p.

Steele, S. J., S. T. Gower, J. G. Vogel, and J. M. Norman. 1997. Root mass, net primary
production and turnover in aspen, jack pine and black spruce forests in Saskatchewan and
Manitoba, Canada. Tree Phys. 17: 577-587.

Tappeiner, J. C., and A. A. Alm. 1975. Undergrowth vegetation effects on the nutrient

content of litterfall and soils in red pine and birch stands in northern Minnesota. Ecol. 56:
1193-1200.

84

 

 

Wells. C . C
Franklin. R

knmt'ledgc

 

 

Wells, C. G., R. E. Campbell, L. F. DeBano, C. E. Lewis, R. L Fredriksen, E. C.
Franklin, R. C. F roelich, and P. H. Dunn. 1979. Effects of fire on soil. A state-of-
knowledge review. USDA For. Serv. Gen. Tech. Rep. WO-7. 34p.

85

 

Chapter 3
Effects of fire on fine roots and stem growth

86

 

 

Introduction

Roots have generally been ignored in studies of tree growth and carbon allocation
until relatively recently. Several studies over the last 20 years have begun to illuminate
their importance as a sink in within-tree carbon allocation. Fine roots also play an
important role in ecosystem carbon dynamics. Fine root turnover is a major source of
carbon for the soil ecosystem. While the role of fine roots has become clearer, very little
is known about the effects of fire on fine roots. For example two recent textbooks on fire
ecology (DeBano et al. 1998, Johnson and Miyanishi 2001) each contain only one page
on fire’s effects on roots. This study attempted to elucidate the role of low-intensity
prescribed fire in the fine root dynamics and carbon allocation of a red pine stand in

southern Lower Michigan.

Hypotheses

Four hypotheses were posed at the beginning of the experiment:

1. A low-intensity prescribed fire in a red pine stand will kill all roots in the
forest floor (FF) and some in the top 2cm of mineral soil.

2. Fine roots will recover to pre-fire levels (root mass or length) within two
growing seasons.

3. The new growth or “flush” of roots will occur slightly below the 0-2 cm depth
(recovering from heat injury and a nutrient pulse will lead to no initiation of new roots at
the surface).

4. Until the fine root system gets re-established, stem diameter growth will be

decreased as more carbon will be allocated to recovering fine roots.

87

Methods

Site description

The study site is located at the WK. Kellogg Experimental Forest of Michigan
State University near Augusta, MI. The site was degraded farmland when planted in
1950 with 2-1 red pine seedlings, at a 3 x 3m spacing. Over the next two years, seedlings
that died were replaced. The current stand was originally part of a thinning study. Five
plots of 0.06ha each were established in 1994 for both the burn and control treatments;
plots were selected based on amount of understory (little to none) and blocked based on
previous thinning treatment and overstory density. As of 1990, basal area ranged from
32.2 — 55.2 mZ/ha (140 to 240 ftZ/ac), with D., of 23.6 — 28.4 cm (9.3 to 11.2 in). Site
index is approximately 19.8m (65 feet) at 50 years. Soils are a Kalamazoo loam and an
Oshtemo sandy loam (Typic Hapludalfs) (Soil Conservation Service 1979). In a separate
study on this site Gerlach (2001) determined the soil texture to be either a sand or loamy
sand (to a depth of 1.5m).
Experimental design

The experiment was designed as a Randomized Complete Block with 5 Control
plots and 5 Burn plots. Root samples analyzed as a split plot based on soil depth.
Repeated-Measures analysis was used to compare changes in the several root variables
over time. An alpha level of 0.10 was used because of the extremely high variability

found in studies of roots (c.f. Steele et al. 1997).

88

Minirhizotron

In the fall of 1994, two 60-cm-long minirhizotron (MR) tubes were placed around
each of three trees in each of the 10 plots (5 Burn and 5 Control), for a total of 6 tubes per
plot. Tubes were inserted into the soil at a 30° angle to the horizontal, to a vertical depth
of approximately 25cm. Frames (1.2 cm x 1.8 cm) were scribed along the length of each
tube, defining the area to be recorded. Images were recorded using a Bartz Technology
Company BTC-lOOX minirhizotron video camera system (Bartz Technology Company,
Santa Barbara, CA). Samples were recorded at approximately monthly intervals,
beginning in the spring of 1996 and continuing through 1997. Images were digitally
analyzed using the ROOTS program developed at Michigan State University. A diameter
of 1 mm was considered to be the cutoff between “fine” and “coarse” roots.

Root cores (described below) were separated into forest floor (FF), 0-2 cm, 2-6
cm, 6-10 cm and 10-25 cm depths. However, roots observed in MR images were
analyzed by only two depths, 0-10 cm and 10-25 cm. No finer separation of depths was
done in the MR images because of the paucity of roots viewed near the surface, a
common problem in MR studies (Vos and Groenwold 1987, Franco and Abrisqueta 1997,
Steele et al. 1997, and references therein). For calculations of root length density, it was
assumed that roots could be viewed up to 2 mm from the tube.

Root lifespan was estimated directly from the MR data. All roots born on a given
date were recorded in the ROOTS database until they were classified as either “missing”
or “dead”. The exact date of birth or death was estimated as halfway between the current

sample date (when the root was first seen or judged to be dead) and the previous sample

89

 

date (c.f. Kosola et al. 1995). Also, certain plots actually had no “New” roots in a given
cohort. For statistical analysis the roots in these plots were given a lifespan of zero days.

Direct estimates of root lifespan were based on a cohort of roots followed over
time, using the lifespan of the “median root” as the estimate of root lifespan. If a given
cohort had an even number of total roots, then two roots had to be considered “median”.
If their lifespans differed, then the true median was estimated as the average lifespan of
those two roots. If there were no new roots in a given depth, in a given plot, then a
lifespan value of “zero” was used in statistical analyses. For those cohorts that had roots
alive at the end of the experiment and the median had not yet been established, the
median was estimated as “greater than” a certain number of days. However, in these
cases the numerical value was used for statistical analyses without the “greater than”
designation.

Occasionally, a “New” root would actually be dead at its first viewing. In those
cases, the lifespan of that root was estimated as half the number of days in the sampling
interval. For example, if the sampling interval was 30 days, a New/Dead root’s lifespan
was estimated as 15 days.

Root turnover was also calculated from the MR data, using the changes in root
length (live or dead) between samples. In some instances, an individual root’s length
decreased between sample intervals, only to increase again at the next sampling.
Therefore, only increases in live root length were accepted in determining root length, i.e.
data were adjusted such that the length of a living root never decreased. This technique
compensated for roots that were less visible on a given date. However, any potential

effects of root herbivory, which can cause root length to decrease even though the root

9O

remains alive, were not accounted for. Therefore, total live root length and increases in
root length may have been overestimated. Some roundworms (assumed to be nematodes)
were occasionally observed in MR images but their abundance was not determined.
Instances where root herbivory was actually occurring — where some animal was in the

process of feeding on a root — were not observed.

Root cores

Root cores were taken concurrently with MR samples, beginning in the spring of
1997. While MR samples were taken six times during 1997, root cores were taken only
during four of those sample dates because the analysis of root cores is incredibly tirne-
consuming. This destructive sampling continued through 1998, but MR sampling
continued only through 1997 because of equipment problems. Cores were taken 4 times
during each growing season. Six cores of 4.76cm (l-7/8in) diameter were extracted in
each plot, to a depth of 25cm. Each core was separated into forest floor (FF), 0-2 cm, 2-6
cm, 6-10 cm, and 10-25 cm. At each depth, the six cores were composited within each
block to reduce variability. Half of the composite (at each depth) was used for analysis
of fine roots while the other half was reserved for soil nutrient analysis (Chapter 2).
Shortly after sampling, the fine roots were washed from each composite using a
hydropneumatic root elutriator (Smucker et al. 1982). Roots were then stored in a
solution of 20% methanol, at ~4 °C until further analysis.

In the lab, each sample was placed in a round tray and all intact (>lcm length)
fine roots were removed and set aside. The remaining material in the tray was mixed and

a 20° metal divider was placed in each of 3 randomly-chosen sections of the tray,

91

resulting in a 25% sample. All broken-off root tips lmm-lcm in length (considered
“dead”) were picked from each of these sections and set aside. Removing all fine roots in
this manner was very labor-intensive, taking 45 minutes to 16 hours per sample. Each
intact fine root was identified as either “live” or “dead” based on color, texture, and shape
of the root (V ogt and Persson 1991). Determining viability was also very labor intensive,
taking 1 to 8 hours per sample. Once roots were separated by viability class, samples
were dried at 70 °C to a constant weight. Preliminary analyses of the 1997 data indicated
no significant differences between treatments at the 10-25 cm depth. Based on this, and
because of the tirne-consuming nature of the analysis, the samples from 1998 at the 10-25

cm depth were not analyzed.

Forest floor‘weight

Several additional FF samples were taken during 1998 (the second post-burn
growing season) to determine the amount of litter that burned, and the amount of duff that
decomposed, following the fires. Three samples, 15.2 x 15.2cm, were taken in each plot
during each sample date. These samples were separated into litter (L) and duff (F + H)

layers. Samples were dried in a 70°C oven to a constant weight and then weighed.

Diameter growth

In the fall of 1998, after the completion of two post-fire growing seasons,
increment cores were taken from 10 trees in each plot. Two cores were taken from each
tree, one from the north side (within the row) and one from the west side (between rows).

Cores were air dried for approximately two weeks, then fixed to mounts. They were

92

thoroughly sanded and then digitized. Ring width for two postfire years and 10 pre-fire
years was measured using WinDendroTM software (Regent Instruments Inc., Quebec,
Canada). Initial analysis indicated that there were no differences in ring width, between
the north and west cores. Therefore, the final analysis used the average ring width for

each tree.

Calculations

Three different methods of calculating turnover from MR data were compared
(Table 3-1) — direct observation, Hendrick and Pregitzer’s (1992) method and Cheng et
al.’s method (1991) as modified by Steele et al. (1997). Standing crop was considered to

be the mass of live (not live + dead) fine roots at any given time.

Table 3-1. Equations used in comparison of methods of calculating fine root turnover.

 

 

 

 

 

 

 

 

 

Method Equation/Calculation
Direct Turnover = l .
Median Lifespan
Hendrick and Pregitzer TI (Turnover Index) = 2 Root (length) loss
(1992) Initial Standing Crop
Cheng et al. (1991) T1 = 2 Root (length) loss
(Initial + Final Standing Crop)/2

 

 

Root core data were used to calculate net primary production (N PP) of fine roots,

according to the equation (V ogt and Persson 1991):

NPPR = ABa-u + AM t2-tl + AD mi, (1)

93

where NPPR = net root production, ABa-n = statistically significant increments of live
roots between two sampling periods, AM :24] = significant increments in dead fine roots
between two sampling periods, and AD 1241 = root decomposition occurring between two
sampling periods. Because decomposition was not measured in this study, only the first
two components of the equation were used in the calculations. Vogt and Persson (1991)

give a more thorough discussion of the methods of calculating NPP.

Fire Conditions

Low-intensity backfires were used to burn the 5 “Burn” plots in early June 1997.
Flame lengths were low as was the rate-of-spread (Table 3-2). On occasion the wind
shifted, turning the backfire to a headfire. However, within approximately one minute

the wind shifted back, retuming to backfire conditions.

Table 3-2. Weather, fuel and fire characteristics, June 1997.

 

 

 

 

Weather Conditions Fire Characteristics

Sunny Fuel loadings 5.0-6.1 Mg/ha
Temperature 24-2 8°C Rate of spread 0.1 m/min

Wind speed 0.7-1.8m/s Flame lengths <0.3m backfire
Relative Humidity 17-25% O.6-O.9m headfire
Days since last rain 7-8 Amount 0.97cm (0.38in)

 

 

Minirhizotron tubes were protected from the flames by covering them with a
piece of fiberglass pipe insulation that had been wrapped in aluminum foil. These tube
protectors were designed to protect the tubes while minimizing protection of the roots in
the ViCinity of the tube. The design worked well; however several tubes were slightly

damaged by the heat and one tube out of 30 was destroyed.

94

Soil temperature during the fires was measured using a system of Type J
thermocouples connected to a Campbell CR-10 datalogger. The system performed well
in pre-fire tests but during the experimental burns some difficulties were encountered.
Data was obtained from one of the burn plots but one of the thermocouples

malfunctioned above 73°C (see below).

The results of minirhizotron analysis and root core analysis are difficult to
compare because samples were taken concurrently only during 1997. Minirhizotron
samples were taken during 1996 also, while root cores were taken during 1998. Also,
MR samples were taken six times per year while root cores were taken only four times

per year. Table 3-3 may help to keep things clear while reading the results and

discussion.

Table 3-3. General timetable of samples. Prescribed fire treatments were applied in June
1997.

l 1998

t core
ree ring

 

RESULTS

Of the nearly 2,000 “new” roots studied in the MR images in this study, 95%
were <O.8mm in diameter, and 97% were <1 .Omm (Figure 3-1). From observations of
roots under the microscope, roots larger than approximately 0.8mm in diameter were tan
in color, While smaller-diameter roots were either white (at the tips) or brown. The larger

roots also had substantially greater tensile strengths than those in smaller diameter

95

 

 

classes. Therefore, approximately 0.8mm was probably the ideal cutoff point between

“fine” and “coarse” roots. However, to compare the results of this study with those found

in the literature, 1.0mm was chosen to distinguish fine from coarse roots.

 

h
C

 

 

 

 

 

 

 

 

 

 

 

 

 

 

35.
g 30. _. __
1Clo-2cm -
E 25 3.2-6cm
‘5 |.6-10cm ’
20 .
‘5 ..10-25cm
o 1 --.
a

 

 

 

 

 

 

<0.3 0.3- 0.4- 0.5- 0.6- 0.7- 0.8- 0.9- 1.0- 1.1- >1.2
0.399 0.499 0.599 0.699 0.799 0.899 0.999 1.099 1.199

Root diameter (mm)

Figure 3-1. Diameter distribution of “new” roots observed in minirhozotron images over
the course of two growing seasons. n = 1964 roots.

Soil temperatures

Soil temperature during the fires was measured in one plot (Figures 3-2 and 3-3).
Temperatures rose above the presumed lethal threshold of 60°C (see Chapter 4) at one
point but (Figure 3-2) but not at the other (Figure 3-3). The two sample stations were less
than 1.5 m apart. The temperature was higher than 60°C only at the surface of the

mineral soil; temperatures at the 2 cm depth increased by less than 5°C.

96

80

 

 

 

 

 

 

 

 

 

Thermocouple errofl

70 .

60 -
8?. 5° 4 , -
'3 40 l 2 ‘3'“
8. _ 6 cm
E 30 . ‘ '
....

20 J

10 -

o -

0 4 8 12 16 20 24 28 32 36 40

Time (min)

Figure 3-2. Soil temperatures during a prescribed backfire in red pine. Surface
temperatures during burning reached the presumed lethal threshold (60°C) and remained

there for approximately 11 minutes at this sample station. The thermocouple located at
the 0 cm depth malfunctioned above 73°C.

 

 

 

25

20
a: 15.
g . —Ocrn
.3. 2cm‘
E10 _ 6cm
5.

5

ol_

 

 

 

0 4 8 12 16 20 24 28 32 36 40
Time (min)

Figure 3-3. Soil temperatures during a prescribed backfire in red pine. Surface

temperatures increased only 6°C as the fire passed. This sample station was less than
1.5m from that described in Figure 3-2.

97

 

Lifespan/Turnover from MR data

Direct estimates of root lifespan did not differ between treatments either as pre-
fire values alone or post-fire values alone (results not shown). Therefore, both pre- and
post-fire data were combined and analyzed (Table 3-4). There were differences among
the sample dates and a significant interaction between depth and time. Not counting the
Sept97 cohort (because of right-censored data), median lifespan ranged from 38 days (0-
10 cm, Aug97 cohort) to 272 days (0-10 cm, June96 cohort) (Figure 3-4). This leads to
direct estimates of turnover from 11.1 to 1.8 times per year, respectively. Root cohorts
born in May, July, August and September 1997 were significantly shorter-lived than all
other cohorts, but did not differ from each other (results not shown). In addition to
lifespan, variability also decreased from 1996 to 1997 (Figure 3-4).

Table 3-4. AN OVA results for root lifespan, from direct estimates from MR images. For
these data, the “depth” values are 0-10 cm and 10-25 cm.

 

 

 

 

 

 

 

 

 

 

 

 

Effect F3 Value I Prob > F
Treatment 0.69 0.4541
Depth 0.14 0.7046
Time 0.16 <0.0001
Treatment x Depth 18.92 0.6875

,. Treatment x Time 0.24 0.9918
Depth x Time 1.99 0.03 70

. Treatment x Depth x Time 1.08 0.3832

 

 

 

98

350

 

      

 

  

 

 

 

 

   
  

 

   

 

             

 

 

 

 

 

 

 

 

 

 

 

 

 

acoua— ___~_r , r,,e_-__ _D __D-_-
2. 250 ...—__... L. a s..- __r__ 3;; .11, ,r_-_l...#.
“’ '4
U . .
" +1
C I?-
1' 1.1.1 M H__.____ ~____~M
g 200 . 1 100406“
" i1 1.10-256m
'3 150 .3 — 5“?“ .H____,_,
111
r: 1,1
.2 .11.:
'2 100 ... D _d , 1, __
1.1
50 .1 __ _ _, fi—
0 . is :i 1 .
May96 June96 July96 Au996 Sept96 Oates Apr97 May97 July97 Au997 Sept97

Date of root appearance

Figure 3-4. Median root lifespan, as determined by direct observation of minirhizotron
images. Arrow indicates date of prescribed fire treatment. Error bars indicate the
standard error (n = 10). There were no significant differences between treatments either

before or after the fire.

As with lifespan estimates, neither root (length) production nor loss were different
between treatments, either before or after fires were set (results not shown). Therefore,
data from both treatments were combined for turnover calculations. There were
significant differences between depths and among times, as well as a significant depth x
time interaction for both production and loss (Table 3-5). Production and loss (m root
length/m3 soil volume) were nearly always greater in the 10-25 cm depth than in the 0-10
cm layer (Figures 3-5 and 3-6). Production was generally higher during 1996 than 1997
and was Very high in June 1996. However, there was a large decrease in production
during September 1996 which was mirrored by a slight increase in loss. There was a
large increase in loss in July97, over twice the amount measured in any other interval.

This was followed by very little production in August 1997.

99

Table 3-5. ANOVA results for root length production and loss calculated from MR data.
Both pre- and post-fire data were used in these analyses.

on Loss
Effect F V ue > ue

reatment 0. . 1 . .
9. 7 <0. . . 1

true .1 . 1 . . 1
Treatment x 1. . . 910
Treatment x lme 0.20 . . .4956

x lme . . . <0.0001
Treatment x 0. . . 0.

 

2000

 

1800 .e_.-._..-,_,__ __. __
16001._.______, ~ __

é

 

 

 

§
1

 

2.10—25cm

é

 

 

 

Production (mlm’)

 

 

 

 

 

 

 

 

 

 

 

 

 

go 010

 

Figure 3-5. Root production (m of root length/m3 soil volume) between depths, over
time. Arrow indicates time of fire. Because there were no treatment differences, both
treatments were combined for this analysis. Error bars indicate stande error (n = 10).

100

E

 

 

1

3500 .

3000 .
"s 2500-
E {EB-”1.03%" "
g 2000 1.10-25cm
8
a:

 

 

 

 

 

 

 

 

Figure 3-6. Root loss (m of root length/m3 soil volume) between depths, over time.
Arrow indicates time of fire. Both treatments were combined for this analysis. Error
bars indicate standard error (n = 10).

There are some parallels in production and loss values compared to the rainfall
from the previous 30 days (Figures 3-7 and 3-8). Although rainfall was relatively low
before the July 1996 sarnple, there may have been quite a bit of water retained in the soil
from the previous month. While quite a bit of rain fell before the October 1996 and
October 1997 samples, production values (Figure 3-7) were relatively not very high.
Root loss (Figure 3-8) appears to be inversely related to rainfall, although the July 1997
sample doesn’t follow this trend. It must be noted, however, that over 6 of the 8 cm of
precipitation in the month before the July 1997 sample came in one storm nearly 3 weeks

before the samples were taken.

10]

 

 

 

 

 

 

 

 

 

- 16 3
1:
- 14 g

- 12 g _ ._-_‘__,

10.; 1:0-100m '

2 E 10-250m -

18 “3 1 .

1:“ 1+Ram (CD1).

- 6 .2 ‘ "——
.-4 =3
11 I” .-
JE E "E l1? i "11k 2 §

 

eeooeo««««««
9 9’ '8' '9 8’ 9 '9
$43” >9 $9909 0° ‘79 s§s~§$v~°ge°q 069

Sample date

Figure 3-7. Root production (m of root length/m3 of soil volume) compared to the
rainfall of the previous 30 days.

 

 

 

 

4000 18

a

3500 -_ — 16 5‘

g. 3000 - ll _- 14 8

s :- __ 12 3
E 2500 i 1; 10 «g 15:10-10cm 1
W w 2 A 1 - 1
3 1,1 -18 9.5, |m10250m1
E '1 g I+Rain (cm) '
;-:= —1 6 .. W“

a .. 4 :1

4 .. 1' I: 1" '5 ’8

i E I: l l's ': w— 2 e

I! I ..F ‘. 9 hi I I _ 0 A

 

 

 

 

Sample date

Figure 3-8. Root loss (m of root length/m3 of soil volume) compared to the rainfall of the
previous 30 days. Although 8 cm of rain fell in the 30 days before the July 1997 sample,
over 6 cm of that came in one storm almost 3 weeks before the samples were taken.

102

 

 

 

 

 

R001 1038, and not pl'OdUCtion, was used in calculations of turnover in my stud)“

However, both determine the amount of carbon entering the soil system. Production is
‘he amount 0f new root length added. Plants can add root length by extending e’dsting

ragga/1070f initiating new roots. The lower production values found in September 1996

find August 1997 (Figure 3-5) are paralleled by initiation of very few roots (F iglll' e 3‘9).

However, as shOwn in Figure 3-4, the cohort initiated in September 1996 had a medium-

\Qéfiyg‘fifif’e Span While the August 1997 cohort had a short lifespan. New root initiation

was espeCiauy high during June 1996. Production outstripped loss during the early part

of this experiment but loss played the greater role starting in September 1996 (Figure 3-
10).

250

 

 

 

z 150 _—~—— -

=.-. W011
3 ~310-25cm
a

E

:3

c

‘3

.—

 

  
  

:-r-- .
- “Writer’s; . ..-
u Athl..4&.'.‘g.."‘_ -_ ' --

    

 

 

Figure 3-9. Total number of “New”
and tubes combined, Arrow indicat

roots in MR images by depth and month, all 131018
es time of experimental fires.

103

 

 

 

   

 

 

 

Figure 3-10. Production vs. loss, 0-10 cm. Similar results were found at the 10-25 Cm
depth- Arrow indicates time of experimental fires. Error bars indicate standard ermr (n =
10).

The methods of Hendrick and Pregitzer (1992) and Cheng et al. (1991) for
calculating T1 were compared to each other, and with my direct estimates of turnover.
The results of calculations for T1 can be sensitive to choice of beginning and ending
dates. Therefore, only those cohorts in which roots were observed for approximately a

full calendar year were used in T1 calculations.

Both methods of calculating TI resulted in lower estimates of turnover (times per
year) than did the direct estimate of root lifespan (Figures 3-1 1 and 3-12). However,
direct estimates of turnover were generally more Variable within one starting date, and
heme“ “fining dates, than estimates from either of the calculation methods, at both the

0-1 0 cm and 10-2 5 cm depths. The Cheng et al. (1 991) method gave more even results at

104

 

1
1.

 

the 0-10 cm depth across sample dates, while the Hendrick and Pregitzer method was
more consistent at 1 0-25 cm. The method of Hendrick and Pregitzer (1992) consistently
resmted in lower estimates of turnover, and therefore longer estimates of root lifespan,
than the Cheng et al. (1991) method. Roots consistently had lower TI estimates at the 10'
2‘12”] depth than at 0-10 cm using the both the Hendrick and Pregitzer (1992) methOd

and the Cheng et al. (1991) method. Turnover based on my direct estimates give mixed

results at the tyvo depths.

 

 

  
 

 

6
a
C
T. 4
O
5 r—* - ~-~ 7-1,
g [D Direct
1: - H&P
O
:L i Bfiheng 9181.
g ,
0
E
:3
‘-.'

 

 

 

 

 

 

 

Figure 3-11. Turnover (times per year) at the 0-10 cm depth, as estimated from direct

observations of root lifespans and as calculated from the methods of Hendrick and
Pregitzer (H&P) (1992) and Cheng et al. (199]).

105

 

 

 

 

1:515:53
1IH&P

 

 

 

 

 

 

 

Figure 3112. Turnover (times per year) at the 10-25 cm depth, as estimated from direct

observations of root lifespans and as calculated from Hendrick and Pregitzer’s ( 1992)
method and Cheng et al.’s ( 1 991) method.

Root mass and NPP

The masses of live fine roots and dead fine roots were analde separately to
determine possible treatment differences and to determine significant changes between
successive samplings. Burning had no effect on mass of live or dead fine roots (Table 3-
6). However, beginning in August 1997 and continuing through 1998, the mass of live

fine roots 1n the forest floor was consistently higher in the control plots than the in the

burn plots (Figure 3-13).

106

 
 

 

Table 3-6. ANQV A results for 1i ve and dead fine roots, core data.

  
   
   
 
  

    
 

We me me

 

Effect

 

reatment

  
 
      
    
   

"he
“cement x
X
X 1me
rea 011th X

   

 

 

 

.0
N
o
|

1- - O - -§Jfi$
1—0—Controlf

L.,..-

 

‘ I

R001 “1333 (M91113)
0

c.
)
O

 

 

 

 

 

0.05

 

 

 

 

 

 

 

0.00 l

f

May97 July97 Au997 Oct97 May98 Juneee fi—AUQQB 0°93
Month

Figure 3-13. Mass of live fine roots in the forest floor over time. Similar results were
found w1th dead fine roots. Arrow indicates time of experimental fires. Error bars

indicate standard error (n = 5).
There was a significant difference among depths for mass of live fine roots (Table
3'6) ROGt mass density (RMD) decreased with depth in the mineral soil (Table 3-7).
That is, the concentration of fine roots was highest at the surface of the mineral soil, and

decreased With depth, RMD was not calculated for the Forest Floor beoause the

107

 

 

 

 

 

thickness of the forest floor, and therefore the upward extent of rooting, was highly

Variable aniong plots.

There were significant differences over time and there was a significant <1er x
Time inter action for live fine roots (Table 3-6). For example, the mass of live fine roots
d‘m’d’fd Sh€111le from May to July 1997 at the 2-6 cm and 6-10 cm depths. (Figure 3'
14) 771iS was followed by an increase in live root mass over the next two sample dates at
these dePthS, although root mass never recovered to May 1997 levels. Live root mass

fixo decreased through summer and fall of 1998 at the 0-2 cm and 2-6 cm depths, but not

at the 6-10 cm depth.

Table 3'7. Mean (Standard error) May 1997
, prefire fine root mass M a and
denSlty (8/m3) across the different depths. n=10 ( g/h ) root mass

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1 Live Fine Roots Dead Fine Roots Total Fine Roots
Depth \Mass \Density Mass Density Mass Density
FF \0.069 0.410 0,478
(0.0151 (0.060) (0.072)
0-2cm E334 1.672 0.767 3.835‘ 17W 5.506
2-6 cm (0:312) (0.210) (0.071) (0.353) (0.104) (0.520)
E) 0 47 1.269 0.863 11$er 3.427
6_10 cm 0£113) (0.107) (0.071) (0.177) (0.106) (0.264)
(0 0: 1.040 0.686 1.716 HOT 2.756
0-10 cm 1:25 3) (0.159) [(0065) (0.161) (0,113) (0.282)
. 8 1.258 12.317 2.317 3.574 3.574
(0125) (”-125><\0-151) (0.151) I (0.270) (0.270)

 

 

 

 

 

 

 

108

 

 

 

 

 

 

 

 

 

 

 

 

l

M8y97 Juty97 Aug97 Oct97 May987
Sampledate

Junesa Au998 0ct98

Figure 3-14. Mass of live fine roots over time. Arrow indicates time of experimental
fires. Error bars indicate standard error (11 = 10).

The changes in mass of dead fine roots almost mirrored the changes in live fine
roots (Figure 3-1 5). Dead root mass increased sharply from May to July 1997 at all
depths and slowly decreased over the next two sample dates. Dead root mass remained at
these relatively low levels through June 1998 and then increased in August and October

1998. Minirhizotron results (Figure 3-10) also showed a large increase in root loss in
July 1997.

109

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

May97 July97 Au997 0ct97 May98

Jun398 Au998 Oct98
Sample data

Figure 3-15. Mass of dead fine roots over time. Arrow indicates time of experimenta]
fires. Error bars indicate standard error (n = 10).

The calculation of NPP is predicated on changes in root mass between successive
sample periods. Time was significant in the ANOVA (Table 3-6), as was the depth x
time interaction for live fine roots. I decided that all depths should be combined for NPP
calculations because the interaction was significant only for live, and not dead fine roots.
The data from both treatments were combined.

- Ideally, data would span one complete year for determining NPP. However, there
are data for only one winter interval in this study (Table 3-8). Therefore, to WhiCh
growing season Should this be added, 1997 or 1998? Including this interval for either
9'0ng $635011 doesn’t change the results, however, as there were no significant

differences for H Ve or dead fine root mass, from October 1997 to May I 998.

110

. . . to
Table 3-8. Significant changes 11111“? and dead fine roots during 1997 and 1998111231]ts
calculate NPP, All depths were Combined and both treatments were combined.

Were summed for each growing season to determine annual NPP.

  
  

Ann

  

e (M
F me

   
  
 
    
     
 
 

)

    
  

1ve 1ne
Roots
-0. 57
. 20

  
   
       
 
     

 
 
  

Roots
1 .

      
   

    

ӣ3

to
to
Aug97 to
0ct97 to Na

to
Iraw to A
to

     

    
  

A 7

 
  
   

   

  

 
  

  
 

A

Diameter Growth

Diameter growth of the red pines in this study was not affected by these fires
(Table 3-9). There were no differences between years (1997 and 1998), nor were there
any interactions. Pre-treatment diameter was related to post-treatment growth, Le. large

trees continued to produce large rings, through 1997 and 1998.

Table 3-9. ANOVA results of ring width data. Pre-treatment diamaer was sigmficantly
related to post-treatrnent ring-width growth,

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Effect F Value Prob > F
Treatment 0.53 0.4672
Diameter ‘ l 7.66 <0.0001

Wear ‘ 0.07 0.7855

[ Treatment x Diameter 0.76 0.3852

[ Treatment x E 0.07 0.7921
Diameter x W 0.12 0%
Treatment XEmeter x Year 0.16 0.6m?“—

 

 

 

lll

 

 

he

 

DISCUSSION

Root lifespan and turnover

Direct estimates of root lifespan ranged from 38 to 272 days, depending on which
cohort was followed. These values are within the range of those reported from the
literature (Table 1-3). Similar to Reid et al. (1993), there were no effects of season of
initiation on root lifespan. However, there were distinct differences in lifespan between
roots born before and during April 1997 and those born afterwards. There was a large
increase in root loss from May to July 1997 (Figure 3-6). Precipitation for July97 was
over 75% below the long-term average of 9.2cm. Growing-season precipitation was 20
and 24% below the long-term average for 1996 and 1997, respectively.

The “bimodal” growth of red pine roots that has been reported by others (Merritt
1968, Wilcox 1968) was not observed in this study, neither in initiation of new roots
(Figure 3-9) nor in root length extension (Figure 3-5). Soil water availability is probably
the main factor responsible for changes in a species’ inherent root growth strategy
(Katterer et al. 1995). Because soil water availability was not measured in this study I
used precipitation as a proxy. Production paralleled growing season rainfall, but only to
some extent (Figure 3-7). For example, there was very high production in June 1996
when there were 16 cm of rainfall in the previous 30 days. Production was similar in July
and August 1996. However, only 1.1 cm of rain fell before the July samples were taken
while 5.7 cm fell before the August samples. It is likely that the high June rainfall
affected root production in July. Although production was lower in July and August

1996, it wasn’t at the lowest point of the experiment. Both rainfall and production were

112

 

 

 

 

f0)?

mill

low for the September 1996 samples (1.8 cm of rain) and August 1997 samples (2.0 cm).
Although over 12 cm of rain fell before the October 1996 and October 1997 samples, root
growth was slightly low. It is probable that root grth was slowing at this point as
temperatures were beginning to fall.

Like production, root loss was also somewhat related to rainfall (Figure 3-8).
Root loss was very low in May 1996, June 1996 and October 1997; these months had 3 of
the 4 highest rainfall totals in this study. The July 1997 spike in loss is somewhat
problematic. However, as noted above, over 6 of the 8 cm of precipitation in the month
before the July 1997 sample came in one storm nearly 3 weeks before the samples were
taken. While taking root cores during that month I was forced to moisten the soil corer
simply to get it into the ground. Although drought may not completely stop root growth,
it can result in reduced root elongation and increased root dormancy (F eil et al. 1988).

As dry soils begin to receive more water (from rain or irrigation), root growth will
quickly recover (Feil et al. 1988). Although soil temperature affects root growth (e.g.,
Hendrick and Pregitzer 1993), root growth may be more closely related to soil water
content in some ecosystems (Lopez et al. 1998).

Direct estimates of root turnover in my study were usually greater than the results
of the Hendrick and Pregitzer (1992) method or the Cheng et al. (1991) method, both of
which are based on changes in root length (Figures 3-11 and 3—12). Results from the
literature (Table 1-3) indicate that direct estimates of root lifespan are usually shorter
than those based on changes in length, though the two types of estimates have not been
presented together for a given study. In my experiment, part of this discrepancy is

because of the adjustments made to the length data; i.e., because roots were never

ll3

 

 

 

 

1:15"

W/

allowed to decrease in length, root loss was underestimated and therefore standing crop
was overestimated. Therefore, TI values were probably too low. The assumption that
roots did not lose length was probably valid for some roots but certainly not for others.
Analyses which determine root herbivory may be able to correct for this. The turnover
calculated by the various methods in this study is comparable to that found in other

ecosystems (Table 3-10).

Table 3-10. Comparison of turnover values in a variety of ecosystems, as calculated by a
variety of methods.

 

 

 

 

 

 

 

 

Species/Ecosystem Turnover Notes Study
(times/year)
Basket willow 4.8 — 8.1 Based on numbers Rytter and Rytter
of roots, not lengths (1998)
Sugar maple 0.46 — 0.65 To 100 cm depth Hendrick and
Pregitzer (1992)
Boreal forests 1.4 — 3.3 Two different Steele et al. (1997)
calculation methods
Northern hardwood 0.8 - 1.2 Burke and Raynal
(1994)
Red pine 1.4 — 4.4 Direct method This study
0.9 — 1.3 Hendrick and
Pregitzer (1992)
0.9 — 2.3 Cheng et al. (1991)

 

 

 

 

Turnover index from the “length” methods was consistently higher, and therefore
estimates of root lifespan were consistently shorter, at the 0-10 cm depth than at 10-25
cm. Presumably, roots in the shallower depth are more sensitive to soil drying and
therefore trees shed the shallower roots much more quickly than deeper roots. The
deeper roots then become more important in water acquisition (Hendrick and Pregitzer

1996, Joslin and Wolfe 1998). For example, Espeleta and Eissenstat (1998) showed that

114

 

 

 

 

mature citrus tree
yet established 0

different betweer

 

al. 1991.1(05018
roots had shorter
while their "deep

It appear.

However season

 

and Pregitzer (l 1,
sum of annual r0
roots initiated in
Wilmer than n“

baSi‘d 5016]}. 0“

R001 mass and

There ‘
10013. Howey
see Chapter 4
Sample 31a“ 0
results in a “1

Base

he FF and 1

mature citrus trees shed surface fine roots more quickly than did seedlings that had not
yet established deep root systems. However, root mortality in seedlings may not be
different between well-watered and drought conditions (c.f., Marshall 1986, Hallgren et
al. 1991, Kosola and Eissenstat 1994). Hendrick and Pregitzer (1992), though, found that
roots had shorter lifespans deeper in the soil, but their “shallow” depth extended to 30cm,
while their “deep” roots were 30-100cm.

It appears that there is a seasonal difference in turnover (Figures 3-11 and 3-12).
However, seasonal changes in turnover actually cannot be determined using the Hendrick
and Pregitzer (1992) and Cheng et al. (1991) methods because their calculations use the
sum of annual root (length) loss. Based on the direct method (inverse of root lifespan),
roots initiated in May, August and September 1996 had consistently higher estimates of
turnover than those initiated in July and October 1996. However, these estimates are

based solely on lifespan.

Root mass and NPP

There were no differences between treatments in amount of live or dead fine
roots. However, soil surface temperatures did reach the presumed lethal levels (>60 °C,
see Chapter 4) at one location (Figure 3-2), but not at another (Figure 3-3). The two
sample stations were less than 1.5m apart. Smith and Sparling (1966) found similar
results in a “cool” fire in jack pine, with surface temperatures above 70°C for 11 minutes.
In this study, lethal temperatures did not extend to 2cm into the mineral soil.

Based on the thermocouple data, lethal temperatures were probably restricted to

the FF and perhaps a small fraction of the mineral soil. Root mass density (RMD)

llS

 

decreases with d
thinning (L0pe7
into the mineral

roots in the FF 11

fine roots in the

 

ranged from 26

may have maskt
been rep01ted b}
crop of red pine

”994) reported '

 

pine. But Santen
“993 OlliVe- an
Another

the hen sample
it Would “01b:
“N {001 ETD“

the MR data c

Sample date .
fme TOOI bio
for at least 5
Sludy with 5

(a)
6.1 C fgr 0V

decreases with depth (Strong and LaRoi 1985), although root density can be affected by
thinning (Lopez et al. 1998, Santantonio and Santantonio 1987). Restricting heat input
into the mineral soil is critical for limiting damage to the fine root system. Some fine
roots in the FF must certainly have been killed. However, the masses of live and dead
fine roots in the FF (Figure 3-4) were highly variable and coefficients of variation (CVs)
ranged from 26 to 68%, (pre-fire data, both live and dead fine roots). This variability
may have masked any real changes caused by burning. Variability of this magnitude has
been reported by others. For example, F ogel (1990) reported the CVs for the standing
crop of red pine fine roots (<2mm) to range from 27 to 34%, while Cropper and Gholz
(1994) reported within-plot CVs often greater than 100% for fine roots (<1mm) of slash
pine. But Santantonio and Hermann (1985) reported CVs of only 11-16% for standing
crops of live- and dead-fine roots (<1mm) of Douglas-fir in Oregon.

Another possibility is that roots were killed by the fire, but replaced by the time
the next samples were taken, approximately 4 weeks later. If this phenomenon occurred
it would not be detected by root coring but a sudden pulse of dead roots and a flush of
new root growth in the burn plots would be shown by MR analysis. However, analysis of
the MR data confirm there was no such difference between treatments in either number of
new roots (p _<_ 0.905), production (p 5 0.467) or root loss (p 5 0.972) during the first
sample date after the fires. Swezy and Agee (1991) did find significant differences in
fine root biomass 1 month after burning in ponderosa pine; these differences continued
for at least S-months after their fires. However, it is difficult to directly compare this
study with Swezy and Agee’s (1991) as the temperature at the soil surface surpassed

60°C for over 5 hours in their experiment.

ll6

 

 

Factors
[ponderosa pine
larger fuels (lar:

result in a high 1
rules out this far
(potential fuel) L

(Figure 3-16). al

fuel consumed;

maximum temp

50000
45000
40000 '
35000
30000
25000
20000
15000
1000C
500t

 

FF weight (ks/m"

Factors that caused root death in other studies — fuel arrangement and smoldering
(ponderosa pine) — were inconsequential in my study. Hungerford et al. (1991) state that
larger fuels (large branches and logs) are what generate the large quantities of heat that
result in a high temperature pulse into the mineral soil. The lack of large fuels on this site
rules out this factor as a potential cause of root mortality. The forest floor weight
(potential fuel) at this site was at the median of values reported for red pine stands
(Figure 3-16), although it was below average. Fire intensity is correlated to amount of
fuel consumed; therefore, less fuel results in lower intensity and therefore lower

maximum temperatures (Johnson 1992).

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

50000
45000 .e...-_fi __,,___ . ,, .1___ _ _.._ .. .__—__—_- l—
40000 ._W _.__. _..____ .-..-.___- as... , 2,7, We __-.- _ .--.___-_-______. __
E 35000 5...-..--.___ __ - - -- 2, _.
E 300001__._ _ _. _l
:5 250004 'mra.e= 2 WI 0 - lfl
; 20000 is. _.
{t 15000 _ _
10000 .. u. re“ a- __
0
9) 0) 0) 6%
(9% (9% C e‘”
. \.
a” a” a” 3“ a
at“ of a“ 5°" at
e e 9 Q0“ 0 8*
«9

Figure 3-16. Mass of the forest floor (FF) in various red pine stands.

A backfire was used in this study, and backfires are usually less-intense than

headfires (Van Wagner 1983). It’s difficult to tell if a headfire would have given

117

 

different results
total amount of
(though usuall}

the maximum St
headfires. Bach
more duff wouh
concluded that i
. l
remams unbum,

lUIure research.

\l‘hile th

 

that the trend rel
if in control pl<
did not burn sul
(May) very litt‘
winter. During
shovm graphic
0Verrode this
If it is

Would the 0\'
SHOW) loss?

We l0w co

different results. While headfires are more intense, they are very fast-moving, and the
total amount of heat released over time, at any one point in space, may be higher or lower
(though usually higher) than that released from a backfire. Heyward (1938) found that
the maximum soil temperature at 0.3 to 0.6cm was usually higher in backfires than
headfires. Backfires, though, do burn deeper than headfires (Hough 1978). Therefore
more duff would remain on the soil surface after a headfire and Van Wagner (1971)
concluded that soil temperature increases very little when greater than 1.27cm of duff
remains unburned. Comparing backfires and head fires would be a good avenue for
future research.

While the mass of live fine roots was not different between treatments, I believe
that the trend represented in Figure 3-13 is real; i.e., there were more live fine roots in the
FF in control plots than in burn plots in 1998. Litter was the main fuel in the fire; duff
did not burn substantially. However, I noticed that during the first sampling in 1998
(May) very little of the duff remained. Much had been decomposed over the previous
winter. During the rest of 1998, there was very little forest floor left in the burn plots. As
shown graphically (Figure 3-13) though, variability was rather high and probably
overrode this trend.

If it is assumed that there were fewer roots in the FF of burned vs. unburned plots,
would the overall root system be harmed substantially and would the trees show any
growth loss? I believe not. The mass of fine roots in the forest floor (Table 3-7) was
quite low compared to other values found in the literature; consequently, the proportion
of fine roots in the mineral soil is higher (Table 3-11). The site used in this study

comprised loamy sand soils, located on abandoned farmland in an area that receives

ll8

 

adequate precip
productive than
proportion of f1
may have suffer
lllay 1997. pre

ofthe total to 2:

I
ldblf 3-ll. C0
fine roots are <

- Root mass l R(

 

‘ 1 1
111FF _‘ m1

 

Becat

06 all} he at“

Rm‘ely Sn

“age of tho

This Study r
”1130‘ hfine‘

“

r

adequate precipitation (944mm/year). Therefore this site may be more fertile and
productive than others reported in the literature. Had the site been less fertile, a larger
proportion of fine roots may have been located in the forest floor, and the root systems
may have suffered some noticeable damage from burning. Fine roots in the forest floor
(May 1997, prefire) represented only 1 1.8% of the total root mass, to 10cm depth (7.7%
of the total to 25cm depth, data not shown).

Table 3-11. Comparison of root mass (Mg/ha) in the forest floor and the mineral soil.
Fine roots are <1mm diameter, unless otherwise noted.

 

 

 

 

 

 

 

 

Root mass Root mass in Depth Species Notes Study

in FF mineral soil (cm)

0.478 3.574 10 red pine This study

0.939 0.399 10 white spruce — Kimmins and

subalpine fir Hawkes (1978)

1.98 1.97 10 slash pine Gholz et al.
(1986)

1.2 — 5.4 3.6 -— 5.6 40 lodgepole pine Xeric sites Comeau and

1.6 — 4.4 2.8 - 4.8 40 Mesic sites Kimmins (1989)

0.82 10 red pine Braekke and

Kozlowski ( 1 977)

2.7 2.8 15 red pine Roots <3mm McClaugherty et
al. (1982)

 

 

 

 

 

 

Because there were no treatment differences in mass of fine roots, there could not
be any treatment differences in NPP. The difference between years (Table 3-8) is
relatively small, less than 10%. The values of NPP found in this study are within the
range of those reported in other studies (Table 1-1), although they are near the low end.
This study only measured roots to 10cm, whereas other studies went to 15cm or deeper.
Also, “fine” roots are only considered to be <1mm in diameter in this study, whereas

“fine” roots were <3mm or <5mm in other studies. The growing seasons (April to

H9

 

 

 

October). 1997 |
rtith the wetter
Hermann ( 1985
than moister sin
liar to July 19‘
root loss over If

Many st

treatments. or b

 

 

Where stands ar
(1993) found th
control plots. 11

Controls.
Estimate
months then So
The October 11
Production am
Grier 1981. H
Vader}. 0f ECO
MR reSuits in
death. to, the
Anon
into Viability-

October), 1997 and 1998, were 24% and 26% below normal (63.6 cm) for precipitation,
with the wetter year (1997) having higher production. The results of Santantonio and
Hermann (1985) are somewhat applicable. They found that drier sites had lower NPP
than moister sites, within one season. The large change in live- and dead-fine roots from
May to July 1997 in my study is backed up by MR results that show a large increase in
root loss over this time period (Figure 3-6).

Many studies compare root production and loss between control and drought
treatments, or between dry and wet years. How do these results compare with those
where stands are intensively managed and water is added via irrigation? Gower et al.
(1992) found that fine root NPP was actually lower in an irrigated plots compared to
control plots. In this situation, water was not limiting yet root growth was still lower than
controls.

Estimates of NPP should cover an entire year. Measuring NPP for only six
months then scaling results up to a full year by doubling the estimate is not appropriate.
The October 1997 to May 1998 interval showed no net growth (Table 3-5). Was the
production and loss during that winter typical of others? Several researchers (Keyes and
Grier 1981, Hendrick and Pregitzer 1992, Steele et al. 1997), using the MR technique in a
variety of ecosystems, found that growth and mortality are minimal in the winter. Also,
MR results indicate no difference between production and loss (p _<_ 0.593) at the 0-10 cm
depth, for the October 1996 to April 1997 interval.

Another source of variability in NPP comes from the classification of fine roots
into viability classes. Based on morphology, some roots were distinctly alive and some

were distinctly dead, but others were difficult to classify either way; they held

120

 

 

characteristics of both live and dead fine roots. It is nearly impossible to visually identify
roots that are senescing, which adds ambiguity into the classification of roots into
viability class (V ogt and Bloomfield 1991). Root classification is important because it is
the significant changes in live or dead root mass between successive samples that are
used in NPP calculations. If root viability is grossly mis-classified then significant
differences between sample dates would be incorrect.

Changes in water availability may also cause changes in fine root dynamics and
growth. These growth changes may be viewed along a continuum of soil water
availability. Results may be explained in terms of overall grth and carbon
partitioning. When water is not limiting, for example where trees are irrigated, fine root
production may actually be lower (e.g. Axelsson and Axelsson 1986, Gower et al. 1992),
although this effect may vary with depth and genotype (Dickmann et al. 1996). When
trees are irrigated they will tend to allocate more carbon towards aboveground growth
where competition for sunlight would be greater. Total (aboveground + belowground)
growth would be higher in this situation. However, water availability is dynamic and
overall growth and carbon allocation will change as limiting factors change. As soils dry
and competition for water becomes more intense, trees will allocate more carbon towards
roots, slightly increasing the rootzshoot ratio (Cermak et al. 1993, Joslin and Wolfe
1998), although this varies by genotype and depth (Rodrigues et al. 1995). This
hypothesis is supported by many seedling studies, but may (Axelsson and Axelsson
1986) or may not hold for stands of forest trees (Joslin et al. 2000). Under drought
conditions water availability may be so low that root production is minimized (Liu and

Dickmann 1992) and root mortality increased such that overall tree growth is decreased.

12]

 

Diameter growth

Although diameter growth did not increase following fire in my study, this is not
surprising. As noted by other researchers (see Literature Review), red pine growth is
usually not affected by low-intensity prescribed burning, although it appears that more
intense wildfires (Roberts and Mallik 1994) do result in a loss of growth. Even repeated
burning over many years doesn’t affect red pine growth (Alban 1977).

One of the main mechanisms resulting in growth loss following fire is crown
scorch (Cain 1985, 1996, Lilieholm and Hu 1987, Mann and Whitaker 1955). Visual
surveys of the burned plots showed that these fires resulted in minimal crown scorch. In
fact, only one overtopped tree appeared to have any damage at all. Increased growth
following burning has been attributed to reduced competition in the overstory (Waldrop
and Lloyd 1988, Lloyd et al. 1995), but there was no mortality due to fires in my study.
Another mechanism of aboveground growth loss may be through root damage. However,
no increased root loss was found in this study (Tables 3-5 and 3-6).

Growth increases after fire have been reported, especially in herbaceous plants
(see Chapter 1). Many researchers suggest that one mechanism behind increased growth
is the sudden mineralization of nutrients following burning. This mechanism is intuitive
and has much supporting evidence in prairie/grassland systems or with greenhouse
studies that used soil from burned land. However, no field studies with trees have
actually found correlations between increased nutrient concentrations and increased
overstory growth following fire. Because there were no significant changes in available

cations following the fires in my study (see Chapter 2), I cannot offer support for, nor

l22

 

 

 

 

61'

C0

evidence against, this idea. Sharrow and Wright (1977), burning in tobosagrass
communities in the rolling plains of Texas, concluded that there were no beneficial
effects of burning during dry years because soil moisture was the limiting growth factor.

Sutherland et al. (1991) found that prescribed burning significantly reduced the
radial growth of ponderosa pine for two years before normal growth resumed. In general,
they found that trees that grew well before the fire grew well after the fire. Results of my
study are similar; only pre-fire diameter was significant in determining post-fire ring
width.

This study only tracked growth for two growing seasons following the fire. It is
possible that there is a delayed reaction to prescribed burning that was not measured.
However, negative effects of fire on grth usually appear within the first growing
season; growth generally recovers within 2-3 years (Chambers et al. 1986). Delayed
positive effects on growth are also a possibility. However, in 8 years of post-fire study,

Sutherland et al. (1991) found no positive growth effects.

Summary

The two methods used to estimate root dynamics in this study, sequential coring
and MR, each have advantages and disadvantages. Measurements taken through coring
are extremely difficult to analyze and interpret (Makela and Vanninen 2000), and
different methods of data analysis have given different estimates of growth and root loss
(N adelhoffer and Raich 1992, Schoettle and Fahey 1994). Any temporal overlap in
production and root loss will cause underestimation of their rates (Kurz and Kimmins

1987). MR gets around these difficulties with direct, non-destructive observations of fine

123

 

roots. However, as shown in many studies (e.g., Vos and Groenwold 1987, Franco and
Abrisqueta 1997, Steele et a1. 1997, and references therein), the distribution of roots in
bulk soil is not accurately reflected in MR images.

A clearer definition of what “root turnover” is, or is not, is definitely needed.
Root turnover is important ecologically because of the role of fine roots in the carbon
cycle. Fine roots are a large source of carbon for the soil ecosystem (Vogt et al. 1991,
Steele et al. 1997). Physiologically, the amount of carbon directed towards root growth
must be balanced by an appropriate amount of aboveground growth. The method in
which turnover is considered to be the inverse of root lifespan (or vice-versa) is probably
not as appropriate as methods based on length changes. Considering root lifespan and
turnover to be inverses is based on the assumption that lifespans are normally distributed
(Hendrick and Pregitzer 1992). This would be valid if all roots were the same size.
However, consider the case where two roots of different lengths have the same estimated
lifespan. In such a situation, both roots contribute the same to estimates of turnover, even
though they put different amount of carbon into the soil. This method also ignores other
factors affecting root longevity, including root order or mycorrhizal status. For example,
Wilcox (1968) found greater mortality in second-order laterals than in primary roots of
red pine. Infection by mycorrhizal fungi may also reduce root lifespan (e.g. Hooker et a1.
1995). Other factors affecting root longevity include by stem density (Santantonio and
Santantonio 1987) or soil depth (Lopez et al. 1998).

The method of Cheng et al. (1991) consistently gave higher values of turnover
than the Hendrick and Pregitzer (1992) method. Both methods calculate turnover as root

(length) loss divided by some “standing crop” (Table 3-1). Loss included roots that were

124

 

 

 

S

part of the initial standing crop as well as those that were produced later during the course
of the year. However, the Hendrick and Pregitzer (1992) method used only the initial
standing crop in its estimate of turnover while the Cheng et al. (1991) method used the
average of initial and final standing crops. Therefore, the Cheng et a1. (1991) method
accounted for some of the production that occurred concurrent with loss throughout the
year. However the standing crop of live roots is dynamic. Using the average of initial
and final standing crop only accounts for some of this variability.

Root turnover is the simultaneous addition to, and loss from, a given population
of roots (Sutton and Tinus 1983). That is, turnover is the concurrent birth and death of
roots. Both the Hendrick and Pregitzer (1992) method and the Cheng et al. (1991)
method use only loss in determining root turnover. Because both production and loss
determine the size of the standing crop and rate of carbon cycling through the roots, a
method which uses both of these measures would be more appropriate.

Therefore I propose that turnover be calculated as follows:

Turnover = ((2 Production + 2 Loss)/2) (2)
Average annual standing crop
None of the four hypotheses I proposed at the beginning of the experiment were
supported. Hypothesis 1 was that roots would be killed in the FF and in the top 2cm of
the mineral soil. Some roots were probably killed in the FF, based on temperature data
(Figures 3-2 and 3-3) but statistically there were no differences. Root mass was highly

variable, and this variability may have over-ridden any real changes caused by burning

125

 

 

 

(c.f. Cropper and Gholz 1994). Fire may reduce FF root mass by simply reducing the
total amount of F F available as a rooting medium.

Besides potentially being killed directly by the heat of the fire, roots might have
been lost because of other mechanisms. Had there been a large amount of crown scorch,
trees would have lost photosynthesizing tissue. This lost productivity would have
resulted in increased death of fine roots, with less replacement, as the source of
carbohydrates is reduced. Lower needle mass would also mean reduced transpiration. A
decreased water demand would mean that the trees have no need for an extensive root
system, so increased death of fine roots, with little replacement until crowns recover
would occur. However, crown scorch was minimal in this study.

Because there was no loss of fine roots due to fire, the ideas that they would
“recover” within two growing seasons (Hypothesis 2) with a flush of growth at a slightly
deeper depth (Hypothesis 3) were moot. Also, the expected nutrient “pulse” into the
mineral soil that might have resulted in lower root mass did not occur (see Chapter 2).
Stem diameter growth (Hypothesis 4) was not affected by these fires.

Neary et a1. (1999) state that changes or removal of aboveground structure can
directly affect belowground systems by: (1) altering nutrient inputs that in turn, affect soil
and litter macro- and microflora and fauna; (2) increasing surface soil temperatures as a
result of increased solar heating; and (3) changing evapotranspiration rates due to losses
in vegetation that, in turn, alter soil moisture availability, etc.

These factors did not occur during this study. Nutrient inputs to this site, directly
from fire and indirectly from increased litter decomposition, were negligible. There was

no crown scorch, so there was not an extra pulse of litter input on burned plots. Post-fire

 

 

 

soil temperatures were not measured in this study. Although increased soil temperatures
have often been measured after burning, Sharrow and Wright (1977) concluded that this
is mainly due to loss of shade from the overstory plants, and not to the blackened soil
surface. Because there was no loss of overstory trees following burning the red pine
stand in my study, surface soil temperatures and evapotranspiration rates were probably
negligibly affected. There was no loss of overstory trees because of fire, and plots were
chosen for this study based on their negligible amount of understory. Therefore, there
were no changes in evapotranspiration rates, and presumably no changes in soil moisture
availability.

Future research on fine roots of red pine — and the effects of fire - should be
concentrated in stands that are in the natural range of red pine, on soils that typically
support this species. The stand that I studied was out of red pine’s natural range and on
potentially productive, though somewhat dry, soils. Also, there were relatively few roots
in the FF of this stand. Because these are the roots most-susceptible to damage by fire,
stands which have a higher proportion of roots in the FF (e.g., older stands — Gholz et al.
1986, Nambiar 1990) and which have larger fuel loads, should be studied. This would
help to give a more complete picture of fire’s effects in red pine ecosystems.

Another aspect of fire that was not addressed in this study was the cumulative
effect of repeated fires. Repeated burning has the potential to reduce site productivity
through nutrient loss (Hunt and Simpson 1985, Wright and Hart 1997), although Alban
(1977) found that repeated burning in red pine had little or no effect on soil nutrient

status. The main effect of repeated burning may be lowered water-holding capacity

127

 

 

 

(Boyer and Miller 1994). And as shown in this study, root growth and loss are sensitive

to soil moisture availability.

Literature Cited

Alban, D. H. 1977. Influence on soil properties of prescribed burning under mature red
pine. USDA For. Serv. Res. Pap. NC-139. 8p.

Alban, D. H. 1982. Effects of nutrient accumulation by aspen, spruce, and pine on soil
properties. Soil Sci. Soc. Am. J. 46: 853-861.

Alban, D. H., D. A. Perala, and B. E. Schlaegel. 1978. Biomass and nutrient distribution
in aspen, pine, and spruce stands on the same soil type in Minnesota. Can. J. For. Res. 8:
290-299.

Arrnson, K. A., and J. R. M. Williams. 1960. The root development of red pine (Pinus
resinosa Ait.) seedlings in relation to various soil conditions. For. Chron. 36: 14-17.

Axelsson, B., and B. Axelsson. 1986. Changes in carbon allocation patterns in spruce
and pine trees following irrigation and fertilization. Tree Phys. 2: 189-204.

Bockheim, J. G., J. E. Leide, and L. E. Frelich. 1989. Red pine growth and chemical
composition of foliage and forest floors across a precipitation-chemistry gradient in
Wisconsin. Can. J. For. Res. 19: 1543-1549.

Bockheim, J. G., J. E. Leide, and D. S. Tavella. 1986. Distribution and cycling of

macronutrients in a Pinus resinosa plantation fertilized with nitrogen and potassium. Can.
J. For. Res. 16: 778-785.

Boyer, W. D., and J. H. Miller. 1994. Effect of burning and brush treatments on nutrient

and soil physical properties in young longleaf pine stands. For. Ecol. and Manage. 70:
311-318.

Braekke, F. H., and T. T. Kozlowski. 1977. Distribution and growth of roots in Pinus

resinosa and Betula papyrifera stands. Med. fra Norsk Instittutt for Skogforskning (Rep.
Norwegian For. Res. Inst.) 33(10): 437-451.

Cain, M. D. 1985. Prescribed winter burns can reduce the growth of nine-year-old
loblolly pines. USDA For. Serv. Res. Note SO-3 l2. 4p.

Cain, M. D. 1996. Growth expectations from alternative thinning regimes and prescribed

burning in naturally regenerated loblolly-shortleaf pine stands through age 20. For. Ecol.
Manage. 81 : 227—241.

128

 

 

 

Cermak, J ., R. Matyssek, and J. Kucera. 1993. Rapid response of large, drought-stressed
beech trees to irrigation. Tree Phys. 12: 281-290.

Chambers, J. L., P. M. Dougherty, and T. C. Hennessey. 1986. Fire: Its effects on ngth
and physiological processes in conifer forests. pp. 171-189 in Stress physiology and
forest productivity. Henessey, T. C., P. M. Dougherty, S. V. Kossuth, and J. D. Johnson

eds.

Cheng, W., D. C. Coleman, and J. E. Box, Jr. 1991. Measuring root turnover using the
minirhizotron technique. Ag. Ecosystems Env. 34: 261-267.

Comeau, P. G., and J. P. Kirmnins. 1989. Above- and below-ground biomass and
production of lodgepole pine on sites with differing soil moisture regimes. Can. J. For.
Res. 19:447-454.

Cropper, W. P., Jr., and H. L. Gholz. 1994. Evaluating potential response mechanisms of
a forest stand to fertilization and night temperature. a case study using Pinus elIiottii.
Ecol. Bull. 43: 154-160.

DeBano, L. F., D. G. Neary, and P. F. Ffolliott. 1998. Fire’s effects on ecosystems. John
Wiley & Sons. New York, NY. 333p.

Dickmann, D. I. 1993. Management of red pine for multiple benefits using prescribed
fire. North. J. Appl. For. 10: 53-62.

Dickmann, D. 1., P. V. Nguyen, and K. S. Pregitzer. 1996. Effects ofirrigation and
coppicing on above-ground growth, physiology, and fine-root dynam1cs of two field-
grown hybrid poplar clones. For. Ecol. Manage. 80: 163-174.

Espeleta, J. F ., and D. M. Eissenstat. 1998. Responses of citrus fine roots to localized
soil drying: a comparison of seedlings with adult fruiting trees. Tree Phys. 18: 113-119.

Feil, W. ,I. Kottke, and F. Oberwinkler. 1988. The effect of drought on mycorrhizal
production and very fine root system development of Norway spruce under natural and
experimental conditions. Plant and Soil 108: 221-231.

Fogel, R. 1990. Root turnover and production in forest trees. HortScience 25: 270-273.

Franco, J. A. ,and J M. Abrisqueta. 1 997. A comparison between minirhizotron and soil
coring methods of estimating root distribution 1n young almond trees under trickle
irrigation. l. Hort. Sci. 72: 797- 805.

Fyles, J. W. ,and B. Cote. 1994. Forest floor and soil nutrient status under Norway
Spruce and red pine in a plantation 1n southern Quebec. Can. J. Soil Sci. 74: 387- 392.

129

h

 

Gerlach, J. P. 2001. A comparison of productivity and related traits for European larclt
(Larix decidua Miller) and red pine (Pinus resinosa Ait.) across a site quality gradient in
the Great Lakes region. Unpublished Masters thesis. Michigan State Universrty, East

Lansing, MI.

Gholz, H. L., L. C. Hendry, and W. P. Cropper, Jr. 1986. Organicmatter dynamics of
fine roots in plantations of slash pine (Pinus elliottii) in north Flonda. Can. J. For. Res.

16: 529-538.

Gower, S. T., K. A. Vogt, and C. C. Grier. 1992. Carbon dynamics of Rocky Mountain
Douglas-fir: Influence of water and nutrient availability. Ecol. Monog. 62: 43-65.

Hallgren, S. W., C. G. Tauer and J. E. Lock. 1991. Fine root carbohydrate dynamics of
loblolly pine seedlings grown under contrasting levels of soil morsture. For. Scr. 37: 766-
780.

Hendrick, R. L., and K. S. Pregitzer. 1992. The demography of fine roots in a northern
hardwood forest. Ecol. 73: 1094-1104.

Hendrick, R. L., and K. S. Pregitzer. 1993. Patterns of fine root mortality in two sugar
maple forests. Nature 361: 59-61.

Hendrick, R. L., and K. S. Pregitzer. 1996. Temporal and depth-related patterns of fine
root dynamics in northern hardwood forests. J. Ecol. 84: 167-176.

Heyward, F. 1938. Soil temperatures during forest fires in the longleaf pine region. J.
For. 36: 478-491.

Hooker, J. B., K. E. Black, R. L. Perry, and D. Atkinson. 1995. Arbuscular mycorrhizal
fungi induced alteration to root longevity of poplar. Plant and 8011 172: 327-329.

Hough, W. A. 1978. Estimating available fuel weight consumed by prescribed fires in
the South. USDA For. Serv. Res. Pap. SE-187. 12p.

Hungerford, R. D., M. G. Harrington, W. H. F randsen, K. C. Ryan, and G. J. Niehoff.

1991. Influence of fire on factors that affect site productivity. pp. 32-50 in Proceedings —
Management and productivity of Westem-Montane forest soils. USDA For. Serv. Gen.

Tech. Rep. INT-280,

Hunt, 3- M., and J. A. Simpson. 1985. Effects of low intensity prescribed fire on the
gr owth and nutrition of a slash pine plantation. Aust. For. Res. 15: 67-77.

Johnson, E. A. 1992, Fire and vegetation dynamics: studies from the North American
boreal forest. Cambridge University Press, Cambridge, England. 129p.

130

 

I

Johnson, E. A., and K. Miyaishi. 2001. Forest fires. Behavior and ecological effects.
Academic Press. San Diego, CA. 594p.

Joslin, J. D., and M. H. Wolfe. 1998. Impacts of water input manipulations on fine root
production and mortality in a mature hardwood forest. Plant and Soil 204: 165-174.

Joslin, J. D., M. H. Wolfe, and J. P. Hanson. 2000. Effects of altered water regimes on
forest root systems. New Phytol. 147: 117-129.

Katterer, T., A. Fabiao, M. Madeira, C. Ribeiro, and E. Steen.1995. Fine-root dynamics,
.5011 moisture and soil carbon content in a Eucalyptus globulus plantation under different
1mgations and fertilisation regimes. For. Ecol. Manage. 74: 1-12.

 

KCYCS, M. R., and C. C. Grier. 1981. Above- and below-ground net production in 40-

gar-old Douglas-fir stands on low and high productivity sites. Can. J. For. Res. 11: 599-

KiI‘l'mlins, J. P., and B. C. Hawkes. 1978. Distribution and chemistry of fine roots in a
fhlte Spruce-subalpine fir stand in British Columbia: implications for management. Can.
'For. Res. 8: 265-279.

Kosola, K. R., D. M. Eissenstat, and J. H. Graham. 1995. Root demography of mature
Citrus trees: the influence of Phytophthora nicotianae. Plant and Soil 171: 283-288.

fOSOIfi, K. R., and D. M. Eissenstat. 1994. The fate of surface roots of citrus seedlings in
'3’ 8011. J. Exp. Bot. 45: 1639-1645.

Kurz, W. A - , and J. P. Kimmins. 1987. Analysis of some sources of error in methods
used to determine fine root production in forest ecosystems: a simulation approach. Can.
J. For. Res. 1 ‘7: 909-912.

Lilieholm, R- J ., and S. C. Hu. 1987. Effect of crown scorch on mortality and diameter
ngth of 19—year-old loblolly pine. South. J. Appl. For. 11: 209-211.

Liu, Z., and D. I. Dickmann. 1992. Responses of two hybrid Populus clones to flooding,

grzo7ught, and nitrogen availability. 1. Morphology and growth. Can. J. Bot. 70: 2265-
0.

Lloyd, R T- s T. A. Waldrop, and D. L. White. 1995. Fire and fertilizer as alternatives to

gand gurgligng in a natural stand of precommercial-sized loblolly pine. South. J. Appl.
or. 1 : - .

Lopez, B» S . Sabaté, and C. Gracia. 1998. Fine roots dynamics in a Mediterranean
forest: effeCtS of drought and stem density. Tree Phys. 18: 601-606.

131

 

Makela, A., and P. Vanninen. 2000. Estimation of fine root mortality and growth from
simple measurements: a method based on system dynamics. Trees 14: 316-323.

Mann, W. F., Jr., and L. B. Whitaker. 1955. Effects of prescribe-burning 4-year-old
planted slash pine. USDA For. Serv. Fire Cont. Notes 16(3): 3-5.

Marshall, J. D. 1986. Drought and shade interact to cause fine-root mortality in Douglas-
fir seedlings. Plant and Soil 91: 51-60.

McClaugherty, C. A., J. D. Aber, and J. M. Melillo. 1982. The role of fine roots in the
Organic matter and nitrogen budgets of two forested ecosystems. Ecol. 63: 1481-1490.

Merritt, C. 1968. Effect of environment and heredity on the root-growth pattern of red
pine. Ecol. 49: 34-40.

NEldelhoffer, K. J ., and J. W. Raich. 1992. Fine root production estimates and
loWground carbon allocation in forest ecosystems. Ecology 73: 1139-1147.

,Namhiar, E. K. S. 1990. Interplay between nutrients, water, root growth and productivity
in y 0ling plantations. For. Ecol. Manage. 30: 213-232.
gems), D. G., C. C. Klopatek, L. F. DeBano, and P. F. Ffolliott. 1999. Fire effects on
Iowground sustainability: a review and synthesis. For. Ecol. Manage. 122: 51-71.

Perala, D. A., and D. H. Alban. 1982. Rates of forest floor decomposition and nutrient

0Ver in aspen, pine, and spruce stands on two different soils. USDA For. Serv. Res.
Pap. No-22 7- 5p.

Reid, J . B., I - Sorensen, and R. A. Petrie. Root demography in kiwifruit (Actim'dia
deliciosa). Plant, Cell Envir. 16: 949-957.

ROberts, B. A., and A. U. Mallik. 1994. Responses of Pinus resinosa in Newfoundland
t0 wildfire. J- Veg. Sci. 5: 187-196.

Rodrigues, M. L., C. M. A. Pacheco and M. M. Chaves. 1995. Soil-plant water relations,

root distribution and biomass partitioning in Lupinus albus L. under drought conditions.
J. Exp. Bot- 46(289): 947-956.

Santantonio, D., and R. K. Hermann. 1985. Standing crop, production, and turnover of

fine roots on dry, moderate, and wet sites of mature Douglas-fir in western Oregon. Ann.
Sci. For. 42: 113-142,

Santantonio, D., and E. Santantonio. 1987. Effect of thinning on production and

mortality 0f fine roots in a Pinus radiata plantation on a fertile site in New Zealand. Can.
J. For. Res. 1 7.- 919-928.

132

 

 

Schoettle, A. W., and T. J. Fahey. 1994. Foliage and fine root longevity of pines. Ecol.
Bull. 43: 136-153.

Sharrow, S. H., and H. A. Wright. 1977. Effects of fire, ash, and litter on soil nitrate,
temperature, moisture and tobosagrass production in the Rolling Plains. J. Range.
Manage. 30: 266-270.

Smith, D. M., and J. H. Sparling. 1966. The temperatures of surface fires in jack pine
barren. I. The variation in temperature with time. Can. J. Bot. 44: 1285-1292.

Smucker, A. J. M, S. L. McBurney and A. K. Srivastava. 1982. Quantitative separation

0f roots from compacted soil profiles by the hydropneumatic elutriation system. Agron.
1. 74: 400-503.

 

Soil Conservation Service. 1979. Soil survey of Kalamazoo County, Michigan. F. R.
Austin, ed. 102p.

Steele, S. J ., S. T. Gower, J. G. Vogel, and J. M. Norman. 1997. Root mass, net primary
pr odflcn'on and turnover in aspen, jack pine and black spruce forests in Saskatchewan and
Manltoba, Canada. Tree Phys. 17: 577-587.

fStrong, W. L., and G. H. LaRoi. 1985. Root density - soil relationships in selected boreal
crests of central Alberta, Canada. For. Ecol. Manage. 12: 233-251.

sutherland, E. K., W. W. Covington, and S. Andariese. 1991. A model of ponderosa pine
gr 0W3! reSponse to prescribed burning. For. Ecol. Manage. 44: 161-173.

Sutton, R. F- , and R. W. Tinus. 1983. Root and root system terminology. For. Sci.
Monograph No. 24. 137p.

SWezy, D. M - , and J. K. Agee. 1991. Prescribed-fire effects on fine-root and tree
mortality in old-growth ponderosa pine. Can. J. For. Res. 21: 626-634.

Tappeiner, J - C., and A. A. Alm. 1975. Undergrowth vegetation effects on the nutrient

Clintent 0f litterfall and soils in red pine and birch stands in northern Minnesota. Ecol. 56:
93-1200.

Van Wagner, C. E. 1971. Temperature gradients in duff and soil during prescribed fires.
For. Abstr. 32: 331 (2686).

Van Wagner, C. E. 1983. Fire behavior in northern conifer forests and shrublands. pp.
65-80 m The role of northern circumpolar ecosystems. Wein, R. W., and D. A. MacLean,
eds. John Wiley & Sons, New York, NY.

133

 

Vogt, K. A., and J. Bloomfield. 1991. Root turnover and senescence. pp. 287-306 in
Plant roots: The hidden half. Eds. Waisel, Y., A. Eschel, and U. Kafkafi. Marcel Dekker,

Inc., New York.

Vogt, K. A., and H. Persson. 1991. Measuring growth and development of roots. pp.
477-501 in Techniques and approaches in forest tree ecophysiology. Lassoie, J. P., and T.
M. Hinckley eds. CRC Press, Boca Raton, FL.

Vogt, K. A., D. J. Vogt, and J. Bloomfield. 1991. Input of organic matter to the soil by
tree roots. pp. 171-190 in Plant roots and their environment. Eds. B. L. McMichael and

H. Persson. Elsevier Publishers.

VOS, J. and J. Groenwold. 1987. The relation between root growth along observation

tubes and in bulk soil. pp. 39-49 in Minirhozotron observation tubes: Methods and

applications for measuring rhizosphere dynamics, ASA Special publication no. 50.
erican Society of Agronomy.

 

Waldmp, T. A., and F. T. Lloyd. 1988. Precommercial thinning of a sapling-sized
IOPIOlly pine stand with fire. South. J. Appl. For. 12: 203-207.

gum", H. 1968. Morphological studies of the root of red pine, Pinus resinosa. I.
r 0W”? characteristics and patterns of branching. Am. J. Bot. 55: 247-254.

{Wright’ R. J ., and S. C. Hart. 1997. Nitrogen and phosphorus status in a ponderosa pine
0’ est after 20 years of interval burning. Ecosci. 4: 526-533.

134

 

Chapter 4
Root thermal death point

 

135

 

Introduction

In conjunction with a larger study on the effects of fire on the fine roots of red
pine trees, a smaller study was undertaken to determine the thermal death point of red
pine roots in a field environment. The instantaneous thermal death point is often
Considered to be 60°C, but tissues can be killed at temperatures less than 60°C, depending
on time-of-exposure. This study was designed to determine if the often-quoted value of

60°C applies to red pine roots, and to determine how time-of-exposure interacts with

temperature.

HYPOtheses
Four hypotheses were proposed at the beginning of the experiment:
1. Any roots exposed to temperatures of 60°C and higher will be killed,
regardless of the time-of-exposure.
2. H eat-induced mortality will begin at 50°C.
3. Between 50°C and 60°C, survival will decrease with increasing time-of-
exposure at any given temperature.
4. Between 50°C and 60°C, survival will decrease with increasing temperature at

any given time-of-exposure.

Methods
Fine roots of mature red pine trees were exposed to high temperatures in the field
in mid'JUIy l 999. The stand used for this experiment was the same one used for the 1997

fire study (Chapters 2 and 3). All measurements were taken in what had been a control

136

 

 

plot, never burned. Only those roots in the forest floor or at the interface of the forest '-
floor and mineral soil that would be exposed to the highest temperatures during the

experimental fires were used (Figure 4-1). Individual roots were located by gently

removing the litter and duff until a root with a white tip was found. Once located, the

root was exposed to high temperatures (see below) or tagged as a “control” root. After a

root was treated, it was re-buried below the duff. The next root to be treated was then

located nearby, often within 0.5m. All roots were located within a circle of radius

 

approximately 15m and therefore probably came from a limited number of trees.
Five roots were exposed in each time-temperature combination (Table 4-1). Most
r OOtS were treated over the course of three days. However, those exposed to 55°C were
“:3th one week before all the other roots, and all were alive after one week. This result
was quite different from those roots that were exposed to higher or lower temperatures
during the following week. Therefore, the 55°C roots were excluded from statistical
analyses.

An el ectrical generator and Neslab 8-liter circulating water bath were taken to the
field in the bed of a utility vehicle that was driven between rows of trees. The water bath
was connected to a “cell” (Figure 4-2) that was used to expose individual roots to high
temperatures - An individual root was identified in the forest floor and then placed in a
foam plug Such that the tip of the root was actually in the empty cell; the foam plug
Prevented leakage of water. Once the root was in place, valves were opened, exposing

the root to heated water. After the appropriate amount of time, the valves were closed

and the 1' 00t was removed from the cell and the foam. A small amount of water was lost

137

 

Temp (°C) 00
N w is 01 93 2 n
._s m f" A

A

Figure 4-
location <
Lhel’mOCc

Table 4-]
\

Temperat

~~~3
m

-
u

“/7

6 s
\

J I

(

 

80 [rhermocouple error I
70 - l

60 -
50 - f"”"‘"“‘
40 - 4 Zem‘
30. ,; 6cml
20 .

10 -.
O

 

 

 

Temp (°C)

 

 

 

0481216202428323640
Time(min)

Figure 4-1. Temperatures at the surface of the mineral soil, 2 cm and 6 cm depths at one
location during low-intensity prescribed burning of a red pine stand, June 1997. The
thermocouple located at the surface malftmctioned above 73°C.

Table 4-1. Time-temperature combinations that were tested (X).

 

Exposure time (minutes)

 

kl!

Temperature (°C)

 

45

 

47.5

 

50

 

52.5

 

><1><><><><4>

55 (Not used)

 

57.5

><><><><><><><N

 

60

 

62.5

 

 

><><><><><><><><><P
><><><><><><><><><~

65

 

 

 

 

 

 

138

 

 

 

Alu‘

3 anz \

 

4.

Figure
“ aler.

 

 

 

 

88:88.85
:23 8&on

 

 

 

 

88 25 53>
win Each

 

 

 

All 835/

 

 

 

 

53 88>»
8 858%

 

 

 

 

 
 

   

 

 

88 820
832 ch

 

 

 

 

53 88>» 88m

 

 

Figure 4-2. Diagrammatic representation of the cell in which roots were bathed in hot

water.

139

from the cel

 

the dufi‘ and
weeks. F in
mortality.

Che
Therefore.
temperatur
temperatui
as the nex
followed 1
W

bath Wag
again det.
all timeg‘
R

a “live"

from the cell at this point. A tag was placed around the root and it was re-buried beneath
the duff and litter. A flag was placed nearby to locate these individual roots in later
weeks. Five “control” roots were also located and tagged to determine background
mortality.

Changing the temperature of the water bath was somewhat time-consuming.
Therefore, a given temperature was chosen at random and all roots to be exposed to that
temperature were individually treated before changing temperature. Within a
temperature, time-of-exposure was also randomized. For example, once I selected 575°C
as the next temperature to be tested, I exposed (individually) all the “4 minute” roots,
followed by the 1 minute treatment, 0.5 minutes, then 2 minutes.

Water temperature was measured directly in the cell and the temperature of the
bath was adjusted accordingly before each group of roots was treated. Temperature was
again determined after all roots had been treated and periodically between samples. At
all times, temperature did not vary from the beginning of the run until the end.

Roots were examined after 1, 2 and 3 weeks had passed. Roots were categorized
as “live” or “dead” based on appearance. Roots which were still firm were considered

alive, even if they had turned brown. Black, shriveled roots were considered “dead”.

Statistical analyses
Because the response variable was binary (alive or dead), logistic regression was
the most-appropriate method of analysis (Myers 1990, Allison 1999). Weeks 1, 2 and 3

were analyzed separately. The logistic regression model is:

I40

 

 

 

ln (1)-“(1'90 "

where p is t?
coefficients

Bas
predict sur

were signi'

Results
Tl
Slll'ViVal a
diagonal -
influence
Otheru'i S
generall§
6'ng SUre

[emperar

1n (p/(l-p)) = B0 + B1 (Time) + B2 (Temperature) + B3 (Time x Temperature)

where p is the‘probability of a root being alive, B0 is the intercept, and Bl-3 are the
coefficients for the other model parameters.

Based on the results of the logistic regression, an idealized model was created to
predict survival based on temperature and time-of—exposure. Only those variables that

were significant in the regression were used in the model.

Results

There were some anomalies in survival after one week (Figure 4-3), the “spike” in
survival at 0.5 min at 65°C is especially problematic. Regression diagnostics (Hat matrix
diagonal and DFBETAS) identified this data point (actually 5 roots) as having high
influence in the model. Therefore, these roots were removed from the analysis.
Otherwise, survival patterns mostly occurred as expected after one week. Survival
generally decreased with increasing temperature above 50°C, within a given time-of-
exposure. Longer times-of—exposure usually resulted in lower survival within a given

temperature above 50°C.

14]

 

 

Hgne4-
C mar.

Signific
lime W
increas
imefac
Inthej

for 4S

 

 

so
60 ES
'2'
40 S
U)

20

Time (min)

    

52.
Temperature (‘C)

Figure 4-3. Survival of fine roots 1 week following hot-water treatment. Temperature
“C” indicates controls.

After 1 week, mortality generally began at 525°C, not counting the drop at
47.5°C/1 min (Figure 4-3). No survival was expected at temperatures of 60°C. However,
after 1 week, all temperature-time combinations at 60°C and 625°C had some living
roots, except for 65°C/1 min. All 5 control roots were still alive after 1 week.

The intercept, temperature and the time x temperature interaction all were
significant in logistic regression of the first week’s data. (Table 4-2). The coefficient for
time was positive, indicating that as time-of—exposure increased, survival actually
increased. However, the coefficients for temperature and the time x temperature
interaction were both negative, which compensate for the effect of time to some extent.
In the idealized model (Figure 4-4), survival increased with increasing time-of-exposure

for 45 and 475°C but decreased with increasing time-of exposure at 525°C and higher.

142

Table 4-3
followiné'
ma=01

m

r———f—'

: lnterce
- Tempe

Time x
g R‘ = 0.

 

 

FlgUre 4
lOgistic 1
“Ere Sig

exmeted
end of tu
With inc;

Table 4-2. Results of the logistic regression analysis of survival of fine roots one week J.
following a hot water treatment. Bold values indicate significant independent variables

ata=0.10.

Probability > X

1me . .11

Temperature
1me x Temperature

 

.100

80
-60 =3
E
40 S
m

20

- 0

Time (min)

 

 

Temperature (°C)

Figure 4-4. Idealized model of root survival after one week, based on the results of
logistic regression. The intercept, temperature and the time x temperature interaction
were significant factors in the model, 1 week after exposure.

Between the first and second weeks survival again generally following the
expected pattern (Figure 4-5). Most roots exposed to 575°C and higher were dead by the
end of two weeks. Within each of the lower temperatures, survival generally decreased

with increasing time-of—exposure, although there were a few unexpected results. At

475°C, survival was not different among exposure times from 1 to 4 minutes. Also,

143

 

survival
not neCc‘

still alix‘

(Figure 4
C ' indi

 

survival at 50°C/0.5 minutes was only 40%. Within a given time however, survival did
not necessarily decrease across temperatures from 45 to 525°C. All control roots were

still alive after two weeks.

 

 

Survival (%)

Time (min)

 

Temperature ('C)

 

 

 

Figure 4-5. Survival of fine roots 2 weeks following hot-water treatment. Temperature
“C” indicates controls.

The logistic regression indicated that all factors were significant in the model
(Table 4-3). The coefficient for time was negative at two weeks. The interaction term
became positive however, which offset some of the main effects of time and temperature.
In the idealized model (Figure 4-6), survival decreased with increasing temperature.

Survival decreased with increasing time-of-exposure only at 55°C and lower.

144

Table 4-3-
following
I at (1 = 0.1

 

; Parame

i lnterce
l Time

r——-—*
Tempe
l-—-—.———
! Time )

l'R‘20

 

 

 

Flgllre
10gisti
exDOS

Table 4-3. Results of the logistic regression analysis of survival of fine roots two weeks
following a hot water treatment. Bold values indicate significant independent variables

at a = 0.10.

Probability > X

1me .
emperature . .0001
Time x Temperature . 1
R = 0.698

 

 

 

 

Survival (%)

Time (min)

 

 

 

Temperature (°C)

Figure 4-6. Idealized model of root survival after two weeks, based on the results of the
logistic regression. All factors were significant in determining survival two weeks after
exposure.

Three weeks after exposure, little additional mortality had occurred above 55°C
(Figure 4-7). Below 55°C, survival dropped more at the lower exposure times (0.5 and 1

minute) than at 2 or 4 minutes. Control roots began to show mortality at this point and

some time-temperature combinations had greater survival than controls.

145

 

 

Intercept and temperature were the only significant factors in the model three

weeks afier treatment (Table 4-4). Maximum predicted survival ranged from less—$39

1% to 82.5% (Figure 4-8).

 

 

 

Survival (%)

Time (min)

 

C Temperature ('C)

_- - If-.———-— — 'wifl- " —T\

 

 

Figure 4-7. Survival of fine roots 3 weeks following hot-water treatment. Tert'q:)e,.a
“C” indicates controls. titre

Table Act-4. Results of the logistic regression analysis of survival of fine roots three Weeks
following a hot water treatment. Bold values indicate significant independent variables

at on = 0.10.

arameter Probability > X

true 1

emperature
1me X emperature
= .7

 

146

 

Time (

legist
deten

Discn

m0;-

eXp

_ 1 CO
-80
so 53:
To
.2
41:: E
(I)
20
0-5

Tithe (min )

 

 

Temperature (°C)

Figure 4-8- I dealized model of root survival after three weeks, based on the res-.1 It s
logistic regression. Intercept and temperature were the o

. nly significant factors i n ofthe
determining survival three weeks after exposure.

Discussion

In this study, root mortality was detected at 525°C (Figure 4-3). This Observation
is consistent with published reports that the onset of heat-induced mortality occurs
between 50 and 55°C. For example, Gentile and Johansen (1956) found almost complete
mortality 0f outplanted slash pine and sand pine seedlings whose root systems had been
exposed to 52°C and higher. According to the literature, complete mortality occurs at
temperatures of 60°C and higher. However, I found that some roots exposed to these

temperatures were still alive after one week in my experiment (Figure 4'3)-

147

 

 

  
 

51mm I
lkcond
tunpnap
H96T)i

46°C f0

ahere
disc 0‘
obser
Once
obse
iflfln
not

till 5

. . . . . re
Two pOSSibilities may explain the unexpected results: (1) red pine roots are mo

. . (3
thel'mOtolerant than other tissues, or (2) red pine roots are not more thermotolermt ’39

the results are an artifact of the methods used in this study
Roots are probably not more thermotolerant than other tissues. For example
Shirley (1 9 3 6) tested shoots and roots of seedlings of several species including
He concluded that roots are more sensitive than above-ground tissues to high
temperature 5. Although heat-induced mortality begins between 50 and 55°C Ursic
(1961) found some mortality of outplanted loblolly pine seedlings after exposure to Only

46°C for ‘5 minutes.

Therefore the methods I used may help to explain why some roots were 81:1 I I al
N

after exposure to 60°C and higher. Immediately following treatment, no roots were
discolored or shriveled at any time-temperature combination. During subseque tit

observations, roots were considered dead if they were black and/or shrunken. HOW
eve
r

once a root tip had shrunk, it could have broken away from the main root betwe en

observations leavmg a brown root. In such a case, the root would be considered a ,

1 De 3,
although the tip had actually been killed by the treatment. In my study, the metal ta

was
not placed at a specific distance from the end of the root; therefore I didn t realize Whe

this situation (broken root tips) had occurred

The logistic regression analysis of the survival after one week indicated that the

coefficient for time was positive which does not fit theory. As time-of-exposure
increases survival should decrease (all other things being equal) There are many cases
wher e suI'Vlval remained the same or even increased with increasing time-of—expesure

(Figure 4 3 ) A lack-of-change in survival would result in a coefficient of zero for time

148

red pine.

 

, i.
1 .
' ,
‘ '5
it.
.5‘~..'.
,‘i‘-.'
53:1
.- 't'

 

 

 

but would

the up ‘

Slil'VlV'cll é

 

 

time beta
Tl
theory. \
increasing
coellicie
combine
that ext
at \hes
2them

“as ]

bee:

QCQ

 

but W0llld probably not change the coefficient from negative to positive. Survival fiends
after two weeks also did not necessarily follow theory. Within a given temperannes
surVivaI di (1 not always decrease with increasing exposure time (Figure 4-5). Howe” ex 9
time became significant in the model.

Three weeks after treatment, survival trends again did not completely F0 110w
theory. W i thin a given temperature, survival again did not necessarily decrease with
increasing exposure time (Figure 4-7), which again had the effect of a non-significant
coefficient for time in the regression. Roots exposed to certain tirne-temperatLue
combinations had higher survival rates than controls after 3 weeks. This woul 6 indicate
that exposing roots in this manner (water bath) was actually beneficial to survi val at least
at these lower temperatures. However, the amount of water lost from the cell was on]
about 20 ml each time. This small quantity may have affected roots for a few d ays’ b 3’
was probably not a factor in roots’ survival after three weeks. Ut

A more likely explanation of the 3-week survival trends is that roots were dyj
because of factors other than the heat treatment. That is, “normal” root death “’83 11g
occurring. Analysis of minirhizotron images (see Chapter 3) indicates that over 2 5% or

“new” roots Were dead within 4-5 weeks of their initial observation (range by cohort 6.
72%, all depths). Of these roots, 5.3% were dead at their initial viewing with estimated
lifespans of less than 2.5 weeks. Median root lifespan can be less than 20 days (Kosola et
al. 1995, Black et al. 1998), although conifers may generally have a longer median feet
lifespan than hardwood species (cf. Rygiewicz et al. 1997, Black et al. 1998)-
The coefficients for time and the time x temperature interaction were always of

Opposite Signs while the coefficient for temperature was always negative. In the idealized

149

 

 

 

 

models. 3

  
 

negative.

increasin:

tempera
third m
Survive

\Emge

(199
Shm

terr

 

. . . . ' %
models, survival increased Wlth increasmg exposure times at lower temperatures duf‘“

oi “‘6

the first week, and at higher temperatures during later weeks. As long as the signs

. . - s
coefficients for time and the interaction are opposrte and that for temperature rema‘“

 

negative, 1' dealized results should closely follow theory, with survival decreasin g with
increasing time. Future work, using many more replications, will help to determine if
this is true -

The only significant factors in the model after three weeks were the intercept and
temperature - The coefficient for temperature remained negative between the second and
third weeks - Because the coefficients of the model factors did not change Sign 9 long-term
survival of fine-roots can probably be determined within two weeks of exposm r—ce to high

temperatures.

Another source of variability in this experiment was time-of-day. C010 131 50

fit a]
(1995 ) reported that roots of black spruce seedlings had higher levels of constiifiati"e
bea
t

to
temperatures of only 5 to 10°C above usual growth temperatures can result in produc
tio
I)

shock proteins (HSPs) in the afternoon compared to the morning. Rapid expo sme

of HSPs (Kozlowski and Pallardy 1997a). Whether or not the results of Colombo et a1

(1995) are applicable to mature trees of red pine is unknown. In this experiment, an TOOts
were treated between approximately 10 am. and 3 pm. If the fine roots of mature red
pine trees have higher levels of HSPs in the afternoon, then time-of-day must be
accounted for in future experiments.
Another question involves the order in which roots were treated. specifically, did

earlier treatment of roots induce heat-hardening in roots from the same tree that were

treated later? The plot in which this eXperiment was performed contained 68 trees, while

150

 

more thm‘

evenly st

 

ilmelon

expen'me
heir exp
minim
acid and
possiblc
temper
\sz c
heat-
Whic

DECC

COT

60‘

El

more than 1 3 5 roots were exposed to high temperatures. Also, the treated roots were fio‘
evenly Spaced throughout the plot but were somewhat clustered near the center.
Therefore, the roots which I used probably came from only a portion of those 6 8 $665.
K0 ppenaal et al.’s (1991) results on heat-hardening are difficult to apply to my
experiment as their pre-treatment (hardening) temperatures were only 34 to 40° C, while
their eXpo s we times ranged from 30 minutes to 360 minutes. Also, their harde :ning
conditions were applied every day for 3 or 6 days. Certain hormones, includin g abscisic
acid and jasmonates, are induced by wounding (Kozlowski and Pallardy 1997 b). It is
possible bat these signal molecules are induced in tissues that are exposed to high
temperatures, then travel to other sites within the tree where they elicit a heat— hardening
type of response. Another signal molecule, brassinosteroid, is particularly imp licated in
heat- shock responses (Kozlowski and Pallardy 1997b). The quantity and speed with
which these signal molecules are produced, and how quickly they are transported in a.
need to be determined. ‘ ees
Future work should also use longer times-of-exposure. During one of the fires
conducted in this study, temperatures were above 45°C for over 15 minutes and above
60°C for over 11 minutes (Table 4-5). The temperatures and times measured at that
location are only one sample of temperature dynamics within a fire. Other fires will have
different temperature profiles, depending on their intensity and rate of spread. The data
presented here should only be used as a guideline. Although there are longer times-of-

exposure listed here (Table 4-5), it should be noted that, at temperatures of 525°C and

higher, mortality began at much shorter exposure times.

151

 

 

 

 

 

Table 4-
during a

 

numbe
COurse
Earlie
“as a
that c
treat;
EXPE-
1r can
at 1h,-
tTeate
time~
Comt-

Sp fin

Taole 4-5. Actual time and temperature information at the surface of the mineral so“
dw‘mg a backfire in a red pine stand. A system of thermocouples connected to a
datalogger was used to record the information.

Temperature 1me at or

given temperature
45

47.

 

10:

1f filnher experiments are attempted, they will have certain difficulties .. Th e
number of replications should probably be increased. All roots must be treated Over the
course of a few days. Those roots exposed to 55°C in my experiment were treated a w
earlier than all others and they reacted quite differently. I believe this was Metre-.11Se lb eek
was a rainstorm (2.67 cm) the day after roots were treated at 55°C. The only preei . ere

that occurred following treatment of the other roots was 0.10 cm that fell 4 days aft 0;;

61-
treatment and 1.09 cm of rain that fell one week after the initial treatment. Future
experiments that treat all roots within a few days while adding replications and longer
treatment times will be challenging. An approach that measures immediate damage/death
at the cellular level may overcome these constraints to some extent. Roots can then be
treated at any time of the growing season if a complete experiment iS done at any one
time. However, Koppenaal and Colombo (198 8) cited several studies that indicate that

conifers are most heat-tolerant during winter dormancy, and most susceptible during the

Spring and early summer, when shoots are actively growing. Whether or not these results

are applicable to roots is unknown.

152

~ "fat
. ‘ .r i
‘ a H'At’W‘R- ’

IQ

7‘!
E3

“5;“

. . my!

 

 

attention

 

 

root. melt
plug and
been trea
were not
approxt
were to e

better st

System
Were e
or eve
OIhErs
Seeclli
r001, ;
Stud}-

5}'5161

 

e
Furthermore, the water bath and cell system used in this study required conS‘afl
a
attention. On occasion the foam plug would detach from the cell during treatment of
. 0am
root, making that root unusable. Caution had to be exercised during removal of the i

plug and root from the cell after treatment. If not done properly, the root that had jUS‘
been treated could detach from the main root, again making that root unusable - Records
were not Kept on how often these failures occurred, but they happened regularl :1, on
approximately 5-10% of attempted treatments. If the failure rate observed in tl—ris study
were to 00¢ 111' while using longer times-of-exposure, much time would be wast cd. A
better system for holding the cell in place during treatment needs to be designs (1.

It i s difficult to directly compare this study with those in which the entire root
systems of seedlings were exposed to high temperatures. Although entire root Systems
were exposed to high temperatures in the other studies, not all roots may have been kill
or even damaged. Some roots survive a given time-temperature combination While ed
others do not. In seedling studies, enough roots may be killed or damaged that 11) e

seedling will not survive. The main focus in this study was at the level of the individua
root, not the: whole plant. Although some roots were certainly killed by the fires in 1
study (Chapter 3), they probably constituted only a small fraction of the entire mat
system.
The temperatures reached at the surface of the mineral soil during low-intensity
Prescribed fire (Figure 4-1) are high enough to kill roots. Higher root mortality will
result in more carbon entering the soil system. Ecologically, this would be beneficial to

the 30” 01‘ ganisms that rely on tree roots as their primary source of energy. If soil

or gamsm pepulations are increased then trees will experience more competition for water

 

 

 

and nutril

teem tr
approprrt
managen

the trees ‘

 

Condusi

and nutrients while trying to rebuild damaged root systems. Where prescribed btnnh‘g as
used to mimic natural disturbance, medium-to-high intensity fires may be more

appropriate to use than low-intensity fires. However, where the overall goal of

management is strictly for trees as a timber resource, low-intensity fires that do not ban“

the trees in any way would be more appropriate.

Conclusio n s

The hypotheses I proposed at the beginning of this experiment were supported
only slightly. Not all roots exposed to 60°C and higher were killed (Hypothes i s 1 ).
However, these results may have been due to the methods employed in this e): petill’lent
Onset of heat-induced mortality occurred at 525°C instead of 50°C (Hypothesi s 2).
Although this hypothesis was not completely supported the results reported he 1.1:: are
similar to those reported in the literature. Increased mortality with increasing ti 11) e‘og
exposure Within a given temperature (Hypothesis 3) was only minimally suppo 118d.
Roots exposed to 525°C or 575°C for 4 minutes had lower survival than those exposed
for only 0.5 minutes (Figure 4-3). The results for 1- or 2-minute exposure times are
highly variable. Decreasing survival with increasing temperature at the same time~ofl

exposure (Hypothesis 4) was also supported to some extent, although the lack of data at

55°C make this conclusion rather tentative.

Literature cited

Alexandrov, V, Ya. 1977. Cells, molecules and temperature. Conformational flexibility
0f macromolecules and ecological adaptation. Springer-Verlag, New York. 330p.

154

 

 

 

Allison, P. D. 1999. Logistic regression using the SAS system: Theory and application.
SAS Institute, Cary, NC 302p.

Baker, F. S. 1929. Effect of excessively high temperatures on coniferous reproduction. J.
For. 27: 949-975.

Bates, C. G., and J. Roeser, Jr. 1924. Relative resistance of tree seedlings to excessive
heat. USDA Bull. 1263. 16p.

Black, K. E., C. G. Harbron, M. Franklin, D. Atkinson and J. E. Hooker. 1998.
Differences in root longevity of some tree species. Tree Phys. 18: 259-264.

Blodgett, F. M. 1923. Time-temperature curves for killing potato tubers by heat
treatments. Phytopath. 13: 465-475.

Bradstock, R. A., A. M. Gill, S. M. Hastings, and P. H. R. Moore. 1994. Survival of
serotinous seedbanks during bushfires: Comparative studies of Hakea species from
southeastern Australia. Aust. J. Ecol. 19:276-282.

Byram, G. M. 1948. Vegetation temperature and fire damage in the southern pines. Fire
Cont. Notes 9(4): 34-36.

Byram, G. M. 1958. Some basic thermal processes controlling the effects of fire on
living vegetation. USDA For. Serv. Southeast. For. Exp. Sta. Res. Notes 114. 2p.

Byram, G. M., and R. M. Nelson. 1952. Lethal temperatures and fire injury. USDA For.
Serv. Southeast. For. Exp. Sta. Res. Notes 1. 2p.

Colombo, S. J., and V. R. Timmer. 1992. Limits of tolerance to high temperatures
causing direct and indirect damage to black spruce. Tree Phys. 11: 95-104.

Colombo, S. J., V. R. Timmer, M. L. Colclough, and E. Blumald. 1995. Diurnal
variation in heat tolerance and heat shock protein expression in black spruce (Picea
mariana). Can. J. For. Res. 25: 369-375.

Connaughton, C. A. 1936. Fire damage in the ponderosa pine type in Idaho. J. For. 34:
46-51.

Daubenmire, R. F. 1943. Soil temperature versus drought as a factor determining lower
altitudinal limits of trees in the Rocky Mountains. Bot. Gaz. 105: 1-13.

Gentile, A. C., and R. W. Johansen. 1956. Heat tolerance of slash and sand pine
seedlings. USDA For. Serv. Southeast. For. Exp. Sta. Res. Notes 95. 1p.

Hare, R. C. 1961. Heat effects on living plants. USDA For. Serv. South. For. Exp. Sta.
Occas. Pap 183. 32p.

155

 

Harrington, M. G., and R. G. Kelsey. 1979. Influence of some environmental factors on

initial establishment and growth of ponderosa pine seedlings. USDA For. Serv. Res. Pap.

INT-230. 26p.

Helgerson, O. T. 1990. Heat damage in tree seedlings and its prevention. New For. 3:
333-358.

Ingram, D. L. 1986. Root cell membrane heat tolerance of two dwarf hollies. J. Am. Soc.

Hort. Sci. 111: 270-272.

Isaac, L. A. 1938. Factors affecting establishment of Douglas-fir seedlings. USDA Circ.
No. 486.

Kappen, L. 1981. Ecological significance of resistance to high temperature. pp. 439-474
in Physiological plant ecology 1: Responses to the physical environment. Encyclopedia

of plant physiology, New series, Vol. 12A. Eds. O. L. Lange, P. S. Nobel, C. B. Osmond
and H Ziegler. Springer-Verlag, New York.

Kayll, A. J. 1966. Some characteristics of heath fires in NE Scotland. J. Appl. Ecol. 3:
2940.

Kolb, P. F., and R. Robberecht. 1996. High temperature and drought stress effects on
survival of Pinus ponderosa seedlings. Tree Phys 16: 665-672.

Koppenaal, R. S., and S. J. Colombo. 1988. Heat tolerance of actively growing, bud-
initiated, and dormant black spruce seedlings. Can. J. For. Res. 18: 1103-1105.

Koppenaal, R. S., S. J. Colombo, and E. Bltunwald. 1991. Acquired thermotolerance of
jack pine, white spruce and black spruce seedlings. Tree Phys. 8: 83-91.

Korstian, C. F., and N. J. Fetherolf. 1921. Control of stem girdle of spruce transplants
caused by excessive heat. Phytopath. 11: 485-490.

Kosola, K. R., D. M. Eissenstat, and J. H. Graham. 1995. Root demography of mature
citrus trees: the influence of Phytophthora nicotianae. Plant and Soil 171: 283-288.

Kozlowski, T. T., and S. G. Pallardy. 1997a. Growth control in woody plants. Academic
Press, San Diego 641p.

Kozlowski, T. T., and S. G. Pallardy. 1997b. Physiology of woody plants. 2nd ed.
Academic Press, San Diego. 411p.

Larcher, W. 1983. Physiological plant ecology. 2"“l ed. Springer-Verlag, New York. 7
303p.

156

 

 

Levitt, J. 1980. Responses of plants to environmental stresses, 2nd ed. Academic Press,
New York. 497p.

Lorenz, R. W. 1939. High temperature tolerance of forest trees. Minn. Ag. Exp. Sta.
Bull. 141. 25p.

McConkey, T. W., and D. R. Gedney. 1951. A guide for salvaging white pine injured by
forest fires. USDA For. Serv. Northeast. For. Exp. Sta. Res. Notes 1]. 4p.

Myers, R. H. 1990. Classical and modern regression with applications. PWS Publishers.
Boston, MA. 488p.

Nelson, R. M. 1952. Observations on heat tolerance of southern pine needles. USDA
For. Serv. Southeast. For. Exp. Sta. Pap. 14. 6p.

Roeser, J ., Jr. 1932. Transpiration capacity of coniferous seedlings and the problem of
heat injury. J. For. 30: 381-395.

Rygiewicz, P. T., M. G. Johnson, L. M. Ganio, D. T. Tingey, and M. J. Storm. 1997.
Lifetime and temporal occurrence of ectomyconhizae on ponderosa pine (Pinus

ponderosa Laws.) seedlings grown under varied atmospheric C02 and nitrogen levels.
Plant and Soil 189: 275-287.

Seidel, K. W. 1986. Tolerance of seedlings of ponderosa pine, Douglas-fir, grand fir, and
Engelmann spruce for high temperatures. Northwest Sci. 60: 1-7.

Shirley, H. L. 1936. Lethal high temperatures for conifers, and the cooling effect of
transpiration. J. Agr. Res. 53: 239-258.

Smith, D. M. 1951. The influence of seedbed conditions on the regeneration of Eastern
white pine. Conn. Ag. Exp. Sta. Bull. 545. 61p.

Toumey, J. W., and E. J. Neethling. 1923. Some effects of cover over coniferous
seedbeds in southern New England. Yale Univ. School of Forestry Bull. 9. 39p.

Ursic, S. J. 1961. Lethal root temperatures of 1-0 loblolly pine seedlings. Tree Pl. Notes
47: 25-29

Vaartaja, O. 1949. High surface soil temperatures: On methods of investigation, and
thermocouple observations on a wooded heath in the south of Finland. Oikos 1: 6-28.

Vines, R. G. 1968. Heat transfer through bark, and the resistance of trees to fire. Aust. J.
Bot. 16: 499-514.

Wright, H. A. 1970. A method to determine heat-caused mortality in bunchgrasses. Ecol.
51: 582-5 87.

157

'7.-

 

Wright, H. A., and A. W. Bailey. 1982. Fire ecology: United States and southern
Canada. John Wiley and Sons, New York, 501p.

158

 

Appendix

159

 

Sample
Date
Sept96
Sept96
Sept96
Sept96
Sept96
Sept96
Sept96
Sept96
Sept96
Sept96

Oct96
Oct96
Oct96
Oct96
Oct96
Oct96
Oct96
Oct96
Oct96
Oct96

Apr97
Apr97
Apr97
Apr97
Apr97
Apr97
Apr97
Apr97
Apr97
Apr97

Burn
Burn
Burn
Burn
Burn
Control
Control
Control
Control
Control

Burn
Burn
Burn
Burn
Burn
Control
Control
Control
Control
Control

Burn
Burn
Burn
Burn
Burn
Control
Control
Control
Control
Control

1

MANN—‘MAWN-d MAUJNHLh-hWN

MANN—‘M-hUJN"

76.9

6.6
26.9
15.6
10.5

7.3
19.3
10.6
10.8
29.2

64.3
13.7
30.4
27.7
22.9

9.8
33.0
21.6
16.0
57.8

47.9
11.5
36.0
22.3
16.4
14.7

8.9
18.4

9.8
47.9

160

Mg2+ Ca2+ K+
TreatmentReplication (ppm) (ppm) (ppm) Roots (g) Roots (g)

516
39
130
99
70
49
114
65
82
166

443

85
190
205
120

55
177
179
103
282

337

71
223
150
111
108

48
124

61
251

Data for 10-25 cm depth, not statistically analyzed

37.7
22.5
26.1
29.7
30.3
21.3
26.6
26.8
27.6
31.9

32.5
21.9
29.2
33.9
33.6
22.0
29.8
29.2
28.7
33.0

32.4
24.7
26.4
31.0
28.5
21.5
19.7
29.0
30.3
40.9

 

Fine Live Fine Dead

Sample

Date

May97
May97
May97
May97
May97
May97
May97
May97
May97
May97

July97
July97
July97
July97
July97
July97
July97
July97
July97
July97

Aug97
Aug97
Aug97
Aug97
Aug97
Aug97
Aug97
Aug97
Aug97
Aug97

Mg2+ Ca2+ K+

Fine Live Fine Dead

TreatmentReplication (ppm) (ppm) (ppm) Roots (g) Roots (g)

Burn
Burn
Burn
Burn
Burn
Control
Control
Control
Control
Control

Burn
Btu'n
Burn
Burn
Burn
Control
Control
Control
Control
Control

Burn
Burn
Burn
Burn
Burn
Control
Control
Control
Control
Control

1

MAWNHm-bb-JN

MAWNt-‘MAUJNH

Lh-fi-UJN—‘M-5LJJNH

36.2
11.3
16.7
18.8
14.6
11.9
41.2
19.8

8.7
28.8

63.7
11.0
21.5
29.0
17.0
17.9
10.6
20.9

5.6
46.0

34.8
14.7
19.9
15.3
19.3
18.5
12.3
14.2
13.0
47.6

161

274
56
114
1 ll
92
86
234
148
48
172

375

69
122
185
114
104

67
151

38
236

237
92
89

102

101

135
59
95
79

234

26.5
24.6
20.8
28.7
25.9
20.6
31.0
28.6
25.7
30.8

37.4
23.3
26.7
31.6
28.9
25.0
21.5
29.5
18.0
37.8

28.6
27.8
21.2
25.9
28.2
24.3
24.5
28.6
22.7
30.8

0.356
0.619
0.365
0.516
0.397
0.509
0.509
0.224
0.513
0.419

0.222
0.184
0.363
0.285
0.205
0.265
0.354
0.195
0.231
0.122

0.242
0.293
0.350
0.401
0.399
0.344
0.412
0.242
0.321
0.330

0.630
0.959
0.680
0.775
0.532
0.900
0.732
0.480
0.874
0.605

1.058
0.973
1.452
0.969
1.003
1.142
1.151
0.832
0.916
1.039

0.755
0.871
1.128
1.012
1.164
1.271
1.067
0.604
0.897
0.785

 

Sample

Date

Sept97
Sept97
Sept97
Sept97
Sept97
Sept97
Sept97
Sept97
Sept97
Sept97

Oct97
Oct97
Oct97
Oct97
Oct97
Oct97
Oct97
Oct97
Oct97
Oct97

Nov97
Nov97
Nov97
Nov97
Nov97
Nov97
Nov97
Nov97
Nov97
Nov97

TreatmentReplication (ppm) (ppm) (ppm)

Burn
Burn
Burn
Burn
Burn
Control
Control
Control
Control
Control

Burn
Burn
Burn
Burn
Burn
Control
Control
Control
Control
Control

Burn
Burn
Burn
Burn
Burn
Control
Control
Control
Control
Control

1

LII-hWNv-‘M-hUJNH MAWN—‘M-bUJN

M-hUJN—‘Lh-bbJN"

162

Mg2+ Ca2+ K+
42.1 295 30.0
18.3 128 30.3
29.4 184 25.5
28.8 170 33.5
11.1 71 27.4

7.5 40 20.5
6.3 42 16.5
9.8 68 24.7
5.3 26 15.8
57.1 239 40.8
63.2 388 40.8
9.5 56 25.6
10.2 60 24.3
17.8 110 25.9
17.8 101 27.6
18.7 146 22.6
16.5 89 22.0
16.1 114 33.4
6.5 41 18.5
30.9 179 27.3
39.2 347 28.8
6.8 35 22.0
8.4 53 19.3
32.7 199 28.6
12.0 66 28.8
12.0 75 18.6
12.8 78 22.5
10.7 65 23.7
4.5 24 15.5
29.3 177 33.5

Fine Live Fine Dead
Roots (g) Roots (g)

0.483
0.424
0.820
0.608
0.367
0.753
0.421
0.424
0.347
0.483

0.761
1.021
1.052
1.018
0.836
1.229
0.929
0.622
0.764
0.842

 

IIIIIIIIIIIIIIIIIIIII

lllllllll