a

.3? .o

x515: . .

.12. I.

I

I} . .

.17.}: . .
1 .

 

 

57‘...

.ma

1: ?.

f
.. S:
a a

.

we,»

mm;

:1
“Em ”

i?

51‘ 19 2 u
.4. 1. .w: ,
a .. is...) r

.43.... 3.." sh.
1‘. . a.) _

.1132“.

This is to certify that the
thesis entitled

Tree Seedling Growth, Survival and Morphology in Response
to Landscape Level Variation ln Soil Resource Availability

presented by

Laura A. Schreeg

has been accepted towards fulfillment
of the requirements for the

 

 

  

Forestry and Ecology.
“gag‘ggf degree in Evolutionary Biology and
Behavior
:\ 7?

42/114,711 [C (aki/ / /// fl// .1"
‘ Major Professor’s Signature v 6""

30 Nov 02,”

Date

MSU is an Affirmative Action/Equal Opportunity Institution

 

LIBRARY
Michigan State
University

 

 

 

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

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 c:/ClRC/DateDue.p65-p.15

 

TREE SEEDLING GROWTH, SURVIVAL AND MORPHOLOGY IN RESPONSE TO
LANDSCAPE LEVEL VARIATION IN SOIL RESOURCE AVAILABILITY
By

Laura A. Schreeg

A THESIS
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
MASTER OF SCIENCE
Departments of Forestry and Ecology, Evolutionary Biology and Behavior

2002

ABSTRACT
TREE SEEDLING GROWTH, SURVIVAL AND MORPHOLOGY IN RESPONSE To
LANDSCAPE LEVEL VARIATION IN SOIL RESOURCE AVAILABILITY
By

Laura A. Schreeg

In the northern lower peninsula of Michigan, landscape-level variation in tree
species composition is associated with glacial landforms which represent an increasing
gradient of soil nutrient and water availability (i.e. outwash < ice contact < moraine). To
investigate the causes of this association, we conducted a reciprocal seedling transplant
experiment with 5 species: Acer saccharum (sugar maple), Fraxinus americana (white
ash), Quercus alba (red oak), Prunus serotina (black cherry) and Q. velutina (black oak).
Fertilizer additions of calcium, nitrogen and calcium x nitrogen were included to test for
limitation of these nutrients. One year after transplanting, for seedlings grown in higher
light plots (14-26% light), a trade-off was found between survival on outwash (poorer
soil resource site) and relative growth rate (RGR) on moraine (richer soil resource site),
suggesting species composition may be determined by maximizing tolerance on outwash
and maximizing competitive ability on moraine. Our results suggest this trade-off may
be underpinned by variation in size and morphology among species. At low light (3-
10%), a trade-off between tolerance to ice—contact and growth on moraine was not
supported by our results. Fertilizer additions were not found to effect survival or growth.
During 2001 , the year the seedlings were harvested, soil moisture was low for much of

the growing season due to a severe drought.

ACKNOWLEDGMENTS

I would like to thankfully acknowledge my advisors, Rich Kobe and Mike
Walters, for their time, effort and financial support. I would also like to thank Phil
Robertson, the third member of my committee, for his advice and suggestions on this
project. I am also grateful for the help I received from the Department of Forestry, the
MSU Tree Research Center, the Rothstein lab and my lab mates, especially Meera Iyer,
Jesse Randall, Justin Kunkle and Laura Marx. In addition, I would like make a special
acknowledgment to Corine Vriesendorp for reliably inspiring conversations and her
boundless support and encouragement. I would like to thank the Organization for
Tropical Studies for the opportunity to participate in their graduate course, which helped
me become a stronger biologist and a more creative teacher. I am also obliged to Sigma

Xi and the Hanover Biology Scholarship program for financial support.

iii

TABLE OF CONTENTS

LIST OF TABLES
LIST OF FIGURES
INTRODUCTION
METHODS
RESULTS
Soil resource availability
Plant survival, growth and morphology
DISCUSSION
Growth and survival
Traits underlying grth and mortality differences
Fertilizer effects
CONCLUSION
APPENDDC

REFERENCES

iv

vi

17
17
18
27
27
30
31
33
34

68

LIST OF TABLES

Table 1 p 35
Soil resource landform characteristics.

Table 2 p 36
Site characteristics.

Table 3 p 37
Soil nutrient availability one month after fertilizer application.

Table 4 p 38
Volumetric soil water availability measured with TDR (0-30 cm) or calculated from
gravimetric and bulk density data (0-10 cm).

Table 5 pp 39-40
Species comparisons within a site.

Table 6 pp 41-42
Site comparisons within species.

Table 7 pp 43-48
Summary of ANOVA/ AN COVA results.

Table 8 pp 49—50
Summary of significant fertilizer effects.

LIST OF FIGURES

Figure l p 51
Percent light availability measured on a plot basis in August 2001 under full canopy
cover.

Figure 2 p 52
Effect of fertilizer on soil resource levels one month after fertilizer application.

Figure 3 p 53
Effect of fertilizer on soil pH approximately one month after May 2001 fertilizer
application.

Figure 4 p 54
Comparison of % soil volumetric water (0- 30cm) between 20- 21 June and 19 July 2001,
estimated with TDR.

Figure 5 pp 55-56
Proportion of seedlings surviving between beginning of June and beginning of September
2001 on a plot basis.

Figure 6 p 57
Species and site differences in relative and absolute grth in high and low light.

Figure 7 p 58
Absolute biomass at end of the growing season.

Figure 8 p 59
Root mass ratios (RMR) graphed as log final root mass versus log final (root + stem)
mass.

Figure 9 p 60
Area of fine roots (<2mm diameter).

Figure 10 p 61
Log transformed fine root surface area versus whole plant mass.

Figure 11 p 62
Specific root area shown as total root surface area versus root biomass (log transformed).

Figure 12 p 63
Approximate maximum rooting depth.

vi

Figure 13 p 64
Rooting depth versus whole plant mass (log transformed).

Figure 14 p 65
Comparison of TDR (0-30cm) measurements among 3 years.

Figure 15 p 66
Trade-off of high soil resource RGR and low soil resource survival.

Figure 16 p 67

Estimated species biomass accumulation (roots + stem) on outwash under high light to
year 2014, assuming constant RGR.

vii

Introduction

The distribution of plant species across soil resource environments has long been
an interest in ecology. Past research has focused on both describing and experimentally
testing edaphic characteristics that are related to discontinuous commmrities (Tansley
1917; Billings 1950; Kruckeberg 1954; Gankin and Major 1964; Goldberg 1985). Other
studies have further contributed to this area of research by investigating how species
traits relate to distributions across soil resource levels (Grime 1977; Tilrnan 1985; Aerts
and Berendse 1989; McGraw and Chapin 1989; Schlesinger et a1. 1989).

Two hypotheses have been proposed to explain variation in species composition
with soil resources: 1) species associated with favorable resource sites are excluded from
poor sites by physiological intolerance of poor resource conditions and 2) species
associated with poor resource sites are competitively excluded from favorable resource
sites by species adapted to these environments (Gankin and Major 1964, Goldberg 1985).
In combination these hypotheses may represent a necessary a trade-off between species
competitiveness on favorable resource sites and tolerance of poor resource sites (Grime
197 7). Species associated with poor sites are physiologically tolerant but they may be
constrained in their ability to grow rapidly in response to more favorable resource
conditions (Grime and Hunt 1975; Chapin et al. 1993). Conversely, species associated
with favorable resource sites are relatively intolerant of poor resource conditions but are
able to grow rapidly in response to favorable environments (Mahmoud and Grime 1976;
McGraw and Chapin 1989).

A trade-off between competitive ability and tolerance limits can be evaluated

through at least two relationships: species rank reversal of relative growth rates (RGR)

between poor and favorable resource and survival in poor versus competitive ability in
favorable resource conditions. Rank reversals of RGR consistent with field distributions
along resource gradients have been reported by several studies (McGraw and Chapin
1989; Lantharn 1992; Lusk et a1. 1997), although none of the studies involving multiple
species have reported an overall negative correlation between high and low resource
RGR, the rank reversals are only for a few species. Rank reversals may not commonly be
reported because studies are limited by their time span. For example, species fi'om high
resource environments may have traits that initially allow greater resource access,
resulting in high initial RGRs in poor resource environments. However, in the longer
term these species may experience cumulative stress, show decreased RGR, and not be
able to persist. In comparison, species native to poor resource sites may have traits
allowing them to deal with the stress of a low resource environment and maintain
consistent RGR. Conversely, overall correlation between high and low resource RGR
may not be reported because other factors are more important in explaining species
distributions than RGR rank reversals. Studies which only investigate RGR often dismiss
individuals that do not survive from their data sets, leading to overestimates of RGR. The
importance of considering survival, over RGR, has been demonstrated by studies on
shade tolerance (Kobe et a1. 1996; Walters and Reich 1996). These studies show that
small differences in growth may have a large effect on survival. It follows that while
species RGR may be a good indicator of competitive ability in a high resource
environment (Grime and Hunt 1975), species survival may be a better indicator of species
tolerance in a low resource environment. Therefore, the competition versus tolerance

trade-off would be supported by a negative correlation between survival in poor resource

conditions versus RGR in high resource conditions.

In systems where composition is related to high (favorable) and low (poor) soil
resource availability, traits which may underpin the proposed trade-off include
adaptations of resource use efficiency, resource access, biomass allocation and
differences in initial size. Species tolerance of low resource environments may be related
to high resource use efficiencies (V itousek 1982; Aerts and Berendse 1989; Schlesinger
et al. 1989) and large initial size, related to large seed size (Stock et al 1990; Milberg et
a1. 1998; Khurana and Singh 2000). However, these traits may incur costs, such as low
rates of nutrient uptake and growth, that limit their ability to respond to favorable
resource environments (Chapin et a1. 1993). In contrast, species associated with high
resource sites may have an adaptive advantage through plasticity to increase their access,
and therefore absorption, of soil resources in high resource environments, translating into
high potential RGR (McGraw and Chapin 1989; Aerts and Chapin 2000). Below-ground
biomass, the ratio of root mass to whole plant mass (root mass ratio (RMR)), the ratio
between root length or area and root biomass (specific root length (SRL) and specific root
area (SRA)) and fine root area have all been proposed as traits related to soil resource
uptake (Lambers et a1. 1998).

Studies investigating the relationship between RMR and whole plant performance
and species distributions are difficult to generalize, while studies considering SRL and
SRA are more consistent with one another. Chapin (1980) and Lambers and Poorter
(1992) report that species associated with high resource environments have lower RMR
than species associated with poor resource environments, when both are grown under

favorable conditions. In addition, species associated with high resource sites will

increase their biomass allocation to roots when in poor resource conditions (Lambers and
Poorter 1992). However, Gunatilleke et a1. (1997) did not find a general trend between
seedling RMR in high nutrient treatments and adult species distributions for 8 Shorea
species in Sri Lanka. In addition, their results did not support the generalizations that
species from nutrient rich sites are more plastic and degree of plasticity is related to
growth response. Huante et a1. (1995) found that all but four of 34 tropical woody
species decreased RMR in response to increased nitrogen, but the relationship of RMR
with RGR was non-significant. Wright and Westoby (1999), a study of temperate
Australian tree seedlings, did find a correlation between RMR and RGR but,
interestingly, it was positive. This study also investigated SRL and found the
hypothesized positive correlation with RGR. These results are similar to those of Reich
et a1. (1998) for boreal tree seedlings and Comas et a1. (2002) for North American
temperate tree seedlings, suggesting that variations in structure have a stronger, more
consistent, influence on whole plant performance than biomass partitioning. Aerts et a1.
(1992) also suggest that allocation ratios may be poor indicators of resource capture and
illustrate, through an experiment with Carex, that it is necessary to consider absolute root
biomass as well.

In order to understand the effect of below-ground resources on species-specific
growth and survival, interactions of below-ground resources and light must be
considered. Light is the most important resource limiting plant growth and survival.
This has been demonstrated consistently for tree seedlings and saplings (Canharn 1988;
Pacala et al. 1994; Kobe et a1. 1995; Walters and Reich 1996). It follows that species-

specific growth rates may be limited by nutrients when in moderate to high light levels

and may have little to no response to nutrient addition in low light environments. This
has been reported for a number of species (Steinbauer 1932; Phares 1971; Lantham 1992;
Grubb et al. 1996; Meziane and Shipley 1999, Walters and Reich 2000). However, other
studies investigating shade tolerant species have reported growth responses in deep shade
to nutrient additions (Peace and Grubb 1982) and to natural variation in nutrient
availability (Walters and Reich 1997). Interestingly, a few studies investigating multiple
species have reported increased mortality in low light! high nutrient environments,
compared to low fertility environments (Hutchinson 1967; Grubb et al. 1996). Increased
water availability has been reported to have the opposite effect, or decrease mortality, for
some species. Casperson and Kobe (2001) report decreased mortality with increasing
water availability in low light environments for saplings of several deciduous hardwood
species common to non-xeric sites, similar results are reported for high light
environments. Sack and Grubb (2002) report parallel results for RGR. Species found on
intermediate and moist sites had depressed RGR in infrequently watered treatments
compared to frequently watered treatments. RGR was reduced by approximately the
same proportion in high and low light conditions, showing that drought does not have a
greater effect in deep shade than in high irradiance conditions. This result is consistent
with that of Holrngren (2000) for tulip poplar.

Previous research provides motivation to investigate a possible trade-off between
species competitiveness on favorable resource environments and tolerance in poor
resource environments, and the morphological traits potentially associated with this
trade-off. Here, we do this with a field experiment of seedlings of five tree species

representing three distinct species communities. Seedlings were grown with and without

fertilizer additions, to explicitly test for nitrogen and! or calcium limitation, on three sites
representing natural gradients of water and nutrient availability and two light categories.

Our study was conducted in Marristee National Forest (MNF) in northern lower
Michigan. Species compositions across the landscape are closely associated with three
glacial landforms. These landforms show strong differences in soil texture (Host et al.
1988), nitrogen (Zak et. al 1989), calcium and late growing season water availability
(Kobe and Schreeg, unpublished). Exchangeable calcium levels of MNF outwash sites
are similar to levels at Hubbard Brook (Likens et al. 1998), a northeastern temperate
forest hypothesized to be calcium limited at least for the regeneration of some tree
species (Kobe et al. 2002). The gradient of soil resource availabilities provides an
excellent opportunity for using a reciprocal transplant experiment to test if species
adaptations confer an ability to succeed on their native sites. In addition, the hypothesis
that soil calcium and nitrogen availability are mediating species distributions is especially
relevant from the perspective of atmospheric deposition of anthropogenic pollutants.
Decreases in exchangeable soil calcium concentrations have been correlated with
increases in atmospheric deposition of acids (Likens et a1. 1996), which may be
accompanied by increases in soil nitrogen (MacDonald et al. 1992).

We asked three main questions focused on understanding the whole-plant traits
underlying landscape variation in forest composition.
1) Is there a trade-off between species competitiveness on rich sites and tolerance on poor
sites?
2) Are species differences in seedling growth, survival and morphology consistent with

the association between species composition and landform?

3) Are nutrients (nitrogen and/ or calcium) limiting for rich species on the poor and

moderate fertility sites?

Methods

Study sites Three study sites, representing each of the three main landforms (outwash,
ice-contact and moraine) in glaciated landscapes, were located in Manistee National
Forest, Wexford and Manistee counties, in the northern lower peninsula of Michigan.
End moraines are a result of accumulated debris at the leading edge of a stagnant glacier,
resulting in unsorted material often associated with relatively high availability of soil
moisture and nutrients. Ice-contact formations originate from material that was water
deposited on, or in, glacial ice, thus leading to sorting of particles and generally lower
nutrient and moisture availability than moraines. Outwash landforms are a result of
coarse material deposition from fast moving meltwaters created from glacial recession
(Ritter et. al. 197 8), and thus tend to be the most sediment sorted, well-drained, and
nutrient poor of the three major landforms. Across landforms in our general study area,
but on different sites, Zak et al. (1989) measured N mineralization rates of 78 kg ha'1 yr”1
in outwash sites to 122 kg ha'1 yr'1 in rich moraines and exchangeable calcium differed
by approximately 1000 ug Ca2+ g’1 soil between outwash and moraine, with ice-contact
levels falling between the extremes (Table 1, from Kobe and Schreeg unpublished data).
Host et a1. (1988) reported larger soil particle diameters, and thus lower soil water
retention capacity for outwash than for ice-contact or moraines. In addition ground flora
and overstory composition are strongly related to the glacial geomorphology with
mesophytes common on moraines, xerophytes more common on outwash, and ice contact
species assemblages intermediate (Host et al. 1988; Host and Pregitzer 1992).
Experimental Setup This experiment included three sites (outwash, ice-contact, moraine),

four fertilizer treatments (control, calcium, nitrogen, calcium and nitrogen) and tree

seedlings of five species. We selected species to include a wide range of landform
affinities, including: black oak (Quercus velutina), most prevalent on outwash and in
lower density on the ice-contact; red oak (Q. alba), most prevalent on ice-contact but also
occurring on moraines and outwash; sugar maple (Acer saccharum) and white ash
(Fraxinus americana), which are largely restricted to moraines and black cherry (Prunus
serotina), which can be found across the landscape but is most prevalent on moraines.
Fenced plots of each fertilizer level were located within each site. Each fertilizer
plot was composed of 3 harvest plots; 1 plot was harvested in May 2001 for initial
biomass and root morphology data and 2 plots were harvested in September 2001 for
final biomass and root data. Each harvest plot was planted with 5 individuals of sugar
maple, white ash, black cherry and black oak, and 2 individuals of red oak (fewer red oak
were available because of low germination). Individuals were randomly placed within
each harvest plot at a spacing of 20 cm. Spacing between harvest plots and from plants to
fencing was approximately 40 cm. Plots were located across the range of extant light
levels at each site.
Seedling establishment and transplanting Seeds for the five species were purchased from
USDA Hardiness Zone 4 or 5 sources. Black oak, red oak and sugar maple were planted
in containers 12 cm deep x 5.5 cm in diameter. Containers used for white ash were 8 cm
deep x 4 cm diameter. Planted black cherry had high mortality so first year gerrninants
collected from MNF were planted directly into field plots. Greenhouse planting began in
early May and progressed as seed continued to germinate. White ash had low
germination success; therefore, newly germinated seedlings were collected, planted in

containers and treated the same as seedlings of other species. Seedlings were grown in a

greenhouse and lathe house and watered regularly with deionized water. Due to crusting
and drainage problems with the soil medium, collected from outwash sites in
Roscommon County, MI, seedlings took along time to acquire adequate biomass for
transplanting to the field. Therefore, deionized water was supplemented with a fertilizer
solution for watering (Greencare 19-4—23-2 Ca) was used. To counteract high soil pH,
seedlings were watered with a solution of sulfuric acid (approximately pH 2), avoiding
contact with foliar tissue. In mid July, seedlings were transported to a location central to
the field sites. The seedlings were kept outside under low - moderate light conditions and
watered with tap water as needed before planting.

Field plots were cleared of vegetation and fenced (1.5 m height) to exclude deer
and deter rodents. After planting, plots were weeded regularly to maintain consistent light
environments. Transplanting to field plots took place between 20 July and 2 August
2000, with 4 plots added to the moraine site on 25 August 2000. Available soil water
was very low in late August; thus, we watered the plots while planting to enhance
transplant success. To account for transplant shock, individuals that died between August
2000 and May 2001 were not included in the study.

Fertilizer was applied in late August 2000 and again in mid May 2001. Fertilizer
was added as calcium sulfate (CaSO4*2HZO) and ammonium sulfate ((NH4)ZSO4).
Application rates were based on the difference between outwash and moraine
concentrations. These differences are 1,000 ug Ca g'1 dry soil, or 500 kg Ca ha'l to 5cm
depth, and 45 kg ha'1 yr'l; translating to application rates of 215 g of CaSO4*2HzO and
21 g of (NH4)2SOZ/ m2. In August 2000, fertilizer addition equaled twice the difference

between the outwash and moraine. The application in May 2001 was at a rate of one

10

times the difference. Sulfate was chosen as the counter ion to the nutrient of interest
because 1) it was unlikely to be limiting in these sites, 2) it was unlikely to be toxic to the
plants in the concentrations applied, 3) both calcium and nitrogen fertilizer are available
in sulfate form, meaning we could maintain the same counter ion between the levels of
fertilizer treatment and 4) it would not lead to confounding increases in pH, in contrast to
other widely applied calcium counter ions (e. g. CO3).
Whole plant harvests, root scanning and biomass measurements A set of plants was
harvested at the beginning (1 1-16 May 2001) and end of the growing season (5-13
September 2001). We carefully excavated entire harvest plots of seedlings, excavating
roots to their maximum depths. The sandy soils minimized loss of fine roots.
Immediately after removal fiom the soil, harvested plants were stored in zip lock bags
and kept in a cooler or refiigerator. During the harvests, field notes were made on
individuals. Although care was taken during transplanting not to J -root the seedlings,
curved roots were common for the oaks.

Harvested seedlings were transported to the lab and stored in a cooler (2°C).
Small groups of seedlings were removed from the cooler, washed using squirt bottles of
deionized water, patted dry with Kimwipes, returned to clean zip lock bags and retumed
to the cooler. Washing was completed within 2 weeks of field harvesting. Harvested
seedlings were divided into sections (root, stem (including petioles) and leaves), placed in
coin envelopes, dried at 65 0C for at least 3 days, and then stored in a dessicator. A
balance accurate to 104 g was used to measure biomass (Model M 310, Denver
Instrument Company, CO).

Prior to drying, a subsarnple of seedlings was scanned and analyzed for root

11

surface area and length using WinRhizo (version 3.10, Regent Instruments Inc., Blaine,
Quebec, Canada). Large roots (approx. >2 mm diameter) had more surface texture and
discoloration than smaller roots. Therefore, large roots had to be analyzed using a filter
and a lower threshold to avoid segmenting the root into artificial sections. Large roots
were separated from smaller roots and arranged on a scanner so the size classes did not
overlap. Roots were scanned at a resolution of 300 dpi; large and small diameter roots
were then analyzed independently. Analysis for large diameter roots used a bark filter
and a threshold of 100, except white ash for which a threshold of 120 was selected
because it had a lighter color. The automatic threshold was selected for the analysis of
the smaller diameter roots. To estimate individual seedling rooting depth, we measured
the length of the longest root of plants harvested in September, excluding those
individuals whose field notes indicated that the longest root was lateral.
Resource measurements Light availability, approximated as percent canopy openness,
was measured during full canopy leaf-out (22-24 August) using a LAI 2000 (LiCor,
Nebraska) in remote mode. One sensor was set in an open field; the other was used to
collect below canopy data. To avoid interference from direct sunlight, measurements
were taken at twilight. Reported data on percent canopy openness were collected from
one of the two final harvest plots within each fenced fertilizer treatment; three
measurements spanning the harvest plot were averaged. A subsarnple showed that
percent light availability among harvest plots within the same fenced plot differed by less
than 0.5%, supporting the extrapolation of harvest plot light data to the entire fenced plot.
To characterize baseline soil resource availability and test the effectiveness of

fertilizer treatments, we collected soil cores from control and fertilized plots one month

12

after fertilizing. In June, five soil cores were taken within the final harvest sub-plot of
each fenced plot and bulked at depths of 0-10 cm and 10-20 cm.

An aerobic lab incubation was used to determine potential net nitrogen
mineralization and nitrification rates. Duplicate 10 g air dried samples were weighed for
initial and incubated extracts, and deionized water was added so soil was at
approximately 60% field capacity. Incubated samples were kept in the dark at 25 °C.
Initial extracts were done after 4 days of wetting to allow time for microbial populations
to re-equilibrate; samples for final extracts were incubated for 30 additional days.
Deionized water was added when necessary to maintain the samples at approximately
60% field capacity. Inorganic nitrogen was extracted using 50 ml of 2 M KCl and a half
hour of mechanical shaking. Solutions were filtered using Whatrrran #2 filter paper that
had been rinsed three times with KCl solution. An Alpkem Series 500 autoanalyzer was
used to determine the concentration of NHH-N and N03'-N in the extract. Potential
nitrogen mineralization was calculated as the final minus initial concentrations of the sum
of NHi-N plus NOg'-N. Potential net nitrification was determined similarly by
subtracting final minus initial concentrations of NOg'-N.

Exchangeable soil calcium was extracted from 5 g sub-samples with 50 m1 of
neutral 0.5 M ammonium acetate solution (Soil and Plant Analysis Council Inc. 2000).
Concentrations were determined using a Direct Current Plasma Atomic Emission
Spectrophotometer (SMI Corp).

Time domain reflectometry (TDR) (Environmental Sensors Inc.) was used to
determine soil volumetric water to 30 cm depth. The calibration of the TDR instrument

was checked by comparing TDR to volumetric measurements, calculated from

13

gravimetric and bulk density data. Regression using 5 points between 2 and 18%
volumetric water in a sandy soil, showed a strong correlation between the two methods
(R2=0.9957; TDR= 0.9383(volumetric) + 0.7948). Gravimetric water content was
determined for the July 2001 0-10 cm samples by using approximately 10 g of field moist
soil. Samples were weighed before and after oven drying (105 °C) to determine water
content. In order to investigate unintended fertilizer effects, soil pH was determined for
0-10 cm and 10-20 cm depth June cores using 5.00 g air dried soil in 5 ml deionized
water.

Calculations and Data analysis Percent canopy openness did not have the same ranges
among sites (Table 2). To avoid confounding site and % canopy openness, we identified a
range of high light levels (14-27% light availability) common to the outwash and moraine
and a range of low light levels (3-10% light availability) common to the ice-contact and
moraine (Figure 1). Using these groups, we analyzed the data as comparisons between
outwash versus moraine at high light and ice-contact versus moraine at low light.

For all analyses, if fertilizer effects were not significant within a light by site
group, fertilizer treatments were pooled. If effects were significant, only control plots
were used for further species and site comparisons.

Differences in survival among species-within sites and among sites-within species
were analyzed with Likelihood Ratio Chi Square tests, also known as G2 tests, using JMP
software (SAS Institute). If species differed significantly within a site, species ranks
were investigated by comparing the observed survival versus that expected if mortality
within a site were randomly distributed among species.

Growth, biomass and approximate rooting depth data were analyzed using a split

14

 

plot model with generalized randomized complete blocks (GRBD) in SAS. Site was the
blocking variable, and species were nested within fertilizer treatments. Significant
interactions were sliced in order to investigate effects within a given treatment level.
Slicing is a quick way to do many contrasts at the same time. For each level of a
treatment in an interaction, slicing compares among all the other levels of other
treatments in the interaction. If fertilizer‘species interactions were significant and site
interactions were not, the model was still sliced by site*fertilizer*species. This was
justified because we were interested in fertilizer effects on each site. For significant
contrasts that investigated treatment effects with more than two levels, significant
differences among individual treatments were tested with Tukey-Kramer multiple
comparison tests (SYSTAT). Whole plant biomass, root biomass and maximum rooting
depth were log transformed to more closely approximate normal distributions. Relative
growth rate (RGR) was calculated as ln(mean mass for species at final harvest) - ln(mean
mass for species at initial harvest). Absolute growth rate was calculated as mean mass
for species at final harvest — mean mass for species at initial harvest. To avoid biasing
comparisons among sites and species, RGR, absolute growth and whole plant biomass
sums do not include leaves because leaves of some species by site combinations,
especially white ash on outwash, were shed before the end of the growing season due to a
severe drought.

Root surface area data were analyzed similarly to biomass data with the exception
that high light moraine seedlings were not included in the subsarnple of individuals that
were analyzed for root surface area. Therefore, high light analysis is for outwash only

using a split-plot model with the whole plot factor (fertilizer) in randomized complete

15

blocks. Fine root area was log transformed to more closely approximate a normal
distribution.

We analyzed the effects of treatments on morphological characteristics
independent of mass to rrrinimize treatment effects on morphology mediated through
mass. Thus we analyzed specific root area as total root area as a function of root mass
(covariate) and the experimental treatments. Root mass ratio was analyzed allometrically
as root mass as a function of whole plant mass and the experimental treatments.
Treatment differences in maximum rooting depth and fine root area were analyzed with
whole plant mass as a covariate. All values were log transformed. Models were
evaluated for covariate interactions and, if these we not significant (P> 0.05), we
proceeded with AN COVAs. If interactions with the covariate plant size were significant,
single species were removed and models were re-run as an attempt to eliminate covariate
interactions. For AN COVAs, significant main effects and interactions were investigated
by slicing, and least squared means and standard errors (a=0.05 level) were used for

multiple comparisons.

l6

Results
Soil resource availability
Soil nutrients Soil samples collected in mid-June, one month after fertilizer application,
show nitrogen fertilizer treatments were effective in increasing soil nitrogen availability
on outwash and ice-contact, although this increase was only detected from 0-10 cm in
nitrogen-amended outwash plots (Table 3, Figure 2a). At 10-20 cm depth, where fewer
of the fine roots occur (personal observation), nitrogen fertilizer effects were found for
nitrogen and nitrogen + calcium amended plots on outwash and ice contact (Table 3,
Figure 2b). Furthermore, values from control plots are consistent with expected
differences in naturally available nitrogen among sites, with moraine sites having two
fold higher extractable nitrogen pools than the outwash site and the ice contact site was
intermediate. Mineralization rates were similar among fertilizer treatments within a site
and differences among sites were consistent with expectations (data not shown).
Extractable calcium was greater for calcium and calcium x nitrogen fertilizer
treatments than for control treatments on outwash and ice-contact sites for mid-June 0-10
cm cores (Fig. 2c), although low sample sizes precluded statistical tests. Soil from
nitrogen amended plots was not analyzed. Calcium treatments on moraine were either
similar or lower than the control treatment, but all treatments fell within the broad range
of exchangeable calcium values that occur naturally on these landforms (Kobe and
Schreeg, unpublished). In addition, site differences among exchangeable calcium levels
in control plots were consistent with expectations and with data from other moraine, ice-
contact and outwash sites.

Soil pH Fertilizer decreased soil prm on outwash and ice-contact but not moraine for

17

both 0-10 cm depth and 10-20 cm depth mid-June samples. On outwash, nitrogen plots
had the lowest, most acidic pH values, for both 0-10 cm and 10-20 cm depth samples
(Table 3, Fig. 3). On ice-contact, for 0-10 cm depth, soil pH 11 was greater in control and
calcium treatments than in nitrogen treatments, whereas for 10-20 cm depth pH of the
control treatment was greater than calcium, calcium x nitrogen and nitrogen treatments
(Table 3). However, although fertilizer generally significantly decreased pH at both
depths, the magnitude of these changes was relatively small with fertilizer decreasing pH
by less than 0.4 pH units.

Volumetric soil water content In mid-June TDR values ranged from 7% volumetric water
content on outwash to approximately 12% on moraine (Table 4, Figure 4). A severe
midsummer drought resulted in extremely low volumetric soil water from the middle to

the end of the growing season (Table 4, Figure 4).

Plant survival, growth and morphology

Survival Fertilizer effects on survival could not be tested with contingency tables
because of low cell counts. However, raw data appear to be very sirrrilar among fertilizer
treatments (Figure 5), providing justification for pooling fertilizer treatments. In high
light on the outwash site, species more common on low soil resource sites (black and red
oak) have higher survival than species common on high soil resource sites (black cherry
and white ash) (G-test, p<0.0001) (Table 5). We did not detect inter-specific differences
in survival on moraine (a= 0.1), with all species showing 95- 100% survival. On
outwash, black oak had higher observed survival than expected assuming that mortality

within a site is randomly distributed among species (i.e. black oak had 65 surviving

l8

 

individuals versus 52.9 expected by random chance). In contrast, white ash had lower
than expected survival (37 surviving individuals were observed while 49 were expected).
Intra-specific site comparisons showed black cherry, white ash and sugar maple had
greater survival on moraine than outwash (G2 test, p= 0.0039, 0.0005, 0.0396), but black
oak and red oak showed no differences between sites in high light (or>0.l in both cases).
Thus, species associated with moraines survived better on moraines than on outwash
while poor species had high survival on both sites.

In low light, species did not differ in survival within ice-contact or moraine (Table
5). Between sites, black cherry and red oak had higher survival (G-test, p=0.0377 and
0.0319) and sugar maple had lower survival (G-test, p= 0.0855) on moraine than ice-
contact. It is interesting to note that red oak and sugar maple show lower understory
survival in the site in which they are most common.
Relative growth rates Fertilizer effects on RGR, calculated as ln(mean mass for species at
final harvest) -— ln(mean mass for species at initial harvest), were not significant in either
high or low light, although it should be noted that high light moraine had n=1 for each
fertilizer treatment. In high light species did not differ in RGR on the outwash site, but
they did on the moraine site (Table 5, Figure 6a), where white ash and black cherry had
higher RGR than black oak and sugar maple. All species, except black oak, had higher
RGR on moraine than outwash (Table 6). These results show that the poor site species,
black oak, does not respond strongly to increases in soil resources, or conversely, that
species associated with rich sites have much reduced growth rates in poor compared to
rich environments. Furthermore, in comparison to poor-site species, species associated

with rich sites, with the exception of sugar maple, have higher RGR in high resource

l9

environments and similar RGR in a low resource environments.

Under low light, there were significant differences among species in RGR within
both sites and rank trends among species were similar (Table 5, Figure 6c), with moraine
species black cherry, white ash and sugar maple having higher RGR than the oaks on
both the moraine and the ice contact site. In addition, there were no significant
differences in low-light RGR between ice-contact and moraine for any of the species
(Table 6), but sugar maple does show a non-significant decline in RGR on ice-contact
(Figure 6c). These results suggest that, low light availability may have been a stronger
constraint to RGR than the variation in soil resources between the ice-contact and
moraine sites. It is possible that under higher light conditions soil resources on ice-
contact would be limiting for moraine species.

Absolute growth and total mass We also examined absolute growth (calculated as mean
mass for species at final harvest — mean mass for species at initial harvest) and total mass,
since these metrics, for seedlings of the same age, are likely as important as RGR as
determinants of competitive ability. Under high light species ranks of absolute growth
were similar for moraine and outwash, with species of larger initial seedling size showing
higher absolute growth. Red oak had the largest increase in biomass, followed by black
oak, white ash, sugar maple and black cherry (Table 5, Figure 6c). White ash, the only
species showing significant site differences, had a greater change in biomass on moraine
versus outwash (Table 6), but all species showed trends of higher absolute growth under
high light at the moraine site. In contrast to high light, under low light all species had
similar absolute growth within both ice-contact and moraine, with the exception of red

oak having a greater change in biomass than black cherry on moraine (Table 5, Figure

20

 

 

6d). Fertilizer effects were not significant in either light category (Table 7, or= 0.1).

In high light, sugar maple, white ash and black cherry had significantly higher
final biomass (sum of roots and stems) on moraine than outwash (Table 6, Figure 7a).
Trends in species ranks for control treatments within each site were similar: red oak >
black oak > white ash > sugar maple > black cherry, these final biomass ranks are similar
to those of initial size (data not shown). All inter-specific comparisons of final biomass
within each site were significant with two exceptions, black oak and white ash did not
have different final biomass on moraine and, sugar maple and white ash final biomass
were not different on outwash. Nitrogen fertilizer treatments depressed final biomass for
white ash on moraine and red oak on outwash (Table 8). Under low light, all species
differed in whole plant biomass on both ice-contact and moraine sites, with the exception
of sugar maple and white ash (Table 5, Figure 7b). Under low light, none of the species
showed intraspecific variation among sites in final biomass (Table 6), likely due to light
being more limiting to growth than soil resources. Sugar maple on calcium plots had
greater biomass in comparison to nitrogen fertilized plots on moraine and ice—contact
(Table 8).

Root biomass In high light, sugar maple and white ash had greater root biomass on
moraine than outwash (Table 6, Figure 7c). Species ranks in root biomass were similar
for all sites in both high and low light and followed the same rankings as whole plant
biomass: red oak > black oak > white ash > sugar maple > black cherry (Table 5). Also
similar to the results for whole plant biomass, there were no site differences in species
final root biomass under low light. In high light, nitrogen fertilizer was found to decrease

root biomass for white ash on moraine and outwash and red oak on outwash, in

21

comparison to other treatments. Similar results were found in low light for sugar maple
on moraine and ice-contact (Table 8).
Root mass ratio Again, whole plant biomass in this study is the sum of roots and stem
mass. Leaves were not included due to early leaf senescence for some site and species
combinations. On outwash, 84% of the white ash individuals, and 28% of sugar maple
alive at the September harvest had shed their leaves. In comparison, no sugar maple and
only 3% of white ash in high light on moraine had early leaf senescence.

Because leaves are not included here, RMR in this study is used to evaluate site
and species differences in root mass when normalized by whole plant mass (stem +
roots), rather than assessing the relative investment to below-ground versus above-ground
resource acquisition. In high light, ANCOVA allometric analysis violated the
homogeneity of slopes assumption of AN COVA (Table 7), which was remedied by
removing black oak from the analysis. Based on this reduced data set, sugar maple root
mass normalized by plant mass effects (i.e. RMR with the effects of plant mass removed)
was greater on outwash than moraine (Table 6). This is the only species that showed an
increased investment in below-ground biomass on the poor versus rich resource site. All
inter-specific comparisons of least squared means and standard errors (or=0.05 level) of
root mass normalized to whole plant mass were significant on outwash with: SM > WA >
BC > R0. On the moraine site, sugar maple and white ash were also found to have larger
root mass, normalized by whole plant mass, than red oak (Table 5, Figure 8). Trends in
RMR are the inverse of those found for absolute root mass suggesting the large absolute
investment in root mass of oaks is a result of their larger sizes and does not originate

fi'om differences in allocation. No fertilizer effects were detected (Table 7).

22

 

ANCOVA of low light data was not possible; there were significant covariate
interactions in the model with all species and in models with single species removed.
Data are shown in Figure 8.

Absolutefine root surface area Roots were not scanned for high light moraine, therefore
site was not included in the high light analysis. For high light control treatments on
outwash, black cherry had significantly lower fine root (<2 mm diameter) area than all
other species (Table 5, Figure 9a).

In low light, sugar maple control treatments had greater fine root surface area on
moraine than ice-contact (Table 6, Figure 9b). White ash shows the same non-significant
trend while red oak shows the Opposite trend (Figure 9b). For inter-specific comparisons
of control plots, all species had greater fine root area than black cherry on both ice-
contact and moraine. In addition, red oak had greater fine root area than white ash and
black oak on ice-contact and sugar maple had greater fine root area than black oak on
moraine (Table 5). Fertilizer effects were significant for red oak and sugar maple on
moraine and white ash on ice-contact (Table 8).

Fine root sufl’ace normalized by whole plant mass Fine root surface area was
significantly correlated with whole plant mass (Table 7). Thus, to examine species and
site effects on root surface area independent of there effects of plant mass, treatment
effects on root surface area were analyzed using whole plant mass as a covariate. In high
light on outwash, moraine species sugar maple and white ash had greater fine root surface
area per whole plant mass than poorer resource site species red oak and black oak. Sugar
maple also had greater normalized fine root investment than black cherry (Table 5, Figure

10a). This suggests sugar maple and white ash have greater ability to access below-

23

ground resources per unit whole plant mass than the other study species. For white ash,
fine root surface area normalized to whole plant mass were significantly lower under
nitrogen amendments in comparison to other treatments (Table 8).

In low light, with all species in the model, species x plant mass interactions on

root surface area were significant . With black oak removed interactions were no longer
significant, thus ANCOVA was appropriate. Species ranks in root surface area
normalized by plant mass were the same on ice-contact and moraine: sugar maple and
white ash > black cherry > red oak (Table 5). Although not analyzed, black oak appears
to rank near red oak (Figure 10 b,c). Similar to high light outwash results, these rankings
suggests sugar maple and white ash have a greater ability to access below-ground
resources per unit whole plant mass. Intra-specific fine root area standardized by plant
mass was not found to differ between sites for any species (Table 6). A separate analysis
using only black oak showed significant covariate interactions and site and fertilizer
effects could not be tested for this species.
Specific root area In high light, unfertilized plots at the outwash site, root surface area
per unit of root biomass (specific root area (SRA) was greater for white ash and sugar
maple than for black oak, red oak or black cherry (Table 5, Figure 11a). Results for fine
root area per whole plant mass were similar and both suggest these moraine species have
traits related to high resource capture. Fertilizer effects on SRA were only found for
white ash on outwash where SRA in plots amended with only nitrogen had significantly
lower SRA than control, calcium or calcium x nitrogen amended plots (Table 8).

In low light, black cherry, sugar maple and white ash had greater SRA on moraine

than ice-contact (Table 6), suggesting that these species can increase their ability to

24

acquire below-ground resources when favorable. Neither of the oaks differed between
site. On moraine, decreasing ranks of SRA were: white ash > sugar maple and black
cherry > red and black oak (Table 5, Figure 11c). White ash, sugar maple and black
cherry also had the highest RGR of any species on moraine in low light, suggesting that
high SRA contributed to high RGR under high soil resource conditions. On ice-contact,
white ash and sugar maple SRAs were greater than red oak, black oak and black cherry.
Red oak SRA was also greater than black cherry (Table 5, Figure 11b).

Maximum rooting depth Black cherry and white ash had greater approximate maximum
rooting depths on moraine compared to outwash in high light (Table 6, Figure 12a).
Within both sites, all species had greater maximum rooting depths than black cherry and,
the rooting depth of red oak was greater than that of white ash. In addition, black oak and
sugar maple maximum root depths were greater than white ash on outwash (Table 5).

In low light, non-fertilized plots, species rankings for maximum rooting depth
were the same for ice-contact and moraine sites: all species were greater than black
cherry (Table 5, Figure 12b). For white ash on ice-contact, control and calcium
treatments had greater rooting depth than calcium x nitrogen treatments (Table 8).
Rooting depth normalized to whole plant mass Approximate maximrurr rooting depth
normalized to whole plant mass showed sugar maple had a greater root length/ whole
plant mass on outwash than moraine (Table 6). This plasticity could enhance survival on
outwash sites, because outwash has lower soil volumetric water than moraine (Table 4).
In addition, on outwash, sugar maple had greater maximum rooting depth standardized by
whole plant mass than all other species, perhaps explaining its high survival on outwash

relative to white ash. Surprisingly, on outwash the normalized rooting depth of red oak

25

 

was significantly lower than all other species and black oak lower than white ash and
black cherry (Table 5, Figure 13), in contrast to absolute rooting depth for which red and
black oak were significantly greater than black cherry and white ash. On the moraine
site, sugar maple, black cherry and white ash had greater normalized rooting depths than
black oak and red oak (Table 5, Figure 13).

Under low light, normalized rooting depths of sugar maple and white ash were
greater than red and black oak in both moraine and ice-contact. On ice-contact,
normalized rooting depth of black cherry was greater than red and black oak (Table 5,
Figure 13). Nitrogen fertilizer depressed normalized rooting depths for black cherry and

white ash on ice-contact (Table 8).

26

Discussion
Growth and survival

We proposed two hypotheses to explain variation in species composition with soil
resources: 1) species associated with high resource sites are excluded from poor sites due
to physiological intolerance at low resource conditions and 2) species associated with
poor resource sites are competitively inferior to rich site species when they are growing
on high resource sites. In other words, there is a trade-off between species
competitiveness on favorable resource sites and tolerance of poor resource sites. We
outlined two relationships that could demonstrate this trade-off: RGR rank reversal and
negative correlation between survival in poor versus growth in favorable resource
conditions.

It appears that the differences in MNF outwash and moraine composition are
influenced by physiological intolerance of moraine species on outwash and higher growth
rates of the rich site species on moraine. We found a negative correlation between
survival in poor versus growth in favorable resource conditions (Figure 15a). This trend
is stronger when our one year survival results are extrapolated over 5 years (Figure 15b),
a reasonable time fi'ame for investigating the effects of survival on forest dynamics.
Assuming a constant survival probability, after five years only 6% of the original cohort
of white ash, 39% of black cherry and 44% of sugar maple would have survived on
outwash, compared to 70% of red oak and 73% of black oak. Lower RGR for the
moraine-affiliated species on the outwash suggests that 5 year survivorship is likely to be
lower than estimated as suggested by strong positive relationships between survivorship

and growth for these species (Kobe et al. 1995). In addition, white ash and sugar maple

27

experienced early leaf senescence on outwash (84% and 29% of the individuals,
respectively), suggesting that cumulative stress also is likely to negatively influence
survival probabilities.

It should be noted that sugar maple appears to be an exception in the trend
between survival in poor versus growth in favorable resource conditions (Figure 15a,b),
showing low outwash survival without the trade-off of high moraine RGR. This may be
related to the high shade tolerance of sugar maple and light levels evaluated for the trade-
off. Sugar maple is known to be a very shade tolerant species (Lorimer 1983; Kobe et al.
1995; Walters and Reich 1996) and to show little response to increased light availability.
This is supported in this study where we found sugar maple showed only 7% increase in
RGR, compared to the all other species, which showed RGR increases of 42-71%
between the high and low light categories on the moraine site. If we also consider that
low light levels are more common on moraine, a rich site with high basal areas, compared
to outwash, a poor resource site with low basal areas, it follows that high light RGR on
moraine may not be the best measure of the competitive ability of sugar maple. Rather, if
we evaluate the trade-off using the most common light levels on each site, sugar maple is
found to fit the trend (Figure 150). The data points are not continuous, however species
are found to be grouped according to whether they are more common on the moraine or
outwash.

The trade-off between survival in poor versus growth in favorable resource
conditions appears to be one explanation for differences in community composition
between moraine and outwash sites in MNF. It is important to note that seedlings in this

study were transplanted to the field plots at an age of 2-3 months, circumventing the very

28

early life stage that is more susceptible to biotic (Augspurger 1983) and abiotic stresses.
The observed trade-off may be operating more strongly for newly germinated seedlings,
which have not yet ligrrified. It should also be noted that seed source limitation provides
an additional explanation for differences in community composition. Kobe (unpublished)
shows sugar maple, white ash and black cherry have much lower seed inputs to outwash
sites than species with adults present on the site. Lower seed abundance together with
lower survivorship may largely underlie the rarity of morainal species as adults on
outwash sites.

We did not find a rank reversal of species RGR between sites. White ash and
black cherry had higher RGR on moraine than black oak. Sugar maple and red oak
showed this same, although non-significant, trend. However, interspecific RGR on
outwash were not significantly different and, black cherry, not black oak, had the highest
mean RGR. RGR ranks on outwash may become more consistent with species
distributions over a longer study period, as a result of cumulative stress on the species
less suited for a poor soil resource environment.

Results from the low light comparison of ice-contact versus moraine do not show
the high soil resource grth versus low soil resource survivorship trade-off, likely
because light is more limiting than soil resources. In support of this interpretation, all
species show increased growth under high versus low light on the moraine site (sugar
maple increased RGR by 7%, white ash 42%, black cherry 48%, black oak 66% and red
oak 71%). In addition, light limitation has been shown to depress the effects of low soil
resource availability for some species (Grubb et al. 1996; Meziane and Shipleyl999;

Kobe unpublished). Moreover, unlike outwash, low light environments compose a large

29

 

percentage of closed forest ice-contact and moraine understories (Figure 1). It may be
the rarer high light environments of forest gaps and clearings that are required for
regeneration of most species in ice-contact and moraine sites (Brokaw 1985; Canham

1988; Runkle 1990).

Traits underlying growth and mortality differences

Sugar maple, white ash and black cherry have traits that would be expected to be

 

advantageous in high resource environments. Sugar maple showed plasticity in biomass
allocation (RMR on outwash > moraine, high light; Table 6) and rooting depth per unit
whole plant mass (outwash > moraine, high light). Sugar maple, white ash and black
cherry showed plasticity in SRA (ice-contact < moraine, low light), demonstrating an
ability to increase below-ground resource access with increased resource availability. In
comparison, black oak and red oak SRA did not respond to site.

Within a common resource environment, sugar maple and white ash consistently
demonstrated a greater ability to access resources, when normalized to plant mass, than
red oak and black oak. Sugar maple and white ash had greater rooting depths per unit
whole plant mass, fine root area per unit whole plant mass and SRA within all site/ light
treatments (Table 5 and Figure 10). Within site trends for black cherry were less
consistent. Plasticity for increased soil resource capture on rich sites may translate into
high RGR compared to species which do not have trait plasticity. It follows that within
moraine, rooting depths per unit whole plant mass, fine root area per unit whole plant
mass and SRA may underpin RGR rank. Within outwash, greater resource access does

not appear to positively influence RGR (Table 5). In this lower soil resource

3O

environment absolute biomass and surface area, rather than data which are normalized to
plant size, may be more important for survival.

Survival has been found to be positively affected by plant size in a number of
studies (Hubbell et al. 2001; Schmidt and Zotz 2001; Horvitz and Schemske 2002). In
addition, high absolute root surface would allow an individual to explore a larger volume
of space, a trait that may be advantageous in a generally low, but heterogeneous soil
resource environment. Red and black oak fine root areas were 3 to 5 times greater than
black cherry and 70 to 150% greater than white ash. Black oak fine root area was similar
to that of sugar maple but red oak fine root area was 60% greater than sugar maple. We
found black oak and red oak root + stem biomass and root biomass are 2-5 times greater
than sugar maple or white ash, and 10-20 times greater masses than black cherry on
outwash (Table 6). The importance of initial size is also apparent when biomass
accumulation is extrapolated over time. Although black cherry RGR is slightly greater
than black or red oak, it would take more than 15 years for black cherry to accumulate
more biomass than black oak and more than 25 years for it to surpass red oak on outwash

(Figure 16), if we assume constant RGR over time.

Fertilizer effects

In general, fertilizer treatments increased nitrogen and/or calcium availability on
outwash and ice-contact sites, but the fertilizer treatments did not affect survival or
growth for any tree species. Fertilizer effects were found for other response variables,
such as whole plant mass, root mass, fine root area and approximate rooting depth.

However, few general patterns across species could be detected (Table 8). It is

31

interesting that in all but one case, nitrogen treatments (nitrogen and/ or calcium x
nitrogen) are found to depress the response variables of whole plant mass, root mass, fine
root area and approximate maximum rooting depth. Nitrogen fertilizer was not found to
affect RGR or absolute growth so the detected effect on whole plant mass may not be
biologically significant. The effect on root mass and morphology suggests that increased
nitrogen availability decreases the need for below- ground investment.

The lack of fertilizer effects on growth and survival may not be surprising given
the strong drought the study area experienced during the mid to end of the growing
season of 2001 . In mid-July, volumetric soil water values were 3-4 times lower than
measurements taken at the end of August 2000, more than one month later in the growing
season. July 2001 outwash and ice-contact means are only slightly less than those
recorded at the beginning of September 1999. However, volumetric soil water at the
moraine site was less than 40% of the September 1999 values (Figure 14). The extremely
low values of soil water availability in July are not typical of these forests and would
clearly have an influence on seedling nutrient uptake and growth. It is reasonable to
believe that the growing season for our seedlings was cut short by more than a month.
Therefore, it is not surprising that fertilizer effects were either not significant or difficult
to interpret (Table 8). Although the drought strongly demonstrated the importance of
water across the MNF landscape, the results of this study do not eliminate the possibility
that calcium and nitrogen are important in other years, when soil water availability is

greater.

32

 

Conclusion

Our results demonstrate that differences in composition between the moraine and
outwash could be explained by a trade-off between tolerance on outwash (poor soil
resource conditions) and grth on moraine (rich soil resource conditions) in high light.
This trade-off is suggested by a negative correlation between survival in poor versus
growth in favorable resource conditions. The trend between survival on outwash and
growth on moraine is stronger when our results are scaled up to 5 years. Results for ice-
contact versus moraine suggest increased RGR of sugar maple, white ash and black
cherry on moraine may be underpinned by greater fine root area per whole plant mass
and a plasticity to increase SRA in response to increased soil resource availability. Black
oak and red oak SRA did not respond to site conditions. The high survivorship of black
oak and red oak on outwash may be related to greater absolute whole plant biomass, root
biomass and fine root area than sugar maple, white ash and black cherry, when grown on
outwash. A trade-off between tolerance to ice-contact and growth on moraine was not
supported by our results. This may be related to the low light environments investigated
in this study. Neither calcium nor nitrogen fertilizer treatments were found to effect
growth or survival. However, 2001 experienced a severe drought extending from the mid
to the end of the growing season. Soil water availability was much lower than in

previous years.

33

 

APPENDD(

34

>30 .5 an .XLV

 

anemia @49va a .15 3a @6382 6533a 5.23.2 5.18:2 seas:
$.33 Ea ads 3a 3.2;? see mew afiawmmmawmv Sadat: Asian: assess
Giana $6.952 9.3:? $.53 58 smwhmfimfiv 5.8va 5.1892 5330

52 w: M 8 “Mach.” 88 Baa... 82 E...
:8 be 7m now—3 m: 533 “Mummy? .x. BEES

 

3855.8.“ 5 @3885 can momma 295m 28 Echo magnum .352: moosoxsom 05 mam: confined 83 0:5on
.388 E8 N a swag 35mm 083 838% .33 casebooa E 380:8 8.80 :8 Eu oTo 80¢ 8a 83 0:5on 28 noumu
Savage—mom £980 08:. wan: 338:3 ma? c895 0529:? “Baum $338835 Eomu§_ 3582 mom A ~35.

35

Egg .XN

 

 

a v §§m§
38.54.8028 - . 32 a .33ng .
$.85." 52¢ $8 2 mzuaz gas Begum 3. r ”2
§Eu<§~
”fizfiwacxowm
gamma...“
. . . as . a fioxfimxa
www.mm” .3wa gm; aziaz Awwrws 2.32382 Sm .3fi
.. 5333:.8
33.3%
. 853m§
a . . g Q .
mefimwwfig awwmw .x. 32 - 32o 3 x8 353 SS pm 5.33
Magmx:
£83 has“? so? 25 ca 9.2 NE
33.3 swag. a 8522:“... m aagasp EOE:
gaggsaofo

owing—Sangbgogmgo—gofiangofifigfiangéEggnog
08a gfiagfiaapgbgfiéfigéfiageag$3”:§~bfiu>oz
agggggssgggofiagafigmégsm 3.3.

36

 

 

 

 

 

 

 

a .39 as: 28
z
a 9.29 8.92, o 0.2
a .89 32, 28
a .39 38 8
a .889 0.8 o o.
9 .x V 089 28
9. .09 m. SN 8
a .m9 m8 o 26
so 2 - o 589$ 25
98 an 0. R089 8
m a .29 2.9 m 8 .39 59 a 8 9.: 58 m a .89 8.8 28
a 9 .29 8.9 m a .29 89 m a .59 QR m a .89 N. 5 z
m 9 .29 89 m 9 .29 88 a a 69 98 m a .89 m2 8
m 9 .29 $9 a 9 .89 26 m 9 N9 98 m a .29 «.2 o 0.2
n 9 .89 8.9 an a .29 B.» m 6 .2. s 8.2 m 9 .89 0.8 28
n a .89 89 n 9, .89 88 m 3 .m. 5 92 m a .m9 m8 2
a 9 .89 8.9 a a .89 59 a 9 .N. 5 8.2 a a .09 mom 8
a a .89 89 a 9 .:9 8.9 a 9 .89 92 a 9 59 39 o o.
m. 9 .89 89 ad 9 .89 29 em. 9 .m. : 92 a 9 .89 8.8 28
a 9 .89 29 o 9 .89 8.8 a 9 .89 2.: m 9 .m. s 2.8 2
n .m 9 .89 8.9 0.9 9 .89 89 o 9 89 «.2 a 9 d9 88 8
m 9 8.9 89 gm 9 .89 89 o... 9 .89 02 a 9 89 ”.8 o 26
.8 8-2 .5 28 so 82 .5 2.8 .m~___tm.._ 8m
.92, In 38 an m \z 9308 2 8E.

, 830555.. 5 9m mwmw,u_..Emm ucm flowamccQw fio "@5310? m -55? 380523
$5.32 E8536 20:3 226... .cozmgaam 555.2 5% 5:8: F §3m=m>m 622:: :8 .m min...

37

 

a C .89 8.8 m c .89 «mm m t .89 E. 3 26. Os.
3 2 .89 3... m 2 .29 S...” n 2 .59 Es 26. o.
m 9 .2. s 8.8 m 9 2.9 2... m 9 .t. s 8. 2 8... 0.2
a .2 .29 Sun a .2 .29 8.. a 2 .89 82 8.; .so
.529 0.852.; 8.. on an: 8.. 2 m9 82. 2 Eu... 2%

 
  

0 «mo E __ w 55.3 22.» 5223 233 2.85... .m 26cm 225.. . Eu 07¢ 2%
£3 25 0:58.55 Ea... 86328 .0 25 8-8 mo... 5.3 2 .wmv 3.3me >Efim=m>m .225 :8 oEwE:_o> .9 min...

38

t2 =0
200 =0
0 >80
U >80
0 25
0 08

0 >80

:00 =0

.00 :0
E0 :0

 

 

 

0 a 0 0 a 0 0 a 0.0 0 Avm N800
00.00 00.8 00.0w 00.0w mm.~m 00.0w N080 002m wofim mmdm gm
0 0 0 a 0 0 0 a 2.-» N82
3.00 3.2. 3.3 <2 00.3 00.02” 3.3 and 2 <2 9.8 8.50 0000 0000 08¢
0.0 0.0 0.0 0.0 0.0 0 0 0 n 0 Avw 800
0.02 0.3 2.2 2.2 0.: 2.0m wAm 2.3 92 0.0m .80: £000 0000
0.0 0 0.0 0.0 0 n 0.0 0 0 a 20800
00.3 3.3 N000 36m $1. mmém Edm mm.m0 002m No.0 0000 0000 0:0
0 0 0 0 n 0 0 0 0 a 2800
M02 fimm 2.2.0. ~.0N 0.0 0.02 2mm 0.3 QNN 0.2 5000 000.~
n n 0 0 0 0 0 0 n— 0 E
mmmmd 0 200.0 300.0 $8.0 $8.0 wumm0 0Nmm.0 50mm.— 0023 300.0 80805 000..
0 0 0 a 0 0 0 0 0 0 30
0000.0 0090.0 003.2 39.0 300.0 300.0 390.0 00mm; 300.2 $02.0 000805 08¢
0.0 0.0 0 0.0 n 0 0 0 0 0 27800000 3
302.0 SRO 0m 2 m0 520 33.0 902.0 32.0 «03.0 023.0 030.0 530% 22022
0 0 a n 0 0 0.0 0 0.0 0 27000000 ..w 3
030.0 300.0 003.0 38.0 500.0 003.0 203.0 203.0 203.0 N350 MOM
00.0 $0 00._ 20.0 00.2 00.0 3.0 5.0 00.0 5.0 .0885 2.
<3 2m OM Om Om <3 2m. OM Om 0m
08802 002000-002 20.4352
20.. 33 W-

w 0...... .1 00...... a. "a. .0

.80800 "00 88200 00; £20 05 8 022— 000 202080 05 8 020208 208 8080002 05. .0008 820 2003 2 00220800:

000 200 020208 8.82 .0235.“ 00 002000.. 000 28000 <>OUZ< .3. "0:20 000: <>002< 4.02320 0000
<>02< 020000 2880 008000220 20:20 08008 500800 2022 8000005 .020 0 8523 2.0050800 02025 .m 030,—.

39

0 >80
0 >80
:00 =0
0000 =0
0 >80
000.0 =0
0 >80
0 .08
000.0 :0

E0 =0
:0 :0

 

 

0 0 .0 0 0 0-0 ~50
00.00 :00 00.00 00.00 00.00 <00
0.0 0 0 0 0.0 A70 N800
00.00 00.00 00.00 00.00 00.00 .8000 0000 0000 000
0 0 0 0 0 0 0 0 0 0 A70 800
0.00 0. 00 0.0 0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 .800 £000 0000
0 0 0 0 0 0 0 0 A70 00
0000.0 0000.0 00 00.0 <2 0000 0000.0 0000.0 0000.0 <2 0000.0 220
0 0.0 0 0.0 0 A0800
0:0 00. 0 0 00.00 00.00 00.0 0000 0000 0:0
0 0.0 0 0.0 0 0 0 0 0 0 A800
0.00 0.00 0.00 0.00 0.0 0 0.00 0.00 0.00 0.00 0.: 00000 0000
0.0 0... a 0.0 0 o o 0 0 0 00
0000.0 0000.0 0000. 0 0000.0 0 000 .0 0000.0 0000.0 0000.0 0000.0 0 0 00.0 000820 0000
0 0 0 0 0 0 0 0 0 0 A00
0000.0 0000.0 0000.0 0000.0 000 0 .0 0000.0 0000.0 0000.0 000 0.0 0000.0 000820 008.0
0 0 0 0 0 0 0 0 0 0 $000000 00
0000.0 0000.0 0000.0 0000.0 0000.0 0000.0 000.0 0000.0 0000.0 003.0 003000 08—000<
0 0.0 0.0 0 0 0 0 0 0 0 $000000 T0 00
0000 .0 0000.0 0000.0 0000.0 00 0 0.0 0000.0 0000.0 0000.0 0000.0 0000.0 MOM
00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 _0>_>.80 .x.
<3 20 OM Om Om <3 20 OM 0m 0m
080002 000380 020> 0080000
200 008

05:80 0 2...;

4O

000.0 000
000.0 =0
0 >80
0 >80
0 >80
0 >80
0 >80
000.0 000

E0 :0

E0 :0

 

 

v v u n v 070 N800
00.00 00.00 00.00 00.00 00.00 00. 0 0 00.00 00.00 00.00 00.00 <00
00.00 00.00 0 0 .00 00.00 00.00 00 .00 <2 <2 00.00 00.00 .8000 0000 0000 08.0
u u u n u 070 800
0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 .8000 .8000 0000
n V H Ill " ANEOV
00.00 00.00 00.00 00.00 00.00 00.00 00.00 00.00 00.0 00.0 0000 0000 08.0
n n .I. n n 0800
0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.0 0.00 00000 0000
0000.0 0000.0 0 000.0 0000.0 0000.0 0000.0 0000.0 0000.0 0000.0 0000.0 0008000 0000
0000.0 0000.0 0000.0 0000.0 0000.0 0000.0 300.0 0000.0 0000.0 000 0 .0 0008000 0080
u H u n n 00-000000 00
000 0 .0 000 0 .0 0000.0 000 0 .0 0000.0 0000.0 000 0 .0 0000.0 0000.0 0000.0 003000 080000<
.I. u u u u 07800000 70 00
0000.0 0000.0 0000.0 0000.0 0000.0 0000.0 0000.0 0000.0 0000.0 0000.0 000
" A V H V
00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00>0>80 °\..
02 00 002 00 002 00 02 00 002 00
<3 020 CM Om 0m
20: 33 00.022.

.0 0000.0.

8 0000000098 000 .0000 000500 0000000000 80008000 3000 00008 05 .0008. 00000 0000000 8503 0000000800 0000 .0 0000.0.

41

0 >80
0 >80
00.0 000
000.0 000
0 >80
00.0 000
0 >80
0 >80
0.00.0 00.0

E0 =0

0.00.0 :0

 

070 N800

 

00.00 0 0.00 00.00 00.00 00.00 <M0m
070 N800
00.00 00. 0 0 00.00 00.00 00.00 .800: 00.00 000.0 0:00
0.8 0.2 0. a 0.00 0.2 0.: 0.00 0.2 30 0.00 0:8 50% 88
u A u " ATM WV
0000.0 0000.0 0000.0 0000.0 00 00.0 0000.0 <2 <2 0000.0 0000.0 g
0.50
E . a 080 00.00 8.00 00.0 80 82 20
V H u H V AEOV
0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.0 0 00000 000.0
V V H H V va
0000.0 0 000.0 0000.0 0000.0 0000.0 0000.0 0000.0 0000.0 0000 .0 0000.0 0008000 000.0
0000.0 0000.0 0000.0 0000.0 0000.0 0000.0 0000.0 0000.0 000 0 .0 000 0 .0 0008000 008.0
V " III N " A—nqomwom mv
0000.0 0000.0 0000.0 0000.0 0000.0 0000.0 0000.0 0000.0 0000 .0 0000.0 00030.00 080000<
v v v u v 07000000 70 00
0000.0 0000.0 0000.0 0000.0 0000.0 0000.0 0000.0 0000.0 0000.0 0000.0 MOM
00.0 00.0 00.0 00.0 8.0 8.0 8.0 .23 8.0 00.0 03.50 0..
002 30 00,0 30 02 BO 002 30 030 30
<3 20 OM Om 0m
50: 008 00.005.

00.2.80 .0 0300.

42

Table 7. Summar of ANOVA/ ANCOVA results.

 

 

 

 

 

RGR (r+s)
Source of variation Num DF Den DF F p value
High light site 1 46 40.24 <.0001
fert 3 46 0.16 0.92
site"fert 3 47 0.22 0.8826
spp 4 46 5.42 0.0012
site"spp 4 46 2.74 0.0397
fert‘spp 12 46 0.99 0.4753
site*fert*spp 1 1 46 0.94 0.5081
Low light site 1 19.4 0.14 0.7159
fert 3 19.6 0.22 0.8805
site"fert 3 19.6 0.05 0.9828
spp 4 69.8 1 1.32 <.0001
site"spp 4 69.8 0.61 0.6575
fert‘spp 12 69.8 1 .03 0.428
site‘fert*spp 12 69.8 0.57 0.856
final log(roots) v log(r+-s)
Source of variation Num DF Den DF F p value
High light site 1 9.85 0.47 0.5105
fert 3 8.28 2.18 0.1654
site"fert 3 8.26 0.78 0.5369
spp 3 33 1 64.71 <.0001
site“spp 3 329 2.01 0.1121
fert*spp 9 329 1.12 0.3513
site*fe1‘t*spp 9 329 1.61 0.1 104
log(r+s) 1 333 5611.7 <.0001
BO eliminated from analysis
Low light site
fert
site"fert
SPP
site“spp
fert‘spp
site‘fert'spp
log(r+s)

43

Table 7 Scont'dz.

 

 

 

 

 

Absolutggrowth

Source of variation Num DF Den DF F p value

High light site 1 10.5 6.92 0.0243
fert 3 10.4 0.16 0.9235

site"fert 3 10.5 0.07 0.9743

spp 4 36.5 20.76 <.0001
site"spp 4 36.3 0.56 0.6963
fert‘spp 12 36.6 0.37 0.9644
site"fert“spp 1 1 36.3 0.55 0.8562
Low light site 1 19.6 0.16 0.6973
fert 3 19.8 0.39 0.7618
site"fert 3 19.8 0.46 0.7132
spp 4 69.9 3.02 0.0233
site"spp 4 69.9 0.19 0.9433
fert‘spp 12 69.9 0.52 0.8959
site"fert*spp 12 69.9 1.19 0.3069

logglongestz v 12g (rt-s)

Source of variation Num DF Den DF F p value
High light site 1 10.7 0 0.9902
fert 3 9.7 0.6 0.6284
site"fert 3 9.69 0.02 0.9954
spp 4 471 22.22 <.0001

site"spp 4 470 3.36 0.01
fert‘spp 12 469 1.05 0.3994

site‘fert‘spp 12 469 0.87 0.575
log(r+s) 1 477 129.63 <.0001
Low light site 1 19.4 2.32 0.1437
fert 3 19.6 0.42 0.7417
site"fert 3 19.7 0.12 0.9443
spp 4 762 26.92 <.0001
site*spp 4 759 1.16 0.3251
fert*spp 12 759 2.13 0.0133
site*fert*spp 12 759 1.19 0.2849
log(r+s) 1 758 674.04 <.0001

44

Table 7 Scont'dz.

 

 

 

 

         
 

 

 

 

 

final log (r+s)
Source of variation Num DF Den DF F p value
High light site 1 9.22 21.61 0.0011
fert 3 8.92 1.04 0.4223
site"fert 3 8.92 0.56 0.6568
spp 4 475 399.65 <.0001
site"spp 4 475 8.86 <.0001
fert*spp 12 475 2.59 0.0024
site*fert*spp 12 475 0.93 0.5157
Low light site 1 20.1 0.03 0.8611
fert 3 20.3 0.34 0.7998
site*fert 3 20.3 0.77 0.5219
spp 4 764 442.77 <.0001
site"spp 4 764 1.34 0.2519
fert*spp 12 764 2.01 0.021
site*fert*spp 12 764 1.41 0.1576
final logjroot area) v log {rt-s)
Source of variation Num DF Den DF F p value
High light site ““7 .. 0 4.41.
fert 3 7.23 3.12 0.0949

 
 
    

 
  

  
 

  
  

° .JH'..~:=4s.u‘.-. 1‘ 052-7».
$1te*fert Z -

 

 

 

   

 

spp 117 19.82 <.0001
site"spp ”g " ”’1 'Egg“i
fert‘spp 12 117 1.43 0.1614
site*fert*spp 5 . . ,._, .-> .
log(r+<.s) 1 120 32.14 <.0001
Low light site 1 10.5 3.23 0.101 1
fert 3 10.6 1.12 0.3833
site"fert 3 10.6 1.24 0.3438
spp 3 176 76.93 <.0001
site“spp 3 174 1.65 0.1788
fert*spp 9 175 1.18 0.3095
site*fert*spp 9 175 1.1 1 0.3546
log(r+s) 1 182 218.92 <.0001

BO eliminated from analysis

45

Table 7 Scont'dz.

 

 

 

       

 

    
   

       
     

 

 

 

 

 

 

 

 

 

 

final logmot mass)
Source of variation Num DF Den DF F J) value
High light site 1 8.82 23.16 0.001
fert 3 8.32 2.04 0.1846
site*fert 3 8.32 0.49 0.6985
spp 4 476 335.92 <.0001
site“spp 4 476 9.33 <.0001
fert*spp 12 476 2.77 0.0012
site*fert*spp 12 476 0.76 0.6966
Low light site 1 20 0.22 0.6466
fert 3 20.2 0.3 0.8216
site*fert 3 20.2 0.94 0.4415
spp 4 764 388.67 <0.0001
site"spp 4 764 1.31 0.2637
fert*spp 12 764 1 .92 0.0293
site*fert*spp 12 764 1 .66 0.0719
final log (root area) v log (root
mass)
Source of variation um DF Den DF F . value
High light site if“. . 4
fert 3 7.35 2.58 0.1326
site"fert [ :7“ in”? ""“ r“ " “ii
spp 4 117 36.69 <.0001
site"spp ““ "1"" ““ :““__ j
fert*spp 2. 0.0269 _
site*fert*spp -- ~ ‘ w. ““:““‘:fi
log(r+s) 1 118 38.18 <.0001
Low light site 1 1 1.5 8.08 0.0154
fert 3 11.4 1.15 0.3716
site*fert 3 1 1.4 0.47 0.7107
spp 4 225 88.62 <.0001
site*spp 4 223 4.3 1 0.0022
fert*spp 12 223 1.01 0.4398
site*fert*spp 12 223 1.06 0.3961
log(r+s) 1 232 67.86 <.0001

46

Table 7 Scont'd).

log (max. mating depth)

 

 

Source of variation Num DF Den DF F p value
High light site 1 9.83 3.54 0.09
fert 3 9.66 0.76 0.5423
site"fert 3 9.66 0.07 0.976
spp 4 470 52.39 <.0001
site"spp 4 470 2.94 0.0202
fert*spp 12 470 1.56 0.1008
site‘fert‘spp 12 470 0.72 0.7328
Low light site 1 20.1 1.13 0.301
fert 3 20.3 0.55 0.6563
site"fert 3 20.3 0.09 0.9631
spp 4 760 63.33 <.0001
site"spp 4 760 0.37 0.8313
fert‘spp 12 760 2.03 0.0197
site"fert‘spp 12 760 0.78 0.6726

47

Table 7 Seont'd).

 

 

 

 

 

final log fine root area)
Source of variation ”Num DF Den DF 7 F p value

     

 

 

 

 

 

High light site
fert
site"fert
SPP
site“spp
fert*8pp
site*fert*spp
Low light site 1 11 0.9 0.3622
fert 3 1 1 1.69 0.2267
site*fert 3 11 0.44 0 732
spp 4 224 93.59 <.0001
site“spp 4 224 0.49 0.7462
fert‘spp 12 224 1 .03 0.4232
site*fert*spp 12 224 2.16 0.0147

48

 

 

c .o 0&3 “8&0 £550 8 32
3o "3% 2 A o 6 .26 <3 30 as was 82 > as 88 .33
A; "2.03 38% snags. 8 32
mod "2% 2 A o .6 .26 <3 30 02 32: Ema 22? > 83 82 2E
33 2A6 <3 2 32
:o.o 55o m89o 2 A26 6 .0 2m 02 32
Rod 226 <3 30 03 6a 38 250
00800005 00: 300
A0 .0 "20.003 08.0.00 03000.50 0: "00.3 $9: 0500 00203 > v.88 0030
wood 08.0 26 Au 6 2m 2 32
”So 6A 0 om B 32
~85 585 ME; 2 .26 .0 A6 :5 oz 32
Sod 68.0 2 A6 6 <3 30 32
Rod 68d m83 2A6 .o .26 om 3o .03
~86 26 <3 02 fie was 830
83 26A6 2m 2 32
83 £35 2A6 2600 5 oz 32
mad asd 26 .26 em 30 $3
$3 26 <3 02 fie 22: Ea 223
0.0 "233 38% 55:50 8 32
0 .o "233 “8&0 £550 8 fie 53on 328.2
2 .o "333 some Efié 8 32
0 .o "233 .80» £08.50 8 $3 mom
0:08, 0 008089800 80800 0000 0.50030 00000 8082030
SEEM 0805.0.

.3 "23.08: 9:52 <>ooz< .3 "2% a: same <.>oz.< 3500030 8.0 -9000 3.30.0 a «saw.

 

49

 

3o "2% z A 26 .o .6 <3 00 32
3o "2% z .26 A o om B 32
A0 .0 "2020000 08.0.00 5N000to0 0: .003 $2: 0500 00203 > £30 0:008 488 #02090
000.0 m000.0 Zao ARV .0 <3 00 300
A0 .0 "9003 08.0.00 030000.50 0: 00000 5000 0:008 .088 08.0.3
0:09» 20 008080800 60800 800 080030 20000 808.080
SEEM 0805.0.

€2.80 00 2.3.

50

Figure 1. Percent light availability measured on a plot basis in August 2001 under
full canopy cover. Plots in 14-26% openness, found on outwash (OW) and
moraine (M0), were grouped in a high light category (0). The low light category
(x) included plots in 3-10% openness, which were found on ice-contact (1C) and
moraine.

 

30 i I l

N
0
I

O CO

% light availability

8

T

W
xx><>§<><

l

 

 

 

OW IC MO
SITE

51

0080:0000. 0:600:00.“ 8.0

0 0000.0. 000 .0000 20:00. a 88.0 600800 00000030 3:000:00:
0:020 0000 .on :0 83200 0.3 8000805080 80000000
000800 :000 503 85200 0.8 00:0: 0003:9005 05 880
0.0 :05 0000000 00:03» 90000000 503 :32? 0.8 00:8
0003:9005 000 0.0800 0.0 00:0 0.0 :003000 00.0 005 00:09»
00200.00 00:0: 0003:9005 05 00:5 0 .0 5503 00.0.0 005
00:09, 0030000 .00 00:8 05 3200. 000000003 xom 6000.850
00:05 00:0 00:0 05 00 0:0 000000 x00 05 50000:: 05 95900
3000 x00 05 :0 0:00 00:00 00.0. .30 00.80 80 00-00 00:0
00 00.80 :8 00-0 8.0 832.00. 0.3 A2002 + Z. +0000 $020
50800: 002:0 0.8.0.0. .500000000 090000.000 000.0 5:9:
0:0 m00>00 00.880: 008 :0 005000000 .00 000mm .0 00:00.0

 

 

 

 

 

 

0.2 O. 050
_ 0 0 0 0
z a

26 0 .. .

60 U [m

o a .00 0..
E05009... u %.
000:5“. .

 

 

_ 0 p 0 _

.000 5 8-2 3

 

0.2 0. 30 o
_ .
26 .
m0 x X
o .0 .. U. m $8
20500:. .0
50:5“. _ .0
3
I -82 e
5
X .r
O.
x .M
1 182 m.
, 52-20
_ _ _ ooow
0.2 o. 26
o _ _ _ T a
to? - -om
.I mm .1 n
.08 I - -8 W
i 1 6
- m -8 m.
.50 m
l— 1 S
I 100 MW
.43. - -
I 500802-20 180

 

 

 

 

P

0

0

 

 

52

OS. 0. >>O 0.2 o. >>O

 

 

 
 

 

 

 

 

 

 

 

 

_ _ _ _ _ d _ _ _ _
Z I
200 m
00 D
o a
j J 1 1
000E009... o
.0055“. I.
O *% 19
I MAW 1 I "w. .1
m .1 Hum
B. an mm
60% as 8-2 3 q 50% go 2-0 a
P r _ _ _ _ _ _ F _

53

 

 

08:80.00? 0:005:06 :0.“ m 050:. 0:0 30E x00 .00 0000:0308
:00 N 200mm 000 0.0—9:00 SV 80 00A: 05 A3 80 2-0 00:0 00030 0:03 00:00 :00 .:00000000
Bum—0:00 Sow >02 0000 5:2: 0:0 3000808090 m0 :8 :0 00000000 .00 Spam .m 0.500...—

Figure 4. Comparison of % soil volumetric water (O-30cm) between
20-21 June and 19 July 2001, estimated with TDR. See Table 4 for

significant differences.

2 I l
E1 a) High light, OW ‘
g1 —
0 1 ‘
.93 1 _
‘é’ _
.g 1 1'
2r. 8 ~
5 6 1 —
9 4 ~
°\° 2 c? "

0_ J 1

June July

2 I 1
A1 c) Low light, IC ‘
0‘ 1
0
b1
.31} 1 0
m
3 1 -
:3 3 -
'6 4 $ ‘
>
0 2 ‘

0 l

 

 

 

 

 

 

 

 

 

 

 

June

July

54

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2 I 1
18" b) High light, MO
16*
,4. l
12—
10* l
8.—
61-
4_ :
2.—
0 l 4
June July
2 l 1
1 d) Low light, MO
1
1 T
12’
10L 1
8— x
6* .
4F— %
2.—
0 June July

 

 

rEL

<>> 2w Om Om 0m

 

 

_ _ fl 0 0

.l l —..

1 02 30: 000: 3 1 NM
1 .. no
r 1 to
1 I 0.0
.. .. 0.0
I 1 3
.I O l m.o
1 , 1 ad

 

 

 

 

 

<2, .20 om Om 00
0 0 fl 0 _

q 0 ..
Q 3000002050 1
fl 0
0 ..
0 mm .. Ar 1
Q in Q0 D I
0 xx _ 0% .0

 

00:05 00:00:08 00

000.000 00 3:000 00000.: :000000: 4 .:00000: x 83200 + £80008 x 400:8 00 00000 000: 0 :0 Sow
000800000 .00 0:0:3000 0:0 0:3. .00 0:0:5000 :003000 0:33.50. 00500000 .00 5000005 .m 0.500..—

od
_..o
Nd
md
06
md
0d
No
0.0
06
o...

angle aunr /eAg|e ides

55

<>> 2w Om

Om Om

 

 

02 £0: 33 e

 

 

od
_.d
Nd
0d
0.0
m d
od

. Nd

0d
0d
0.0

<>>

2w

 

 

/

\.

 

0: 00 00
_ 0 _
Q
0: 50: 33 0
0
x0.
4 x..
x O

 

 

.3880 m 2:0:

od
rd
Nd
md
0d
md
od
Nd
0d
0d
d...

eAgle aunr /6Al|81d9$

56

RGR (g 9’1 season“)

final — initial (roots + stem) (9)

Figure 6. Species and site differences in relative and absolute growth in high
and low light. RGR in high light (a) and low light (b) was calculated as ln (final
roots+stem)- ln (initial roots+stem). Absolute growth in high light (c) and low
light ((1) was calculated as final (roots+stem) — initial (roots+stem). Initial
biomass values are from the harvest in early May 2001, final biomass values are
from the harvest in early September 2001. Box plots show pooled individuals
from all fertilizer treatments (see Figure 2 for explanation of box plots). See
Tables 5 and 6 for significance differences.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2.0 I I I I I 2.6 I l I I I
a) RGR, high light b) RGR, low light
10* g " 1.5- ‘
1.0“" I g " 1.0" '-
I .

0.5 g 0.5 at "
0.0” * ,. r 0.0- . .. -
_ OW _ _ IC

-0.5 [:1 M0 -0.5 G M0
-1 l J J J l _1 l I l l l
' BC BC R0 SM WA ' BC BC R0 SM WA
2.0 I I I I I 2. I I I I I
c) Absolute growth, high light (1) Absolute growth, low light
1.5- - 1.5' '
1.0— 0'" - 1.0- * -
I g *
0.5- fil - 0.5h -
. 9 ¢ .
*
col- 4" . wk 1' - 0.0- 1" $ 4: $$ -*
_ * OW _ _ .. IC
'05 0 MO '0-5 0 Mo
-1 l l l l l _1 l l I l l
' BC BO R0 SM WA ' BC BC R0 SM W

57

Figure 7. Box plots of biomass at end of the growing season. Final whole plant
biomass, from the early September 2001 harvest, is graphed as high light (a) and
low light (b). F inal root biomass is grouped as high light (c) and low light ((1).
Only data from non-fertilized (control) plots are included. See Tables 5 and 6
for significant differences.

 

 

:I:

T

4 l I I I I 4 I T I I I
_a) Final whole plant biomass, . _ b) Final whole plant biomass,
high light low light at:

a 3 - - 3 -' "
v g OW '0
E 0 MO ‘ ' D MO ‘
9 2 — * §| - 2 - 1 -
m
*6
o
a:

_ t j: .0
1 g F1 $5:
ofillll o‘flllfi

‘%

O

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

BC BO R0 SM WA BC BC R0 SM WA
2.5 I I I I I 2.5 I I I I I
~ c) Final root I (1) Final root
biomass, _ _ biomass,
2'0? highlight ow 2'0 low light IC
3 C] MO . i D MO
.915- . ~1.5- 1* -
o
O L . .
m III
1.0- r 1.0- r
i i! * ‘ * . :: -
0.5— i -o.5— —
00 g i I J I 00 L I I
' BC BC R0 SM WA ' BC BC R0 SM WA

58

A9 mmmE A605 + £00: mo.
Fo TN-.. 0 TN; 0 TN; 0 TN-.. 0 TN-

d 1 d 1 q d. 1 I 1 I i

u 4 - 1.. - #v #1 I... IN.-

0
v .v 0 a. 4a A. .
B
1.. I: 41 1.. IF.
v v A

 

 

 

 

 

I 1'

. o A. ab L. .V .. N .
Ham—Bax. .... I. Q I... o 1.. IFI
58:00-03 . : : N 4 \ : A

w .v k. ‘v .v A

 

 

,
l
l
.4
l
hi
I
(6) ssetu too: 60'

is... a... .... k. ... a. .... a. t-

65802

 

 

. .v . A. O A. .
f A? ll 11 Al I
I h riih i P P D h t n r i b t b b L P I h P P i L h i P i P i b
d 1 1 d 1 d d 1 u 1 a 1 1 d 1 d d 1 u d 1 d 1 d u d 1 u 1 d
.I A! 1' 1.! Al 0 1“.

Es... M a. N. ..,. ._.. .... A
58350 X

biBifi F I Pl- D bib h h i .- D bihh PL} hi 5 b

<>> 2w Om Om om
.mooeocobmo “53¢?me he.“ 0 28 m 833. com .83 Sufism vflooa 30% £3me
=< .39: 35% + Soc Rem wB 3&3 mag: 88 Ram wo_ mm gamma 226 838 BE: “com .m «BME

 

 

 

 

 

 

 

 

 

 

b 5 hi

 

59

Figure 9. Area of fine roots (<2mm diameter). Only non-fertilized individuals

included for both (a) and (b). See Figure 2 for explanation of box plots and Tables 5
and 6 for significant differences.

 

 

 

 

 

 

 

 

 

 

 

120 l l l I I
100 _ a) High light, Outwash —
“g 80 — —
7;
e
«I 60 _ —
*5
e
.2 4o — —
Ll.
20 __ ._
*
EH
0 l l I l 1
BC BC R0 SM WA
120 I I I l I
b) Low light, 1c and MO
100 — IC ~
* [:1 Mo
"s 80 — _
9.
(U
9
g 60 — —
O
2
."s’ 40 —- —
Ll.
20 — —
0 I I A l l

 

 

60

 

 

 

 

:3 $9.55 €on + woo: mo.

 

 

 

N._. wd vd 0.0 v.0- mdi N._.-.N.—. wd Yo 0.0 To- md- N._.-.N
_ _ . q _ 00 O _ _ . _ 7 00 o
<>> D Em: 3o: 6:882 Au <>> D Em: >5— JoSeooooH 3
. 2m 4 into um M into
Om + Om x
Om x 0 0m 0 00
i 0m 0 O .36 o .o
O X AW 0
D . X .
i x imuw %x «D N F
«My 4 D
.. xi Xmas -8; + cw? «so 3
+ fly 1+ d
— _ AV — _ _ F.N _ I+H _ _ _ F.N

 

 

.F md v.0 od v.0- m6- «.7.
— _ _ q . 00
<3 s a»: .32 £838 a
I 55 4 IN?
Om +
Om x
I 0m 0
B m.
If C
I X vaX % N.
ITX X Qv¢§b
D
n ++x «ado m.
LI x
IT
b _ L _ _ F

 

 

 

.3820:an “:85de .8: e was m 833. com .on 05808 Em: 32 98 3V 88590”: Em: 32 A3 Emmi—8
Em: EmE .8: £56365 3553-5: .«o meme ESQ 20:3 mews.» ~on 083m 88 on: 38.8%wa m3 .3 9.5mm..—

0'

c5

(awe) Bale 1001 am; 60|

o

F

F

.N

61

 

 

:9 30:83 000. mo.

 

 

o T N- o _.- N- r o T N-
. _ . _ . .o . _ . s . od . _ . 4 . o.o
<>> D Em: 30: 05802 G <2, D Em: 32 408000-00: 3 <>> D Em: EmE £8330 Q
.5 a .3 am M 0. -md - Em a .3
OK + o om x 0 WC om +
om x 000% 0m 0 %0 m0 om x wmw
0m 0 a lo; 0 o. .3 - 0m 0 Mo -3
OWOQVD 00
xi“. Women, Im.—. x xxx .(MDM 0 lm._. r. + x now, D lm.—.
x h. 0% xxm-vnu-m %B x + x 0% D
”Ma-TX DJ .7.” @kb x +iux d 0
++++++¢on so Io.N Hemm. c JON I H... x IoN
+ w 0
. _ F p . m.N r _ F L . «HWN . _ h h _ WTN

 

 

 

 

 

 

.mooeouobfl 88:36 8: c 05 m 838. 00m
$3365 BEEe a. 8205 3 oases Ba 3 33832 :0: 33 23305 80:50-5: :5 832: 3
E5380 Em: Ema Avonbomflab me: $285 88 member «can oomtsm 008 :38 mm 850% «0.8 008 0:6on .: 0.5mm:

(zwo) 9919 902an 1001 Ietot 60|

62

Figure 12. Approximate maximum rooting depth. High light (a) includes all
individuals and low light (b) includes only non-fertilized individuals. See
Tables 5 and 6 for significant differences.

Root depth (cm)

 

 

 

 

 

 

63

 

 

 

 

H T I I j I 60 I I I *I I

i a) High light * OW ‘ t 1)) LOW light IC

i“ 0 MO ‘ 50 o 0 MO

I . * - 40L

. * q k

-— - 30l—

I- g! g. i.

i- * Q 3. .- 20i-

F I I-

r' g- - 10r-

- J I I 1 I 0. 1
BC 80 R0 SM WA BC 30 R0 SM WA

 

:9 $0805 E90 + £00: mo.
For-N-.. oF-N-F oF-N-r oF-N-F or-N-

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

d I q I i i d d I u 1 q I d I a I I q l 1 1 H I q I q I u I q o
I V J .I I j n I j o o
a g v . . v . . L
1 I I I I 1 fl I I 1 m-o
. . v 0 . . J . . .
“£m_— 30— I o o I. I I I I I I I Q I. OIP
mamgoz . . v 0 A a kuv A . mm . v .
I § 1 t 0fi~w®o n V 0 j I 1 I I moF
. . v . v . v A w .
I i r n I nIA In I n I n I I I I n I I n In I niI i I n1 1P I I I I nI o N
In I d I q I q u I q I q I an In I q I a I n In 1 d I q I 0.. I: I a I q 1 1... Ono
. g . . . g . . v .
. . o i . . . . . . md
£9.30. . o g. o : : : o . I
o I I o I I I m I I 1 0.? m
65:88— ' &@u g fimoo . . $ . fi @00 . v . II
.I 1 U I. I o I_ I n I I. nor 0
. g . o . v 0 x v . v A m
In I P I n I b4 In I P I n I n1 In I n I P I nI 1n I n I n I n1 In i n I n I b1 o o N w
I. I q 1‘ d I «I I1 I d I d I «I 1. I q l d I «I #1 I u I q .4 Id i q I q 1 - 00° 1
. . v . v . f A v m c m
I 1 I I I I I I I I m o ‘06-
. 4 . . a . v . v 4 )
20. .52 . o . . o . . 9
05905.. @m I o .:. : €68 :. an ”or Iw-
o I 4 I I I % I I $ I I I. m-v
. A . . v . v . v x
In I P I n I n; I! I n in i nm In n I I I n1 1 Ii P I I I n1 In I n I n I a DIN
Id 1 q I q I an I: I J1 I a I .1 In I u 1 u i A I. I d I d I 1.. i I u I d I :1 Ono
. A . A . . . L f A
I A I I I I I I O I m-o
v o . v . fi . v . v A
3029; . a. : : o .. : . 3
stCE—O . % .. .. .. kf .
t I I I I 1 I L 1 I m.—‘
. . . o . fi- . . A v .
In P n I b t n1 Ih I n I n P n n I n I P i nI In I n . n I b IF I n I n I n; com

 

<>> 2w Om Om om
.mooeonommv EeéEmmm 8: o 05 m 8:an 00m .920 60:00 80.:
$3365 $205 :8 as 05808 23 av 3980-02 Em: 33 assumes confine as see manages oases
3v 05808 05 A8 E3330 Em: Emmm $080888“ mos meme Edi 20:3 3%? 5000 mcuoom .m— 0.5mm:

64

 

 

 

 

 

 

 

 

 

 

 

 

 

 

8. 300 8. 02 3. 22.
_ _ I A _ C
I .

.— o

— H 15..
A
. 02

05802 A0
_ _ _ n F AON

8. 500 8. 02

5. 22.

 

 

 

 

 

 

 

L b _ P

3:388 3

L

16—.

 

5m

3. 300 8. 02 3. 22,

 

 

1
L0

nos omewmm %

 

 

 

 

 

£3350 Q .
LON

.800» .050 8 00.80800

 

L _ L _ r

32 020 0.8 008000.000 00.0 £00380 :0 300800 80003 Sow ES. .002 8008080m 05 coon 23.. 8 00.80800
68808 00 32 bfiaouqooxo mm 88 b3. 8 80003 050822, 8080A $080 85 8 000: 35 82: 0:0 880b€
0 80b 0.3 80m 08“.. 003 80.“ 800 08802 .800» m mac—5 0808080008 A8002: me .«0 0850800 .3 0.5»?—

65

Figure 15. Trade-off of high soil resource RGR and low soil resource survival. For
high light, the trade-off is stronger when extrapolated to 5 years (a versus b). Sugar
maple is an exception to the trend seen in (a) and (b), showing low survival on
outwash without the trade-off of high RGR on moraine. When the most common
light levels for each site are considered (high light on outwash, low light on
moraine), sugar maple then appears to show an RGR versus survival trade—off (c).

 

 

1.5 I I I I 1.5 I I I I

a) data from 2001 O O J
95 V V
0: 1.0b ' 1.0" -
E
g A + - I. A + -I
.I:
.23 0 BC X X
0 0.5" X 30 ‘ 0,50 _
2 + RO

A SM . . .

v WA b) 5 year estimates

 

 

 

 

 

 

O. l l l l O l l l l
8.0 0.2 0.4 0.6 0.8 1.0 8.0 0.2 0.4 0.6 0.8 1.0
OW high light survival probability (OW high light survival probability)"5

 

1.0I I I I I
c) low light RGR versus
high light survival
2:9 V 0A
(E
g, 0.5I -
E
O I
2 l +x

 

 

 

0. l l J l
8.0 0.2 0.4 0.6 0.8 1.0
(OW high light survival probability)"5

66

September (roots + stem) (9)

Figure 16. Estimated species biomass accumulation'(roots +
stem) on outwash under high light to year 2014, assuming
constant RGR..

 

100

90

8O

70

60

50

4O

30

20

1O

 

 

 

 

 

Year

67

REFERENCES

Aerts, R., DeCaluwe, H., Konings, H. (1992) Seasonal allocation of biomass and nitrogen
by four Carex species fiom mesotrophic and eutrophic fens as affected by nitrogen
supply. Journal of Ecology, 80, 653-664.

Aerts, R., Chapin, F .S. III. (2000) The mineral nutrition of wild plants revisited: A re-
evaluation of processes and patterns. Advances in Ecological Research, 30, 1-67.

Aerts, R., Berendse F. (1989) Above-ground nutrient turnover and net primary
production of an evergreen and a deciduous species in a heathland ecosystem. Journal of
Ecology, 77, 343-356.

Augspurger, CK. (1983) Seed dispersal of the tropical tree, Platypodium elegans, and the
escape of its seedlings from fungal pathogens. Journal of Ecology, 71, 759-771.

Billings, W.D. (1950) Vegetation and plant growth as affected by chemically altered
rocks in western Great Basin. Ecology, 31, 62-74.

Brokaw, N. (1985) Treefalls, regrowth and community structure in tropical forests. In
The ecology of natural disturbance and patch dynamics. Edited by S.T.A. Pickett and P.
White. Academic Press, Orlando. pp 53-69.

Canham, C. (1988) Growth and canopy architecture of shade tolerant trees: the response
of Acer saccharum and F agus grandifolia to canopy gaps. Ecology, 69, 786-795.

Casperson, J .P., Kobe, R.K. (2001) Interspecific variation in sapling mortality in relation
to grth and soil moisture. OIKOS, 92, 160-168.

Chapin, F. S. (1980) The mineral nutrition of wild plants. Ann. Rev. Ecol. Syst, 11, 233-
260.

Chapin, F.S., Autumn, K., Pugnaire, F. (1993) Evolution of suites of traits in response to
environmental stress. The American Naturalist, 142, S78-S92.

Comas, L.H., Boumas T.J., Eissenstat, D.M. (2002) Linking root traits to potential
growth rate in six temperate tree species. Oecologica, 132, 34-43.

Gankin, R., Major, J. (1964) Arctostaphylos myrtzfolia, its biology and relationship to the
problem of endemism. Ecology, 45, 792-808.

Gleeson, S. K., Tilman, D. (1994) Plant allocation, growth rate and successional status.
Functional Ecology. 8, 543-550.

Goldberg, DE. (1985) Effects of soil pH, competition, and seed predation on the
distributions of two tree species. Ecology. 66, 503-511

68

Grime, J.P. (1977) Evidence for the existence of three primary strategies in plant and its
relevance to ecological and evolutionary theory. The American Naturalist. l l 1, 1169-
1 194.

Grime, J .P., Hunt, R. (1975) Relative growth-rate: its range and adaptive significance in
local flora. Journal of Ecology, 63, 393-422.

Grubb, P.J., Lee, W.G., Kollmann, J ., Wilson, J .B. (1996) Interaction of irradiance and
soil nutrient supply on growth of seedlings of ten European tall-shrub species and Fagus
sylvatica. Journal of Ecology, 84, 827-840.

Gunatilleke, C.V.S., Gunatilleke, I.A.U.N., Perera, G.A.D., Burslem, D.F.R.P., Ashton,
P.M.S. (1997) Responses to nutrient additions among seedlings of eight closely related
species of Shorea in Sri Lanka. Journal of Ecology, 85, 301-311.

Holrngren, M. (2000) Combined effects of shade and drought on tulip poplar seedlings:
trade-off in tolerance or facilitation? OIKOS, 90, 67-78.

Horvitz, C.C., Schemske D.W. (2002). Effects of plant size, leaf herbivory, local
competition and fruit production on survival, growth and future reproduction of a
neotropical herb. Journal of Ecology. 90, 279-290.

Host, G.E., Pregitzer, K.S., Ramm, C.W., Lusch, D.P., Cleland, D.T.. (1988) Variation in
overstory biomass among glacial landforms and ecological land units in northwestern
Lower Michigan. Can. J. For. Res. 18, 659- 668.

Host, G.E., Pregitzer, KS. (1992) Geomorphic influences on ground-flora and overstory
composition in upland forests of northwestern lower Michigan. Can. J. For. Res. 22,
1547-1555. ‘

Huante, P., Rincén, E., Acosta, I. (1995) Nutrient availability and growth rate of 34
woody species from a tropical deciduous forest in Mexico. Functional Ecology. 9, 849-
858.

Hubbell, S.P., Ahumada, J .A., Condit, R., Foster, RB. (2001) Local neighborhood effects
on long-term survival of individual trees in a neotropical forest. Ecological Research. 16,

859-875.

Hutchinson, TC. (1967) Comparative studies of the ability of species to withstand
prolonged periods of darkness. Journal of Ecology, 55, 291-299.

Khurana, E., Singh, J .S. (2000) Influence of seed size on seedling growth of Albizia
procera under different soil water levels. Annals of Botany, 86, 1185-1192.

69

Kobe, R.K. (1996) Intraspecific variation in sapling mortality and growth predicts
geographic variation in forest composition. Ecological Monographs, 66, 181-201.

Kobe, R.K., S.W. Pacala, J .S. Silander Jr. (1995) Juvenile survivorship as a component of
shade tolerance. Ecological Applications, 5, 517-532.

Kobe, R.K., G.E. Likens, C. Eagar. (2002) Tree seedling growth and mortality responses
to manipulations of calcium and aluminum in a northern hardwood forest. Canadian
Journal of Forest Resources, 32, 954-966.

Kruckenberg A. R. (1954) The ecology of serpertine soils. III Plant species in relation to
serpentine soils. Ecology, 35, 267-274.

Lambers, H., Poorter, H. (1992) Inherent variation in growth rate between higher plants:

a search for physiological causes and ecological consequences. Advances in Ecological
Research, 22, 187-261.

Lambers H., T.L. Pons, F .S. Chapin. (1998) Plant physiological ecology. Springer
Verlag.

Lantham, RE. (1992) Co-occurring tree species change rank in seedling performance
with resources varied experimentally. Ecology, 73, 2129-2144.

Likens, G. E., Driscoll, C. T., Buso, D. C. (1996) Long-Term Effects of Acid Rain:
Response and Recovery of a Forest Ecosystem. Science, 272, 244-246.

Likens, G.E., Driscoll, C.T., Buso, D.C., Siccama, T.G., Johnson, C.E., Lovett, G..M.,
F ahey, T.J., Reiners, W.A., Ryan, D.F., Martin, C. W. & Bailey, SW. (1998) The
biochemistry of calcium at Hubbard Brook. Biogeochemistry, 41, 89-173

Lorimer, CG. (1983) A test of the accuracy of shade-tolerance classifications based on
physiognomic and reproductive traits. Canadian Journal of Botany, 61, 1595-1598.

Lusk, C.H., Contreras, 0., Figueroa, J. (1997) Growth, biomass allocation and plant
nitrogen concentration in Chilean temperate rainforest tree seedlings: effects of nutrient
availability. Oecologia, 109, 49-58.

MacDonald, N., Burton, A., Liechty, H., Witter, J ., Pregitzer, K., Mroz, G., Richter, D.
(1992) Ion leaching in forest ecosystems along a Great Lakes air pollution gradient.
Journal of Environmental Quality, 21, 614-623.

Mahmoud, A., Grime, J. P. (1976) An analysis of competitive ability in three perennial
grasses. New Phytol. 77, 431-435.

McGraw, J .B., Chapin, RS. (1989) Competitive ability and adaptation to fertile and
infertile soils in two Eriophorum species. Ecology. 70, 736-749.

70

Meziane, D., Shipley, B. (1999) Interacting components of interspecific relative growth
rate: constancy and change under differing conditions of light and nutrient supply.
Functional Ecology, 13, 611-622.

Milberg, P., Pérez-Feméndez, M.A., Lamont, BB. (1998) Seedling growth response to
added nutrients depends on seed size in three woody genera. Journal of Ecology, 86, 624-
632.

Pacala, S.W., Canham, C.D., Silander, J .A., Jr., Kober, R.K. (1994) Sapling growth as a
function of resources in a north temperate forest. Can. J. of F or. Res., 24, 2172-2183.

Peace, W.J.H., Grubb, P.J. (1982) Interaction of light and mineral nutrient supply in the
growth of Impatiens parviflora. New Phytol. 90, 127-150.

Phares, RE. (1971) Growth of red oak (Quercus rubra L.) seedlings in relation to light
and nutrients. Ecology, 52, 669-672.

Reich, P.B., Walters, M.B., Tjoelker, M.G., Vanderklein, D., Buschena, C. (1998)
Photosynthesis and respiration rates depend on leaf and root morphology and nitrogen

concentration in nine boreal tree species differing in relative growth rate. Functional
Ecology, 12, 395-405.

Ritter, D. F ., Kochel, R. C., Miller, J. R. (1978) Process Geomorpholgy. Third edition.
McGraw Hill. Boston, Massachusetts.

Runkle, J .R. (1990) Gaps dynamics in an Ohio Acer—Fagus forest and speculations on the
geography of distributions. Can. J. For. Res., 20, 632-641.

Sack, L., Grubb, P.J. (2002) The combined impacts of deep shade and drought on the
growth and biomass allocation of shade-tolerant woody seedlings. Oecologia, 131, 175-
185.

Schlesinger, W.H., DeLucia, E.H., Billings, W.D. (1989) Nutrient-use efficiency of
woody plants on contrasting soils in the western Great Basin, Nevada. Ecology, 70, 105-
113.

Schmidt, G., Zotz, G. (2001). Ecophysiological consequences of differences in plant size-
in situ carbon gain and water relations of epiphytic bromeliad, Vriesea sanguinolenta.

Plant, Cell and Environment, 24, 101-112.

Soil and Plant Analysis Council Inc. (2000) Soil Analysis: Handbook of Reference
Methods. CRC Press. Boca Raton.

Steinbauer, GP. (1932) Growth of tree seedlings in relation to light intensity and
concentration of nutrient solution. Plant Physiology, 7, 742-745.

71

Stock, W.D., Pate, J .S., Delfs, J. (1990) Influence of seed size and quality on seedling
development under low nutrient conditions in five Australian and South Afiican members
of Proteace. Journal of Ecology, 78, 1005-1020.

Tansley, A. G. (1917) On competition between Galium saxatile L. and Galium Sylvestre
Poll. On different types of soil. Journal of Ecology, 5, 173-179.

Tilrnan, D. (1985) The resource-ratio hypothesis of plant succession. The American
Naturalist, 125, 827-852.

Vitousek, P. (1982) Nutrient cycling and nutrient use efficiency. The American
Naturalist, 119, 553-572.

Walters, M. B., Reich, P. B. (1996) Are shade tolerance, survival, and growth linked?
Low light and nitrogen effects on hardwood seedlings. Ecology, 77, 841-853.

Walters, M.B. Reich, RB. (1997) Growth of Acer saccharum seedlings in deeply shaded
understories of northern Wisconsin: effects of nitrogen and water availability. Can. J.
For. Res. 27, 237-247.

Walters, M.B., Reich, RB. (2000) Seed size, nitrogen supply, and growth rate affect tree
seedling survival in deep shade. Ecology, 81, 1887-1901.

Wright, I.J., Westoby, M. (1999) Differences in seedling growth behaviour among
species: trait correlations across species, and trait shifts along nutrient compared to
rainfall gradients. Journal of Ecology, 87, 85-97.

Zak, D.R., Host, G.E., Pregitzer, KS. (1989) Regional variability in nitrogen

mineralization, nitrification, and overstory biomass in northern Lower Michigan. Can. J.
For. Res. 19, 1521-1526.

72