Directory UMM :Data Elmu:jurnal:T:Tree Physiology:Vol15.1995:

Tree Physiology 15, 317--325
© 1995 Heron Publishing----Victoria, Canada

Belowground carbon allocation in unfertilized and fertilized red pine
plantations in northern Wisconsin
BRENT E. HAYNES and STITH T. GOWER
Department of Forestry, University of Wisconsin-Madison, 120 Russell Laboratories, 1630 Linden Drive, Madison, WI 53706, USA
Received February 22, 1994

Summary We estimated carbon allocation to belowground
processes in unfertilized and fertilized red pine (Pinus resinosa
Ait.) plantations in northern Wisconsin to determine how soil
fertility affects belowground allocation patterns. We used soil
CO2 efflux and litterfall measurements to estimate total belowground carbon allocation (root production and root respiration)
by the carbon balance method, established root-free trenched
plots to examine treatment effects on microbial respiration,
estimated fine root production by sequential coring, and developed allometric equations to estimate coarse root production.
Fine root production ranged from 150 to 284 g m −2 year −1 and
was significantly lower for fertilized plots than for unfertilized
plots. Coarse root production ranged from 60 to 90 g m −2 year −1
and was significantly lower for fertilized plots than for unfertilized plots. Annual soil CO2 fluxes ranged from 331 to 541 g

C m −2 year −1 and were significantly lower for fertilized plots
than for unfertilized plots. Annual foliage litterfall ranged from
110 to 187 g C m −2 year −1 and was significantly greater for
fertilized plots than for unfertilized plots. Total belowground
carbon allocation ranged from 188 to 395 g C m −2 year −1 and
was significantly lower for fertilized plots than for unfertilized
plots. Annual soil CO2 flux was lower for trenched plots than
for untrenched plots but did not differ between fertilized and
unfertilized trenched plots. Collectively, these independent
estimates suggest that fertilization decreased the relative allocation of carbon belowground.
Keywords: carbon balance, fertilization, litterfall, Pinus resinosa, root production, soil respiration, trenched plots.

Introduction
Despite the importance of root production in forest carbon
budgets, it is still unclear how soil fertility affects the allocation of carbon to roots. In many western conifers, the absolute
amount of carbon allocated to fine root production is inversely
related to site quality (Keyes and Grier 1981, Vogt et al. 1983,
Santantonio and Hermann 1985, Vogt et al. 1987, Comeau and
Kimmins 1989, Kurz 1989). A fertilization study has also
shown that increased nutrient availability decreased fine root

production of a Douglas-fir forest in New Mexico (Gower et
al. 1992). However, Nadelhoffer et al. (1985) reported that, in
forests in Wisconsin, belowground carbon allocation is posi-

tively correlated to nitrogen availability and aboveground productivity.
It is difficult to reconcile the differing effects of nutrient
availability on carbon allocation patterns because different
methods of estimating fine root production have been used to
study forests in different climates. Furthermore, all of the
methods used to estimate fine root production require assumptions that are difficult to test, and belowground spatial heterogeneity is large. Most estimates of fine root production are
based on changes in fine root biomass in sequential soil cores.
However, width and depth of soil cores, frequency of sampling, root separation methods and classification of roots as
live or dead can all significantly affect estimates of fine root
productivity (Vogt et al. 1986, Vogt et al. 1989, Publicover and
Vogt 1993). In-growth cores, where root production is measured as colonization of root-free soil (Persson 1983, Ahlström
et al. 1988), generally give low estimates relative to sequential
coring and carbon budgets (Nadelhoffer and Raich 1992, but
see Neill 1992). The N budget technique uses a mass balance
approach to estimate root production and requires accurate
measurement of several N fluxes (Nadelhoffer et al. 1985).

An alternative explanation for the different patterns of belowground carbon allocation in response to nutrient availability is that there is an interaction of climate and nutrition on
carbon allocation. For example, several researchers have reported a negative relationship between belowground carbon
allocation and latitude (Schlesinger 1977, Vogt et al. 1986) or
mean annual temperature (Gower et al. 1994), implying a
strong influence of climate.
Fine and mycorrhizal root primary production is only one of
the major processes determining the total amount of carbon
allocated to roots. Total belowground carbon allocation can be
separated into two primary components: root and mycorrhizal
respiration, and root and mycorrhizae production. Because
direct measurements of root respiration are difficult to obtain,
total soil CO2 efflux is often used as an index of root and
mycorrhizal respiration (Raich and Nadelhoffer 1989). Various
factors influence soil respiration, but soil temperature and
water content are the main abiotic factors (Singh and Gupta
1977). Soil respiration for a Pinus radiata D. Don forest in
Australia has been successfully modeled with these two variables alone (Carlyle and Than 1988).
The usefulness of soil respiration as an index of below-

318

HAYNES AND GOWER

ground carbon allocation is greatly improved if the proportions
of respiration attributable to roots and mycorrhizae and freeliving microbes are determined. In situ measurements of root
respiration are difficult to obtain (Vogt et al. 1989); however,
comparison of soil CO2 flux between root-free trenched plots
and untrenched plots in forests may provide a simple approximation of the contribution of root and mycorrhizal turnover
and respiration to total soil CO2 efflux. For example, Bowden
et al. (1993) and Ewel et al. (1987b), using this method, have
estimated that roots account for 33--62% of soil CO2 efflux.
Raich and Nadelhoffer (1989) have proposed that soil respiration, together with aboveground litterfall, can be used to
estimate total belowground carbon allocation (root production
+ root respiration) in forest ecosystems by the equation:

edges of the treated buffers of the fertilized and unfertilized
plots was > 15 m. Fertilized plots received 150 kg N ha −1 in
the spring and fall of 1990, 1991 and 1992, and spring 1993.
Nitrogen was applied in equal proportions of NH +4 and NO −3 ,
and other essential macro- and micronutrients were applied in

the ratio 24/6/12/1/1/1.2 (N/P/K/Ca/Mg/S). In October 1990,
one 2.5 × 2.5 m trenched plot was established in the buffer strip
of each plot. To prevent including a tree in the trenched plot,
we selected locations where seedlings had died during early
stand development. The trenches were dug to a depth of least
1.5 m, lined with two layers of 4.5 mil black polyethylene
sheeting, and backfilled. The trenched plots were maintained
free of vegetation for the duration of the project by monthly
weeding.

Rs − Pa ≈ Pb + Rr,

Field measurements
where Rs is the annual soil respiration, Pa is the aboveground
detritus production, Pb is the belowground detritus production,
and Rr is the root respiration. For this case, soil carbon content
is assumed to be at steady state, and leaching losses are assumed to be negligible. Raich and Nadelhoffer (1989) used this
approach to show that belowground carbon allocation is positively correlated to aboveground litterfall, and thus aboveground productivity, for forests of the world.
The objective of our study was to use two independent
methods to estimate total belowground carbon allocation (by

the carbon balance approach) and fine and coarse root productivity in unfertilized and fertilized red pine plantations. The
study was conducted in Wisconsin because Nadelhoffer et al.
(1985), using the N budget technique to estimate fine root
production, found that belowground carbon allocation to fine
roots is positively correlated to nitrogen availability in Wisconsin forests. We reasoned that if several independent methods to
estimate belowground allocation showed a similar effect of
fertilization, the results would clarify the effects of nutrient
availability on carbon allocation patterns in forests.
Methods

Soil respiration
We measured soil respiration monthly from July 1990 to October 1993, except for months when the site was under continuous snowpack, by the soda-lime chamber method (Edwards
1982). To avoid a disturbance effect during the measurement
period, we inserted 28-cm diameter plastic collars into the
ground at least 2 weeks before soil CO2 flux was measured. At
the time of measurement, each collar was replaced with a
28-cm diameter plastic chamber (12.6 l) under which was
placed an uncovered tin containing soda lime. All soil CO2 flux
measurements were made for 24 h. The ratio of the tin surface
area to chamber surface area exceeded the 6% minimum suggested by Raich and Nadelhoffer (1989). Soil CO2 flux was

measured at 10 random locations on each untrenched plot and
at two locations on each trenched plot. Three covered tins were
left on the soil surface of each plot during this same period as
blanks. We calculated actual CO2 absorbance as the difference
Table 1. Physiographic, climatic and structural characteristics of unfertilized and fertilized red pine plantations in northern Wisconsin (as
of 1990). Climate characteristics are from long-term regional weather
station data.

Study site
The study was conducted in a 31-year-old (age in 1990) red
pine (Pinus resinosa Ait.) plantation located approximately 10
km northwest of Boulder Junction, WI (46°10′ N, 89°40′ W)
(Table 1). Seedlings were planted at a 2 × 2 m spacing, and the
stand was never thinned. Soil parent material was glacial
outwash, and the soil was classified as a sandy, mixed, frigid,
entic Haplorthod (USDA 1988). The surface soil texture was a
loamy fine sand. During the winter (December--March), the
soil was usually frozen to a depth of 50 cm (Haynes, unpublished data).
The experiment was based on a completely randomized
block design with a 25 × 25 m fertilized plot and an unfertilized

control plot in each of three blocks. A treated 5-m buffer was
maintained around each plot for the duration of the study,
although all the data reported in this study are based on the 25
× 25 m plots. Within each block, the distance between the outer

(A) Physical and climate characteristics
Elevation (m)
Slope (%)
Average January air temperature (°C)
Average July air temperature (°C)
Growing season precipitation (mm)
Average annual snowfall (mm)
(B) Stand characteristics (± 1 SE)
(from Gower et al. 1993)

500
0--5
−10.4
19.5
586

2660
Unfertilized

Trees.ha −1
2106 ± 229
42.4 ± 2.7
Basal area (m2 ha −1)
Average stem diameter (cm) by canopy class
Dominant
19.1 ± 0.3
Codominant
14.0 ± 0.4
Intermediate + suppressed
8.9 ± 0.4

Fertilized
2016 ± 180
38.8 ± 3.1
19.6 ± 0.4
13.9 ± 0.2

8.4 ± 0.6

BELOWGROUND CARBON ALLOCATION IN RED PINE

between the weight gain for each tin and the mean weight gain
of the three corresponding blanks on the plot, corrected for
water loss on CO2 absorption (Edwards 1982). The soda lime
was replaced after each field season.
Monthly soil respiration rates were calculated by assuming
the measured daily respiration rate was the mean for the
month. Soil respiration rates for April 28, 1993, when soil
temperatures were ~ 0 °C, were multiplied by the time of
continuous snowpack (approximately 5 months each year) to
obtain estimates of winter respiration, and these were added to
the monthly totals to obtain annual estimates of soil respiration.
From June to October 1993, we made direct comparisons of
measurements of soil respiration by the soda-lime and IRGA
methods. At each sampling period (n = 8), we selected four
soda-lime chambers per treatment plot for the comparison. We
placed three 10.15-cm diameter PVC collars 2 cm into the

ground around each selected soda-lime chamber (external collars), and placed one collar inside each chamber (internal
collar). Soil respiration from these collars was measured with
an LI-6200 infrared gas analyzer (Li-Cor Inc., Lincoln, NE)
with a prototype of the LI-6000-09 soil respiration cuvette
following the measurement protocol described by Norman et
al. (1992).
Measurements on the internal collars were taken immediately before and after each 24-h soda-lime absorption period.
Two sets of measurements were taken on the external collars
during each absorption period to account for spatial and temporal variability. To compare the two techniques, we developed
a regression equation relating soda-lime soil respiration measurements to their corresponding IRGA measurements for all
measurement periods.

319

The sampling scheme was based on the fine root phenology of
red pine reported by Aber et al. (1985) and from root periscope
data from this study (Haynes, unpublished data). The corer was
41.5 mm in diameter and 57 cm in length. Cores were kept at
2 °C until processed.
Intact roots (> 1 cm in length) were removed from the cores,
sorted into live and dead roots by visual inspection under a
dissecting microscope, and further sorted into diameter classes
(< 1 mm and 1--5 mm). For the first four sampling periods, root
tips < 1 cm in length were separated from each soil core with
a hydropneumatic root elutriator (Smucker et al. 1982). Organic matter from the elutriator screens was placed in a shallow round water-filled pan (40-cm diameter) and
homogenized. Root tips were removed from one quarter of the
pan area. Visual inspection of the collected tips showed that
nearly all were dead. The five categories of fine roots (> 1 mm
live, < 1 mm live, > 1 mm dead, < 1 mm dead, and dead tips)
from each core were dried at 70 °C and weighed. Root categories were compiled by plot, ground in a Wiley mill, and subsamples were ashed at 425 °C in a muffle furnace to determine
the percent ash content of samples by weight.
Fine root productivity was determined on a plot basis by the
max--min and decision matrix methods (McClaugherty et al.
1982, Publicover and Vogt 1993). Root tip biomass did not
change substantially between any of the first four sampling
periods and was therefore excluded from calculations. Annual
fine root production for 1991 was based on changes in live fine
root biomass between April 1991 and May 1992, and production for 1992 was based on changes in root biomass between
May and October 1992. We assumed that no root growth
occurred during the winter months (November to April or
May).

Soil temperature
A datalogger was used to measure soil temperature at a depth
of 10 cm on Plots 1 (fertilized) and 2 (unfertilized) from
September 1992 to October 1993. During each soil respiration
measurement period, we also measured the soil temperature at
a depth of 10 cm adjacent to each chamber with digital soil
thermometers (Fisher Scientific).
Litterfall
Ten 0.25-m2 litter screens (8.7 cm deep, raised on 10 cm legs)
were randomly placed on each plot in May 1990. Litterfall was
collected monthly from July 1990 to October 1993, except for
periods of continuous snowpack. Litter samples were taken to
the laboratory, sorted into pine foliage and miscellaneous tissue, dried at 70 °C and weighed. Annual litterfall mass was
calculated by summing all monthly collections plus the first
collection of the following year, which consisted of litter shed
the previous year that remained lodged in the canopy. Foliage
litterfall was assumed to be 48% carbon (Raich and Nadelhoffer 1989).
Fine root biomass
Fine root biomass was estimated from cores taken in April and
July 1991 (n = 20), and May, July and October 1992 (n = 10).

Coarse root biomass
In October 1992, we excavated the coarse root systems of
seven unfertilized red pine trees, located less than 100 m from
the study plots, using water pressure from a fire truck. The
harvested trees represented a range of stem diameters (7.5 to
21.7 cm) similar to those found on the plots. Each root system
was taken to the laboratory and separated into five diameter
classes (1.0--1.5, 1.5--2.5, 2.5--5.0, 5.0--10.0 and > 10 cm, or
stump). All root samples were washed and dried at 70 °C to
constant weight. A subsample of each root sample was ground
in a Wiley mill and ashed at 425 °C to determine the ash
content by weight. Regression equations relating coarse root
diameter class biomass and total coarse root biomass to diameter at breast height were developed for the seven excavated
trees, using the REG procedure of SAS (SAS Institute Inc.,
Cary, NC).
Diameters at breast height of all trees on each plot were
measured in April 1990 and March 1993. Total coarse root
biomass of each plot at each time was calculated by applying
the regression equation to each tree on each plot and summing
the biomass values. Annual coarse root productivity for each
plot was determined by subtracting April 1990 coarse root
biomass from March 1993 biomass and dividing by three.

320

HAYNES AND GOWER

Statistical analyses
Main effects of fertilization on soil respiration and root
biomass for each sampling period, as well as annual soil
respiration, root production and litterfall totals, were determined by an analysis of variance on treatment means. Blocking effects were rarely significant at α = 0.05 and rarely
affected the significance of treatment effects. Therefore, all
analyses of treatment effects were made assuming a completely randomized design. Determinations of trenching main
effects on respiration were made using a conservative F-test
with reduced degrees of freedom to account for unequal subsample numbers. Interaction between fertilization treatment
and trenching was determined by analysis of variance on
differences between main and trenched plot means. All analyses were performed with the GLM procedure of SAS. An α =
0.10 level was used to detect statistically significant differences in all analyses.

significantly greater 1--5 mm dead root biomass in July 1991
and May 1992 than unfertilized plots. Fertilized plots also had
significantly greater (P < 0.05) < 1 mm dead root biomass in
May 1992, as well as greater total dead fine root biomass in
April 1991 (P < 0.09) and October 1992 than unfertilized plots.
Fine root net primary production was significantly lower in
fertilized than in unfertilized plots in 1991 by both methods of
calculation (P < 0.03 max--min, P < 0.09 decision matrix), but
the fertilizer effect was not significant in 1992 (Table 2).
Coarse root biomass was positively correlated with stem
diameter for all diameter classes (Table 3). Total coarse root
biomass averaged 2260 g m −2 for fertilized plots and 2460 g
m −2 for unfertilized plots in March 1993 (Table 4). Total coarse
root productivity averaged 60 and 90 g m −2 year −1 for fertilized
and unfertilized plots, respectively, and the treatment effect
was significant.
Soil respiration

Results
Root biomass and production
Total live fine (< 5 mm diameter) root biomass ranged from 59
to 431 g m −2, whereas total intact dead root biomass ranged
from 74 to 141 g m −2 (Figure 1). Fertilized plots had significantly lower < 1 mm and total live fine root biomass and

Excluding the trenched plots, soil respiration rates ranged from
0.097 (April 1993, fertilized treatment) to 0.462 (August 1993,
unfertilized treatment) g CO2 m −2 h −1. Soil respiration was
significantly lower for fertilized plots than for unfertilized
plots for most measurement periods in each growing season
(Figure 2). Soil respiration rates for the trenched plots ranged
from 0.079 (April 1993, unfertilized treatment) to 0.426 (May
1991, fertilized treatment) g CO2 m −2 h −1. Soil respiration rates
did not differ significantly between unfertilized and fertilized
trenched plots (Figure 3); however, soil respiration was significantly lower in trenched plots than in untrenched plots for most
of the 1991 and 1992 growing seasons for both fertilized
Table 2. Mean (± 1 SE) annual fine root production (g m − 2 year − 1). An
asterisk denotes a significant difference between treatments (P <
0.10).
Max--min

Decision matrix

1991
Unfertilized
Fertilized

251 ± 39*
94 ± 26

284 ± 39*
150 ± 44

1992
Unfertilized
Fertilized

180 ± 32
194 ± 32

182 ± 31
208 ± 39

Table 3. Allometric equations for ash-free coarse root component
biomass (n = 7) (log10(biomass) = a + b[log10(dbh)], biomass in
grams, dbh in centimeters).

Figure 1. Mean (n = 3) intact live and dead fine root (< 5 mm diameter)
biomass (± 1 SE) for fertilized and unfertilized plots in 1991 and 1992.
An asterisk denotes a significant difference (P < 0.10).

Component

a

b

SEE

r2

Stump
10.0--5.0 cm
5.0--2.5 cm
2.5--1.5 cm
1.5--1.0 cm
Total

0.454
−3.992
0.054
0.895
0.415
0.821

2.654
5.795
2.706
1.775
1.970
2.636

0.165
1.065
0.241
0.127
0.157
0.139

0.981
0.732
0.962
0.975
0.969
0.986

BELOWGROUND CARBON ALLOCATION IN RED PINE

321

Table 4. Mean (± 1 SE) coarse (> 1.0 cm diameter) root biomass and annual production estimates (n = 3). An asterisk denotes a significant difference
between treatments (P < 0.10).
Treatment

April 1990 coarse root biomass
(g m − 2)

March 1993 coarse root biomass
(g m −2)

Annual coarse root productivity
(g m − 2 year − 1)

Unfertilized
Fertilized

2190 ± 100
2070 ± 130

2460 ± 90
2260 ± 130

90 ± 7*
60 ± 6

Figure 2. Mean untrenched plot soil respiration rates (± 1 SE) for
individual sampling periods, 1990--1993 (n = 3). An asterisk denotes
significantly (P < 0.10) different means.

Figure 3. Mean trenched plot soil respiration rates (± 1 SE) for
individual sampling periods, 1991--1993 (n = 3). An asterisk denotes
significantly (P < 0.10) different means.

(Figure 4) and unfertilized (Figure 5) treatments.
Annual soil respiration rates based on the soda-lime method
ranged from 374 to 541 g C m −2 year −1 (Table 5). Fertilization
significantly (P < 0.05) decreased annual soil respiration in the
untrenched plots in 1992 and 1993, but not 1991. Annual soil
respiration did not differ significantly between trenched unfer-

Figure 4. Mean fertilized plot soil respiration rates (± 1 SE) for
individual sampling periods, 1991--1993 (n = 3). An asterisk denotes
significantly (P < 0.10) different means.

Figure 5. Mean unfertilized plot soil respiration rates (± 1 SE) for
individual sampling periods, 1991--1993 (n = 3). An asterisk denotes
significantly (P < 0.10) different means.

tilized and trenched fertilized plots in any year. Soil respiration
in the fertilized trenched plots was significantly lower than in
the fertilized untrenched plots in 1992 (P < 0.10), whereas soil
respiration in the unfertilized trenched plots was significantly
lower than in the unfertilized untrenched plots in 1991 and
1992 (P < 0.10).
The relationship between the soda-lime and IRGA measure-

322

HAYNES AND GOWER

Table 5. Mean ± 1 SE (n = 3) annual soil respiration (g C m − 2 year − 1)
by treatment. Different lower case letters denote significant differences (P < 0.10) among treatment means for each individual year and
method.

1992, when we hypothesize that advective forces may control
CO2 flux (soil warmer than air), the model is (respiration rate)
= 0.129 × 100.02870(temperature) (r2 = 0.80, P < 0.0001, Q10 =
1.94).

Treatment

Litterfall

Soda lime

IRGA corrected

1991
Unfertilized untrenched
Unfertilized trenched
Fertilized untrenched
Fertilized trenched

541 ± 30 a
430 ± 40 b
478 ± 17 ab
461 ± 26 ab

904 ± 85 a
563 ± 75 b
696 ± 97 a
783 ± 203 ab

1992
Unfertilized untrenched
Unfertilized trenched
Fertilized untrenched
Fertilized trenched

483 ± 20 a
336 ± 39 bc
374 ± 7 b
331 ± 17 c

769 ± 18 a
391 ± 31 bc
451 ± 17 b
369 ± 15 c

1993
Unfertilized untrenched
Unfertilized trenched
Fertilized untrenched
Fertilized trenched

505 ± 20 a
395 ± 56 ab
404 ± 6 b
379 ± 17 b

901 ± 18 a
534 ± 100 bc
505 ± 17 b
442 ± 16 c

ments of soil respiration rates was similar to that reported by
Ewel et al. (1987a) (Figure 6). Using this relationship, we
calculated annual soil respiration corrected for soda-lime bias
(Table 5). The correction factor increased annual soil respiration by 11 to 78%, and made the trenching effect significant
for fertilized (P < 0.06) and unfertilized (P < 0.03) plots in
1993.
Relationship between soil temperature and soil respiration
We found a significant relationship between soil temperature
and soil respiration. The best model was an exponential relationship (respiration rate = 0.158 × 100.02266(temperature), P <
0.0001, Q10 = 1.685) that explained 54% of the variation in soil
respiration. Excluding data for October 1990 and November

Figure 6. Soda-lime soil respiration rates versus the natural logarithm
of simultaneous IRGA measurements. The solid line (ln(IRGA) = 3.94
+ 0.00718(soda lime), r2 = 0.585) represents data from this study. The
dashed line (ln(IRGA) = 3.98 + 0.00653(soda lime)) from Ewel et al.
(1987a).

Mean annual foliage litterfall mass ranged from 150 to 387 g
m −2 (Figure 7). Annual foliage litterfall was significantly
higher in the unfertilized plots in 1990, the first year of fertilization treatment, but foliage litterfall was significantly greater
in the fertilized plots in 1991, 1992 and 1993. Annual miscellaneous litterfall ranged from 68 to 154 g m −2 and was significantly (P < 0.052) greater for fertilized plots than for
unfertilized plots in 1992.
Carbon balance analysis
When we used the carbon balance approach to estimate maximum belowground carbon allocation, we found that total belowground carbon allocation was significantly lower for
fertilized plots than for unfertilized plots for 1991, 1992 and
1993 (Table 6). Total belowground carbon allocation ranged
from 188 to 395 g m −2 year −1. However, when we based our
calculations on the IRGA-corrected respiration rates, the range
was 253 to 791 g C m −2 year −1, and the significant treatment
effect for 1991 was lost.
Discussion
Our maximum estimate of live fine root biomass (431 g m −2)
is comparable to values reported for other red pine forests in
Wisconsin (441 g m −2, Aber et al. 1985) and Massachusetts
(510 g m −2, McClaugherty et al. 1982), and is typical for
cold-temperate conifers. Our estimates of annual fine root
production are also similar to those derived by Nadelhoffer et
al. (1985) (120 g m −2 year −1) and Aber et al. (1985) (200 g m −2
year −1) for red pine plantations in Wisconsin. The estimates
based on the decision matrix are higher than those from the
max--min method (cf. Publicover and Vogt 1993). Although the

Figure 7. Mean (n = 3) annual foliage and miscellaneous litterfall (± 1
SE) for fertilized and unfertilized main plots, 1990--1993. The 1990
totals are partial year, and the 1993 totals are corrected for estimated
winter litterfall.

BELOWGROUND CARBON ALLOCATION IN RED PINE

323

Table 6. Maximum belowground carbon allocation (± 1 SE) (sensu Raich and Nadelhoffer 1989) for unfertilized and fertilized red pine plantations
for 1991, 1992 and 1993. An asterisk denotes a significant difference (P < 0.10) between treatments.
Treatment

1991
Unfertilized
Fertilized
1992
Unfertilized
Fertilized
1993
Unfertilized
Fertilized
1

Soil CO2 flux

Foliage litterfall

Soda lime
(g C m − 2 year − 1)

IRGA corrected
(g C m −2 year −1)

(g C m

541 ± 30
478 ± 17

904 ± 85
696 ± 97

483 ± 20*
374 ± 7
505 ± 20*
404 ± 6

−2

−1

year )

Belowground carbon allocation
Soda lime
(g C m −2 year −1)

IRGA corrected
(g C m − 2 year − 1)

146 ± 3*
162 ± 3

395 ± 33*
317 ± 14

758 ± 89
535 ± 95

769 ± 18*
440 ± 15

151 ± 3*
187 ± 4

332 ± 19*
188 ± 12

618 ± 21*
253 ± 18

901 ± 18*
505 ± 17

110 ± 21*
164 ± 121

395 ± 19*
240 ± 13

791 ± 18*
342 ± 5

Values are projected totals.

treatment difference was significant only in 1991, the estimates corroborated our soil respiration results.
Our coarse root biomass estimate is also similar to those
from conifer stands of similar age and density summarized by
Santantonio et al. (1977); however, few estimates of coarse
root production are available for comparison. The coarse root
net primary production rate (90 g m −2 year −1) of our unfertilized plots is similar to that reported for roots > 5 mm in
diameter of Pinus contorta var. latifolia Engelm. on a mesic
site in British Columbia (Comeau and Kimmins 1989), but
their estimate for a xeric site (40 g m −2 year −1) is significantly
lower. Gower et al. (1992) reported high coarse root production
values (136 to 187 g m −2 year −1) in a New Mexico Douglas-fir
(Pseudotsuga menziesii (Mirb.) Franco) forest, but they also
reported a negative effect of fertilization and irrigation on
coarse root production.
The soda-lime soil respiration rates on our plots (0.097 to
0.462 g CO2 m −2 h −1) are similar to those reported by Bowden
et al. (1993) for a temperate hardwood stand in Massachusetts
and Ewel et al. (1987a) for a warm-temperate Pinus elliottii
Engelm. forest in Florida. Our annual estimates are also well
within the reported range for temperate conifer forests
(Schlesinger 1977, Raich and Nadelhoffer 1989), but annual
estimates for the unfertilized red pine plantation are low compared to the prediction (604 g C m −2 year −1) for our site based
on latitude (Schlesinger 1977). The low annual soil CO2 flux
for our site may be due to the small amount of organic matter
in the sandy soil.
We feel that our correction of annual respiration totals for
winter respiration is valid. Although CO2 diffusion may be
restricted by frozen soil and snowpack, soil microbes and deep
roots may have a basal respiration rate that we did not measure.
Sommerfeld et al. (1993) measured mean respiration rates of
0.031 to 0.114 g CO2 m −2 h −1 from snow in the Rocky Mountains. Our soil respiration measurements for April 1993, when
most of the soil was still frozen < 10 cm deep, are probably
reasonable estimates of soil respiration rates for winter conditions. Because soil respiration during the winter months can
comprise up to 25% of the annual totals, it is unwise to assume
winter soil respiration is negligible.

The relationship between soda-lime and IRGA measurements is similar to that derived by Ewel et al. (1987a), although
the r2 value is lower (0.59 versus 0.80). These results suggest
that the relationship between the soda-lime and IRGA measurements may be applicable under a range of conditions.
Fertilization significantly decreased soil respiration in the
red pine plantation. This response may be explained by one of
two mechanisms. First, several scientists have reported that
fertilization reduces microbial activity (Foster et al. 1980, Fog
1988). However, this explanation does not appear to be valid
for this study, because annual soil surface CO2 flux did not
differ significantly between trenched fertilized and trenched
unfertilized plots. The alternative explanation is that fertilization decreased root and mycorrhizal growth or respiration, or
both. This explanation is consistent with the fine root biomass
data for the unfertilized and fertilized forests in this study.
Comparison of annual respiration from trenched and untrenched plots may provide a first approximation of the proportion of soil respiration that can be attributed to roots and
mycorrhizae. Our soda-lime data suggest that roots and mycorrhizae contributed 21 to 30% of total soil respiration for unfertilized red pine plantations, and 4 to 12% for fertilized red pine
plantations, depending on the year. When corrected for sodalime bias, root contribution to total soil respiration was 38 to
49% for unfertilized plots. The IRGA-corrected values showed
fertilized trenched plots had 13% greater respiration in 1991,
and 17 and 13% lower respiration in 1992 and 1993, respectively. We speculate that the decay of severed roots in the
fertilized trenched plots increased soil respiration in 1991.
Our estimate of root contribution to soil respiration in unfertilized plots for 1992 (30%) is similar to the values of 33% for
a mixed-hardwood stand in Massachusetts (Bowden et al.
1993) also derived using alkali absorption, and 35% for a
mixed-hardwood stand in Tennessee (Edwards and Sollins
1973), derived using a respirometer. However, our IRGA-corrected estimates of the contribution of root respiration in unfertilized plots are closer to the 58--62% determined by Ewel
et al. (1987b) for a slash pine stand in Florida (also IRGA
corrected) and 50% estimated for Japanese red pine (Nakane
et al. 1983). If the IRGA values of soil respiration rates are

324

HAYNES AND GOWER

correct, then we can state that roots generally contribute 40 to
65% of total soil respiration for pine forests.
The slightly higher soil temperature of the unfertilized plots
compared with the fertilized plots may have affected soil
respiration rates (Haynes, unpublished data). To test this, we
applied the derived 10-cm Q10 value (1.685) to observed mean
differences in soil temperature between treatments for each
sampling period in 1993. We found that temperature accounted
for only a 4.9% difference in respiration between the fertilized
and unfertilized plots.
In conclusion, based on two independent methods to estimate belowground carbon allocation in unfertilized and fertilized red pine forests, we found that belowground carbon
allocation was inversely related to nutrient availability (Kurz
1989, Gower et al. 1992). Our estimates of total belowground
carbon allocation by the carbon balance method showed significantly lower belowground carbon allocation in fertilized
red pine forests than in unfertilized red pine forests in 1991,
1992 and 1993. Compared with the unfertilized plots, total soil
respiration was significantly lower in 1992 and 1993, and
foliage litterfall was significantly higher in fertilized plots each
year. Similar results have also been reported for unfertilized
and fertilized P. radiata plantations in Australia (Pongracic
1993). These results conflict with the hypotheses that belowground carbon allocation is positively correlated to nutrient
availability (Nadelhoffer et al. 1985) and aboveground litterfall (Raich and Nadelhoffer 1989). Although these hypotheses
may hold true on a global scale, it appears that within a given
ecosystem, belowground carbon allocation is negatively related to nutrient availability and aboveground productivity.
Although our fine root production estimates did not always
show a significant treatment effect, we always observed a shift
in the proportion of carbon allocated belowground. Aboveground net primary production estimates for the site showed
significantly greater aboveground production for fertilized
plots than for unfertilized plots (568 versus 431 g C m −2 year −1
in 1991) (Gower et al., unpublished observations). Therefore,
even if we assume fine root net primary production did not
differ (as in 1992), a lower proportion of carbon was allocated
belowground.

Acknowledgments
The authors thank Mr. Ralph Hewett and Mr. Phillip Theiler, Wisconsin Department of Natural Resources, Trout Lake, WI, for their help
in locating the study site and for the use of their fire truck. Special
thanks are also extended to Tom Steele, Director of Kemp Natural
Resources Station, Woodruff, WI, for his invaluable assistance with all
phases of this research. We also thank Dr. Erik Nordheim for his
valuable statistics advice. Finally, we thank Daniel Olson, Karin Fassnacht, Charles Brooks, Joseph House and the many other student
workers who made this study possible. This research was supported by
an NSF grant (BSR-8918022) to S.T. Gower, S.W. Running and H.L.
Gholz, and an NSF doctoral dissertation improvement grant (DEB9212873) to B.E. Haynes and S.T. Gower.

References
Aber, J.D., J.M. Melillo, K.J. Nadelhoffer, C.A. McClaugherty and
J. Pastor. 1985. Fine root turnover in forest ecosystems in relation
to quantity and form of nitrogen availability: a comparison of two
methods. Oecologia 66:317--321.
Ahlström, K., H. Persson and I. Börjesson. 1988. Fertilization in a
mature Scots pine (Pinus sylvestris L.) stand----effects on fine roots.
Plant Soil 106:179--190.
Bowden, R.D., K.J. Nadelhoffer, R.D. Boone, J.M. Melillo and J.B.
Garrison. 1993. Contributions of aboveground litter, belowground
litter and root respiration to total soil respiration in a temperate
mixed hardwood forest. Can. J. For. Res. 23:1402--1407.
Carlyle, J.C. and U.B. Than. 1988. Abiotic controls of soil respiration
beneath an eighteen-year-old Pinus radiata stand in south-eastern
Australia. J. Ecol. 76:654--662.
Comeau, P.G. and J.P. Kimmins. 1989. Above- and below-ground
biomass and production of lodgepole pine on sites with differing
soil moisture regimes. Can. J. For. Res. 19:447--454.
Edwards, N.T. 1982. The use of soda-lime for measuring respiration
rates in terrestrial systems. Pedobiologia 23:321--330.
Edwards, N.T. and P. Sollins. 1973. Continuous measurement of
carbon dioxide evolution from partitioned forest floor components.
Ecology 54:406--412.
Ewel, K.C., W.P. Cropper Jr. and H.L. Gholz. 1987a. Soil CO2 evolution in Florida slash pine plantations. I. Changes through time. Can.
J. For. Res. 17:325--329.
Ewel, K.C., W.P. Cropper Jr. and H.L. Gholz. 1987b. Soil CO2 evolution in Florida slash pine plantations. II. Importance of root respiration. Can. J. For. Res. 17:330--333.
Fog, K. 1988. The effect of added nitrogen on the rate of decomposition of organic matter. Biol. Rev. 63:433--462.
Foster, N.W., E.G. Beauchamp and C.T. Corke. 1980. Microbial activity in a Pinus banksiana Lamb. forest floor amended with nitrogen
and carbon. Can. J. Soil Sci. 60:199--209.
Gower, S.T., H.L. Gholz, K. Nakane and V.C. Baldwin. 1994. Production and carbon allocation patterns of pine forests. In Structure and
Productivity of Pine Forests: A Synthesis. Eds. H.L. Gholz, S.
Linder and R. McMurtrie. Ecol. Bull. 43:115--135.
Gower, S.T., B.E. Haynes, K.S. Fassnacht, S.W. Running and E.R.
Hunt Jr. 1993. Influence of fretilization on the allometric relations
for two pines in contrasting environments. Can. J. For. Res. 23:
1704--1711.
Gower, S.T., K.A. Vogt and C.C. Grier. 1992. Carbon dynamics of
Rocky Mountain Douglas-fir: influence of water and nutrient availability. Ecol. Monogr. 62:43--65.
Keyes, M.R. and C.C. Grier. 1981. Above- and belowground net
production in 40-year-old Douglas-fir stands on low and high productivity sites. Can. J. For. Res. 11:599--605.
Kurz, W.A. 1989. Net primary production, production allocation, and
foliage efficiency in second growth Douglas-fir stands with differing site quality. Ph.D. Dissertation. University of British Columbia,
Vancouver, British Columbia, Canada, 224 p.
McClaugherty, C.A., J.D. Aber and J.M. Melillo. 1982. The role of
fine roots in the organic matter and nitrogen budgets of two forested
ecosystems. Ecology 63:1481--1490.
Nadelhoffer, K.J., J.D. Aber and J.M. Melillo. 1985. Fine roots, net
primary production and soil nitrogen availability: a new hypothesis.
Ecology 66:1370--1390.
Nadelhoffer, K.J., and J.W. Raich. 1992. Fine root production estimates and belowground carbon allocation in forest ecosystems.
Ecology 73:1139--1147.

BELOWGROUND CARBON ALLOCATION IN RED PINE
Nakane, K., M. Yamamoto and H. Tsubota. 1983. Estimation of root
respiration rate in a mature forest ecosystem. Jpn. J. Ecol. 33:397-408.
Neill, C. 1992. Comparison of soil coring and in-growth methods for
measuring belowground production. Ecology 73:1918--1921.
Norman, J.M., R. Garcia and S.B. Verma. 1992. Soil surface CO2
fluxes and the carbon budget of a grassland. J. Geophys. Res.
97:18845--18853.
Persson, H. 1983. The distribution and productivity of fine roots in
boreal forests. Plant Soil 71:87--101.
Pongracic, S. 1993. Estimating belowground carbon allocation in
forests. Abstr. Bull. Ecol. Soc. Am. 74:396.
Publicover, D.A. and K.A. Vogt. 1993. A comparison of methods for
estimating forest fine root production with respect to sources of
error. Can. J. For. Res. 23:1179--1186.
Raich, J.W. and K.J. Nadelhoffer. 1989. Belowground carbon allocation in forest ecosystems: global trends. Ecology 70:1346--1354.
Santantonio, D. and R.K. Hermann. 1985. Standing crop, production,
and turnover of fine roots on dry, moderate, and wet sites of mature
Douglas-fir in western Oregon. Ann. Sci. For. 42:113--142.
Santantonio, D., R.K. Hermann and W.S. Overton. 1977. Root
biomass studies in forest ecosystems. Pedobiologia 17:1--31.
Schlesinger, W.H. 1977. Carbon balance in terrestrial detritus. Annu.
Rev. Ecol. Syst. 8:51--81.
Singh, J.S. and S.R. Gupta. 1977. Plant decomposition and soil respiration in terrestrial ecosystems. Bot. Rev. 43:449--528.

325

Smucker, A.J.M., S.L. McBurney and A.K. Srivastava. 1982. Quantitative separation of roots from compacted soil profiles by the
hydropneumatic elutriation system. Agron. J. 74:500--503.
Sommerfeld, R.A., A.R. Mosier and R.C. Musselman. 1993. CO2,
CH4 and N2O flux through a Wyoming snowpack and implications
for global budgets. Nature 361:140--142.
USDA. 1988. Soil survey of Vilas County, Wisconsin. Soil Conservation Service, Madison, WI, 156 p.
Vogt, K.A., C.C. Grier and D.J. Vogt. 1986. Production, turnover and
nutrient dynamics of above- and belowground detritus of world
forests. Adv. Ecol. Res. 15:303--377.
Vogt, K.A., E.E. Moore, D.J. Vogt, M.J. Redlin and R.L. Edmonds.
1983. Conifer fine root and mycorrhizal root biomass within the
forest floor of Douglas-fir stands of different ages and site productivities. Can. J. For. Res. 13:429--437.
Vogt, K.A., D.J. Vogt, E.E. Moore, B.A. Fatuga, M.R. Redlin and R.L.
Edmonds. 1987. Conifer and angiosperm fine-root biomass in relation to stand age and site productivity in Douglas-fir forests. J. Ecol.
75:857--870.
Vogt, K.A., D.J. Vogt, E.E. Moore and D.G. Sprugel. 1989. Methodological considerations in measuring biomass, production, respiration and nutrient resorption for tree roots in natural ecosystems. In
Applications of Continuous and Steady-State Methods to Root
Biology. Eds. J.G. Torrey and L.J. Winship. Kluwer Academic
Publishers, Dordrecht, The Netherlands, pp 217--232.