Download Simulating Picea schrenkiana forest productivity under climatic

Forest Ecology and Management 246 (2007) 273–284
www.elsevier.com/locate/foreco
Simulating Picea schrenkiana forest productivity under climatic
changes and atmospheric CO2 increase in Tianshan Mountains,
Xinjiang Autonomous Region, China
Hongxin Su a, Weiguo Sang a,*, Yunxia Wang b, Keping Ma a
a
Laboratory of Quantitative Vegetation Ecology, Institute of Botany, The Chinese Academy of Sciences, 20 Nanxincun,
Xiangshan, Beijing 100093, PR China
b
China University of Mining and Technology, Beijing 100083, PR China
Received 22 November 2005; received in revised form 8 March 2007; accepted 10 April 2007
Abstract
A process-based model, BIOME-BGC, was used to investigate the response of Picea schrenkiana forest to future climate changes and
atmospheric carbon dioxide (CO2) concentration increases in the Tianshan Mountains of northwestern China. The model was validated by
comparing simulated net primary productivity (NPP) under current climatic conditions with independent field-measured data. Then the model was
used to predict P. schrenkiana forest productivity response to the different climatic and CO2 change scenarios. The results showed that NPP
increased moderately (about 18.6%) for all sites when both temperature and precipitation changes were taken into account. The main factor
contributing to NPP increase was an increase in precipitation, which tended to alleviate moisture stress for P. schrenkiana forest growth. NPP
increase was relatively low (about 2.7%) when the physiological fertilizing effect of the doubled atmospheric CO2 concentration was considered
alone, in part because of nitrogen limitation and low temperatures. When both climatic changes and doubling of atmospheric CO2 concentration
were taken into account, NPP increased more dramatically (from 26.4% to 37.2%). The results also indicated that the interactive effects of climate
and CO2 changes were not a simple additive combination of the individual responses. The results provided support for the view that forest response
to climatic change depends on local site conditions.
# 2007 Elsevier B.V. All rights reserved.
Keywords: Picea schrenkiana forest; Global change; Net primary productivity (NPP); Tree ring width standard index (RWSI); BIOME-BGC
1. Introduction
Since the industrial revolution, atmospheric carbon dioxide
(CO2) concentration has increased markedly, from 280 parts per
million by volume (ppmv) in preindustrial times to 368 ppmv in
2000 (Keeling and Whorf, 2002), and the global mean
temperature increased by 0.6 0.2 8C during the 20th century
(Jones and Moberg, 2003). Those changes have had significant
impacts on the structure and function of terrestrial ecosystems
(Cao and Woodward, 1998; Alley et al., 2003). Now most
climate models predict that during the 21st century, air
temperature will rise by 1.5–4.5 8C and that major changes will
occur in precipitation and cloud patterns (Cubasch et al., 2001).
Forests cover nearly one-third of the earth’s land area and
* Corresponding author. Tel.: +86 10 62836278; fax: +86 10 82599519.
E-mail address: [email protected] (W. Sang).
0378-1127/$ – see front matter # 2007 Elsevier B.V. All rights reserved.
doi:10.1016/j.foreco.2007.04.010
contain up to 80% of the total above-ground terrestrial carbon
and 40% of below-ground carbon (Dixon et al., 1994).
Therefore, understanding how forest ecosystems will respond
to global change in the future is imperative. Kirschbaum and
Fischlin (1996) have suggested that the first response of forests
to climate change and atmospheric CO2 concentration increase
will likely be an increase in productivity. Although there is a
long history of forest net primary productivity (NPP) studies, it
is not yet clear how forest NPP will respond to global changes at
ecosystem scale.
In recent years, numerous models have been developed to
estimate NPP. Regression-based models estimate NPP using
only climate variables such as temperature, precipitation, and
evaporation (Zhou and Zhang, 1996). However, the potential
application of regression models for future projection is
limited, because most of them do not consider the fertilizing
effect of increased atmospheric CO2 concentration and the
possible limitations of nutrient availability (e.g., nitrogen) on
274
H. Su et al. / Forest Ecology and Management 246 (2007) 273–284
productivity (Melillo et al., 1993). Unlike regression-based
models, process-based models combine scaled-up representations of basic plant biology with ecosystem dynamics and
functions to simulate ecosystem-scale processes, including
canopy photosynthesis, transpiration, litter dynamics, changes
in soil moisture regime, and carbon and nutrient dynamics
(VEMAP Members, 1995). Process-based models coupled with
field studies of single plants or of whole-ecosystem NPP can
help quantify the impacts of global changes on forest ecosystem
productivity at various scales (Thornton et al., 2002; Hanson
et al., 2004; Law et al., 2004).
P. schrenkiana (Picea schrenkiana var. tianschanica) forest
is the most dominant and widespread boreal forest type on
the north slopes of the Tianshan Mountains. It is also one of
the most important zonal forms of vegetation in the study
region, representing 60.8% of the timber growing stock and
54.0% of forest land in the Xinjiang area (Zhang and Tang,
1989). The biomass and NPP of P. schrenkiana forest
have been studied over the past two decades (Zhang et al.,
1980; Luo, 1996; Wang and Zhao, 1999). The geographical
patterns of the P. schrenkiana forest have also been analyzed
using data from forest inventories and direct field measurements (Wang and Zhao, 2000; Ni, 2004). Until now, no work
has been carried out to predict P. schrenkiana forest
productivity under future scenarios of climatic change and
CO2 increase.
In this study, a process-based model, BIOME-BGC
(Thornton et al., 2002), was used to assess the effects of
potential climatic changes and atmospheric CO2 concentration
increases on the NPP of P. schrenkiana forest. First, the model
was validated by comparing independent field observed data
with simulated NPP values. Then the BGC model was used to
predict the NPP of P. schrenkiana forests under the different
scenarios of climate change and atmospheric CO2 concentration increase in the future.
2. Materials and methods
2.1. Study sites
The forested zone in the Tianshan Mountains is limited to
altitudes between 1500 and 2700 m where sufficient moisture
and warmth exist. The thermal tree line lies at 2700 m, where
the average temperature during the warmest month is about
10 8C, and the areas below 1500 m are generally too dry to
support boreal forest (Zhang and Tang, 1989). The forest
structure is multilayered, with different types of plants
including trees, shrubs, ferns, grasses, and moss. The canopy
is dominated by P. schrenkiana. Four sites representing
different climatic conditions in the middle elevations on
the northern slopes of the Tianshan Mountains were selected
for this study: west Tianshan = Zhaosu (ZS), central
Tianshan = Tianchi (TC) and Xiaoquzi (XQZ), and east
Tianshan = Yiwu (YW) (Fig. 1). The forests are relatively
undisturbed by either natural or anthropogenic factors
and tend to be quite homogeneous within each site.
General characteristics of the forests and sites are given in
Table 1.
2.2. Field measurements
A total of 37 plots were surveyed in the four sites (Table 1),
each of size 20 m 20 m. Measurements were made of tree
diameter at breast height (DBH, 1.30 m above the ground) and
of height for all trees >4 cm DBH. Smaller trees were included
in the survey as shrubs. Six P. schrenkiana trees in different
diameter classes without visible signs of damage were
randomly selected as sample trees to obtain increment cores
for each plot. Two increment cores for each selected tree were
sampled at 1.30 m above the ground with a Presler increment
borer and stored in paper straws.
Fig. 1. Location of the four studied sites, and the corresponding meteorological stations in the Tianshan Mountains, Xinjiang Autonomous region, China.
H. Su et al. / Forest Ecology and Management 246 (2007) 273–284
275
Table 1
General characteristic of the four study sites with current climate (mean annual temperature and total precipitation)
Sites
No. of
plots
Alt. rang
(m)
Soil
typea
SOMb
(%)
Total
Nb (%)
Density (trees ha1)
DBHc
(cm)
Height (m)
Temperature
(8C)
Precipitation
(mm)
ZS
XQZ
TC
YW
7
9
12
9
2130–2230
1940–2090
1900–2320
2310–2440
MLGS
MTGS
MTGS
MCGS
25.31
23.03
23.03
14.34
0.81
0.75
0.75
0.46
1123(700–2200)
1269(825–1700)
1975(1050–3450)
1583(975–3250)
17.9 1.5
17.1 1.4
15.6 1.0
17.5 1.4
11.0 1.1
13.5 1.0
11.4 0.7
11.9 1.0
2.3 0.1
2.3 0.1
2.1 0.1
1.2 0.1
657.7 17.7
540.1 16.0
533.2 18.9
417.0 11.7
a
b
c
MLGS: mountain leaching gray-cinnamon forest soil, MTGS: mountain typical gray-cinnamon forest soil, MCGS: mountain carbonate gray-cinnamon forest soil.
SOM: soil organic matter, total N: total nitrogen. Only organic matter content and total nitrogen in the top 30 cm were analyzed by Wang and Zhao (2000).
DBH: diameter at breast height. DBH, height, temperature and precipitation are all presented: mean S.E.
2.3. Tree-ring chronology development
were:
Tree cores were mounted on grooved boards and sanded to
facilitate identification of the tree rings. Tree rings were then
crossdated using a skeleton plot (Stokes and Smiley, 1968),
and the ring widths were measured to the nearest 0.001 mm
using the WinDENDROTM 2001b apparatus (Université du
Québec à Chicoutimi, Canada). Crossdating and measurement accuracies were checked and confirmed by correlating
overlapping 50-year segments of all measured series using
the COFECHA software developed by Holmes (1983), and
the cores that could not be crossdated were discarded. Treering chronologies were constructed using conventional
analytical techniques (Fritts, 1976; Cook and Kairiukstis,
1990). To focus on the common growth variations associated
with climate, growth trends were removed from individual
series using conservative detrending (negative exponential
curves or straight lines) (Fritts, 1976). Indices for each
series were derived by taking the ratio of the measurement
over the fitted value in each year. After standardizing
each chronology, we used the ARSTAN software (Cook and
Holmes, 1986) to average the standardized ring widths
and create the standard chronology for the each site.
Descriptive statistics were calculated for each site chronology, including mean sensitivity (MS) and standard deviation
(S.D.) to assess the high-frequency variations (Fritts, 1976),
and the first-order autocorrelation (R1) was used to detect the
interdependence among successive-year radial growth
values.
In this study, it was assumed that in natural forest sites,
annual tree height growth and annual tree diameter growth
are highly correlated (Graumlich et al., 1989; Rathgeber
et al., 2000). Thus, the annual growth of tree rings, which is
sensitive to year-to-year climatic variations, can be used to
assess the annual biomass increment of standing woody
plants to provide a ‘‘productivity measure’’ which is related
to NPP.
W stem ¼ 0:047465712ðD2 HÞ
2.4. Field-based estimation of NPP
Using field survey data, biomass values for tree stems,
branches, and needles were estimated from DBH (cm) and H
(m) following the allometric relationship derived from P.
schrenkiana forests adjacent to the central Tianshan study
sites (Zhang et al., 1980). The allometric equations used
W branch ¼ 0:00189147ðD2 HÞ
0:88217
1:03981
0:78914
W needle ¼ 0:01451443ðD2 HÞ
(1)
(2)
(3)
where Wstem, Wbranch, and Wneedle are the biomass values of tree
stems, branches, and needles, respectively. Stem and branch
NPP were estimated from the mean radial growth increment
over 10 years (1991–2000), tree dimensions, and species (the
last using specific allometric equations on a tree-by-tree basis)
and were summed to the plot level. The arithmetic average
radial increment was estimated from all the increment cores for
each site. The annual net increment of leaf was derived from
leaf biomass divided by leaf age (residence time, which is 5
years for P. schrenkiana). The belowground biomass and NPP
were not measured in the field in this study. Root biomass was
assumed to account for 18.6% of total tree biomass and the root
NPP to be 26.9% of total tree NPP, in accordance with the work
of Wang and Zhao (2000).
Biomass of shrubs and grasses was estimated by using ratios
based on the biomass of trees. The average NPP of the shrub and
grass layers was estimated by dividing their biomass by their
average age (7 years for shrubs and 1 year for grasses in this
study) (Luo, 1996). Litter and biomass losses to herbivores were
not considered. Total forest NPP was estimated by summarizing
annual net increments of tree stems, branches, leaves, and roots
and the annual net increments of shrubs and grasses.
The field-based NPP data were converted to grams of carbon
per square meter per year (g C m2 year1), where 1 g carbon
is equivalent to 2.2 g oven-dry organic matter (Ni, 2003), for
purposes of comparison with model-predicted NPP values.
2.5. Simulation of NPP using BIOME-BGC model
2.5.1. Model description
The BIOME-BGC (Version 4.1.1) model has been described
in detail by Thornton et al. (2002), and therefore only a
summary of important components of the model relating to the
NPP and carbon-cycle processes is briefly presented here. NPP
was computed as the difference between gross photosynthesis
and autotrophic respiration. Gross photosynthesis on a unit
projected leaf area basis for C3 plants was estimated
independently for the sunlit and shaded canopy fractions using
276
H. Su et al. / Forest Ecology and Management 246 (2007) 273–284
Table 2
Ecophysiological parameterization of Picea schrenkiana forest for BIOME-BGC model
Value
Unita
Parameter description
0.3
0.3
0.25
0.70
0.005
0.005
1.0
2.2
0.1
0.3
0.5
42.0
93.0
42.0
50.0
729.0
0.32
0.44
0.24
0.30
0.45
0.25
0.76
0.24
0.041
0.5
2.6
12.0
2.0
0.04
0.003
0.00001
0.08
0.6
2.3
930.0
4100.0
DIM
DIM
1 year1
1 year1
1 year1
1 year1
Ratio
Ratio
Ratio
Ratio
DIM
kg C/kg N
kg C/kg N
kg C/kg N
kg C/kg N
kg C/kg N
DIM
DIM
DIM
DIM
DIM
DIM
DIM
DIM
1/LAI/d
DIM
DIM
m2/kg C
DIM
DIM
m/s
m/s
m/s
MPa
MPa
Pa
Pa
Transfer growth period as fraction of growing season
Litterfall as fraction of growing season
Annual leaf and fine root turnover fraction
Annual live wood turnover fraction
Annual whole-plant mortality fraction
Annual fire mortality fraction
(Allocation) new fine root C: new leaf C
(Allocation) new stem C: new leaf C
(Allocation) new live wood C: new total wood C
(Allocation) new root C: new stem C
(Allocation) current growth proportion
C:N of leaves
C:N of leaf litter, after retranslocation
C:N of fine roots
C:N of live wood
C:N of dead wood
Leaf litter labile proportion
Leaf litter cellulose proportion
Leaf litter lignin proportion
Fine root labile proportion
Fine root cellulose proportion
Fine root lignin proportion
Dead wood cellulose proportion
Dead wood lignin proportion
Canopy water interception coefficient
Canopy light extinction coefficient
All-sided to projected leaf area ratio
Canopy average specific leaf area (projected area basis)
Ratio of shaded SLA:sunlit SLA (specific leaf area)
Fraction of leaf N in Rubisco
Maximum stomatal conductance (projected area basis)
Cuticular conductance (projected area basis)
Boundary layer conductance (projected area basis)
Leaf water potential: start of conductance reduction
Leaf water potential: complete conductance reduction
Vapor pressure deficit: start of conductance reduction
Vapor pressure deficit: complete conductance reduction
a
DIM: dimensionless.
a biochemical model (Farquhar et al., 1980), with kinetic
parameters from Woodrow and Berry (1988) and de Pury and
Farquhar (1997) and substitution from the CO2 diffusion
equation to eliminate the explicit dependency on intracellular
CO2 concentration. The maximum rate of carboxylation was
calculated as a function of the specific activity of the Rubisco
enzyme (itself a function of leaf temperature), the weight
fraction of nitrogen in the Rubisco molecule, the fraction of
total leaf nitrogen in the Rubisco enzyme, the specific leaf area,
and the leaf C:N ratio. Autotrophic respiration was the sum of
maintenance and growth respiration of the different parts of the
plant (canopy, stem, and roots). Maintenance respiration of
each plant component was calculated as a function of tissue
mass, tissue nitrogen concentration, and tissue temperature.
Growth respiration was a simple proportion of total new carbon
allocated to growth. Thus, BIOME-BGC was able to simulate
the effects of a number of abiotic and biotic controls on forest
productivity.
Table 3
Site-specific parameterization of the study sites for BIOME-BGC model
Sites
ZS
XQZ
TC
YW
a
Latitude (8)
43.23
43.49
43.88
43.40
Data from Sun (1995).
Longitude (8)
81.11
87.11
88.12
93.87
Alt. (m)
2007.1
2018.9
1942.5
2395.6
Soil texture a
Soil depth (m)
Sand (%)
Silt (%)
Clay (%)
19.79
26.17
26.17
28.18
66.21
51.67
51.67
50.62
14.00
22.16
22.16
21.20
1.50
0.85
0.85
1.00
H. Su et al. / Forest Ecology and Management 246 (2007) 273–284
Consistency between model results and field measurements is
essential for establishing the credibility of a model such as
BIOME-BGC. Although the logic of the BIOME-BGC model
has undergone widespread testing and validation (Hunt et al.,
1991; Korol et al., 1991; Thornton et al., 2002; Churkina et al.,
2003; Hanson et al., 2004; Law et al., 2004; Schmid et al., 2006.),
the performance of the model was further tested in this research
for P. schrenkiana forests in the Tianshan Mountains by
comparing modeled outputs to independent field observations.
2.5.2. Model parameterization
The BIOME-BGC model requires parameters relating to
vegetation, site condition, and daily climate data to simulate the
NPP of an ecosystem. The model was initialized for the four P.
schrenkiana forest sites with a single set of ecophysiological
parameters (Table 2) and with four site-specific sets of
parameters (Table 3).
The standard daily meteorological input file, including
maximum temperature, minimum temperature, precipitation,
solar radiation, vapor pressure deficit, and day length, was
generated by a computer program for microclimate simulation,
MT-CLIM (Version 4.3, www.forestry.umt.edu/ntsg). Daily
meteorological records (daily air temperature and precipitation
for 1961–2000) were obtained from the China National
Climatic Data Center (NCDC). The XQZ and TC stations
are centrally located in the study region and fall within the
elevation range of the plots under study, so the weather data
could be used without interpolation. MT-CLIM initialized the
flat surface conditions of the two sites to the same elevation as
the original station data to compute the variables not present in
standard weather station records for the BIOME-BGC model.
However, the ZS and YW stations are some distance from the
study sites (about 10 and 25 km, respectively), which may
introduce some uncertainty into the climate data used for the
simulations. Therefore, the daily data were adjusted for site
conditions using MT-CLIM. In addition, the coefficient to
adjust daylight average temperature (TEMCF), which was set
to 0.45 in the original model program, was set to 0.11, 0.12,
0.12, and 0.07 for ZS, TC, XQZ, and YW, respectively,
based on the available data from these stations.
No records of atmospheric CO2 concentrations in the
Tianshan Mountains region are available, so the data for
277
Table 5
General statistics of the standardized chronology for four P. schrenkiana forest
sites
Sites
Samples/trees
Period
MS
S.D.
R1
ZS
XQZ
TC
YW
33/33
26/25
61/32
18/18
1941–2002
1929–2002
1900–2000
1851–2002
0.16
0.22
0.19
0.23
0.15
0.18
0.20
0.22
0.23
0.28
0.07
0.20
MS: mean sensitivity, S.D.: standard deviation, R1: the first-order autocorrelation.
1961–2000 obtained from direct observations at the Mauna Loa
Observatory were used in the simulations (Keeling and Whorf,
2002).
2.5.3. Climate and CO2 scenarios
The future climate scenario was obtained from a secondgeneration regional climate model (RegCM2) downscaled to
the study sites (Gao et al., 2001, 2003). The grid resolution of
this model was approximately 60 km 60 km. A control run of
the RegCM2 model was available which covered a 10-year
period (January 1981 to December 1990), and the doubled CO2
scenario was simulated over 5 years. The simulated climatic
changes for a doubled CO2 scenario were very similar at each of
the four sites. The same climate change scenario was therefore
used for all sites, taking the mean change for each variable
(temperature +2.6 8C and precipitation +25%). To help separate
out the relative importance of doubled CO2 concentration and
climate change, the three primary driving variables, each at two
levels, were used to define eight combinations as climatic
scenarios (Table 4).
2.5.4. Model simulation
Unless initial values for the model’s state variables are
available from measurements, model simulations (startup runs)
are required to initialize them. In the startup run, the model is
run to steady state to obtain the size of the ecosystem’s C and N
pools under the assumption that the ecosystem is in
Table 4
The future climatic scenarios obtained from a second-generation regional
climate model (RegCM2) for the BIOME-BGC run
Climatic
scenariosa
Ambient CO2
concentration
Temperature
Precipitation
C0T0P0
C0T0P1
C0T1P0
C0T1P1
C1T0P0
C1T0P1
C1T1P0
C1T1P1
No change
No change
No change
No change
Doubled
Doubled
Doubled
Doubled
No change
No change
+2.6 8C
+2.6 8C
No change
No change
+2.6 8C
+2.6 8C
No change
+25%
No change
+25%
No change
+25%
No change
+25%
a
The first simulation (C0T0P0) was realized with the current meteorological
data and an atmospheric CO2 concentration fixed at 355 ppmv.
Fig. 2. Comparisons between the simulated and field-based NPP
(g C m2 year1). The bars represent the mean S.E. during the period of
1991–2000.
278
H. Su et al. / Forest Ecology and Management 246 (2007) 273–284
equilibrium; the 40-year climate record was also replicated as a
requirement. Using the endpoint of the startup run as an initial
condition, the model first reconstructed the NPP series for the
last 40 years (using the observed climatic data and atmospheric
CO2 concentrations during the period from 1961 to 2000) to
provide validation. Then, in order to assess the impact of
climatic change and atmospheric CO2 concentration increases,
the model was run with each scenario in Table 4 for each site.
3. Results
3.1. Tree-ring standard chronologies and field-based NPP
Tree-ring chronologies were developed for P. schrenkiana
forest in each site and the general statistics are shown in Table 5.
The higher values of MS and S.D. are exceptional for arid site
forest (e.g., plot YW). Based on field survey data, the mean
Fig. 3. Comparison between the NPP (g C m2 year1) estimated by BIOME-BGC and the tree ring-width standard index (RWSI) for four Picea schrenkiana forest
sites. (a) Chronological comparison between NPP (*) and RWSI (*) from 1961 to 2000. (b) Relationship between NPP and RWSI.
H. Su et al. / Forest Ecology and Management 246 (2007) 273–284
NPP of P. schrenkiana forests was 630.7 g C m2 year1 in ZS,
524.4 g C m2 year1 in XQZ, 446.0 g C m2 year1 in TC,
and 342.4 g C m2 year1 in YW (Fig. 2).
3.2. Model validation
For the climate data and atmospheric CO2 concentrations for
1991–2000, the BIOME-BGC estimations of mean NPP were
close to the mean field-based estimates of NPP for the forests in
ZS and XQZ, and 13.7% and 11.7% higher than the field
estimates for the forests in TC and YW, respectively (Fig. 2).
However, all the model-derived estimates were within the
ranges of statistical validity established from field surveys at
each site.
The NPP series produced by the model were significantly
correlated (R values ranging between 0.35 and 0.54) with
the independent RWSI series, which was used as an index of
279
long-term forest productivity variations under natural environmental conditions at the four sites (Fig. 3). Although the model
showed some bias in the estimation of the RWSI series and
predicted lower-than-observed values in some years (e.g., 1995
in ZS, 2000 in YW), the model matched quite well the year-toyear frequency component of productivity variation, which
corresponds to interannual climatic variations.
3.3. Response of NPP to changes in climate and
atmospheric CO2 concentration
Using the simulated results with current values of 355 ppmv
CO2 (under the C0T0P0 scenario) as the baseline for comparison,
the predicted NPP of the P. schrenkiana forests increased by
14.5% when precipitation increases alone were taken into
account (under the C0T0P1 scenario). Nevertheless, some
differences could be observed among the different sites: the
Fig. 4. Change in site NPP under the different climate and atmospheric CO2 change scenarios. The histograms represent the mean for the years studied:
[(scenariox C0T0P0)/C0T0P0] 100. The bars represent the mean S.E. calculated on the same period.
280
H. Su et al. / Forest Ecology and Management 246 (2007) 273–284
productivity increases ranged from 10.4% (for site ZS) to 17.8%
(for site YW) (Fig. 4). The effects of temperature increases alone
(under the C0T1P0 scenario) were positive for all the studied sites,
and the simulated increase was about 6.4%. The increase is not
uniform and ranges between 1.8% (for site XQZ) to 12.1% (for
site YW). With changes in both temperature and precipitation
(under the C0T1P1 scenario), NPP increased by about 18.6%,
with a range from 13.3% (for site ZS) to 29.2% (for site YW). In
contrast to the results for climate change, the simulated results
indicated that NPP showed relatively little response to doubled
atmospheric CO2 concentration alone (under the C1T0P0
scenario); NPP increased by about 2.7%, and only the XQZ
site showed a significant increase (8.6%). With changes in both
temperature and atmospheric CO2 concentration (under the
C1T1P0 scenario), productivity increased by about 17.9% (with a
range from 13.7% for site TC to 19.8% for site XQZ). Changes in
both precipitation and atmospheric CO2 concentration (under the
C1T0P1 scenario) increased NPP by about 18.6% (with a range
from 12.3% for site ZS to 23.4% for site XQZ). When all climatic
changes and doubled atmospheric CO2 concentration were taken
into account (under the C1T1P1 scenario), NPP increased
dramatically, with an average increase of about 30.4% and a
range from 26.4% (for site ZS) to 37.2% (for site YW) (Fig. 4).
4. Discussion
4.1. Tree-ring standard chronologies
In a dendroclimatological study, it is necessary to select sites
carefully to maximize the potential climate sensitivity within
tree-ring samples, so most field work in this area has been done
in extreme tree-growth environments such as timberlines. Such
areas are near the biotic ecological limits of the tree species that
occur there, where the trees may react more strongly to climatic
changes (Graumlich, 1991). However, the dendrochronological
variable RWSI was used in this study as an estimator of forest
productivity, and all the samples were obtained within the
center of the forest zone, where competition among trees may
have had significant effects on growth. As a result, the MS
(ranging from 0.16 to 0.23) and S.D. (ranging from 0.15 to 0.22)
of the tree-ring standard chronologies were relatively low
(Table 5). Nevertheless, these values agree with previously
published tree-ring standard chronologies for the north slopes
of the Tianshan Mountains (Li, 1989). R1 values ranged from
0.07 to 0.28, suggesting some autocorrelation between the
previous year’s and the current year’s growth rings, mostly
because of tree biology (Berninger et al., 2004; Table 5). MS,
S.D., and R1 for each chronology all suggested that the four
chronologies constructed in this study are well suited for
evaluating the impacts of climatic change and atmospheric CO2
concentration increases on forest productivity.
4.2. Field-based NPP estimates
Field-based NPP estimates obtained in this study agreed
with previously published estimates of productivity in the north
slopes of the Tianshan Mountains. Based on forest inventory
data for 1991–1996, Wang and Zhao (2000) estimated that the
NPP of P. schrenkiana forests was 466.8 g C m2 year1 (with
a range from 246.65 to 863.9 g C m2 year1) in west
Tianshan, and 328.8 g C m2 year1 (with a range from
162.6 to 636.9 g C m2 year1) in central and east Tianshan.
Ni (2004) found that the NPP of coniferous forests was
495.0 215.0 g C m2 year1 (with a range from 110.0 to
915.0 g C m2 year1) using data derived from a national
forest inventory for 1956–1970 at 56 plots in 14 forest sites in
the Tianshan Mountains. Although much work has already been
done, there are still some uncertainties in the NPP estimates
derived from field measurements in this study. The problems
relate to biomass measurements and the methodologies used for
NPP estimation. Developing site-specific allometric equations
for forest trees is laborious and expensive, sometimes
impossible. Therefore, existing allometric equations were used
in this study. Because of the potential for intersite variation in
factors such as tree architecture and wood density (dependent
on site conditions and climate variability), this decision may
have caused some errors in the estimated forest biomass. For
this reason, in any future work, it will be important to test the
allometric equations used by measuring and harvesting some
individual trees on each study site. The measured productivity
was obtained by summing the increments from individual trees
in a unit area. The 10-year radial growth increment was
estimated from all the increment cores for each site. This
approach may overrate the production from old trees. The
magnitude of this error depends largely on the age structure of
P. schrenkiana forest and the composition of the tree-ring
sample.
The results of the field survey also suggested that the
productivity gradually decreased from west to east on the north
slopes of the Tianshan Mountains (Fig. 2). This result agreed
with the work of Sun (1994) and of Wang and Zhao (2000).
These researchers all indicated that two main environmental
gradients control the NPP of P. schrenkiana at a regional scale:
productivity decreased with annual temperature and with
precipitation.
4.3. Model validation
Using site-specific parameters, the BIOME-BGC predictions showed good agreement with the independent field-based
NPP values, being of similar magnitude and presenting similar
temporal dynamic patterns to those of the RWSI at the four sites
(Figs. 2 and 3). This demonstrated the ability of BIOME-BGC
to simulate accurately the P. schrenkiana forest in the Tianshan
Mountains and provided the first step toward a future regionalscale analysis of P. schrenkiana forest NPP or carbon budgets
using this process-based model.
The main discrepancies between field-based NPP estimates
and the simulated results may be associated with the effect of an
inadequate representation of the impact of site age (or
successional stage) on growth of P. schrenkiana forests in
the BIOME-BGC model. Based on the study by Zhang and
Tang (1989), productivity of P. schrenkiana forest rapidly
increases initially with site age and reaches its maximum in
H. Su et al. / Forest Ecology and Management 246 (2007) 273–284
about 100 years. Productivity then gradually decreases with age
until the trees are about 200 years old. This suggests that site
age is a critical factor determining forest ecosystem productivity (Simith and Long, 2001). However, an accurate age
structure of the P. schrenkiana forests in the study areas was not
available, and the model simulations were performed for the
same number of years, equal to the time span of climate records,
with all C and N pools initialized from the steady-state
condition (startup run representing a mature site) in this study.
Because of this simplification, the actual simulated NPP values
may not be representative, though they are comparable with
observed values (Churkina et al., 2003). In any future work, the
site age structure should be taken into consideration when
simulating NPP using the BIOME-BGC model. At the same
time, the P. schrenkiana forests in the four study sites were
assumed in this study to have the same ecophysiological
characteristics. However, this assumption is not necessary true,
because tree physiology has been shown to be site-dependent,
varying with water and nitrogen availability, stand density, and
other environmental variables (White et al., 2000). This is one
more possible reason for the observed discrepancies. With a
view to future regional applications of the model, it would be
important to test this assumption.
The dendrochronological variable RWSI is a good estimator
of tree radial growth, which is related to the annual biomass
increment of stem wood under natural environmental conditions. In this study, the correlations between the observed RWSI
series and the simulated NPP series were not very high.
However, this result is reasonable and acceptable, because it is
difficult to estimate how much of the NPP should be allocated
to stem wood, as this depends on a number of biotic and abiotic
factors (Running and Coughlan, 1988). It is especially
noteworthy that the trends in the simulated NPP were consistent
with the annual fluctuations in the independently observed
RWSI series at all sites. This confirmed that the process-based
model is sensitive to the year-to-year variations in climate
represented in the annual growth rings of trees. This study has
also suggested that unless interannual variables for ecosystem
productivity are available from long-term measurements, the
RWSI can be used as a good estimator of interannual NPP
variations under natural environmental conditions to validate
process-based models for the temporal dynamics of productivity simulations (Korol et al., 1991; Rathgeber et al., 2003). The
very large size of the existing tree-ring database for Xinjiang
offers a wide scope for possible applications (Li et al., 2000).
4.4. Climatic change scenarios
For a doubled CO2 scenario, the RegCM2 model simulates a
significant increase in annual air temperature and precipitation
for the P. schrenkiana forest ecosystems in the Tianshan
Mountains. These simulations agreed well with other atmospheric general circulation model (AGCM) simulations, which
suggested that the annual air temperature will increase by
2.2 8C and that precipitation will increase by 28.0% in Xinjiang
under IPCC95 scenarios (Zhao et al., 2002). However, in this
study, the climate change scenarios do not include seasonal
281
changes. This simplification may cause some uncertainty in the
climate data used for the simulations, but the climatic changes
in most growth seasons and the overall annual changes
remained comparable, and this variability does not alter the
general trends.
4.5. Climate change versus NPP variations
The predicted NPP responses to the scenario with
temperature increase alone (C0T1P0) are positive in this study
(Fig. 4). In the BIOME-BGC model, the specific activity of the
Rubisco enzyme depends on air temperature (Farquhar et al.,
1980). Similarly, litter and soil organic matter (SOM)
decomposition supplies most of the available soil mineral N
for tree growth, which depends on the size of the litter and SOM
pools and their decomposition rates. These rates depend on soil
temperature and soil moisture. Therefore, increased temperature may increase NPP through metabolically enhanced
photosynthesis as well as by increasing nutrient availability
through higher rates of decomposition. Moreover, some
nutrients can be stored to support growth during the following
growing season, as represented by the phenology module in the
BIOME-BGC model. In addition, the increased temperatures
may increase NPP by increasing the length of the growing
season, which strongly correlates with site mean annual
temperature in mid- and high latitudes. On the other hand,
evaporation and transpiration in the BIOME-BGC model are
estimated using the Penman–Monteith equation, and a decrease
in soil moisture is caused by increased evaporative demand
because of higher temperatures. Thus increased temperatures
may also decrease NPP by decreasing water availability and
enhancing plant respiration (see the earlier description of the
BIOME-BGC model). In this study, the results suggest that the
positive effects of the temperature increase more than
compensate for the negative effects on the NPP of the P.
schrenkiana forest in the Tianshan Mountains region.
An increase in precipitation tends to alleviate the effects of
moisture stress on P. schrenkiana forest growth and hence has
positive effects on the NPP (Fig. 4). This result agreed with the
studies of water-limited regions by Nemani et al. (2003) and
Mohamed et al. (2004) and also showed that the direct effect of
precipitation increase has a greater influence on forest
productivity than do temperature changes (Fig. 4). This
indicates that water availability may be the main limiting
factor for NPP at the current atmospheric CO2 concentration.
This result agreed with the evidence from P. schrenkiana tree
rings in this study region (Yuan and Li, 1995; Yuan et al., 2001;
Zhu et al., 2004).
4.6. Atmospheric CO2 concentration increase versus NPP
variations
In the BIOME-BGC model, the responses of intercellular
CO2 and leaf nitrogen concentration to a doubling of CO2
concentration are included. Intercellular CO2 increases with a
rise in atmospheric CO2 because these two quantities are
linearly related in the model. Similarly, the BIOME-BGC
282
H. Su et al. / Forest Ecology and Management 246 (2007) 273–284
model represents the observation that increased CO2 reduces
leaf nitrogen concentration by prescribing a 40% reduction of
leaf nitrogen in conifers for a doubling of atmospheric CO2.
Thus, carbon assimilation increases with increasing intercellular CO2 and decreases with decreasing leaf nitrogen
concentration. Except at the XQZ site, a doubling of
atmospheric CO2 concentration had no significant effect on
NPP (Fig. 4). This result agreed with the study by Melillo et al.
(1993), which found that increasing simulated atmospheric
CO2 levels from 312.5 to 625.0 ppmv did not change the NPP of
the boreal woodland and boreal forest as predicted by the
terrestrial ecosystem model (TEM) (Raich et al., 1991).
Compared with published CO2 enrichment experiments, the
simulated NPP increase predicted by the BIOME-BGC model
in this study was substantially low. Curtis and Wang (1998)
carried out a meta-analysis on 102 measurements of tree
biomass response to elevated CO2 and found a 29% increase.
Similarly, Ainsworth and Long (2005) reported a 28% increase
in aboveground dry matter production for trees grown under
elevated CO2 concentration, based on the data collected from
120 primary, peer-reviewed articles describing tree physiology
and production in the 12 large-scale free-air-CO2-enrichment
(FACE) experiments (with simulated CO2 concentrations from
475 to 600 ppmv). So the question arises: are the representations of CO2 effects in the model used here reasonable, and why
do they not agree with what is known from direct observations
and experimentation? The response of forest productivity to
CO2 enrichment is local, and there have been no contemporary
experiments or observations on the P. schrenkiana forests, so it
is difficult to provide a direct answer to this question. The large
difference between the results of the two kinds of studies might
be due, in part, to the BIOME-BGC model prediction that the P.
schrenkiana forests, which are boreal ecosystems, have a low
capacity to incorporate elevated CO2 concentrations into
production because of nitrogen limitation and low temperatures
(Melillo et al., 1993). Most of the experimental studies were
performed under fertile and warm conditions, and trees are
assumed to react more strongly in these environments.
Furthermore, most of the experimental studies were shortterm, and the reported responses of photosynthesis and growth
to environmental factors do not fully predict the true responses
of plant productivity (Lloyd and Farquhar, 1996) because the
long-term changes in ecosystem properties in response to CO2
enrichment are largely regulated by indirect, interactive, and
feedback effects (Luo et al., 2004).
4.7. Climate change and CO2 increase versus NPP
variations
As productivity in boreal forest ecosystems is substantially
limited by low temperatures and N availability, P. schrenkiana
forest NPP in Tianshan region increased dramatically (by 26.4–
37.2%) under the C1T1P1 scenario (Fig. 4). By comparison,
Zhou and Zhang (1996) suggested that the NPP of the boreal
coniferous forest zone in China would increase by 26.9% under
conditions of doubled CO2 concentration and associated
climate changes (temperature +2 8C and precipitation
+20%). Using the TEM (Version 4.0) for climate change
simulation at 519 ppmv CO2, Xiao et al. (1998) showed that the
annual NPP of boreal forest in China would increase by 22.2%
for the Goddard Institute for Space Studies (GISS) climate
scenario, 21.0% for the Oregon State University (OSU)
scenario, and 23.9% for the Geophysical Fluid Dynamic
Laboratory (GFDL-q) scenario. These studies together with the
present study all indicate that the projected global climate
change would increase boreal forest NPP in China and that the
forest region is likely to become a major potential carbon sink.
However, a more detailed and accurate calculation of carbon
storage and flux in these boreal forests has not been performed
here.
Comparison of the results of various scenarios in this study
indicated that the effects of precipitation and temperature
change were simply additive, but that the synergy between the
effects of climate change and doubled CO2 was important,
since the feedback loops associated with the water and nitrogen
cycles ultimately influence the carbon assimilation response.
The simulated results presented here are consistent with the
work of Shaw et al. (2002), which suggested that ecosystem
responses to realistic combinations of global changes are not
necessarily simple combinations of the responses to the
individual factors.
4.8. Site location versus NPP variations
An important aspect of analyses of ecosystem responses to
global climatic change is dependency on local conditions
(Shaver et al., 2000). In this study, the effects of doubled
atmospheric CO2 concentration on NPP were higher at the QXZ
site than at the YW site, with intermediate values at the ZS and
TC sites (Fig. 4). It is suggested that the effects of CO2
fertilization on productivity may be dependent on climate at the
study site. Under the C0T1P1 scenario, the NPP response is
lowest for the westernmost site and highest for the easternmost
site (Fig. 4). It is noteworthy that the site with the lowest
productivity increases, ZS, was the warmest and wettest site,
and hence the least likely to benefit from climate change, while
the site with the harshest conditions (YW) experienced the
greatest productivity increases under the C0T1P1 scenario
(Fig. 4). These results are in agreement with Xiao et al. (1998),
who suggested that the NPP would show a large increase in dry
and cool regions and only a slight increase in humid and warm
regions, in their study of NPP change of various vegetation
types in China under conditions of climate change using
different AGCMs. A wider variety of sites and climate types
should therefore be taken into account in future studies to make
a proper assessment of how ecosystems respond to global
changes at larger scales.
5. Conclusions
The predictions of the BIOME-BGC model are in reasonably close agreement with the magnitude and the temporal
dynamics of the independent productivity variables in four P.
schrenkiana forests on the north slopes of the Tianshan
H. Su et al. / Forest Ecology and Management 246 (2007) 273–284
Mountains. Therefore, the model can be used in future studies
as a diagnostic tool to explain changes in forest growth as a
result of global changes.
For the study sites, the model simulations showed that the
effects of climate change (increase in NPP from 13.3% to
29.1% under the C0T1P1 scenario) are more significant than the
effect of doubled atmospheric CO2 concentration on the NPP of
the P. schrenkiana forests in the Tianshan Mountains. Climatic
changes and atmospheric CO2 concentration increase together
led to a marked increase in NPP (from 26.4% to 37.2% under
the C1T1P1 scenario), but with variations in magnitude from
place to place. The simulations also indicated that climate
change and elevated atmospheric CO2 concentration had
strongly interactive effects on NPP. These model simulations of
various scenarios provide valuable predictions that can be
helpful in understanding how ecosystems respond to simultaneous increases in atmospheric CO2 concentration and climate
changes.
Experimental investigations (Van Breemen et al., 1998) and
model-based simulations (Ni, 2002) have strongly indicated
that boreal forests would change significantly, both in
geographical range and in their biogeochemical processes,
under the influence of global warming and atmospheric CO2
concentration enrichment. Future studies using BIOME-BGC
will evaluate the sensitivity of NPP to vegetation redistribution
as a part of investigations into the long-term effects of elevated
atmospheric CO2 concentrations and associated climate
changes on forest ecosystems. In addition, given the focus of
this study on the effects of climate change and atmospheric CO2
concentration on NPP, NPP variability associated with other
environmental factors such as N, O3, and UVB was not
investigated. These and similar topics should be the subject of
further research.
Acknowledgements
This study was supported by grants from the Chinese
Ecosystem Research Network Foundation, Chinese Academy
of Sciences, to the Beijing Forest Ecosystem Research Station,
and from the National Natural Science Foundation of China
grant No. 90102009 to W. Sang. The BIOME-BGC (Version
4.1.1) and MT-CLIM software packages were provided by the
Numerical Terradynamic Simulation Group (NTSG) at the
University of Montana. We are grateful to Dr. O. Sun and Z. Lu
for their help in preparing the manuscript, and to two
anonymous reviewers who provided comments and suggestions
about this paper as well as much constructive advice.
References
Ainsworth, E.A., Long, S.P., 2005. What have we learned from 15 years of freeair CO2 enrichment (FACE)? A meta-analytic review of the responses of
photosynthesis, canopy properties and plant production to rising CO2. New
Phytol. 165, 351–372.
Alley, R.B., Marotzke, J., Nordhaus, W.D., Overpeck, J.T., Peteet, D.M., Pielke
Jr., R.A., Pierrehumbert, R.T., Rhines, P.B., Stocker, T.F., Talley, L.D.,
Wallace, J.M., 2003. Abrupt climate change. Science 299, 2005–2010.
283
Berninger, F., Hari, P., Nikinmaa, E., Lindholm, M., Meriläinen, J., 2004. Use of
modeled photosynthesis and decomposition to describe tree growth at the
northern treeline. Tree Physiol. 24, 193–204.
Cao, M., Woodward, F.I., 1998. Dynamic responses of terrestrial ecosystem
carbon on global climate change. Nature 393, 249–251.
Churkina, G., Tenhunen, J., Thornton, P., Falge, E.M., Elbers, J.A., Erhard, M.,
Grünwald, T., Kowalski, A.S., Rannik, Ü, Sprinz, D., 2003. Analyzing the
ecosystem carbon dynamics of four European coniferous forests using a
biogeochemistry model. Ecosystems 6, 168–184.
Cook, E.R., Kairiukstis, L.A., 1990. Methods of Dendrochronology: Applications in the Environmental Sciences. Kluwer Academic Publishers, Dordrecht, The Netherlands.
Cook, E.R., Holmes, R.L., 1986. Users Manual for Program ARSTAN.
Laboratory of Tree-Ring Research. University of Arizona, Tucson, USA.
Cubasch, U., Meehl, G.A., Boer, G.J., Stouffer, R.J., Dix, M., Noda, A.,
Senior, C.A., Raper, S., Yap, K.S., 2001. Projections of future climate
change. In: Houghton, J.T., Ding, Y., Griggs, D.J., Noguer, M., van der
Linden, P.J., Dai, X., Maskell, K., Johnson, C.A. (Eds.), Climate Change
2001: The Scientific Basis, Cambridge University Press, Cambridge UK,
pp. 527–582.
Curtis, P.S., Wang, X., 1998. A meta-analysis of elevated CO2 effects on woody
plant mass, form, and physiology. Oecologia 113, 299–313.
de Pury, D.G.G., Farquhar, G.D., 1997. Simple scaling of photosynthesis from
leaves to canopies without the errors of big-leaf models. Plant Cell Environ.
20, 537–557.
Dixon, R.K., Brown, S., Houghton, R.A., Solomon, A.M., Trexler, M.C.,
Wisniewski, J., 1994. Carbon pools and flux of global forest ecosystems.
Science 263, 185–190.
Farquhar, G.D., Caemmerer, S.V., Berry, J.A., 1980. A biochemical model of
photosynthetic CO2 assimilation in leaves of C3 plants. Planta 149, 78–80.
Fritts, H.C., 1976. Tree Rings and Climate. Academic Press, London.
Gao, X., Zhao, Z., Ding, Y., Huang, R., Filippo, G., 2001. Climate change due to
greenhouse effects in China as simulated by a regional climate model. Adv.
Atmos. Sci. 18 (6), 1224–1230.
Gao, X., Zhao, Z., Ding, Y., 2003. Climate change due to greenhouse effects in
northwest China as simulated by a regional climate model. J. Glaciol.
Geocryol. 25 (2), 165–169 (in Chinese, with English abstract).
Graumlich, L.J., 1991. Subalpine tree growth, climate, and increasing CO2: an
assessment of recent growth trends. Ecology 72, 1–11.
Graumlich, L.J., Brubaker, L.B., Grier, C.C., 1989. Long-term trends in forest
net primary productivity: Cascade Mountains, Washington. Ecology 70,
405–410.
Hanson, P.J., Amthor, J.S., Wullschleger, S.D., Wilson, K.B., Grant, R.F.,
Hartley, A., Hui, D., Hunt Jr., E.R., Johnson, D.W., Kimball, J.S., King,
A.W., Luo, Y., McNulty, S.G., Sun, G., Thornton, P.E., Wang, S., Williams,
M., Baldocchi, D.D., Cushman, R.M., 2004. Oak forest carbon and water
simulations: model intercomparisons and evaluations against independent
data. Ecol. Monogr. 74, 443–489.
Holmes, R.L., 1983. Computer-assisted quality control in tree-ring research.
Tree-Ring Bull. 43, 69–75.
Hunt Jr., E.R., Martin, F.C., Running, S.W., 1991. Simulating the effects of
climatic variation on stem carbon accumulation of a Ponderosa pine stand:
comparison to annual growth increment data. Tree Physiol. 9, 161–171.
Jones, P.D., Moberg, A., 2003. Hemispheric and large-scale surface air temperature variations: an extensive revision and an update to 2001. J. Climate
16, 206–223.
Keeling, C.D., Whorf, T.P., 2002. Atmospheric CO2 records from Sites in the
SIO Air Sampling Network/Trends: A Compendium of Data on Global
Change. Carbon Dioxide Information Analysis Center, Oak Ridge National
Laboratory, U.S. Department of Energy, Oak Ridge, Tennessee.
Kirschbaum, M.U.F., Fischlin, A., 1996. Climate change impacts on forest. In:
Watson, R.T., Zinyowera, M.C., Moss, R.H., Dokken, D.J. (Eds.), Climate
Change 1995. Impacts, Adaptations and Mitigation of Climate Change:
Scientific-Technical Analyses. Cambridge. University Press, Cambridge,
UK, pp. 95–130.
Korol, R.L., Running, S.W., Milner, K.S., Hunt Jr., E.R., 1991. Testing a
mechanistic carbon balance model against observed tree growth. Can. J.
Forest Res. 21, 1098–1105.
284
H. Su et al. / Forest Ecology and Management 246 (2007) 273–284
Law, B.E., Turner, D., Campbell, J., Sun, O.J., Van Tuyl, S., Ritts, W.D., Cohen,
W.B., 2004. Disturbance and climate effects on carbon stocks and fluxes
across Western Oregon. U.S.A. Global Change Biol. 10, 1429–1444.
Li, J., 1989. Study of Dendroclimatology and Dendrohydrology in Xinjiang.
Meteorological Press, Beijing, China (in Chinese).
Li, J., Yuan, Y., You, X., 2000. Study of Tree-Ring Hydrology and its
Applications. Science Press, Beijing China (in Chinese).
Lloyd, J., Farquhar, G.D., 1996. The CO2 dependence of photosynthesis, plant
growth responses to elevated atmospheric CO2 concentrations and their
interaction with soil nutrient status. 1. General principles and forest
ecosystems. Funct. Ecol. 10, 4–32.
Luo, T., 1996. Patterns of net primary productivity for chinese major forest
types and their mathematical models. PhD Dissertation, Commission for
Comprehensive Survey of Natural Resources, Chinese Academy of
Sciences, Beijing, China, pp. 34–60 (in Chinese, with English abstract).
Luo, Y., Su, B., Currie, W.S., Dukes, J., Finzi, A., Hartwig, U., Hungate, B.,
McMurtrie, R., Parton, W.J., Pataki, D., Shaw, R., Zak, D.R., FieldF C.B.,
2004. Progressive nitrogen limitation of ecosystem responses to rising
atmospheric CO2 concentration. Bioscience 54, 731–739.
Melillo, J.M., McGuire, A.D., Kicklighter, D.W., Moore III, B., Vorosmarty,
C.J., Schloss, A.L., 1993. Global climate change and terrestrial net primary
production. Nature 363, 234–240.
Mohamed, M.A.A., Babiker, I.S., Chen, Z., Ikeda, K., Ohta, K., Kato, K., 2004.
The role of climate variability in the inter-annual variation of terrestrial net
primary production (NPP). Sci. Total Environ. 332, 123–137.
Nemani, R.R., Keeling, C.D., Hashimoto, H., Jolly, W.W., Piper, S.C., Tucker,
C.J., Myneni, R.B., Running, S.W., 2003. Climate-driven increases in global
terrestrial net primary production from 1982 to 1999. Science 300, 1560–
1563.
Ni, J., 2002. Effects of climate change on carbon storage in boreal forests of
China: a local perspective. Climat. Change 55, 61–75.
Ni, J., 2003. Net primary productivity in forests of China: a scaling-up of
national inventory data and comparison with model predictions. For. Ecol.
Manage. 176, 485–495.
Ni, J., 2004. Forest productivity of the Altay and Tianshan Mountains in the dry
land, northwestern China. For. Ecol. Manage. 202, 13–22.
Raich, J.W., Rastetter, E.B., Melillo, J.M., Kicklighter, D.W., Steudler, P.A.,
Peterson, B.J., Grace, A.L., Moore III, B., Vorosmarty, C.J., 1991. Potential
net primary productivity in South America: application of a global model.
Ecol. Appl. 1, 399–429.
Rathgeber, C., Nicault, A., Guiot, J., Keller, T., Guibal, F., Roche, P., 2000.
Simulated responses of Pinus halepensis forest productivity to climatic
change and CO2 increase using a statistical model. Global Planet. Change
26, 405–421.
Rathgeber, C., Nicault, A., Kaplan, J.O., Guiot, J., 2003. Using a biogeochemistry model in simulating forests productivity responses to climatic change
and [CO2] increase: example of Pinus halepensis in Provence (South-east
France). Ecol. Model. 166, 239–255.
Running, S.W., Coughlan, J.C., 1988. A general model of forest ecosystem
processes for regional applications. I. Hydrologic balance, canopy gas
exchange and primary production processes. Ecol. Model. 42, 125–154.
Schmid, S., Zierl, B., Bugmann, H., 2006. Analyzing the carbon dynamics of
central European forests: comparison of Biome-BGC simulations with
measurements. Reg. Environ. Change 6, 167–180.
Shaver, G.R., Canadell, J., Chapin, F.S., Gurevitch, J., Harte, J., Henry, G.,
Ineson, P., Jonasson, S., Melillo, J., Pitelka, L., Rustad, L., 2000. Global
warming and terrestrial ecosystems: a conceptual framework for analysis.
Bioscience 50, 871–881.
Shaw, M.R., Zavaleta, E.S., Chiariello, N.R., Cleland, E.E., Mooney, H.A.,
Field, C.B., 2002. Grassland responses to global environmental changes
suppressed by elevated CO2. Science 298, 1987–1990.
Simith, F.W., Long, J.N., 2001. Age-related decline in forest growth: an
emergent property. For. Ecol. Manage. 144, 175–181.
Stokes, M.A., Smiley, T.L., 1968. An Introduction to Tree Ring Dating.
University of Chicago Press, Chicago.
Sun, J., 1994. The productivity of forest stand and the distribution regularity of
forest types and in Tianshan forest. Arid Zone Res. 11 (1), 1–6 (in Chinese,
with English abstract).
Sun, J., 1995. Soil of the Tianshan Mountain northern aspect. The Forestry
Department of Xinjiang Uygur Autonomous Region, Forest Soils of Xinjiang Mountain Region. Xinjiang Science Technology and Hygiene Publishing House, Xinjiang, China, pp. 152–212 (in Chinese).
Thornton, P.E., Law, B.E., Gholz, H.L., Clark, K.L., Falge, E., Ellsworth, D.E.,
Goldstein, A.H., Monson, R.H., Hollinger, D.Y., Falk, M., Chen, J., Sparks,
J.P., 2002. Modeling and measuring the effects of disturbance history and
climate on carbon and water budgets in evergreen needleleaf forests. Agric.
For. Meteorol. 113, 185–222.
Van Breemen, N., Jenkins, A., Wright, R.F., Beerling, D.J., Arp, W.J., Berendse,
F., Beier, C., Collins, R., Van Dam, D., Rasmussen, L., Verburg, P.S.J.,
Wills, M.A., 1998. Impacts of elevated carbon dioxide and temperature on a
boreal forest ecosystem (CLIMEX Project). Ecosystems 1, 345–351.
VEMAP Members, 1995. VEMAP: a comparison of biogeography and biogeochemistry models in the context of global climate change. Global
Biogeochem. Cy. 9, 407–437.
Wang, Y., Zhao, S., 1999. A study on biomass and net productivity of Picea
schrenkiana var. tianshanica forest. Chin. J. Appl. Ecol. 10 (4), 389–391 (in
Chinese, with English abstract).
Wang, Y., Zhao, S., 2000. Productivity pattern of Picea schrenkiana var.
tianshanica forest. Acta Phytoecol. Sin. 24, 186–190 (in Chinese, with
English abstract).
White, M.A., Thornton, P.E., Running, S.W., Nemani, R.R., 2000. Parameterization and sensitivity analysis of the Biome-BGC terrestrial ecosystem
model: net primary production controls. Earth Interact. 4, 1–85.
Woodrow, I.E., Berry, J.A., 1988. Enzymatic regulation of photosynthetic
CO2 fixation in C3 plants. Ann. Rev. Plant Physiol. Plant Mol. Biol. 39,
533–594.
Xiao, X., Melillo, J.M., Kicklighter, D.W., Pan, Y., McGuire, A.D., Helfrich, J.,
1998. Net primary production of terrestrial ecosystems in China and its
equilibrium responses to changes in climate and atmospheric CO2 concentration. Acta Phytoecol. Sin. 22 (2), 97–118.
Yuan, Y., Li, J., 1995. The transfer functions of tree-ring chronologies in
western Tianshan Mountains. J. Glaciol. Geocryol. 17 (2), 170–177 (in
Chinese, with English abstract).
Yuan, Y., Li, J., Hu, R., Liu, C., Jiao, K., Li, Z., 2001. Reconstruction of
precipitation in the recent 350a from tree-rings in the middle Tianshan
Mountains. J. Glaciol. Geocryol. 23 (1), 34–40 (in Chinese, with English
abstract).
Zhang, Y., Tang, G., 1989. Picea schrenkiana forest. Editorial Committee of
Xinjiang Forest. Xinjiang Forest, vols. 121–149. Xinjiang People’s Press/
Chinese Forestry Press, Urumchi, China/Beijing, China (in Chinese).
Zhang, Y., Wang, X., Zhou, L., 1980. Primary study on biomass of Picea
schrenkiana. J. Xinjiang August lst Agric. College 3, 19–25 (in Chinese,
with English abstract).
Zhao, C., Gao, X., Tang, M., Xu, Y., 2002. Prediction of climate change. In:
Ding, Y. (Ed.), Prediction of Environment Changes in Western China.
Science Press, Beijing, China, (in Chinese), pp. 16–47.
Zhou, G., Zhang, X., 1996. Study on NPP of natural vegetation in China under
global climate change. Acta Phytoecol. Sin. 20, 11–19 (in Chinese, with
English abstract).
Zhu, H., Wang, L., Shao, X., Fang, X., 2004. Tree ring-width response of Picea
schrenkiana to climate change. Acta Geograph. Sin. 59 (6), 863–870 (in
Chinese, with English abstract).