全 文 ：Received 20 May 2004 Accepted 10 Aug. 2004
Supported by Beijing Forest Ecosystem Research Station and the National Natural Science Foundation of China (90102009).
* Author for correspondence. Tel (Fax): +86 (0)10 82599519; E-mail: .
http://www.chineseplantscience.com
Acta Botanica Sinica
植 物 学 报 2004, 46 (11): 1281-1291
Simulations and Analysis of Net Primary Productivity in Quercus
liaotungensis Forest of Donglingshan Mountain Range in
Response to Different Climate Change Scenarios
SU Hong-Xin, SANG Wei-Guo*
(Institute of Botany, The Chinese Academy of Sciences, Beijing 100093, China)
Abstract: An ecosystem process model, BIOME-BGC, was used to explore the sensitivity of net primary
productivity (NPP) of an oak (Quercus liaotungensis Koidz) forest ecosystem in Beijing area to global
climate changes caused by increasing atmospheric CO2 concentrations. Firstly we tested the model, and
validated the modeled outputs using observational data; the outputs of BIOME-BGC model were consistent
with observed soil water content and annual NPP. Secondly the potential impacts of climate change on the
oak forest ecosystem were predicted with BIOME-BGC model. We found that the simulated NPP was much
more sensitive to a 20% precipitation increase or a doubling of atmospheric CO2 from 355 to 710 µmol/mol
than to a 2 ℃ temperature increase. Our results also indicated that the effects of elevated CO2 and
climate change on the response of NPP were not interactive.
Key words: BIOME-BGC model; climate change; oak forest; net primary productivity (NPP)
Net primary productivity (NPP) is a key ecosystem vari-
able and an important component of the global carbon cycle.
It plays a key role in our understanding of carbon exchange
between biota and atmosphere, both currently and under
climate change conditions caused by the human-induced
increase in atmospheric CO2 concentration (Woodward
et al., 1995; Melillo et al., 1996). The practical importance
of NPP is in its utility as a measure of crop yield and forest
production (Milner et al., 1996), as well as other economi-
cally and socially significant products of vegetation growth.
Many models have been developed to study the responses
in terms of primary productivity (Lieth, 1975; Melillo et al.,
1993; Field et al., 1995; VEMAP Members, 1995; Friend et
al., 1997; Zheng et al., 1997; Bondeau et al., 1999; Cramer
et al., 1999). Among them, the biogeochemical models (such
as BIOME-BGC (BioGeochemical Cycles), CENTURY, and
Terrestrial Ecosystem Model (TEM)) address the problems
related to the response of primary productivity in terms of
detailed eco-physiological mechanisms, which tend to sim-
plify the structural aspect of responses (Parton et al., 1988;
1989; Running and Coughlan, 1988; Melillo et al., 1993;
Running and Hunt, 1993).
Human actions affect many aspects of the Earth system.
The composition of the atmosphere, climate, abundance of
invasive species, and areas of managed landscapes have
all undergone significant changes (Vitousek et al., 1997).
Globally, the atmospheric CO2 concentration increased by
more than 30% in the last century; temperature,
precipitation, and deposition of biologically available ni-
trogen (N) also increased in large regions (IPCC, 2001). Fur-
ther increases of them in the future are almost certain. The
elevated atmospheric CO2 and climate changes can lead to
changes in terrestrial ecosystems. Nemani et al. (2003) sug-
gested that global changes in climate eased several critical
climatic constraints to plant growth, such that NPP in-
creased 6% (3.4 petagrams of carbon (C) over 18 yr) glo-
bally from 1982 to 1999. Gao et al. (2003) improved the re-
gional dynamic vegetation model by Yu et al. (2002) to
include land use as a constraint to spatiotemporal vegeta-
tion dynamics and applied to the North-South transect of
eastern China (NSTEC) to examine the combined effects of
climate and land use on vegetation distribution, primary
production and N cycling. The simulation result indicated
that the effects of doubling CO2 and climatic changes on
NSTEC were to produce an increased NPP at equilibrium
for all seven general circulation models (GCM) scenarios.
The biogeochemical models can be used to simulate
effects on NPP under different climate scenarios. In this
study, we evaluated the feasibility of a biogeochemical
model, BIOME-BGC model (version 4.1.1), by comparing
field measurements with the modeled outputs under cur-
rent climate conditions (1993-2001), and simulated the re-
sponses of NPP to the different climatic change scenarios
projected by GCMs on Quercus liaotungensis forest, one
of the typical temperate forests in the mountainous areas
of Beijing.
Acta Botanica Sinica 植物学报 Vol.46 No.11 20041282
1 Methods
1.1 BIOME-BGC model
BIOME-BGC model, a multi-biome generalization of
FOREST-BGC model, is a general ecosystem process model
designed to simulate daily biogeochemical and hydrologic
processes from stand to global scales. BIOME-BGC logic
has been described by Running and Coughlan (1988) and
Running and Hunt (1993) in detail. BIOME-BGC was ex-
tensively modified and reported by Thornton (1998).
White et al. (2000) provided an account of the source for
parameters in BIOME-BGC, and assessed the sensitivity
of NPP to independent variation in every parameter. The
temporal framework of the BIOME-BGC model (version
4.1.1) (Thornton et al., 2002) consists of a dual discrete
time step approach, which is slightly different from the
implementation in previous versions of the BGC logic
(Running and Hunt, 1993; Hunt et al., 1996). In the latest
version of BIOME-BGC, most simulated ecosystem activ-
ity occurs at a daily time step, driven by daily values for
maximum and minimum temperatures, precipitation, solar
radiation, and air humidity. Examples of processes assessed
daily are soil water balance, photosynthesis, allocation
(which was assessed annually in the previous version),
litterfall, and C and N dynamics in the litter and soil. Phe-
nological timing of the budburst is determined once a year
using long-term thermal sums and soil moisture status for
deciduous forests and grasslands (White et al., 1997).
For evergreen vegetation, current climatic conditions
determine the beginning and end of the growing season.
Finally, the model is designed to require only standard
meteorological data¡ªnamely, maximum and minimum
daily air temperature (℃), precipitation (cm), daylight av-
erage partial pressure of water vapor (Pa), daylight aver-
age shortwave radiant flux density (W/m2) and daylength
(s)¡ªso that the model may be applied beyond those sites
with sophisticated instrumentation. A summary of the im-
portant components of BIOME-BGC relating to the pre-
diction of daily C allocation and exchange is given in Fig.
1, and some of the principle physical and biological pro-
cesses represented in BIOME-BGC were summarized
below.
1.1.1 Canopy radiation The plant canopy leaf area is
divided into sunlit and shaded fractions on the basis of a
radiation extinction coefficient that varies with leaf
geometry. All plant physiological processes are calculated
separately for the sunlit and shaded canopy fractions. Dif-
ferences in leaf physiology between the sunlit and shaded
fractions are parameterized as differences in specific leaf
area (sla), with the mass-based N concentration and con-
trols on stomatal conductance constant between sunlit and
shaded fractions.
1.1.2 Photosynthesis Assimilation (A) on a unit pro-
jected leaf area basis for C3 plants is estimated indepen-
dently for the sunlit and shaded canopy fractions, using a
biochemical model (Farquhar et al., 1980, with kinetic pa-
rameters from Woodrow and Berry (1988), de Pury and
Farquhar (1997)) and substitution from the CO2 diffusion
Fig.1. BIOME-BGC model carbon dynamics (modified by Thornton, 1998). The sole input to the carbon budget in BIOME-BGC is
the photosynthetic fixation of CO2 by the vegetation canopy. Outputs are all in the form of respired CO2, coming either from plant
tissues due to growth or maintenance respiration, or from the litter and soil carbon pools as the result of heterotrophic respiration. GPP,
gross primary production; GR, growth respiration; HR, heterotrophic respiration; MR, maintenance respiration; NBP, net biome
production; NEP, net ecosystem production; NPP, net primary production; PSN, photosynthesis.
SU Hong-Xin et al.: Simulations and Analysis of Net Primary Production in Quercus liaotungensis Forest of Donglingshan
Mountain Range in Response to Different Climate Change Scenarios 1283
equation to eliminate the explicit dependency on intracellu-
lar CO2 concentration. The maximum rate of carboxylation
(VC,max) is calculated as a function of the specific activity of
the Rubisco enzyme (act, itself a function of leaf
temperature), the weight fraction of N in the Rubisco mol-
ecule (fnr), the fraction of total leaf N in the Rubisco
enzyme (flnr), sla, and the leaf C:N ratio (C:Nleaf ) as
follows:
VC,max =
The model is very sensitive to the value for flnr, and
when data is available we optimize this parameter by fitting
to A-Ci curves. This approach requires knowing the leaf
temperature as well as sla and C:Nleaf for the measured
leaves. One advantage of this formulation is that it makes
explicit the dependence of VC,max on sla and C:Nleaf . Values
for fnr and act, as well as the temperature dependence of
act, are assumed constant across all species.
1.1.3 Stomatal conductance A form of the Leuning model
is used, which makes actual conductance a function of a
minimum value and a series of multiplicative reductions
based on incident radiation, vapor pressure deficit, leaf water
potential, and night minimum temperature (Running and
Coughlan, 1988). There is no direct effect of changing at-
mospheric CO2 concentration on stomatal conductance,
which is in agreement with recent studies for woody veg-
etation (Norby et al., 1999). One practical benefit of this
formulation is that it is not necessary to iterate between the
equations for CO2, water, and energy transfer at the leaf
surface, as is the case, e.g. with the Ball-Berry model in
which stomatal conductance is an explicit function of car-
bon assimilation.
1.1.4 Evaporation and transpiration Both processes are
estimated using the Penman-Monteith equation (Monteith,
1973). These computations were based on absorbed pho-
tosynthetically active radiation, daily average surface tem-
perature (soil or canopy), and surface (soil or canopy) re-
sistances to the transport of sensible heat and water vapor,
respectively. In the transpiration function, canopy resis-
tance to sensible heat was dependent on canopy bound-
ary layer conductance. Canopy resistance to water vapor
was calculated as the inverse of stomatal conductance. In
the soil evaporation function, constant soil resistance to
water vapor and latent heat were used to calculate poten-
tial evaporation, which was consequently modified by a
function of time since rain or snowmelt event. Canopy
evaporation was dependent on interception and the poten-
tial evaporation rates.
1.1.5 Autotrophic respiration Autotrophic respiration
was the sum of maintenance and growth respiration of the
different parts of the plant (canopy, stem and roots). Main-
tenance respiration (Rm) of each plant compartment was
computed as a function of tissue mass, tissue N
concentration, and using an exponentially increasing func-
tion of respiration with temperature as described by Amthor
(1986). Growth respiration (Rg), which is a simple propor-
tion of total new C allocated to growth, was incurred re-
gardless of current assimilation rate, and was calculated on
the basis of construction costs by plant compartment. Dif-
ferent construction costs were applied to woody and
nonwoody plant tissues. Here daily growth respiration was
not determined explicitly by the model, but was computed
as a proportion (32%) of the daily difference between gross
primary production (GPP) and Rm (Penning de Vries et al.,
1974).
1.1.6 Estimating NPP GPP represents the total gain of C
to the system by net photosynthesis and is defined as the
daily sum of gross photosynthesis and daily foliar
respiration. GPP was calculated based on absorbed photo-
synthetically active radiation, atmospheric CO2
concentration, air temperature, vapor pressure deficit,
precipitation, atmospheric N deposition, LAI, and avail-
able N content in soil. NPP represents the net accumulation
of C by the stand and is determined as the difference be-
tween GPP and the sum of the maintenance (Rm) and growth
(Rg) respiration components. Therefore, BIOME-BGC cap-
tures effects of a number of abiotic (temperature, vapor
pressure deficit, soil water, solar radiation, and CO2
concentration) and biotic (leaf area index, leaf, and root N
contents) controls on NPP.
1.2 Study site
The study site lies in the Dongling Mountain Range
(western Beijing, 39o5748 N, 115o2529 E, 1 250 m a.s.l,
Slope 22o) in eastern China, where Q. liaotungensis is domi-
nant canopy tree species with average tree heights of 12 m
and age of 45 yr old. Canopy coverage is about 80%. The
shrub layer is composed of Lespedeza bicolor, Spiraea
pubescens, Abelia biflora, Deutzia grandiflora, and
Rhododendron mucronulatum. On the forest floor, herba-
ceous species are abundant. The soil in this area belongs
to mountainous brown soil, ~ 50 cm in depth.
1.3 Model initiation
The BIOME-BGC model requires daily climate data and
site conditions to estimate NPP of the ecosystem. In this
study, the daily climate data were adjusted for the site con-
ditions (for example slope, aspect, albedo, elevation, pre-
cipitation pattern) using a microclimate simulation model,
act× f lnr
fnr× sla×C:Nleaf
Acta Botanica Sinica 植物学报 Vol.46 No.11 20041284
MT-CLIM (Running et al., 1987; Kimball et al., 1997;
Thornton and Running, 1999; Thornton et al., 2000). In the
simulation of the MT-CLIM, we changed the coefficient to
adjust daylight average temperature (TEMCF) from 0.45 in
the original model to -0.10 based on the available data of
the daily temperature observations in the study area. Cur-
rent climate data for the stand were simulated by MT-CLIM
based on the data from 1993 to 2001 recorded at the Beijing
Forest Ecosystem Research Station’s meteorological sta-
tion (1 150 m a.s.l, 39o57 N, 115o26 E). The data for 2000
were incomplete, and were removed from the dataset. The
future climatic scenarios were made based on the outputs
of the General Circulation Models (GCMs) for continental
China (Zhang, 1993; Gao et al., 1997). Nine climatic sce-
narios used in this study are displayed in Table 1.
The other critical parameters used for the model initial-
ization are shown in Table 2. Data pertaining to site-spe-
cific parameters for soil and plant values (including
mortality, allocation, plant labile, cellulose, lignin fraction,
C:N, canopy water interception, light extinction, flnr, sla)
were obtained from previous studies. When data were not
available for oak, parameters were obtained from related
genera under similar environmental conditions.
2 Test of BIOME-BGC
The BIOME-BGC model was developed to simulate the
changes of ecosystem water, C, and N pools over time.
Although the C dynamics of the BIOME-BGC were under-
gone testing and validation widely (Korol et al., 1991; Hunt
et al., 1991; Keyser et al., 2000; Thornton et al., 2002;
Churkina et al., 2003), we further tested the performance of
the BIOME-BGC for the Dongling Mountain Range by com-
paring modeled outputs with the field observations. As the
BIOME-BGC model relies primarily on the hydrologic cycle
and the control of water availability on C uptake and stor-
age (VEMAP members, 1995), we selected soil water con-
tent (mm) and annual total NPP (gC.m-2.yr-1) for the pur-
pose of model validation.
2.1 Soil water
Figure 2 presents the simulated soil water fraction in
1996 and 1997 in which the annual amounts of rainfall were
773.8 mm and 476.3 mm respectively. Comparison between
the simulated and observed values showed that the daily
time-step of the model provided a realistic water balance.
The simulated daily soil water resembled the extent and the
seasonal patterns of observed values (Wan, 1997). Simu-
lated soil water explained only 78.14% (SE =16.9 mm) of the
variance in the observed values by Wan (1997). Simula-
tions tended to overestimate soil water content by 16.67%
(Fig. 3).
2.2 NPP and the effects of double CO2
During the 1993-2001 period, the mean annual NPP simu-
lated was 729.3 gC.m-2.yr-1 (SE =39.9) with a range from
528.9 to 893.3 gC.m-2.yr-1, which was within the range of
observations by Jiang (1997) for the same region (349.3 to
1 064.2 gC.m-2.yr-1). Especially, the simulated annual NPP
was 669.2 gC.m-2.yr-1 for 1993, a value close to that of
671.3 gC.m-2.yr-1 calculated by Sun (1997) when only the
growth seasonal flux (May to Oct.) was taken into account.
Melillo et al. (1993) used the TEM to estimate the annual
NPP for the potential vegetation in the terrestrial biosphere
at an atmospheric CO2 concentration of 355 mmol /mol. Their
results showed that the average NPP of the temperate de-
ciduous forest was 620 gC.m-2.yr-1, ranging from 81 to
978 gC.m-2.yr-1. Xiao et al. (1998) used the same model to
estimate NPP in China for contemporary climate and sug-
gested the average NPP of the temperate deciduous for-
est was 715 gC.m-2.yr-1, ranging from 502 to 1 036
Table 1 The future climatic scenarios in this study
Climatic scenarios The potential change of climate
C0T0P0 Current climatic scenario (atmospheric CO2 concentration was 355 µmol /mol);
C0T0P2 Temperature unchanged, precipitation increased by 20%;
C0T2P0 Precipitation unchanged, temperature increased by 2 ℃;
C0T2P2 Temperature increased by 2 ℃ and precipitation increased by 20%;
C1T0P0 Doubled CO2, current precipitation and temperature;
C1T0P2 Doubled CO2, temperature unchanged, precipitation increased by 20%;
C1T2P0 Doubled CO2, temperature increased by 2 ℃, precipitation unchanged;
C1T2P2 Doubled CO2, temperature increased by 2 ℃, and precipitation increased by 20%;
C1T4P2 Doubled CO2, temperature increased by 4 ℃, and precipitation increased by 20%.
The C0T0P0 represents the current climatic condition and ambient CO2 concentration. The results of the scenarios of C0T0P2, C0T2P0, C1T0P0
illustrate the effects of individual factors, respectively. The C1T0P2 indicate the combined effects of precipitation increase and doubling
atmospheric CO2 concentration. The C1T2P0 indicate the combined effects of temperature increase and doubling atmospheric CO2 concentration.
The C0T2P2, C1T2P2, C1T4P2 indicate the combined effects of temperature and precipitation increases, with and without doubling atmospheric
CO2 concentration, respectively.
SU Hong-Xin et al.: Simulations and Analysis of Net Primary Production in Quercus liaotungensis Forest of Donglingshan
Mountain Range in Response to Different Climate Change Scenarios 1285
gC.m-2.yr-1. Recently, Li and Ji (2001) estimated the NPP
of global terrestrial ecosystem by an Atmosphere-Vegeta-
tion Interaction Model (AVIM), which suggested that the
NPP of the temperate deciduous forest was 808 gC.m-2.
yr-1. The simulated values of annual NPP by BIOME -BGC
here are within the range of observed values and others
simulations.
Elevated level of atmospheric CO2 is a primary constitu-
ent of global climate change. When atmospheric CO2 was
doubled from 355 µmol /mol to 710 µmol /mol without cli-
mate change, equilibrium simulations of the model projected
an increase of NPP ranging from 11.36% to 16.11% with an
average of 14.06% (SE = 0.70) during 8 yr for the oak forest
in this study (Fig.4). This result agreed with the study by
Pan et al. (1998), who found that doubled atmospheric CO2
would increase NPP for the conterminous United States
temperate deciduous forest by 15.50% when simulated by
the BIOME-BGC model. The results presented here showed
the model was sensitive to the elevated atmospheric CO2
and could be used to simulate the response of NPP of oak
forest ecosystem in Dongling Mountain areas under the
global change scenarios.
Table 2 Values of BIOME-BGC parameters for oak
Parameters Value Unit
Site parameters
Effective soil depth 0.45 m
Sand percentage by volume in rock-free soil 20 %
Silt percentage by volume in rock-free soil 50 %
Clay percentage by volume in rock-free soil 30 %
Site elevation 1 350 m
Site shortwave albedo (snow free) 0.15 %
Wet+dry atmospheric deposition of N 0.002 kgN.m-2.yr-1
Symbiotic+asymbiotic fixation of N 0.006 kgN.m-2.yr-1
Ecophysiological parameters
Annual whole-plant mortality fraction 0.005 1/yr
Annual fire mortality fraction 0.0025 1/yr
New fine root C : new leaf C 0.8 ratio
New stem C : new leaf C 2.0 ratio
New live wood C : new total wood C 0.15 ratio
New root C : new stem C 0.23 ratio
C:N of leaves 20.6 kgC/kgN
C:N of leaf litter 49.6 kgC/kgN
C:N of fine roots 64.7 kgC/kgN
C:N of live wood 67.5 kgC/kgN
C:N of dead wood 109.9 kgC/kgN
Leaf litter labile proportion 0.30 DIM
Leaf litter cellulose proportion 0.44 DIM
Leaf litter lignin proportion 0.26 DIM
Fine root labile proportion 0.29 DIM
Fine root cellulose proportion 0.18 DIM
Fine root lignin proportion 0.53 DIM
Dead wood cellulose proportion 0.66 DIM
Dead wood lignin proportion 0.34 DIM
Canopy water interception coefficient 0.021 1/LAI/d
Canopy light extinction coefficient 0.7 DIM
Canopy average specific leaf area 30.0 m2/kgC
Fraction of leaf N in Rubisco 0.085 DIM
Maximum stomatal conductance 0.006 5 m /s
Cuticular conductance 0.000 01 m /s
Boundary layer conductance 0.01 m /s
Leaf water potential: start of conductance reduction - 0 . 6 Mpa
Leaf water potential: complete conductance reduction - 2 . 3 Mpa
Vapor pressure deficit: start of conductance reduction 930.0 Pa
Vapor pressure deficit: complete conductance reduction 4 100.0 Pa
Default values are considered for the other parameters of the model. DIM, dimensionless; LAI, leaf area index.
Acta Botanica Sinica 植物学报 Vol.46 No.11 20041286
3 Results and Discussion
3.1 Meteorological characteristic
The daily meteorological data simulated by MT-CLIM
for the study site showed that it was a strong monsoon
climate, with concentrated seasonal precipitation profiles
in summer. During the period of 1993 to 2001, annual pre-
cipitation ranged from 398.0 mm to 776.5 mm, with an aver-
age of 590.1 mm (SE = 56.2). Annual average air temperature
was 5.5 ℃ (SE = 0.2). The large SE values indicated great
inter-annual climatic variations, which would result in
marked inter-annual NPP variations in the simulation (Table
3).
3.2 Response of NPP to changes in climate and ambient
CO2 concentration
The simulated annual NPP responses to the different
climatic scenarios are shown in Table 3. The simulated
results indicated NPP was not sensitive to the temperature
change alone, decreasing by only 0.36% under the C0T2P0
scenario (Table 3). This result is consistent with the analy-
sis of Hunt and Running (1992). Both suggest that although
NPP is sensitive to temperature sensitivity of photosyn-
thetic efficiency, the effect in BIOME-BGC is small. Simu-
lated NPP increased by 9.82% under the C0T0P2 scenario.
With changes in both temperature and precipitation, NPP
increased by 9.78% (under C0T2P2). It is well established
that temperature and precipitation are dominant controls
on plant photosynthesis (Lieth, 1975; Dai and Fung, 1993).
Climate change will affect NPP in a number of different ways
(Houghton and Woodwell, 1989). Elevated temperatures
may increase NPP through metabolically enhanced photo-
synthesis as well as by increasing nutrient availability
through higher rates of decomposition. E leva ted
temperatures, however, may also decrease NPP by
Fig.2. a, b. Soil water content in 1996 and 1997, respectively. c, d. Precipitation in 1996 and 1997, respectively. The model predictions
of soil water generally agreed with field micrometeorological measurements made in 1996 and 1997.
Table 3 The simulated annual net primary productivity (NPP) (gC.m-2.yr-1) responses to the different climatic scenarios
Year C0T0P0 C0T0P2 C0T2P0 C0T2P2 C1T0P0 C1T0P2 C1T2P0 C1T2P2 C1T4P2
1993 669.2 729.5 673.0 733.7 761.1 827.4 773.8 849.1 850.0
1994 788.3 847.9 782.3 845.9 914.1 980.6 924.6 993.1 998.0
1995 842.3 869.4 864.3 890.5 968.2 993.2 1 014.9 1 032.5 1 046.4
1996 893.3 925.0 898.6 935.5 1 005.5 1 046.4 1 018.7 1 078.7 1 086.4
1997 715.6 794.4 694.6 782.8 800.0 892.9 794.8 892.2 875.5
1998 709.6 800.3 705.9 797.4 823.0 902.6 832.0 914.1 915.8
1999 686.8 744.7 684.2 746.6 764.8 839.3 778.2 847.4 839.4
2001 528.9 650.7 518.1 636.5 614.1 738.1 613.1 728.5 724.2
Mean value 729.3 795.2 727.6 796.1 831.4 902.6 843.8 917.0 917.0
Abbreviations are the same as in Table 1.
SU Hong-Xin et al.: Simulations and Analysis of Net Primary Production in Quercus liaotungensis Forest of Donglingshan
Mountain Range in Response to Different Climate Change Scenarios 1287
decreasing soil moisture and enhancing plant respiration.
And an increase in precipitation tends to alleviate the mois-
ture stress for plant growth, and hence has positive effects
on NPP in water-limited region. The sensitivity analysis
revealed that simulated NPP by BIOME-BGC here is much
more sensitive to a 20% precipitation increase (P < 0.017)
than to a 2 ℃ temperature increase (P = 0.964) (Table 4). In
other words, the available water was the main limiting fac-
tor for NPP under the current ambient CO2 concentration in
the study area. The result is consistent with the analysis of
Churkina and Running (1998). With no CO2 fertilization ef-
fect included in the simulations, the results of Churkina
and Running (1998) indicated that the deciduous broadleaf
forests productivity is largely controlled by water avail-
ability (64%). Environmental controls other than climate
(such as nutrient availability, plant characteristics, and natu-
ral and human disturbances, as well as many other factors)
were of secondary importance (17%), and the temperature
the third (only 11%) for deciduous broadleaf forest. Nemani
et al. (2003) suggests that water availability most strongly
limits vegetation growth over 40% of Earth’s vegetated
surface. From 1982 to 1999, modeled NPP increased the
most (6.5%) in water- and radiation-limited regions, followed
by temperature- and radiation-limited regions (5.7%) and
temperature- and water-limited regions (5.4%). NPP in-
creased significantly (P < 0.01) over 25% of the global veg-
etated area, with a mean rate of 6.3 gC.m-2.yr-1.
Doubling atmospheric CO2 concentration under the
C1T0P0 scenario increased simulated NPP, with 14.05%.
This result agreed with the study of Pan et al. (1998). While
there is still uncertainty over the physiological response to
increased CO2 (Ward and Strain, 1999), some commonalties
in experimental response are emerging. In general, net as-
similation and growth rate increase under elevated (double
the current concentration) CO2, while dark respiration and
leaf nitrogen concentration decrease (Curtis and Wang,
1998). Increasing CO2 concentration has been found to
decrease the stomatal conductance to H2O, and then im-
prove the water use efficiency of plants, thus tends to en-
hance the primary production. However, the response var-
ies with species and ontogeny (Tjoelker et al., 1998a). Other
studies suggest that acclimation (down-regulation) in rates
of photosynthesis occurs via changes in sla, leaf nitrogen
concentration, and amount of non-structural carbohydrates
(TNC) (Tjoelker et al., 1998b). The results by BIOME-BGC
here suggested NPP increased significantly (P < 0.000) to a
doubling of atmospheric CO2 (Table 4).
With changes in both climate and a doubling of atmo-
spheric CO2, the model generated increases in NPP of
24.60% for C1T0P2, 15.62% for C1T2P0, 26.35%for C1T2P2,
and 26.25% for C1T4P2, respectively. In comparison, Zhou
and Zhang (1996) suggested that the NPP of the warm tem-
perate and deciduous broad-leaved forest zone in China
increased by 34.26% and 38.36% under the C1T2P2 and
C1T4P2 respectively based on data of 50 climatic stations.
Fig.4. Comparing net primary productivity (NPP) under the
current atmospheric CO2 with that under doubled CO2 without
climate change.
Fig.3. Linear regression results between simulated and observed
1996 and 1997 daily soil water.
Table 4 The results of the Univariate Analysis of Variance: tests
of between-subjects effects based on simulated net primary pro-
ductivity (NPP)
Source F Significant
P 2*T 2 0.002 0.969
C1*T2 0.059 0.809
C1*P2 0.008 0.931
C1*P2*T2 0.000 0.996
C1, doubled atmospheric CO2 concentration; P2, precipitation in-
creased by 20%; T2, temperature increased by 2 ℃; *, show
interaction.
Acta Botanica Sinica 植物学报 Vol.46 No.11 20041288
The large difference between the two studies might be due,
in part, to the static model (BIOME-BGC). The model did
not consider the changes in vegetation types under doubled
atmospheric CO2 concentration and associated climate
changes. But, based on the study of Nemani et al. (2003),
changes in climate (with constant vegetation) directly con-
tributed only about 40% of the total increase in global NPP
from 1982 to 1996, and changes in vegetation (with con-
stant climate) over the same period contributed ~60% of
total NPP increase, possibly as a result of climate-vegeta-
tion feedbacks, changes in land use, and growth stimula-
tion from other mechanisms. So it is important to involve
the change of vegetation distribution when considering
the long-term effects of elevated atmospheric CO2 concen-
trations and associated climate changes on forest
ecosystems.
The results of the Univariate Analysis of Variance: tests
of between-subjects effects based on simulated NPP by
the SPSS (10.0) are shown in Table 4. It is clear that the
combined climate and CO2 change had no significant effect
on NPP. In other words, the NPP responses of BIOME-
BGC to changes in both climate and CO2 are essentially
additive (VEMAP, 1995). This result is due, in part, to the
mechanisms that control CO2 responses on NPP in BIOME-
BGC (Pan et al., 1998). In the model, the NPP response to
doubled CO2 is directly affected by increased intercellular
CO2 concentration and the efficiency of quantum yield,
reduced canopy conductance, and leaf N concentration.
The effects of these factors on NPP do not vary with climate.
The patterns of NPP responses along the climate gradients
are controlled by the change in transpiration associated
with reduced leaf conductance to water vapor. This change
affects soil water, then leaf area development, and finally
NPP. However, ecosystem responses to realistic combina-
tions of global changes are not necessarily simple combi-
nations of the responses to the individual factors (Shaw et
al., 2002). Accurate predictions of ecosystems responses
to suites of global changes depend on successful integra-
tion across a range of processes and time scales. Multifac-
tor experiments on ecosystems that are easy to manipulate
can provide a rich source of examples as well as test beds
for exploring hypotheses with the potential to explain the
responses of a wide range of ecosystems (Oechel et al.,
1994; Schimel et al., 2000; McGuire et al., 2001; Shaw et al.,
2002). The ability of ecological model to accurately simu-
late ecosystems responses to future climate change would
be improved by incorporated the results of those multifac-
tor experiments.
4 Summary and Conclusion
A general ecosystem process model, BIOME-BGC, was
used to simulate the NPP response to the global changes at
stand scale in the Dongling Mountain Range of the Beijing
area. The main conclusions can be drawn from our analy-
ses as following:
(1) Validation of the model simulations was carried out
using measurements in oak forest of Dongling Mountain at
different time steps of the model (daily and annual). These
results showed that the BIOME-BGC model could be ap-
plied to the Dongling Mountain at stand scale.
(2) Simulated NPP was much more sensitive to a 20%
precipitation increase and a doubling of atmospheric CO2
from 355 µmol /mol to 710 µmol /mol than to a 2 ℃ tempera-
ture increase. These results also showed clearly that the
effects of elevated CO2 and climate change on the response
of NPP were not interactive.
The accuracy of our predictions was affected by two
facts: first, there is a lack of sufficient data for validating
and running the model. For example, there were only 8-yr
climatic data to run the model and few quantitative data in
annual NPP of oak forest ecosystems in Dongling Moun-
tain to validate the model; second, there were some con-
straining factors from the BIOME-BGC model itself (as men-
tioned earlier in Results and Discussion). Despite the wide-
spread application of the process model to predict the re-
sponse of forest ecosystems to CO2 increase and climate
change, it is important to acknowledge that there are sev-
eral key processes still not well understood. Stomatal
conductance, tissue and soil respiration, and resource allo-
cation are currently still represented in a very empirical way.
Other processes, such as nutrient uptake, tissue mortality,
fruiting and competition, presented similar problems
(Magnani and Matteucci, 2001). With all the limitations as
explained previously, it is important to recognize that the
ability of a model to simulate current ecological conditions
does not validate its ability to accurately simulate responses
to future climate change (VEMAP, 1995).
Acknowledgements: BIOME-BGC (version 4.1.1) and MT-
CLIM were provided by the Numerical Terradynamic Simu-
lation Group (NTSG) at the University of Montana. We are
grateful to Osbert J Sun for his assistance in preparing the
manuscript.
References:
Amthor J S. 1986. Evolution and applicability of a whole plant
respiration model. J Theor Biol, 122: 473-490.
Bondeau A, Kicklighter D W, Kaduk J, the Participants of the
SU Hong-Xin et al.: Simulations and Analysis of Net Primary Production in Quercus liaotungensis Forest of Donglingshan
Mountain Range in Response to Different Climate Change Scenarios 1289
Potsdam NPP Model Intercomparison. 1999. Comparing glo-
bal models of terrestrial net primary productivity (NPP): im-
portance of vegetation structure on seasonal NPP estimates.
Global Change Biol, 5 (Suppl. 1): 35-45.
Cramer W, Kicklighter D W, Bondeau A, Moore III B, Churkina
G, Nemry B, Ruimy A, Schloss A, the Participants of the
Potsdam NPP Model Intercomparison. 1999. Comparing glo-
bal models of terrestrial net primary productivity (NPP): over-
view and key results. Global Change Biol, 5 (Suppl. 1): 1-15.
Churkina G, Running S W. 1998. Contrasting climatic controls on
the estimated productivity of global terrestrial biomes.
Ecosystems, 1: 206-215.
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 Eu-
ropean coniferous forests using a biogeochemistry model.
Ecosystems, 6: 168-184.
Curtis P S, Wang X. 1998. A meta-analysis of elevated CO2 ef-
fects on woody plant mass, from and physiology. Oecologia,
113: 299-313.
Dai A, Fung I Y. 1993. Can climate variability contribute to the
“missing” CO2 sink? Global Biogeochem Cy, 7: 599-609.
de Pury D G G, Farquhar G D. 1997. Simple scaling of photo-
synthesis from leaves to canopies without the errors of big-
leaf models. Plant Cell Environ, 20: 537-557.
Farquhar G D, von Caemmerer S, Berry J A. 1980. A biochemical
model of photosynthetic CO2 assimilation in leaves of C3
species. Planta, 149: 78-90.
Field C B, Chapin F S, Matson P A, Mooney H A. 1992. Re-
sponses of terrestrial ecosystems to the changing atmosphere:
a resource-based approach. Annu Rev Ecol Syst, 23: 201-235.
Field C B, Randerson J T, Malmstrom C M. 1995. Global net
primary production: combing ecology and remote sensing.
Remote Sens Environ， 51: 74-88.
Friend A D, Stevens A K, Knox R G, Cannell M G R. 1997. A
process-based, terrestrial biosphere model of ecosystem dy-
namics (Hybrid v.3.0). Ecol Model, 95: 249-287.
Gao Q , Li X-B，Yang X-S. 2003. Responses of vegetation and
primary production in north-south transect of eastern China to
global change under land use constraint. Acta Bot Sin ，45:
1274-1284.
Gao Q, Yu M, Zhang X-S, Guan . 1997. Dynamic modeling of north-
east China transect responses to global change — a regional
vegetation model driven by remote sensing information. Acta
Bot Sin,39: 800-810. (in Chinese with English abstract)
Houghton R A, Woodwell G M. 1989. Global climatic change. Sci
Am, 260(4): 36-44.
Hunt E R Jr, Martin F C, Running S W. 1991. Simulating the
effect of climatic variation on stem carbon accumulation of a
ponderosa pine stand: comparison with annual growth incre-
ment data. Tree Physiol, 9: 161-172.
Hunt E R Jr, Piper S C, Nemani R R, Keeling C D, Otto R D,
Running S W. 1996. Global net carbon exchange and intra-
annual atmospheric CO2 concentrations predicted by an eco-
system process model and three-dimensional atmospheric
transport model. Global Biogeochem Cy, 10: 431-456.
Hunt E R Jr, Running S W. 1992. Simulated dry matter yields for
aspen and spruce stands in the North American boreal forest.
Can J Remote Sens, 3: 126-133
IPPC. 2001.Third Assessment Report. Houghton J T, Ding Y,
Griggs D J, Noguer M, Linden P J van der, Dai X, Maskell K,
Johnson C A. Climate Change 2001: The Scientific Basis.
Cambridge：Cambridge University Press. 1-94.
Jiang H. 1997. Study on biomass of Quercs liaotungensis and
Betla dahurica forest in Dongling Mountain. Chen L-Z, Huang
J-H. Studies on Structure and Functions of Forest Ecosystem
in Warm Temperate Zone. Beijing: Science Press. 104-115. (in
Chinese with English abstract)
Keyser A R, Kimball J S, Nemani R R, Running S W. 2000. Simu-
lating the effects of climate change on carbon balance of North
American high — latitude forests. Global Change Biol, 6: 1-
11.
Kimball J S, Running S W, Nemani R. 1997. An improved method
for estimating surface humidity from daily minimum
temperature. Agr For Meteorol, 85: 87-98.
Korol R L, Running S W, Milner K S, Hunt E R Jr. 1991. Testing
a mechanistic carbon balance model against observed tree
growth. Can J For Res, 21: 1098-1105.
Li Y-P, Ji J-J. 2001. Simulations of carbon exchange between
global terrestrial ecosystem and atmosphere. Acta Geogr Sin,
56: 379-389. (in Chinese with English abstract)
Lieth H. 1975. Modeling primary productivity of the world. Lieth
H, Whittaker R H. Primary Productivity of the Biosphere.
Vol. 14. New York: Springer-Verlag. 237–263.
Magnani F, Matteucci G. 2001. Modelling the impact of green-
house gases on forests. Karnosky D F, Ceulemans G E. The
Impact of Carbon Dioxide and Other Greenhouse Gases on
Forest Ecosystem. Wallingford: CABI Publishing. 269-295.
McGuire A D, Sitch S, Clein J S, Dargaville R, Esser G, Foley J,
Heimann M, Joos F, Kaplan J, Kicklighter D W, Meier R A,
Melillo J M, Moore B, Prentice I C, Ramankutty N, Reichenau
T, Schloss A, Tian H, Williams L J, Wittenberg U. 2001.
Carbon balance of the terrestrial biosphere in the twentieth
century: analyses of CO2, climate and land use effects with
four process-based ecosystem models. Global Biogeochem
Cy, 15: 183-206.
Melillo J M, McGuire A D, Kicklighter D W, Moore Ⅲ B,
Vorosmarty C J, Schloss A L. 1993. Global climate change and
Acta Botanica Sinica 植物学报 Vol.46 No.11 20041290
terrestrial net primary production. Nature, 363: 234-240.
Melillo J M, Prentice I C, Schulze E D, Farquhar G, Sala O. 1996.
Terrestrial biotic responses to environmental change and feed-
backs to climate. Houghton J T, Meira Filho L G, Callender B
A, Harris N, Kattenberg A, Maskell K. Climate Change 1995:
The Science of Climate Change. Cambridge: Cambridge Uni-
versity Press. 445-482.
Milner K S, Running S W, Coble D W. 1996. A biophysical soil-
site model for estimating potential productivity of forested
landscapes. Can J For Resour, 26: 1174-1186.
Monteith J L. 1973. Principles of environmental physics. Lon-
don : Edward Arnold. 241.
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.
Norby R J, Wullschleger S D, Gunderson C A, Johnson D W,
Ceulmans R. 1999. Tree responses to rising CO2 in field
experiments: implications for the future forest. Plant Cell
Environ, 22: 683-714.
Oechel W C, Cowles S, Grulke N, Hastings S J, Lawrence B,
Prudhomme T, Riechers G, Strain B, Tissue D, Vourlitis G.
1994. Transient nature of CO2 fertilization in Arctic tundra.
Nature, 371: 500-503.
Pan Y D, Melillo J M, McGuire A D, Kicklighter D W, Pitelka L
F, Hibbard K, Pierce L L, Running S W, Ojima D S, Parton W
J, Schimel D S, Other VEMAP members. 1998. Modeled re-
sponses of terrestrial ecosystems to elevated atmospheric CO2:
a comparison of simulations by the biogeochemistry models
of the vegetation/ecosystem modeling and analysis project
(VEMAP). Oecologia, 114:389-404.
Parton W J, Cole C V, Stewart J W B, Ojima D S, Schimel D S.
1989. Simulating regional patterns of soil C, N, and P dynam-
ics in the U.S. central grasslands region. Clarholm M,
Bergström L. Ecology of Arable Lands. Amsterdam, The
Netherlands: Kluwer Academic Publishers. 99-108.
Parton W J, Stewart J W B, Cole C V. 1988. Dynamics of C, N,
P and S in grassland soils: a model. Biogeochemistry, 5:109-
131.
Penning de Vries F W T, Brunsting A, van Laar H H. 1974.
Products, requirements and efficiency of biosynthesis: a quan-
titative approach. J Theor Biol, 45: 339-377.
Running S W, Coughlan J C. 1988. A general model of forest
ecosystem processes for regional applications. Ⅰ. Hydro-
logic balance, canopy gas exchange and primary production
processes. Ecol Model, 42: 125-154.
Running S W, Hunt R E. 1993. Generalization of a forest ecosys-
tem process model for other biomes, BIOME-BGC, and an
application for global-scale models. Ehleringer J R, Field C B.
Scaling Physiologic Processes: Leaf to Globe. San Diego: Aca-
demic Press. 141-158.
Running S W, Nemani R R, Hungerford R D. 1987. Extrapolation
of synoptic meteorological data in mountainous terrain and its
use for simulat ing forest evapotranspiration and
photosynthesis. Can J For Resour, 17: 472-483.
Schimel D, Melillo J, Tian H, McGuire A D, Kicklighter D, Kittel
T, Rosenbloom N, Running S, Thornton P, Ojima D, Parton
W, Kelly R, Sykes M, Neilson M, Rizzo B. 2000. Contribu-
tion of increasing CO2 and climate to carbon storage by eco-
systems in the United States. Science, 287: 2004-2006.
Shaw M R, Zavaleta E S, Chiariello N R, Cleland E E, Mooney H
A, Field C B. 2002. Grassland responses to global environ-
mental changes suppressed by elevated CO2. Science, 298:
1987-1990.
Sun X-F. 1997. The energy flow and efficiency of solar energy
conversion in a warm temperate, deciduous broad- leaved forest.
Chen L-Z. Huang J-H. Studies on Structure and Functions of
Forest Ecosystem in Warm Temperate Zone. Beijing: Science
Press. 137-162. (in Chinese with English abstract)
Thornton P E. 1998. Description of a numerical simulation model
for predicting the dynamics of energy, water, carbon, and
nitrogen in a terrestrial ecosystem. Missoula MT. Ph.D.
dissertation, University of Montana, 280.
Thornton P E, Hasenauer H, White M A. 2000. Simultaneous
estimation of daily solar radiation and humidity from observed
temperature and precipitation: an application over complex
terrain in Austria. Agr For Meteorol, 104: 255-271.
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 bud-
gets in evergreen needleleaf forests. Agr For Meteorol, 113:
185-222.
Thornton P E, Running S W. 1999. An improved algorithm for
estimating incident daily solar radiation from measurements
of temperature, humidity, and precipitation. Agr For Meteorol,
93: 211-228.
Tjoelher M G, Oleksyn J, Reich P B. 1998a. Temperature and
ontogeny mediate growth response to elevated CO2 in seeding
of five boreal species. New Phytol, 140: 197-210.
Tjoelher M G, Oleksyn J, Reich P B. 1998b. Seedlings of five
boreal tree species differ in acclimation of net photosynthesis
to elevated CO2 and temperature. Tree Physiol, 18: 715-726.
VEMAP Members. 1995. Vegetation/ecosystem modeling and
analysis project: comparing biogeography and biogeochemistry
models in a continental-scale study of terrestrial ecosystem re-
sponses to climate change and CO2 doubling. Global
Biogeochem Cy, 9: 407-437.
SU Hong-Xin et al.: Simulations and Analysis of Net Primary Production in Quercus liaotungensis Forest of Donglingshan
Mountain Range in Response to Different Climate Change Scenarios 1291
Vitousek P M, Mooney H A, Lubchenco J, Melillo J M. 1997. Hu-
man domination of earth’s ecosystems. Science, 277: 494-499
Wan S-Q . 1997. Some hydrological aspects of forest ecosystems
at Dongling Mountains. Chen L-Z . Huang J-H. Studies on
Structure and Functions of Forest Ecosystem in Warm Tem-
perate Zone. Beijing: Science Press. 46-57. (in Chinese with
English abstract)
Ward J K, Strain B R. 1999. Elevated CO2 studies: past, present
and future. Tree Physiol, 19: 211-220.
White M A, Thornton P E, Running S W. 1997. A continental
phenology model for monitoring vegetation responses to
interannual climatic variability. Global Biogeochem Cy, 11:
217-234.
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 Interactions, 4 (3): 1-85.
Woodrow I E, Berry J A. 1988. Enzymatic regulation of photo-
synthetic CO2 fixation in C3 plants. Ann Rev Plant Physiol
Plant Mol Biol, 39: 533-594.
Woodward F I, Smith T M, Emanuel W R. 1995. A global land
primary productivity and phytogeography model. Glob
Biogeochem Cy, 9: 471-490.
Xiao X-M , Melillo J M, Kicklighter D W, Pan Y-D, McGuire A D,
Helfrich J. 1998. Net primary production of terrestrial ecosys-
tems in China and its equilibrium responses to changes in cli-
mate and atmospheric CO2 concentration. Acta Phytoecol Sin,
22: 97-118.
Yu M, Gao Q, Liu Y H, Xu H M, Shi P J. 2002. Responses of
vegetation structure and primary production of a forest transect
in eastern China to global change. Global Ecol Biogeogr, 11:
223-236.
Zhang X-S (Chang Hsin-Shih). 1993. A vegetation-climate classi-
fication system for global change studies in China. Q Sci, 2:
157-169. (in Chinese with English abstract)
Zheng Y-R， Zhou G-S ， Zhang X-S ，Wang J-L， Tai H-J .
1997. An agricultural net primary productivity model. Acta Bot
Sin ， 39: 831-836. (in Chinese with English abstract)
Zhou G-S, Zhang X-S . 1996. Study on NPP of natural vegetation
in China under global climate change. Acta Phytoecol Sin , 20:
11-19. (in Chinese with English abstract)
(Managing editor: HAN Ya-Qin)