ecosystem and landscape scales: (1) CO 2 flux measurements by eddy covariance
technique (Goulden et al. 2011; Moureaux et al. 2008; Verbeeck et al. 2008);
(2) process-based biogeochemical models that incorporate a number of physiological processes and use climate data as inputs (Matsushita et al. 2004; Running
1993); and (3) Production Efficiency Models (PEM) that use the principle of
radiation-use efficiency (RUE) or light-use efficiency (LUE) with the inputs of
satellite images and climate data (Peng and Gitelson 2012; Potter et al. 1993; Prince
and Goward 1995a, b; Running et al. 1999; Sims et al. 2008; Sims et al. 2006a; Xiao
et al. 2005a, b).
The first year-long continuous measurements of net ecosystem CO 2 exchange
(NEE) from the eddy covariance technique were conducted at the Harvard Forest
site in Massachusetts in 1990 (Wofsy et al. 1993). The integrated CO 2 flux
measurements available at CO 2 flux tower sites cover footprints with various sizes
and shapes, which range from hundreds of meters to several kilometers, depending
on tower heights, canopy physical characteristics, and wind velocity (Baldocchi
et al. 1996). Continuous measurements of NEE between terrestrial ecosystems and
the atmosphere from eddy flux towers at half-hour intervals allow for more
detailed study of ecosystem respiration (R e ) and GPP at ecosystem and landscape
scales (Wofsy et al. 1993). NEE data can be gap-filled and partitioned into R e and
GPP with different methods (Papale et al. 2006; Reichstein et al. 2005); however,
there are still large uncertainties in estimating seasonal dynamics and spatial
variation of R e and GPP at the canopy and landscape scales due to spatial heterogeneity within the footprints of flux measurements in a flux tower.
Process-based biogeochemical models describe the energy conversion in the
vegetation growth process, including photosynthesis and respiration. A number of
process-based biogeochemical models have been developed, such as the BioGeochemical Cycles model (BIOME-BGC) (Running and Gower 1991), the
Terrestrial Ecosystem Model (TEM) (McGuire et al. 1995), the CENTURY model
(Parton et al. 1993), the Carbon Exchange in the Vegetation-Soil-Atmosphere
model (CEVSA) (Cao and Woodward 1998a, b; Woodward et al. 1995), and the
Atmosphere-Vegetation Interaction Model (AVIM) (Ji 1995). These models often
have a number of state variables and a large number of parameters that describe
the responses of various biogeochemical processes to climate, soils, and water.
Model calibration is essential and is often done before these models are applied to
simulate the carbon dynamics of terrestrial ecosystems at landscape and regional
scales (Chen and Zhuang 2012; Chiesi et al. 2012).
The PEM was first proposed to estimate NPP of vegetation by Monteith
(Monteith 1972; Monteith 1977), based on the theory of RUE or LUE. In simple
terms, NPP is estimated as the product of absorbed photosynthetically active
radiation (PAR) and RUE or LUE. Based on this concept, a number of PEMs have
been developed to estimate gross and net primary productions with the use of
satellite image data and climate data (Field et al. 1995; Potter et al. 1993; Prince
and Goward 1995; Running et al. 1994, 2004; Xiao et al. 2004c), such as the
Global Production Efficiency Model (GLO-PEM) (Prince and Goward 1995),
the Carnegie-Ames-Stanford Approach model (CASA) (Potter 1999; Potter et al.
128
X. Xiao et al.
technique (Goulden et al. 2011; Moureaux et al. 2008; Verbeeck et al. 2008);
(2) process-based biogeochemical models that incorporate a number of physiological processes and use climate data as inputs (Matsushita et al. 2004; Running
1993); and (3) Production Efficiency Models (PEM) that use the principle of
radiation-use efficiency (RUE) or light-use efficiency (LUE) with the inputs of
satellite images and climate data (Peng and Gitelson 2012; Potter et al. 1993; Prince
and Goward 1995a, b; Running et al. 1999; Sims et al. 2008; Sims et al. 2006a; Xiao
et al. 2005a, b).
The first year-long continuous measurements of net ecosystem CO 2 exchange
(NEE) from the eddy covariance technique were conducted at the Harvard Forest
site in Massachusetts in 1990 (Wofsy et al. 1993). The integrated CO 2 flux
measurements available at CO 2 flux tower sites cover footprints with various sizes
and shapes, which range from hundreds of meters to several kilometers, depending
on tower heights, canopy physical characteristics, and wind velocity (Baldocchi
et al. 1996). Continuous measurements of NEE between terrestrial ecosystems and
the atmosphere from eddy flux towers at half-hour intervals allow for more
detailed study of ecosystem respiration (R e ) and GPP at ecosystem and landscape
scales (Wofsy et al. 1993). NEE data can be gap-filled and partitioned into R e and
GPP with different methods (Papale et al. 2006; Reichstein et al. 2005); however,
there are still large uncertainties in estimating seasonal dynamics and spatial
variation of R e and GPP at the canopy and landscape scales due to spatial heterogeneity within the footprints of flux measurements in a flux tower.
Process-based biogeochemical models describe the energy conversion in the
vegetation growth process, including photosynthesis and respiration. A number of
process-based biogeochemical models have been developed, such as the BioGeochemical Cycles model (BIOME-BGC) (Running and Gower 1991), the
Terrestrial Ecosystem Model (TEM) (McGuire et al. 1995), the CENTURY model
(Parton et al. 1993), the Carbon Exchange in the Vegetation-Soil-Atmosphere
model (CEVSA) (Cao and Woodward 1998a, b; Woodward et al. 1995), and the
Atmosphere-Vegetation Interaction Model (AVIM) (Ji 1995). These models often
have a number of state variables and a large number of parameters that describe
the responses of various biogeochemical processes to climate, soils, and water.
Model calibration is essential and is often done before these models are applied to
simulate the carbon dynamics of terrestrial ecosystems at landscape and regional
scales (Chen and Zhuang 2012; Chiesi et al. 2012).
The PEM was first proposed to estimate NPP of vegetation by Monteith
(Monteith 1972; Monteith 1977), based on the theory of RUE or LUE. In simple
terms, NPP is estimated as the product of absorbed photosynthetically active
radiation (PAR) and RUE or LUE. Based on this concept, a number of PEMs have
been developed to estimate gross and net primary productions with the use of
satellite image data and climate data (Field et al. 1995; Potter et al. 1993; Prince
and Goward 1995; Running et al. 1994, 2004; Xiao et al. 2004c), such as the
Global Production Efficiency Model (GLO-PEM) (Prince and Goward 1995),
the Carnegie-Ames-Stanford Approach model (CASA) (Potter 1999; Potter et al.
128
X. Xiao et al.