where G = 2.5, C 1 = 6, C 2 = 7.5, and L = 1, and q NIR , q red , and q blue are the
land surface reflectance of the NIR, red, and blue bands, respectively. EVI is more
sensitive in medium-to-high biomass areas than NDVI (Huete et al. 1997). The
following sections describe detailed calculation methods of FPAR in different
PEMs.
LUE is the rate at which absorbed radiation is converted into dry matter. It can
be estimated with various methods: a constant conversion efficiency or the product
of a maximum (optimum) constant adjusted by environmental stress scalars
(Ruimy et al. 1999). Various environmental stress scalars, such as air temperature
and soil moisture, result in downregulation of the maximum LUE (McCallum I.
2009). The GR and TG models estimate the parameter e g using remote sensing
vegetation indices (Table 5.1). The TURC model assumes maximum light-use
efficiency for all biome types without considering the environmental stresses. The
other six PEMs calculate the parameter e g by adjusting maximum light-use efficiency e 0 with soil moisture, air temperature, or CO 2 concentration scalars.
Moreover, the CASA, TURC, C-Fix, and EC-LUE models use the same e 0 across
all vegetation types. GLO-PEM, VPM, and PSN models assign specific e 0 for
different biome types. A brief description of the unique variable calculations of
PEMs is listed below.
5.3.1 The Carnegie-Ames-Stanford Approach (CASA) Model
The CASA model is a simple and mechanical process model, and it describes the
processes of carbon dynamics in terrestrial vegetation and soils (Potter et al. 1993).
The CASA model directly estimates NPP, avoiding complicated calculation of
autotrophic respiration.
In the earlier version of the CASA model, FPAR canopy is estimated with NDVI
and biome type (Potter et al. 1993). The maximum value of FPAR canopy is 0.95.
FPAR canopy is defined as below:
FPAR canopy ¼ min
SR À SR min
SR max À SR min
; 0:95
!
ð5:15Þ
SR ¼
1 þ NDVI
1 À NDVI
ð5:16Þ
where SR min is 1.08 for bare land surface, and SR max is determined by biome type,
ranging from 4.14 to 6.17.
The updated version of the CASA model uses EVI to estimate FPAR chl (Potter
2012), following the work suggested by Xiao et al. (2004a):
FPAR chl ¼ EVI
ð5:17Þ
132
X. Xiao et al.
land surface reflectance of the NIR, red, and blue bands, respectively. EVI is more
sensitive in medium-to-high biomass areas than NDVI (Huete et al. 1997). The
following sections describe detailed calculation methods of FPAR in different
PEMs.
LUE is the rate at which absorbed radiation is converted into dry matter. It can
be estimated with various methods: a constant conversion efficiency or the product
of a maximum (optimum) constant adjusted by environmental stress scalars
(Ruimy et al. 1999). Various environmental stress scalars, such as air temperature
and soil moisture, result in downregulation of the maximum LUE (McCallum I.
2009). The GR and TG models estimate the parameter e g using remote sensing
vegetation indices (Table 5.1). The TURC model assumes maximum light-use
efficiency for all biome types without considering the environmental stresses. The
other six PEMs calculate the parameter e g by adjusting maximum light-use efficiency e 0 with soil moisture, air temperature, or CO 2 concentration scalars.
Moreover, the CASA, TURC, C-Fix, and EC-LUE models use the same e 0 across
all vegetation types. GLO-PEM, VPM, and PSN models assign specific e 0 for
different biome types. A brief description of the unique variable calculations of
PEMs is listed below.
5.3.1 The Carnegie-Ames-Stanford Approach (CASA) Model
The CASA model is a simple and mechanical process model, and it describes the
processes of carbon dynamics in terrestrial vegetation and soils (Potter et al. 1993).
The CASA model directly estimates NPP, avoiding complicated calculation of
autotrophic respiration.
In the earlier version of the CASA model, FPAR canopy is estimated with NDVI
and biome type (Potter et al. 1993). The maximum value of FPAR canopy is 0.95.
FPAR canopy is defined as below:
FPAR canopy ¼ min
SR À SR min
SR max À SR min
; 0:95
!
ð5:15Þ
SR ¼
1 þ NDVI
1 À NDVI
ð5:16Þ
where SR min is 1.08 for bare land surface, and SR max is determined by biome type,
ranging from 4.14 to 6.17.
The updated version of the CASA model uses EVI to estimate FPAR chl (Potter
2012), following the work suggested by Xiao et al. (2004a):
FPAR chl ¼ EVI
ð5:17Þ
132
X. Xiao et al.