GPP and maximum light use efficiency estimates using different approaches over a rotating biodiesel crop

Abstract

This paper presents: (a) results of gross primary production (GPP) 8-d estimated values using a lightuse efficiency model (LUE) in a non-irrigated rotating rapeseed crop in the upper Spanish plateau, and(b) inter-comparison results of observed GPP with those concurrently retrieved by MODIS. The rotation scheme over the four-year study comprised rapeseed, wheat, peas and rye. Rapeseed, peas and, in part, rye grew under well-watered conditions whereas wheat was dominated by drought.Input data for the LUE model were the fraction of PAR absorbed (FPAR) 8-d products supplied by MODIS(FPARMODIS), in situ photosynthetic active radiation (PAR) measurements and a scalar f varying between 0and 1, to take into account the reduction of the maximum PAR conversion efficiency (ε0LUE) under limiting environmental conditions. In this study, f values were assumed to be dependent on air temperature (T)and the evaporative fraction which was considered a proxy of water availability. ε0LUE, a key parameterin LUE models, which varied according to land use, was derived through the results of a linear regressionfit between observed GPP and concurrent GAPAR estimates defined as the product of PAR, FPARMODISandf. Overall, the LUE model provided satisfactory results, R2= 86.3%, significantly improving GPP MODIS estimates (GPPMODIS), R2= 71.8%. GPPMODIS uncertainties have primarily been attributed to differences inthe f stress factor involved in its formulation (fMODIS) depending on vapour pressure deficit and T which did not fully describe the environmental stress conditions at the measuring site.Overall, ε0LUE yielded 3.33 ± 0.10 g C MJ−1although this varied depending on crop architecture, phenology and prevailing meteorological conditions. Crop-to-crop ε0LUEranged from 2.74 ± 0.17 to3.95 ± 0.19 g C MJ−1for peas and rye, respectively, yielding intermediate values for rapeseed andwheat, 2.92 ± 0.18 and 2.86 ± 0.23 g C MJ−1, respectively. ε0MODIS, derived from the linear fit of GPPversus GPPMODISestimates, yielded 2.13 ± 0.10 g C MJ−1and crop-to-crop ranged from 1.28 ± 0.17 to2.41 ± 0.12 g C MJ−1for wheat and rapeseed, respectively. The best linear fits corresponded to crops grow-ing under well-watered conditions, rapeseed and peas, and the worst fits were for wheat, affected bydrought. GPP annuals were 1680, 710, 730 and 1410 g C m−2 for rapeseed, wheat, peas and rye, respectively.