Estimation of crop gross primary production (GPP): I. impact of MODIS observation footprint and impact of vegetation BRDF characteristics

Abstract

a b s t r a c tAccurate estimation of gross primary production (GPP) is essential for carbon cycle and climate change studies. Three AmeriFlux crop sites of maize and soybean were selected for this study. Two of the sites were irrigated and the other one was rainfed. The normalized difference vegetation index (NDVI), the enhanced vegetation index (EVI), the green band chlorophyll index (CI green ), and the green band wide dynamic range vegetation index (WDRVI green ) were computed from the moderate resolution imaging spectroradiometer (MODIS) surface reflectance data. We examined the impacts of the MODIS observation footprint and the vegetation bidirectional reflectance distribution function (BRDF) on crop daily GPP estimation with the four spectral vegetation indices (VIs -NDVI, EVI, WDRVI green and CI green ) where GPP was predicted with two linear models, with and without offset: GPP = a × VI × PAR and GPP = a × VI × PAR + b. Model performance was evaluated with coefficient of determination (R 2 ), root mean square error (RMSE), and coefficient of variation (CV). The MODIS data were filtered into four categories and four experiments were conducted to assess the impacts. The first experiment included all observations. The second experiment only included observations with view zenith angle (VZA) ≤ 35• to constrain growth of the footprint size,which achieved a better grid cell match with the agricultural fields. The third experiment included only forward scatter observations with VZA ≤ 35• . The fourth experiment included only backscatter observations with VZA ≤ 35• . Overall, the EVI yielded the most consistently strong relationships to daily GPP under all examined conditions. The model GPP = a × VI × PAR + b had better performance than the model GPP = a × VI × PAR, and the offset was significant for most cases. Better performance was obtained for the irrigated field than its counterpart rainfed field. Comparison of experiment 2 vs. experiment 1 was used to examine the observation footprint impact whereas comparison of experiment 4 vs. experiment 3 was used to examine the BRDF impact. Changes in R 2 , RMSE,CV and changes in model coefficients "a" and "b" (experiment 2 vs. experiment 1; and experiment 4 vs. experiment 3) were indicators of the impacts. The second experiment produced better performance than the first experiment, increasing R 2 (↑0.13) and reducing RMSE (↓0.68 g C m −2 d −1 ) and CV (↓9%). For each VI, the slope of GPP = a × VI × PAR in the second experiment for each crop type changed little while the slope and intercept of GPP = a × VI × PAR + b varied field by field. The CI green was least affected by the MODIS observation footprint in estimating crop daily GPP (R 2 , ↑0.08; RMSE, ↓0.42 g C m −2 d −1 ; and CV, ↓7%). Footprint most affected the NDVI (R 2 , ↑0.15; CV, ↓10%) and the EVI (RMSE, ↓0.84 g C m −2 d −1 ). The vegetation BRDF impact also caused variation of model performance and change of model coefficients. Significantly different slopes were obtained for forward vs. backscatter observat...

Estimation of crop gross primary production (GPP): II. Do scaled MODIS vegetation indices improve performance?

“…The coefficientsε 0 andε and LUE chl of each crop per field are listed in Table 2. Corn LUE chl is ∼1.6 times of soybean LUE chl (Table 2), which agrees with the expectation that C4 plants have higher LUE than C3 plants (e.g., Prince, 1991), and explains why maize displays a wider daily GPP range (∼34 g C m −2 d −1 ) than soybean (∼19 g C m −2 d −1 ) (Zhang et al, 2014b). The coefficientsε 0 andε were applied to estimate crop daily GPP.…”

“…However, since the empirical regressions between the VI*PAR products and GPP measured locally at flux towers do not pass through the origin (i.e., the zero X-Y value for regressions) and produce offsets, they are somewhat difficult to interpret and apply Sims et al, 2006;Zhang et al, 2014b). This is considered to be a source of error affecting the accuracy and reliability of remote sensing GPP estimates based on VIs.…”

“…The impact of MODIS observation footprint size resulting from variable view zenith angle (VZA) on crop daily GPP estimation for these sites was recently reported elsewhere (Zhang et al, 2014b). In order to eliminate the potential bias due to large VZAs, only observations with VZA ≤ 35 • were included in this study.…”

“…The MODIS gridding procedure for the standard MOD09 product does not ensure the gridded surface reflectance covers the entire grid (Wolfe et al, 1998). A modified gridding procedure was used for this study (Zhang et al, 2014b), whereby the centers of the three 500 m grids were matched to the centers of the three fields, respectively. The L1B radiance data from each swath were gridded at 500 m resolution for MODIS bands 1-7 with area weight of each MODIS observation.…”

“…Zhang et al (2014a) presented the performance of fAPAR chl and MOD15A2 FPAR in crop GPP estimation, and concluded that fAPAR chl is superior to MOD15A2 FPAR. Zhang et al (2014b) investigated the performance of original un-scaled VIs in GPP estimation, and suggested that further investigation on the performance of scaled VIs should be carried out.…”