An improved indicator of simulated grassland production based on MODIS NDVI and GPP data: A case study in the Sichuan province, China

Grassland monitoring is important for both global change research and regional sustainable development. Gross primary production (GPP) is one of the key factors for understanding grass growing conditions. Methods for estimating GPP are plentiful, and the light use efficiency (LUE) model based on remote sensing data is widely used. The MODIS GPP product, which is employed by the National Aeronautics and Space Administration (NASA), is calculated using the LUE model and the surface reflection data from the Moderate Resolution Imaging Spectroradiometer onboard the Terra/Aqua satellite. The MODIS GPP product harbors its own uncertainties arising from the sources and parameters, such as FPAR and light use efficiency (epsilon). In this study, we propose an improved indicator for monitoring grassland based on MODIS GPP and NDVI data. Fractional vegetation coverage and the percentage of grass area (1 km(2)) were used to reduce the mixed pixel effect. A function of NDVI was used to simulate the light use efficiency and FPAR. The modified GPP data were calculated and validated with in situ measured data from the Sichuan province, China, 2011. The results indicated that the modified GPP data were a more accurate indicator for monitoring grassland than previous indicators, and the precision of grass production simulated by SsGPP(ndvi) reached 85.6%. Spatial statistic results were consistent with the practical condition in most cases. Since MODIS data are available twice a day, the improved indicator can meet the actual requirement of grassland monitoring at regional scale. (C) 2014 Elsevier Ltd. All rights reserved