Integrating remote sensing-based process model with environmental zonation scheme to estimate rice yield gap in Northeast China

Abstract

Yield gap (YG) analysis is a powerful method to reveal the possible opportunities for fulfilling the growing food demand. It used to be restricted to a small geographic area, and thus the scope of untapped yield often remains unclear at macro scale. To this end, this study presents an improved remote sensing approach to assessing regional rice YG in northeast China (NEC) during the period from 2006 to 2017. A satellite-based biophysical model (BEPS) was used to derive the actual yield (Ya) and its spatial patterns. A machine-learning zonation scheme was further developed to estimate the potential yield for farmers (Yp) within domains having similar climatic, geomorphic, and edaphic context. Results indicate that the BEPS model can provide reliable estimates of rice yields in this region, with RMSE below 20 % at the county level; and the novel zonation scheme enables better portrayal of the spatial variation in Yp. To identify areas with the greatest potential to narrow the YG, we proposed a quantitative method for dividing NEC into four parts with different priorities for future yield improvement. In general, the exploitable YG in NEC was 2599 kg ha−1, amounting to 24.7 % of Yp. Southern NEC possessed substantial YG with the primary priority to be explored; whereas in the north, Ya already approximating Yp, further yield increase was limited and challenging. Growing degree days (≥10 °C), cumulative solar radiation and elevation were all significant limiting factors of Yp. This study demonstrates the ability of remote sensing approach to assessing regional YG. Meanwhile, this regional assessment can support planners to set practical yield target and to prioritize regions and needs for future exploitation and researches.