Production Capacity Evaluation of Farmland Using Long Time Series of Remote Sensing Images

Farmland is a crucial resource for the survival and evolution of humans. The accurate evaluation farmland production capacity (FPC) great significance planting structure optimization, improvement low-yield sustainable utilization. objective this study to quantitatively evaluate FPC at county scale using time series remote sensing (RS) images. Taking winter wheat as benchmark crop, relations between annual yield Normalized Difference Vegetation Index (NDVI) were established by multiple linear regression algorithm. mean standard deviations (SD) multi-year used its instability parcels basic unit. results show that estimation model performed best in 2011. R2 modeling sample was 0.93, RMSE testing 368.1 kg/ha. grades south north area relatively high with good stability, while those center low poor stability. There certain correlation soil organic matter (SOM), coefficient 0.525 (p < 0.01). In study, taking parcel unit instead pixel, long multi-source RS images medium resolution monitor per crops then FPC. This can provide method rapid scale.