Rice Yield Estimation Based on an NPP Model With a Changing Harvest Index

Abstract

Paddy rice is one of the most widely planted crops in Asia. Yield estimation of paddy rice is crucial to food security. Gross primary productivity or net primary productivity (NPP) models are some of the most commonly used methods for crop yield estimation as they have a theoretical basis and are simple to use. The harvest index (HI) of paddy rice, one of the input parameters in yield estimation models, has been increasingly used with the improvement of paddy rice cultivars over the past four decades. However, in many previous researches, this parameter was arbitrarily determined to fit the observed yield data values, without considering its changing trend. In the present study, we developed an NPP-based rice yield estimation model with HI being a changing parameter. The accuracy of the proposed method is tested over the Jiangsu Province, Southeast China. The results showed that the rice HI increases linearly in study area. Compared with yield estimation with a fixed HI, yield estimations using a changing HI are greatly improved, with the average estimation accuracy greater than 96%, and the relative errors within ±5%. The results proved that the NPP-based yield estimation model with a changing HI can be a promising alternative for rice yield estimation.

Keywords

• Gross primary productivity (GPP)
• harvest index (HI)
• net primary productivity (NPP)
• paddy rice yield estimation
• rectangular hyperbolic model