Early season prediction of within-field crop yield variability by assimilating CubeSat data into a crop model

Matteo G Ziliani,Muhammad U Altaf,Bruno Aragon,Rasmus Houborg,Trenton E Franz,Yang Lu,Justin Sheffield,Ibrahim Hoteit,Matthew F Mccabe

doi:10.1016/j.agrformet.2021.108736

Abstract

Accurate early season predictions of crop yield at the within-field scale can be used to address a range of crop production, management, and precision agricultural challenges. While the remote sensing of within-field insights has been a research goal for many years, it is only recently that observations with the required spatio-temporal resolutions, together with efficient assimilation methods to integrate these into modeling frameworks, have become available to advance yield prediction efforts. Here we explore a yield prediction approach that combines daily high-resolution CubeSat imagery with the APSIM crop model. The approach employs APSIM to train a linear regression that relates simulated yield to simulated leaf area index (LAI). That relationship is then used to identify the optimal regression date at which the LAI provides the best prediction of yield: in this case, approximately 14 weeks prior to harvest. Instead of applying the regression on satellite imagery that is coincident, or closest to, the regression date, our method implements a particle filter that integrates CubeSat-based LAI into APSIM to provide end-of-season high-resolution (3 m) yield maps weeks before the optimal regression date. The approach is demonstrated on a rainfed maize field located in Nebraska, USA, where suitable collections of both imagery and in-situ data were available for assessment. The procedure does not require in-field data to calibrate the regression model, with results showing that even with a single assimilation step, it is possible to provide yield estimates with good accuracy up to 21 days before the optimal regression date. Yield spatial variability was reproduced reasonably well, with a strong correlation to independently collected measurements (R2 = 0.73 and rRMSE = 12%). When the field averaged yield was compared, our approach reduced yield prediction error from 1 Mg/ha (control case based on a calibrated APSIM model), to 0.5 Mg/ha (using satellite imagery alone), and then to 0.2 Mg/ha (results with assimilation up to three weeks prior to the optimal regression date). Such a capacity to provide spatially explicit yield predictions early in the season has considerable potential to enhance digital agricultural goals and improve end-of-season yield predictions.

Highlights

The delivery of timely crop monitoring and accurate crop yield es timates, at both the field and regional scales, remains a precision agri culture and food security goal (Fritz et al, 2019)
It should be noted that even though our study focuses on yield predictions using Agricultural Production System Simulator (APSIM), the approach is generic, and other models could be used with minor framework modifications
Our results demonstrate the benefit of integrating CubeSat data into the APSIM crop model to enhance yield prediction at both the field and within-field scales

Summary

Introduction

The delivery of timely crop monitoring and accurate crop yield es timates, at both the field and regional scales, remains a precision agri culture and food security goal (Fritz et al, 2019). Accurate pre-harvest predictions of crop yield at the field scale would help farmers to tailor site-specific management decisions regarding crop inputs (i.e. timing and amount of fertilizer and irrigation), and to forecast profit based on spatially explicit yield estimates (Tewes et al, 2020). Are increasingly being used to predict crop yield Models such as the Agricultural Production System Simulator (APSIM) (Holzworth et al, 2014) or the Decision Support System for Agrotechnology Transfer (DSSAT) (Jones et al, 2003) have been applied to describe the temporal development of key crop processes throughout the growing season, providing insight into photosynthesis, soil dynamics, biomass parti tioning, and yield accumulation, using climate, soil, and field manage ment information (Hoogenboom, 2000). Crop models are employed for farming operations at a range of scales (Tewes et al, 2020), they were originally developed for point-based applications that neglect spatial variation in yield forecast

Methods

Results

Discussion

Conclusion