Heavy metal contamination in crops is a worldwide problem that requires accurate and timely monitoring.
This study is aimed at improving the accuracy of monitoring heavy metal stress levels in rice utilizing
remote sensing data. An assimilation framework based on remote sensing and improved crop growth
model was developed to continuously monitor heavy metal stress levels over the entire period of crop
growth based on the growth law of crops and the stress mechanism. Compared with other physiological
indices, dry weight of rice roots (WRT) was selected as the best indicator to estimate heavy metal
stress levels. The World Food Study (WOFOST) model, widely used for the description of crop growth,
was improved by incorporating stress factors with overall consideration for the changes in physiological
status under heavy metal stress. Three scenarios were put forward based on the stress factors fDTGA and
fCVF, which, respectively, correspond to the daily total gross assimilation of CO2 and carbohydrate-to-dry
matter conversion coefficient, and were analyzed for their efficiency of simulating WRT. A method of
assimilating the leaf area index (LAI) retrieved from remotely sensed data into the improved WOFOST
model was applied to optimize fDTGA and fCVF. The results suggested that the scenario using both factors
can simulate WRT under heavy metal stress more accurately, with a relative percent error (RPE) lower
than 14%. Based on the RS-WOFOST assimilation framework, continuous spatial-temporal evaluation of
heavy metal stress levels based on WRT can be accomplished.