Publications

Cao, H; Zhao, RK; Xia, L; Wu, SR; Yang, P (2025). Trends in crop yield estimation via data assimilation based on multi-interdisciplinary analysis. FIELD CROPS RESEARCH, 322, 109745.

Abstract
Context or problem: Data assimilation technology coupling crop growth model and remote sensing inversion, is an important scientific tool for crop growth simulation, which realizes dynamic spatio-temporal monitoring of crop yield. Because data assimilation originated from meteorological, marine and land surface simulation, it lacks consideration of crop growth mechanism and temporal and spatial heterogeneity of field when applied to crop yield estimation. Objective or research question: The objective of this review is to summarize the existing research, reveal the current problems, and propose improved approaches and future directions, based on three important components of crop assimilation system: bridging parameters, assimilation grids and assimilation algorithms, focusing on a multidisciplinary perspective of plant physiology (photosynthetic succession, etc.), crop science (plant morphological structure, etc.), meteorology (new assimilation algorithms, etc.) and other disciplines. Methods: Based on the literature retrieval platform of www.webofscience.com, this study searched the literature in recent 20 years by using the keyword combinations of crop growth model, remote sensing, data assimilation and agriculture, and screened out 110 papers in the field of crop assimilation and yield estimation for reading, trend analysis and summary. Results: The current state of crop yield estimation by assimilation: 1) Leaf area index (LAI) is one of the most commonly used crop canopy state variables, but photosynthesis of non-foliar green organs is usually ignored. 2) The existing researches generally regard the remote sensing image pixels or resampled pixels as the unit of regional yield simulation, lacking the consideration of spatial heterogeneity. 3) The assimilation algorithm has difficulty solving the problem of a lack of observations. Conclusions: In the future research, the improvement direction and measures can be put forward: 1) Considering non-foliar green organs as an integral part of state variables in assimilation system. 2) By setting irregular grids, establishing the division standard of regional heterogeneity, and the optimal assimilation unit grid is obtained. 3) Using the hybrid assimilation algorithm, constructing a variable time window to adapt to the spatio-temporal heterogeneity. Implications or significance: This review is expected to make the assimilation yield estimation system more agronomic and further improve the accuracy of crop yield estimation.

DOI:
10.1016/j.fcr.2025.109745

ISSN:
1872-6852