Optimizing Crop Yield Estimation through Geospatial Technology - A Comparative Analysis of a Semi-Physical Model, Crop Simulation, and Machine Learning Algorithms

Aug 5, 2026

Accurate crop yield information is critical for national food security and export planning, especially for wheat at the Gram Panchayat level in Bareilly district, Uttar Pradesh. The study estimates yields by integrating Sentinel-2 time-series and ground observations to build crop type maps and derive spatial variations in growth stages and LAI. Spectral matching, crop-cutting experiment site selection, soil and weather inputs improve precision. A comparison of three yield models shows trade-offs across homogenous areas, data resolution sensitivity, and spatial transferability, supporting near-real-time local decision-making.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages17

Reading Time43 min

AccessFree to download

Introduction

Wheat yield prediction needs and applications

Methodology

Geospatial data integration and crop type mapping

Crop-cutting experiment site selection

Comparative Models

Semi-physical model vs. crop simulation vs. machine learning algorithms

Model performance and transfer limitations

Implications

Support for policymakers and near-real-time local estimates

AgriEngineering Article Optimizing Crop Yield Estimation through Geospatial Technology: A Comparative Analysis of a Semi-Physical Model, Crop Simulation, and Machine Learning Algorithms Murali Krishna Gumma 1, *, Ramavenkata Mahesh Nukala 2, Pranay Panjala 1, Pavan Kumar Bellam 1, Snigdha Gajjala 1, Sunil Kumar Dubey 3, Vinay Kumar Sehgal 4, Ismail Mohammed 1 and Kumara Charyulu Deevi 1 Citation: Gumma, M.K.; Nukala, R.M.; Panjala, P.; Bellam, P.K.; Gajjala, S.; Dubey, S.K.; Sehgal, V.K.; Mohammed, I.; Deevi, K.C. Optimizing Crop Yield Estimation through Geospatial Technology: A Comparative Analysis of a Semi-Physical Model, Crop Simulation, and Machine Learning Algorithms. AgriEngineering 2024, 6, 786–802. [https://doi.org/10.3390/](https://doi.org/10.3390/)[ ](https://doi.org/10.3390/)agriengineering6010045 Academic Editor: Luis A. Ruiz Received: 15 January 2024 Revised: 25 February 2024 Accepted: 4 March 2024 Published: 11 March 2024 Copyright: © 2024 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license ([https://](https://)[ ](https://)[creativecommons.org/licenses/by/](creativecommons.org/licenses/by/)[ ](creativecommons.org/licenses/by/)[4.0/](4.0/)) . 1 Geospatial Sciences and Big Data, International Crops Research Institute for the Semi-Arid Tropics (ICRISAT), Hyderabad 502324, India 2 Faculty of Geo-Engineering, Andhra University, Visakhapatnam 530003, India; [nukalma@yahoo.com](nukalma@yahoo.com) 3 Mahalanobis National Crop Forecast Centre, Delhi 110012, India 4 Indian Agricultural Research Institute, Delhi 110012, India * [Correspondence: muralikrishna.gumma@icrisat.org](Correspondence: muralikrishna.gumma@icrisat.org) Abstract: This study underscores the critical importance of accurate crop yield information for national food security and export considerations, with a specific focus on wheat yield estimation atthe Gram Panchayat (GP) level in Bareilly district, Uttar Pradesh, using technologies such as machine learning algorithms (ML), the Decision Support System for Agrotechnology Transfer (DSSAT) crop model and semi-physical models (SPMs) . The research integrates Sentinel-2 time-series data and ground data to generate comprehensive crop type maps. These maps offer insights into spatial variations in crop extent, growth stages and the leaf area index (LAI), serving as essential components for precise yield assessment. The classification of crops employed spectral matching techniques (SMTs) on Sentinel-2 time-series data, complemented by field surveys and ground data on crop management. The strategic identification of crop-cutting experiment (CCE) locations, based on a combination of crop type maps, soil data and weather parameters, further enhanced the precision of the study. A systematic comparison of three major crop yield estimation models revealed distinctive gaps in each approach. Machine learning models exhibit effectiveness in homogenous areas with similar cultivars, while the accuracy of a semi-physical model depends upon the resolution of the utilized data. The DSSAT model is effective in predicting yields at specific locations but faces difficulties when trying to extend these predictions to cover a larger study area. This research provides valuable insights for policymakers by providing near-real-time, high-resolution crop yield estimates at the local level, facilitating informed decision making in attaining food security. Keywords: crop yield; DSSAT; ML algorithms 1. Introduction Accurate information regarding the productivity of staple crops at their scale is highly essential for successful national planning as well as for ensuring food security at the country level. The application of satellite-based remote sensing emerges as a practical and costeffective strategy for thorough crop monitoring, both at regional and national l

FAQs

What data sources are used to estimate wheat yield at the Gram Panchayat level?

The approach integrates Sentinel-2 time-series data with ground observations to generate crop type maps and capture spatial variation in growth stages and leaf area index (LAI).

How are crop type maps and crop-cutting experiment locations determined?

Crop type mapping uses spectral matching techniques on Sentinel-2 time-series imagery, supported by field surveys and ground crop management data. CCE locations are selected using crop type maps together with soil data and weather parameters.

What are the main differences among the three yield estimation models?

Machine learning models perform well in homogeneous areas with similar cultivars, the semi-physical model accuracy depends on the resolution of the input data, and DSSAT predicts yields well at specific locations but struggles to extend predictions over larger areas.

Aug 5, 2026

Upload by Emma Wilson