Interpretable Machine Learning for Multi-Crop Yield Prediction in Semi-Arid Regions: A Hierarchical Approach to Handle Climate Data Sparsity

Rachid Ed-daoudi; M'barek El Haloui

doi:10.22194/JGIAS/25.1694

Author(s)

Rachid Ed-daoudi LyRICA: Laboratory of Research in Informatics, Data Sciences and Artificial Intelligence, School of Information Sciences,B.P. 604, Rabat-Instituts, Rabat,Morocco

M'barek El Haloui LyRICA: Laboratory of Research in Informatics, Data Sciences and Artificial Intelligence, School of Information Sciences,B.P. 604, Rabat-Instituts, Rabat,Morocco

Abstract

This study develops a hierarchical machine learning framework to address the challenges of multi-crop yield prediction in semi-arid regions, focusing on sparse climate data, model interpretability, and heterogeneous climate-crop interactions. The framework integrates Random Forest (RF) techniques for crop-specific modeling, exploiting their robustness to nonlinear relationships in small datasets, while long short term memory (LSTM) networks extract long-term climate trends, and Convolutional Neural Network (TCNN) with dilated kernels capture multi-scale temporal dependencies, creating a complementary hierarchy that leverages the strengths of each approach. Missing weather data (gaps of 8.3-45.2%) are addressed via Multiple Imputation by Chained Equations (MICE), incorporating climatological constraints to preserve physical consistency. Model interpretability is achieved through SHapley Additive exPlanations (SHAP) analysis and uncertainty decomposition, quantifying the contributions of data variability, temporal dynamics, and model ensembles. Validated on 13 years of agricultural data (2010-2022) from the semi-arid Ouarzazate region of Morocco, the framework achieved R² values of 0.77 (cereals), 0.73 (vegetables), and 0.70 (tree crops), representing a 12-18% improvement over conventional single-model approaches which typically achieve R² values of 0.65-0.68 in similar semi-arid conditions with complete datasets. SHAP analysis identified crop-specific critical thresholds, including maximum summer temperatures (>40°C) for vegetables and winter precipitation (<30mm) for cereals, while uncertainty quantification revealed that transition seasons (spring/autumn) are high-risk periods with prediction variances exceeding ±1.5 T/ha. Performance declines during extreme weather events (e.g., unprecedented droughts) that are absent from historical data, and soil moisture dynamics remain excluded. Future work will address these limitations through integration of soil moisture sensors, development of transfer learning approaches for extreme weather events. Practical applications include optimizing irrigation schedules using SHAP-derived temperature thresholds and prioritizing resource allocation during periods of high uncertainty. This work represents a novel unified framework combining imputation, temporal modeling, and crop-specific interpretability for semi-arid systems, going beyond black-box models or single-crop approaches.

Publication Details

Page(s) 1217-1228

DOI 10.22194/JGIAS/25.1694

Published Journal: Journal of Global Innovations in Agricultural Sciences, Volume: 13, Issue: 3, Year: 2025

Request the full text from PASTIC

Keywords

missing meteorological data., shap interpretability, Hierarchical machine learning, multi-crop systems, agricultural yield prediction

References

Adraoui I.,Jaafar B. .2024 .Sustainable management of water resources and assessment of the vulnerability of Moroccan oases to climate change. Environmental Science and Pollution Research, 31 : 17981-17993.

Altieri M.A C.I.,Nicholls C.I. .2017 .The adaptation and mitigation potential of traditional agriculture in a changing climate. Climatic Change, 140 : 33-45.

Attri I.,Awasthi L.K.,Sharma T.P.,Rathee P. .2023 .A review of deep learning techniques used in agriculture. Ecological Informatics, 75 : 102217.

Badakhshan E. ,Vaunat J.,Veylon G. .2024 .Meteorological and vegetation effects on the thermal analysis of slopes. Renewable and Sustainable Energy Reviews, 196 : 114352.

Baig F. ,Ali L.,Faiz M.A.,Chen H.,Sherif M. .2024 .How accurate are the machine learning models in improving monthly rainfall prediction in hyper arid environment. Journal of Hydrology, 633 : 131040.

Bantchina B.B. ,M. Qaswar S.,Arslan Y.,Ulusoy K.S.,Gündoğdu Y.,Tekin A.M.,Mouazen A.M. .2024 .Corn yield prediction in site-specific management zones using proximal soil sensing, remote sensing, and machine learning approach. Computers and Electronics in Agriculture, 225 : 109329.

Cravero A ,Pardo S.,Sepúlveda S.,Muñoz L. .2022 .Challenges to use machine learning in agricultural big data: a systematic literature review. Agronomy, 12 : 748.

Dhakshayani J. ,Surendiran B.,Jyothsna J. .2024 .Artificial intelligence in precision agriculture: a systematic review on tools, techniques, and applications. , : 37-57.

Dong X.,Li B.,Yan Z.,Guan L.,Huang S.,Li S.,Yang H. .2024 .Impacts of high temperature, relative air humidity, and vapor pressure deficit on the seed set of contrasting maize genotypes during flowering. Journal of Integrative Agriculture, 23 : 2955-2969.

Drogkoula M.,Kokkinos K.,Samaras N. .2023 .A comprehensive survey of machine learning methodologies with emphasis in water resources management. Applied Sciences, 13 : 12147.

Ed-Daoudi R.,Alaoui A.,Ettaki B.,Zerouaoui J. .2023 .Improving crop yield predictions in Morocco using machine learning algorithms. Journal of Ecological Engineering, 24 : 162769.

El Kenawy A.M. .2024 .Hydroclimatic extremes in arid and semi-arid regions: status, challenges, and future outlook. In Hydroclimatic extremes in the Middle East and North Africa. Elsevier, : 1-22.

Gautam S. ,Siddharth K.,Kashyap P.,Kamani P.,Kanojia M.,Wazir S.,Malik K.,Sehgal V.K.,Dhakar R. .2021 .Revolutionizing agriculture: a comprehensive review of artificial intelligence techniques in farming. In Artificial Intelligence in Agriculture. Taylor & Francis, : 101-121.

Gürkan H. O.,Eskioğlu Y.,Çalık M.,Ekici S.,Yağan H.,Eroğlu H.,Doğan H. .2022 .Assessment of growing degree days (GDDGROWN) index with climate projections. , : .

Houssni M.,Kassout J.,El Mahroussi M.,Chakkour S.,Kadiri M.,Ater M.,Petrisor A.I. .2023 .Evaluation and structuring of agrodiversity in oases agroecosystems of southern Morocco. Agriculture, 13 : 1413.

Inouye D.W. .2022 .Climate change and phenology. Wiley Interdisciplinary Reviews: Climate Change, 13 : e764.

Ippolito M.,D. De Caro M.,Cannarozzo G.,Ciraolo G. .2024 .Evaluation of daily crop reference evapotranspiration and sensitivity analysis of FAO Penman-Monteith equation using ERA5-Land reanalysis database in Sicily, Italy. Agricultural Water Management, 295 : 108732.

Jabed M.A M.A.A.,Murad M.A.A. .2024 .Crop yield prediction in agriculture: a comprehensive review of machine learning and deep learning approaches, with insights for future research and sustainability. Heliyon, 10 : e40836.

Khaldi L. ,El Bilali A.,El Abed A.,Krakauer N.,Khanchoufi A. El .2025 .Developing an explainable and interpretable machine learning model for flood susceptibility mapping. Ecological Engineering and Environmental Technology, 1 : 201-215.

Khangamwa G. ,C.J. van Alten. .2022 .Towards a methodology for addressing missingness in datasets, with an application to demographic health datasets. arXiv 2211, : .

Mandal N. ,Chanda K. .2023 .Performance of machine learning algorithms for multi-step ahead prediction of reference evapotranspiration across various agroclimatic zones and cropping seasons. Journal of Hydrology, 620 : 129418.

Marcinkevičs R. ,Vogt J.E. .2023 .Interpretable and explainable machine learning: a methods‐centric overview with concrete examples. Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery, 13 : e1493.

Naorem A. .2023 .Soil constraints in an arid environment - challenges, prospects, and implications. Agronomy, 13 : 220.

Nayak H.S. J.V.,Silva C.M.,Parihar T.J.,Krupnik D.R.,Sena S.K.,Kakraliya T.B.,Sapkota T.B. .2022 .Interpretable machine learning methods to explain onfarm yield variability of high productivity wheat in Northwest India. Field Crops Research, 287 : 108640.

Office Régional de Mise en .2025 .. Official website of the ORMVA of Ouarzazate. Retrieved January 20, : .

Park S.J. C.S.,Hwang P.L.G.,Vlek P.L.G. .2005 .Comparison of adaptive techniques to predict crop yield response under varying soil and land management conditions. Agricultural Systems, 85 : 59-81.

Praveen B. and P.,Sharma B. and P. .2020 .Climate variability and its impacts on agriculture production and future prediction using autoregressive integrated moving average method (ARIMA). Journal of Public Affairs, 20 : e2016.

Saini P. ,Nagpal B. .2024 .Analysis of missing data and comparing the accuracy of imputation methods using wheat crop data. Multimedia Tools and Applications, 83 : 40393-40414.

Singh P. ,Sehgal V.K.,Dhakar R.,Neale C.M.,Goncalves I.Z.,Rani A.,Dubey S.K. .2024 .Estimation of ET and crop water productivity in a semi-arid region using a large aperture scintillometer and remote sensingbased SETMI model. Water, 16 : 422.

Sudhamathi T.,Perumal K. .2024 .Ensemble regression based Extra Tree Regressor for hybrid crop yield prediction system. Measurement: Sensors, 35 : 101277.

Thulare E. ,Ajoodha R.,Jadhav A. .2021 .An empirical analysis and application of the expectation-maximization and matrix completion algorithms for varying degrees of missing data. IEEE, : 1-7.

Tursunalieva A. ,Alexander D.L.,Dunne R.,Li J.,Riera L.,Zhao Y. .2024 .Making sense of machine learning: a review of interpretation techniques and their applications. Applied Sciences, 14 : 496.

Tzoumpas K.,Estrada A.,Miraglio P.,Zambelli P. .2024 .A data filling methodology for time series based on CNN and (Bi)LSTM neural networks. IEEE Access, 12 : 31443-31460.

Zhang S. ,Duan J.,Qi X.,Gao Y.,He L.,Liu L.,Feng W. .2024 .Combining spectrum, thermal, and texture features using machine learning algorithms for wheat nitrogen nutrient index estimation and model transferability analysis. Computers and Electronics in Agriculture, 222 : 109022.

Zhang Y. ,Thorburn P.J. .2022 .Handling missing data in near real-time environmental monitoring: a system and a review of selected methods. Future Generation Computer Systems, 128 : 63-72.