pipette
ENEnglish

An Input-Frugal Deep Learning Framework for Weather-Driven National Crop-Yield Forecasting: A Case Study of Brazilian Soybean

Fernando Dupin da Cunha Mello, Prashant Kumar, Erick G. Sperandio Nascimento Stricto Sensu Department, SENAI CIMATEC University, Salvador, Bahia, Brazil, Global Centre for Clean Air Research, School of Engineering, Civil, Environmental Engineering, Faculty of Engineering, Physical Sciences, University of Surrey, Guildford GU2 7XH, United Kingdom, Institute for Sustainability, University of Surrey, Guildford, GU2 7XH, United Kingdom, Surrey Institute for People-Centred AI, Faculty of Engineering, Physical Sciences, University of Surrey, Guildford GU2 7XH, UK)

PreprintAfirmaciones fuertes, leer con cuidadoUso en el mundo real

En palabras de los autores

Reliable, timely crop-yield forecasts are essential for market stability and risk management, yet many approaches rely on costly or hard-to-scale inputs. We present a frugal, transferable, and architecture-agnostic deep learning framework that uses routine weather as the only time-varying input plus two lightweight static context inputs (crop year and an agro-environmental label) to capture long-run change and regional heterogeneity, while supporting multiple sequence encoders under identical data requirements. Using a 20-season Brazilian soybean case study (2001/02-2020/21) with leave-one-year-out cross-validation, we benchmark MLP, CNN, LSTM, CNN-LSTM, a Transformer encoder and the Mamba state-space model against linear ridge regression and a five-year moving-average "farmer" baseline. All deep learning variants outperform ridge, and all sequential encoders surpass the non-sequential MLP. The Transformer achieves the best national accuracy (RMSE 149 kg ha^-1; rRMSE 5.3%; R^2 = 0.784), reducing error by 47.6% relative to the farmer baseline. In-season forecasts improve monotonically from early- to late-season issuance, reaching approximately 50% lower error than the baseline at the latest forecast point. Ablations indicate that the agro-environmental label and spatial instance expansion (multiple grid-node weather sequences per municipality-year) contribute positively without increasing input complexity. SHAP diagnostics suggest crop year explains most of the long-run trajectory, whereas within-season weather and agro-environmental context primarily drive interannual deviations, with moisture/cloud and thermal-demand variables dominating. Overall, the framework is straightforward to deploy across other crops and geographic regions and is naturally compatible with operational weather forecasts for routine monitoring.

Resultado principalEl resumen no menciona limitaciones.

Apareció: jueves, 1 de octubre. arXiv. Preprint, todavía sin revisión por pares.