发表机构
SENAI CIMATEC University; University of Surrey（SENAI CIMATEC大学; 萨里大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
提出一种仅用常规天气和轻量静态输入的通用深度学习框架，在巴西大豆20季数据上验证，Transformer模型达最优精度，误差较基线降低47.6%，且易推广至其他作物和区域。
AI 中文摘要
可靠、及时的作物产量预测对于市场稳定和风险管理至关重要，然而许多方法依赖于成本高昂或难以扩展的输入。我们提出了一种节约、可迁移且架构无关的深度学习框架，该框架仅使用常规天气作为唯一的时间变化输入，外加两个轻量级静态上下文输入（作物年份和农业环境标签），以捕捉长期变化和区域异质性，同时在相同数据需求下支持多种序列编码器。利用巴西大豆20个生长季（2001/02-2020/21）的案例研究，采用留一年交叉验证，我们将MLP、CNN、LSTM、CNN-LSTM、Transformer编码器和Mamba状态空间模型与线性岭回归和五年移动平均“农民”基线进行基准比较。所有深度学习变体均优于岭回归，所有序列编码器均超过非序列MLP。Transformer实现了最佳的全国精度（RMSE 149 kg ha^-1；rRMSE 5.3%；R^2 = 0.784），相对于农民基线误差降低了47.6%。季节内预测从早期到晚期发布单调改善，在最新预测点误差比基线低约50%。消融研究表明，农业环境标签和空间实例扩展（每个市-年多个网格节点天气序列）在增加输入复杂性的情况下贡献积极。SHAP诊断表明，作物年份解释了大部分长期轨迹，而季节内天气和农业环境背景主要驱动年际偏差，其中湿度/云量和热需求变量占主导。总体而言，该框架易于部署到其他作物和地理区域，并自然兼容用于常规监测的业务天气预报。
英文摘要
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.
