LSTM比Transformer更适配无观测流域的上游流量预测。
Evaluating Transformer and LSTM Frameworks for Prediction in Ungauged Basins
- 用NWM回溯模拟对比LSTM与纯编码器Transformer性能
- 下游信息使中位数NNSE提升超60%，显著改善预测
- 适合关注水文序列建模与模型架构偏见的研究者
流域网络具有汇聚型拓扑结构，多条支流汇入下游河道，整合了上游多样化的水文过程。在无观测流域中，缺乏直接观测数据导致不确定性增加，限制了极端事件的预测能力。本研究评估了在有限水文信息条件下，仅编码器的Transformer是否优于LSTM用于上游流量推断，基于美国国家海洋和大气管理局（NOAA）国家水文模型（NWM）的回溯模拟。在仅上游和联合配置下，LSTM整体表现均优于Transformer。引入下游信息后，所有模型性能均显著提升，中位数归一化均方误差（NNSE）提升超过60%。我们不将此视为排行榜式比较，而是将其作为对水文序列推断中架构归纳偏置的检验。结果表明，循环记忆机制仍比纯编码器Transformer更契合上游重构任务，而下游水文上下文则提供了强有力的辅助约束，显著提升各类架构的预测能力。
原文摘要 · Abstract (English)
Watershed networks exhibit convergent topologies in which multiple tributaries merge into downstream channels,integrating diverse upstream hydrological processes. In ungauged basins, the absence of direct observations increases uncertainty and limits the ability to anticipate extreme events. This study evaluates whether an encoder-only Transformer provides an advantage over an LSTM for upstream streamflow inference under limited hydrologic information, using retrospective simulations from the NOAA National Water Model (NWM). Across both upstream-only and combined configurations, the LSTM showed stronger overall performance than the Transformer model across the two configurations. Incorporating downstream information further boosted performance for all models, increasing median NNSE by more than 60%. Rather than treating this as a leaderboard-style comparison, we interpret the experiments as a test of architectural inductive bias for hydrologic sequence inference. The results indicate that recurrent memory remains better aligned with this upstream reconstruction task than an encoder-only Transformer, while downstream hydrologic context provides a strong auxiliary constraint that substantially improves prediction skill across architectures
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。