arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2203.14827cs.LG
收藏

具有物理输出的可区分、可学习、区域化的基于过程的模型可以接近最先进的水文预测精度

Differentiable, learnable, regionalized process-based models with physical outputs can approach state-of-the-art hydrologic prediction accuracy

The Pennsylvania State University(宾夕法尼亚州立大学)

Dapeng Feng, Jiangtao Liu, Kathryn Lawson, Chaopeng Shen

摘要:

整个水循环中水文变量的预测对于水资源管理以及下游应用(如生态系统和水质建模)具有重要价值。最近,纯数据驱动的深度学习模型(如长短期记忆(LSTM))在降雨径流和其他地球科学变量建模方面表现出似乎难以克服的性能,但它们无法预测未经训练的物理变量,并且仍然具有挑战性解释。在这里,我们展示了可微分、可学习、基于过程的模型(这里称为δ模型)可以接近LSTM的性能水平,用于区域化参数化的密集观测变量(流量)。我们使用一个简单的水文模型HBV作为骨干,并使用嵌入式神经网络,它只能在一个可微的编程框架中进行训练,以参数化,增强或替换基于过程的模型模块。在不使用集合或后处理器的情况下,δ模型可以为Daymet强迫数据集获得美国671个流域的中值Nash Sutcliffe效率为0.732,而具有相同设置的最先进的LSTM模型为0.748。对于另一个强迫数据集,差异甚至更小:0.715 vs. 0.722。同时,所得到的可学习的基于过程的模型可以输出一整套未经训练的变量,例如,土壤和地下水储存,积雪,蒸散和基流,后来受到他们的观察。模拟的蒸散量和基流排放量与其他估计值一致。通用框架可以处理具有各种过程复杂性的模型,并为从大数据中学习物理开辟了道路。

英文摘要:

Predictions of hydrologic variables across the entire water cycle have significant value for water resource management as well as downstream applications such as ecosystem and water quality modeling. Recently, purely data-driven deep learning models like long short-term memory (LSTM) showed seemingly-insurmountable performance in modeling rainfall-runoff and other geoscientific variables, yet they cannot predict untrained physical variables and remain challenging to interpret. Here we show that differentiable, learnable, process-based models (called δ models here) can approach the performance level of LSTM for the intensively-observed variable (streamflow) with regionalized parameterization. We use a simple hydrologic model HBV as the backbone and use embedded neural networks, which can only be trained in a differentiable programming framework, to parameterize, enhance, or replace the process-based model modules. Without using an ensemble or post-processor, δ models can obtain a median Nash Sutcliffe efficiency of 0.732 for 671 basins across the USA for the Daymet forcing dataset, compared to 0.748 from a state-of-the-art LSTM model with the same setup. For another forcing dataset, the difference is even smaller: 0.715 vs. 0.722. Meanwhile, the resulting learnable process-based models can output a full set of untrained variables, e.g., soil and groundwater storage, snowpack, evapotranspiration, and baseflow, and later be constrained by their observations. Both simulated evapotranspiration and fraction of discharge from baseflow agreed decently with alternative estimates. The general framework can work with models with various process complexity and opens up the path for learning physics from big data.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑