Exploring the exceptional performance of a deep learning stream temperature model and the value of streamflow data
Exploring the exceptional performance of a deep learning stream temperature model and the value of streamflow data
复制标题
DOI：
10.1088/1748-9326/abd501
复制
发表时间：
2021-02-01
影响因子：
6.7
通讯作者：
Shen, Chaopeng
中科院分区：
文献类型：
--
作者：
Rahmani, Farshid;Lawson, Kathryn;Shen, Chaopeng
河流水温（Ts）是水生生态系统健康的一个至关重要的变量。T-S受到地下水-地表水相互作用的强烈影响，这可以从径流记录中了解到，但由于参数等同性，以前这种信息很难有效地吸收基于过程的模型。基于长短期记忆（LSTM）深度学习架构，我们开发了一个以流域为中心的集总日平均T-s模型，该模型在美国境内118个数据丰富的流域（没有大型水坝）上进行了训练，并显示出了良好的结果。在全国范围内，我们得到了0.69摄氏度的中位数均方根误差，纳什-萨克利夫模型的效率系数为0.985，相关系数为0.994，这是显着的改善，比以前的值在文献中报道。作为模型输入的径流观测值的增加大大提高了该模型的性能。在没有测量流量的情况下，我们证明了可以使用两阶段模型，其中来自预训练LSTM模型（Q（sim））的模拟流量仍然有利于T-s模型，即使没有新信息直接引入T-s模型的输入。该模型间接使用的信息，从径流观测提供的Q（SIM）的训练过程中，有可能提高物理意义的变量的内部表示。我们的研究结果表明，流域平均强迫变量，流域属性和T-S之间存在很强的关系，可以通过一个单一的模型训练的数据在大陆尺度上模拟。
Stream water temperature (T-s) is a variable of critical importance for aquatic ecosystem health. T-s is strongly affected by groundwater-surface water interactions which can be learned from streamflow records, but previously such information was challenging to effectively absorb with process-based models due to parameter equifinality. Based on the long short-term memory (LSTM) deep learning architecture, we developed a basin-centric lumped daily mean T-s model, which was trained over 118 data-rich basins with no major dams in the conterminous United States, and showed strong results. At a national scale, we obtained a median root-mean-square error of 0.69 degrees C, Nash-Sutcliffe model efficiency coefficient of 0.985, and correlation of 0.994, which are marked improvements over previous values reported in literature. The addition of streamflow observations as a model input strongly elevated the performance of this model. In the absence of measured streamflow, we showed that a two-stage model could be used, where simulated streamflow from a pre-trained LSTM model (Q(sim)) still benefited the T-s model even though no new information was brought directly into the inputs of the T-s model. The model indirectly used information learned from streamflow observations provided during the training of Q(sim), potentially to improve internal representation of physically meaningful variables. Our results indicate that strong relationships exist between basin-averaged forcing variables, catchment attributes, and T-s that can be simulated by a single model trained by data on the continental scale.