Predicting lake surface water phosphorus dynamics using process-guided machine learning
Predicting lake surface water phosphorus dynamics using process-guided machine learning
复制标题
DOI：
10.1016/j.ecolmodel.2020.109136
复制
发表时间：
2020-08-15
影响因子：
3.1
通讯作者：
Kumar, Vipin
中科院分区：
文献类型：
--
作者：
Hanson, Paul C.;Stillman, Aviah B.;Kumar, Vipin
湖泊中的磷 (P) 正在降低全球水的质量和可用性。需要准确预测湖泊磷动态，以了解整个生态系统磷预算，以及改变湖泊磷浓度对水质的影响。然而，湖泊内复杂的生物物理过程以及有限的观测数据，对我们再现水质预测所需的短期湖泊动态以及了解 P 湖大范围控制所需的长期动态的能力提出了挑战。在这里，我们使用过程引导机器学习 (PGML) 建模中的新兴范例，为门多塔湖（美国威斯康星州）生成磷预算，并准确预测几天到几十年的时间范围内的表湖磷。在我们的 PGML（我们称之为过程引导循环神经网络（PGRNN））的实现中，我们将 P 湖基于过程的模型与循环神经网络相结合，然后用生态原理约束预测。我们独立测试基于过程的模型、循环神经网络和 PGRNN 以评估整体方法。基于过程的模型解释了 P 湖观测到的大部分模式；但它错过了湖泊 P 的长期趋势，并且在预测地表水冬夏季 P 方面表现最差。基于过程的模型、循环神经网络和 PGRNN 的均方根误差 (RMSE) 分别为 33.0 μg P L-1、22.7 μg P L-1 和 20.7 μg P L-1。所有模型在夏季均表现较好，三个模型（同一顺序）的 RMSE 值分别为 14.3 μg P L-1、10.9 μg P L-1 和 10.7 μg P L-1。尽管 PGRNN 在夏季的 RMSE 只稍好一些，但它的偏差较低，并且再现了其他两个模型错过的 P 湖的长期下降。对于所有季节和所有年份，循环神经网络比单独处理具有更好的预测，均方根误差 (RMSE) 分别为 23.8 μg P L-1 和 28.0 μg P L-1。 PGRNN 的输出表明，需要与水温、热分层和外部载荷的长期变化相关的新过程来改进过程模型。通过使用生态知识以及复杂数据的信息内容，PGML 显示出作为一种在混乱的现实世界生态动态中进行准确预测的技术的前景，同时提供了可以提高我们对过程的理解的有价值的信息。
Phosphorus (P) loading to lakes is degrading the quality and usability of water globally. Accurate predictions of lake P dynamics are needed to understand whole-ecosystem P budgets, as well as the consequences of changing lake P concentrations for water quality. However, complex biophysical processes within lakes, along with limited observational data, challenge our capacity to reproduce short-term lake dynamics needed for water quality predictions, as well as long-term dynamics needed to understand broad scale controls over lake P. Here we use an emerging paradigm in modeling, process-guided machine learning (PGML), to produce a phosphorus budget for Lake Mendota (Wisconsin, USA) and to accurately predict epilimnetic phosphorus over a time range of days to decades. In our implementation of PGML, which we term a Process-Guided Recurrent Neural Network (PGRNN), we combine a process-based model for lake P with a recurrent neural network, and then constrain the predictions with ecological principles. We test independently the process-based model, the recurrent neural network, and the PGRNN to evaluate the overall approach. The process-based model accounted for most of the observed pattern in lake P; however it missed the long-term trend in lake P and had the worst performance in predicting winter and summer P in surface waters. The root mean square error (RMSE) for the process-based model, the recurrent neural network, and the PGRNN was 33.0 mu g P L-1, 22.7 mu g P L-1, and 20.7 mu g P L-1, respectively. All models performed better during summer, with RMSE values for the three models (same order) equal to 14.3 mu g P L-1, 10.9 mu g P L-1, and 10.7 mu g P L-1. Although the PGRNN had only marginally better RMSE during summer, it had lower bias and reproduced long-term decreases in lake P missed by the other two models. For all seasons and all years, the recurrent neural network had better predictions than process alone, with root mean square error (RMSE) of 23.8 mu g P L-1 and 28.0 mu g P L-1, respectively. The output of PGRNN indicated that new processes related to water temperature, thermal stratification, and long term changes in external loads are needed to improve the process model. By using ecological knowledge, as well as the information content of complex data, PGML shows promise as a technique for accurate prediction in messy, real-world ecological dynamics, while providing valuable information that can improve our understanding of process.