In the Great Plains, playas are critical wetland habitats for migratory birds and a source of recharge for the agriculturally important High Plains aquifer. The temporary wetlands exhibit complex hydrology, filling rapidly via local rain storms and then drying through evaporation and groundwater infiltration. Using a long short-term memory (LSTM) neural network to account for these complex processes, we modeled the probability of playa inundation for 71,842 playas in the Great Plains from 1984 to 2018. At the level of individual playas, the model achieved an F1-score of 0.522 on a withheld test set, displaying the ability to predict complex inundation patterns. When simulating playa inundation over the entire region, the model is able to very closely track inundation trends, even during periods of drought. Our results demonstrate potential for using LSTMs to model complex hydrological dynamics. Our modeling approach could be used to model playa inundation into the future under different climate scenarios to better understand how wetland habitats and groundwater will be impacted by changing climate.

中文翻译：

---

使用长短期记忆神经网络预测 Playa 洪水

在大平原，滩涂是候鸟的重要湿地栖息地，也是农业重要的高平原含水层的补给源。临时湿地表现出复杂的水文特征，通过局部暴雨迅速填充，然后通过蒸发和地下水渗透干燥。使用长短期记忆 (LSTM) 神经网络来解释这些复杂的过程，我们对 1984 年至 2018 年大平原 71,842 个海滩被淹没的概率进行了建模。在单个海滩的水平上，该模型实现了 F1 - 在保留的测试集上的得分为 0.522，显示了预测复杂淹没模式的能力。在模拟整个地区的海滩淹没时，该模型能够非常密切地跟踪淹没趋势，即使在干旱时期也是如此。我们的结果证明了使用 LSTM 模拟复杂水文动力学的潜力。我们的建模方法可用于模拟不同气候情景下未来的海滩淹没情况，以更好地了解气候变化将如何影响湿地栖息地和地下水。