ElSaadani Mohamed, Habib Emad, Abdelhameed Ahmed M, Bayoumi Magdy
Department of Civil Engineering and Louisiana Watershed Flood Center, University of Louisiana at Lafayatte, Lafayette, LA, United States.
Department of Electrical and Computer Engineering, University of Louisiana at Lafayatte, Lafayette, LA, United States.
Front Artif Intell. 2021 Mar 4;4:636234. doi: 10.3389/frai.2021.636234. eCollection 2021.
Soil moisture (SM) plays a significant role in determining the probability of flooding in a given area. Currently, SM is most commonly modeled using physically-based numerical hydrologic models. Modeling the natural processes that take place in the soil is difficult and requires assumptions. Besides, hydrologic model runtime is highly impacted by the extent and resolution of the study domain. In this study, we propose a data-driven modeling approach using Deep Learning (DL) models. There are different types of DL algorithms that serve different purposes. For example, the Convolutional Neural Network (CNN) algorithm is well suited for capturing and learning spatial patterns, while the Long Short-Term Memory (LSTM) algorithm is designed to utilize time-series information and to learn from past observations. A DL algorithm that combines the capabilities of CNN and LSTM called ConvLSTM was recently developed. In this study, we investigate the applicability of the ConvLSTM algorithm in predicting SM in a study area located in south Louisiana in the United States. This study reveals that ConvLSTM significantly outperformed CNN in predicting SM. We tested the performance of ConvLSTM based models by using a combination of different sets of predictors and different LSTM sequence lengths. The study results show that ConvLSTM models can predict SM with a mean areal Root Mean Squared Error (RMSE) of 2.5% and mean areal correlation coefficients of 0.9 for our study area. ConvLSTM models can also provide predictions between discrete SM observations, making them potentially useful for applications such as filling observational gaps between satellite overpasses.
土壤湿度（SM）在确定特定区域发生洪水的可能性方面起着重要作用。目前，土壤湿度最常使用基于物理的数值水文模型进行模拟。对土壤中发生的自然过程进行建模很困难，需要做出假设。此外，水文模型的运行时间受到研究区域范围和分辨率的高度影响。在本研究中，我们提出了一种使用深度学习（DL）模型的数据驱动建模方法。有不同类型的深度学习算法，用于不同的目的。例如，卷积神经网络（CNN）算法非常适合捕捉和学习空间模式，而长短期记忆（LSTM）算法则旨在利用时间序列信息并从过去的观测中学习。最近开发了一种结合了CNN和LSTM功能的深度学习算法，称为卷积长短期记忆网络（ConvLSTM）。在本研究中，我们调查了ConvLSTM算法在美国路易斯安那州南部一个研究区域预测土壤湿度的适用性。这项研究表明，在预测土壤湿度方面，ConvLSTM明显优于CNN。我们通过结合不同的预测变量集和不同的LSTM序列长度来测试基于ConvLSTM的模型的性能。研究结果表明，对于我们的研究区域，ConvLSTM模型可以预测土壤湿度，平均区域均方根误差（RMSE）为2.5%，平均区域相关系数为0.9。ConvLSTM模型还可以在离散的土壤湿度观测值之间提供预测，使其在填补卫星过境之间的观测空白等应用中具有潜在用途。
