Assessment of a Spatiotemporal Deep Learning Approach for Soil Moisture Prediction and Filling the Gaps in Between Soil Moisture Observations.
Assessment of a Spatiotemporal Deep Learning Approach for Soil Moisture Prediction and Filling the Gaps in Between Soil Moisture Observations.
复制标题
评估时空深度学习方法预测土壤水分和填补土壤水分观测之间的空白。
DOI：
10.3389/frai.2021.636234
复制
发表时间：
2021
影响因子：
4
通讯作者：
Bayoumi M
中科院分区：
文献类型：
--
作者：
ElSaadani M;Habib E;Abdelhameed AM;Bayoumi M
土壤湿度(SM)在决定特定地区发生洪涝灾害的可能性方面起着重要作用。目前，SM最常用的模拟方法是基于物理的数值水文模型。对土壤中发生的自然过程进行建模是困难的，需要假设。此外，水文模型的运行时间受研究领域的范围和分辨率的影响很大。在这项研究中，我们提出了一种使用深度学习模型的数据驱动建模方法。有不同类型的DL算法用于不同的目的。例如，卷积神经网络(CNN)算法非常适合于捕获和学习空间模式，而长短期记忆(LSTM)算法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CNN和LSTM能力的DL算法，称为ConvLSTM。在这项研究中，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SM的适用性。这项研究表明，ConvLSTM在预测SM方面显著优于CNN。我们通过组合使用不同的预测器集和不同的LSTM序列长度来测试基于ConvLSTM的模型的性能。研究结果表明，ConvLSTM模型对研究区SM的平均面积均方根误差(RMSE)为2.5%，平均面积相关系数为0.9。ConvLSTM模型还可以提供离散SM观测之间的预测，使其在填补卫星立交桥之间的观测差距等应用中具有潜在的实用价值。
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.
登录
查看更多内容
使用长短期记忆网络方法进行降雨径流模拟的深度学习
影响因子：
3.4
作者：
Hu, Caihong;Wu, Qiang;Lou, Zhengzheng
通讯作者：
Lou, Zhengzheng
影响因子：
56.9
作者：
Koster, RD;Dirmeyer, PA;Yamada, T
通讯作者：
Yamada, T
影响因子：
8
作者：
Zhang, Jian;Howard, Kenneth;Kitzmiller, David
通讯作者：
Kitzmiller, David
影响因子：
5.2
作者：
Fang, Kuai;Shen, Chaopeng;Yang, Xiao
通讯作者：
Yang, Xiao
影响因子：
3.8
作者：
Xia, Youlong;Ek, Michael B.;Quiring, Steven M.
通讯作者：
Quiring, Steven M.