[论文解读] Nowcasting-Nets: Deep Neural Network Structures for Precipitation Nowcasting Using IMERG

Mohammad Reza Ehsani, Ariyan Zarei+3 more|arXiv (Cornell University)|Aug 16, 2021

Meteorological Phenomena and Simulations参考文献 67被引用 13

一句话总结

Nowcasting-Nets 提出两种基于深度学习的架构——基于循环神经网络和基于卷积神经网络的模型——利用IMERG卫星降水数据，将短时降雨预报的提前期提升至最多4.5小时。带有残差头的卷积型临近预报网络（CNC-R）在美国区域的降水临近预报精度上相较基线模型提升25%至46%。

ABSTRACT

Accurate and timely estimation of precipitation is critical for issuing hazard warnings (e.g., for flash floods or landslides). Current remotely sensed precipitation products have a few hours of latency, associated with the acquisition and processing of satellite data. By applying a robust nowcasting system to these products, it is (in principle) possible to reduce this latency and improve their applicability, value, and impact. However, the development of such a system is complicated by the chaotic nature of the atmosphere, and the consequent rapid changes that can occur in the structures of precipitation systems In this work, we develop two approaches (hereafter referred to as Nowcasting-Nets) that use Recurrent and Convolutional deep neural network structures to address the challenge of precipitation nowcasting. A total of five models are trained using Global Precipitation Measurement (GPM) Integrated Multi-satellitE Retrievals for GPM (IMERG) precipitation data over the Eastern Contiguous United States (CONUS) and then tested against independent data for the Eastern and Western CONUS. The models were designed to provide forecasts with a lead time of up to 1.5 hours and, by using a feedback loop approach, the ability of the models to extend the forecast time to 4.5 hours was also investigated. Model performance was compared against the Random Forest (RF) and Linear Regression (LR) machine learning methods, and also against a persistence benchmark (BM) that used the most recent observation as the forecast. Independent IMERG observations were used as a reference, and experiments were conducted to examine both overall statistics and case studies involving specific precipitation events. Overall, the forecasts provided by the Nowcasting-Net models are superior, with the Convolutional Nowcasting Network with Residual Head (CNC-R) achieving 25%, 28%, and 46% improvement in the test ...

研究动机与目标

• 通过在IMERG数据上应用实时临近预报系统，减少遥感降水估算中的延迟。
• 通过深度学习模型应对降水系统混沌且快速演变的特性。
• 开发并评估能够实现最高达4.5小时提前期的高精度短期降水预报的深度神经网络。
• 将 Nowcasting-Nets 的性能与传统机器学习模型（随机森林、线性回归）及持续性基准进行比较。
• 利用独立的IMERG观测数据，在美国东部和西部区域验证模型性能。

提出的方法

• 设计两种深度神经网络架构：基于循环神经网络的临近预报网络（RNN-based）和基于卷积神经网络的临近预报网络（CNN-based），并引入残差头（CNC-R）。
• 在美国东部大陆区域（CONUS）的IMERG降水数据上训练模型，通过反馈环路机制将预报时间延长至1.5小时以上。
• 采用时空建模方法，通过序列处理卫星反演的降水场，捕捉不断演变的降水模式。
• 在CNC-R模型中引入残差学习头，以稳定训练过程并提升高分辨率降水场中的特征表示能力。
• 实施反馈环路机制，使模型在每一步的预测输出作为下一步预报的输入，从而实现最高达4.5小时的延长预报。
• 使用独立的IMERG观测数据作为真实值进行评估，应用标准指标，包括偏差、相关系数和命中率指数（CSI）

实验结果

研究问题

• RQ1 当应用于IMERG卫星数据时，深度神经网络能否有效降低降水临近预报中的延迟？
• RQ2 RNN和CNN架构在预测美国区域短期降水时的性能表现如何比较？
• RQ3 反馈环路机制在保持预报精度的前提下，能将预报提前期延长至1.5小时以上多远？
• RQ4 Nowcasting-Nets 与传统机器学习模型（随机森林、线性回归）及持续性基准相比，在预报精度方面表现如何？
• RQ5 模型在不同类型的降水事件和地理区域（东部CONUS与西部CONUS）中的性能表现如何？

主要发现

• 在1.5小时提前期，带有残差头的卷积型临近预报网络（CNC-R）相较持续性基准模型，命中率指数（CSI）提升25%。
• 在1.5小时提前期，CNC-R相较随机森林模型CSI提升28%，相较线性回归模型提升46%。
• 反馈环路机制成功将预报提前期延长至4.5小时，尽管不确定性逐渐增加，但预报质量仍保持合理水平。
• CNC-R模型在所有统计指标和具体对流事件的案例研究中均优于所有基线模型。
• 东部CONUS区域的性能持续优于西部CONUS区域，可能由于东部地区降水系统更频繁且更具组织性。
• 本研究证明，基于IMERG数据训练的深度学习模型可显著提升实时降水临近预报性能，降低对延迟卫星处理的依赖。