Two-Stage Spatiotemporal Context Refinement Network for Precipitation Nowcasting

Niu, Dan; Huang, Junqi; Zang, Zengliang; Xu, Liujia; Che, Hongshu; Tang, Yang

doi:10.3390/rs13214285

Cited by 14 publications

(21 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The STConvS2S consists of only convolutional layers, and the authors showed that STConvS2S worked faster than PredRNN for the radar echo forecasting task. Niu et al (2021) proposed a two-stage spatiotemporal context refinement network (2S-STRef), which is two-fold as the name suggests. The first stage is a spatiotemporal prediction network that gives a first-stage prediction using a spatiotemporal RNN module incorporated into an encoder-decoder framework.…”

Section: D Rainfall Forecastingmentioning

confidence: 99%

A Systematic Review of Deep Learning Applications in Interpolation and Extrapolation of Precipitation Data

Sit¹,

Demiray²,

Demir³

2022

Preprint

View full text Add to dashboard Cite

With technological enhancements, the volume, velocity, and variety (3Vs) of the raw digital Earth data have increased in recent years. Due to the increased availability of computer resources and the growing popularity of deep learning applications, this data has been a crucial source for data-driven studies that have transformed the fields of climate and earth science. One of the critical data sources is precipitation supporting climate and earth science studies on modeling, forecasting, and preparedness for extreme events (i.e., floods, droughts, pollution transport). In this study, we worked on an extensive review of manuscripts focusing on use of deep learning methods to tackle challenges either to improve the quality or extrapolate (forecast) rainfall datasets. The purpose of this study is to summarize the most recent developments in deep learning approaches for forecasting rainfall or improving rainfall datasets, as well as highlighting issues, shortcomings, and open questions with insightful recommendations for future directions.

show abstract

Section: D Rainfall Forecastingmentioning

confidence: 99%

A Systematic Review of Deep Learning Applications in Interpolation and Extrapolation of Precipitation Data

Sit¹,

Demiray²,

Demir³

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…In recent times, there has been a lot of interest directed toward this timeframe as a result of the commencement of the Subseasonal to Seasonal Prediction Experiment. The linked processes of the atmosphere and the ocean are essential for annual forecasting, and the occurrence of El Nino in the equatorial region of the Pacific Ocean is a classic example [9]. Many large datasets connected to this program include S2S projections from tactical and academic simulations, which are useful for possible ML requests.…”

Section: Introductionmentioning

confidence: 99%

Proper Weather Forecasting Internet of Things Sensor Framework with Machine Learning

Turukmane,

Pande

2024

EAI Endorsed Trans IoT

View full text Add to dashboard Cite

Recent times have seen a rise in the amount of focus placed on the configurations of big data and the Internet of Things (IoT). The primary focus of the researchers was the development of big data analytics solutions based on machine learning. Machine learning is becoming more prevalent in this sector because of its ability to unearth hidden traits and patterns, even within exceedingly complicated datasets. This is one reason why this is the case. For the purpose of this study, we applied our Big Data and Internet of Things (IoT)-based system to a use case that involved the processing of weather information. We put climate clustering and sensor identification algorithms into practice by using data that was available to the general public. For this particular application, the execution information was shown as follows:every single level of the construction. The training method that we've decided to use for the package is a k-means cluster that's based on Scikit-Learn. According to the results of the information analyses, our strategy has the potential to be utilized in usefully retrieving information from a database that is rather complicated.

show abstract

“…Traditional methods for radar echo extrapolation include cross-correlation methods [9,10], centroid tracking methods [11,12] and optical flow-based methods [13,14]. The cross-correlation methods calculate the correlation of each sub-region in two consecutive radar echo images to get the motion vectors [15]. These methods have low prediction accuracy regarding the condition that echoes evolve rapidly [16].…”

Section: Introductionmentioning

confidence: 99%

CEMA-LSTM: Enhancing Contextual Feature Correlation for Radar Extrapolation Using Fine-Grained Echo Datasets

Yang¹,

Liu²,

Wu³

et al. 2023

Computer Modeling in Engineering &Amp; Sciences

View full text Add to dashboard Cite

Accurate precipitation nowcasting can provide great convenience to the public so they can conduct corresponding arrangements in advance to deal with the possible impact of upcoming heavy rain. Recent relevant research activities have shown their concerns on various deep learning models for radar echo extrapolation, where radar echo maps were used to predict their consequent moment, so as to recognize potential severe convective weather events. However, these approaches suffer from an inaccurate prediction of echo dynamics and unreliable depiction of echo aggregation or dissipation, due to the size limitation of convolution filter, lack of global feature, and less attention to features from previous states. To address the problems, this paper proposes a CEMA-LSTM recurrent unit, which is embedded with a Contextual Feature Correlation Enhancement Block (CEB) and a Multi-Attention Mechanism Block (MAB). The CEB enhances contextual feature correlation and supports its model to memorize significant features for near-future prediction; the MAB uses a position and channel attention mechanism to capture global features of radar echoes. Two practical radar echo datasets were used involving the FREM and CIKM 2017 datasets. Both quantification and visualization of comparative experimental results have demonstrated outperformance of the proposed CEMA-LSTM over recent models, e.g., PhyDNet, MIM and PredRNN++, etc. In particular, compared with the second-ranked model, its average POD, FAR and CSI have been improved by 3.87%, 1.65% and 1.79%, respectively on the FREM, and by 1.42%, 5.60% and 3.16%, respectively on the CIKM 2017.

show abstract

Two-Stage Spatiotemporal Context Refinement Network for Precipitation Nowcasting

Cited by 14 publications

References 45 publications

A Systematic Review of Deep Learning Applications in Interpolation and Extrapolation of Precipitation Data

A Systematic Review of Deep Learning Applications in Interpolation and Extrapolation of Precipitation Data

Proper Weather Forecasting Internet of Things Sensor Framework with Machine Learning

CEMA-LSTM: Enhancing Contextual Feature Correlation for Radar Extrapolation Using Fine-Grained Echo Datasets

Contact Info

Product

Resources

About