Filters








295 Hits in 1e+01 sec

Deep Learning for Precipitation Nowcasting: A Benchmark and A New Model [article]

Xingjian Shi, Zhihan Gao, Leonard Lausen, Hao Wang, Dit-Yan Yeung, Wai-kin Wong, Wang-chun Woo
2017 arXiv   pre-print
To address these problems, we propose both a new model and a benchmark for precipitation nowcasting.  ...  Recently, the Convolutional LSTM (ConvLSTM) model has been shown to outperform traditional optical flow based methods for precipitation nowcasting, suggesting that deep learning models have a huge potential  ...  Acknowledgments This research has been supported by General Research Fund 16207316 from the Research Grants Council and Innovation and Technology Fund ITS/205/15FP from the Innovation and Technology Commission  ... 
arXiv:1706.03458v2 fatcat:d4f7q5yhyvexzfepy3bsgdmmfy

Distributed Deep Learning for Precipitation Nowcasting [article]

Siddharth Samsi, Christopher J. Mattioli, Mark S. Veillette
2019 arXiv   pre-print
In this paper, this problem is investigated in the context of precipitation nowcasting, a term used to describe highly detailed short-term forecasts of precipitation and other hazardous weather.  ...  By leveraging multiple GPUs, we show that the training time for a given nowcasting model architecture can be reduced from 59 hours to just over 1 hour.  ...  The authors thank Michael Jones on the LLSC team for his assistance with the machine learning software infrastructure setup.  ... 
arXiv:1908.10964v1 fatcat:w44bfj2iljdwvaz3mpcqnripjm

Multimodal Semisupervised Deep Graph Learning for Automatic Precipitation Nowcasting

Kaichao Miao, Wei Wang, Rui Hu, Lei Zhang, Yali Zhang, Xiang Wang, Fudong Nian, Jia-Bao Liu
2020 Mathematical Problems in Engineering  
In this study, we propose a novel multimodal semisupervised deep graph learning framework for precipitation nowcasting.  ...  A majority of existing deep learning-based techniques realize precipitation nowcasting by learning a deep nonlinear function from a single information source, e.g., weather radar.  ...  nowcasting, indicating that deep learning models have a huge potential for solving this problem.  ... 
doi:10.1155/2020/4018042 fatcat:3xspod2xanezhgiahu54376nie

Performance Comparison between Deep Learning and Optical Flow-Based Techniques for Nowcast Precipitation from Radar Images

Marino Marrocu, Luca Massidda
2020 Forecasting  
In this article, a nowcasting technique for meteorological radar images based on a generative neural network is presented.  ...  Both methods have been validated using a public domain data set of radar images, covering an area of about 10 4 km 2 over Japan, and a period of five years with a sampling frequency of five minutes.  ...  Acknowledgments: The authors would like to thank Ryoma Sato of Kyoto University for sharing his scripts, necessary to download the dataset Conflicts of Interest: The authors declare no conflict of interest  ... 
doi:10.3390/forecast2020011 fatcat:jlgofn7eezhnpoqwzo47cz2khm

FDNet: A Deep Learning Approach with Two Parallel Cross Encoding Pathways for Precipitation Nowcasting [article]

Bi-Ying Yan and Chao Yang and Feng Chen and Kohei Takeda and Changjun Wang
2021 arXiv   pre-print
To the best of our knowledge, this is the first network architecture with flow and deformation separation to model the evolution of radar echoes for precipitation nowcasting.  ...  The radar echo extrapolation approaches for precipitation nowcasting take radar echo images as input, aiming to generate future radar echo images by learning from the historical images.  ...  For precipitation nowcasting, x t is a 2D radar echo map.  ... 
arXiv:2105.02585v1 fatcat:wgq4i2dlsvaetn5bb3syly2mru

Nowcasting-Nets: Deep Neural Network Structures for Precipitation Nowcasting Using IMERG [article]

Mohammad Reza Ehsani, Ariyan Zarei, Hoshin V. Gupta, Kobus Barnard, Ali Behrangi
2021 arXiv   pre-print
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  ...  , 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.  ...  The dataset used in this study and the Nowcasting-Nets are freely accessible for non-commercial use at https://github.com/ariyanzri/Nowcasting-Nets accessed on 08 August 2021.  ... 
arXiv:2108.06868v1 fatcat:ry2iin4vabfvpfvlbyqagks4vm

Deep learning‐based precipitation bias correction approach for Yin–He global spectral model

Yi‐Fan Hu, Fu‐Kang Yin, Wei‐Min Zhang
2021 Meteorological Applications  
In this paper, a data-driven bias correction approach based on deep learning is proposed, which is appropriate for the Yin-He global spectral model (YHGSM) re-forecasting.  ...  The results revealed that U-Net-based models could reduce the root mean squared error (RMSE) and improve the threat scores (TSs), especially for heavy precipitation and rainstorms.  ...  The value of the new indicators ranges from 0 to 1; they can be used in deep learning models.  ... 
doi:10.1002/met.2032 fatcat:jqtq4j7ibrhfnoykk3fbrm5v2q

MS-nowcasting: Operational Precipitation Nowcasting with Convolutional LSTMs at Microsoft Weather [article]

Sylwester Klocek, Haiyu Dong, Matthew Dixon, Panashe Kanengoni, Najeeb Kazmi, Pete Luferenko, Zhongjian Lv, Shikhar Sharma, Jonathan Weyn, Siqi Xiang
2022 arXiv   pre-print
We present the encoder-forecaster convolutional long short-term memory (LSTM) deep-learning model that powers Microsoft Weather's operational precipitation nowcasting product.  ...  This model takes as input a sequence of weather radar mosaics and deterministically predicts future radar reflectivity at lead times up to 6 hours.  ...  Acknowledgments and Disclosure of Funding The authors thank the many engineers at Microsoft who helped make the production model a reality. References  ... 
arXiv:2111.09954v2 fatcat:nwjwho5t3vcmnawulpzeyi26uq

NOWCAST

2004 Bulletin of The American Meteorological Society - (BAMS)  
"Continued observations and modeling efforts will be necessary to learn their implications for climate."  ...  When they do, short-scale air-sea interaction phenomena documented here and elsewhere will serve as a benchmark for these models to reproduce.  ...  Understanding the processes that control the termination of El Nino is essential to representing them properly in dynamical prediction schemes, and to determining how the duration of events may change  ... 
doi:10.1175/1520-0477-85.8.1047 fatcat:dwm2brwym5bynfmhzv23vl2er4

NOWCAST

2009 Bulletin of The American Meteorological Society - (BAMS)  
But other Brazilians-such as Elci Dalgalo, a farmer in the drought-stricken southern state of Parana-believe the extreme weather events in the country are natural: "As my grandfather and father would say  ...  many years, prompting authorities to declare a state of emergency in more than 300 towns.  ...  "Every part of the new system has been upgraded to allow for a high rate of production and cost reduction."  ... 
doi:10.1175/1520-0477-90.10.1431 fatcat:c3e4oezubrhqhh2pkwrv6b622q

Cloud Cover Nowcasting with Deep Learning [article]

Léa Berthomier, Bruno Pradel, Lior Perez
2020 arXiv   pre-print
Following recent deep learning successes on multiple imagery tasks, we applied deep convolutionnal neural networks on Meteosat satellite images for cloud cover nowcasting.  ...  Nowcasting is a field of meteorology which aims at forecasting weather on a short term of up to a few hours.  ...  Larvor and M. Sorel from METEO FRANCE for their help and advice.  ... 
arXiv:2009.11577v2 fatcat:6yoesbkshvewvoywrs7h7bm2ua

Deep Orthogonal Decompositions for Convective Nowcasting [article]

Daniel J. Tait
2020 arXiv   pre-print
In this work we demonstrate that their remains an important role to be played by physically informed models, which can successfully leverage deep learning (DL) to project the process onto a lower dimensional  ...  real world datasets including sea surface temperature and precipitation.  ...  a U-Net [21] for both our basis feature extractor, and in its own right as a benchmark DL method.  ... 
arXiv:2006.15628v1 fatcat:7vt2qgv36fbupd7lyewn6kjjjy

RainNet v1.0: a convolutional neural network for radar-based precipitation nowcasting

Georgy Ayzel, Tobias Scheffer, Maik Heistermann
2020 Geoscientific Model Development  
In this study, we present RainNet, a deep convolutional neural network for radar-based precipitation nowcasting.  ...  Its design was inspired by the U-Net and SegNet families of deep learning models, which were originally designed for binary segmentation tasks.  ...  Georgy Ayzel would like to thank the Open Data Science community (https://ods.ai, last access: 10 June 2020) for many valuable discussions and educational help in the growing field of deep learning.  ... 
doi:10.5194/gmd-13-2631-2020 fatcat:xxgtmszdirffxbqyhglie3jkzy

Accurate and Clear Precipitation Nowcasting with Consecutive Attention and Rain-map Discrimination [article]

Ashesh, Buo-Fu Chen, Treng-Shi Huang, Boyo Chen, Chia-Tung Chang, Hsuan-Tien Lin
2021 arXiv   pre-print
We propose a new deep learning model for precipitation nowcasting that includes both the discrimination and attention techniques.  ...  Precipitation nowcasting is an important task for weather forecasting.  ...  Combining the discriminator and the attention mechnism results in a novel deep learning model, which, to the best of our knowledge, is the first deep learning model that tackles the extreme precipitation  ... 
arXiv:2102.08175v1 fatcat:33nqjxuqwbdcnertptwy5tdhra

Machine Learning: New Potential for Local and Regional Deep-Seated Landslide Nowcasting

Adriaan L. van Natijne, Roderik C. Lindenbergh, Thom A. Bogaard
2020 Sensors  
Machine Learning and satellite remote sensing products offer new opportunities for both local and regional monitoring of deep-seated landslide deformation and associated processes.  ...  Nowcasting and early warning systems for landslide hazards have been implemented mostly at the slope or catchment scale.  ...  Discussion The limited roots in physics of many machine learning algorithms provide new potential for the nowcasting of deep-seated landslides.  ... 
doi:10.3390/s20051425 pmid:32151069 pmcid:PMC7085549 fatcat:rqizvjduknduzojyswazmdtjxy
« Previous Showing results 1 — 15 out of 295 results