Evaluation of Deep-Learning-Based Very Short-Term Rainfall Forecasts in South Korea

Oh, Seok‐Geun; Park, Chan-Il; Son, Seok‐Woo; Ko, Jihoon; Shin, Kijung; Kim, Sun‐Young; Park, Junsang

doi:10.1007/s13143-022-00310-4

Cited by 5 publications

(7 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…southern cases, which encompass relatively broader regions (Kim and Hong, 2022a;Oh et al, 2023). To further improve the performance of nowcasting models, enhancing the prediction capabilities for isolated precipitation patterns is crucial.…”

Section: Modelmentioning

confidence: 99%

“…Recently, deep learning-based models have been studied to predict the non-linear evolution of precipitation patterns (e.g., Shi et al, 2015;Shi et al, 2017;Agrawal et al, 2019;Lebedev et al, 2019;Ayzel et al, 2020;Sønderby et al, 2020;Ravuri et al, 2021;Kim and Hong, 2022a;Kim and Hong, 2022b;Choi and Kim, 2022;Ko et al, 2022;Oh et al, 2023). These models are data-driven and physics-free, as they can be trained using high-resolution radar data that contain spatiotemporal information on instantaneous rainfall.…”

mentioning

confidence: 99%

“…S-band radar has a spatiotemporal resolution of approximately 10-min intervals and 1-km scale (Thorndabl et al, 2017), enabling sufficiently high-resolution predictions within 0-6h lead times. In particular, models based on convolutional long short-term memory (Conv-LSTM) and U-Net have previously been proposed, wherein multiple past radar images are used as input data to train the model and predict future timestep images (Shi et al, 2015;Agrawal et al, 2019;Lebedev et al, 2019;Ayzel et al, 2020;Sønderby et al, 2020;Kim and Hong, 2022a;Ko et al, 2022;Oh et al, 2023). These models learn the complex evolution of rainfall patterns over time, resulting in better predictive performance than baseline models, such as NWP, radar echo tracking, and Eulerian persistence.…”

mentioning

confidence: 99%

See 2 more Smart Citations

Advancing very short-term rainfall prediction with blended U-Net and partial differential approaches

Ha,

Park

2024

Front. Earth Sci.

Self Cite

View full text Add to dashboard Cite

Accurate and timely prediction of short-term rainfall is crucial for reducing the damages caused by heavy rainfall events. Therefore, various precipitation nowcasting models have been proposed. However, the performance of these models still remains limited. In particular, the current operational precipitation nowcasting method, which is based on radar echo tracking, such as the McGill Algorithm for Precipitation Nowcasting by Lagrangian Extrapolation (MAPLE), has a critical drawback when predicting newly developed or decayed precipitation fields. Recently proposed deep learning models, such as the U-Net convolutional neural network outperform the models based on radar echo tracking. However, these models are unsuitable for operational precipitation nowcasting due to their blurry predictions over longer lead times. To address these blurry effects and enhance the performance of U-Net-based rainfall prediction, we propose a blended model that combines a partial differential equation (PDE) model based on fluid dynamics with the U-Net model. The evaluation of the forecast skill, based on both qualitative and quantitative methods for 0–3-h lead times, demonstrates that the blended model provides less blurry and more accurate rainfall predictions compared with the U-Net and partial differential equation models. This indicates the potential to enhance the field of very short-term rainfall prediction. Additionally, we also evaluated the monthly-averaged forecast skills for different seasons, and confirmed the operational feasibility of the blended model, which contributes to the performance enhancement of operational nowcasting.

show abstract

Section: Modelmentioning

confidence: 99%

mentioning

confidence: 99%

mentioning

confidence: 99%

See 1 more Smart Citation

Advancing very short-term rainfall prediction with blended U-Net and partial differential approaches

Ha,

Park

2024

Front. Earth Sci.

Self Cite

View full text Add to dashboard Cite

show abstract

“…Although the accuracy of nowcasting can be tested using the CSI value, it can be higher when nowcasting overestimates the area and intensity of the precipitation fields. We conducted separate evaluations of the CSI, POD, and FAR values at the thresholds of 1 and 10 mm/h [14].…”

Section: Quantitative Evaluationmentioning

confidence: 99%

“…For instance, convolutional long short-term memory models (Conv-LSTM) have been employed to predict sequential output from sequential inputs [7][8][9]. In addition, U-Net convolutional neural networks have also been used (e.g., [10][11][12][13][14]). Models based on convolutional neural networks outperform strong baselines such as optical flow methods and numerical weather prediction [9].…”

Section: Introductionmentioning

confidence: 99%

Enhancing Rainfall Nowcasting Using Generative Deep Learning Model with Multi-Temporal Optical Flow

Ha,

Lee

2023

Remote Sensing

View full text Add to dashboard Cite

Precipitation nowcasting is critical for preventing damage to human life and the economy. Radar echo tracking methods such as optical flow algorithms have been widely employed for precipitation nowcasting because they can track precipitation motions well. Thus, this method, including the McGill algorithm for precipitation nowcasting by Lagrangian extrapolation (MAPLE), was implemented for operational precipitation nowcasting. However, advection-based methods struggle to predict the nonlinear motions of precipitation fields and dynamic processes, such as the growth and decay of precipitation. This study proposes an enhanced optical flow model using a multi-temporal optical flow field and a conditional generative adversarial network (cGAN). We trained the proposed model using a 3-year radar dataset provided by the Korean Meteorological Administration and performed forecast skill evaluations using both qualitative and quantitative methods. In particular, the model featuring multi-temporal optical flow enhances prediction accuracy for the nonlinear motion of precipitation fields, and the model’s accuracy can be further improved through the use of the cGAN structure. We have verified that these improvements hold for 0–3 h lead times. Based on this performance enhancement, we conclude that the multi-temporal optical flow model with cGAN has a potential role in operational precipitation nowcasting.

show abstract

深層学習降水予測の精度改善と現状の詳細評価

KANEKO,

YOSHIMURA

2024

Japanese Journal of JSCE

View full text Add to dashboard Cite

Evaluation of Deep-Learning-Based Very Short-Term Rainfall Forecasts in South Korea

Cited by 5 publications

References 39 publications

Advancing very short-term rainfall prediction with blended U-Net and partial differential approaches

Advancing very short-term rainfall prediction with blended U-Net and partial differential approaches

Enhancing Rainfall Nowcasting Using Generative Deep Learning Model with Multi-Temporal Optical Flow

深層学習降水予測の精度改善と現状の詳細評価

Contact Info

Product

Resources

About