Wang-chun Woo

Mr. Wang-chun Woo

Hong Kong Observatory
orcid

Research Focus & Metrics

Academic Career Length

12Years (2015-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
25 February 2017 in Atmosphere
Verified
Journal Article
Open access
PDF file
WWWang-chun Woo
Wang-chun Woo
WWWai-Kin Wong
Wai-Kin Wong
Hong Kong Observatory has been operating an in-house developed rainfall nowcasting system called “Short-range Warning of Intense Rainstorms in Localized Systems (SWIRLS)” to support rainstorm warning and rainfall nowcasting services. A crucial step in rainfall nowcasting is the tracking of radar echoes to generate motion fields for extrapolation of rainfall areas in the following few hours. SWIRLS adopted a correlation-based method in its first operational version in 1999, which was subsequently replaced by optical flow algorithm in 2010 and further enhanced in 2013. The latest optical flow algorithm employs a transformation function to enhance a selected range of reflectivity for feature tracking. It also adopts variational optical flow computation that takes advantage of the Horn–Schunck approach and the Lucas–Kanade method. This paper details the three radar echo tracking algorithms, examines their performances in several significant rainstorm cases and summaries verification results of multi-year performances. The limitations of the current approach are discussed. Developments underway along with future research areas are also presented.

19 September 2015 in arXiv
Verified
Preprint
Open access
PDF file
XSXingjian Shi
Xingjian Shi
ZCZhourong Chen
Zhourong Chen
HWHao Wang
Hao Wang
DYDit-Yan Yeung
Dit-Yan Yeung
WWWai-Kin Wong
Wai-Kin Wong
WWWang-chun Woo
Wang-chun Woo
The goal of precipitation nowcasting is to predict the future rainfall intensity in a local region over a relatively short period of time. Very few previous studies have examined this crucial and challenging weather forecasting problem from the machine learning perspective. In this paper, we formulate precipitation nowcasting as a spatiotemporal sequence forecasting problem in which both the input and the prediction target are spatiotemporal sequences. By extending the fully connected LSTM (FC-LSTM) to have convolutional structures in both the input-to-state and state-to-state transitions, we propose the convolutional LSTM (ConvLSTM) and use it to build an end-to-end trainable model for the precipitation nowcasting problem. Experiments show that our ConvLSTM network captures spatiotemporal correlations better and consistently outperforms FC-LSTM and the state-of-the-art operational ROVER algorithm for precipitation nowcasting.

5 October 2017 in arXiv
Verified
Preprint
Open access
PDF file
XSXingjian Shi
Xingjian Shi
ZGZhihan Gao
Zhihan Gao
LLLeonard Lausen
Leonard Lausen
HWHao Wang
Hao Wang
DYDit-Yan Yeung
Dit-Yan Yeung
WWWang-chun Woo
Wang-chun Woo
With the goal of making high-resolution forecasts of regional rainfall, precipitation nowcasting has become an important and fundamental technology underlying various public services ranging from rainstorm warnings to flight safety. 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 for solving the problem. However, the convolutional recurrence structure in ConvLSTM-based models is location-invariant while natural motion and transformation (e.g., rotation) are location-variant in general. Furthermore, since deep-learning-based precipitation nowcasting is a newly emerging area, clear evaluation protocols have not yet been established. To address these problems, we propose both a new model and a benchmark for precipitation nowcasting. Specifically, we go beyond ConvLSTM and propose the Trajectory GRU (TrajGRU) model that can actively learn the location-variant structure for recurrent connections. Besides, we provide a benchmark that includes a real-world large-scale dataset from the Hong Kong Observatory, a new training loss, and a comprehensive evaluation protocol to facilitate future research and gauge the state of the art.

12 February 2021 in Atmosphere
Verified
Journal Article
Open access
PDF file
MCMan-Lok Chong
Man-Lok Chong
YWYat Chun Wong
Yat Chun Wong
WWWang-chun Woo
Wang-chun Woo
ATAmos Tai
Amos Tai
WWWai-Kin Wong
Wai-Kin Wong
Short-lived and intense rainfall is common in Hong Kong during the wet season and often causes disruption to daily lives. As global numerical weather prediction (NWP) models are progressively improved, representation of sub-synoptic or mesoscale systems associated with intense rainfall is better parametrized and resolved than ever before, but their quantitative precipitation forecasts (QPFs) still tend to underestimate the magnitude of intense rainfall. This study calibrated model QPFs over the region of Hong Kong by two frequency-matching methods. In both methods, conversion schemes between the direct model output (DMO) and calibrated forecasts were first established by matching the cumulative distribution to that of the observed data. The “Adaptive Table” method updated the conversion scheme whenever the latest observation fell out of its expected range in the existing scheme, whereas the “Sliding Window” method reconstructed the conversion scheme using data from the most recent two years. The calibration methods had been verified against actual rainfall events with different thresholds, and it was found that both methods could improve model performance for moderate and heavy rainfall in short-range forecasts with similar effectiveness. They were also able to reduce the systematic bias of precipitation forecasts for significant rainfall throughout the verification period.