Seasonal prediction of winter extreme precipitation over Canada by support vector regression
For forecasting the maximum 5-day accumulated precipitation over the winter season at lead times of 3, 6, 9 and 12 months over Canada from 1950 to 2007, two nonlinear and two linear regression models were used, where the models were support vector regression (SVR) (nonlinear and linear versions), no...
Published in: | Hydrology and Earth System Sciences |
---|---|
Other Authors: | , , , |
Format: | Article in Journal/Newspaper |
Language: | English |
Published: |
2011
|
Subjects: | |
Online Access: | https://doi.org/10.5194/hess-15-65-2011 |
id |
ftncar:oai:drupal-site.org:articles_25200 |
---|---|
record_format |
openpolar |
spelling |
ftncar:oai:drupal-site.org:articles_25200 2024-04-28T08:09:08+00:00 Seasonal prediction of winter extreme precipitation over Canada by support vector regression Zeng, Zhen (author) Hsieh, W. W. (author) Shabbar, A. (author) Burrows, W. R. (author) 2011-01-06 https://doi.org/10.5194/hess-15-65-2011 en eng Hydrology and Earth System Sciences--Hydrol. Earth Syst. Sci.--1607-7938 articles:25200 doi:10.5194/hess-15-65-2011 ark:/85065/d7474fgz Copyright author(s). This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License. article Text 2011 ftncar https://doi.org/10.5194/hess-15-65-2011 2024-04-04T17:33:50Z For forecasting the maximum 5-day accumulated precipitation over the winter season at lead times of 3, 6, 9 and 12 months over Canada from 1950 to 2007, two nonlinear and two linear regression models were used, where the models were support vector regression (SVR) (nonlinear and linear versions), nonlinear Bayesian neural network (BNN) and multiple linear regression (MLR). The 118 stations were grouped into six geographic regions by K-means clustering. For each region, the leading principal components of the winter maximum 5-d accumulated precipitation anomalies were the predictands. Potential predictors included quasi-global sea surface temperature anomalies and 500 hPa geopotential height anomalies over the Northern Hemisphere, as well as six climate indices (the Nino-3.4 region sea surface temperature, the North Atlantic Oscillation, the Pacific-North American teleconnection, the Pacific Decadal Oscillation, the Scandinavia pattern, and the East Atlantic pattern). The results showed that in general the two robust SVR models tended to have better forecast skills than the two non-robust models (MLR and BNN), and the nonlinear SVR model tended to forecast slightly better than the linear SVR model. Among the six regions, the Prairies region displayed the highest forecast skills, and the Arctic region the second highest. The strongest nonlinearity was manifested over the Prairies and the weakest nonlinearity over the Arctic. Article in Journal/Newspaper Arctic North Atlantic North Atlantic oscillation OpenSky (NCAR/UCAR - National Center for Atmospheric Research/University Corporation for Atmospheric Research) Hydrology and Earth System Sciences 15 1 65 74 |
institution |
Open Polar |
collection |
OpenSky (NCAR/UCAR - National Center for Atmospheric Research/University Corporation for Atmospheric Research) |
op_collection_id |
ftncar |
language |
English |
description |
For forecasting the maximum 5-day accumulated precipitation over the winter season at lead times of 3, 6, 9 and 12 months over Canada from 1950 to 2007, two nonlinear and two linear regression models were used, where the models were support vector regression (SVR) (nonlinear and linear versions), nonlinear Bayesian neural network (BNN) and multiple linear regression (MLR). The 118 stations were grouped into six geographic regions by K-means clustering. For each region, the leading principal components of the winter maximum 5-d accumulated precipitation anomalies were the predictands. Potential predictors included quasi-global sea surface temperature anomalies and 500 hPa geopotential height anomalies over the Northern Hemisphere, as well as six climate indices (the Nino-3.4 region sea surface temperature, the North Atlantic Oscillation, the Pacific-North American teleconnection, the Pacific Decadal Oscillation, the Scandinavia pattern, and the East Atlantic pattern). The results showed that in general the two robust SVR models tended to have better forecast skills than the two non-robust models (MLR and BNN), and the nonlinear SVR model tended to forecast slightly better than the linear SVR model. Among the six regions, the Prairies region displayed the highest forecast skills, and the Arctic region the second highest. The strongest nonlinearity was manifested over the Prairies and the weakest nonlinearity over the Arctic. |
author2 |
Zeng, Zhen (author) Hsieh, W. W. (author) Shabbar, A. (author) Burrows, W. R. (author) |
format |
Article in Journal/Newspaper |
title |
Seasonal prediction of winter extreme precipitation over Canada by support vector regression |
spellingShingle |
Seasonal prediction of winter extreme precipitation over Canada by support vector regression |
title_short |
Seasonal prediction of winter extreme precipitation over Canada by support vector regression |
title_full |
Seasonal prediction of winter extreme precipitation over Canada by support vector regression |
title_fullStr |
Seasonal prediction of winter extreme precipitation over Canada by support vector regression |
title_full_unstemmed |
Seasonal prediction of winter extreme precipitation over Canada by support vector regression |
title_sort |
seasonal prediction of winter extreme precipitation over canada by support vector regression |
publishDate |
2011 |
url |
https://doi.org/10.5194/hess-15-65-2011 |
genre |
Arctic North Atlantic North Atlantic oscillation |
genre_facet |
Arctic North Atlantic North Atlantic oscillation |
op_relation |
Hydrology and Earth System Sciences--Hydrol. Earth Syst. Sci.--1607-7938 articles:25200 doi:10.5194/hess-15-65-2011 ark:/85065/d7474fgz |
op_rights |
Copyright author(s). This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License. |
op_doi |
https://doi.org/10.5194/hess-15-65-2011 |
container_title |
Hydrology and Earth System Sciences |
container_volume |
15 |
container_issue |
1 |
container_start_page |
65 |
op_container_end_page |
74 |
_version_ |
1797577577697640448 |