| Home  | About ScienceAsia  | Publication charge  | Advertise with us  | Subscription for printed version  | Contact us  
Editorial Board
Journal Policy
Instructions for Authors
Online submission
Author Login
Reviewer Login
Volume 50 Number 1
Volume 49 Number 6
Volume 49 Number 5
Volume 49S Number 1
Volume 49 Number 4
Volume 49 Number 3
Earlier issues
Back

Research Article

ScienceAsia 32 (2006): 107-114 |doi: 10.2306/scienceasia1513-1874.2006.32.107

Ultraviolet Forecasting in Thailand


Sumridh Sudhibrabha,a* Robert Harold Buchanan Exella and Dusadee Sukawatb

 
ABSTRACT: Total ozone and erythemal UV models have been created for UV Index forecasting in Thailand. Versions of an ozone model for Chiangmai, Ubon Ratchathani, Bangkok and Songkhla and a UV model for Bangkok and Songkhla are available for operation on a daily basis. The procedure is to forecast ozone from upper air data, and then calculate the forecast erythemal UV irradiance from the ozone forecast. In the ozone models, a linear regression technique was used with fifteen coefficients for temperature and dynamic height at the 100 hPa and 50 hPa levels where the ozone exists and daily balloon observations are made. Simple ozone models were also developed for use in the areas without upper air observations. The data are in the form of time series over three consecutive days. The UV models were obtained from Brewer UV data from Bangkok and Songkhla fitted to a non-linear three dimensional equation which is a function of ozone and its air mass to obtain five coefficients for the empirical formula. The models are used to estimate cloud-free values of the UV Index over the different areas of the country. The output of the erythemal UV for a clear sky is converted to the UV Index. The UV Index is then used to select one of five categories for issuing to the public every noon. Reductions of the UV Index for cloudy skies are also given. The average accuracy of the models is acceptable as the squares of the correlation coefficients (R2) are 0.88 and 0.99 for the ozone and UV models respectively, while the mean absolute percentage errors are 1.5 % and 7.5 % for the ozone and UV forecasting respectively.

Download PDF


a Joint Graduate School of Energy and Environment, King Mongkut’s University of Technology Thonburi, Bangkok 10140.
b Department of Mathematics, Faculty of Science, King Mongkut’s University of Technology Thonburi, Bangkok 10140.

* Corresponding author, E-mail: sumridh@metnet.tmd.go.th

Received 13 May 2005, Accepted 27 Dec 2005