INVESTIGATION OF ATMOSPHERIC WATER VAPOUR IN THAILAND

Abstract

This study aims to investigate the nature of water vapour in Thailand. The study comprises five parts. The first part is to develop a simple empirical model to estimate the precipitable water (PWV). In developing the model, the upper air data from four meteorological stations situated in the main regions of Thailand namely Chiang Mai (18.78°N, 98.98°E), Ubon Ratchathani (15.25°N, 104.87°E), Bangkok (13.67°N, 100.60°E) and Songkhla (7.2°N, 100.6°E) were used to estimate the precipitable water (PWV). Then the PWV was correlated to surface relative humidity (RH), ambient air temperature (Ta) and saturated water vapour pressure (pvs) at these stations to obtain the model. This model performs reasonably when tested against an independent data set. The second part of the study use the Artificial Neural Network (ANN) as an alternative method to estimate monthly average PWV. The input layer of this ANN comprises RH, Ta, pvs and month (m), and the output layer consists of only one parameter namely PWV. These data were collected from Chiang Mai, Ubon Ratchathani, Songkhla and Bangkok. A five-year period (2009-2013) of the input and output data were used to train the ANN and another two-year period (2014-2015) of the input and output data were used to evaluate the performance of the trained ANN. It is found that PWV derived from the ANN agrees well with those obtained from sunphotometers. The third part of this study deals with the investigation of spatial distribution of water vapour in the country. To accomplish this, the data from the water vapour channel of MTSAT-1R satellite and PWV data from sunphotometers at the four stations, were collected and processed. Then a model relating the satellite-derived brightness temperature to the PWV was created. After the validation, the model was used to calculate PWV over the country and the results were displayed as monthly and yearly PWV maps. The maps reveal that PWV varies strongly with the seasons and the geographical regions of the country. The fourth part of this work examines spatial and temporal changes in monthly PWV over Thailand using 37 years of monthly ERA-Interim re-analysis data (1981-2017). PWV from the ERA-Interim were compared with that derived from radiosonde observations at the four stations namely Bangkok (13.67°N; 100.61°E), Singapore (1.37°N; 103.98°E), Kuala Lumpur (2.72°N; 101.70°E) and Danang (16.07°N; 108.35°E and it was shown to have an average RMSE and MBE of 0.264 cm and -0.137 cm, respectively. Then, PWV from ERA data was used to estimate spatial patterns of monthly average PWV for the winter (Nov, Dec, Jan), summer (Feb, Mar, Apr, May) and rainy (Jun, Jul, Aug, Sep, Oct) seasons as well as the yearly averages. Significant increasing trend in PWV are found in the rainy season. Significant increasing trend is also found over the southern half of Thailand during the winter season. The final part of the study concerns the investigation of vertical variation of water vapour in the upper atmosphere. It was found that the water vapour in the upper troposphere varies with the seasons and the systematic variation in the lower stratosphere was not observed.

-