METHODOLOGY
3.1 Rainfall data collection and analysis
Monthly rainfall data during the year 1976 to 2004 from 308 stations of the Thai Meteorological Department
(TMD) stations distributed throughout the study area were collected and digitally encoded into GIS database. Mean
annual rainfall of each station was calculated for 29 years. Median of mean annual rainfall of each year was
selected as a representative of each station during this study periods. Thus, there are 308 values preparing for
interpolation.
Exploratory analysis of rainfall data was examined to explore the frequency distribution of rainfall and the local
characteristics of spatial autocorrelation of rainfall in different location over the study period. The interpolation of
mean annual rainfall for the 29 years period was performed using the Kriging Method to get a spatial pattern of
rainfall data. The appropriate variogram model for rainfall data was the Gaussian model because the variation was
very smooth and the nugget variance was very small compared to the spatially dependent random variation
(Burrough et al, 1998). Cross validation was used to check for measuring the residual between the point and the
surface created.
3.2 Calculation of coefficient of variance (CV)
A measurement of reliability of rainfall at each station for the study periods was calculated using CV. High CV
values indicate a large variation in rainfall over the study period (Saenjan et al, 1990). The formula for CV is;
Where s = standard deviation and x = mean
3.3 SPI calculation
The SPI is the number of standard deviations of which the observed value deviated from mean. Computation of
SPI is firstly done by transforming precipitation value into a normal distribution. SPI is calculated by fitting a
Gamma probability density function to a given frequency distribution of precipitation totals for a station. Then,
each probability density function is transformed into the standardized normal distribution with mean of zero and
variance of one, which is the value of the SPI (Loukas et al, 2004).
METHODOLOGY
3.1 Rainfall data collection and analysis
Monthly rainfall data during the year 1976 to 2004 from 308 stations of the Thai Meteorological Department
(TMD) stations distributed throughout the study area were collected and digitally encoded into GIS database. Mean
annual rainfall of each station was calculated for 29 years. Median of mean annual rainfall of each year was
selected as a representative of each station during this study periods. Thus, there are 308 values preparing for
interpolation.
Exploratory analysis of rainfall data was examined to explore the frequency distribution of rainfall and the local
characteristics of spatial autocorrelation of rainfall in different location over the study period. The interpolation of
mean annual rainfall for the 29 years period was performed using the Kriging Method to get a spatial pattern of
rainfall data. The appropriate variogram model for rainfall data was the Gaussian model because the variation was
very smooth and the nugget variance was very small compared to the spatially dependent random variation
(Burrough et al, 1998). Cross validation was used to check for measuring the residual between the point and the
surface created.
3.2 Calculation of coefficient of variance (CV)
A measurement of reliability of rainfall at each station for the study periods was calculated using CV. High CV
values indicate a large variation in rainfall over the study period (Saenjan et al, 1990). The formula for CV is;
Where s = standard deviation and x = mean
3.3 SPI calculation
The SPI is the number of standard deviations of which the observed value deviated from mean. Computation of
SPI is firstly done by transforming precipitation value into a normal distribution. SPI is calculated by fitting a
Gamma probability density function to a given frequency distribution of precipitation totals for a station. Then,
each probability density function is transformed into the standardized normal distribution with mean of zero and
variance of one, which is the value of the SPI (Loukas et al, 2004).
การแปล กรุณารอสักครู่..