Some Statistical Characteristics of Solar Radiation and Sunshine Variations in Tropical Area. A Case Study of the Kiang Valley Region, Malaysia

Abstract

The purpose of this study is to investigate the variations and trends in monthly, seasonal and annual solar radiation in the Kiang Valley Region for the period 1975-2002. The least square method was used for the trend analysis. The linear trend is used mainly to test the change in solar radiation and sunshine to set limits on the rate of change. Trend line and values and significance levels of the slopes have been found The seasonal and the annual average values were computed from the monthly average radiation data. The seasonal and annual average solar radiation and sunshine values were designated as dependent variables, and thus, were fitted linearly for season and annual means for each station. The results showed that the mean of maximum incoming global radiation in September with a value of 21.1 MI M-2 at Petaling Jaya, while the mean minimum in November and December with values of 10.7 and 10.9 MJ ni2 at Petaling Jaya. The low amounts of solar radiation received in November and December are due to greater cloudiness during the period coinciding with the northeast monsoon season. On rainy days, very little global solar radiation is received The overall average rate of change in global solar radiation during 1975-2002 and 1977-2000 is represented by the slope of the linear regression was small (-0.126 and -0.314 MI m-2 per year for Subang Airport and Petaling Jaya respectively). The sunshine results revealed that there is a maximum in May of 7.0 hours at Subang Airport and a minimum in November with a value of 4.5 hours at Universiti Kebangsaan Malaysia. The coefficient of variation (C. V.) ranged from 4.7% in April to 22.3% in December. The coefficient of variation has a wide range during April and December, suggesting large variations from month to month. The statistically fitted linear trend at Subang Airport and Petaling Jaya over the period 1970- 1999 were -4.30 and 0.22 hour per year and statistically not significant. At Universiti Kebangsaan Malaysia the linear trend over the shorter period of measurement available, 1979-2000 was -13.57 hour per year, and was significant at 0.05 level.