Computational approach to identify the acid rain patterns by adopting satellite imagery & data mining technique

Acid rain is a major provincial ecological problem around the globe. To control acid rain pollution and to protect the ecological environment, it is a major need to identify the happening of acid rains. This paper presents a methodology for identifying the occurrence of acid rains using satellite imagery of precipitation and SO₂. The present analysis applies k-means clustering followed by Haar wavelet transform on the satellite imagery in order to compute normality for precipitation and SO₂. By using normality, pH value is computed and if the value lies in the range of 1 ≤ pH ≤ 5, it is identified as acid rain. The proposed methodology is addressed for the first time using data mining and image processing. The resulted outcome indicates that the amount of rainfall with the concentration of SO₂ have a strong influence on the occurrence of acid rains.