Principal Component Analysis to study spatial variability of errors in the INSAT derived quantitative precipitation estimates over Indian monsoon region

摘要

In this paper, Principal Component Analysis has been applied to investigate the spatial variability of errors in the INSAT derived quantitative precipitation estimates (QPE) over the Indian monsoon region, using daily rainfall analysis (at the same resolution) for the period from 1 June to 30 August of summer monsoon 2001. The study shows that the QPE errors have certain spatial variability. The orographic rainfall is significantly underestimated along the Western Ghats and along the foothills of the Himalayas, where the root mean square errors are also very large. Otherwise, the performance of the QPE is reasonably good over the rest of the region. The first principal component, which explains about 5.1% of the variance, corresponds to the onset phase of the monsoon during June, when strong positive loadings dominate over the southern parts of the country. The second principal component explaining about 4.2% of the variance, has strong positive loading in the intermittent presence of the monsoon low pressure system over the east central parts of the country. The third principal component which explains 3.3% of the variance is associated with the monsoon trough at the normal position, and the fourth principal component which explains 3.1% of the variance is associated with the monsoon trough at the southern position.