Clock hour correction effect on extreme value analysis of rainfall on Western Coast of India

ABSTRACT

Increasing trend of number and intensity of the extreme precipitation events in India has been reported in literature. Majority of the Indian weather stations have historical rainfall data recorded at a fixed time (0830 hours) of the day, which could be a limitation for correct estimation of extremes. To overcome this, clock hour correction factor (CHCF) in extreme value analysis (EVA) assumes importance. With hourly rainfall data from a site on western coast of India, CHCF was estimated and compared with those reported in literature. The CHCF estimated in this study was higher than those suggested for India in literature or codes. Further, the effect of the CHCF on EVA was examined for India taking CHCF from literature and Indian standards. Towards this, EVA was performed with fixed window (FW) maxima (with CHCF) and sliding window (SW) maxima obtained from the hourly rainfall records of the site. The extreme rainfall from EVA with FW monthly maxima (68 entries) along with CHCF was conservative, compared to those with SW monthly maxima. Annual FW data (17 entries) resulted in unacceptable extreme values (with or without CHCF). This could be due to the insufficient data length and different distribution shape for annual FW data. Treating each monsoon month as a block in Block Maxima approach or adopting suitable threshold in Peak-Over-Threshold approach could help to increase data length for EVA from shorter record length of rainfall data.

KEYWORDS:

• Rainfall
• extreme value analysis
• clock hour correction factor
• time discretization

Acknowledgments

Author thanks the team involved in the weather station for providing the rainfall data. The author sincerely appreciates the contributions from anonymous reviewers for providing critical and insightful review comments which helped to immensely improve the technical content and clarity of the article.

Data Availability

All data used in the study are confidential in nature and were available to the author only for research purposes with restrictions. The author is not authorized to share the data.

Disclosure statement

No potential conflict of interest was reported by the author.