Weighted ensemble analysis of extreme precipitation under climate change

Zhang, Chen

Permalink

https://hdl.handle.net/2142/90646 Copy

Description

Title

Weighted ensemble analysis of extreme precipitation under climate change

Author(s)

Zhang, Chen

Issue Date

2016-04-28

Director of Research (if dissertation) or Advisor (if thesis)

Cai, Ximing

Department of Study

Civil & Environmental Eng

Discipline

Civil Engineering

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

M.S.

Degree Level

Thesis

Keyword(s)

• Ensemble analysis
• Extreme precipitation event
• Climate change

Abstract

The frequency or intensity of heavy precipitation has likely increased in North America since 1950s. In order to analyze climate change impacts on extreme precipitation events in Chicago area, historical (1961-2000) and projected (2046-2065, 2081-2100) daily precipitation data are calculated from 13 statistical downscaling general circulation models under 3 CMIP3 emission scenarios: A1B, A2 and B1, as well as from 17 stations in NCDC and CCPN rain gage network. Then precipitation events of different recurrence intervals are calculated through regional frequency analysis and based on average deviation of climate model estimates from observation estimates, tricube weight function is used to assign weights to climate model ensemble. This weight result is further applied to projected quantile estimates to derive weighted expected values and confidence intervals of future extreme precipitation events under different emission scenarios, these results are further compared with current available estimates from NOAA Atlas 14. Finally, maximum entropy method (MEM) is applied to assign weights and the results are compared with those from weighted ensemble method (WEM). It is found that intensity and the confidence intervals of heavy precipitation is likely to increase significantly for about 20% from now to 2050s under all emission scenarios (A1B>A2>B1), afterwards, this increase trend will slow down (B1>A1B>A2). As for the performance of expected value projection based on MEM, it can also provide accurate estimates with high computational efficiency.

Graduation Semester

2016-05

Type of Resource

text

Permalink

http://hdl.handle.net/2142/90646

Copyright and License Information

Copyright 2016 Chen Zhang

Owning Collections