Degree Type

Thesis

Date of Award

2015

Degree Name

Master of Science

Department

Agronomy

First Advisor

Raymond W. Arritt

Abstract

We analyze changes in extreme daily precipitation frequency using atmosphere-ocean global climate model (AOGCM) driven regional climate simulations. Nested regional climate model simulations were conducted using RegCM4.4 over the CORDEX-North America domain with 25 and 50 km grid spacing. Initial and lateral boundary conditions are taken from the HadGEM2-ES and GFDL-ESM2M AOGCMs (for RCP8.5 emissions scenario) to simulate present and future climate (1971-2000, 2041-2070). For each run, RegCM4 uses two different convection schemes: the Emanuel scheme, and a Mixed scheme which uses the Emanuel scheme over water and Grell scheme over land. These models were compared for the time period 1971-2000 to observations from the Global Historical Climate Daily Network.

The models produced results relatively consistent with observations. The light precipitation events were overestimated in frequency while the extreme events were underestimated in frequency. There was good agreement between models and observation for intensities between 20 and 40 mm/day. The 25 km model simulations showed less agreement with observations and lower frequencies of extreme precipitation than the 50 km models. The increase in resolution was an attempt to better simulate extreme precipitation; therefore, this result was unexpected. Future simulations (2041-2070) showed increases in the frequency of heavy rainfall events in all models. The 99th percentile showed an average increase of about 9% while the 99.9th percentile had a greater increase of about 15%.

Copyright Owner

Ariele Rhiane Daniel

Language

en

File Format

application/pdf

File Size

53 pages

Share

COinS