Uncertainties of Assessing Projected Changes in Precipitation Extremes
Abstract
Water resource managers share a common challenge in understanding what climate change could mean for future hydroclimate extremes. Understanding the uncertainty of projected changes in extremes is critical to making decisions about whether to invest in adaptation measures today or delay until more credible information becomes available. Uncertainties arise from several methodological choices including, including criteria that drive selection of global climate projection information to frame the assessment, whether and how to bias-correct global projection information, and how to represent local controls on how to spatially downscale translations of these projections. This presentation highlights such uncertainties, focusing on projected changes in precipitation indicated by two metrics: annual total and annual maximum daily amount. Attention is given to metric conditions varying from typical (i.e. metrics having 0.50 cumulative probability) to extreme (i.e. annual totals having 0.01 and 0.05 cumulative probabilities, which are relevant to drought, and annual maximum daily amounts having 0.95 and 0.99 cumulative probabilities, which are relevant to floods). The assessment is informed by an ensemble of 53 daily CMIP3 precipitation projections from the "Bias Corrected and Downscaled WCRP CMIP3 Climate Projections" web-archive (see URL), regridded over the contiguous United States from native climate model resolution to a common 2° grid and reported during 1961-2000, 2046-2065 and 2081-2100. Focusing on changes between 20-year periods, evaluations include (a) assessing changes in typical metric conditions and determining whether changes in metric distributions are statistically significant, (b) characterizing metric extremes using parametric techniques and assessing for changes in metric extremes, (c) assessing how uncertainties in projected typical and extreme metrics associate with three sources of global climate projection uncertainty (emissions scenario, global climate model, internal variability), and (d) characterizing how projected changes may be sensitive to spatial downscaling (focusing on one non-dynamical scheme). Presentation will provide an overview of methods and highlight key results. (For more information on (c), see Barsugli et al. submitted to GC session "Climate Extremes 2. Challenges in Understanding and Modeling Regional Climate Extremes.")
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2011
- Bibcode:
- 2011AGUFM.H23J..01B
- Keywords:
-
- 1807 HYDROLOGY / Climate impacts;
- 1817 HYDROLOGY / Extreme events;
- 1854 HYDROLOGY / Precipitation;
- 1880 HYDROLOGY / Water management