Accounting for the changing frequency of heavy storms in Illinois Dr. Momcilo Markus Principal Research Scientist, Illinois State Water Survey Research Assoc. Prof., Dep. of Nat. Res. and Environ. Sci. Research Assoc. Prof., Dep. of Agric. and Biol. Eng. University of Illinois at Urbana-Champaign
BEYOND THE BASICS SEMINAR 2020: DESIGNING FOR AN UNCERTAIN FUTURE
Acknowledgments Work on updating Bulletin 70 was supported by the Illinois Department of Commerce and Economic Opportunity under Grant No. 08-355061 and funded by the U.S. Department of Housing and Urban Development’s Community Development Block Grants Award No. B-08-DI-17-0001. Coordinated with IDNR Office of Water Resources. Sally McConkey, David Kristovich, Brian Kerschner, Mary Richardson, Wes Cattoor, Loren Wobig, Kexuan Ariel Wang, Lu Jin, Shaoxuan Guo, Shailendra Singh, Tom Over, Annie Peiyong Qu, Francina Dominguez, Ryan Shriver, and Lisa Sheppard
Dr. Jim Angel Illinois State Climatologist
What is Bulletin 70? • Report containing the expected rainfall amounts for selected storm durations and return periods (for example, the 24-hour, 100-year storm for a site in Illinois). • Many state and local agencies in Illinois required the use of Bulletin 70 for design work after its publication in 1989. • The original Bulletin 70 needed to be updated due to the observed increases in frequency and amounts of heavy rainfall events.
IDCEO funded project on updated Bulletin 70 resulted in two reports Angel, J. and M. Markus. 2019. Frequency Distributions of Heavy Precipitation in Illinois: Updated Bulletin 70 . Illinois State Water Survey Contract Report 2019-05, Champaign, IL. Markus M., J. Angel, K. Wang, B. Kerschner, and S. Singh. 2019. Frequency Distributions of Heavy Precipitation in Illinois: Spatiotemporal Analyses . Illinois State Water Survey Contract Report 2019-10, Champaign, IL.
WHAT IS UPDATED BULLETIN 70?
“Updated Bulletin 70” = Two reports published in 2019
“Updated Bulletin 70” = Two reports published in 2019 BULLETIN 75
“Updated Bulletin 70”= Bulletin 75 (to be published in March 2020)
Historical evolution of rainfall frequency standards • TP-40 (Hershfield, 1961) • Bulletin 70 (Huff and Angel, 1989) • Bulletin 71 (Huff and Angel, 1992) • NOAA Atlas 14 (Bonnin et al. 2004) • Angel, J. and M. Markus. 2019. Frequency Distributions of Heavy Precipitation in Illinois: Updated Bulletin 70 . Illinois State Water Survey Contract Report 2019-05, Champaign, IL. • Markus M., J. Angel, K. Wang, B. Kerschner, and S. Singh. 2019. Frequency Distributions of Heavy Precipitation in Illinois: Spatiotemporal Analyses . Illinois State Water Survey Contract Report 2019-10, Champaign, IL. • Bulletin 75
Maximum daily precipitation in inches 10 12 14 16 18 0 2 4 6 8 (none of the 10 largest values at Aurora were observed prior to 1950) 1900 1905 1910 Annual Maximum Daily Rainfall 1915 1920 1925 1930 1935 1940 1945 Aurora 1950 1955 TP40 1960 1965 1970 1975 1980 1985 1990 ISWS Bulletin 70 1995 2000 NOAA Atlas 14 2005 2010 2015 ISWS Bulletin 75
Maximum daily precipitation in inches 10 12 14 16 18 0 2 4 6 8 (none of the 10 largest values at Aurora were observed prior to 1950) 1900 1905 1910 Annual Maximum Daily Rainfall 1915 1920 1925 1930 1935 1940 1945 Aurora 1950 1955 TP40 1960 1965 1970 1975 1980 1985 1990 ISWS Bulletin 70 1995 2000 NOAA Atlas 14 2005 2010 2015 ISWS Bulletin 75
Change in Annual Precipitation 1986-2015 minus 1901-1960 NCA4 https://science2017.globalchange.gov/
Change in the top 1% of extreme precipitation (the 99 th percentile) https://science2017.globalchange.gov/
Cedar Rapids, IA, July 2008
Grand Forks, ND, April 1997
On average, over the past 20 years in the U.S., flooding has claimed over 90 lives annually Des Plaines River April 18, 2013 Courtesy of Clayton Ballerine, ISWS
Stationarity of heavy rainfall • Traditional assumption: Future variability will be like past variability. This assumption is called “stationarity.” • In the light of climate change, assumption of stationarity may not be appropriate for precipitation frequency analysis.
Nonstationarity of heavy rainfall
Nonstationarity of heavy rainfall 2020
Our Solution to Observed Changes • Use 1948-2017 data to better represent the current, wetter climate • Three times as many stations are available from 1948 onward • Include a Bulletin 70 style adjustment by giving more weight to the second half of the record
L-Moments Software 24-Hour Precipitation Annual Maximum Frequency Analysis Results for Aurora, IL. Observed GLO (Single Site) WAK (Single Site) 40 GLO (6 Sites) WAK (6 Sites) Precipitation (cm) GLO (12 Sites) 30 WAK (12 Sites) 20 10 0 .1 .2 .5 1 2 3 5 10 20 30 40 50 60 70 80 90 95 99 99.9 Non-exceedence probability Markus, M., Angel, J.R., Yang, L. and Hejazi, M.I., 2007, Changing estimates of design precipitation in Northeastern Illinois: Comparison between different sources and sensitivity analysis, J. Hydrology, 347(1-2):211-222.
10 Regions in Illinois
176 Daily Precipitation Stations 1948-2017
73 Hourly Precipitation Stations 1948-2017
Process • Obtained and QC’d the data • Selected stations based on availability and length of record • Calculated the expected precipitation at selected return periods for 1 to 10 days using L-moments
Quality Control https://www.scirp.org/journal/PaperInformation.aspx?paperID=67664&
Process • Adjusted the results from the annual maximum series into a partial duration series using a standard approach ( Langbein’s equation (1949)) • Converted the constrained to unconstrained using standard conversions
Process … • Averaged the station frequency values into a regional frequency analysis (RFA)
TP-40 (Hershfield, 1961)
NOAA Atlas 14 (Bonnin et al. 2004)
Process • Calculate frequencies for sub-daily durations using conversion factors due to limitations of hourly data
Adjustment for Non-Stationarity • Ratio of the 1983-2017 RFA divided by the 1948-1982 RFA
Confidence Intervals
Old and New 100-Yr, 24-Hour Storm OLD NEW
New vs Bulletin 70 Differences in inches
New vs NOAA Atlas 14 Differences in inches
• Volume I of the NCA4 • Precipitation will continue to increase (medium confidence) • Heavy precipitation events will increase in frequency and amounts (high confidence) https://science2017.globalchange.gov/
Nonstationarity of heavy rainfall 2020
Climate modeling A climate model = an atmosphere model, an ocean model, a land model, and a sea ice model. Inputs = the concentration of greenhouse gases, the intensity of sunlight, the amount of deforestation, and volcanoes that should erupt during the simulation. Outputs = atmospheric pressure, ocean salinity, forest cover, temperature, precipitation, clouds,…
Climate Modeling Inputs Outputs Model The concentration of greenhouse gases, An atmosphere Atmospheric the intensity of model, an pressure, ocean sunlight, the amount ocean model, a salinity, forest of deforestation, and land model, cover, temperature, volcanoes that should and a sea ice precipitation, clouds,… erupt during the model. simulation
100-yr 24-hour rainfall Consistency with other approaches
100-yr 24-hour rainfall Consistency with other approaches
Differences between projected 100-year, 24-hour isohyetals for late-21st century and those based on NOAA Atlas 14 These results are not designed for operational use, nor do they replace the existing sources
Projected flooding in Cook County, IL Managing Increases in Extreme Precipitation In Cook County, IL: The 100-year rainfall in 2050 is The 165-year rainfall today The 100-year rainfall in 2100 is The 250-year rainfall today SRES A2. 24-hour Precipitation The 1% Annual Chance (100-year) floodplains of Upper Salt Creek (left) and Stony Creek (right) respond very differently to changes in extreme precipitation, highlighting the need for climate modeling and mapping of impacts at a local scale. SRES A2 for Mid-Century (2046-2065) and to Late-Century (2081-2100).
Final thoughts on Bulletin 75 • The frequency estimates in the bulletin represent the present time. Climate studies indicate that the trends will continue in the future. • As a result of the increasing rainfall, the structures designed based on the 2020 Bulletin 75 may not be adequate for future (e.g. 2050) rainfall magnitudes. • Thus, similar studies using the projected climate modeling data need to be performed to determine future rainfall frequencies.
Questions/comments?
European Experiences
Recommend
More recommend
