Untitled-2.png Back Home Again in Indiana THE USE OF HAZUS TO ASSESS POTENTIAL IMPACTS OF CLIMATE CHANGE ON FUTURE FLOOD LOSSES: AN UPPER MISSISSIPPI RIVER VALLEY CASE STUDY Flooded_Farm.JPG The Flood Model Session Jonathan Remo Ph.D., P.G. Post Doctoral Research Fellow and Asst. Adj. Professor Southern Illinois University, Carbondale SIULOGO.bmp Untitled-2.png Climate Modeling Not Complete C:\Users\Jonathan Remo\AppData\Local\Microsoft\Windows\Temporary Internet Files\Content.IE5\AG94JI6O\MC900423165[1].wmf SIULOGO.bmp Methods + Present Day HAZUS Flood Loss Estimated + Comparative Analysis = C:\Users\Jonathan Remo\AppData\Local\Microsoft\Windows\Temporary Internet Files\Content.IE5\ZJCLXIPB\MC900423171[1].wmf C:\Users\Jonathan Remo\AppData\Local\Microsoft\Windows\Temporary Internet Files\Content.IE5\ZJCLXIPB\MC900423171[1].wmf Problem Statement –Flood frequency studies and consequently flood risk studies often do not account for potential changes in physical processes such as climate change (i.e., non-stationarity). –Potential resulting in the systematic underestimation of flood risk disseminated in U.S. Policy and to the public. –Underestimation of flood risk has been in part to continue the trend in floodplain development SIULOGO.bmp Research and Modeling Goals •Develop a framework from which to incorporate from which to incorporate climate change into flood risk and flood loss assessment. •Determine the potential change in magnitudeof flood losses for likelyfuture climate scenarios. SIULOGO.bmp Use HAZUS-MH provided database Level 1 Level2 Level3 Modify building inventory using local data Integrate output from other models and community-specific damage functions Levels of Analysis in HAZUS Expert Supplied Data Levels of Analysis and CredibleModeling Default Database User Modified Data SIULOGO.bmp SIULOGO.bmp Credibility Credibility Developing CreditableModels Quantifying Uncertainty Modeling Goals Assumptions and Limitations Clear & attainable modeling goals Clear explanation of model assumptions and limitations Perform calibration, validation, and sensitivity analysis Credibility Credibility SIULOGO.bmp Modeling Team me.jpg Dr. Katharine Hayhoe is an atmospheric scientist with expertise in climate change issues related to global modeling and regional climate impacts. Dr. Keith Cherkauer is a hydrology and water resources engineering who areas of expertise include cold climate hydrology, remote sensing & large scale data analysis. Jonathan is a mild mannered fluvial geomorphologist whose areas of expertise include hazard modeling, fluvial geomorphology, hydraulic modeling, hydrology, and geographic information systems. SIULOGO.bmp Climate Modeling Models Description HadCM3 Met Office Hadley Centre, UK GFDL NOAA–Geophysical Fluid Dynamics Laboratory PCM DOE -Parallel Climate Model Scenarios A2High population growth, slow economic developmentand slow technological changes. A1B Very rapid economic growth, global population peaks mid century, new and more efficient technologies B1 Same populationas A B Rapid chancein economic structures toward a service and information economy SIULOGO.bmp Climate Modeling SIULOGO.bmp Projected surface temperature changes for the early and late 21stcentury relative to 1980-1999 Change in Precipitation Relative changes in precipitation for the period 2029 –2099, relative to 1980 to 1999 -Scenario A1B SIULOGO.bmp Hydrologic Modeling with VIC Changes in Peak Flows SIULOGO.bmp Finally -Flow Frequency Analysis Study Region: Upper Mississippi River Basin Upper_Mississippi_Basin.jpg Study Reach Study_Reach.jpg Starting Point Assumptions 1.No significant channel change 2.Levees protection levels remain the same 3.Assumes no flood fighting 4.Levees do not fail but overtop 5.Sufficient length of flood flow duration to fill levee cells SIULOGO.bmp Hydraulic Modeling Levee Challenge 100_year_flood_v2.jpg Levee vs. Current 100-yr Flood Profile 340 360 380 400 420 440 460 480 35 45 55 65 75 85 95 105 115 125 135 145 155 165 175 185 195 205 215 225 235 245 255 265 275 285 295 305 Elevation (ft NGVD 1929) River Mile Illinois Levees Missouri Levees 100-year Flood Elevation SIULOGO.bmp Columbia Levee District Ste. Genevieve County #2 Levee HAZUS RUNs •100 and 500 Year Floods •1/3-Arc Second DEMs •Flood Depth Grids –Generated from water surface elevations from the Upper Mississippi River System Flow Frequency Study (USACE, 2004) outsideof HAZUS. •Used HAZUS’s “stock” structure database –Update Critical Facilities in 5 Illinois Counties SIULOGO.bmp Present Day 100-Year Flood Lossesusing HAZUS ‘s “stock” structure database 100_year_Flood_losses.jpg Levee vs. Current 500-yr Flood Profile 340 360 380 400 420 440 460 480 35 45 55 65 75 85 95 105 115 125 135 145 155 165 175 185 195 205 215 225 235 245 255 265 275 285 295 305 Elevation (ft NGVD 1929) River Mile 500 -Year Flood Elevation Illinois Levees Missouri Levees Hannibal, MO St. Louis Illinois River Missouri River Chester, IL Thebes, IL SnyIsland Levee Columbia Harrisonville Stringtown Fort Charters Ivy Landing PrairieDu Rocher & Modoc Levees Kaskaskia Degognia & Grand Big Five Levees Ste. Genevieve#2 Bois Brule Metro East and St. Louis Present Day 500-Year Flood Lossesusing HAZUS’s “stock” structure database 500_year_flood_v3.jpg Flood Lossesfor Worse Case Scenario using HAZUS’s “stock” structure database 500_year_flood_Levees_Fail.jpg HAZUS Flood Loss Estimates $342 $546 $5,001 $669 $1,094 $11,771 0 2,000 4,000 6,000 8,000 10,000 12,000 14,000 100-year 500-year Worse Case Losses in Millions of Dollars Buidling Losses Depreciated Replacement Value Full Replacement Value Structure Data Comparison “Sensitivity” Analysis UDF_Verse_Aggergate.jpg Aggregate vs. User Defined Analysis Building Data ComparisonUser Defined vs. HAZUS AggregateJackson County, IL $46.0 $78.8 $28.9 $44.8 $20.4 0 10 20 30 40 50 60 70 80 90 User Defined Analysis (Local Data) Stock Aggregate Full Replacement Value (Level I +) Stock Aggregate Depreciated Replacement Value (Level I +) Updated Aggregate Full Replacement Value (Level III) Updated Aggregate Depreciated Replacement Value (Level III) Building Loss in Millions of Dollars Building Data ComparisonUser Defined vs. HAZUS Aggregate $15.7 $24.6 $46.0 $959.4 $56.0 $55.4 $78.8 $1,231.1 $25.0 $27.0 $28.9 $543.1 0 200 400 600 800 1000 1200 1400 Alexander County Union County Jackson County St. Clair County* Building Losses Milllions of Dollars Building Losses for 500 Year Flood User Defined Analysis (Local Data) Stock Aggregate Full Replacement Value (Level I +) Stock Aggregate Depreciated Replacement Value (Level I +) * Assumes Levee Failure Average Values for Residential Structures Local vs. HAZUS’s stock data $41,215 $57,920 $42,640 $49,441 $72,544 $61,776 $126,041 $117,698 $40,770 $42,804 $44,967 $62,338 $0 $20,000 $40,000 $60,000 $80,000 $100,000 $120,000 $140,000 Alexander Union Jackson St. Clair Local Floodplain Ave. for Residential HAZUS Ave. Residential (Replacement) HAZUS Ave. Residential (Depreciated) Percent Difference in Modeled Flood Losses: HAZUS Stock Data and Local Assessors Data 72.0 55.6 41.6 22.1 47.8 21.2 37.2 8.8 -59.2 -76.6 -22.5 54.2 -100.0 -80.0 -60.0 -40.0 -20.0 0.0 20.0 40.0 60.0 80.0 100.0 Alexander County Union County Jackson County St. Clair County* Average Standard Deviation Percent Difference Precent Difference Aggregate Full Replacement Value vs. UDF (Local) Precent Difference Aggregate Depreciated Replacement Value vs. UDF (Local) * Assumes Levee Failure Range of Potential Flood LossesDepreciated Replacement Cost $156.6 $250.1 $2,290.6 $342.0 $546.1 $5,001.3 $527.3 $842.1 $7,712.0 $0 $1,000 $2,000 $3,000 $4,000 $5,000 $6,000 $7,000 $8,000 $9,000 100-year 500-year Worse Case Building Losses Millions of Dollars Lower Bound Depreciated Replacement Value Upper Bound Range of Potential Flood LossesFullReplacement Cost $521.6 $1,044.5 $9,181.2 $668.8 $1,093.7 $11,770.8 $815.9 $1,325.5 $14,360.4 $0 $2,000 $4,000 $6,000 $8,000 $10,000 $12,000 $14,000 $16,000 100-year 500-year Worse Case Building Loss Miillions of Dollars Lower Bound Full Replacement Value Upper Bound Conclusions •Comparison of building losses for HAZUS’s aggregate flood analysis employing “stock” database vs. user defined flood analysis employing local assessor data show reasonable agreement. •Difference in these loss estimated can be used to provide a realistic range of flood losses. •Comparison of HAZUS flood loss estimates for present day and future climate scenarios should provide a reasonable assessment of potential changes in flood losses attributable to climate change. Thank You Belleau _Creek.JPG Contact Info: Jonathan Remo Department of Geology Southern Illinois University, Carbondale Carbondale, IL 62901-4324 E-mail diamict@siu.edu Phone 618.453.7370 Web: http://mypage.siu.edu/diamict image004.gif