Mapping Flood Risk in the Upper Fox River Basin: Vulnerable Populations and Adverse Health Effects Presented by: Angelina Hanson
STUDY AREA: Wisconsin's Upper Fox River Basin Total Population 139,309. (2010 Census Bureau Data) This water rich basin has 15 watersheds drained by 1,257 miles of rivers and streams. It contains 154 lakes larger than 10 acres in size and 10.5% of its land cover is wetland. (WI DNR) The Upper Fox River Basin is an area known for its flooding. There were 39 days of flooding between January, 1996 and June, 2015. The last significant flooding event was in 2008. The most active year, 2004, experienced 10 days of flooding. The cumulative damage caused by these events was an estimated $46.5 million in property damage and $183.0 million in crop damage. (NCDC Storm Events Database) Reference: http://www.friendsofthefox.org/explore/geography/ 2
Objective and Background The objective of this study is to evaluate the flood risk in Wisconsin s Upper Fox River Basin, identify the social and economic background of the communities living within or close to the floodplain, and to investigate the adverse health effects related to flooding events. It is a student internship project supported by the Wisconsin Department of Health Services through their BRACE grant. Other contributors are the Wisconsin Emergency Management and Shane Hubbard at the Space Science and Engineering Center at UW-Madison. 3
Study Details Map the Flood Risk: Generate a flood scenario (floodplain) to understand the location and severity of possible flooding. Locate the Risks and Vulnerabilities: Using the Federal Emergency Management Agency s (FEMA) Hazus-MH software, estimate building damages and economic loss. Using US Census Bureau statistics, assess the socioeconomic status of those communities living within or close to the floodplain. Investigate adverse health effects related to flooding Literary Review via PubMed. Preliminary focus on severity of flooding related to Mental Health outcomes. Estimate the potential impacts of climate change Contrast and Compare 100 year and 500 year scenarios; e.g. risks and vulnerabilities. 4
FEMA s Hazus-MH software FEMA s Hazus-MH software is a standardized methodology that estimates losses from earthquakes, hurricanes, and riverine and coastal flooding. It uses Geographic Information Systems (GIS) to map and display hazard data and provides damage and economic loss estimates caused by these types of disasters. (Department of Homeland Security http://www.fema.gov/hazus-mh-overview accessed 10.16.2015) This study uses Hazus-MH to calculate damages, but also provides demographic data that assists in identifying vulnerable populations. 5
Generating the Flood Scenario (Floodplain) December 2014 No current flood depth grid for this area. Hazus-MH must have a flood depth grid to calculate damage and loss estimates. Level of flooding = how damaged are the buildings and estimate economic loss to repair or rebuild. April 2015 FEMA s Hazus-MH (Level 1 Analysis). Failed to capture the floodplain accurately. Boundary was inconsistent with the DFIRM. Issue: building damage / economic loss. Communities of interest (Census Statistics). Enhanced Quick Look Hazus-MH tool in ArcGIS also failed to capture the floodplain accurately. 6
Generating the Flood Scenario (Floodplain): Alternative ArcGIS Methodology Download FEMA s Digital Flood Insurance Rate Map (DFIRM) for the study area. Gathered and use Flood Insurance Study reports for each county (most recent) to build on the DFIRM. Produced an edited DFIRM that was then subtracted from a USGS Digital Elevation Model (DEM) layer. RESULTS 7
Alternative Method Results Produced a very accurate flood scenario. 8
100 Year Flood Results: Global Statistics Report General Building Stock Damage Hazus-MH estimates 335 buildings will be at least moderately damaged. This is over 6% of the total number of buildings in the flood scenario. Building Value (thousands of dollars) Total Study Region: 17,537,802 Essential Facility Damage Classification Total At Least Moderate At Least Substantial Fire Stations 39 0 0 0 Hospitals 5 0 0 0 Police Stations 18 0 0 0 Schools 103 0 0 0 Loss of Use Significant finding: Residential buildings damaged 41-100%; 138 of the total. 9
100 Year Flood Results: Census Bureau Statistics Census Blocks that Intersect the Floodplain 1099 Census Blocks 19,365 people (14% of total population) 8720 households (16% of all households) 3694 people under 16 yrs old (13%) 3740 people over 65 yrs old (17% ) 1380 households with income 20K or less (15%) Study Region Census Statistics 7918 Census Blocks 139,174 people 54,934 households 28,795 people under 16 yrs old 22,481 people over 65 yrs old 9318 households with income 20K or less 10
Health and Vulnerable Populations Vulnerable Populations Borrowing from the concept of Social Vulnerability Indexing (SOVI), Social vulnerability is represented as the social, economic, demographic (emphasis mine), and housing characteristics that influence a community s ability to respond to, cope with, recover from, and adapt to environmental hazards. Reference: Hazards and Vulnerability Research Institute, University of South Carolina The Elderly and the Young Difficult to get out of harms way. Susceptible to Illness. Dependent on others for help. 11
Mental Health and Vulnerable Populations Study: An exploration of factors affecting the long term psychological impact and deterioration of mental health in flooded households Jessica Elizabeth Lamond et al., April 2015. Pre-existing conditions: age, household income level, occupation of main income earner, number of people living in the household Severity of flooding event: Flood water level Injury Death of a relative Property loss Their model suggests that those on very low incomes are eight times more likely to report severe mental health deterioration than those on higher incomes. Study: Resilience and Vulnerability to the Psychological Harm From Flooding: The Role of Social Cohesion Giles Greene et al., September 2015. Connection to the severity of the flooding event: Following the summer floods in England in 2007, we found that the prevalence of mental health symptoms (psychological distress, anxiety, depression, and post-traumatic stress disorder) was up to fivefold higher among individuals affected by floodwater in the home. 12
100 year versus 500 year flood scenario WHAT IF the 500 year flood scenario becomes the new 100 year scenario in 50 years from now? 100 year scenario 1099 Census Blocks 19,365 people (14% of total population) 8720 households (16% of all households) 3694 people under 16 yrs old (13%) 3740 people over 65 yrs old (17% ) 1380 households with income 20K or less (15%) 500 year scenario 1221 Census Blocks 21624 people (15% of total population) 9621 households (17% of all households) 4162 people under 16yrs old (14%) 4006 people over 65 yrs old (18%) 1542 households with income 20K or less (16%) The percentages below indicate a % living in a census block that intersects the floodplain. THIS DOES NOT NECESSARILY MEAN THEY ARE LIVING WITHIN THE FLOODPLAIN. 13
Further Research Mental Health: Perform Network Analysis on the location of mental health facilities relative to census blocks that have the largest number of households with income less than 20k/yr and the greatest number of substantially damaged residential buildings. Investigate additional adverse health effects associated with flooding: Mortality Injuries Gastrointestinal Illnesses Skin Infections Mold Exposure Chemical Exposure Add additional layers for overlay analysis: Major employers Financial Institutions Day Cares Hazardous Facilities Hospitals / Clinics 14
LinkedIn: https://www.linkedin.com/in/angelinahansonmadisonwi Email: angelina.hanson@wisconsin.gov ahanson2662@gmail.com SPECIAL THANKS Wisconsin Department of Health Services Bureau of Environmental and Occupational Health BRACE Program Wisconsin Emergency Management Shane Hubbard Assistant Researcher Space Science and Engineering Center 15