Rebuilding a Community Post Hurricane Katrina (document link)
Rebuilding a Community Post Hurricane Katrina
By: Vernessa Shih
My topic of study was seeing the effects of the incredibly damaging Hurricane Katrina on the State of Louisiana and particularly a five parish area that was geographically closest to where the hurricane made landfall and saw the most damage. I also wanted to see how Louisiana and the five parish area recovered from this natural disaster and what areas still needed focused attention. My goal was to spatially show the areas in need of greater assistance and perhaps hypothesize about abnormalities or patterns shown in the maps.
When I began my project I identified a few questions that I wanted to address.
- What happened?
- What were the rebuilding goals?
- What were parish priorities? (particularly those in the hardest hit region)
- What areas saw the most success in rebuilding efforts?
- What areas need the most attention as of NOW?
I started with looking at what actually happened by creating a map to show the exact path of Hurricane Katrina and the strength category of the hurricane as is progressed from August 23 2005 to August 31 2005. Hurricane Katrina began as a category 1 tropical depression, peaked at a category 6 hurricane over the Gulf of Mexico before making landfall in Louisiana as a category 4 hurricane with a high of 140 miles per hour wind speed.
Area of Study: 5 Parishes
Orleans, St. Tammany, St. Bernard, Jefferson, and Plaquemines Parishes
As you can see below, Plaquemines parish, St. Bernard parish and St. Tammany parish took a direct hit from the hurricane path with neighboring areas Orleans and Jefferson parish also in very close proximity to the hurricane path.
The effects of Hurricane Katrina have deeply affected not just the residents of Louisiana or any other southern state that was physically affected, but it called into question how our nation as a whole responds to geographic disasters. The actions of FEMA and the President were criticized as being too slow and not drastic enough to account for projected damage that would result from this hurricane. The levee failures that increased the amount of flooding to up to 90% in some parishes called into question the work of the US Army Engineers Corps, who is tasked with building and maintaining levees all across America.
While accurate summaries of damage estimates are difficult to find due to changes in how government agencies were categorizing damage, I complied a few generally agreed upon facts. Over 200,000 homes were severely damaged, 1.4 million people were displaced with 15 million total people affected by the hurricane, over 1,800 deaths, extreme coastal erosion (up to 50 years of coastal damage within one week) and an estimated $75 to $110 billion dollars worth of damage. Hurricane Katrina definitely earned the title of “Costliest Hurricane in History” and this is only the monetary estimate, there is no possible was to estimate the damage it has done the the culture and community of all the affected areas, as well as the questions and doubt that arose from the mishandling by FEMA.
When the waters receded and determined citizens began looking to rebuild, the state of Louisiana formed the Louisiana Recovery Authority which produced “Louisiana Speaks – Long Term Recovery Plan”. This joint effort plan held community meetings in 25 southern Louisiana parishes over the course of two days and asked them a series of questions to rank their desired priorities of action. Overwhelming, 98% of the parishes strongly agreed that they should build back differently to address issues of poverty, hurricane/flood risk and environmental risk. 74% of surveyed respondents agreed that some places in Louisiana are too at risk to rebuild. And one perfectly divisive issue with a 50/50 split was whether or not everyone who wanted to return could come back to their original home site. Overall, when the respondents were asked to rank their most important priorities from a selection of about 20 options they favored building better levees, encouraging development, improving schools, increasing business and job opportunities as well as devising a workable evacuation plan.
One of the major criticisms of the handling of Hurricane Katrina was regarding how FEMA reacted to the hurricane. It took multiple days to evacuate those in affected areas and then several more days to get the necessary supplies to the Superdome which sheltered up to 25,000 hurricane survivors. In the map below, I aimed to show the approved FEMA evacuation routes for Louisiana and highlight what I believe to be one massive error. The coastal areas of Louisiana are those which are most susceptible to damage and regardless of the lower population density, those regions should be more heavily covered with evacuation routes.
And as difficult as it was to track displaced citizens, I was able to calculate a breakdown of where residents were tracked going through FEMA aid requests. It seems a majority of those that left the state went to neighboring Texas (about 90,000 people) and Mississippi and Georgia (about 18,00 people per state). Those moving instate favored Baton Rouge with about 34,000 displaced residents. Because of the transient nature of disaster victims, I chose to represent their immediate preferences (between 2006-2007) with these pie charts.
In further study of displaced residents, I wanted to see how many parishes recovered their lost residents by looking at population changes and changes in occupied housing between the years of 2000 and 2010. As you can see with the Population Change map, St. Bernard and Orleans parish lost a significant percentage of their population with obvious gains in neighboring St. Tammany, Tangipahoa, Livingston, and Ascension parish. Further evidence of this population shift is seen in the change in occupied housing. Again, it is Orleans and St. Bernard Parish which saw the most drastic losses with gains in neighboring parishes that suffered less damage.
Considering that infrastructure and improvements in schools were listed as major priorities to the entire state of Louisiana as well as the 5 parish study area, I chose to look at the prevalence of schools and hospitals in Louisiana. While initially, it seemed that the spread of schools and hospitals was pretty even across the state of Louisiana, I decided to show a distance buffer of 5 and 10 miles over the current state population to analyze whether there were under served communities.
Looking at the Service Area for Hospitals map, I can identify many areas that are under served including much of Jefferson, St. Bernard, and Plaquemines parish. There is also a significant under served population in highly populated Alexandria area. When I created the Service Area for schools map, I saw that there was much more complete coverage, except in certain areas in St. Bernard and Plaquemines parish. While theoretically, it makes sense for there to be more schools than hospitals, as hospitals have higher start up capital and associated costs, there are some highly under served communities that will require increased investment in hospital expansion. However, it is important to note that up to 30 hospitals were affected or damaged by Hurricane Katrina with about 10 closing permanently due to damage.
I then chose to look at dropout rates for students between 7th and 12th grade as an indicator for improvement of schools. In order to do this I had to create an entirely new dataset that had specific dropout numbers for each of the individual schools in the five parish study area. After geocoding each of these schools, I was able to compare the number of students who dropped out in 2005 and the number of students that dropped out in 2010 to see which areas saw the greatest decrease in dropout students. I was personally interested to see if there would be a correlation between dropout rates and race and as Orleans Parish in particular has a high rate of African American residents. Therefore, I chose to overlay African American population data to see if there were any outstanding patterns. The largest pattern I saw was that the majority of reduced dropout rates came from downtown metropolitan New Orleans City which has a densely populated African American population. There actually seems to be no strong correlation between population density of African American residents and reduction or increase in dropout rates.
The final map I compiled was to look at multiple attributes and create an index to see the areas that I believe require the most focus and attention for additional recovery and growth. I began by collecting data for Louisiana as a whole, for residents living below the poverty line, the number of vacant housing units per parish, the number of unemployed workers in the labor force and the number of high school dropouts per parish. I created a raster for each of these attributes and reclassified them to show areas of highest percentages to be of higher risk and thus importance. I then used map algebra to create a Priority Need Index for the entire state of Louisiana. The findings from this map show that Orleans, Jefferson, Lafayette, Calcasieu, Baton Rouge, Ouachita, and Caddo parishes have the highest priority on the index. While I understand that these regions might have been identified due to the economic downturn which increased unemployment rates across the entire nation, it is important to note that Louisiana still has devastating lingering affects due to Hurricane Katrina that needs to be addressed. As more disasters occur, we cannot forget the communities already affected by disaster and call for change and growth with our nation’s disaster response, management and recovery.
I found geospatial mapping to be extremely useful in showing dramatic areas of need or priority. Using rasters and change maps, it is easy to quickly identify differences across multiple areas in a far more dynamic method than showing changes in data tables. That being said, I identified several areas across that state of Louisiana that should be hailed for their decrease in dropout rates or low percentage of unemployed residents. However, I also identified several areas that are struggling to this day. Orleans and Jefferson parish were hit particularly hard by Hurricane Katrina and while they have made great strides to recovery, we can see with the Need index that they still have a long road ahead.
I ran into several complications with data regarding access to databases and clearinghouses that have since been closed to public access. Disaster data by nature is quite unreliable due to the constant changes in population and displaced residents changing locations.
For the creation of the base maps and attribute tables, I used geoprocessing clipping and joining skills, as well as boundary sub-set selection, aggregating attribute fields and creating custom shape files to create the 5 parish map and joining for attaching attribute data. To highlight features or points I used point/line graduated symbols and KML files. I used images and pie charts to highlight additional information that was not shown spatially. I then created original data by geocoding addresses and did distance analysis by creating concentric buffers. I used spatial analysis and model building to create and reclassify rasters in order to show hotspot analysis for a Priority Need index.
- Esri Tiger for Geographic Shapefiles
- Social Explorer/Census for attribute data
- Google earth – Hurricane Katrina Route
- Louisiana Site Selection for Hospital and School data
- Louisiana Department of Education for 7-12th grader dropout rates
- Greater New Orleans Community Data Center for comparison data, Orleans Facts
- FEMA.gov for FEMA evacuation routes and post Katrina reports
Model: Used Model builder to turn features to rasters, I then reclassified the rasters and used map algebra to create an index of Priority Need