New Orleans Flood Impact Analysis
with GIS and Remote Sensing
Introduction
On August 29, 2005
A Category 4 storm, Hurricane Katrina, struck Southeast Louisiana with winds of 135 mph (217 km/h).
Katrina made landfall near New Orleans, breaching levees and causing catastrophic damage to the city and significant loss of life.
"Katrina was the most destructive and expensive natural disaster in the history of the United States."
(FEMA, 2006)
Katrina Impact
- 80% of the city was flooded
- More than 1,000 fatalities
- Estimated total economic losses in excess of $125 billion
- Insured losses of $35 billion
(FEMA, 2006; Masozera & Bailey & Kerchner, 2007; Pistrika & Jonkman, 2009)
However, Hurricane Katrina was not the sole disaster that attacked the city.
"In fact, in the last 20 years, New Orleans has seen 42 flood events."
(City of New Orleans, n.d.)
My research question is:
How could New Orleans become resilient to recurrent flood events?
Study Area
New Orleans is a city in Louisiana state, located near the Gulf of Mexico and shaped by the waterways of the Mississippi River and Lake Pontchartrain. It has a humid subtropical climate with mild winters and hot summers. Its population is about 400,000.
Satellite Image
Here is the georeferenced high-resolution natural-color satellite image from Google Earth, taken on November 15, 2019.
New Orleans is long east to west, and you can see urbanized areas in the western half of the city, while the eastern half is mostly wetlands.
Slope
Data Source: U.S. Geological Survey (2017)
This map shows the Slope layer of the city with the Hillshade layer on it. This almost single light color describes how flat the city is.
Population Density
Data source: Esri. U.S. Census Bureau. (2020)
This is the population density around the city. It is reasonable that the density is higher in the southwestern urbanized residential areas.
Methods
I used three general methods for this project:
- Data collection
- GIS Analysis and Remote Sensing
- Literature Review
For Data Collection, I downloaded datasets, such as Flood Hazard Areas, Per Capita Income, and hurricane tracks from the City of New Orleans GIS Open Data sources, ArcGIS Online, and Living Atlas.
I downloaded satellite imagery from USGS EarthExplorer and Google Earth Pro.
Regarding technical methods, I first performed the Mosaic tool and then the basic DEM, Slope, and Hillshade analyses.
I also performed Georeferencing, Band Combinations, Land-Use Classification, and Change Detection analyses.
Results
Elevation (in meters)
Data Source: U.S. Geological Survey (2017)
This is the elevation of the city with a DEM raster. The dark blue and the navy colors indicate areas that are lower than the mean sea level and thus especially vulnerable to flood events.
Note that the gray areas in the north and south of the urban areas have a relatively high elevation above sea level.
Flood Hazard Areas
Data source: FEMA (2020)
These high-elevation areas shown in yellow have levees and walls to reduce flood risk, but Katrina breached some of them, resulting in the majority of the city flooded.
Elevation Map (3D)
Data source: esri_3d (2017)
I also combined the 3D buildings polygon feature with the elevation layer, which may help you grasp how extensively the urban areas could get damaged when a hurricane event occurs. Again, in the case of Hurricane Katrina in August 2005, approximately 80% of the city was flooded.
The Day After Katrina
Data source: OrbView-3 (2005)
This black-and-white satellite image was taken on the day after Hurricane Katrina attacked the city. One thing to note is the flooded areas shown in black here approximately correspond to the areas that are lower than the mean sea level shown in the elevation map. Swipe the imagery to compare them.
Impact and Income Level
Data source: cisaac_UMN (2020)
This map above shows per capita income by ward simply from low to high. The purple color describes high-income communities.
However, this comparison of the two images reveals that even some high-income communities are exposed to the hurricane flood impacts due to the city’s flat grounds and low elevation.
Evacuation Routes
Here I put evacuation routes on the DEM raster. Some of them are placed in the lowest-elevation areas and thus may not work in times of disaster.
Data source: jcpoe_NCSU (2020)
Grocery Store Location
I also put grocery store locations on the elevation map, and you can see many of them are located in low elevation areas.
Data source: City of New Orleans (2021)
From the viewpoint of food security, there have to be food storages in high elevation areas, which are relatively safe, enough to feed the entire population at least for a few days after a serious flood event occurs, given that some evacuation routes may not work, potentially making it difficult for the city to receive food aid from outside of the city.
Photo: "Bourbon Street, French Quarter" by Pedro Szekely is licensed under CC BY-SA 2.0
Leaving the urban area, let us take a look at the wetlands area in the northeastern portion of the city.
Land-Use Classification (2000 vs 2017)
Data sources: (Left) U.S. Geological Survey (USGS) Earth Resources Observation and Science (EROS) Center (2016a) / (Right) U.S. Geological Survey (USGS) Earth Resources Observation and Science (EROS) Center (2017a)
These are Landsat unsupervised land-use classification maps. The light blue areas describe wetlands in the left 2000 Landsat 7 image and the right 2017 Landsat 8 image. Based on peer-reviewed articles, coastal wetlands in Louisiana are supposed to be dramatically decreasing because of a combination of natural and human activities. However, contrary to my anticipation, the results of this classified imagery did not necessarily show the tendency.
Katrina Track (2005)
Data source: IDM801814833_ohiostate (2021)
Then I started to look at other wetlands areas while considering major hurricane tracks because past studies point out that wetlands work as crucial buffer zones to hurricanes and mitigate their impacts on New Orleans (Barbier et al., 2013).
This map shows Hurricane Katrina track in 2005, and you can see Katrina crossing New Orleans from south to north.
Most historical hurricanes have attacked South Louisiana from south to north after getting stronger above the Gulf of Mexico.
Wetlands in South Louisiana
Data source: U.S. Geological Survey (USGS) Earth Resources Observation and Science (EROS) Center (2016b)
Thus, I decided to add the vast wetlands area located southeast of New Orleans, considering it as a crucial buffer zone to hurricanes that hit the city.
Natural Color Bands Comparison (2000 vs 2017)
Data sources: (Left) U.S. Geological Survey (USGS) Earth Resources Observation and Science (EROS) Center (2016b) / (Right) U.S. Geological Survey (USGS) Earth Resources Observation and Science (EROS) Center (2017b)
I downloaded satellite imagery of 2000 and 2017 from Landsat 7 and 8, respectively, and first compared them with the Natural-Color band combination. Though the images’ brightness is quite different, making it a bit hard to compare them, it looks like the central part of the Landsat 8 image is more eroded than that of the Landsat 7 image.
Land-Water Interface Bands Comparison (2000 vs 2017)
Next, I applied the Land-Water Interface Bands for both images, which is suitable for identifying land-water boundaries clearly. Now it’s obvious that the wetlands in 2017 on the right are sparse, and you can observe huge water areas around the center of the image, which you cannot see in the left image of 2000.
Land-Use Classification Comparison (2000 vs 2017)
Additionally, I performed the unsupervised land-use classification for the imagery. These classified images may also help you see the difference in wetland spaces between the two periods of time.
Land-Use Change between 2000 and 2017
Finally, here is the image describing land-use change between 2000 and 2017, emphasizing the erosion of the coastal wetlands. I have observed a stable decrease in wetlands in this area on Satellite imagery year-by-year, so I suppose the change shown here is not because of particular natural disasters.
While Louisiana has approximately 40% of the U.S. wetlands, over 90% of the total coastal marsh in the country has been lost in the state. Some studies say Louisiana is losing a football field of coastal wetlands per hour (Boesch et al., 1994; Couvillion et al., 2017).
Oil and Gas Wells Locations
Data source: agroviz (2020)
Here I put oil and gas wells locations on the land-use change map. Many studies and articles say the oil and gas industry has significantly damaged Louisiana’s coastal wetlands by drilling wells, canal construction, and waste mismanagement (McGill, 2020).
Discussions
We have seen a variety of maps describing New Orleans’ vulnerability to flooding events, mainly due to its low elevation in relation to mean sea level.
We also recognized significant wetlands loss in other parts of Southeastern Louisiana over time.
The rapid loss of crucial buffer zones could imply further exacerbation of the city’s vulnerability.
In order to enhance the city’s resilience, South Louisiana’s coastal wetlands restoration should be the priority.
Conclusion
Initial Conclusion:
As my initial conclusions, I would say wetlands restoration effort will be the key to mitigating future storm surge impacts on New Orleans, considering the entire city’s unchangeable low elevation.
Other Measures:
Other practical measures could be incentives for elevating existing or new structures or levee upgrades. However, researchers claim that levee construction could damage coastal wetlands. Therefore, careful planning is needed.
Next Steps:
The next steps in this research will be to dig into other major causes of wetlands loss, such as coastal excavation, hydrological alteration, sea-level rise, and saltwater intrusion, and potential mitigation measures (USGS, 1996; Glick et al., 2013).
Photo: "Bourbon Street, French Quarter" by Pedro Szekely is licensed under CC BY-SA 2.0