Lab 8

Before beginning my research on the fire spread of the summer of 2009 in Los Angeles County, I used my previous knowledge of the local terrain to generate a hypothesis of how I believed the fire spread pattern would look. Knowing that the areas most prone to fire are located near steep topographic features and highly-flammable land cover, I thought that the fire pattern would remain somewhat contained on steeper topology with sparse vegetation. As the fire spread down the mountains and hills, I then believed the fire pattern would reach its maximum spread near the base of the steep features where the vegetation cover increased and the topology became flat, thus more prone to quickly spread the fire. I also believed that these flatter areas of topology would be located closely to urbanized, populated areas and therefore would pose an increased safety risk during the late summer months as the fires continue to increase in size and spread across larger areas.

In order to test my theories, I first began my research by downloading a map of LA County from the National Map Seamless Server (USGS, 2010). I chose to download the county map from this site because it highlights the topology of regions, thus would allow me to examine the terrain encompassed by the fire spread. After downloading this map, I then retrieved a shapefile of the Los Angeles County fire spread from August to September 2009 from the Los Angeles County Government’s GIS department site (Los Angeles County Enterprise GIS, 2009). Knowing that I not only wanted to map the fire spread in relation to the terrain/topology, but that I also wanted to map fire spread patterns in relation to their location in and around urbanized, populated areas in the county, I then went to UCLA’s MapShare site in order to download shapefiles based on 2000 Census data related to urbanized areas, populated areas, and census tracts in Los Angeles County (ESRI, 2008).

After downloading all of the shapefiles necessary to conduct my research, I then began to upload them into ArcView. I first uploaded the shapefile from the National Map Seamless Server in order to generate a hill shade map of Los Angeles County’s terrain. Creating the hill shade map would tell me which areas have the steepest terrain and which areas have the flattest. Next, I added all of the other census layers to the DEM in order to map the relations of urban/populated areas and the topology of the county. After placing all of the census overlays to the DEM, I then uploaded the fire-spread data over a four-day period. Finally when I completed inserting the overlay layers, I changed the color ramps and fills of each individual layer in order to make them all visible on the map.

After I completed the map, I then looked to analyze the pattern of the fire-spread upon the affected topology. Looking at the first day of fire spread (Fire Spread 1), I noticed that due to the light blue undertones of the magenta polygon, this fire was beginning to spread from the top of a steep topographic feature within the Angeles National Forest. I was also able to know that this was the top of a topographic feature because when looking at the center of the Fire Spread 1 polygon, I noticed that the topographic light blue lines converged into a distinct point, therefore indicating the top of a mountain or hill. This first day of fire-spread pattern reinforced my original hypothesis for according to the map the fire was concentrated in a small area on the steepest part of the terrain. When considering the spread of the fire over the terrain for the next three days, I also came to the conclusion that the data presented in my map reinforced other parts of my hypothesis for a number of reasons. First, when looking at the second day of fire, or Fire Spread 2, it is apparent that the fire was spreading down the mountain, as I originally had thought it would, and was still spreading in a pattern consistent with the first day, thus still continuing to spread in a contained manner toward the North. When looking at the third day of fire spread (Fire Spread 3), however, I noticed a change in the pattern of the fire. Again backing up my original theory, after analyzing the topology of the map I realized that this change was due to a change in the shape of the steep topographic features upon which the fire originally started. Looking at the map I realized around the third day, the fire spread had expanded towards the northeast and in no other direction as it had done in the previous days. This may have occurred because when considering the topology in the map, the topological features affected during day three of the fire spread are still located in steep areas (seen in the light blue undertones of Fire Spread 3), while those areas affected by the fire during day four have darker undertones indicating a flatter terrain. These differences in topology allowed me to conclude that before the fire could spread to the flatter areas in the northeast, it first needed to travel down the steep slopes of the terrain towards the northwest. Looking at the expansive spread pattern of Fire Spread 4, my hypothesis is once again reinforced. The darker undertones located under the orange-colored spread indicate that these areas are mostly flat terrain since the darker pink color on the DEM color-scheme indicates flat topology.

When analyzing the fire spread patterns in relation to locations of urbanized and populated areas, I was surprised to see how few census tracts the fires in summer of 2009 actually affected. Looking at the results on the map, my hypothesis was proven incorrect, for those urbanized/populated areas most affected by the fires were not only harmed during the late summer, but during the early and middle parts of the summer as well. All of the affected communities are located on the northern border of the populated and urbanized zones between the Valley and the Angeles National Forest in Los Angeles County. Considering the spread of the fire, it looked as if during the first day the fire may spread in a way that would support my hypothesis since it only harmed a small section of a one or two census tracts. As the spread expanded, however, it began to affect more and more communities each day, therefore refuting my suggestion that communities would only become affected on the last day of the spread once the fire reached flat terrain.

All of these areas affected by the fires qualified as both populated and urbanized regions. This characteristic allowed me to conclude that they have larger amounts of people per square mile than those areas that are classified as only populated and therefore can be considered high-risk communities “identified within the wildland urban interface (where structures and other human development meet or intermingle with undeveloped wildland)” (FRAP, 2010). In the state of California, these high-risk communities are required to develop Community Wildfire Protection Plans that “must be collaboratively developed (with agreement among local government, local fire departments and the state agency responsible for forest management), identify and prioritize areas for hazardous fuel reduction treatments, and recommend measures that homeowners and communities can take to reduce the ignitability of structures” (FRAP, 2010). During the 2009 Station Fire shown in the above map, one of the largest safety concerns for the affected communities was the risk of debris flows and flash floods occurring after the fire ceased. According to US News and World Report, the Station Fire of 2009 was one of the largest fires in the area’s watershed history and the park rangers and engineers were concerned that larger than normal quantities of debris would be captured by the flood control system and increase the systems stress and likelihood to burst under the pressure (US News and World Report, 2009, p.1). If the floodgates were to burst, the thousands of homes at the foot of the forest’s edge could be washed away and many larger animals that survived the fires could be forced to migrate into the communities on the flatter terrain. All of these events further increase the risk of living in the communities at the foot of the Angeles National Forest and reinforce the need for comprehensive wildfire/debris flow planning to take place in high-risk urbanized communities.


Works Cited:

Enhance Public Benefits from Trees and Forests: Planning for and Reducing Wildfire Risks. (2010). Fire and Resource Assessment Program (FRAP). Retrieved May 22, 2010, from California Department of Forestry and Fire Protection (FRAP) website: http://frap.fire.ca.gov/assessment2010/3.3_wildfire_planning.html

ESRI. (2008). Los Angeles County Census Tracts [Data file]. Retrieved from
http://gis.ats.ucla.edu//Mapshare/Default.cfm

ESRI. (2008). Los Angeles County Census Urbanized Areas [Data file]. Retrieved from
http://gis.ats.ucla.edu//Mapshare/Default.cfm

ESRI. (2008). Los Angeles County Populated Place Areas [Data file]. Retrieved from
http://gis.ats.ucla.edu//Mapshare/Default.cfm


Los Angeles County Enterprise GIS. (2009). All Station Fire Perimeters (as of September 2, 07:02) – Complete set [Data file]. Retrieved from http://gis.lacounty.gov/eGIS/?p=1055

Post-Wildfire Worries: Floods, Damaged Ecosystem. (2009, September 8). Retrieved May 22, 2010, from US News and World Report website: http://www.usnews.com/science/articles/2009/09/08/post-wildfire-worries-floods-damaged-ecosystem.html?PageNr=2

USGS. (2010). National Map Seamless Server [Data file]. Retrieved from http://seamless.usgs.gov/

Lab 6

I chose to map the topology of the Hawaiian Islands Lanai, Molokai, and Maui. I believe these Hawaiian Islands are very interesting to look at in terms of topology because they are volcanic and mountainous, thus vary completely from the surrounding sea-level topographic characteristics of the Pacific Ocean. In terms of decimal degree coordinates, this group of islands is located at (.00028, .00028). Looking at latitude and longitude, the Hawaiian Islands of Lanai, Molokai and Maui are located solely in the Northern and Western Hemispheres in UTM zone 4N. Their locations range from 21.18 degrees to 20.76 degrees in the North to -157.06 to -156.5 degrees in the West. These topographic maps use the spatial references of GCS_North_American_1983 as well as the North American Datum of 1983.

Map Projections

Map projections are important for many reasons. One of the main reasons is that map projections are necessary for creating maps. They take images from the 3D geoids of Earth and project them onto a 2D plane, thus making flat maps which are easier to carry, use, and digitize for GIS purposes. It is important that there is not only one type of map projection, but instead a variety of map projections since a different projection will be required depending on the purpose of the map. The maps purpose will also help decide what characteristics need to be preserved and what characteristics can be distorted, thus helping to find the proper projection. For example, for navigational purposes, one would want to use a conformal projection because it preserves direction, and for distance measuring, one would most likely want to use an equi-distant projection due to its equally spaced standard lines. For certain maps, however, it does not matter which projection you use especially if the purpose of the map is for thematic or illustrative purposes. One may also pick a different map projection based on what part of the world they are looking at, for different projections preserve and distort different parts of the globe.

The Mercator projection is a conformal projection that preserves direction and shape but varies in distance and area. In this projection, the N/S run of the map is longer than the E/W portion, thus possibly leading to more distortion in distance. When looking at the Mercator projection above, I believe this may be one of the reasons that the distance between Washington D.C. and Kabul is much larger than the true distance (10,098.61 miles [Mercator] v. 7,000 miles [True Distance]). Another reason that the projected distance on the Mercator map may be so different from the True Distance is that Mercator projections are many times used to examine Polar Regions and the cities under study in this map is located outside this region, thus causing a distortion in the distance. The Mercator projection does preserve distance, however, so this map could be very useful to a sailor or pilot for navigational purposes between the two cities. The other conformal map shown above is the Gall Stereographic projection. This projection is azimuthal, preserves angles and direction, while distorting area, object size and distance. Looking at the examples above, I thought that it was interesting that even though the Gall Stereographic projection distorts distance, the projected distance on the map was still very close to the True Distance (7,175.43 miles [Gall Stereographic] v. 7,000 miles [True Distance]). I believe this may be due to the lack of large distortions found at the poles, causing the continents to be properly spaced from one another and not squished together like in the Mercator Projection.

Equal Area maps preserve the area of the projected map surface. The Bonne projection preserves size, and has evenly spaced parallels that are true to scale along the concentric arcs. This projection also lacks distortion along the central meridians and parallel. I believe the Bonne projection’s distance (6,716.91 miles) is so closely related to the True Distance (7,000 miles) between the two cities due to the projections ability to preserve size. As seen in the conformal projections, the size of objects on the map greatly affects the spacing of the continents on the projections by causing them to converge or diverge. With the preserved size, however, all objects on the map are true to scale, thus allow for a more accurate measurement between the two cities. The Sinusoidal Projection is also an equal area projection. It is actually a special case Bonne projection. The Sinusoidal Projection preserves distances along horizontal lines, and all of its parallels are standard lines. This projection becomes distorted toward the poles and is slightly distorted along the equator. I believe the distance of the Sinusoidal Projection is relatively close to the True Distance between the two cities (8,100 miles [Sinusoidal] v. 7,000 miles [True Distance] for a couple of reasons. One of the main reasons is that both the N/S scale and the E/W scale are equal, therefore making distances more accurate. Also, in a Sinusoidal Projection, the true distance between two points on the same meridian corresponds to different points on the map between two parallels. This creates a larger projected distance between the two points on the map, hence explaining the above results.

Equidistant Projections preserve the distance between points on a projected surface. In the above examples, this is best seen in the Equidistant Conic projection, where due to its ability to preserve distance, out of all the other maps, this map’s projected distance is the most closely related to the True Distance on the ground (6,728.47 miles [Equidistant Conic] v. 7,000 miles [True Distance]). The Plate Carree equidistant projection, however, is a completely different story. Even though it is supposed to be an equidistant projection, the Plate Carree’s projected distance is much larger than the True Distance, coming in at 10,247.44 miles instead of around 7,000 miles (True Distance). I believe this may have occurred for a few reasons. First, the Plate Carree does not preserve shape or size. This may have caused the shapes of the continents to stretch out when projected, thus increasing the distance between the two cities. The Plate Carree’s inability to preserve size and shape are one of the main reasons it cannot be used for navigational purposes. Second, the Plate Carree also does not deal with complex relations between points on the map and their relation to points on the Earth, but rather uses simple relations which may make the distance less accurate.