My spatial exploration began in Spring of 2013 when I began my firsts steps toward achieving the GIS Certificate at UMW; since, I have become increasingly excited by the technology. As my skills and abilities grow, my understanding of the vitalness of spatial analysis and applications intensifies.
PAHs - Behavior and Distribution
An ongoing undergraduate research project at the University of Mary Washington investigates the spatial and temporal distribution of polycyclic aromatic hydrocarbons (PAHs) in the lower Chesapeake Bay Basin. Initial investigation was done to propose a potential concentration gradient accross the study area. Future work aims at relating this model to the actual concentration gradient seen in sediment and fish liver and muscle tissues throughout the study area.
Through geocoding, identified sources of PAHs were compiled into spatial feature classes. A linear relationship was hypothesized between PAH concentration and source location; thus, Euclidean distance raster files were created for each source type. A weighted overlay was completed for these sources, such that priority was given to petroleum power plants and major and minor roads (due to the serious impacts of coal tar).
Model builder was used to stream-line the process. This method can be improved by adding wind to the model, as this is attributed to atmospheric behaviors of PAHs. Additionally, more weight should be given to areas most proximal to source areas. This is especially true for sources with extremely concentrated carbon burning. Additionally, consideration should be given to water flow within the system, a medium of PAH travel.
LMW PAHs are known to source from atmospheric deposition while HMW PAHs tend to be deposited quickly through sorption to particulates. Interestingly we see pinker color in the west; these areas of few identified sources are being greatly impacted by eastern point sources.
Through geocoding, identified sources of PAHs were compiled into spatial feature classes. A linear relationship was hypothesized between PAH concentration and source location; thus, Euclidean distance raster files were created for each source type. A weighted overlay was completed for these sources, such that priority was given to petroleum power plants and major and minor roads (due to the serious impacts of coal tar).
Remote Sensing
The basics of imagery interpretation and remote sensing are being explored currently at the University of Mary Washington, Fall 2014. This course has exposed me to fundamental principles of image analysis, photogrammetry, concepts of RADAR and LiDAR, and the associated applications of GPS and GIS.
Supervised with 285 training sample, the City of Fredericksburg is classified into 10 land use covers. Average probability of fit was 6.5.
A lidar point cloud of Charleston, SC was interpolated into an elevation surface via Kriging. Differences between this interpolated surface and elevations determined in a DSM model are shown in blue.
The process of pan-sharpening was explored. It is evident that the spatial resolution of the panchromatic image is being applied to the higher spectrally resolved MSI. Details within the 10m cells can be observed, though it is not obsolete.
Supervised with 285 training sample, the City of Fredericksburg is classified into 10 land use covers. Average probability of fit was 6.5.
Obesity Varying with Poverty, Development, and Marital Status (2007-2010)
A semester long capstone project was completed Fall of 2013 at the University of Mary Washington. This project spatially analyzed relations between obesity rates, poverty rates, development, and marital status in the Commonwealth of Virginia at the Health District level over a four year time span. Degree of development was inferred to be highest closest to major interstates. A static measure of development would have been more appropriate, i.e. percent impervious cover or population density. Marital status was observed via accumulated rates of widowing, divorce, and separation. Intense data management was required to gain functional data both in tabular and spatial form.
It is observed that health districts closer to major interstates have higher rates of obesity. This is most strongly seen in the Southwest along I-81, along the I-95 corridor between Richmond and D.C., and in districts bordering I-64; these areas have highest rates of obesity. It is hypothesized that regions closer to major interstates are more developed. The fast-food industry of urban areas and outdoor center lifestyles of rurally located citizens may relate to obesity.
It was predicted that areas with higher rates of poverty would coincide with areas of higher rates of obesity. This is based off the notion that people of poverty have less access to organized exercise facilities, health care, and the means to purchase healthier, but pricier, foods. It is observed that there is no consistent relationship between obesity and poverty rates. This may be due to humanitarian efforts and the recent change is legislation.
A positive relationship was predicted. It was hypothesized that individuals experiencing marriage ending would be more prone to depressed feelings. The positive correlation between depression and over eating supports this prediction. No relationship was observed. Rather, it can be argued that a positive relationship exists between marriage endings and poverty. This may be due to a lack of access to counseling services in poverty stricken individuals or a lack of community support.
It is observed that health districts closer to major interstates have higher rates of obesity. This is most strongly seen in the Southwest along I-81, along the I-95 corridor between Richmond and D.C., and in districts bordering I-64; these areas have highest rates of obesity. It is hypothesized that regions closer to major interstates are more developed. The fast-food industry of urban areas and outdoor center lifestyles of rurally located citizens may relate to obesity.