Thursday, April 27, 2017

Spatial Analysis of Vector & Raster Data/Vector Analysis

This is a spatial analysis map in which potential campsites have been highlighted and displayed based upon specific criteria. Each campsite must be within 150 meters of a stream or 500 meters of a lake, within 300 meters of a road, and must be outside of conservation areas. In order to classify these areas and group them accordingly, I had to use the overlay toolset in ArcMap (erase, intersect, and union functions) to manipulate my campsite selection to fit the parameters and to erase unwanted area such as the conservation areas. The display was edited to blend in with the basemap image and essential map elements were added for additional clarification.

Tuesday, April 18, 2017

Ground Truthing

This is an updated version of the previous LULC map. It is essentially the same MXD file as the other map with one added shapefile point layer representing randomly assigned points that were then ground truthed by using the google maps as a reference. The points are either marked with a green pentagon for a 'true' correctly classified feature, or a red pentagon for a 'false' incorrectly classified feature.

Friday, April 14, 2017


This is a land use/ land classification map of Pascagoula, Mississippi. The classification scheme is set into several different classes: residential (light green), commercial (orange), industrial (red), mixed (grey), retail (yellow), deciduous forest (dark green), mixed forest (brown), bare soil (light blue), riparian landscape (dark blue), lakes (purple), and canals (pink). Essential map elements were also added for clarity.

Thursday, April 13, 2017

Network Analysis/ Model Builder

This is a basemap with Lake County, Fl highlighted with a clear outline around its borders representing EMS locations and a route between three specific ones.Using provided data, both from UWF and online, I geocoded the addresses by setting up an address locator in arcmap and matching them either to the provided data or manually using google maps. By adding three random stops from location on the map in the network analysis window, adding the parameters,  and specifying route options, I found the optimal route and added it to the map as a new data frame for the inset map. i then added essential map elements to clear everything up.

Thursday, April 6, 2017

Georeferencing, editing, and ArcScene

 The first map document contains two versions of the same image showing UWF campus in relation to an eagle nesting site. The first image contains two separate raster files that have been georeferenced with the roads and buildings layers in order for them to line up correctly in the final map (all shapefiles in the image are projected in NAD1983 HARN State Plane Florida North). The second image contains a portion of the previous image and extends the map to the extent of the eagle nest, showing the distance of the nest from the main campus (as both images have the same scale). Essential map elements are included for reference and a written description is included to account for the RMS error when georeferencing the UWF images.
The second map is a 3-d map made in arcscene that has been reformatted to fir the 2-d format in arcmap. The basemap images and features have been set to 'float' above the digital elevation model, and an exaggeration of 5 has been set to emphasize the elevation within the map. While buildings and roads are included, specifically, two features are highlighted with different colors to indicate UWF gym and Campus Lane and essential map elements have been added for reference, but there is no spatial reference system, therefore no north arrow or scale bar.

Friday, March 3, 2017

NDVI Index

The following three images display features that were identified given a discrepancy in pixel values for a given layer or layers. The images represent these features in a way that is easily identifiable by the human eye.
I found the feature that causes a pixel spike in layer 4 between values 12 and 18 is a riparian wetland. I inquired the area with the described values and tried several different band combinations, until finally deciding the best color to distinguish the wetlands from the outlying green (NIR) area was a purple that happened to be displayed in the R-3,G-4,B-5 combination.
The second feature has a small spike in layers 1-4 around pixel value 200, and a large spike between pixel values 9 and 11 in layer 5 and layer 6. Upon finding the feature, it was hard to identify and seemed out of place in the image. With further investigation appeared to be an isolated water body which explains the drastic difference in color because it had a different composition than that of the darker (probably salt) water body in the upper left corner of the image. I used several different band combinations to differentiate the lake from the rest of the landscape that involved turning the lake various shades of yellow and purple, but ultimately found that the natural teal blue color was the best way to isolate the feature. The other water was left in the picture for reference, but was left  true color with the rest of the image as it was difficult to find a band combination that isolated the different water features from each other, as well as the rest of the landscape.
The third feature (snow) was identified given the high values in layers 1-3 and the pixel values remained relatively unchanged in layers 5-6. Given the information provided in the feature description and some slight experimentation, I thought the obvious combination choice was R-1,G-5,B-6. Because the wavelengths in layers 5-6 were so similar between bandwidths, I chose to represent them with blue and green to display the surrounding area and vegetation as a dark green and bare extraneous land as blue. The white color allowed the snow to be extremely visible in any layer 1-3, so I chose layer 1 to be represented with red, in order to highlight the snow and distinguish it from the surrounding blue/green area.

Thursday, March 2, 2017

Thermal Imagery & Analysis Interpreting Thermal Infrared Radiation

The image above is two separate maps displaying the same area, the first with a multi-spectral classification scheme and the second with a light-dark color scheme with one layer displaying TIR energy only. I decided to display the distribution of thermal energy in an area where several different classification themes came together (from left to right in the image: forested land integrated with streams/bayous, dense urban area, river, island surrounded by river, other side of the river, and residential/ agricultural land running off the right of the data viewer). The multiple bandwidths in the first map properly classify the image and clarify its features by creating a simple scheme (urban areas=purple, vegetation=green, water=blue, thermal energy=red). With the added bandwidth information, the TIR values can be difficult to make out, which is why the second map displaying only the TIR layer was added for clarification. For instance: given the TIR information on the second map, you can deduct that urban areas are purplish/ pink due to the mixing between the blue/grey urban wavelength and the red TIR wavelength.