The complicated water rights system in California has led to a hierarchy for access to water, particularly affecting farmers in the Central Valley. Senior or junior water right status depends on the date users first claimed their water use or applied for a water diversion permit. The state’s Electronic Water Rights Information Management System (eWRIMS) online database contains related information, including water right type and points of diversion. During periods of drought, water allocations are adjusted and junior water rights holders are restricted more than seniors. Variable water supplies likely affect cropland health from year to year and influence farmers’ decisions to grow certain crops, but could also have other indirect impacts such as land subsidence due to increased groundwater pumping. This research project examines satellite imagery, vegetation indices, cropland classifications, and subsidence data to determine if cropland and land surface dynamics differ between the two water rights groups in the Sacramento and San Joaquin climate divisions. These data layers were sampled at points within buffered regions of the points of diversion. Statistical t-tests showed a significant difference (p < 0.05) between seniors and juniors for characteristics including cropland diversity and maximum NDVI (normalized difference vegetation index) over the past eight to fifteen years based on data availability. Land subsidence was limited to a small area in southern Central Valley, and only showed a significant difference in one of the two hot spots. Future analysis will replace the assumptive buffers with actual places of water use when data becomes available.

The goal of this project is to create a Python tool to automate a common workflow of image analysis for preliminary assessment of land cover and vegetation mapping through supervised classifications in Pima County, Arizona. This tool is primarily designed to aid in assessing the distribution of resources, natural and non-natural, at multiple scales. Execution at multiple scale levels is necessary to illustrate that Remote Sensing project design is extremely important in successful classification of imagery. This project uses Landsat-5 TM imagery to conduct an image classification of twelve unique vegetation types in Saguaro National Park West - Tucson Mountain District. Landsat-5 TM imagery is also utilized at a smaller scale to conduct a land use classification using the same automated workflow. Results of the image classification process are evaluated by conducting accuracy assessments on the classified rasters. The Saguaro National Park West – Tucson Mountain District classification had a much lower overall accuracy than the Pima County land use classification due to how specific the classification was and the lack of spectral resolution present in the Landsat-5 TM imagery. The classified rasters are visualized in an interactive web application built with Web AppBuilder by consuming geoprocessing services allowing the processing to be completed in the client. This tool should be used as preliminary analysis for the area of interest, and is an effective classification workflow if paired with remotely sensed imagery that supports the project goals of a particular study. 

Madrean desert grasslands are subject to numerous environmental pressures and, without conservation efforts, biological diversity will fade quietly.  Bearded Chinchweed (Pectis imberbis) is a Sonoran Desert grassland plant that faces many threats that include grazing, off road vehicles, and road expansion.  This study creates a habitat suitability model for Pectis imberbis that improves conservation and reclamation efforts in Southern Arizona.  This study creates a habitat suitability model for Pectis imberbis using maximum entropy modeling software, SAS statistical analysis, and ArcGIS. MaxEnt takes environmental layers such as climate, soils, digital elevation models, and vegetation rasters as well as georeferenced point locations and creates the range of a given species. MaxEnt is a powerful modeling platform that can create relationships between as few as 10 species point locations, model relationships between environmental rasters and point locations, create test statistics.  Results show that Pectis imberbis conservation efforts should be concentrated around the U.S. Mexico border near Nogales, Arizona and on the southern edge of the Atascosa Mountains.  In addition, temperature annual range, mean temperature of the coldest quarter, and warm season precipitation are factors that drive healthy Pectis imberbis habitat.  With complete soils data and continuous vegetation data, a future study would be able to create a model that even more accurately predicts Pectis imberbis range.  With more studies, biological diversity conservation can become more efficient with interactive cartographic products that can pinpoint exact areas that need conservation.  

Soil erosion is an important issue worldwide, especially in arid environments.  Erosion can be accelerated through several factors, including wildfires.  The 2011 Wallow Fire in Arizona and New Mexico burned approximately 535,000 acres making it the largest wildfire in Arizona history.  In this study, I evaluated soil erosion rates before and after the Wallow Fire to study the effects of the fire on soil loss.  To explore this problem, I used the Revised Universal Soil Loss Equation (RUSLE).  The vegetation cover factor used by the equation was developed based on the Normalized Difference Vegetation Index (NDVI) from Landsat 5 images from before and after the fire.  Results showed a slight decrease in soil loss for the entire area of the Wallow Fire.  However, the amount of change in soil loss is dependent on in burn severity of the fire.  Higher burn severity areas showed high increases in soil losses, while area of low burn severity showed a decrease or negligible change in soil loss.  Results would be useful to post-fire soil erosion mitigation efforts.

Mental maps are used to elicit perceptions of the environment from members of the public. These maps can be used to interpret perceptions of pavement conditions in the form of ratings for selected street segments. Pavements are a part of the built environment and function to assist in the conveyance of the urban population. The City of Casa Grande is located in central Arizona between the cities of Phoenix and Tucson. City governments must consider multiple factors when determining what roads require maintenance above others. In addition to budget and the force and degradation factors, governments must assume that there will usually be public opinion on what streets take precedence over others. This study determines how perceptions of pavement conditions in 2015 from different stakeholder groups relate to the actual pavement conditions used by the City of Casa Grande Government. Statistical analyses between the actual pavement condition values and the perceived values were used to evaluate the ability of the stakeholder groups’ accuracy to predict the existing pavement conditions. In addition, a cumulative opinion surface was created to compare areas with low PCI values and areas that the public thought had low PCI values. Results collected from this study indicate that the input of the public was not able to completely or accurately predict the existing conditions of the roads. However, public opinion and how the opinions differ amongst stakeholder groups is a factor that cannot be ignored when determining the maintenance schedule for the road network.

This project examines the history of Tucson, Arizona from 1860 to 2013 using an online web atlas. Comprehensive street-level maps spaced roughly a decade apart show the growth of Tucson from a small agricultural settlement to a moderately sized town and finally to a sprawling modern metropolis. The maps are modeled after online road map resources such as Google Maps, and include features such as municipal boundaries, parks, schools, hospitals, civic buildings, airports, railways, and notable historic and cultural sites using a standard symbology across all maps. The maps are hosted on an ArcGIS Online server as tiled raster imagery with zoom and pan capabilities. Data are constructed by taking modern GIS data acquired from governmental sources and altering them through intensive comparison of scanned historic road maps, topo maps, and surveys. This web atlas provides the first comprehensive digital resource for viewing the growth of a city in the Southwestern United States.

The Wallow Fire was a mega-fire that occurred in Arizona and New Mexico in 2011. Satellite data is available from Landsat for the 19 years before and the year after the fire occurred. 1992 to 2011 are the years before the fire occurred, and 2012 is the year after the fire occurred. This study analyzes the Normalized Burn Ratio (NBR) for those 20 years and compares the changes in vigor to the impacts of climate change. By performing a measurement of NBR for different vegetation types, a supervised classification was created. The NBR was then graphed for each vegetation type. The results show the vegetation class most effected prior to the Wallow fire was Spruce-Alpine Forest. Overall from 1992 to 2012, the NBR has showed a negative trend for all vegetation types and that climate change has impacted all the vegetation types in the area prior to the Wallow Fire occurring.

Desperately trying to locate a hard to find shutoff valve during a dark and stormy night is a City Plumber’s nightmare.  The plumber’s difficulties would be lessened using a new tool: a GIS application on his smart phone.   This study explores whether there is reasonable return on investment if the City of Tucson Facilities and Communication Maintenance Division utilizes GIS for external infrastructure with a Collector for ArcGIS application, and whether it would help field technicians and improve efficiencies.  It is proposed that rather than creating a new GIS database of all infrastructure, it is more economically feasible to map a subset, such as plumbing exterior infrastructure.  The upfront expenses of creating a partial GIS database will be offset by less time spent locating assets, such as cleanout and shutoff valves.  The data for this analysis was collected using the Collector for ArcGIS application.  This study documents the time difference in locating shutoffs and cleanouts with and without GIS assistance.  The results show an overall average 40 percent reduction of time spent finding cleanout and shutoff valves with GIS assistance versus without GIS assistance.  This time savings is calculated as a percentage of labor hours booked into the Division’s work order system and is presented as a future efficiency solution for the Division.