Matching Items (2)
Filtering by
- Creators: Fox, Peter
- Status: Published
Description
Safe, readily available, and reliable sources of water are an essential component of any municipality’s infrastructure. Phoenix, Arizona, a southwestern city, has among the highest per capita water use in the United States, making it essential to carefully manage its reservoirs. Generally, municipal water bodies are monitored through field sampling. However, this approach is limited spatially and temporally in addition to being costly. In this study, the application of remotely sensed reflectance data from Landsat 7’s Enhanced Thematic Mapper Plus (ETM+) and Landsat 8’s Operational Land Imager (OLI) along with data generated through field-sampling is used to gain a better understanding of the seasonal development of algal communities and levels of suspended particulates in the three main terminal reservoirs supplying water to the Phoenix metro area: Bartlett Lake, Lake Pleasant, and Saguaro Lake. Algal abundances, particularly the abundance of filamentous cyanobacteria, increased with warmer temperatures in all three reservoirs and reached the highest comparative abundance in Bartlett Lake. Prymnesiophytes (the class of algae to which the toxin-producing golden algae belong) tended to peak between June and August, with one notable peak occurring in Saguaro Lake in August 2017 during which time a fish-kill was observed. In the cooler months algal abundance was comparatively lower in all three lakes, with a more even distribution of abundance across algae classes. In-situ data from March 2017 to March 2018 were compared with algal communities sampled approximately ten years ago in each reservoir to understand any possible long-term changes. The findings show that the algal communities in the reservoirs are relatively stable, particularly those of the filamentous cyanobacteria, chlorophytes, and prymnesiophytes with some notable exceptions, such as the abundance of diatoms, which increased in Bartlett Lake and Lake Pleasant. When in-situ data were compared with Landsat-derived reflectance data, two-band combinations were found to be the best-estimators of chlorophyll-a concentration (as a proxy for algal biomass) and total suspended sediment concentration. The ratio of the reflectance value of the red band and the blue band produced reasonable estimates for the in-situ parameters in Bartlett Lake. The ratio of the reflectance value of the green band and the blue band produced reasonable estimates for the in-situ parameters in Saguaro Lake. However, even the best performing two-band algorithm did not produce any significant correlation between reflectance and in-situ data in Lake Pleasant. Overall, remotely-sensed observations can significantly improve our understanding of the water quality as measured by algae abundance and particulate loading in Arizona Reservoirs, especially when applied over long timescales.
ContributorsRussell, Jazmine Barkley (Author) / Neuer, Susanne (Thesis advisor) / Fox, Peter (Committee member) / Myint, Soe (Committee member) / Arizona State University (Publisher)
Created2018
Description
C.C. Cragin Reservoir’s location in the Coconino National Forest, Arizona makes it prone to wild fire. This study focused on the potential impacts of such a wild fire on the reservoir’s annual thermal stratification cycle impacts and water quality. The annual thermal stratification cycle impacted the reservoir’s water quality by increasing hypolimnion concentrations of magnesium, iron, turbidity, and specific ultraviolet absorbance (SUVA) values, as well as resulting in the hypolimnion having decreased dissolved oxygen concentrations during stratified months. The scarification process did not affect the dissolved organic carbon (DOC) concentrations in the reservoir or the total/dissolved nitrogen and phosphorous concentrations. Some general water quality trends that emerged were that phosphorous was the limiting nutrient, secchi disk depth and chlorophyll a concentration are inversely related, and no metals were found to be in concentrations that would violate an EPA drinking water maximum contaminant level (MCL). A carbon mass model was developed and parameterized using DOC measurements, and then using historic reservoir storage and weather data, the model simulated DOC concentrations in the reservoir following four hypothetical wild fire events. The model simulated varying initial reservoir storage volumes, initial flush volumes, and flush DOC concentrations, resulting in reservoir DOC concentrations varying from 17.41 mg/L to 8.82 mg/L.
ContributorsFlatebo, Theodore (Author) / Westerhoff, Paul K (Thesis advisor) / Fox, Peter (Committee member) / Perreault, Francois (Committee member) / Arizona State University (Publisher)
Created2018