Document Type

Thesis - Open Access

Award Date

1975

Degree Name

Master of Science (MS)

Department

Biology

Abstract

The possibility of predicting prairie lake water quality using remote-sensing data during the 1974 open water season was investigated. Water quality samples were collected to coincide with satellite (LANDSAT-1) overpasses ad low-altitude aircraft flights. Twelve different remote-sensing estimates were taken from LANDSAT-1 imagery, low-altitude aerial photographs, ground-based radiometer, and ground-level photographs. These were correlated with prairie lake water transparency (secchi depth) and algae abundance (chlorophyll ‘a’ and total cell concentrations). For predictive purposes multiple regression analysis were performed using the three water quality parameters as independent variables. The correlations between the three water quality parameters and physical factors (rainfall, solar energy, wind velocity, water temperature, and water depth) were also investigated using multiple regression analyses. All twelve remote-sensing estimates gave significant correlations with the water quality parameters measured. Film densities of ground photos taken directly over the water significantly correlated with the selected water parameters. Infrared films were most capable of detecting variations of chlorophyll ‘a’ and total algae while black and white film with a green filter best predicted water transparency for the aircraft data. Of the four spectral regions (bands) received by the LANDSTA-1 sensors, band 4 (0.5 to 0.6 um) best predicted chlorophyll ‘a’ concentrations and Band 5 (0.6 to 0.7 um) best predicted both total algae and water transparency. All water quality parameters measured were significantly correlated with each other, with nutrient levels, and with some of the physical factors measured. Decreased water depth particularly was associated with higher nutrient levels, increased total algae and chlorophyll ‘a’, and decreased water transparency. Nitrate appeared to be the principal nutrient positively correlated with algal abundance. Extensive ground-level photographic or radiometric measurements are impractical in many operational programs since the time involved on the ground would justify collection of actual water samples. Low altitude aerial photographs offer finer resolution of small areas but are more expensive to obtain than satellite data. LANDSAT imagery offers practical potential for measuring prairie lake water quality. It provides frequent and total coverage of a region at an inexpensive cost.

Library of Congress Subject Headings

Water quality

Remote sensing

Format

application/pdf

Number of Pages

56

Publisher

South Dakota State University

Share

COinS