Date of Award

1985

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Abstract

The objective of this investigation was the development of quantitative criteria based on the estimation of maximum levels of oxygen demand associated with specific levels of nutrients in algal dominated streams. 1582 intensive survey records were obtained from the Louisiana Department of Environmental Quality (LA DEQ). A sub-set of 225 records was defined by a minimum chlorophyll a of 10.0 ug/l and influential observation analysis to restrict the data to streams where algal processes were likely to be dominant. Nitrogen, as measured by total Kjeldahl nitrogen (TKN) was identified as the principle nutrient of concern in the prediction of algal associated oxygen demand in the streams represented in the data sets. A linear model describes the relationship between TKN and BOD(,20) in the region below 6.0 mg/l TKN. Variability in observed data about the model line required that the model be expressed in a probabilistic format. The probabilistic model relates the risk of exceeding any BOD standard for any given TKN concentration up to 6.0 mg/l. In order to examine the generality of the developed nutrient criteria model, an independent data set of 7227 observations from 36 states was obtained from the STORET database for comparison purposes. A similar process as for the LA DEQ data was used to define a comparison data sub-set of 199 observations where algal processes were likely to be dominant. Of the 5 states well represented in the STORET subset, Ohio and a "No State Reported" group showed no significant difference from the LA DEQ model, while Delaware, Pennsylvania and Minnesota showed significantly different unit BOD to TKN values in the range of 4.7 to 8.4. There was insufficient data to identify sources of regional differences. The developed nutrient criteria model represents ubiquitous processes, minimizing reliance on site specific parameter estimations, and maximizing the value of historical databases. All evidence suggests that the model is general and will not change substantially as additional verification data becomes available.

Pages

139

DOI

10.31390/gradschool_disstheses.4141

Share

COinS