『Abstract
Several stream monitoring programs aim to measure nonpoint source
pollutant loading to the major river systems in the Twin Cities
Metropolitan Area (TCMA) of Minneapolis and St. Paul, Minnesota.
However, due to cost and logistical considerations, only a portion
of the total nonpoint source load can be effectively monitored.
Regression models were developed relating nonpoint source pollutant
yields to landscape characteristics in order to estimate the total
nonpoint source contributions of nutrients and suspended sediments
to the Mississippi, Minnesota, and St. Croix Rivers from the region.
The regression models were generally strong with r-squared values
ranging from 0.57 for total suspended solids yield to 0.90 for
nitrate yield. The model factors included both land cover variables,
such as the percent row crop or the percent urban land as well
as soil variables, such as the percent organic matter or the percent
clay. These results highlight the importance of considering both
land cover and solids when estimating nonpoint source pollutant
loads. Using the fitted regression models, the estimated annual
average total suspended solids, total phosphorus, nitrate, and
total Kjeldahl nitrogen loads for the TCMA are 104,000 metric
t/yr, 354 t/yr, 3,580 t/yr, and 1,760 t/yr, respectively.
Keywords: geographic information systems; landscape; modeling;
nonpoint; nutrients; sediment; regression』
1. Introduction
1.1. Study area description
2. Methods
2.1. Monitored pollutant data
2.2. Landscape data
2.3. Statistical analysis methods
3. Results
3.1. Monitored NPS pollutant yields and loads
3.2. Landscape characteristics
3.3. Principal component analysis
3.4. Regression model results
4. Discussion
4.1. Yields
4.2. Loads
4.3. Model strength
4.4. Model factors
Acknowledgements
References