Sources of Nutrient Input to an
Urbanazing Watershed
Kazuhiro Sonoda
Ph.D. in Environmental Science
Portland State University
Presented November 8, 2002
Abstract
I investigated spatial and temporal relationships among surface and subsurface watershed attributes including regional groundwater, near-stream groundwater, and riparian soil chemistry and stream nutrient concentrations (Phosphorus and Nitrogen) in urbanizing Johnson Creek watershed in Portland, Oregon. Furthermore, I investigated relative contribution of P loading to the stream via various hydrological pathways within a small urbanized sub-watershed using a mass balance approach. My null hypotheses were that: (1) streamwater nutrient concentration would not vary according to watershed attributes, and (2) P loading had no significant surface hydrological pathways due to urbanization within a sub-watershed.
Streamwater and near-stream groundwater samples were collected from 13 locations along Johnson Creek and analyzed for P and N. Soil samples were collected from the same sample locations and analyzed for pH, P, Al and Fe. The mass balance of P loading to the urban stream was developed using daily discharge and P data collected at the beginning of the stream, end of the stream, 2 storm drains, near-stream groundwater and direct precipitation to the stream.
On the basin scale, I found a significant correlation between N and non-urban landuse. P correlated with urban landuse. Within non-urban areas, near-stream groundwater P correlated with streamwater P; however, there were no such correlations within urban landuse areas indicating significant surface inputs of P in addition to groundwater sources. Both Al and Fe were inversely correlated with near-stream groundwater P, suggesting the formation of Al/Fe-P complexes in the non-urban area reducing availability of P. Based on a mass balance analysis, within a small urbanized sub-watershed, storm drains contributed 37% of SRP and 25% of TP during rain.
These findings suggest that elevated levels of P within urbanizing areas of Johnson Creek watershed is most likely controlled by two different mechanisms: riparian soil chemistry and urbanization. Significantly lower Al and Fe contents in soil within urban areas of Johnson Creek likely reduce P retention, hence increasing P availability to stream. Increase in storm runoff input via street and storm drains within urbanized areas of Johnson Creek is likely anthropogenic source of P to the stream. Kazuhiro Sonoda