Agricultural activities and residential development contribute to widespread increases in the concentrations of nutrients in stream ecosystems, which frequently leads to ecological impairment. Further, changes in nitrogen (N): phosphorus (P) ratio can lead to changes in biological community structure as predicted by resource-ratio theory. Little is known about the influence of land cover on stream N: P ratios. Concentrations of total dissolved nitrogen (TDN), dissolved inorganic nitrogen (nitrate + ammonium + nitrite, (DIN)) total dissolved phosphorus (TDP) and dissolved inorganic phosphorus (phosphate, (DIP)) were measured monthly using standard methods over two years in 15 Georgia Piedmont headwater streams in an area with extensive poultry and cattle production and rapid residential development. Study watersheds were selected along a gradient of land use and organic waste inputs from poultry and cattle operations. Robust non-detects data techniques were used to calculate medians. Robust linear regression techniques were used to evaluate relationships between land use in the entire watershed and in the stream buffer zone (90 m), and median TDN: TDP and DIN: DIP atomic ratios. Number Cruncher Statistical Software was used for all statistical analyses. Land cover metrics were calculated with a geographic information system (GIS) and National Land Cover Data (NLCD) information.
Results/Conclusions
Individual TDN: TDP ratios were highly variable, ranging from 5.3 to 292, whereas two-year site median TDN: TDP ratios were less variable ranging from 14.6 to 220. Individual DIN: DIP ratios were also highly variable ranging from 5.5 to 470, whereas two-year site median DIN: DIP ratios were less variable ranging from 16 to 234. The lower end of both the two-year site median TDN: TDP and DIN: DIP ratios are near the Redfield N: P ratio of 16. N: P ratios of major anthropogenic organic matter sources to the study watersheds are 2.5 for poultry litter and from 5 to 28 for septic tank effluents. The robust linear regression analyses showed that two-year site median TDN: TDP ratios were marginally correlated to watershed forest land cover (negative, R2 = 0.22, p = 0.055) and pasture land cover (positive, R2 = 0.18, p = 0.087). Two-year site median DIN: DIP ratios were more strongly related to buffer open water (positive, R2 = 0.48, p = 0.006). These results suggest significant land cover control of stream nutrient N: P ratios and that relatively simple land cover analysis may prove useful for predicting N: P ratios in other watersheds.