Evaluation of Time Proportional Sampling Strategies for Estimating Annual Nutrient Fluxes at the Outlets of Coastal Plain Watersheds

Excess nutrient loads have been recognized to be the major cause of serious water quality problems recently encountered in the North Carolina estuaries and coastal waters. There has been a particular concern in coastal watersheds because agricultural and forested lands are located in close proximity to recreational and environmentally sensitive. Best Management Practices have been proposed to reduce nutrient emissions at the field and the watershed scale. BMP implementations have been coupled with nutrient flux monitoring programs for effectiveness assessment purposes. Nutrient fluxes are usually estimated from continuous flow measurements and discontinuous concentration values obtained after laboratory analysis from discrete water samples. Because of the discontinuity on the concentration information, there is an error made in flux estimations compared to the actual fluxes. This error may be, in some cases, within the same range as the expected water quality improvements. Actual improvements may thus very well remain undetected because of unsuited sampling strategies. To our knowledge, there are no existing references to1) estimate those errors and 2) propose sampling guidelines for minimizing those errors. In this paper, we have examined, using reference databases, the possible errors that can be made on the estimation of annual nutrient fluxes at the outlet of the lower coastal plain watersheds, using time proportional sampling. Our results show that significant errors can be made using even weekly sampling. To detect water quality improvements, our results suggest it is desirable to design sampling on at least a 3-day basis for nitrate and total nitrogen, between 1to 2-day basis for TOC and less than 1-day for total phosphorus. However, our results also show that there can be significant differences between watersheds. One should expect the magnitude of errors to decrease with increasing watershed size, as long as there is a significant correlation between flow and concentration. One should also expect annual nutrient fluxes to be underestimated for nutrients or chemicals exhibiting a concentration effect during flow events, the opposite being true for nutrients or chemical exhibiting dilution effects. Time proportional sampling thus seems relatively ill designed for monitoring, with reasonable uncertainties, annual nutrient flux at the outlets of small coastal plain watersheds. In particular, increasing sampling frequencies from monthly to biweekly or even weekly does not decrease significantly possible errors on flux estimations.