Skip to main content
ARS Home » Plains Area » Temple, Texas » Grassland Soil and Water Research Laboratory » Research » Publications at this Location » Publication #139303


item Harmel, Daren
item King, Kevin

Submitted to: Applied Engineering in Agriculture
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 5/13/2003
Publication Date: 11/20/2003
Citation: Harmel, R.D., King, K.W., Slade, R.M. 2003. Automated storm water sampling on small watersheds. Applied Engineering in Agriculture. 19(6):667-674.

Interpretive Summary: Many water quality projects now require monitoring water quality during storm events. In the past, monitoring focused on low flow sampling to characterize point source pollution (discharged from specific locations such as factories and waste water treatment plants), but now sampling projects must often include storm flow monitoring to characterize nonpoint source pollution (discharged from diffuse sources such as urban areas, farms, and silvicultural operations). When excessive amounts of these pollutants enter water bodies, they degrade aquatic ecosystem health, increase water treatment costs, and diminish recreational and aesthetic value. Automated storm sampling equipment is commercially available, but guidance on equipment operation is limited. Therefore, we wrote this paper to discuss several sampling strategy components and to share some of the lessons we have learned in storm sampling studies. Based on these studies and experience, we recommend setting samplers to trigger sampling at low flow levels using small flow intervals on which to collect samples after the sampler is triggered (for instance sample every 0.04 inches of runoff). Also, to limit numbers samples to a manageable level, composite several samples into each collection bottle. These are general recommendations and may need adjustment based on watershed conditions and monitoring project goals.

Technical Abstract: Typical automated storm sampling involves setting a minimum threshold to initiate sampling and setting a time or flow interval on which to collect samples after the sampler is triggered. However, little guidance is currently available to assist in making these settings for storm sampling strategies in small watersheds. Therefore, appropriate guidance is needed to develop automated sampling strategies that accurately characterize storm water quality and storm loads within project budget, laboratory, and personnel limitations. In this paper, several important sampling strategy components such as: minimum flow thresholds, time- and flow-weighted sampling, and discrete and composite sample collection are explored. These issues are important because they determine the size of storms sampled, the number of samples taken during each storm, and the accuracy of the sampled storm load compared to the true load. In addition, pertinent findings and lessons learned from previous field and modeling studies are reported. Based on these studies and experience, we recommended setting low enable levels and using flow-weighted sampling. When attempting to limit samples to a manageable number, use composite sampling, which tends to introduce less error than raising minimum flow thresholds or increasing sampling intervals. These broad recommendations, however, may need to be adjusted because of the variation in watershed conditions and monitoring goals. Therefore, knowledge of sampling strategy components described is also important. Monitoring program developers informed with some knowledge of watershed characteristics and these reported findings should be able to implement a successful storm water quality monitoring program for small watersheds.