R.M. Slade

Cumulative uncertainty in measured streamflow and water quality data for small watersheds

The scientific community has not established an adequate understanding of the uncertainty inherent in measured water quality data, which is introduced by four procedural categories: streamflow measurement, sample collection, sample preservation/storage, and laboratory analysis. Although previous research has produced valuable information on relative differences in procedures within these categories, little information is available that compares the procedural categories or presents the cumulative uncertainty in resulting water quality data. As a result, quality control emphasis is often misdirected, and data uncertainty is typically either ignored or accounted for with an arbitrary margin of safety. Faced with the need for scientifically defensible estimates of data uncertainty to support water resource management, the objectives of this research were to: (1) compile selected published information on uncertainty related to measured streamflow and water quality data for small watersheds, (2) use a root mean square error propagation method to compare the uncertainty introduced by each procedural category, and (3) use the error propagation method to determine the cumulative probable uncertainty in measured streamflow, sediment, and nutrient data. Best case, typical, and worst case “data quality” scenarios were examined. Averaged across all constituents, the calculated cumulative probable uncertainty (±%) contributed under typical scenarios ranged from 6% to 19% for streamflow measurement, from 4% to 48% for sample collection, from 2% to 16% for sample preservation/storage, and from 5% to 21% for laboratory analysis. Under typical conditions, errors in storm loads ranged from 8% to 104% for dissolved nutrients, from 8% to 110% for total N and P, and from 7% to 53% for TSS. Results indicated that uncertainty can increase substantially under poor measurement conditions and limited quality control effort. This research provides introductory scientific estimates of uncertainty in measured water quality data. The results and procedures presented should also assist modelers in quantifying the “quality” of calibration and evaluation data sets, determining model accuracy goals, and evaluating model performance.

Estimating storm discharge and water quality data uncertainty: A software tool for monitoring and modeling applications

Uncertainty estimates corresponding to measured hydrologic and water quality data can contribute to improved monitoring design, decision-making, model application, and regulatory formulation. With these benefits in mind, the Data Uncertainty Estimation Tool for Hydrology and Water Quality (DUET-H/WQ) was developed from an existing uncertainty estimation framework for small watershed discharge, sediment, and N and P data. Both the software and its framework-basis utilize the root mean square error propagation methodology to provide uncertainty estimates instead of more rigorous approaches requiring detailed statistical information, which is rarely available. DUET-H/WQ lists published uncertainty information for data collection procedures to assist the user in assigning appropriate data-specific uncertainty estimates and then calculates the uncertainty for individual discharge, concentration, and load values. Results of DUET-H/WQ application in several studies indicated that substantial uncertainty can be contributed by each procedural category (discharge measurement, sample collection, sample preservation/storage, laboratory analysis, and data processing and management). For storm loads, the uncertainty was typically least for discharge (+/-7-23%), greater for sediment (+/-16-27%) and dissolved N and P (+/-14-31%) loads, and greater yet for total N and P (+/-18-36%). When these uncertainty estimates for individual values were aggregated within study periods (i.e. total discharge, average concentration, and total load), uncertainties followed the same pattern (Q < TSS < dissolved N and P < total N and P). This rigorous demonstration of uncertainty in discharge and water quality data illustrates the importance of uncertainty analysis and the need for appropriate tools. It is our hope that DUET-H/WQ contributes to making uncertainty estimation a routine data collection and reporting procedure and thus enhances environmental monitoring, modeling, and decision-making. Hydrologic and water quality data are too important for scientists to continue to ignore the inherent uncertainty.

Automated storm water sampling on small watersheds

Few guidelines are currently available to assist in designing appropriate automated storm water sampling strategies for small watersheds. Therefore, guidance is needed to develop strategies that achieve an appropriate balance between accurate characterization of storm water quality and loads and limitations of budget, equipment, and personnel. In this article, we explore the important sampling strategy components (minimum flow threshold, sampling interval, and discrete versus composite sampling) and project -specific considerations (sampling goal, sampling and analysis resources, and watershed characteristics) based on personal experiences and pertinent field and analytical studies. These components and considerations are important in achieving the balance between sampling goals and limitations because they determine how and when samples are taken and the potential sampling error. Several general recommendations are made, including: setting low minimum flow thresholds, using flow-interval or variable time-interval sampling, and using composite sampling to limit the number of samples collected. Guidelines are presented to aid in selection of an appropriate sampling strategy based on user’s project -specific considerations. Our experiences suggest these recommendations should allow implementation of a successful sampling strategy for most small watershed sampling projects with common sampling goals.