Accurate and representative precipitation and stormwater-flow data are crucial for use of highway- or urban-runoff study results, either individually or in a regional or national synthesis of stormwater-runoff data. Equally important is information on the level of accuracy and representativeness of this precipitation and stormwaterflow data. Accurate and representative measurements of precipitation and stormwater flow, however, are difficult to obtain because of the rapidly changing spatial and temporal distribution of precipitation and flows during a storm. Many hydrologic and hydraulic factors must be considered in performing the following: selecting sites for measuring precipitation and stormwater flow that will provide data that adequately meet the objectives and goals of the study, determining frequencies and durations of data collection to fully characterize the storm and the rapidly changing stormwater flows, and selecting methods that will yield accurate data over the full range of both rainfall intensities and stormwater flows.
To ensure that the accuracy and representativeness of precipitation and stormwater-flow data can be evaluated, decisions as to (1) where in the drainage system precipitation and stormwater flows are measured, (2) how frequently precipitation and stormwater flows are measured, (3) what methods are used to measure precipitation and stormwater flows, and (4) on what basis are these decisions made, must all be documented and communicated in an accessible format, such as a project description report, a data report or an appendix to a technical report, and (or) archived in a State or national records center.
A quality assurance/quality control program must be established to ensure that this information is documented and reported, and that decisions made in the design phase of a study are continually reviewed, internally and externally, throughout the study. Without the supporting data needed to evaluate the accuracy and representativeness of the precipitation and stormwater-flow measurements, the data collected and interpretations made may have little meaning.