Accuracy of flow measurement or "the closeness of agreement between measurement and the value of the measurand" is both a function of a flow meter's precision and ability of the measurer to recognize, avoid, and correct for bias errors.  This process begins during the site investigation and monitor selection stages.  Sites are chosen based on their ability to provide a hydraulic environment which would introduce as little bias error as possible.  This is accomplished by avoiding highly turbulent flows such as those immediately downstream of bends, weirs, and manholes with multiple feeds.

Ideal velocities are preferred and are exhibited at those sites where velocities are high enough to prohibit grit and/or silt deposition, yet not so high as to create irregular surface turbulence (levels).  Once the proper site is agreed upon, it is critical that the diameter of the meter installation site be measured as accurately as possible.  This is a very important part of the pre-installation procedure since this measurement will be used in the final flow calculations and can be the source of a constant bias error if measured incorrectly or assumed from sewer maps and prints.  At this point, it is necessary to prepare the flow meter for installation.

At Drnach Environmental, we emphasize the need for attention to precision during the pre-calibration and installation stages of every flow study.  Prior to installation, each flow meter is bench calibrated in a custom controlled flow calibration flume to guarantee that the calibrated depth and velocity are as accurate as possible.  Following initial depth calibration, submersible pressure transducers are held in the calibration line for a minimum of three minutes to assure that there is no sensor drift (sometimes observed in older sensors) prior to removal for installation.  Upon installation, a field measurement is taken and compared to the actual meter output.  If the readings of the field measurement and actual output correspond, the site is then secured and left to be revisited within the week for service and a second field measurement.  Following the second site visit, the first week's head/velocity data is applied to one of our computerized spreadsheets where it is reviewed for H/V hydrograph relationships, free flow and total flow scatter graph characteristics, and diurnal flow patterns.  H/V hydrographs typically shed light on those readings well outside the norm (illegitimate errors).  These errors are evidenced by spikes and drop-outs of individual raw head and/or velocity readings.

Total and free flow scatter graphs are viewed as each pertains to specific conditions.  Tight and almost linear free flow head/velocity relationships are plotted and the R² valueis determined.  Although we are aware that flow variations in open channel environments are not truly linear, we have observed the tendency during free flow of many sites to exhibit an apparent linear relationship between head and velocity readings prior to that point where increasing friction, pipe capacity limitations, and downstream backwater effects are observed.  For this reason, free flow data is plotted along with the total flow scatter graph.  The total flow scatter graph is useful, as it will indicate not only sensor stability (repeatability), but also hydraulic changes at different heads and velocities (i.e. low flow silting, surcharges, downstream obstacles, backwater effects, and temporary downstream bypasses).  Diurnal flows are plotted in order to identify the norm and provide for the identification of abnormal flows over the course of the study.

The review and analysis of data is dynamic and ongoing following weekly site visits, measurements, and meter downloads.  With the addition of precipitation data to the previously mentioned practices and procedures, we believe that our QA/QC program minimizes bias and precision errors and thus establishes the accuracy and quality of flow measurement our clients have come to expect from us.