How Accurate are My Flow Data? Water Level Sensor Evaluation to Assess Confidence in Compliance Assessments

Date / Time:
Wednesday, Oct 9 11:45am to 12:10pm
Location:
Marriott - San Carlos III
Track / Session:
Monitoring, Science, and Data Management Track / Accurately Quantifying Non-Stormwater Flows
Description/Abstract: 

The San Diego Regional MS4 Permit requires responsible agencies to conduct a wide variety of water quality monitoring. This can include routine storm water characterization monitoring, TMDL compliance monitoring, dry weather flow elimination monitoring, special studies for model development or BMP design, and many other programs. Nearly all monitoring includes collection of reliable flow data. Without flow data, analytical data are minimally useful since there is no information on total constituent mass (i.e., load) or context for the water quality information. In some cases, such as TMDL compliance monitoring, load is required as part of the compliance assessment. In other cases, such as BMP sizing and design, total flow and total load play a critical part in the design and success of the BMP.

In the case of dry weather flow reduction monitoring, reliable flow data are at the core of the compliance assessment since data are compared between years to assess the effectiveness of management measures and track progress towards accomplishing reduction goals. High levels of uncertainty in water level data make it is difficult to determine the associated flow rates.

To assess the precision and accuracy of water level data collected during previous monitoring programs using a variety of water level sensors, the County initiated an equipment evaluation study testing multiple water level sensors that are commonly used monitor flow. The water level sensors in this evaluation were used in conjunction with compound weirs as a control structure, and include: a non-vented pressure transducer (PT), three different brands of vented PTs (one also measures temperature and conductivity), two different brands of bubblers, and an ultrasonic lookdown sensor. Time-lapse cameras were also installed at each test site to serve as a quality assurance (QA) step to help verify any high flow occasions, weir blockages, or other site conditions that may cause erroneous or anomalous water level measurements affecting the resulting flow calculations, but not be explainable without visual confirmation.

The same of water level sensors and cameras were installed at three MS4 outfall sites in the field, and in a laboratory setting with an in-house, custom variable flow test pipe. Data were collected over a minimum of two weeks. Throughout the test period, multiple manual level and flow measurements were collected for calibration and validation. Data sets were evaluated visually for precision (or “noise”), drift, and accuracy. Accuracy was also evaluated by calculating the root mean square error (RMSE) between recorded data and manual measurements.

This presentation summarizes the results of the County’s equipment evaluation. Sharing the results of this evaluation is intended to help other agencies and contractors make informed decisions on their water level monitoring equipment based on real-world implementation. Improved and reliable water level data collection will allow agencies to maintain a high level of confidence in their flow data, and ultimately in their compliance assessments, be it for load calculation, BMP sizing, or dry weather flow reductions.

This presentation addresses the conference theme by showcasing a study that assesses confidence in water level monitoring and associated flow data. The notion of “Why we do what we do” can be answered as improving the environmental and ecological conditions around us. Understanding if improvement is occurring is achieved at a fundamental level through monitoring. Without reliable flow data, dependably assessing impacts from storm water and other water quality issues is quite challenging.

Primary Speaker:
Jeremy Burns, Wood Environment & Infrastructure Solutions, Inc.
Jeremy Burns is a Senior Associate Scientist and Project Manager at Wood E&IS.