Stormwater Monitoring

The MWMO monitors water quantity and quality at several stormwater outfalls into the Mississippi River, as well as some upstream stormwater pipe sites. Monitoring stormwater is necessary to help us understand the effect that our highly urbanized watershed has on the amount and quality of the water that runs off the landscape and into the Mississippi River.

The MWMO watershed can be broken up into pipesheds, areas of the urban landscape that drain to one stormwater outfall. Land use in each pipeshed affects water quality and the amount of impervious surface and potential pollution differs between industrial, residential, and commercial land uses. The locations that the MWMO monitors were chosen, in part, because they are the most extensive pipesheds within the watershed, they are accessible for equipment installation. Some sites were selected to provide data for specific research or modeling projects.

Five of the MWMO’s monitored locations are long-term sites and collect continuous water level and velocity measurements as well as water quality samples during both storm and baseflow (dry weather) conditions. Stormwater monitoring equipment has been installed on a shorter-term basis at other stormwater pipe locations.


Water Level and Velocity Monitoring

Typical area/velocity sensor (black rectangle) and cable and tubing configuration in a monitored stormwater tunnel.
Typical area/velocity sensor (black rectangle) and cable and tubing configuration in a monitored stormwater tunnel.

Water level in a stormwater pipe is very different from water level in the Mississippi River. Stormwater pipes respond quickly to rainfall, so water levels may rise many feet within a few minutes, depending on the size and intensity of the storm event. Some stormwater pipes only contain water during precipitation events, while others have baseflow throughout the year. Most stormwater pipes monitored by the MWMO have varying levels of baseflow year round. Baseflow in stormwater pipes is typically from groundwater or permitted discharges into the stormwater system.

Five long-term stormwater monitoring sites are equipped with area/velocity sensors installed within the pipe as well as automated water quality samplers. The area/velocity sensor is connected to the automated sampler via a cable. The sensor measures water level and velocity in the pipe and these measurements are recorded every 5 minutes in the sampler where the level and velocity measurements, along with information about the pipe’s dimensions, are used to compute the discharge or flow rate of water through the pipe. A typical automated stormwater site setup is shown below.

In 2008, the MWMO, in collaboration with scientists at the University of Minnesota’s St. Anthony Falls Laboratory, designed and deployed a continuous automated water monitoring network at the long-term stormwater monitoring sites. The network was built to enable the remote display of the data in real time and allow for the control of the automated samplers from a remote location. The system uses a network of data-loggers and radios to collect and transmit data from stormwater sites to a web service that creates graphs and tables describing current and past conditions. With this network, automated samplers can be controlled from afar and their sampling programs changed based on site conditions and weather forecasts. Details about the specific hardware and software used in the network are included in the 2013 MWMO Annual Monitoring Report (PDF).

The MWMO has recently installed area/velocity sensors at various upstream locations within pipesheds. Most of these sites currently collect water quantity data only and are used to provide a more detailed look at runoff from smaller subwatersheds than the long-term monitoring sites mentioned above. These data have contributed to the development of stormwater models for pipesheds within the MWMO. Such models are used to predict areas of flooding and infrastructure requirements during a given storm event. The data are also useful in calibrating stormwater quality models which are used to inform best management practices and pinpoint locations within the watershed where projects can have the greatest impact on improving water quality.

Sample Collection, Handling, and Preservation

Example of a typical MWMO stormwater monitoring cabinet setup, including automated sampling and continuous automated monitoring equipment.
Example of a typical MWMO stormwater monitoring cabinet setup, including automated sampling and continuous automated monitoring equipment.

To collect stormwater quality data, grab and composite samples are collected from the five long-term stormwater sites in the MWMO watershed. For the majority of parameters, samples are collected in laboratory-cleansed (non-sterile) plastic bottles. Samples are either collected directly into the bottle as grab samples or with automated samplers as described below.

The automatic samplers house 24, 1-liter plastic bottles for sample collection. Samplers are programmed so that once the water level in the pipe reaches a certain value above baseflow, the sampler is enabled to start sampling. Once enabled, the sampler rinses the sample tubing before drawing the sample into the containers. Samples are collected on a flow-paced basis, meaning a bottle is filled each time a specified amount of water has flowed past the sensor. Once collected, the bottles are composited by a monitoring specialist by pouring an equal amount of water from each sampler bottle into a bottle for laboratory submittal.

Stormwater composite samples are collected for a minimum of three precipitation events per month, as long as that many events occur. If baseflow conditions are present, grab samples are collected twice per month, year-round.

Sampling Quality Control

Blank samples of deionized (DI) water are submitted to laboratories quarterly to verify that sample containers are clean and samples are not contaminated during travel. In addition, 10 percent of samples are collected in duplicate to verify that sampling and laboratory procedures do not jeopardize the data. Equipment blanks are also collected on a quarterly basis where DI water is pumped through the automated sampler tubing and submitted to ensure that equipment conditions are not affecting sample results.

Laboratory Analysis

Samples are analyzed at the Metropolitan Council Environmental Services (MCES) laboratory. The laboratory follows strict protocols for quality assurance and quality control. Information regarding laboratory protocol is available from MWMO staff. Refer to Table A.1 in Appendix A in the annual monitoring reports for a list of sample parameters, the laboratories used for analysis, the analysis methods, and information regarding certification.

Data Analysis

The following data cleaning techniques are used to ensure quality data:

  • Suspect data are flagged and verified with the laboratory
  • Statistical regression techniques are used to interpolate automated flow data that are missing due to equipment problems

Stormwater Water Quality Monitoring Results

The Minnesota Pollution Control Agency (MPCA) does not have water quality criteria for stormwater drainage systems; therefore, data are not compared with standards. The MWMO monitors stormwater to characterize surface runoff in the watershed and determine land contributions to water quality in the Mississippi River. Samples are collected for bacteria, nutrients, sediment, inorganics, organics and metals analyses.

The MWMO does not draw conclusions or make assumptions based on this data until several years of accurate flow-weighted composite data are available.The MWMO has begun calculating annual pollutant loads at stormwater monitoring sites where several years of data do exist. Water quality data are submitted to the MPCA’s water quality database (EQuIS) and are available upon request from MWMO monitoring staff. Stormwater bacteria data are discussed in the Bacteria Monitoring section of this website.