Today’s trends in stormwater monitoring and sampling are focusing on research as well as how to better manage stormwater onsite, industry observers say. Meeting National Pollutant Discharge Elimination System (NPDES) permit requirements continues to be a focus of many stormwater monitoring efforts. Instrument manufacturers are rolling out integrated systems designed to help end users achieve their many goals.
Monitoring in Portland
In Portland, OR, the Bureau of Environmental Services engages in many stormwater monitoring programs, including those for industrial customers that discharge to the treatment plant, as well as surface water monitoring for compliance with NPDES permits.
“We used to do quite a bit of performance evaluations of stormwater BMPs, but we moved away from that recently and are now doing more investigations,” says Matt Sullivan, environmental specialist.
One such investigation centers on a Superfund site created 10 years ago, where the Bureau of Environmental Services is doing source tracing.
“We also do biological monitoring of the streams and water bodies throughout the city of Portland,” adds Sullivan.
The overall focus of the stormwater monitoring is getting inputs for a contamination loading evaluation model for the Columbia Slough, which is one of the more contaminated water bodies in the city of Portland, he says. The Columbia Slough is 12 miles long and a very slow-moving water body.
“Water takes up to a week to exit from the system,” says Sullivan. “Even though we’re pretty far inland, it’s still highly affected. Because it’s a slow-moving water body, a lot of sediment stays within the system.”
Portland has used In-Situ’s Troll 9500 equipment to collect data at a few different points within the Columbia Slough. “Over the course of time, we collected enough data to be able to get a good model of the water quality in the slough,” says Sullivan. “We have a good grasp of the whole system, and now we’re focusing our attention on individual stormwater outfalls that the city owns.”
Bureau staff has seen contamination in the sediment.
“We’re trying to see if the current stormwater is clean enough that the associated sediment will cover over and provide natural attenuation of the contaminated sediment and provide a layer of cleaner sediment on top of it, or if it’s still actively contributing contaminants of concern,” says Sullivan.
Portland does continuous monitoring. “We try to capture over the course of an entire storm or as much of a storm as we can,” he says. The staff visits the site to retrieve the data.
Additionally, the bureau has a sediment trap installed inside a stormwater line to continuously capture sediment over an entire wet season. It is hooked up to flow meters; positive flows trigger the trap to start collecting samples.
The city also has an in-house lab. “At the conclusion of a storm event, we’ll go out there, retrieve the data off the flow meter, pull the samples, and then submit them to our laboratory,” says Sullivan.
PCBs are one of the main constituents for which the Bureau staff is looking; metals are another.
“We’ve been having trouble achieving detection limits,” says Sullivan of the PCBs. “The detection limits are so low that we’ve had to be innovative in our sample collection to be able to achieve those detection limits. We’ve done modified approaches to look at really low levels to achieve that.”
Sullivan says he finds the low detection limits the most interesting part of the stormwater monitoring project.
“We did a lot of high-volume sampling as well, where we would pump a large amount of water and filter it to retain the solid components in order to achieve the low detection limits, and then with that calculate the actual concentration in the water,” he says. “PCB levels are too low to be seen in a water sample, but if we can extract the sediment from the stormwater with the lab analytical methods, we can detect in the sediment that’s left behind and then calculate the concentration in the stormwater.”
The program has been in effect for two years.
“We did a lot of sediment coring to look at the sedimentation rate of the receiving water of the Columbia Slough and went through several approaches until we ended up with some equipment that would work,” says Sullivan. “The sediment is really fine material, and some of the coring approaches we had tried to use didn’t work initially, so we had to refine those techniques.”
The Troll 9500 is one of many stormwater sampling and monitoring tools offered by In-Situ. The portable water quality instrument is a “set and forget” system used for groundwater and surface water monitoring and houses up to nine water quality sensors, internal power, and optional data logging capabilities.
Jon Firooz, vice president with In-Situ, says the top issues his company hears about from its customers is a lack of time and cost and budget restrictions.
“Additionally, they also have to deal with human resources, so there’s a lot of turnover that takes place,” he says. “Every time you have a new person on staff, you have to deal with training and getting them up to speed, making sure they’re doing things properly.”
He says users also are concerned about regulatory requirements. “They also require flexibility because they never know exactly what monitoring is going to be needed at any given site, so they want flexibility in terms of the types of equipment they use,” points out Firooz.
In-Situ approaches those problems from a perspective of both technology and service. The company recently released its smarTroll Multiparameter (MP) Handheld, combining water-quality sensors with smartphone mobility. The intuitive In-Situ application runs on an iPhone, iPod Touch, or iPad device.
The smarTroll MP Handheld is connected to a battery pack, the In-Situ application is launched, and immediate results are produced. The In-Situ application guides its user through spot checks, calibrations, and data management. The device measures up to 14 parameters. Chemical parameters include dissolved oxygen (DO), pH, oxygen-reduction potential (ORP), conductivity (actual or specific), salinity, total dissolved solids, resistivity, and density. Physical parameters include air and water temperature, barometric pressure, water level, and water pressure.
“It can be used for spot-checking or profiling,” says Firooz, adding it requires no training. “We’ve also designed it to take advantage of a lot of features that are native to smartphone platforms. For instance, you can tag GPS locations for a site to make it easy to come back and confirm that you are in the right location. One of the best features is the ability to e-mail this instantly back to colleagues or others on your team.”
pollutants of concern.
All work can be done from the field. Links are provided to an online store for those who need to order a new calibration solution.
End users also find the company’s Rugged Dissolved Oxygen Pro Probe helpful, Firooz says, adding that it is approved by EPA for Clean Water Act monitoring requirements.
When the probe initiates a reading, an LED emits blue light, exciting lumiphore molecules in the sensing element. Excited lumiphore molecules emit red light, which is detected by a photodiode. Oxygen molecules quench the excited lumiphore molecules and prevent the emission of red light through the “dynamic luminescence quenching” process. Determination of DO concentration by luminescence quenching has a linear response over a range of concentrations.
The technology was designed to be more durable and maintenance free than some of the legacy electro-chemical technologies, helping “significantly” in handling some of the abrasion and the harshness of the environments with which those who monitor stormwater deal on a regular basis. EPA approval allows use without the need for special regulatory approval or alternative test procedure clearance, Firooz says.
The desire for flexibility is addressed in many of the company’s products, including the relatively new multiparameter Aqua Troll 400, which offers support for standard open protocols so that end users can connect the instrument with different loggers or automated sampling systems on the market.
“We have a number of customers who will connect our instruments to a Campbell Scientific data logger or a Teledyne Isco automated sampling system,” says Firooz, adding that users appreciate the flexibility.
The need for real-time data access is another trend noted by In-Situ. In response, the company has provided telemetry solutions that enable customers to use a cellular network or a satellite telemetry system in extremely remote areas. The solutions are designed for customers who want to get data any time, eliminating the need for labor costs associated with grab samples and other manual data retrieval techniques.
“Budgets are being squeezed; the biggest costs tend to be labor,” says Firooz. “If we can reduce the number of trips they have to take, that can give end users a lot more time at staying within their budget constraints.”
Comparing Farming Techniques
At The Ohio State University, researchers are engaged in a multidisciplinary project that combines physical science data collection to study water quality and social science survey work. The study focuses on the differences in farming practices between Amish and non-Amish farmers.
“We’re doing a comparison of the water quality to support the work that the social scientists are doing at the sites,” says Deana Hudgins, a research associate. “We’re looking at farming practices and how decisions are made, as well as water quality, because these areas are heavily farmed.”
The comparisons are within the East Branch–South Fork Sugar Creek watershed in north central Ohio, one of the most polluted watersheds in the state from nonpoint-source pollution.
East Branch Sugar Creek is the non-Amish site; the South Fork Sugar Creek is the Amish site. Both are part of the Sugar Creek watershed, which is part of the Muskingum River watershed.
“In the Amish watershed, their practices are different. They’re not using mechanized systems. They often are farming smaller areas. They do a lot of seeding by hand,” says Hudgins. “The comparison was driven much more by the social science aspect rather than the water sampling aspect.”
Amish farmers are doing what many mainstream, non-Amish farms are doing, she adds.
“They are using similar crops. You see GMO [genetically modified crops] on occasion; it depends on whether it’s approved in their church district. You see fertilizers, pesticides. There is manure application in both watersheds. There is a large dairy in the non-Amish watershed that is downstream of where the sample site is located. Some farmers in both watersheds are doing organic farming. It’s a mixture in the area,” says Hudgins.
Most watersheds in the area tend to look very similar, she says.
“They are very flat areas with very eroded stream banks, a lot of straightening of streams,” says Hudgins.
The instrumentation was installed in late spring 2013. Researchers are using two SonTek-IQ Plus flow meters with standard mounts and Teledyne Isco 24-bottle samplers.
The SonTek-IQ is designed with a custom flow algorithm. Four velocity beams profile water velocity in 3D vertically and horizontally. A built-in pressure sensor and vertical acoustic beam operate in tandem to measure water level.
The Isco bottle samplers collect 1-liter samples every 24 hours. Researchers take those samples back to a lab for analyses for phosphorous, nitrate, ammonia, total nitrogen, and total phosphorous.
The research team is doing continuous flow monitoring as well. “The Isco is gathering approximately 250 milliliters every eight hours. We collect the samples every two and a half to three weeks and do the analyses on those. They are acid-fixed samples; the unit is not refrigerated. We have ongoing remote access to our flow data.”
The research team also has alarms set up for the rare flooding situation.
“There are very deep channels from drawdown in erosion, but we are able to set up alarm systems because it’s roughly an hour away from where any of us lives. If we did have a flooding situation, we’d need a lot of notice to be able to pull equipment in a timely manner,” says Hudgins. “We have it set up so we can get the data anywhere we have an Internet connection.”
Data are collected every 15 minutes. A rising water level triggers the flow monitor to collect data every minute, with a set threshold of 3 feet.
The next step in the project will be for researchers to submit the findings to a hydrologist affiliated with the National Research Institute in Agriculture in France–a counterpart to the United States Department of Agriculture–who will analyze the data and collaborate with social scientists.
Xylem and its many brands–which includes SonTek/YSI–offers a range of instrumentation that focuses on water quality and quantity, notes the company’s Chris Heyer. He says Xylem is noticing more concerns within the industry about what is contained in stormwater.
“They want to know when there is a storm event how is it increasing turbidity levels, total suspended solids, and nutrient loading into whatever the subsequent waterways are beyond the stream, the river, the pond–whatever they might be monitoring,” he says. “We are seeing a large growing trend around increased loadings to waterways and what the impacts of those loadings are in terms of low-dissolved oxygen and potential fish kill–any number of things.”
That ties into both urban and agricultural environments, although the components those monitoring stormwater are monitoring are vastly different, Heyer says.
The company offers sensors that focus on basic ambient water-quality parameters such as temperature, conductivity, turbidity, pH levels, and DO. Instruments for water quantity measure both standard water level as well as velocity.
Integrated Systems and Solutions is a full custom solutions division offering options for stormwater monitoring.
One multi-parameter water-quality instrument is typically used alongside a rain gauge and has an autosampler for water samples, says Heyer.
“The autosampler will draw a water sample from the stream, river, or stormwater drain and store it in different compartments of the sampler,” he says. “Some of the samples might be 24 samples over a 24-hour period. Those are often event-triggered either by flow or by rainfall and can be picked up later by a technician.”
Telemetry also is incorporated into the system, allowing technicians to see in real time that an event occurred as well as the ambient water-quality parameters, rainfall, water level, or water flow for that event.
End users receive notification that their sampler is actively triggering it so they can retrieve the water samples for further analysis in a laboratory for such factors as total suspended solids.
Xylem’s brand companies build solutions such as a turnkey system with solar power, which might include an EXO water-quality multiparameter sonde, a rain gauge sensor, an autosampler, telemetry, a water level sensor, and a water velocity sensor, Heyer says. The YSI EXO system is designed to offer calibration and redeploy in the time span of a typical sample interval; it features wireless communications, onboard diagnostics, copper alloy parts, and anti-fouling wipers.
YSI also provides flood alert monitoring to the industry through standard system and custom-built systems. These systems include components that monitor water levels, log precipitation, calculate water discharge, and transmit information.
Studying Temperature
Representatives from St. Anthony Falls Laboratory at the University of Minnesota have conducted field studies for the Minnesota Pollution Control Agency to collect the data necessary to support the formulation and validation of a temperature simulation model for urban stormwater detention ponds.
In May 2011, the University of Minnesota prepared a report for the Minnesota Pollution Control Agency (MPCA), authored by Michael P. Weiss, William R. Herb, and Heinz G. Stefan of the St. Anthony Falls Laboratory, regarding the stormwater detention pond water temperature data collection and interpretation. The researchers indicated the purpose of the data collection effort was to determine the impact wet detention ponds have on runoff water temperature, especially if the outflow drains into a cold, class A trout stream.
The data were used to develop and to validate a model for predicting outflow temperature and to offer recommendations for the reduction of stormwater pond outflow temperatures.
Detailed field data were collected on 17 ponds in the metropolitan area. The study was considered important because there are several cold-water trout streams in the periphery of the Minneapolis/St. Paul metropolitan area and in the city of Duluth.
Researchers also collected additional data to document salinity profiles in stormwater detention ponds in connection with the use of road salt in the Twin Cities area.
Land use was important in the field study. Researchers sought a stormwater detention pond in an industrial or commercial development or residential area with high percentage of impervious area.
Other desirable pond attributes included ponds used in previous studies, unshaded pond surfaces, one inlet as opposed to multiple inlets, a single outlet, a pond surface area of 1 to 5 acres, a continually flooded wet pond with an open water area 3 to 10 feet deep, no deterioration or silt, and a drainage area easy to determine.
The urban stormwater detention pond simulation model is included in the MINUHET (MINnesota Urban Heat Expert Tool) model that computes runoff temperatures for typical residential and commercial watersheds, simulating single rainfall events or continuous periods of several months.
The simulated runoff temperatures and volumes are used to estimate the heat loading from urban surface runoff to cold-water streams. To support the simulations, weather data and urban runoff temperature data were collected to serve as model inputs and to validate model outputs.
In 2005 and 2006, researchers studied a pond for detailed instrumentation and data collection on the former site of the State Farm Insurance Company Headquarters, near I-94 and Radio Drive in Woodbury, MN. The manmade wet pond featured an outflow structure and one major storm sewer inflow from two parking lots and the roof. Its surface area is 1.32 acres. Its drainage collection area of 43.5 acres is 52% impervious. It is nearly completely unshaded. The permanent pool depth is 2.4 meters.
The installed instrumentation measured and recorded weather data, temperature stratification data in the pond, surface inflow and outflow data, pavement temperature, and pavement runoff temperature data during and after rainfall events. The instrumentation in the pond was operated from June 3 to August 25, 2005.
The pond has a clay liner to prevent water loss by infiltration. An outlet structure stops the outflow from the pond when the maximum pond water depth has dropped to about 2.43 meters. When it is not overflowing, the pond’s surface area is about 1.2 acres.
Surface water inflow is from the former State Farm Insurance Company office complex, including an upper and lower asphalt parking lot.
A weather station comprising an anemometer, a wind direction vane, and a tipping bucket rain gauge was placed in the middle of the pond. A thermistor chain consisting of six temperature sensors attached to the pole and a pressure sensor was placed below the water surface to record pond water level. All instruments were connected to a Campbell Scientific data logger CR-10 that was attached to the pole above the water.
The weather parameters and the water temperatures in the pond were measured every minute, with averages recorded on the Campbell data logger every 10 minutes. Water temperatures were measured with YSI Model 55032 thermistors with a time constant of about 10 seconds.
Wind speed was measured by an R.M. Young model 03001 anemometer and wind direction instrument. Water level was measured with an Instrumentation Northwest model 9805 pressure sensor.
Onset Hobo temperature loggers were used to record water temperatures in Celsius at one-minute intervals at the pond inlet and outlet structures and at one minute initially, then two-minute intervals at two stormwater catchments in the parking lots. Onset Hobo miniloggers were used to replace loggers of another manufacturer that had an upper recording limit less than what Onset could provide.
Two additional temperature loggers were buried in the surface of the asphalt parking lot to record pavement temperature in Celsius at two-minute intervals initially, and then five-minute intervals.
Water transparency in the pond was measured manually using a Secchi disk.
The report indicates that a numerical simulation model was developed to simulate the hydraulic and heat transfer properties of a stormwater detention pond. The model is dynamic and based on basic principles of hydraulics and heat transfer, driven by hourly climate and weather data.
At the former State Farm site, the pond model field data were calibrated and validated. The relationship between pond inflow and outflow rates to precipitation was effectively calibrated using continuously recorded pond level.
Algorithms developed for surface heat transfer in lakes were found to be applicable to the pond with some modification, according to the report. A significant diurnal thermal stratification was simulated and measured in the pond. Temperature differences from top to bottom were as high as 13°C (55°F) during daytime hours.
The outflowing water temperature was essentially equal to the pond surface temperature because the outlet was located near the pond surface, researchers pointed out. Outflow water temperatures were calculated with a RMSE of 1.40°C (55.40°F). Water clarity had little effect on the pond outflow temperatures; however, the pond bottom temperature was found to be highly sensitive to water clarity.
Researchers concluded that for pond designs with outlet structures that take subsurface water, water clarity will introduce uncertainty to simulations of the pond temperature profile and the pond outlet temperature.
Onset’s data loggers are primarily used in research for those trying to develop new ways of managing stormwater, notes the company’s Paul Gannett. “I’m hearing of a lot of efforts going on relative to handling the stormwater at the site better rather than just having it run off and dealing with it somewhere else,” he says.
Green roofs are a prime example. “You have a green roof that absorbs the water, allows it to dissipate through evapotranspiration or maybe collects it for use somewhere, so the result is less runoff,” says Gannett. Permeable pavement is another example.
Data loggers are used to measure the amount of rainfall.
“The areas being studied are the amount of rain falling onto the roof and into the parking lots,” Gannett says. “The data loggers monitor that rainfall–not just the amount of rainfall, but when it fell and how it behaves is sometimes part of the study. You want that time record of the rainfall. On the other side, you want to measure how much water runs off from that.”
There are three ways to measure it, Gannett says.
One is through runoff gauges, which can be rainfall gauges in themselves. “A lot of times, they’ll take the runoff, run it through a rain gauge or some sort of runoff gauge using rain gauges and event loggers,” he says. One of the most used tools for that purpose is the HOBO Data Logging Rain Gauge-RG3. The rain gauge records up to 160 inches of rainfall at rates up to 5 inches per hour. The battery-powered system includes a HOBO Pendant Event data logger with a tipping-bucket rain gauge to collect rainfall, time, and duration data as well as temperature when used with an optional solar radiation shield. A base station or shuttle is required.
Another technique uses water-level loggers that are deployed in collection tanks that collect the rainwater for another use or until it can dissipate into the ground. The water loggers deal with quantity information–the amount of rainfall coming in and the flow out of the system or collected within the system. The 13-foot HOBO Water Level Data Logger is used to monitor water levels and temperatures in wells, streams, lakes, and wetlands. For saltwater use such as brackish wetlands and tidal areas, the HOBO U20 Water Level Titanium is used. The system features lightning protection. HOBOware Pro software provides conversion to accurate water level reading, fully compensated for barometric pressure, temperature, and water density. Multiple-rate sampling allows faster sampling at critical times such as when pumping starts or stops.
A third common technique is examining stormwater inundation from storm surges along coastal areas. Case in point: Hurricane Sandy. Examining the inundation helps in mapping out rebuilding strategies.
“There are areas where you shouldn’t rebuild at all, areas where you can rebuild but you should plan on inundation for this amount of time, or build on stilts–those kind of strategies,” points out Gannett. “There are a lot of deployments of water-level loggers in those coastal areas by organizations like the US Geological Survey.”
For green roofs, the U30 remote monitoring systems are especially important because they can monitor incoming rainfall and accept a range of sensors, says Gannett.
“They’ll accept soil moisture, which is important for a lot of these studies. You want to monitor how much water is being absorbed into the soil and available for plants, because it’s important for the plant’s health to have the right amount of soil moisture,” he adds.
In some cases, it’s used to manage irrigation systems on the green roofs, says Gannett. “Sometimes you get too much rain, and there are other times when you’re not getting enough rain. You need to do some sort of irrigation to keep a healthy, well-functioning green roof to keep those plants alive up there,” he points out. “Sometimes there’s a lot of experimentation with collecting the rainwater on the rooftop and using that to irrigate the rooftop during periods of low rainfall. Or in some cases, they’ll just have an irrigation system that’s tied into the water supply that can be used to irrigate the roof in times of low rainfall.”
In the past year, Onset has started to bundle its existing products that had been available for the U30 remote monitoring system to make it more convenient to put together a system for an application such as a green roof, Gannett says.
The U30 remote monitoring system is a web-enabled system that can tie into a building’s WiFi; users can post the data from the green roof on the Internet.
“That’s very popular because a lot of the green roofs have a double benefit,” says Gannett. “You’re reducing the amount of runoff. Also, a lot of companies want to show how they’re taking steps to be more green. When you can publish that information, show that on a website and provide some real tangible data, it sets a positive corporate message for what they’re doing to be green, reduce runoff, and create a better environment.”