Stormwater | Remote Water-Quality Monitoring

Remote Water-Quaity Monitoring

By Carol Brzozowski

This is water monitoring of the past: After a heavy rainfall in the middle of the night, a technician would drag himself out of bed to take a "first flush" sample. Time-consuming routine sampling had been done by hand and involved such challenges as navigating hazardous areas to acquire access.

This is water monitoring today: Automated equipment, coupled with data management systems, are getting sampling results to labs, stormwater program managers, even the public via Web sites - and it's all done remotely.

Throughout North America, remote water monitoring devices are being used to get a snapshot of water quality: in Chicago, IL, to study one of the world's largest water reclamation districts; in California, to monitor municipal stormwater; in Colorado, to study the impact of wastewater and mining practices; in Alaska, to examine a glacier and its surrounding ecosystem; and in Oregon, even to predict volcanic activity.

National Pollutant Discharge Elimination System (NPDES) Phase I permits require numeric water-quality monitoring. Developing total maximum daily loads (TMDLs) also requires careful monitoring of water quality over time. NPDES Phase II permits generally do not require numeric water-quality monitoring, although some stormwater managers in Phase II cities will do it in an effort to determine if the best management practices (BMPs) being used are effective.

The parameters being tracked depend on several factors, including:

whether a lake or river is considered "impaired" by some type of pollution,
if the water is used for drinking or swimming,
concern over aquatic habitat,
if an upstream event such as construction is causing immediate problems.

Today's equipment may include options to monitor many of these parameters:

Turbidity or total suspended solids
Temperature: Influenced by soil erosion and greater impervious surface in the watershed, increased temperatures can increase biological/biochemical oxygen demand (BOD), decrease oxygen, and decrease aquatic diversity
pH level: Influenced by acid mine drainage and particulates from automobiles and coal-fired power plants; can result in increased toxicity from metals and decreased diversity
Dissolved oxygen (DO): Lower volumes, increasing temperature, sewage effluent, and fertilizer runoff can result in decreased diversity
BOD: Urban, fertilizer, and agricultural runoff and industrial and municipal inputs result in decreased oxygen, changes in speciation, and decreased diversity
Phosphorus: Influenced by sewage effluent, animal waste, urban runoff, and agricultural considerations, results include excessive plant growth and decreased oxygen
Nitrogen: Influenced by sewage effluent, animal waste, agricultural runoff, contaminated groundwater, and urban runoff, increased nitrogen can result in excessive plant growth, decreased oxygen, and "blue baby syndrome"
Metals, including mercury, copper, lead, zinc, and aluminum
Toxic organic compounds
Fecal coliform contamination and other pathogens: Sewage effluent, animal waste, and urban runoff can result in pathogens that cause many diseases in humans

Pollution Control in Chicago

In the world's largest water reclamation district, Chicago-area scientists are using remote water monitoring systems to determine how successful its pollution control methods are with respect to 220 mi. of natural and manmade waterways that serve more than 150 communities. The Metropolitan Water Reclamation District of Greater Chicago also is getting a bird's eye view of its waterways' ecosystems and biology through a system that was deployed in August 1998.

The prevailing belief about the ecosystem was that it was perturbed and not very complex. But through monitoring, the ecosystem has manifested itself as a dynamic system in which pollution factors, flow, and water quality are always changing, necessitating continuous monitoring. Some 20 YSI 6920 multiparameter sondes in custom-built, stainless steel enclosures to protect against theft and boat traffic are deployed weekly.

Each unit logs 168 measurements of DO, temperature, and conductivity. It is retrieved by a boat, replaced with a newly maintained sonde, and returned to the lab, where the data are downloaded into Microsoft Access and the unit is calibrated and cleaned.

The Access report generates for the district's research and development department weekly summary statistics and line drawings of all data points for each sonde, which are passed on to maintenance and operation engineers for their use. In the long term, the water-quality reports are being used to efficiently allocate public works resources.

Combined sewer overflows, storms, and other environmental events are chronicled to such detail that the district has been able to solve some of its water-quality mysteries:

One summer, a fish kill occurred in a spot with four monitors. The district was able to determine when the overflow occurred and how the plume moved downstream.
Diurnal changes in DO from algae and aquatic vegetation and the mixing that takes place in the water as it flows through the system was revealed as a result of continuous monitoring. A matter that had not been addressed in scientific literature, the explanation for the DO changes was found in the process of cold-water algal blooms in the river.
The district embarked on a study centered on a concern over how much water the area can take in from Lake Michigan through the locks system. "We are limited to a volume that we can take in from the lake, and we've noticed that in certain times of the year there's actually flow reversal where the flow that would normally be going away from the lake seems to flow toward the lake. You can see that in the dissolved-oxygen readings that we've taken," says Sam Dennison, a biologist with the reclamation district.

Photometer in lab application

Additionally, a suite of YSI 6600 sondes are utilized as an upgrading of a decades-old annual survey of water-quality grab samples along 160 nautical mi. of the Illinois Waterway System from Chicago to Peoria. The equipment enables district staff to measure DO, water temperature, conductivity, pH, turbidity, and chlorophyll with a single monitor from a boat.

"We're trying to have a more efficient interpretation and dissemination of continuous dissolved-oxygen data," relates Dennison. "Part of this has to do with NPDES and part of it has to do with determining locations of our side stream elevated pool aeration stations based on what changes we see in the dissolved oxygen."

Hourly measurements have been taken ever since the remote water monitoring began. "We find that as long as we are able to exchange the monitors with another monitor at the end of every weekly period, we can allow them to have continuous data," he states.

Generally the monitors are serviced weekly. Personnel retrieve each monitor from the field following seven days of continuous monitoring. Before retrieval, a water sample for DO is collected next to the protective housing where the unit is located, and another monitor that previously had been calibrated and serviced in the lab is deployed to replace the retrieved monitor.

When the monitors are in the laboratory for data downloading, they receive external cleaning, servicing, and calibration of the DO sensors. The sample that was manually taken in the field during the switch of monitors is subjected to the Winkler method to verify the results of the automated system.

According to Dennison, considering the travel time of a boat going up and down the Chicago river system, the time and labor savings using remote water monitoring is most likely "considerable." He believes Chicago will expand the system as needed.

"Right now we're covering the entire waterway system," he points out. "For example, the 34^th monitor was added recently to be involved with a project that's going on in a fairly small area having to do with repumping of water back through a pumping station. That's just an experimental project, but whenever projects like that come up, we might have to add a monitor."

Designing Monitoring Programs in California

Pat Kinney of Kinnetic Labs in Santa Cruz, CA, has helped develop automatic sampling systems out of available components, to which controllers and telemetry are attached. The systems are equipped with flow meters. Kinnetic uses Campbell Scientific products.

"We usually need to get flow composite sampling," Kinney explains. "We can turn the samplers on remotely; we can then change all the sampling parameters. We can tell what the voltages are and if everything is working or not remotely."

During a storm event, information is transmitted by telemetry to a central control area where someone is managing the samplings. The samples themselves are brought back to the laboratory for analysis. Field crews are available to handle situations when something needs to be done at a station - for example, if it rains harder than usual or sample bottles are overfilling or something isn't working correctly. "The field crews can go there instead of just running around at random," Kinney notes.

Kinnetic Labs has monitored parameters in some 1,000 station events over the last few years and also uses absorption media to get a reading of absorption of parameters over time. The lab tests the samples for such parameters as toxicity, organics, and metals.

"We've done municipal stormwater monitoring programs for more than 20 large urban areas; a lot of them would be Phase I communities. We've also done a lot for the California Department of Transportation," Kinney relates. "We've done a lot of structural BMP analysis - sand filters, detention basins, bioswales, and others."

Kinnetic Labs also does periodic manual sampling to verify results obtained from the automated system. Routine maintenance is done monthly. During heavy rain periods, affected sites are monitored constantly, with human intervention if necessary to adjust the sampling rates or change the bottles.

"Oftentimes we're doing a lot of toxicity monitoring where we may need 20 to 60 liters of water for all the tests that have to be done, so you have to set the sampling rate so you get the amount of water you want," Kinney describes. "But if it rains a lot more than you've predicted and your sampling rate is not set right, then you have to do something about that during the storm."

Watershed Monitoring in Colorado

In Westminster, CO, remote water monitoring technology is being used at the site of what water-quality officials there call "the most monitored watershed in the country."

Andrew Cross is a water resources technician for the City of Westminster. While he is engaged in many water sampling programs, Upper Clear Creek watershed is more under his watch than other areas. According to Cross, there are a number of communities in the mountains from which Clear Creek runs down with snowmelt.

"A lot of those communities are old mining towns and may have their own wastewater plants. There was a time when they were largely unmonitored," Cross says. "Since the water's coming down to the cities here on the front range, some of the communities got a little concerned and started a monitoring program to cajole the mountain communities to clean up their sewage. You combine that with all the old mines that are up there, which tend to discharge metals and things of that ilk, and when you put it all together, they decided in the 1980s to monitor the watershed."

The RUSS water-quality monitoring system

Until recently, water monitoring had been done by a personal visit, but Westminster has augmented that by the use of the RUSS water-quality monitoring system. The buoy system features an Apprise Technologies' RePDAR (Remote Programming, Data Acquisition and Retrieval) onboard system that works in conjunction with a cellular telephone modem and is powered by a marine battery recharged by an onboard solar panel. Cross sees some benefit to upgrading to a 900-MHz transceiver or another radio-based system as opposed to the analog cellular technology.

Connected to the buoy by an umbilical cable is the Profiler, which Cross describes as a submarinelike device with two cylinders: a dry cylinder hosting the electronics and pump and a wet cylinder hosting an air bladder and into which water is pumped in and out.

There also is a YSI9000 sonde onboard the RUSS. While the Profiler is a conveyance device, the sonde does the actual monitoring, Cross points out. The sonde is hooked up to the Profiler, which is connected to the buoy, all by cables. An antenna onboard the buoy allows Westminster water officials to have the data transmitted.

Water-quality parameters being monitored by the probes include pH, oxidation reduction potential (ORP), turbidity, chlorophyll, conductivity, and DO.

Cross changes the sondes every two weeks so he can recalibrate them. "Otherwise the data start to stray too much from reality," he notes, adding that he uses a laptop when he goes out to check the buoy.

"I hook up directly to the buoy, then when I'm doing the switch, I make sure it's communicating properly," Cross explains. "While it hasn't happened yet, if something were to go wrong or if we weren't able to communicate with it properly from the base station - because there are problems with cell coverage occasionally - I can imagine sometime someone saying they need to see the data immediately and I can go out there on a boat and directly download them from there if necessary." Cross performs manual sampling to verify the results when he changes the sonde.

The buoy is anchored in a reservoir, which is about 25 m deep. "That's a little deeper than the cable is capable of reaching, so we've pretty much maxed its capability out as far as the depth goes," Cross states. He notes that the RUSS is pulled out in the colder months as a result of a RUSS in Fort Collins taking a beating during the winter.

The system, which was deployed in April, was financed by three different cities that use the reservoir, including North Glenn and Thornton. It is programmed to monitor water quality twice daily as he gathers large quantities of data in order to develop baselines.

"We do have plenty of data over the years, but nothing quite this consistent on a day-to-day basis," Cross notes. "It's always been two weeks to a month apart per monitoring event. Now we've got something going daily; we're getting thousands of percent more data than we used to.

"What I will do is especially watch the DO levels, because the current outlets in Standley Lake are in the bottom of the lake and the lake goes anoxic on the bottom first. When that happens, manganese is released. I try to use it as an early warning system for the operators of the treatment plant."

Another application is to subject the data to statistical analysis in an effort to identify trends. "Are we seeing a warming trend? That's hard to tell anyway because the lake fluctuates so much, but as far as other quality parameters go, we might be able to actually see some patterns," Cross says. "We want to develop a decent picture of what's happening in the lake."

He adds that the lake is undergoing major renovations. The Standley Lake Dam, one of the largest urban dams in the country, has houses below it. "There is some concern that having outlet works going through the dam isn't necessarily the best way of dealing with the outlet works. Right now they're microtunneling underneath the lake and putting in new outlet works to go around the dam. They're going to retire the old outlet works."

When the RUSS was placed in the lake, it was done between the old outlet works and the new outlet works with the idea being that once the new outlet works are used, there will be a discernible difference in the draw going through the lake.

"Right now the inlet to the lake is closer to the current outlet than the new one is, so we're hoping that we might be able to see some of the effect of having changed the outlet works position in the lake," Cross says. "The current outlet works are only at one depth - pretty much the bottom - and the new outlet works are going to be up a little bit higher. One of them is slightly higher than the old outlet works and the other one is even higher than that. So they'll be dual levels, and we're hoping that by using the RUSS, we'll be able to see anoxia and that sort of thing happening in the lake, so we can predict manganese coming through and determine [whether] it would be a good time to use the upper or lower inlets depending on what the RUSS profile is showing us."

Studying the Bering Glacier

Altarum Institute staff take basic water-quality measurements of Berg Lake, which is part of the Bering Glacier watershed in Alaska.

Hanging out of a helicopter to take water-parameter measurements is a twist on the concept of remote water monitoring. But that's exactly the approach taken by Robert Shuchman, a senior vice president and chief technical officer for Altarum, a not-for-profit research-and-development institute that stemmed from the University of Michigan in the early 1970s and addresses a variety of issues for federal, state, and local governments.

Shuchman and his colleagues have been participating in a water-quality study of Bering Glacier and several surrounding alpine lakes in Alaska on behalf of the Bureau of Land Management (BLM), a division of the United States Department of the Interior.

The program is an ongoing scientific activity at what is the largest glacier in North America, which sits atop BLM land. BLM is considering how remote sensing satellites, as well as in-situ measurements, can best characterize not only the status of the glacier but the surrounding ecosystem. It is also studying how the glacier affects the local ecosystem, part of which includes indigenous fish in the glacial lakes that are a result of the Bering Glacier having existed for the last 20,000 years, Shuchman states.

The glacier is a "poster child" for remote water monitoring. "Bering Glacier is about a hundred miles from nowhere," remarks Shuchman. "It's halfway between Cordova and Yakutat. You could go in by a slow boat to China or fly in by light aircraft on the landing strip. Everything that's done on the glacier and the surrounding water bodies is done by Zodiac [inflatable boat] or helicopter."

Shuchman and his colleagues are dealing not only with the glacier but also with Vitas Lake, the volume of which is nearly the size of Lake Erie, Shuchman points out. Altarum is doing profiles in Vitas Lake to address the question of the nature of the fish population and why there is a set of harbor seals that inhabits the Vitas Lake region.

"There are more than 1,000 seals reported there," he says. "Why are they there? Are they eating fish out of the lake? There are some very interesting ecology problems that the BLM needs to understand so they can better manage this wilderness area."

Because some of the higher alpine lakes surrounding the glacier area are inaccessible short of backpacking for a few days to gain access, Shuchman and his colleagues have had to remove a door from a helicopter and position the helicopter approximately 10 m above some of the glacial lakes to sample the lakes while the helicopter is in hover mode.

"You can't be afraid of heights, but you can do it," he says. "We found that it was a fairly efficient way to do some of this sampling because some of these ëfish experts' would say there are absolutely no fish in them [the glacier lakes], and sampling shows there are all sorts of fish in there, which is fascinating. There are a whole bunch of wild theories on how these lakes got populated with fish. But at this point, our job is to characterize and to do a benchmark of water-property measurements."

In doing so, Altarum is using Horiba Instruments's U-22XD water-quality monitoring system. Although monitoring of Bering Glacier has been ongoing for more than a decade, the measurements haven't been as comprehensive as with the Horiba instrumentation, Shuchman points out.

"It was a pH here and temperature there," he recalls. "The neat thing about the Horiba, particularly with dissolved oxygen, for example, is that we're now getting a comprehensive suite of measurements as a function of depth. This used to take years to do with water sampling and titration; now, literally within 15 minutes, we're getting a comprehensive look not only at the surface water but the water column. It's a much more efficient way to do this."

The system offers measuring parameters that include pH, DO, conductivity, salinity, total dissolved solids, seawater specific gravity, temperature, turbidity, water depth, and ORP.

"We're not only getting those on the surface, but we drop it down one meter at a time and do profiles. We've done a couple all the way down to 100 meters," Shuchman reports. "We've got 100-meter slices of Vitas Lake, which they never did in the past."

Altarum has written a program whereby the Horiba values are fed into a geographic information system, and Altarum automatically receives a three-dimensional map of the lake with several of the measurements as a function of depth. Every profile is tagged with latitude and longitude.

"It allows you to do slices, and you can rotate the lake around so that you can look at layers," Shuchman explains. "We're finding that a very useful scientific visualization tool."

Because of logistics and safety concerns, the Altarum team goes out in the first two weeks of August when they have the most access to the ice-free areas. "These measurements don't happen in five seconds," Shuchman notes. "Each helicopter event is a half day. You fly up to two lakes and back, and you're basically exhausted after hanging out of a helicopter without a door."

The Altarum team carries new sensor heads so that when they are engaged in parameter measurings, they can change a sensor head that needs it.

Ichthyologists are doing more classical water sampling at the surface of pH, DO, and total dissolved solids; their results are compared to the measurements acquired through the Horiba. When the Altarum team first used the system last year, it had 12 stations; it plans to add another 30 stations. The fish species and water-quality reports are expected to be completed this year.

"Without this instrument, we wouldn't be making these measurements; they would be too complicated at this point to bother with," Shuchman maintains. "It's 125 miles from nowhere. If you had to do this the old classical chemistry way, it would be way too complicated to titrate things in the field, so I don't believe without an instrument of this type we would have hatched this scientific issue [of relating] water-quality measurements to fish species that are in these surrounding lakes."

Volcanic Activity in Oregon

Studying water-quality parameters as an indicator of potential volcanic unrest is one of the more unusual ways In-Situ Inc.'s remote water monitoring equipment is being used. The US Geological Survey (USGS) is monitoring temperature, pH, stream depth, and conductivity in a stream draining the Three Sisters volcanic area in central Oregon.

"The main parameter we can relate to volcanic unrest is conductivity, but we need to monitor the other parameters in order to relate the conductivity to the amount of chloride discharged in the stream," explains Bill Evans of the USGS. "Chloride is released from magma as it rises to shallow depths within the crust. There are no other likely sources of chloride in this pristine environment, so an increase in chloride in the stream would indicate an increase in magmatic degassing."

For instance, the first step in using the data is to remove the effect of temperature on conductivity by converting to specific conductance. The temperature coefficient of Separation Creek water is determined by laboratory heating experiments.

The USGS has been monitoring the stream since April 2002, following detection of crustal uplift on the west side of Three Sisters. "That uplift is still ongoing and likely indicates intrusion of magma at depth," Evans notes.

The USGS is using two In-Situ Multi-Parameter TROLL 9000 systems to monitor the waters of Lower Separation Creek. Other equipment being used: a 25-ft. QuickConnect cable and communication cable, a laptop computer, dye dilution equipment for gauging stream flow, and a 6-ft. section of PVC pipe for a stilling well in which stage measurements are performed.

In deploying the equipment, the first probe was placed inside a plastic pipe that serves as a stilling well and provides protection from floating debris. The second probe was installed in a small tributary high in the Separation Creek drainage where a stilling well is unnecessary. The QuickConnect cable is securely fastened above the high-water mark. Conductivity and pH sensors are calibrated at each visit to ensure data accuracy. Hourly data are downloaded weekly at the site via a laptop.

Carol Brzozowski is a journalist in Coral Springs, FL.

SW November/December 2003

Home + About + Subscribe + News + Calendar + Glossary
Talk + Images + Advertise + Contact Us + Search + Register + Services