Braving the midday blaze on a particularly hot Front Range Friday, Evan Smart and a friend lingered near their inner tubes, beer cans in hand, after a refreshing float down Boulder Creek.
“Over my whole life of coming into the creek, I’ve never heard of anybody getting sick,” said Smart, a 26-year-old paramedic. I’ve always heard that it’s got E. coli in it – don’t drink the water – but I’ve never heard anything about it.”
But public health officials are taking this fecal bacterium quite seriously, as summer temperatures make Colorado’s waterways ideal breeding grounds for Escherichia coli. Policymakers and scientists across the state are working to decipher which types of microbes are lurking in the water, and whether they actually pose a significant threat to human health.
Smart said for at least two decades he’s been cooling off in Boulder Creek, where a sign now warns that bacteria levels are sometimes high enough to cause illness. And while he said he likely would wash his hands before eating anything – though he admitted he had opened his beer can without doing so – he is “totally fine” tubing.
“Realistically, everybody’s gone in the creek,” Smart said. “I’ve gone in it for a long time. If you drink a ton of water, you’re going to be sick, but if you just get it in your mouth, I think you’re fine.”
100 waterways considered “impaired” by E. coli
The Colorado Department of Public Health and Environment has registered more than 100 waterbody segments on its impaired waters list due to alarmingly high E. coli levels. While only certain strains of E. coli cause illness in humans, officials do not yet have the capacity to pinpoint in any real-time fashion where and when these strains congregate.
Among the newest segments on the list is the stretch of Boulder Creek between the mouth of Boulder Canyon and 13th Street, where Smart and dozens of others were tubing without a second thought at noon that Friday. This stretch of the creek, which includes Eben G. Fine Park and urban green spaces near the Boulder Public Library, also is a popular place for parents eager to splash with their children.
Prior to the latest update in January, the CDPHE had considered only the portion from below 13th Street to its confluence with South Boulder Creek to be impaired.
Like all states, Colorado must provide the Environmental Protection Agency with updated records of waterbodies that fail to meet water quality standards every two years, as required by amendments to the Clean Water Act. Each such biennial submission is based on the five most recent years of water quality data collected in a given location.
The city of Boulder isn’t sure where the contamination is coming from, but a team led by Candice Owen, the stormwater quality supervisor, is trying to figure it out. She and her team will be taking more frequent dry weather discharge samples toward the end of the recreation season, when E. coli concentrations are typically highest, she said. The city also recently began posting precautionary signs along the creek, in English and Spanish, indicating the periodic presence of bacteria.
“Based on our analysis, E. coli in the creek hasn’t really changed over time,” she said. “It hasn’t gotten significantly worse. Unfortunately, it hasn’t really gotten better.”
“It’s how we are assessing the creek that has changed,” she added.
Owen was referring to a 2016 update to the state’s listing methodology that made E. coli contamination assessments much stricter.
The CDPHE’s Monitoring and Evaluation list, or M&E list, includes waterways in which two, three, or four water samples have exceeded the EPA’s recreational-waters standard of 126 colony-forming units (cfu) per 100 milliliters. For more serious violations, in which there is “overwhelming evidence” of contamination, waterways end up on the state’s list of impaired waters, officially known as 303(d). The Water Quality Control Division defines overwhelming evidence as exceeding water quality standards by more than 50%.
While EPA standards consider recreational waters to be impaired if E. coli levels exceed 126 cfu per 100 mL – as opposed to 235 cfu per 100 mL necessary for swim beach closures – the CDPHE warns that risk of becoming ill still exists in these waters.
CDPHE recommends that people take precautions if they choose to swim in impaired waterways, mainly by avoiding swallowing water and washing their hands upon exiting. To minimize further contamination of the waterways, the department advises showering before entering, taking children on frequent bathroom breaks and staying out of the water when ill with gastrointestinal symptoms.
“Determining that a waterbody is impaired initiates steps toward identifying the source of the impairment so the state and community can make progress toward restoring water quality,” CDPHE spokeswoman MaryAnn Nason said in a written statement. “Elevated levels of E. coli are not uncommon in urban areas in Colorado and around the United States. E. coli is higher in these areas because there are more animals in urban areas.”
A body of water can get off the impaired waters list if a minimum of five samples collected in the same calendar months in which the original breach occurred remain under the EPA standard for at least two years in a row.
Flowing water makes E. coli hard to track
Tracking E. coli is particularly difficult in urban streams, as the microbes in question have typically washed far away from their sampling site by the time officials have test results, Owen explained. Due to that in-stream flow factor, taking reliable samples can prove much more challenging than doing so in comparatively stagnant swim beaches at lakes and reservoirs.
“The ability to assess that piece is much more challenging,” Owen said.
Although Colorado’s swim beaches are sometimes closed due to high E. coli levels, these sites do not tend to face the same type of perennial problems that plague many urban creeks, rivers and streams.
Boulder Reservoir has faced sporadic closures as a result of high E. coli concentrations, but the city has not opened the reservoir to swimmers this season as part of its COVID-related precautionary measures.
Last July, meanwhile, Rock Canyon Swim Beach at Lake Pueblo State Park closed for a week, following a rain event that washed animal fecal matter into the Arkansas River, said Bill Vogrin, spokesman for the Colorado Parks and Wildlife’s Southeast Regional Office. It was the first such E. coli-related closure at Rock Canyon since August 2010.
Cherry Creek Reservoir also temporarily closed in mid-July last year due to elevated E. coli levels, but this was also a relatively unusual event. Cherry Creek Reservoir is now open to visitors, and the reservoir has not exceeded E. coli limits this year, said Jason Clay, spokesman for Colorado Parks and Wildlife’s Northeast Region Office. However, the streams and tributaries to Cherry Creek Reservoir are listed as impaired.
“Streams are more shaded than lakes by trees and overpasses, so the UV light that kills the E. coli has less opportunity to reach the E. coli in streams,” Clay said. “When the water in the streams washes into Cherry Creek Reservoir, its travel speed slows down, and there is no shade, so the E. coli is fully exposed to UV light.”
MORE: Colorado Springs’ downtown creek has long been viewed as a blight. Then one man started catching trout in it.
The City of Boulder has begun implementing an initial mitigation plan for Boulder Creek – which is also heavily shaded – and made its first progress report to the Water Resources Advisory Committee on Monday.
But creek advocates say they already know what the problem is.
“The major discharger – the discharger who doesn’t seem to care and is not taking it seriously — is the University of Colorado,” said Art Hirsch, a retired environmental engineer and a watershed activist for the Boulder Waterkeeper organization.
He praised the city for taking E. coli reduction seriously, but said the stormwater is discharged continuously from several locations on the campus into the creek between 17th and Folsom streets. Although he acknowledged that the university technically is in compliance with its stormwater discharge permit, known as a MS4 permit, Hirsch argued that the permit itself is obsolete.
Joshua Lindenstein, a CU spokesman, said the university is working closely with the CDPHE to ensure best practices and compliance with the permit, which he said was issued in 2008.
“CU Boulder has for years done more than is required under our current MS4 permit as it relates to monitoring for, and mitigating against, E. coli in storm drains that discharge into Boulder Creek,” Lindenstein said.
The state is in the process of revising “nonstandard MS4” permits, like the one CU holds, but a backlog has caused delays in making such changes, said Nason, the CDPHE spokeswoman.
CU has both increased sampling frequency and further restricted animal access to storm drains – the latter of which he said would likely be addressed in the new permit, Lindenstein said. The university monitors sanitary sewers in order to pinpoint possible leaks and has promptly repaired any such problems that have been identified, he added.
“We continue to collaborate with the city of Boulder on E. coli assessment and mitigation efforts in Boulder Creek,” Lindsenstein said.
Boulder Creek is far from alone in its E. coli problems – with quite a formidable competitor at Confluence Park in Denver, where the South Platte River and Cherry Creek come together.
When storm drains undergo flushing or sediment in streams is stirred up, so, too, are the E. coli lurking in these spaces, explained Jon Novick, environmental administrator at the Denver Department of Public Health and Environment. Like Owen, Novick said pinpointing the bacteria’s exact sources is difficult, but he noted that raccoons congregate near the park and homeless individuals also camp along the river.
“No matter what you do to improve it, you’re fighting an uphill battle,” he said.
Still, Denver has launched a number of initiatives aimed at tackling the problem – particularly within the stormwater outfalls where “urban drool,” like irrigation return flows and other untreated water tends to accumulate, Novick said. For example, he said, the city has installed UV filtration systems that are quite effective in eliminating E. coli from sewage during dry weather.
“However, we have really struggled to demonstrate that this has any impact on E. coli in-stream,” Novick said. “It’s partly because there’s such a large reservoir of E. coli out there already.”
While water quality improvements are rarely detectable immediately, Novick said significant progress has occurred in the 40 years that Denver has been battling E. coli in the South Platte and Cherry Creek.
MORE: After long regarding the South Platte as not much more than a sewer, is metro Denver ready to love its river?
“If you look at the bacteria levels that we were seeing in the late ‘60s and early ‘70s, and even into the early ‘80s and ‘90s, there have been orders of magnitude of improvements,” he said.
In order to make water quality data more accessible to the public, the DDPHE developed a water quality app designed to help the public see where aquatic activity is safe and view other data, although the tool is currently malfunctioning.
Confluence Park samples collected on July 14 indicated that E. coli levels were above recreational standards at both the Cherry Creek and South Platte River testing sites, which Novick attributed to that day’s storm. The Cherry Creek spot is typically the greater offender of the two, due to its shallow water level, sandy bottom and shaded environment, he said.
“It’s where we see most people playing in the water,” he said.
Nonetheless, Novick acknowledged that officials don’t really know whether exposure to E. coli in impaired waters actually leads to illness. Public health agencies do not typically survey bathers to find out if swimming in the creek has made them sick, he said.
“Also, what happens if they ate a raw hamburger?” Novick posited. “No one is doing the epidemiology as to what’s the source.”
Researchers in Colorado and around the world are trying to circumvent this roadblock by devising accessible methods to identify whether human fecal pathogens are actually present in water recreation hotspots.
While traditional sampling methods involve testing for the presence of E. coli, scientists have come to the conclusion that only the human-associated strains actually cause disease in people, explained Annika Mosier, an assistant professor in the Department of Integrative Biology at CU Denver. E. coli is a common microbe that is able to thrive in many places due to its flexibility and hardiness – even taking up residence in waterways and replicating on its own.
“Our approach – and other people’s approach – is to look at a different indicator, but also look at the DNA sequence,” Mosier said.
Since the strains of E. coli that make humans ill are difficult to identify, researchers have begun sampling for human-associated members of the genus Bacteroides, another fecal organism that is much easier to identify in testing, she explained. Using an approach called “quantitative real-time polymerase chain reaction,” or qPCR, scientists are able to analyze Bacteroides’ DNA for specific genes linked to humans.
“When you see a positive hit for a human-associated Bacteroides, then you can be more confident that there is human fecal matter in that sample,” Mosier said. “If there’s fecal matter there, you don’t want to be swimming in the water because there’s a whole host of illnesses you could be exposed to.”
Scientists are also exploring a “community fingerprint type of approach” that involves sequencing bacterial communities in water and sediment samples to identify microbial signatures similar to those found in fecal matter, Mosier said. By comparing their results to a library of samples from animals that might inhabit streams, she continued, they can determine whether the community bacterial profile of a particular spot more likely matches that of an animal or a human.
As far as the widespread implementation of these methods is concerned, Mosier said the EPA is working to standardize sampling for Bacteroides and has developed some draft protocols. Such standards could be used to help waterway managers determine when – and if – they need to implement mitigation efforts.
“If you had enough data, you could pinpoint a particular area where human fecal matter is more problematic than another area,” Mosier said.
The promise of such technologies has given municipal officials some optimism that this elusive problem might not persist indefinitely. Reflecting on the state of Confluence Park, and whether this hotspot might make it off the impaired waters list any time soon, Novick expressed cautious optimism, tinged with realism.
“It’s possible,” he said. “I don’t think it will happen any time in the near future – probably not while I’m still working.”