U of I online tool protecting communities’ drinking water from wildfires

When a large wildfire recently burned the forested slopes surrounding Walla Walla, Washington’s primary drinking water source, city officials turned to University of Idaho researchers Erin Brooks and Mariana Dobre to assess potential water-quality risks and help guide the community’s response.

The College of Agricultural and Life Sciences faculty members enhanced U.S. Forest Service software to make the Water Erosion Prediction Project (WEPPcloud), which is an online interface that simulates erosion, runoff and sediment delivery in watersheds. Their user-friendly watershed modeling platform has become a critical asset for cities that still rely on unfiltered surface water for their drinking supply.

Brooks and Dobre collaborate with a growing list of communities, planning proactive projects that minimize the severity of wildfires within surface drinking water source watersheds and provide insight into the best mitigation measures to minimize impacts to drinking water after a wildfire has occurred.

“The main goal of the tool is to protect water resources, especially after wildfires, but it also helps managers prepare and plan for measures such as forest thinning and prescribed fires to reduce wildfire risk,” Dobre said.

Other cities that have worked closely with Brooks and Dobre to protect sources of unfiltered drinking water include Ketchum; Seattle and Bremerton, Washington; Portland, Salem, Eugene and Medford, Oregon; and Victoria, Canada.

Paradise fire

During the summer, the massive Paradise Fire burned 45% of the forested land surrounding Mill Creek, which supplies nearly 90% of Walla Walla’s drinking water. The city disinfects Mill Creek water with chlorine and ultraviolet light and sometimes blends it with well water to ensure it meets standards, but officials worry a large storm in the burn area could wash excessive ash and sediment into the stream.

In the wake of the Paradise Fire, Adam Klein, water engineer with Walla Walla’s public works department, has been in near-daily contact with Brooks and Dobre to understand the current risk.

“U of I’s help is incredibly valuable,” Klein said. “They’re bringing expertise that the city of Walla Walla just doesn’t have.”

Prior to the fire, Walla Walla had been working with the U of I researchers on planning thinning projects and controlled burns surrounding Mill Creek to minimize their wildfire risk. Many areas slated for fuel-reduction projects went up in flames during the wildfire, so the city has shifted toward using WEPPcloud for post-fire risk assessment.

Brooks and Dobre will run their program to analyze a Forest Service burn-severity map surrounding Mill Creek. The researchers have also installed sensors and auto-water samplers in the watershed to monitor water quality. Klein hopes the threat will prove to be minimal, buying time for the city to find alternate water sources or allow the Mill Creek drainage to recover. In the worst-case scenario, the city may have to build a surface-water treatment plant, which could cost upwards of $75 million.

“We’re one of the last unfiltered systems, so we are a lot more sensitive to surface water impacts,” Klein said. “We haven’t seen any tangible impacts to water quality yet.”

Brooks and Dobre have brought in nearly $10 million in grants since 2007 to develop and apply WEPPcloud and expand its capabilities, including funding from NASA, the National Science Foundation, the U.S. Forest Service, the U.S. Geological Survey, the USDA Agricultural Research Service (ARS) and the USDA National Institute of Food and Agriculture.

Their team of U of I scientists also includes three postdoctoral researchers, and they plan to add two additional postdoctoral researchers and a doctoral student.

WEPPcloud’s evolution

The USDA-Agricultural Research Service developed the baseline model in the late 1980s and 1990s to estimate primarily agricultural erosion. Research leads at the Forest Service’s Rocky Mountain Research Station in Moscow adapted the model to estimate erosion from single hillslopes or road segments in the late 1990s.

In the early 2000s, Brooks partnered with researchers from the Rocky Mountain Research Station to modify it to help predict erosion from large road networks in the South Fork Clearwater River Basin.

In 2013, Brooks and Dobre, who was a postdoctoral researcher at the time, began a three-year project to develop WEPP into a model capable of evaluating erosion in multiple large watersheds in an effort to plan forest thinning surrounding Lake Tahoe, which was rapidly seeing reductions in water clarity due to fine sediment loading from contributing forested watersheds.

“What the tool provided for Lake Tahoe was an assessment of, ‘What are the soils you want to avoid? Maybe you can do work on steep slopes in some areas and not in others,’” Brooks said. “Targeted management really works — 80% of the erosion often happens from 20% of the land.”

The online WEPPcloud tool was born when Brooks brought in a computer modeler, Roger Lew with U of I’s College of Art and Architecture, as part of the Lake Tahoe project to incorporate the model into an online watershed interface with access to global satellite imagery, soil maps and other data.

“We found as we developed this tool that we can use it to make these assessments and actually predict different scenarios,” Brooks said.

BEAR teams

Forest Service Burned Area Emergency Response (BEAR) teams that are assigned to develop post-fire response plans for the federal agency have also relied on WEPPcloud to guide their decisions. Brooks and Dobre work with Rocky Mountain Research Station staff to offer regular WEPPcloud training sessions for BEAR teams.

They further refined the tool to help the city of Ketchum compare the risk of sediment from forest thinning projects with the potential consequences of a large fire.

In 2019, they collaborated with scientists at Washington State University, University of Nevada Reno and the Rocky Mountain Research Station and submitted a proposal to NASA, seeking funding to further develop the tool to evaluate the risk of wildfires affecting communities’ surface water supplies in the Western Cascades. NASA officials initially turned down the project but asked the collaborative team to resubmit their application and ultimately funded it following the 2020 Labor Day fires, which burned over 800,000 acres and destroyed thousands of homes and businesses throughout the Western Cascades.

With the current model, it takes users just five minutes to run simulations and generate maps that formerly required months of work.

“Our tools are going to get better and better the more we work with communities that are going through this,” Brooks said. “If some community in the state of Idaho is affected by a fire, we’re going to be better prepared because we went through these projects.”

Lew continues to aid in technical upgrades to WEPPcloud, which is administered on a university server and is free for the public to use. In addition to helping cities and agencies such as the Forest Service, WEPPcloud has been broadly used by students, scientists and officials from throughout the world. The software now has more than 4,700 registered users in nearly every U.S. state, as well as several locations in Canada, South America and Europe.