Bigfork Eagle: Success of stream restoration projects go largely unknown

Dollars spent on restoring America's rivers are increasing exponentially, but does anyone know how effective river restoration projects really are?

Finding the answer to that question prompted Duke University Assistant Professor Emily Bernhardt to participate in a survey study. The results of that study were presented July 27 in a lecture titled "Measuring, managing and restoring freshwater ecosystem services" as part of the University of Montana Biological Station Summer Seminar series Thursday evenings at Yellow Bay.

Despite the fact that river systems represent just one percent of all freshwater on earth, Bernhardt said 90 percent of U.S. rivers have been strongly impacted by channel manipulation and fragmentation, dams, reservoirs, diversions or irrigation. In addition, 20 percent of freshwater fishes are threatened or extinct and freshwater species represent 47 percent of all endangered species in the U.S.

Changes to river systems occur in large part due to an increase in the number of people in an area, Bernhardt said. Agriculture, high density housing and impervious surfaces such as road, parking lots and rooftops increases the amount of nutrients, particularly nitrates, phosphates and petroleum products entering rivers. Where water once percolated slowly into soil, impervious surfaces send it rushing off a landscape into stream channels and eroding them. Macro-invertebrates and other aquatic species in those streams are reduced. Slow moving streams with a diversity of substrate habitats, such as deep pools, slow moving water and riffles become uniform. Streamside riparian vegetation and wetlands are affected when channel modifications disconnect streams from their floodplains and wetland habitats.

Bernhardt said a number of stream restoration practices are common throughout the country. They include bank stabilization by planting trees, reconnecting channelized streams to adjacent wetlands, creating stormwater retention ponds to capture stormwater runoff coming off developments with impervious surfaces, channel reconfiguration, dam removal and using hard materials such as rock to armor stream banks.

As part of the ongoing, 40-year-old National River Restoration Science Synthesis project, Bernhardt's study was a collaborative effort conducted in North Carolina between ecologists, morphologists and hydrologists focused on determining whether stream restoration projects were effective and successful. Finding that answer meant determining whether individuals actively monitored projects after completion. Bernhardt's findings suggested that most did not monitor projects, thus the project was largely unable to determine individual project successes.

Despite more than 37,000 restoration projects in just seven regions across the U.S., each costing an average of $380,000, Bernhardt said there is no federal program requiring record keeping of restoration projects. Preliminary assessment of the 37,000 projects determined that fewer than 10 percent contained records indicating intentional monitoring of project effectiveness. Bernhardt said most projects were funded by a combination of state, federal, nonprofit organization and private landowner money.

Bernhardt and her associates created a sub-sample from the large number of restoration projects in order to survey project managers to assess project successes. The team conducted telephone interviews with 317 project managers. It found less than half of the projects had a stated goal, as most projects were funded by funds allotted for streambank stabilization. Bernhardt said improving water quality and wildlife habitat were the most common stated goals, but said they were often a guise. Of the 317 projects, only 33 percent evaluated project success using monitoring data, and 32 percent conducted before and after restoration work monitoring. Less than 30 percent had information on how big the project was in terms of feet or stream miles.

Of the 317 project managers interviewed, Bernhardt said 63 percent said their project was completely successful, 30 percent said the project was partially successful, five percent said it was hard to tell and two percent said the project was not at all successful. However, because 65 percent of "successful" reports were based on photographs that were easy to manipulate, such as taking pictures at different seasons in the year, the team determined the rate of success likely is not as high as reported.

The study concluded successful projects had a few things in common. The projects differed little in cost, size or number of funders, and each displayed a higher level of public involvement and were more likely to have an advisory committee overseeing them. Therefore, the more accountability the project faced the more likely it was to be successful.

In conclusion, Bernhardt said she would like to see a portion of a project's funds allocated for assessment and long-term monitoring. Statistical data of water quality and subsequent invertebrate and aquatic species populations, stream substrate and instream habitat types, nutrients and other factors determining stream health is necessary to determine project success and to justify expenditures. In addition, using a successful project as a guide; having a goal of ecologically improving the stream; using natural, self-sustaining channel improvements rather than hard armoring with concrete or rip-rap; and ensuring more good than harm is done to the stream area additional criteria Bernhardt said should be considered in any stream restoration project.

For more information on river restoration across the U.S. and the National River Restoration Science Synthesis project, visit www.nrrss.umd.edu.

The final FLBS Summer Series lecture is Thursday, Aug. 3 in the biological station's Elrod Lecture Hall. That lecture is titled "Interpretations of climate change for the northern Rocky Mountains."