Recipient Organization
UNIVERSITY OF MONTANA
COLLEGE OF FORESTRY AND CONSERVATION
MISSOULA,MT 59812
Performing Department
College of Forestry and Conservation
Non Technical Summary
The removal of beaver from the landscape has been a major contributor to increased stream channel incision and loss of connection between the stream channel and its floodplain. These changes have been linked to lowering of the water table, less water storage, decreased base flows, warmer water temperatures, and reduced channel complexity. Throughout the US, stream restoration practitioners are increasingly using Beaver Dam Analogs (BDAs) and beaver re-establishment as a technique for restoring degraded streams. Consistent monitoring and the full range of impacts to the aquatic ecosystem are not well understood, resulting recent calls for research as the practice is ahead of the scienceThis study stems out of this call to better understand the magnitude of water storage and quantify other ecosystem effects that with influence the riparian zone plant community and fish habitat associated with these activities. In addition, concerns about whether these structures and subsequent beaver establishment on a novel landscape with lower base flows, warmer temperatures, and ubiquitous nonnative species might result in more harm than good in certain watersheds for the native biota. This research will address specific concerns about increased surface area from impoundments increasing temperatures to a problematic range for native fish, dams restricting fish passage to upstream spawning habitat and cool water refugia, and impoundments altering habitat that promotes nonnative species over native species. If tradeoffs are present, we may make different management decisions about where these stream restoration practices are most appropriate.
Animal Health Component
80%
Research Effort Categories
Basic
20%
Applied
80%
Developmental
0%
Goals / Objectives
Many human activities, including overgrazing, loss of riparian vegetation, and the removal of beaver, have exacerbated the occurrence of stream channel incision disconnecting the channel from its floodplain. The effect of channel incision may include lowering of the water table, decreased water storage and base flows, warmer water temperatures, and reduced channel complexity (Bouwes et al. 2016, Goldfarb 2018). Beaver Dam Analogs (hereafter, BDAs) are a rapidly growing stream restoration technique to improve eroded systems. In addition hydrologists are promoting beaver dam analogs to promote water storage on the landscape longer in the season, ideally increasing late season flows. Increasing hydrologic connectivity (laterally and vertically) has been shown to lead to increased riparian vegetation as well as increases in mesic vegetation resilience to climatic variability (Silverman et al. 2019). The installation of BDAs and encouragement of natural beaver activity is becoming more common throughout the United States, but consistent monitoring and the full range of impacts to the aquatic ecosystem are not well understood, resulting recent calls for research as the practice is ahead of the science (Pilliod et al. 2018; Lautz et al. 2019).This uncertainty of the effects of BDAs has resulted in a controversy in the Rocky Mountain region regarding the benefits and costs of BDAs and subsequent beaver reestablishment. Many biologists are concerned that the positive impacts of BDAs and beaver reestablishment for restoring eroded and incised streams and increasing water storage will be minor compared to the negative potential impacts on aquatic life, specifically native fish. Native fish require cold, clear, connected complex habitat. There are benefits of BDAs and beaver in producing complex habitat for fishes. But with increased years of warmer summer time temperatures and more drought, there are concerns associated with BDAs and beaver ponds warming water temperatures beyond temperatures optimal for native fish, the dam structures could impede fish movement with less precipitation and likely lower flows, and finally this restoration technique and the reestablishment of beaver may promote nonnative fish species over native fish species.Our overarching goal of the research is to inform this debate through both through a replicated Before-After-Control-Impact study design, as well as, evaluation of broader landscape patterns. Goal 1: We will use a Before-After-Control-Impact study design to evaluate changes to water storage and riparian community, stream thermal profiles, organic matter retention, and instream habitat complexity across multiple watersheds in MT. We have just completed the "before" year of data collection and with UM undergraduate and graduate student volunteers, as well as the Clark Fork Coalition installed 26 BDA structures across three project sites in October 2019. Goal 2: Evaluate what characteristics of beaver dam analog structures and beaver dams lead to resistance to fish movement across a typical western MT fish community and seasonal flows. We do not expect most beaver dams to impact upstream migration to spawning grounds for spring-spawning (and spring-migrating) trout species as flows are typically high creating a more permeable riverscape. The largest impacts will likely be observed in summer time movements and to spawning habitat for fall-spawning fishes. Goal 3: Evaluate whether this stream restoration approach promotes nonnative fish species, specifically brook trout on the landscape in favor of native fish of conservation concern in Montana (and throughout the Rocky Mountain region). If stream restoration creates habitat characteristics (warm, pool habitat) that promote non-native fish over native fish in their headwater refuge habitat, we may be storing water but it will come at a cost of native fish distributions. If tradeoffs are present, we may make different management decisions about where these stream restoration practices are most appropriate.Our project also addresses the three primary goals of the The McIntire-Stennis Cooperative Forestry Research Program 2017 Strategic Plan. (Goal 1, through graduate education create the future generation) Specifically, Andrew Lahr, PhD candidate, is leading the project as a part of his dissertation. In the first year we have incorporated two senior thesis students into the project (one ESR student and one WBIO student). Funds in this project will go towards hiring an undergraduate field assistant. These students are also helping with the restoration itself, learning from USFS hydrologists & practitioners working with nonprofit organizations resulting in an enhancement of training for graduate students and future graduate students (through senior theses). (Goal 2, build and sustain strategic relationships) This project is currently being done in partnership with Lolo National Forest, Clark Fork Coalition, The Nature Conservancy, and with the help of Montana Wildlife and Parks. Having such a diverse partnership is helping build strategic relationships between scientists and practitioners as well as among various organizations. The current program's broad involvement is a great start to building and sustaining strategic relationships to improve the effectiveness and continue to build support for the project. (Goal 3, spread the work about the success of M-S) Working with these groups, we have already written outreach documents, given talks and tours, and finally A. Lahr will be required to communicate his research to both a scientific and a broader audience. This will provide the avenue for the success of the M-S Program to be communicated not just to scientists but across a broader audience. Finally, some of our sites are on private land resulting in regular communication and our NGO collaborators regularly work with private landowners ensuring that the results of the research and the success of the M-S program will get communicated to these stakeholders as well.
Project Methods
Methods:Goal 1: We will use a Before-After-Control-Impact study design to evaluate changes to water storage and riparian community, stream thermal profiles, and instream habitat complexity across multiple watersheds in MT. We have established paired sites across three watersheds in western MT (Teepee and NF Howard in the Lolo Creek watershed; Lost Horse and Lost Prairie in the Clearwater watershed; and Fish and Little Fish Creek in the Blackfoot watershed). We have just completed the "before" year of data collection and with UM undergraduate and graduate student volunteers, as well as the Clark Fork Coalition installed 26 BDA structures across three project sites in October 2019.In order to document any changes to flow and thermal regimes, we will establish a gaging station downstream of the anticipated BDA installation (and appropriate location in the control reaches) and evaluate temperature profiles at the BDA site and longitudinally along the tributaries before and after installation and compare them to nearby control streams. We will establish gaging stations in appropriate, stable locations, measure the cross-section, set-up pressure transducers, and develop a rating curve for monitoring discharge at each site. Twelve to fifteen temperature loggers will be longitudinally distributed per site to evaluate both the reach scale changes in temperature associated with BDAs as well as changes downstream. The reach level data (with BDA placement) will be analyzed for changes to growing degree days and maximum temperatures. We will measure the temporal and spatial extent of thermal buffering by comparing changes in the maximum and minimum temperatures longitudinally along the watershed for the next 2 to 3 years post installation.To examine the ecosystem level effects of BDAs, we will quantify particulate organic matter, concentration and loads of organic carbon, and dissolved organic matter quality in all streams. All streams will be sampled at an upstream and downstream location, which will then fall above and below BDAs after their installation in those streams receiving BDAs. We will sample coarse and fine woody debris yearly at all sites using the line-intersect method. Stove-pipe core sampling will be used to collect non-woody coarse particulate organic matter (CPOM, > 1 mm), algal biomass, and fine particulate organic matter (FPOM; 0.45 µm - 1 mm). Subsamples will be measured for chlorophyll a, with the remainder to be dried, ground, and analyzed on a CHN element analyzer to determine carbon and nitrogen concentrations. We will sample water for dissolved organic carbon and nutrients monthly. Water samples will be collected and filtered in the field through a 0.45 µm filter. Samples for dissolved organic matter quality measurement will be brought back to the lab on ice and kept at 4 ºC prior to analysis. Dissolved organic carbon will be analyzed on a TOC analyzer using persulfate oxidation. Dissolved organic matter quality will be assessed using both UV-Vis absorbance and fluorescence excitation emission matrix measurements. From absorbance and fluorescence spectra, we can examine the relative abundance of different light absorbing and fluorescing components of dissolved organic matter that can give insights into molecular weight, chemical complexity, and provide insight into the degree to which it has been processed. Load will be calculated for dissolved organic carbon by taking their concentrations and discharge, and modeling them using USGS-R/loadflex each year.Finally, each summer at each site we will complete a habitat survey during summertime base flow. We will complete modified R1/R4 fish habitat surveys (Overton et al.1997) including quantification of habitat types, large woody debris (same collection described above), bank erosion, and substrate. This will allow us to link the restoration action, BDAs to changes in riparian and fish habitat responses. These variables will continue to be measured through the next 2 to 3 years of post-installation monitoring.Goal 2: Fish will be sampled annually using backpack electrofishing beginning July 2019 in 6-8 creeks in the Lolo Creek drainage. Streams will vary in beaver dam abundance as well as whether restoration action (BDAs) took place. Every summer each stream will be longitudinally sampled over the course of one to two days. All trout captured will be identified to species, measured for length, and tagged with a unique ID PIT tag for mark recapture purposes and movement tracking. Mobile RFID backpack surveys will be used to monitor fish movement throughout study reaches bi-weekly from May-November each year from 2019-2022. Mobile RFID backpack detections will be georeferenced.Dam (natural or beaver analog dams) height, width, presence of side channel, path of dominant downstream flow, and position (river kilometer) will be determined for each beaver dam on each study system. Stream flow will be monitored by creating a rating curve at the downstream end of each study reach. Rating curves will be developed by taking bi-weekly flow measurements using a Marsh McBirney electromagnetic velocimeter and matching measurements up to continuous water depth data collected on onset HOBO water depth sensors. Stream temperature will also be measured using OnSet HOBO water temperature loggers placed at regularly intervals along the study reach length.Fish movement will be characterized by mean individual linear movement, average movement rate, and probability of dam passage. We will examine how probability of dam passage is influenced by spawning season (yes/no), stream temperature, discharge, dam height and porosity, presence of side channels, location of dam in reference to the stream miles and other dams in the system, and fish length.Goal 3: To address some of the concerns regarding whether or not beaver facilitate the expansion of non-native trout range, we will be working with Taylor Wilcox and Mike Young at the USFS National Genomic Center to identify approximately 380 sites across 20-30 watersheds that have been longitudinally sampled approximately 1 km apart for eDNA (already in the eDNA atlas). Each site will be analyzed for presence of brook and brown trout. GIS and remote sensed data on basin characteristics previously determined to be predictive of brook and brown trout such as slope, watershed size (stream width), as well as measure distance from beaver impoundments. I will investigate the relationship between presence/absence of these non-native salmonids, the presence of impoundments and environmental variables such as temperature, flow, valley width and slope. Results from this model should reveal whether or not presence of beaver impoundments are stronger predictors for brook trout and brown trout invasion when compared other more well-known environmental drivers such as slope, flow (derived from models), and average temperatures (derived from regional datasets). We will also specifically examine if mid and high elevation beaver impoundments result in these nonnative species being more broadly distributed, higher in the watershed. Once general landscape characteristics are considered, if detections of non-native fish appear higher in watersheds with more mid and high elevation beaver activity, this would indicate that beaver may be a vector for expanded range of non-native trout.