Beaver Dam Analog Monitoring

In conjunction with The Nature Conservancy-Montana (TNC), Montana Conservation Corps (MCC), and World Wildlife Fund (WWF), our team is creating a protocol to evaluate the impacts of beaver dam analogs that have been installed in central Montana. This protocol will leverage GIS and remote sensing technologies, challenging our team to identify metrics that can be detected remotely and provide valuable information regarding the natural resources impacted. Beaver dam analogs (BDAs) are human-made structures that are placed in streams to mimic the ecological and physical impacts of a true beaver dam. The western partners have collaborated with local ranches to install these analogs to increase herbaceous productivity and drought resilience to benefit ecosystem function and forage availability for livestock. The ability to keep forage production elevated through periods of drought is becoming more important as the frequency of extreme weather events increases. We will provide a user-friendly and cost-efficient method to understand the impacts of BDAs across multiple years, providing insight into overall drought resilience in the area after the installation of BDAs. 

The developed tool utilizes Google Earth Engine to analyze satellite images derived from Sentinel data. This tool outputs computations that allow the user to understand how NDVI and NDWI change on a temporal scale. The next user will be able to simply input a shapefile of their area of interest and the tool in turn will provide analysis of how that surfaced has changed from 2019-2022. Below you will find an example taken from the "Beaver Trap" site. This will show a graph describing climate patterns occurring at this specific area of interest. Please see the video at the bottom of this section for a demonstration on how to operate this tool.

Demonstrational Materials 

r.watershed Tutorial

QGIS Tool Tutorial

Download the QGIS tool here: 

Google Earth Engine Tool Tutorial

The Google Earth Engine application can be found here

Recorded meeting with UCROSS, TNC, WWF, and MCC can be found here

As this research continues for the foreseeable future, our team has compiled a list of potential recommendations. Please see below.

• If timing and funding permits, it would be invaluable to monitor an area of interest before a beaver dam analog is installed. Similarly to the BMR Monitoring study, being able to understand the underlying environmental variables before a treatment is given would allow the team to output more significant results. Having the ability to directly compare factors such as groundwater elevation, surface water temperature, and flow rate across a variety of years would add validity to the data, as well as support arguments for the impact of beaver dam analogs. 

• If funding were available, it would be useful to collect our own satellite imagery data in order to build more complex vegetation and wetness indices.  The current resolution of Sentinel data is helpful for identifying large scale trends, however if areas of interest cover a small parcel of land, it would be difficult to understand patterns without a finer resolution. Therefore, if drone data collection techniques become feasible, the ability to draw more intricate conclusions from the data would greatly increase. 

• If resources were available, it would be useful to have trained volunteers or staff periodically go into the field to identify present flora and fauna. While satellite data can capture overall greenness, it doesn't capture the variety of diversity present in these landscapes. Having folks on the ground to understand how species presence is changing over a multi-year period would offer new insight. 

• Delineating the area of interest is a vital first step to take for establishing a baseline and any following monitoring practices. Depending on resource availability, it would be useful to have a standardized method for drawing AOI boundaries. The watershed tool developed by Jason Hanlon would be an excellent option to utilize as it bases its calculations on nearing watersheds, thus indicating water availability. Whichever method chosen for delineation, our team recommends that it is standardized for all future research in the area for the sake of replicability and scientific validity.

Our team explored a variety of analytical techniques to implement during this year long research period. For varying reasons, these options were not pursued, though they should be kept in mind for future research projects depending on changes in data availability during the upcoming years. 

1) StreamStats: StreamStats is a very interesting website that generates a report including basin characteristics and scenario flows based upon a delineated point in any stream. There is historical data available for each state. While this would potentially be a useful resource, there is very little data available for eastern Montana, therefore the statistics at this time are not significant for determining water availability over time. However, each county is being updated from month to month, so hopefully next year there would be data to use for analysis.

2) Creating our own NDVI: This would be an interesting option to explore as it would allow us to have a high resolution vegetation index of our study area. The reason we are not pursuing this option this year is because we do not have high quality satellite data to work with. If in the future we are able to use drones in our study areas to capture our own satellite data, then creating our own NDVI would be practical. 

3) Creating Topographic Wetness Index: Similarly to NDVI, creating our own Topographic Wetness Index (TWI) would allow the team to understand, at a higher resolution, where water in accumulating based on the landscapes elevation. Again, if we were able to use drones to capture specific spectral data, then this option would be practical. 

For further information on how to utilize and build these options, please see Explored Options Document.

To explore the results of these explored options, please see our original google site. 

The section below details a literature review conducted by the students in September and October of 2022. Findings from each study provide useful implementation ideas which could be potentially utilize for future research purposes. Many of these studies have access to specific equipment, educated volunteers for species identification, and access to drones for specialized spectral analyses. To see the original brainstorm document, click here.

Valley Country Long Range Plan | Montana NRCS Glasgow Field Office

1) Soil type and topography are important to consider for draining. Different types of soil have varying capacities for holding water. Understanding how water moves and where it is likely to hold will be useful for rancher's grazing patterns. 

2) The Valley Plan warned to be cognizant of pipelines in the study area, as there is potential of them contaminating waterways. Before installing analog structures, it would be useful to understand where pipelines exist so the researcher can intentionally avoid those areas.

3) Be aware of any externalities that may create barriers to mesic restoration. Ex. local airport, roadways that block crossing, noise pollution. These barriers deter organisms from moving within the area of interest. This study found that fencing has been linked with soil quality degradation and low plant productivity. 

4) It is useful to monitor at-risk flora to see if dams rebound their populations. Below are at risk species in Glasgow Montana.

• • Alkali Marsh Aster

  • Hot Spring Phacelia

  • Scarlet Ammannia

  • Silver Bladderpod

  • Lead Plant

  • Slender branched popcorn flower

  • River Bulrush

  • Platte Cinquefoil

  • Chaffweed

  • Dwarf woolly-head

  • Bracless blazingstar

  • Scribner’s Ragwort

  • Slim pod Venus’ looking glass

5) It is useful to monitor at-risk fauna to see if dams rebound their populations. Below are at risk species in Glasgow Montana.

• Pallid Sturgeon

• Whooping Crane

• Interior Least Tern

• Piping Plove

• Northern Long-Eared Bat

• Red Knot

Valley Plan Document

Hydrologic Monitoring Recommendations for Beaver-Mimicry Stream Restoration | Montana Department of Natural Resources and Conservation

A study conducted at Alkali Creek monitored sites two months prior to installing the beaver dam analogs. By doing so, they were able to create a baseline describing the environment. This was incredibly useful for any future monitoring purposes as they were able to directly compare progress. Once the installed the beaver dam analogs, the team monitored variables such as: groundwater elevation (piezometer), surface water temperature, stream flows, stream stage, and riparian vegetation.

Below is a photo depicting their study area. They have four control regions: upstream, downstream, reference, and treatment. The upstream control measure groundwater elevation, and surface water temperature. The reference area included surface-water stations and meansures groundwater elevation. The treatment region had installed 6 beaver dam analogs and simultaneously measured groundwater elevation as well as surface water temperature. Finally, the downstream control measured both groundwater elevation and surface water temperature. 

This study was useful to understand how they delineated their area of interest and what types of variables would be significant to measure.

BMR Monitoring Recommendations Document

Beaver Trap Example

The example shown below demonstrates an area of interest that has been analyzed using the BDA geospatial tool. Below is a region called Beaver Trap. The black lined polygon demonstrates the boundaries of this AOI, which is composed of 29 acres. The red dots represent the installed beaver dam analogs. These structures were installed of October 2021, therefore our team has not been able to find any significant patterns of their impact on the landscape. That process will most likely take years to conclude whether or not the beaver dam analogs are effecting water availability. Instead, this example is simply to show the types of analyses that can be completed in the future. The satellite images shown below represent NDVI (Normalized Difference Vegetation Index) to understand the abundance of vegetation in the region. This NDVI data comes from Sentinel and has been clipped to only include the growing season, which has been defined as May 15th to September 15th. 

While our team will have to wait a few growing seasons to determine whether or not the BDA's are having an impact in the designated areas of interest, this tool still offers much potential for understanding the baseline environmental conditions, which will benefit monitoring.