Montana is the 4th largest state in the United States, covering more than 147,000 square miles (380,800 square km) and is larger than most countries. When caving groups and federal, state, tribal, and nongovernmental organizations became concerned about detecting and mitigating the threat of Pseudogymnoascus destructans, also known as White Nose Syndrome (WNS), these stakeholders met to do something about it.
Goal: Establish Baseline Information to Track WNS Advancement in the West
According to the US Fish and Wildlife Service (USFWS), WNS has killed an estimated 5.7M-6.7M bats in the eastern United States, and the fungus continues to advance west.
With the threat of White Nose Syndrome (WNS), in 2010 the groups in the state of Montana committed to an initial 4‐5 year (2011‐2016) collaborative statewide effort to assess year-round distribution and activity of bats. They wanted to establish baseline information on bat activity prior to WNS becoming present in their region. Funding for the purchase of detectors and other gear was provided through a variety of small grants available within agencies or organizations. Major funding for the management and analysis of the data came from Federal and State agencies including the Bureau of Land Management, US Fish and Wildlife Service, US Forest Service, the Montana Department of Environmental Quality, and Montana Fish, Wildlife, and Parks.
The Role of Song Meter Detector/Recorders
A key component of this project was the agreement to use automated bat recorders. These recorders were deployed and left unattended to detect and record bat calls. In this way, large amounts of bat acoustic data was acquired in often hard to reach places without a surveyor being present. The data was later collected for analysis.
Spearheading this effort was the Montana Natural Heritage Program whose staff analyzed most of the recordings, identifying the calls to species. The Song Meter Ultrasonic Bat Detectors from Wildlife Acoustics, Inc. were chosen for the project. "The Song Meters are a weatherproof detector that can be deployed for long time periods and were the most affordable detector along these lines." Explains Bryce Maxell, Program Coordinator. To date, there are over 75 Song Meters deployed for this project. The Map in Figure 2. illustrates deployment sites across Montana as well as some of the peripheral states. In selecting the deployment sites for the Song Meters, the group considered the following factors:
- Does the location fill a gap in the statewide acoustic monitoring scheme?
- Does the location provide information to inform local management decisions?
- Is roost habitat available in the nearby landscape, preferably year‐round?
- Is surface water available in the nearby landscape, preferably year‐round?
- Is there adequate solar exposure for charging a battery to power the detector/recorder?
- Is the site free of vegetation or other biotic and abiotic sources of ultrasonic noises (e.g. whitewater, electric lines, shrubs, immediate vicinity of bat roost)?
- Is the Song Meter at risk of damage from vandalism, cows, or other hazards?
Advice for Deploying Song Meters and Analyzing Data
For advice to other customers considering a wide-scale monitoring effort, Bryce recommends "that the deployment is done by only a small number of individuals and maybe an additional 20 individuals with a good checklist to help check the detectors to send in data cards and put fresh cards in the Song Meters. For processing and analyzing the data, having a small brain trust of about 4 persons is important because they can discuss issues and bounce ideas to develop a call determination rule set, supporting documentation, etc." To help process the data, Bryce's team chose to use two programs for species analysis, including Wildlife Acoustics' Kaleidoscope Pro software. All calls, however, were hand reviewed for confirmation of species ID.
Bryce confirms that, "The biggest issue is dealing with the volume of data that a network like this will produce, especially if you are going to combine it with weather data. You need to have very database savvy people that can write python or other code in order to create routines that will batch load and process the analysis outputs. Dealing with data this size on a file by file basis using a GUI approach is just too much. They will also need a very fast computer or set of computers. We often had 4 or 5 computers processing or running analyses around the clock."
Results to Date
One of the main goals of the deployment was to summarize monthly and nightly bat activity levels, as measured by the number of bat passes detected, regardless of species and then correlate the data with temperature and, where possible, wind speed and barometric pressure. As of September 2015, over 44,747 nights have been sampled, over 6M bat passes recorded and over 40,000 call sequences analyzed to species ID. By understanding where the bats are, where they are migrating and bat pass counts, the group can monitor the changes of bat activity as WNS penetrates west into the region.
There are other benefits as well to this monitoring program, including resource management plans and local project level decisions to guide timing of work associated with habitat alterations. As a result of the program data, the Montana Natural Heritage Program has some great information that the wind power industry can use to guide power generation at wind turbines to reduce or eliminate bat mortality. This will be important for implementing the United States Environmental Protection Agency's Clean Power Plan and Montana's implementation under that effort.
For the future, the Montana Natural Heritage Program still has over 2,500+ hours of files needing to be hand confirmed to individual species. The team is working on developing full reports for each individual long term deployment and will work on a publication for the larger regional dataset as a whole. There are more habitats to sample (e.g., alpine habitats) and still more management questions to be answered such as, What is bat activity in intact vs. insect killed vs. harvested vs. burned forest, or How does bat activity vary with distance from oil drilling pads. For now, the team has a lot to do and will continue the project.
*Thanks to Bryce Maxell and his collaboration for this case study as well as his documents, "Montana Bat and White‐Nose Syndrome Surveillance Plan and Protocols" and "Montana Bat Acoustic Surveillance Summary 20151215" which provided the majority of the content.