Methods
Summary
All the methods applied in this project are completely harmless to animal communities.
- Camera trap. It has been proven very useful in detecting cryptic or rare species. 2 or 3 camera traps are deployed in every 0.04º grid cell to maximize detection and ensure methodological comparability with data from other areas. Cameras will remain in place for at least one full year to account for seasonal variations in detection probability.
- Local herders survey. Questionnaires will be conducted in a conversational setting, ensuring full respect for local culture and customs. The dialogue will take place in the participants' native language and will be free of judgment. Explicit permission will be obtained for all interviews, allowing the collected information to be used in this research. Survey responses will be transferred onto a local map, digitized, and incorporated into the model parameters.
Challenges
Camera trap deployment and interviews account with several intrinsic limitations.
1. The terrain is highly heterogeneous, and many locations will be suboptimal for Himalayan wolves or inaccessible for camera trap placement and animal detection. To address this limitation, we have stratified our study area to ensure an optimal camera trap distribution, maintaining the desired camera density and maximizing coverage.
2. As this is a human-dominated, touristic landscape, we anticipate a certain degree of camera loss. Theft and camera malfunctions are the most common factors affecting deployment efficiency.
3. The Yartsa Gunbu (caterpillar fungus) collection season introduces a temporary data gap from May 1st to the end of the month. The high influx of collectors in the area increases the risk of camera loss. To mitigate this, we will temporarily retrieve the cameras during this period to minimize potential losses.
4. Interview limitations: Some individuals may hold personal biases or simply not wish to participate in the survey. In such cases, no pressure will be applied, and their decision will be fully respected. Ensuring a voluntary and respectful approach is fundamental to the integrity of our research.
Pre Analysis Plan
Camera trap data will be processed using the WildEye Traptagger AI platform, which will facilitate the transfer of animal detections into operational matrices and numerical datasets.
A single-season modeling approach will be applied using the "unmarked" package in R. This analysis will yield two key outputs: (1) Himalayan wolf occupancy levels and (2) detection probability for each pixel. By fitting the model, we will identify the most significant factors influencing these two aspects.
Our fundamental hypothesis is that in human-dominated landscapes, human touristic activities shape Himalayan wolf occupancy patterns. To test this hypothesis, we will compare our results from the Annapurna Conservation Area with our existing data from Humla, Nepal, a non-human-dominated landscape.
Protocols
Browse the protocols that are part of the experimental methods.
