This document provides a description of the data used in Penjor et al 2022. Contrasting effects of anthropogenic disturbance on the interaction among sympatric carnivores. Proceedings of the Royal Society B ##### GENERAL INFORMATION ##### 1. Title of Dataset: Data and BUGS code for the manuscript 'Contrasting effects of anthropogenic disturbance on the interaction among sympatric carnivores'. 2. Author Information i. Name: Ugyen Penjor Institution: Wildlife Conservation Research Unit, Department of Zoology, University of Oxford Address: Tubney House, Abingdon Road, Tubney, OX13 5QL AND Nature Conservation Division, Department of Forests and Park Services Ministry of Agriculture and Forests Taba, Thimphu, Bhutan Email: ugyenpenjor.bt@gmail.com 3. Geographic location of data collection: Bhutan 4. Funding sources that supported data collection: Royal Government of Bhutan, IDA-World Bank, WWF-Bhutan, and the Bhutan Foundation ##### SHARING/ACCESS INFORMATION ##### 1. Links to other data and R scripts: https://github.com/ugyenpenjor1/Multispecies-Interaction-Model. ##### METHODOLOGICAL INFORMATION ##### 1. Description of methods used for collection of data: We used camera traps to collect the data. A total of 1129 unbaited camera stations were established. We placed a pair of infrared motion-triggered camera traps in each grid cell along human or animal trails (except in absence of trails, when cameras were placed randomly) at a height of ~45 cm above the ground. For logistical convenience, the country was divided into two blocks and camera traps were deployed during the dry season. Data were processed in MS Excel spreadsheet and R programming language. We provide more details in the paper that used this dataset. 2. Methods for processing data: We used a multi-species occupancy model for two or more interacting species (Rota et al., 2016). We used detection/non-detection data to estimate margina and conditional occupancy probability while accounting to imperfect detection. We modelled occupancy probability as a function of human settlement density, forest cover, disturbance, river, slope and prey. Detection was modelled as a function of disturbance, effort and trail. 3. Instrument or software used: Dell Precision 3630 Tower, Intel(R) Core(TM) i7-8700K CPU @ 3.70GHz 3.70 GHz, 64 GB installed RAM, Windows 10 Pro version 20H2. R version 4.0.0, JAGS 4.3.0, R packages unmarked, jagsUI, loo. ##### DATA-SPECIFIC INFORMATION ##### ##### WORKSHEETS ##### Worksheet 1:covariates.csv 1. Number of cases/rows: Each row contains site-level attributes. There are 850 rows and 12 columns. 2. Variable List: site - site number and code trail - whether camera placed on (indicated as trail) or off-trail (indicated as no) block - study blocks (north abd south) effort - number of active camera trap days for each site (camera station) Disturb - mean daily encounter rates of humans and livestock (dogs, cattle, and horses) river - density of river within 4 km radius of camera station settlement - density of human settlement within the 4 km radius of camera station slope - focal mean of slope extracted from DEM forest - global tree cover data (from Hansen et al., 2013) muntjac - relative abundance of muntjac estimated by N-mixture model using camera trap data gaur - relative abundance of gaur estimated by N-mixture model using camera trap data serow - relative abundance of serow estimated by N-mixture model using camera trap data Data 1: species_detection_history.RData Each species has 850 rows and 5 columns site - site number and code (site identity) V1-V5 - number of sampling occasions (15-day binned into one sampling occasion) species - study species (i.e., tiger, leopard, dhole) R SCRIPTS Script 1: interaction_model_BUGS_code.txt BUGS (Bayesian analysis using Gibbs sampler) code for multispecies occupancy model for two or more interacting species.