### Description of data and scripts for analyses in "Harvest and density-dependent predation drive long-term population decline in a northern ungulate"
### Authors: Robby R. Marrotte
### February 9th, 2022

I have set up the repositories so that anyone can run the code using the original data in the manuscript.
The Dryad repository contains the RData file (moose_data_dryad.RData) and the Zenodo repository contains two R scripts (run_jags_model.R and gompertz_jags.R). Download all 3 files and put them all in the same directory.

1) Data: moose_data_dryad.RData
This is a Rdata file that anyone can load into R using the load() function.
All Moose Aerial Survey (MAI), Hunter Survey data and Snow data were collected by the Ontario Ministry of Natural Resources and Forestry (https://www.ontario.ca/page/moose-population-management)
This R data has the following 4 objects: jags_data, params, wmus, years 

- jags_data is a list() of all the predictors used in the model. Description of each:

	-indic.mai: Indicates if there was an aerial survey or not. A 0 = there was survey, and 1 = there was not
	-mu.mai: Log transformation of the number of moose counted during an aerial survey
	-tau.mai: Precision of the number of moose counted during an aerial survey 
	-mu.mai.init: Log transformation of the number of moose counted during an aerial survey in 1999 
	-tau.mai.init: Precision of the number of moose counted during an aerial survey in 1999
	-Seen.moose: Number of moose seen by hunters        
	-Ndays.moose_s: Standardized number of days hunted
	-Snow_hunt_s: Standardized snow severity index        
	-deer_per_day_s: Standardized number of deer seen per day
	-predators.per.day_s: Standardized number of canids seen per day	
	-success.bull: The success rate for harvesting a bull for hunters that sent back their questionnaires       
	-success.cow: The success rate for harvesting a cow for hunters that sent back their questionnaires  
	-success.calf: The success rate for harvesting a calf for hunters that sent back their questionnaires         
	-miss_tags.bull: Number of hunters that were issued a bull tag but didn't send back their questionnaires
	-miss_tags.cow: Number of hunters that were issued a cow tag but didn't send back their questionnaires      
	-miss_tags.calf: Number of hunters that were issued a calf tag but didn't send back their questionnaires
	-kill.moose_Bull: Number of bulls harvested    
	-kill.moose_Cow: Number of cows harvested  
	-kill.moose_Calf: Number of calves harvested
	-kill.moose_Z: Number of moose harvested that hunters did not identify the sex or age     
	-nwmu: Number of Wildlife Management Units (WMUs). There were 55 out of 69 investigated
	-nyears: Number of years in the data. It started in 1999 and ended in 2018             
	-bear_density: Standardized bear density per 100 km^2
	-Missing_bear: Indicates if there was a bear density measurement or not in each WMU.

- params was the parameters that we tracked within the MCMC.

- wmus is the Wildlife Management Unit (WMU) identifiers. There were 55.

- years is the years investigated [1999-2018]

2) Script to run the model: run_jags_model.R 

This is the R script used to load the data, load the libraries, and run the MCMC. I included the original version of the jags command that we used in the manuscript. It takes about 4 days to run with 3 chains. I also included a version that uses the default values for n.chains, n.adapt, n.iter, n.burnin, and n.thin. This last one only takes a few minutes but is incorrect (Posterior does not converge). You'll need to run the MCMC with the original parameters to get convergence.

3) The jags code for the model likelihood: gompertz_jags.R

This last file is the jags code used for the likelihood of the Gompertz population model