Portal Forecasts 1/6/2017

This is a forecast made for the upcoming January 6 - 8 rodent trapping at the long-term Portal research site in southern Arizona. Four models were used to achieve forecasts with 90% prediction intervals. There are two forecast tables. One is for the site as a whole (total means across all plots, all treatment types), named "All". The other is for the 4 long-term Control plots only ("Controls").
Two naive models using the defaults in the forecast package in R were used to create forecasts for the total granivore captures (species="Total") over the next 12 trapping periods, the first being tomorrow.
The "Forecast" modelforecast(abundances$total,h=12,level=0.95,BoxCox.lambda(0),allow.multiplicative.trend=T)
The "AutoArima" modelforecast(auto.arima(abundances$total,lambda = 0),h=12,level=0.95,fan=T)
Species-level predictions, as well as another total captures prediction, were made using the tscount package in R.
The "NegBinom Time Series" model, by species (species codes are the first letter of the genus and species)
model=tsglm(abundances[[s]],model=list(past_obs=1,past_mean=11),distr="nbinom")
predict(model,12,level=0.9) 
The "Poisson Environmental" model, by species (species codes are the first letter of the genus and species), using monthly minimum temperature, maximum tempurature, mean temperature, total precipitation from our on-site weather station, and NDVI from a compiled LANDSAT MODIS estimate. Model selection was based on AIC and performed for each species.
model=tsglm(abundances[[s]],model=list(past_obs=1,past_mean=11),distr="poisson",xreg=weather)
predict(model,12,level=0.9,newdata=weatherforecast)