bowhead.hmm.pars: Greenland bowhead whale availability HMM parameters.

In david-borchers/hmltm: Line transect estimation with hidden Markov availability models

Greenland bowhead whale availability HMM parameters.

Description

Hidden Markov model (HMM) parameter estimates for bowhead whale availability obtained from fitting HMMs (using library HiddenMarkov) to data from electronic depth-recording tags that were attached to eight bowhead whales from the West Greenland population. The tags generated 8 time series of durations between 2.6 and 53 hours, with depths recorded every second (see Laidre et al., 2007). Following previous practice (Heide-Jorgensen et al., 2007), animals were considered to be available for detection only when within 2 m of the surface. The time series were accordingly converted into binary availability time series and HMMs were fitted to these. This dataset was used in the analyses of Borchers et al. (2013) - albeit with survey data that had no forward distances.

Usage

bowhead.hmm.pars

Format

A list with the following three elements.

Pi:

a 2x2x8 array containg the 8 HMM transition probability matrices (one for each tagged whale). States can be interpreted as behavioural states, one of which being a state in which the animal is more likely to be available than when in the other state.

pm:

a 2x8 matrix containing the 8 vectors of state-dependent Bernoulli availability parameters. pm[1,i] is the probability of whale i being available given state 1 (the less available behavoural state), and pm[1,i] is the probability of whale i being available given state 1 (the more available behavoural state).

delta:

a 2x8 matrix containing the stationary distribution of Pi for each whale. So delta[1,i] is the probability that whale i is in behavioural state 1 when observation starts, and delta[2,i] is the probability that it is in behavioural state 2 when observation starts.

@seealso The depth time series data are in object bowhead.depths and the binary presence/absence data obtained from these are in object bowhead.adat.

Source

Laidre et al. (2007).

References

Borchers, D.L., Zucchini, W., Heide-Jorgenssen, M.P., Canadas, A. and Langrock, R. 2013. Using hidden Markov models to deal with availability bias on line transect surveys. Biometrics.

Heide-Jørgensen, M. P., Laidre, K., Borchers, D. L., Samarrra,F., and Stern, H. 2007. Increasing abundance of bowhead whales in west greenland. Biology Letters 3, 577–580.

Laidre, K., Heide-Jørgensen, M. P., and Nielsen, T. 2007. Role of bowhead whale as a predator in West Rreenland. Marine Ecology Progress Series 346, 285–297.

Examples

 data(bowhead.hmm.pars)

david-borchers/hmltm documentation built on Oct. 29, 2023, 9:07 p.m.