Hi all – hopefully this is a suitable place to ask these questions
Based on this post
(https://groups.google.com/d/msg/unmarked/izJU61k9hKI/sBNr9I-tnWYJ)
I have been reading about the models described in Chandler and Royle 2013 and Ramsey 2015. Have there been any further developments of these models since 2013/2015? Have they been more thoroughly tested/validated? I have not yet found other papers that use or develop these models further; a few review papers that mention them but no real studies as far as I can find.
I’m also hoping to get some study design advice in two situations, one regarding coyotes in a very urban environment (the Bronx) and another with white-tailed deer in a suburban forest preserve. In both cases the C&R and/or Ramsey models are attractive to us in terms of adapting our current camera surveys to (attempt to) actually measure abundance. What scale is appropriate for these models, in terms of both population size and area? Also, how would one interpret the resulting activity centers when looking at social or herding animals?
Additionally, how do these models handle social or family-grouped species, where animals often are photographed in groups of 3 or more? As far as I can tell, the model does not discriminate between, eg, a single photograph/observation of 3 deer vs. 3 observations of a single deer. Seems to me then that an activity center in this situation would be an activity center for a group or family, not just one animal, and therefore density ought to be higher than what the model might produce. Or the model results should be interpreted as # of social groups, not number of individuals.
If you are curious for more detail on what we’re hoping to do:
1. The deer study is probably a simpler situation so I’ll start there. We have been using a camera trap-based method of estimating deer densities based on an updated version of this protocol (http://onlinelibrary.wiley.com/doi/10.1002/wsb.64/abstract) to measure the impact of ongoing deer management. I’d like to try C&R’s model this year and, if possible, on our prior years’ datasets and compare the two methods.
The area we are looking at is approximately 5 sq. km and in the past we’ve used 20 or so cameras. I can up this to 30+ cameras now, if that would enable us to use C&R’s 2013 model to greater precision. This is much smaller than the simulated 15x15 grid used in the 2013 paper, but given the area we’re interested in, 30 cams would represent a grid of 300m, 30% of an assumed 0.3km2 home range for suburban wt deer. Is this (~3-5km2) an appropriate study area for these models, or is it too small?
2. The coyote question involves measuring abundance of coyotes across a handful of urban parks of various sizes. Each park, as best we can tell with camera observations and some genetic sampling, has one breeding pair/family group with a few-to-several transient animals that come and go. There are likely never more than ~10 coyotes in each park (but I could be wrong; thus why I’d like to try these models). The parks range around 400 – 800 ha (1,000 to 2000 acres) but some potential study areas are as small as 200ha. We’re looking this summer to place cameras at high densities in a few of these occupied parks to, hopefully, measure abundance/density. In the larger parks a grid of 500m would be 20 – 35 cameras.
So again my question is whether this scenario is an appropriate one for these models – I suspect individual parks might be too small and/or the populations of coyotes are too small, or the number of cameras I can deploy would be too few even using very tight grids relative to the animals’ home ranges, given the sizes of our sites. I could run the entire borough as one big study area, but there would be substantial empty spaces between parks where I cannot place cameras on city streets.
Thanks for any advice you can send my way.
--
You received this message because you are subscribed to the Google Groups "unmarked" group.
To unsubscribe from this group and stop receiving emails from it, send an email to unmarked+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.