inclusion of multi-year data for pre-/post-construction

77 views
Skip to first unread message

bbielfelt

unread,
Dec 3, 2014, 7:42:08 PM12/3/14
to distance...@googlegroups.com
Hello All,
 
I am analyzing raven density pre- and post-construction, but to do so, I need to estimate density using Program DISTANCE.  I am having trouble determining how to import the data for year.
 
Here is how sampling occurred:
 
point count surveys conducted 12 times per year at each location, 6 visits in the spring and 6 visits in the fall. 
conducted pre-construction in 2011 and post-construction in 2014.
 
I haven't decided if season is important should be pooled for each year or if the density should be estimated separately.  My thought would be to treat each season/year separately (i.e., spring 2011, fall 2011, spring 2014, fall 2014) so that any differences in the ability to detect ravens between seasons is accounted for.
 
My main question is whether this data belongs in the "region" or "sample/point transect" layer? 
 
Reading through the help topics on stratification and post-stratification for DISTANCE, it seems like the "region" is more for geographic data.  However, construction is altering the landscape, so I can envision a scenario where year is a function of "region" rather than the "sample". 
 
A follow-up question would be how to make sure DISTANCE recognizes that there are multiple visits to the same point transect in each year?  I know how to do this for when there is only 1 year of data, but not when there are multiple visits and multiple years.

Thank you in advance,
 
Brian J. Bielfelt

Eric Rexstad

unread,
Dec 4, 2014, 4:14:12 AM12/4/14
to bbielfelt, distance...@googlegroups.com
Brian:

My suggestion is that you take the simplest approach to the analysis, unless the simplest approach is flawed for some reason.

You are correct that "region" or stratum, often carries a geographical definition, but it is not necessary for that to be the case. Based upon the information you have provided, the simplest approach would be to enter your data with 4 strata that you identified (S11, F11, S14, F14) and estimate stratum-specific detection functions and density estimates.  I hope that with a study of ravens you do not have a shortage of sightings.

For each point transect in each stratum you associate a measure of effort; for points effort is measured in visits rather than in length of transect.  Hence you would enter "6" in the Survey Effort field for the points; if some points were not surveyed the full 6 times, you enter the number of times they were visited.  There are point count projects available as sample projects shipped with Distance; have a look at the "Songbird" or the "Amakihi" projects.
--
You received this message because you are subscribed to the Google Groups "distance-sampling" group.
To unsubscribe from this group and stop receiving emails from it, send an email to distance-sampl...@googlegroups.com.
To post to this group, send email to distance...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/distance-sampling/31b4bc59-54ac-4417-b056-4cb85457db1d%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

-- 
Eric Rexstad
Research Unit for Wildlife Population Assessment
Centre for Research into Ecological and Environmental Modelling
University of St. Andrews
St. Andrews Scotland KY16 9LZ
+44 (0)1334 461833
The University of St Andrews is a charity registered in Scotland : No SC013532
Reply all
Reply to author
Forward
0 new messages