Population/Abundance Estimate from Site Observation Percentage?

[Noel Jonathan Ellis]

Given data on bird counts with many species from several sites, 10 points at each site and numbers observed at different distance bands, and trying to decide how much time to allocate to it.

I'm wondering if population or abundance estimates could be done by inputting the % of points at each site that each species was seen at alongside the total number within each distance band? 

e.g. Site 1: 

Bird 1 seen at 30% of points

•  2 in a 0-50m band
• 15 in a 50-100m band 
• 7 in a 100-150m band

Bird 2 at 70% of points .... etc.

Or does it need to be disagregated to the observations at each point, which would of course take longer to format and input.

e.g. Site 1 Bird 1 Point 1

• 0 in 0-50m band
• 3 in 50-100m band
• 0 in 100-150m band

Site 1 Bird 1 Point 2

• 0 in 0-50m band
• 0 in 50-100m band
• 0 in 100-150m band

Site 1 Bird 1 Point 3

• o in 0-50m band
• 2 in 50-100m band
• 1 in 100-150m band ...

etc. for all points and birds at each site.

Sorry if this is obvious but I've been looking at some documentation and don't have a clear answer. Mostly self-taught with data analysis so I might be missing key background knowledge to figure this out without trial and error.

[Jeffrey Royle]

hi Noel,

 I think your question is a bit unclear, and it may not be related to unmarked , so you may not get much out of this group.

 That said, if you have bird point count distance sampling data then you could use the distsamp() or gdistsamp() functions to estimate population density for each species. I suggest taking a look at the help files and examples for that function, in addition to the relevant material in the Applied Hierarchical Modeling book Applied Hierarchical Modeling in Ecology: Analysis of Distribution, Abundan... - Google Books  where we have some worked examples and other information about the models.

regards

andy

--
*** Three hierarchical modeling email lists ***
(1) unmarked (this list): for questions specific to the R package unmarked
(2) SCR: for design and Bayesian or non-bayesian analysis of spatial capture-recapture
(3) HMecology: for everything else, especially material covered in the books by Royle & Dorazio (2008), Kéry & Schaub (2012), Kéry & Royle (2016, 2021) and Schaub & Kéry (2022)
---
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+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/unmarked/ace6584d-b773-4fce-8288-bc3725e4fa14n%40googlegroups.com.

[Noel Jonathan Ellis]

I have been using gdistsamp actually, looking through the main documentation and examples from wherever I can find. 

What I could find has the data entered for each bird and point individually, which would take more time. I was hoping that I could enter site data (summary of all points at a site) directly and still get population estimates, Given that the question itself seems unclear, it seems like the answer is no.

Thanks for the help anyway.

- Regards

Noel

You received this message because you are subscribed to a topic in the Google Groups "unmarked" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/unmarked/WCOCBx4gwdM/unsubscribe.
To unsubscribe from this group and all its topics, send an email to unmarked+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/unmarked/CAE%2BwgF3AwhqQau84_i1mKFvdD-6XhNw9XCghMx8BeQ5L9VnBcw%40mail.gmail.com.

[Jeffrey Royle]

hi Noel,

   The data structure for unmarked should have all sites included as a single matrix for a given (single) species.  So for distance sampling data, you should have a matrix structured like this:

             dist1   dist2 .... distD

 site1       x11  x12 ....  x1D

 site2       x21  x22 ...   x2D

   ....

   ...

  siteM      xM1  xM2 ... xMD

where the columns of this matrix (dist1, dist2, ..., distD) represent the distance categories you collected data in, and  xij = frequency of detections at site i in distance class j.

Once you have this matrix for each species, which is easy to create in R most of the time, then you can implement the models directly using distsamp or gdistamp.

If that's not clear, then I suspect you are having trouble manipulating your observation-level spreadsheet into these summaries perhaps?  If that's the case you should provide some specific info on the structure of your data and someone on here might be able to give advice.

regards

andy

To view this discussion on the web visit https://groups.google.com/d/msgid/unmarked/CACCkcYfVhQUaP1qCokVaY%3DKnb1su9YpT-WAo0VURd-VhO8R86A%40mail.gmail.com.

[Noel Jonathan Ellis]

Thanks Andy. Seems like I won't be getting around entering all the data. 

What I was hoping to do was essentially compress all the sites in an area to one line of the matrix, and give the proportion of sites with detections. It would look something like this:

Area 1 

                dist1      dist2 ..... distD          Detection%

all sites     x87     x191         x18                  70

Area 2

                dist1       dist2        distD            Detection%

all sites    x13         x85           x6                    30%

But as I there is nowhere to add this additional column, its pretty clear that I need to actually input what was seen at each site. In the time it took to ask I've already got through most of the data either way.

- Noel