Detection variable of soil type

questions concerning analysis/theory using program PRESENCE

Detection variable of soil type

Postby dmh1454 » Wed Nov 13, 2019 2:11 pm

I am conducting research on the occupancy of semi-aquatic mammals via sign survey. For a detection rate, I am looking at the soil type any tracks are found in for each replicate. However, because I didn't always find tracks (and some tracks were in snow) I am unsure how I could use this data in my models.

Currently the data looks like this
Soil Type (Silt; Clay; Sand)
R1 R2 R3 R4 R5
- - - 50, 40, 10; 40, 30, 30
- - - - -
- - - - -
- - - - -
- - - - -
- - - - -
- - - - -
- - - - -
- - - - -
- - - - -
- - - 50, 50, 0 -
- - - - -
40, 60, 0 - - - -
- - - - -
- - - - -
Frozen Ground - - - -

Should I put the all of my lack of data as 0's or should I give each soil type a different data set?
dmh1454
 
Posts: 11
Joined: Fri Aug 02, 2019 12:35 pm

Re: Detection variable of soil type

Postby jhines » Thu Nov 14, 2019 12:50 pm

Your data should contain 0's only if you visited a site and did not detect sign of the species on a survey. If you didn't visit the site, the data should be '-'.

If you're interested in detection probabilities which vary by soil type, you'll need to define your sites such that each site is of a particular soil type. So, you would have 3 times as many sites if all sites have all 3 soil types.
Then, you could use site covariates to model detection.

If each site consists of the 3 soil types, you could use the multi-method model. Then each survey would have 3 columns of data for each site. The first 3 columns would indicate detection of the species for survey 1 in each of the 3 soil types, the 2nd 3 columns for survey 2, ...etc.

If the numbers in the sample data are percentages of tracks observed in each of the soil types, I don't think you can use that to estimate detection probabilities for the 3 soil types, since you would need also have information about the proportion of area each soil type.
jhines
 
Posts: 599
Joined: Fri May 16, 2003 9:24 am
Location: Laurel, MD, USA


Return to analysis help

Who is online

Users browsing this forum: No registered users and 13 guests