Need help on best choice of survey design for primates study

[Ambrose Bockarie Kanneh]

Dear Distance users,

I am new to using distance software, and thanks to this house for the support in terms of difficulties. 

I am undertaking a study to assess primates (with focus on Diana monkey and Black and White Colobus) abundance or encounter rate within a small Island of area 13.5 km2. I have generated two line transects survey designs but I am not sure which one is the most appropriate design to conduct the survey and that will make analysis more valid. I kindly need help on this and any other recommendation is highly welcomed. 

Below are the two designs across the study area. 

A)     The distance between one transects and the other is 500m and the length of each transect varies according to the shape of the study area (Island)

B)     The transects have equal lengths of 400m each (400m is the smallest length that can fit for the narrowest part of the island and so I decided to maintain it), and the distance between transects are still 500m

The idea is to survey each transect once from the most appropriate design. 

Thanks to whoever will come to my aid. 

Kind regard,

Ambrose Bockarie Kanneh. 

[Eric Rexstad]

Ambrose

Good to see you are making good progress designing your survey.  From a statistical point of view, I think either design would be fine.  Design A has 22 replicate transects (if I can count correctly); that is a sufficient number of replicate transects to assess encounter rate variance.

Design B has a greater number of replicates at the expense of introducing more transit effort between transects.  I don't feel Design B is any better or worse than Design A. 

Does each survey have roughly the same amount of total transect length; I'm guessing the answer is yes, so nothing to suggest one over the other.  Have you done a small pilot study, in which you can approximate the encounter rate for your focal species?  If you have done a small study, you can use the formula below to estimate the total amount of transect length needed to produce a measure of precision (coefficient of variation) that you desire.

--
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 view this discussion on the web visit https://groups.google.com/d/msgid/distance-sampling/CADuW%3Dch0q14%3D2HAv9DxMVw2T9%3Dh3sE4Kwj9EbJiDLAsn7ccMkw%40mail.gmail.com.

-- 
Eric Rexstad
Centre for Ecological and Environmental Modelling
University of St Andrews
St Andrews is a charity registered in Scotland SC013532

[Ambrose Bockarie Kanneh]

Dear Eric,

Very welcoming feedback from you. 

No pilot surveys have been conducted as of now. This is the first attempt for such study on the Island. For the first design, the total transect length is 38786.571 Meters and the second design is 29800.524 Meters, generated under the survey plan results.  

Thanks. 

Best regards,

Ambrose. 

[Eric Rexstad]

Ambrose

Ask yourself what level of precision in your abundance estimate would be necessary/useful in understanding your species.  You could then consider spending a day or two walking several of the proposed transects, say 3-5km, recording the total effort and the number of detections.  Use the information from that pilot survey in conjunction with the formula provided earlier to assess what precision you might expect from your proposed surveys of 30-40km of effort.

[Stephen Buckland]

I would tend to favour design A, as for narrow regions like this, there is a tendency to undersample near the edge of the region when using design B.  However, I would not expect to see big differences, and practical considerations may dictate which of the two you use.

 

Steve Buckland

To view this discussion on the web visit https://groups.google.com/d/msgid/distance-sampling/CADuW%3Dcjfv17wxp3ofnP0j5QstEL-bLfwG3_7_scOsQJL66rj8w%40mail.gmail.com.

[Ambrose Bockarie Kanneh]

Thank you Erick and Steve for all the recommendations, well noted.