Synchronised vs successive surveys

[Jérôme Couvat]

Dear all,

We finished our boat-based line transect survey for cetaceans in the Lesser Antilles 2 weeks ago and we are planning the next one in automn for the wet season. We have 2 survey areas (Northern Islands and Martinique-Guadeloupe, about 100 NM from each other) with one boat in each, surveying the areas in sync.

For next automn chances are that we won't be able to mobilise 2 boats, meaning that areas will have to be covered one after the other (each area takes about three weeks to cover).

How much of a problem will that be in terms of analysis and interpretation? Covering both areas at the same time reduces the probability of seeing the same animals in both areas but since the survey lasts for three weeks, animals have time to move from one area to the other and be sighted twice anyway.

Although synchronised surveys will always be best, how much could successive surveys affect the robustness of the results?

Thank you.

Regards,

[Eric Rexstad]

Jérôme

As you note, synchronisation will present the fewest difficulties for your work.  The question "how many difficulties will be caused by lack of synchrony" is difficult to answer.  The answer depends upon what you are trying to address with your surveys.

Do you wish to make temporal inference separately for the two areas?  Are you trying to make spatial inference between the areas over time?  Is there sufficient effort and adequate detections within each area that they can be "self-sufficient" or do the data need to be pooled across areas to obtain sufficient detections for fitting detection functions?

I don't think there is a simple answer regarding the threat to the robustness of the results.  But the questions above form a sort of 'checklist' I would use to formulate my ideas on the subject.

--
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/3db79382-c2af-41a3-a12f-e62b81c974aen%40googlegroups.com.

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

[Jérôme Couvat]

Thank you Eric for your feedback.

Unfortunately it seems that animal density in our study areas is quite low so it is very likely that I will need to pool the data across areas to have the enough detections for fitting detection functions. Therefore I cannot afford to analyse areas separately.

Comparing data across seasons would be good enough for now.

Jerome