Seasonal clustering bias

14 views
Skip to first unread message

Adolfo Garcia

unread,
Oct 19, 2021, 5:24:43 PM10/19/21
to distance-sampling
Hi! Again I am communicating in this excellent support group regarding a possible source of bias. 

A study will be carried out during breeding season, in which the species exhibits the behavior of moving in groups (actually, perching in groups, so estimating cluster size or the distance from a line to each individual would't be a problem). In your opinion, is this seasonal clustering problematic in terms of over estimation of densities?

Thank you very much 

Stephen Buckland

unread,
Oct 20, 2021, 8:42:40 AM10/20/21
to Adolfo Garcia, distance-sampling

No reason why that would cause problems, although you can expect lower precision when animals form groups than when they are spread through the study area – and consquently, you may need to have greater survey effort when animals are in groups.  It is also more impotant to have an adequate number of lines or points in the design.

 

Steve Buckland

--
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/ea9ad191-88c6-4a00-a187-1d64e8cb01c4n%40googlegroups.com.

Adolfo Garcia

unread,
Oct 20, 2021, 10:26:20 AM10/20/21
to distance-sampling
Professor Buckland, 

Thank you very much for your answer. I just have a final question. Given that a sufficient number of observations is needed in order to estimate the detection function. Why would a group of animals (in case samplers are sufficient, and assumptions are met), which are naturally distributed in a given area, lower the precision of the estimates?

Stephen Buckland

unread,
Oct 20, 2021, 11:08:17 AM10/20/21
to Adolfo Garcia, distance-sampling

The number of independent detections is number of groups detected when animals are in groups, and number of animals detected when they are not in groups.  So the sample size (in terms of number of independent detections) is much smaller when the animals form groups.  Small sample size equates to low precision.  In terms of number of animals, encounter rate (number of animals per unit distance travelled for line transects, or per point for point transects) is a lot more variable when animals are in groups.

 

Steve

Reply all
Reply to author
Forward
0 new messages