ordering of transects for use with dht2 er_est = S2 or O2

16 views
Skip to first unread message

Janet Mossman

unread,
Feb 6, 2026, 1:22:47 PM (5 days ago) Feb 6
to distance-sampling
Thanks for the response to my previous question- I have a follow up question about the spatial ordering of sample.label required within dht2 to use er_est =S2 or O2. Our survey design was systematic, zigzag transects oriented within multiple spatial strata but with the same transect spacing. We used 9 iterations of the same survey design for our 9 surveys conducted within different different months of the year across 3 years. Again, our goal is to get seasonal abundance estimates for 3 species so we will be pooling surveys by season (not by year). I want to see if S2 or O2 would improve our abundance CI's to account for potential patchiness of animals between transects. If I have sample.label named by "surveyID_transectID" with unique numbering for each geographic strata (ie restart the transect numbering within each spatial strata), is this sufficient for dht2 to understand the "meaningful ordering" of transects required for these ER variance estimators? 
Thanks
Janet

Eric Rexstad

unread,
Feb 9, 2026, 6:08:36 AM (3 days ago) Feb 9
to Janet Mossman, distance-sampling
Janet

You are correct that the numbering scheme for your transects is fundamental to using the S2 or O2 encounter rate variance estimators. Information embedded in the sample.label is assumed to measure proximity of transects. I think the numbering scheme will achieve that proximity measure.

One note of caution, consider coding sample.label as a numeric rather than character (replace "_" with ".") as there is a sort performed on sample.label as part of the calculations. Also use leading zeros; e.g. 003.010 for the tenth transect in stratum 3.

From: distance...@googlegroups.com <distance...@googlegroups.com> on behalf of Janet Mossman <pacif...@gmail.com>
Sent: 06 February 2026 18:22
To: distance-sampling <distance...@googlegroups.com>
Subject: [distance-sampling] ordering of transects for use with dht2 er_est = S2 or O2
 
Thanks for the response to my previous question- I have a follow up question about the spatial ordering of sample.label required within dht2 to use er_est =S2 or O2. Our survey design was systematic, zigzag transects oriented within multiple spatial strata but with the same transect spacing. We used 9 iterations of the same survey design for our 9 surveys conducted within different different months of the year across 3 years. Again, our goal is to get seasonal abundance estimates for 3 species so we will be pooling surveys by season (not by year). I want to see if S2 or O2 would improve our abundance CI's to account for potential patchiness of animals between transects. If I have sample.label named by "surveyID_transectID" with unique numbering for each geographic strata (ie restart the transect numbering within each spatial strata), is this sufficient for dht2 to understand the "meaningful ordering" of transects required for these ER variance estimators? 
Thanks
Janet

--
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 visit https://groups.google.com/d/msgid/distance-sampling/5c7853bf-1f59-4a17-a428-2c2b41e87750n%40googlegroups.com.

Janet Mossman

unread,
Feb 10, 2026, 12:29:15 PM (yesterday) Feb 10
to distance-sampling
Thanks for your reply. Is a way to indicate that transects from separate surveys should not be associated with each other when using er_est=s2 pr 02? I haven't been able to find anything regarding this and it seems that there would be danger of dht2 accidentally pairing transects from different surveys that are not spatially close to each other just based on their sequential order by sample.label?
Thanks
Janet

Eric Rexstad

unread,
2:20 AM (17 hours ago) 2:20 AM
to Janet Mossman, distance-sampling
That's a good point Janet.  I thought having surveyID as part of your sample label would deal with this, but on second thought, that might not work. I believe these alternative encounter rate variance estimators were developed for the situation with a single stratum/survey. There will be some "edge" cases where the pairing will be spatially inaccurate; if you have lots of transects, those inaccurate pairings might not have much of an effect on the variance estimate.

Sent: 10 February 2026 17:29
To: distance-sampling <distance...@googlegroups.com>
Subject: Re: [distance-sampling] ordering of transects for use with dht2 er_est = S2 or O2
 
Reply all
Reply to author
Forward
0 new messages