Possible overestimation of population size?

[Adolfo Garcia]

Hello, 

It is a pleasure to greet you in this productive distance sampling support community.

I am new to distance sampling and doing some initial trials. I am very interested to know if there is a risk of overestimation of the population, for example, when traveling a line transect "out and back" (walking a transect in a "round trip" fashion) on the same day, as part of a protocol that consists of a monthly visit (round trip for a certain number of trails) for 6 months. Given the short time between these replicas, is it likely that the same individuals will be registered on more than one occasion, resulting in an overestimation of the population size and therefore of the population density?

I understand that surveying a transect in this way implies doubling the distance, and therefore, doubling the surveyed area when calculating the density, but does this eliminate the risk of over estimation?

Thank you

Adolfo G.

[Eric Rexstad]

Greetings Adolfo

While your survey design is not the best, it does not produce an overestimate of population density.  Distance sampling estimation used the encounter rate (number of detections per unit of survey effort) in the estimation of density.  View the standard density estimation formula this way:

The middle term on the right hand side is encounter rate. 

Pretend there is one animal within the covered region of your transect, which is 1km long.  You detect that animal on your outbound leg of your walk: encounter rate from that walk would be 1 detection per kilometer.  Pretend you also detect that animal on your return leg: encounter rate for the round trip is 2 detections per 2 kilometers.

Two issues in your design are of greater concern for the integrity of your work:

• by walking the same transect twice, you are not increasing spatial replication of your sampling.  It is spatial replication that strengthens your inference when making inference from the area you sampled to the entire study area; more spatial replicates is better
• you mention walking trails.  This also compromises the strength of your inference because trails are unlikely to be representative of the study area.  Trails are not randomly distributed on the landscape.  In addition, the animals you are studying know of these trails and are either attracted or repelled by them.  In either case, bias in your estimates can result.

If possible, try not to place your transects on trails and attempt to increase the spatial replication to make your inference stronger.

---

From: distance...@googlegroups.com <distance...@googlegroups.com> on behalf of Adolfo Garcia <addie...@gmail.com>
Sent: 12 October 2021 01:43
To: distance-sampling <distance...@googlegroups.com>
Subject: {Suspected Spam} [distance-sampling] Possible overestimation of population size?

 

--
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/c638011b-725b-43b3-90c2-db162715c1b0n%40googlegroups.com.

[Adolfo Garcia]

Dear Eric Rexstad
Thank you for such a quick and informative answer. I'm so glad I came here.