Weird collated index and lower IC

[SarahVray]

Hi @RetoSchmucki, that's me again ^^" For several species in some years, I obtain a very low collated index and/or a very low CI.

• The low collated indices always correspond to a year with 0 specimen recorded, so maybe this very low value is normal?
• The low CIs always correspond to years with very few occurrences, not enough to build a flight curve, but I am wondering why the lower CI is so low whereas the upper CI looks normal?

Here is the example of A. agestis: Aricia agestis_collated index.csv Here are the counts per year:

Here is another example, P. c-album: Polygonia c-album_collated index.csv In 2012, the collated index looks OK but the lower CI is very low. In 2017, the collated index is very low although it is the year with the highest counts: But there is clearly an issue with the computed flight curve for 2017 (see issue #21).

Thank you very much for your help!

[RetoSchmucki]

Hi @SarahVray The CI is computed by resampling your sites. If you have a species that is only present in very few sites, there is a chance that the resampling (with replacement) might not include the site where it is observed, resulting in a collated index of "0". If this happens more than 5% of the time, the "0" will be in your 95%CI. Unfortunately, this is the limit of the bootstrap CI, and it does not perform well with species on rare sites.

[SarahVray]

Hi @RetoSchmucki, I see, it makes sense. Thank you for your explanation!