What makes a bird "rare"? (1 Viewer)

Hello all, I'm living and Birding in India, and an avid eBirder. I often look at the handy target lists they can generate, and it's made me really curious how eBird, local reviewers, and rarity committees make the cutoff on the arbitrary distinction of a a bird is considered "rare."

In places like India where distribution is not as well known, I find this to be a particularly interesting question, but also in places where now is known. For instance, does a know how ABA codes are determined?

Any input or discussion is welcome!

In Britain:-

"The starting point for defining a national rarity is either 100 or fewer individuals in the most recent ten-year period, or ten or fewer individuals in at least seven of the last ten years. Any taxon that meets, or is close to, either of these criteria, will then be considered for inclusion as a BBRC rarity."

Of course there is still some arbitrariness as taxon vary in ease of identification and ease of detection.

For ABA, there is this:

http://listing.aba.org/checklist-codes/

of course these apply to the entire ABA region (which in 2017 at some point will also include Hawaii). Some birds who are overall code 1 for the ABA may be extraordinary rare for a given state or time of year, and many codes are not updated as regularly as they maybe should be.

A species or subspecies could informally be considered 'rare' when it is very thinly widespread over a larger area, which can happen towards the limits of a bird's distribution, or when habitat has been considerably fragmented.

Many species are in serious decline, and so may be rare in areas where once they were common: Jerdon's Courser is an example in India.
MJB

It's also about context.
For example a Red Grouse would be stupendously rare in Salford docks but a few miles away in the Pennines, one might almost expect to see one or more individuals.

Is there an Indian site/forum/organisation that might be able to help our new friend specifically with the subcontinent context?

John

brd said:

Hello all, I'm living and Birding in India, and an avid eBirder. I often look at the handy target lists they can generate, and it's made me really curious how eBird, local reviewers, and rarity committees make the cutoff on the arbitrary distinction of a a bird is considered "rare."

In places like India where distribution is not as well known, I find this to be a particularly interesting question, but also in places where now is known. For instance, does a know how ABA codes are determined?

Any input or discussion is welcome!

Click to expand...

Here is an article on how eBird rarity filters are set and the eBird review process.

http://help.ebird.org/customer/port...ing-the-ebird-review-and-data-quality-process

In the US, what observations are set for review for the filter (i.e. are considered rare or unusual) is left to the discretion of a birder with extensive experience in the local area. Also, sometimes review is required where a species may be expected, but can be easily confused with another species.

Jim M. said:

Here is an article on how eBird rarity filters are set and the eBird review process.

http://help.ebird.org/customer/port...ing-the-ebird-review-and-data-quality-process

In the US, what observations are set for review for the filter (i.e. are considered rare or unusual) is left to the discretion of a birder with extensive experience in the local area. Also, sometimes review is required where a species may be expected, but can be easily confused with another species.

Click to expand...

I'm not that familiar with the differences in approach but the variability that I have found to date is quite interesting. Recently when inputting some historic records for Fair Isle in 1992, Yellow-browed Warbler and Barred Warbler did not come up as Rare which was understandable whereas Woodchat Shrike, Pallas's Warbler and Olive-backed Pipit did. However, I was quite surprised that Icterine Warbler came up as Rare and then really surprised that Barnacle Goose on 28th September did so (on date?).

For those unfamiliar with the system, there is a prompt for additional details where the species is flagged up as Rare either because of the species itself, the number seen or the time of year. It is quite good really and presumably it is a system that can build and develop over time. The Reviewer can then choose whether to contact you for additional information.

Photos are flagged up on searches if unconfirmed though some reviewers relatively promptly confirm on the system that they are happy with the identification (without contact) such as on the attached:-

https://ebird.org/ebird/view/checklist/S33114731

Edit - trying to get used to the Breeding Codes. I put the bird below down as 'Carrying Nesting Material'. :-O

All the best

Paul

brd said:

Hello all, I'm living and Birding in India, and an avid eBirder. I often look at the handy target lists they can generate, and it's made me really curious how eBird, local reviewers, and rarity committees make the cutoff on the arbitrary distinction of a a bird is considered "rare."

In places like India where distribution is not as well known, I find this to be a particularly interesting question, but also in places where now is known. For instance, does a know how ABA codes are determined?

Any input or discussion is welcome!

Click to expand...

The eBird software shows RARE in red letters whenever a species is triggering the filter set by the reviewer. You can see a flock of 13 Bar-headed Geese, but if the filter is set to 12, the software sees it the same way as if you saw 13 Californian Condors or 1 Californian Condor (in India).

Now there are all kinds of reasons why the filter is set lower for a particular species:

-a species may be out of its range (wind-blown migrant) so you are not expected to see any; if you have seen it, great, it is a rarity; but you have to provide evidence such as clear photos in order for it to be validated by the committee;

-a species was split into two, and you are trying to enter the other new species that now lives only in Indonesia, not the one which lives in India; this problem appears when someone is using an old field guide book, made before the taxonomic changes.

-two species are very difficult to tell apart, and while experienced birders make great effort to carefully scrutinize each photo pixel by pixel, and discuss it in committees and forums, some new birders may just assume it is the more common species and enter it in eBird as such. Setting the filter values low causes all records of this species to be caught by filter and the reviewer will see them.

-as already said, if there are separate filters for a mountain area and a nearby lowland, the montane species recorded in the lowland marsh will be flagged and a coastal species on mountain top will be flagged as well. If there are no separate filters, the value for a marginal species may be set lower so inexperienced eBirders will not report it unnoticed by the reviewer. I have set the filter low for several mountain-peak species. By now I have already discussed the situation with the few eBirders who live in mountains and they know that they can report their neighborhood bird every day and I will validate it under "Known to be at location", while birders from the lowlands will have to provide some evidence if they claim they saw it.

-you just saw a larger flock than expected. This is also OK, but the filter will stay low and your record will be accepted if you have evidence (or if it is just 22 birds instead of 20 etc.; if you saw 450 birds instead of 30 you need much more evidence).
etc.

. . .
Outside of eBird, birds are rated according to IUCN criteria on level of threat. The handbook for these criteria is several pages long (and complicated and tedious for the general birder; you would need results from various studies in order to assess a species according to each criterion) and special committees on national level meet periodically to review each species. They also send their results to international committees so we get a joined list such as the Red List of Birds of Europe.
It is quite possible to see a large flock of a species high on the threat list. Maybe it just happens to be a flocking species, but the number of such flocks is now much smaller than before as wintering or breeding grounds are destroyed in part of its range. One such species is Common Pochard. We all see the large flock of pochards every time we get to the bank of Danube in December but that is not indicative of its global rarity status.

Thank you all for the interesting feedback. The ABA and BBRC guidelines were specifically what I was looking for: a definition reconciling the vagueness of the word "rare" with the precision of numbers.

And I understand that ebird filters are down to county level in many places, and also account for number and season, but the ebird link provided still didn't answer my main ebird question, which is how reviewers are expected to define the filters. In a place like the us, that's easier, as there are records committees and local definitions of rare. But if you're a reviewer for a less documented place, is rare just a feeling in your gut? Reviewers in these places are not expected to get the filters right all the time, of course, so I understand when I see one bird that is listed as rare and another on the checklist that would be much more rare, I'm just curious what those guidelines are.

Also please excuse my typos. Phone posting...

brd said:

Thank you all for the interesting feedback. The ABA and BBRC guidelines were specifically what I was looking for: a definition reconciling the vagueness of the word "rare" with the precision of numbers.

And I understand that ebird filters are down to county level in many places, and also account for number and season, but the ebird link provided still didn't answer my main ebird question, which is how reviewers are expected to define the filters. In a place like the us, that's easier, as there are records committees and local definitions of rare. But if you're a reviewer for a less documented place, is rare just a feeling in your gut? Reviewers in these places are not expected to get the filters right all the time, of course, so I understand when I see one bird that is listed as rare and another on the checklist that would be much more rare, I'm just curious what those guidelines are.

Also please excuse my typos. Phone posting...

Click to expand...

Reviewers for eBird (I'm one) are basically just using their personal knowledge combined with eBird data to set filters. If neither of those things exist, then yes the filters may be pretty much guesses or set overly broadly. If you think you know better, then email the reviewer, as I'm sure they'll be happy for the feedback!

As far as I know, there is no hard guidelines for how the filters should be set, and a lot is left up to the personal preference of the reviewer. It's a very difficult thing to figure out objectively.

As far as I know, there is no hard guidelines for how the filters should be set, and a lot is left up to the personal preference of the reviewer. It's a very difficult thing to figure out objectively.

Click to expand...

Agree with this. Also as stated above, notice that numbers will trigger the same response, so that a species expected in numbers of 5-10 will trigger the "Rare" flag is you see one more than the number set in the filter. So it is the observation of that number and not the species per see that is rare in such an instance. Additionally notice that filters are set at different numbers at different times of the year and that not all filters are refines to the point that others are. Finally notice that, as stated already, a less common species can artificially be set with a lower filter number if a more numerous species is difficult to separate in the field (if that is what the reviewers want).

Final remark: some areas have a community of reviewers, other areas have only one.

Niels

Also note that the eBird filters are a fixed number threshold. They don't take into account if you've been counting for 5 minutes or 5 hours, 50 meters or 50 kilometers - it's the same threshold for all.

njlarsen said:

Final remark: some areas have a community of reviewers, other areas have only one.

Click to expand...

And I think some have none.

I agree with Zheljko on the use of the alternate reference of IUCN Red List. In this case, 'rarest' is Critically Endangered (Extinct in the Wild). Generally at the next level of Critically Endangered bird numbers are less than 250 (as mostly mature adults).

The IUCN definitions are for a species despite geographic spread so the birds may be 'rarer' more so in one country than other locations - hence the need to define endemics vs transients with registers like eBirder. From a birding perspective, low numbers of transients can be considered rare. This is a definition more so for the location than the species itself but it is still important for birder's Life Lists!

Some bird populations have always been known in low, steady numbers and therefore are a high risk to changing conditions but until this occurs are considered a lower risk of decline.

The reason for my comments here is that I find the word 'rare' is used too often to describe birds without substantiation / definition (especially in news media). I think birders/general public need metrics to interpret the severity of the risk to species. IUCN fulfils the need and other systems should also adhere to a standard criteria.

Andrew Esposito said:

The reason for my comments here is that I find the word 'rare' is used too often to describe birds without substantiation / definition (especially in news media). I think birders/general public need metrics to interpret the severity of the risk to species. IUCN fulfils the need and other systems should also adhere to a standard criteria.

Click to expand...

The trickiest situation is when the species is Vulnerable (VU) as it may be something that everyone has seen, so they may not *believe* that it is "rare" and then question the credibility of the whole "rare species" idea. One example is Green Woodpecker, which breeds in every urban forest in our city, so it is seen and heard by everyone going for a weekend picnic. However, it needs mature trees for nesting (which are rare and after each storm or insect infestation we are left with fewer such trees) and there is generally just that one pair in a given picnic area that everyone has seen.