Remote Nocturnal Counts

[John Kearney]

Hi All,
This forum is a great idea. Thanks for setting it up.

I have one issue that I don't expect to have resolved in the near future since I discussed it with the eBird staff at Cornell a couple of years ago via their NFC-listserve. That discussion led to the resolution of some issues but there remains a couple of critical ones for me.

I don't report my remote nocturnal acoustic monitoring data to eBird because of two main issues. The first is that counts have to be recorded on just one date. This means the calls/birds for one night have to be divided into two portions, one for before midnight and the other for after midnight and reported separately on two different dates. Second, the counts have to be further subdivided on those dates between the calls recorded between civil sunset and astronomical sunset and from astronomical sunset to midnight. Likewise on the morning portion on the following date from midnight to astronomical sunrise and from astronomical sunrise to civil sunrise. Making all these divisions is a lot of work. Even if there was a way to devise a spreadsheet that would calculate these divisions and could quickly export the data into eBird, it still seems to me as a dubious exercise since it does not make biological sense to divide a night migration into two separate dates. Since remote nocturnal monitoring requires a separate eBird account anyway, why not use a night date system rather than a day date system? I also feel the astronomical/civil night distinctions add a lot of latitudinal bias into the counts. Astronomical night is much shorter during peak spring migration in Nova Scotia compared to peak spring migration in Louisiana.

Hopefully there will be more people doing remote nocturnal monitoring (as well as remote diurnal monitoring) in the future, and I believe that eBird will have to become more user friendly if it wishes to tap into this new source of data.

Best regards,
John

[Paolo Matteucci]

yes the Nocturnal Flight Call Count protocol seems complicated, with the pm/am distinction and also the astronomical twilight requirement to quality for that protocol. But instead of not putting any data at all, couldn't one simply put 'stationary' as protocol and enter the actual start time + duration? this is obviously not ideal but i do no see a big problem with that either.

[John Kearney]

Thanks for the suggestion Paolo. There is a specific protocol required for night flight call data that can be found here: http://help.ebird.org/customer/portal/articles/1010492-entering-nocturnal-flight-call-counts.
John

[David Bell]

Hey John,

Depending on what format your detection data are in (excel spreadsheet?), it would likely be possible to break them into hourly checklists, similar to what official stationary counts (seawatch, lakewatch, and hawkwatch sites) are now doing. (Just noticed it mentions this in the NFC eBird protocol page) This would probably make more sense than arbitrarily splitting the night into 2 'halves', and make the data more useful for determining peak calling activity times, etc. You could then just 'bin' detection data by hour to avoid the 'midnight spanning' issue, and any partial hours would go in as such. Example:

Start at 9:28pm - one checklist from 21:28 to 21:59:59, one from 22:00:00 to 22:59:59, etc. Ending at 5:48am would give you 7 full-hour checklists and 2 partial checklists on either end. It'd take more coding to break up the twilights but it'd probably still be doable.

 It'd definitely be great to get some of this nocturnal count data in to the database!

David

[John Kearney]

This is definitely a step in the right direction. The flight calls in my Excel spreadsheets are already coded by hour and twilight period. It will be more of a problem for the bird estimates that are only coded for the entire night. Each flight call for each species during the night is given a number. Any calls estimated as belonging to the same bird are given the same number. For example if I have 150 Blackpoll Warbler calls in a night, there might be 110 unique numbers entered for that species during the night; thus, an estimate of 110 birds giving at total of 150 calls.So I need to find a formula for parsing and giving a count of these id numbers by hour and twilight period.
It will be important to find a data entry solution that doesn't substantially add to an already time consuming data processing regime.
Thanks David!
John