Memo

To: Northern Saw-whet Owl Banders

Date: 28 March 2000 (revised 9/19/00)

From: David F. Brinker

RE: Sex Criteria for Northern Saw-whet Owls

Since last autumn numerous people have asked me for the sexing criteria that I have developed for Northern Saw-whet Owls. While I had distributed older versions in the past, I was always distributing them "for bander’s information only" and had not intended for them to be used when submitting schedules to the Bird Banding Lab (BBL). In this respect I guess that I was taking the conservative scientist’s approach in the belief that widespread use of these criteria should not be promoted until I had published them in a peer reviewed journal. The growth of interest in banding migrant Northern Saw-whet Owls, when combined with the magnitude of this past autumn’s flight, has forced me to be less conservative. To facilitate continued growth of Project Owlnet and assist other banders I have pursued official use of these guidelines prior to publishing them.

Thus, I have sought, and received approval from the BBL to provide these guidelines for use by any interested banders. As part of this approval the BBL has asked, and I wholly concur, that anyone using the criteria include the following remark on banding schedules for all birds sexed with the draft criteria. The remark is: Sex determined using the wing-mass DF available from Project Owlnet. This will allow the BBL to identify owls sexed using these criteria in the unlikely event that future research proves these criteria inadequate. As an aside, one of Mary Gustafson’s comments was "… The difference that I see with the saw-whets is that banders are still using the worthless, useless wing chord cut-offs! …" The old wing chord criteria (Buckholtz et al. 1984) do not work; do not use them!

I’m not going to take the time now to describe the derivation of the criteria, nor defend their validity and accuracy. Suffice it to say that they are based upon a reasonably large, and growing sample, that they have been tested, and that the error rate is about 5% or less. You can either contact me privately or wait until I publish them to get more details. Most of the information necessary for identifying sex in Northern Saw-whet Owls comes from weight (mass). Wing chord is necessary to reduce the unknown region where females and males overlap (a zone of about 10-12 grams). Please measure wing chords carefully and maintain a true arc to the wing. I generally find the difference between wing chord and flattened wing measurements to be at least 5 mm and generally it runs 8-10 mm. Poorly measured wing chords will affect accuracy!

I’ve attached the latest draft of the criteria. If you have old copies, especially ones using the older 90% cut offs – throw them out and replace them with the attached table. I do not expect these to be modified or revised until I publish on this. I’ve also attached two figures. Figure 1 shows the distribution of all known sex Northern Saw-whet Owls in the sample by mass and wing chord. Figure 2 expands the lower end of the distribution to illustrate the demarcation of the overlap zone more clearly.

Figure 1 - click for a larger image	Figure 2 - click for a larger image

Click here for a more information

A few comments on the use of the criteria seem appropriate. When sexing owls, generally, you should only need to use the table for owls with masses from 79 to 92 grams. Any owl with a mass of 93 grams or over is a female and any owl with a mass of 78 grams or less is a male. So far, the lower limit on the overlap zone is set by several females with masses less than 86 grams (one as little as 80 grams). Low mass females, assumed to be near starving or otherwise very poor condition individuals, are the important individuals here and they set the lower limit of the unknown zone. There is no way to identify these individuals short of a blood test and genetic analysis to determine their sex. Extremely low mass females appear to be the main reason that the unknown zone is necessary. In the sample these owls are very infrequent and have only occurred during invasion years (1995 and 1999). As you apply these criteria be on the watch for small winged owls (chord of 130 mm or less) with masses of 85 grams or more. Likewise, on the other end of the distribution, expect long winged owls (wing chords of 140 or more) with masses of 85 grams or less to also be rare. I expect individuals of these sizes to relatively infrequent, but it would be useful to know how often they occur.

For banders using Band Manager I recommend setting the wing chord and weight ranges as follows:

These ranges will pass most of the owls that you assign sex to with only the occasional owl generating a heads up message from Band Manager. In most cases heads up notices will be for weights that are out of range rather than wing chords. Also, in older versions of Band Manager the age/sex verification will fail for virtually all owls that you assign sex to (if you band owls during February through July it will accept females, but not males). The BBL revised the verification table beginning with all updates of Band Manager supplied after mid April 2000. If you have not downloaded an update to Band Manager more recently than mid April 2000, I strongly suggest doing so before routinely assigning sex to Northern Saw-whet Owls.

If you have any questions about these criteria or their application, please do not hesitate to contact me at dbrinker@bellatlantic.net or (410) 744-3246.

Boundaries for assignment of sex in Northern Saw-whet Owls*

> 95% probability of correct sex for any individual.

*When using these criteria to sex owls as part of banding schedule preparation you must include the following remark for each owl. Sex determined using the wing-mass DF available from Project Owlnet.