Bob Ringler wrote:

> Fred,
>    In Stewart & Robbins the extreme date of spring departure for Woodcock is listed as Apr. 11,
   a record from Frederick Co. in 1952.  This is the basis for setting 
the early safe date in mid April.  Since Woodcock continue to display 
long after this it was felt that safe date was not a problem.
>    The peak migration time appears to be mid March with some variation from year to year.  When Don Jewell was collecting window kills at the State Office Complex in Baltimore he found Woodcock on Mar. 12 and Mar. 21 in 1979.  When I was doing the same a few years later, and one of the Baltimore Peregrines was roosting there, I found the head of a Woodcock on the corner of Eutaw and Preston Streets on the morning of Mar. 15, 1983.  Though we stopped collecting specimens it was not unusual to see a dead Woodcock in March.  There are always stragglers in migration.  If you assume a bell-shaped abundance curve for pure migrants during the spring then it is reasonable to expect some migrants in April.  Since the Atlas safe dates were designed intentionally to be conservative for all species to eliminate these stragglers from consideration the safe date for Woodcock was set at April 15.
>    As to whether woodcock display in migration I offer my own experience as an example.  On Feb. 7, 1999 I heard a Woodcock displaying above my house in a suburban neighborhood.  This is the only time I have heard Woodcock here so I conclude that it would be a migrant.
> 
> Bob Ringler
> Eldersburg MD
> 
>  
> ---- "Frederick W. Fallon" <> wrote: 
> 
>>This question and Jim Stasz' reply touches on an important problem - how 
>>to map out the breeding distribution of woodcocks. This is especially 
>>relevant now, when so many of us are engaged in the MD Breeding Bird Atlas.
>>
>>So I asked several leading woodcock authorities throughout the US 
>>whether males display during migration, and whether migrants are still 
>>going through MD as late as the Atlas "safe date" of Apr 15. Replies 
>>came back all over the block, leaving me with the impression that the 
>>question is still very much open. (A general concensus, at least, 
>>emerged that Apr 15 was too strict a safe date and it has in fact been 
>>pulled back to Apr 10).
>>
>>Something that might shed some light on the question: birders often come 
>>across woodcock on the Mall, or downtown Baltimore, or other 
>>non-breeding locales. Has anyone ever heard them display in such places? 
>>And what is the latest date for such sightings?
>>
>>Fred Fallon
>>PG Coordinator, MD-DC Breeding Bird Atlas
>>Bowie MD
>>


Bob -

Thanks for the information. It is from actual data like this that the 
time profile of the migration can be better determined.

I know of one other instance (from the 30's) of "peenting" observed from 
a locale where the Woodcock could not have been breeding, so that's why 
I ask if there are any others. But I grant it must happen at least 
occasionally.

In the last several years of Osprey postings there have been several 
accounts of sightings on the mall and some from Baltimore, in addition 
to the ones you mention. Do you know if anyone has gathered these in one 
place?  Most were from March and none from April. So the single record 
in Stewart and Robbins for April seems a thin thread to hang the Apr 15 
safe date on.

There are rigorous statistical methods that can take a series of dates 
and, assuming a sharp cut-off, estimate when the cut-off is. Or, 
assuming a Normal distr, estimate when 90% (or some fraction) of the 
distr has accumulated. This is something I would do if I could find 
sufficient data.

Confounding the problem is the possibilty that migration dates have 
shifted earlier over time. I suspect this is so from my own records, but 
I need a bigger database to test the hypothesis. One difficulty is the 
increasing  number and sophistication of birders. Another is the 
practice of recording "early arrival" and "late departure" dates with 
none in between. So, e.g., a BLPW heard on May 1 is recorded but the 10 
heard on May 20 are not. This makes determining the distribution much 
less certain - tho' the problem is still amenable to statistical 
treatment. Perhaps in time the Cornell e-bird program, which takes 
everything thrown into it, will remedy that defect.

Meantime maybe Jim Stasz has the answer - wherever there's a consistent 
woodcock display site, just search over a 300' radius and find the nest. 
If you find it the problem is solved.

- Fred

-- 
Fred Fallon
PG Coordinator, MD-DC Breeding Bird Atlas
Bowie MD