Date: 12/19/12 4:29 am
From: danny bystrak <dbystrak...>
Subject: Re: [MDBirding] KLZ files [google earth] for count circles


Rick and all,

Actually there is an issue with probably all of the CBC count centers.� For many years I was the "official" (volunteer) approver of all new CBC circles for Audubon.� This was back in the days when maps were on paper and determining center coordinates was challanging in most cases.� I had access to all the 7 1/2 minute topographic maps at work, so I would do my best to get them as accurate as possible.� But the problem was not one of accuracy, it was one of precision.� Audubon only wanted the coordinates to the nearest minute of latitude and longitude, so in most cases they were off by as much as a half mile or so even if they were completely accurate.� When they put the data on the web, they used those 'nearest minute' coordinates to calculate the decimal degrees. The decimal degrees were calculated to as many as 4 decimal places (remember 'significant digits?'), giving the impression of great precision.� But because they were based on the
original rounded minutes, they rarely pinpoint you at the actual center.� This is very obvious when the Annapolis, Bowie and Triadelphia circles are plotted.� We positioned the Bowie count to exactly touch the other two circles, but not overlap.� When you plot the three, there is overlap.� Also, the center of the Annapolis count is the south end of the rte 50 bidge over the�South River.� The coordinates at Audubon are .58 mi. WSW�of there. So�looking at the edge of a circle for what is in or out can be tricky.� Of course, it can be argued that a half mile is not a big deal, but many counts were positioned to capture specific areas at the edges.

Now that we have Google Earth and other such resources, it would probably be wise for count compilers to re-figure center coordinates to an accurate 4-decimal place figure and inform Audubon to correct them.� Audubon should also have a disclaimer about this and somehow�differentiate corrected centers from the "incorrect" ones.

Danny Bystrak
Bristol A A Co.


________________________________
From: Rick Hollis <xiboia...>
To: MdDC Birds <mdbirding...>
Sent: Tuesday, December 18, 2012 6:44 PM
Subject: [MDBirding] KLZ files [google earth] for count circles



The NAS has count circle files which are downloadable at�http://birds.audubon.org/faq/where-can-i-see-cbc-circles-map

http://www.nemesisbird.com/2012/12/mapping-the-christmas-bird-count/ has a discussion�

BUT there is a issue with some of the centers. The Iowa City Christmas Count Circle was off by 1.4 mile in the data release in 2011. �The files they released this year is even farther off, 1.67 miles.

I am not sure what the NAS chart options descried in the nemisisbird post above, gives the wrong center for our count.

Having said that, the Google Earth views are extraordinary -- far better than any old fashion maps. �If you locate your real center and draw your circle, you can find out whether bushes are in count circles -- the photos for our circle are within the past couple months. �My Mom's house in Balto has 2010, where as my Sis [Surgarloaf Circle] has 2011 images.

Rick Hollis
North Liberty IA & Arbutus MD--

--