Tuesday, September 03, 2013

Some analyses of 40-year bird count data in Central Massachusetts


Two years ago, in a post with a cheesy title, I’d written about seeing cardinals frequently. I continue to come across them: it is still a thrill to see a sharp movement of red in the branches of trees, and to hear the bird’s distinctive calls. Inspired by these sightings, I find myself now drawn to other birds, mammals and insects that live in the wooded areas of New England.

On May 27th this year, I visited a tract of land preserved by the Massachusetts Audubon Society near Worcester: the Wachusetts Meadow Wildlife Sanctuary. There, I met naturalist Joe Choiniere. We talked for a while. Joe knew well the birds in the area -- details of their range, habitat, prevalence, appearance, calls etc. As the current property manager, he, like others before him, organizes a bird count each year at the meadow. I told him about my interest in looking at the data -- an interest that has grown ever since I began to teach probability and statistics. 

Joe said could share the data with me.  To observe how the data was collected, he also invited me to attend the count this year, on June 9th. It was a somewhat cloudy Sunday morning, but visibility was good. Around 10-15 people, most of them expert birders, broke into groups and went on different trails that wound through the hill forests, ponds and marshes of the sanctuary. I followed one group along the main dirt road. Each person carried a paper listing of all species and marked the numbers as soon as something was spotted. Close to 500 species of birds have been observed in Massachusetts, so one has to be a keen and experienced observer of birds to get the identifications right.

Late morning, everyone returned to a room in the Visitor’s Center. A coordinator called out the name of each species; the birders around the table responded with their counts, and a tally for 2013 obtained. This process remains the same each year. Minutes of the discussion are maintained, and journal published each year presents the tally. Variables such as the number of observers, weather and temperature can of course change somewhat, but the trails on which the counts are made have mostly remained the same. Most importantly, despite year to year variations, the 50-year duration means that some general inferences on population levels of particular species can be made. 

Joe sent me an Excel Spreadsheet with the count for all species at the meadow from 1964 to 2003 (the data for 2003-2013 is also available, but has not yet been entered in Excel). I started working with the data recently and made some preliminary graphs. In this post, I will illustrate, with examples, whether or not the annual bird count at Wachusetts Meadow tallies with other statewide trends. 

So let’s take a look at the cardinal count for 40 years since 1964, when the count first started at the meadow. The x-axis is the year. The y-axis is the number of cardinals observed by birders at Wachusetts Meadow.  We notice the variability from year to year (except in the early years when no cardinals were seen). But it’s pretty clear that there is an increase in the number of cardinals seen, even though there are still years when none are observed.

Note that the image is only of the male cardinal (somewhat sexist, you could say!); the female is more gray than red, but females are of course part of the count.
What do the statewide counts compiled by the Massachusetts Audubon tell us on the prevalence of cardinals? There are three such counts: the Bird Breeding Atlas (conducted from 1974-1979 and again from 2007-2012); the Bird Breeding Survey (an annual effort on specific roadways); and the Christmas Bird Count (a 114 year tradition that is kept going by enthusiastic citizens).  Trends in the cardinal population based on these counts can be seen here. Notice that all counts suggest either a "likely" or "strong" increase in the number of cardinals. This agrees with the graph above, based on the Wachusetts Meadow annual count.


It is of course also possible that Wachusetts cardinal count may have only coincidentally agreed with other counts. So, for further validation, let’s look at the 40-year trend at Wachusetts for two other birds: the cliff swallow, and the house finch. 

I am fascinated by swallows: this July, I saw hundreds of rock swallows, their appearance similar to the above image, at the desolate ruins of the ancient city of Ani, along the river that separates Turkey from Armenia.
Sadly, for the cliff swallow we see a precipitous decline and counts have been zero for most of the 80s and 90s. This agrees with three statewide counts posted at the Massachusetts Audubon Society: they confirm a “strong decline” in numbers. The brief note says: “Today, the Cliff Swallow occupies less than half of the distribution it held in 1979. Loss of nesting structures, such as old barns and bridges, along with nesting competition from introduced House Sparrows are among the factors accounting for this restricted distribution.” 

For the house finch, we see no sightings at all until the 1980s. It turns out that the house finch was introduced into this region in the previous decade: “The introduced House Finch arrived in Massachusetts during the 1970s and never looked back. It can now be found living alongside humans over much of the state, and as a breeder it is nearly ubiquitous.” In the above figure, we don't notice high numbers in the years leading up to 2003, but the early years validate well with the available knowledge on house finches.

In both these cases, the Wachusetts Meadow trends again seem to match reasonably well with other statewide trends. Still, this is just preliminary evidence. Birders may be influenced by what they hear from others in their community or what they read in journals; during counts, they might unconsciously seek for a particular species or ignore others, thus introducing some “unnatural” variation. Yet, there is no way to completely escape such biases in a field study.

What impressed me most was the perseverance of anonymous individuals participating to keep such censuses alive all over the country. The comprehensive bird counts listed in the Massachusetts Audubon Society, all depend on the efforts of such individuals; the Christmas Bird Count has been going on for 114 years!