Latest update: 5 January 2010
TRENDS IN NUMBERS OF BREEDING BIRDS
IN ORANGE, CHATHAM, AND DURHAM COUNTIES, NC
1999 -- 2009

Click on the links below to see graphs (charts) of the changes in the numbers of birds on the
Mini Breeding Bird Surveys in Orange, Chatham, and Durham Counties in the past decade.

R. Haven Wiley and all NCMBBS participants (2010).
http://www.unc.edu/~rhwiley/mbbs/trends/

These graphs show the total number of individuals of each species recorded on all routes in one county in each year.

Each species has a four-letter code (usually the first two letters of its first name and the first two letters of the its second name (for instance, EABL for Eastern Bluebird).

Some species show trends (either upward or downward) in their numbers over the period of the surveys.   A straight line on a graph indicates a trend that is statistically significant (not likely to be an erroneous conclusion) or almost statistically significant.

** A double asterisk in the upper right corner of a graph indicates a highly significant trend (with a probability of error less than 1%).

* A single asterisk in the upper right corner indicates a significant trend (with a probability of error less than 5% but greater than 1%).

A line but no asterisk indicates a possible trend (with a probability of error less than 10% but greater than 5%).

For more information about the N. C. MBBS, see the
MBBS homepage

For a list of the species (with four-letter codes), see
MBBS birds

Please send suggestions to Haven Wiley!

Some details . . .

The lines for trends were calculated with procedure lm{stats} and the graphs were prepared with plot{graphics} in R 2.9.2 for Mac (GUI 1.29) (The R Foundation for Statistical Computing 2009).

The data are available in a text (ASCII) file as a tab-delimited table which can be read into a data frame in R at MBBS data

and the R script is available in a text (ASCII) file at MBBS analysis

Although the volunteers who conducted these surveys did a great job of covering nearly each route every year, there were a few gaps, when a route was not surveyed one year.   Because the numbers of routes in each county were not high (8-14 routes per county), these missed routes could have a noticeable effect on the numbers of each species recorded that year.   To reduce this effect, the missing values were filled with the average number of each species recorded on that route during years when it was not missed.   This procedure would slightly reduce any trend in the numbers of each species.   The numbers of routes missed during the survey were small (1 in Orange County, 2 in Chatham County, and 1 in Durham County, each in one year only).

There is a graph for every species except the rare ones (those species encountered on the surveys fewer than 5 times on average each year in each county).   These birds were too scarce to analyze.

For information about reading these graphs, see the sidebar to the left.

Notice the downward trends in Chimney Swifts, Red-eyed Vireos, Barn Swallows, and Wood Thrushes.

There are upward trends in Fish Crows in Chatham and Durham Counties.

Among the species holding their own, despite nationwide concerns about their populations, are Eastern Meadowlark and many (but not all) warblers and tanagers.

Check the graphs for some surprises!

Orange County 1999-2009

Chatham County 2000-2009

Durham County 2002-2009

For information about how these surveys are conducted, see the Mini Breeding Bird Survey home page.

In summary, the surveys consist of randomly located routes in each county (12 in Orange, 14 in Chatham, 8 in Durham, the smallest of the three counties).   Observers stop to count birds every half mile along randomly located 10-mile routes on secondary roads.   Although these surveys do not detect all the birds in a county, they do provide a reasonaby dense random sampling of the birds.   They thus are a basis for evaluating trends in numbers of birds.   Because the surveys take place on mornings in May and June, they focus on breeding birds.   Because they are made from roads in the first couple of hours after sunrise, they detect some species more reliably than others.   For instance, nocturnal birds (owls and whip-poor-wills), soaring birds (most hawks and vultures), and birds of deep forest are no doubt underestimated.   Nevertheless, even these birds are sampled consistently from year to year.