Thursday, December 22, 2011


A weak but negative relationship exists between primary forest productivity and the number of bird species that inhabit forests.

A stronger, positive winter relationship exists between temperature (greater elevation and latitude equals lower average temperature) and the number of species that inhabit forests.

The U. S. Geological Survey is collaborating with Bird Conservation Research, Inc. as we continue to analyze the year-round distributions of southern New England’s forest birds. USGS is providing technical assistance in the field of satellite imagery.

We have already demonstrated that a significant winter relationship exists between forest bird diversity and winter energy availability (see above) in that our region’s birds move to warmer, coastal forests in winter. This is likely because birds can live near the coast for fewer calories/day. In winter, when food is limited, reducing energy costs translates into a greater probability of survival. Other of our analyses demonstrate that particularly our wintering permanent resident species congregate toward the coast.

Despite such a relationship, summer average temperatures (as measured by elevation and latitude) exhibits little relation to where birds are found. This is likely because during summer, energy in the form of forest productivity becomes available. This productivity is ultimately what produces the insects, fruits and seeds that birds rely on for food.  

Measuring the productivity of forests is possible with satellite imagery. Satellites routinely make a measurement called the normalized difference vegetative index, or NDVI, which is a measure of the relative greenness of forests. Research has shown that forest greenness serves as a measure of primary productivity.  NDVI data are gathered by the U. S. Geological Survey, so we have teamed with them to seek relationships between their data and our large scale surveys of bird distributions. 

To date, our findings have not been consistent with expected trends.  As noted, theory predicts that NDVI should relate positively to the number of bird species. Instead, our data show a weak negative relationship between NDVI and species (see above). The reason for this pattern is the subject of ongoing investigations.


Wednesday, November 30, 2011


The Carolina Wren is a frequent year-round resident of particularly southern Connecticut

A number of bird species are thought of as being permanent residents in southern New England.  However, although the species are present year-round, the same individuals are not necessarily present.  In fact, populations of resident species often change considerably between seasons.  In a number of instances, short distance migrations occur among some individuals.  In eastern Connecticut, ongoing research conducted by Bird Conservation Research, Inc. has demonstrated that species like the Tufted Titmouse, Blue Jay and Northern Cardinal show population declines from summer to winter as birds appear to leave the region.  In contrast, the Black-capped Chickadee shows population increases that appear to be driven by movement of chickadees into the region from further north.  A number of wintering species also concentrate their numbers near the coast, where winter temperatures average milder than in northern areas.  These findings are detailed in the paper, Seasonal Shifts in Population Distributions and Habitat Use by Permanent Resident Forest Birds in Eastern Connecticut.

Monday, November 7, 2011


Some years ago, I sent off to the venerable Connecticut Warbler an article entitled Saving the Red-eyed Vireo.  After receiving no review or reply of any kind, we at BCR decided to publish the paper in our own Contributions series (, where it has since received enormous worldwide readership.  The paper is a popularized version of a scholarly work, entitled Endangered Species, Provincialism and a Continental Approach to Bird Conservation (, which appeared in the journal Avian Conservation and Ecology.  Even now, six years after its initial appearance, this paper remains among the journal's most frequently downloaded. 

These articles asked the question, what should our conservation priorities be at the regional level in light of continental concerns and local practicalities?  I argued that we in the Northeast live in one of the most well preserved examples of the temperate forest biome (one of limited distribution to begin with) left on the planet, and this is where we must focus our limited conservation resources.  It is the seat of our greatest regional biodiversity and the one for which we can exert our greatest worldwide conservation impact.  I further demonstrated by examining state endangered species programs that, instead of using this kind of approach, states were often disappating efforts into programs where little consequential conservation gain was likely to be achieved.

The rub with this kind of viewpoint was that it identified investing conservation efforts into species like northeastern grassland birds as non-viable enterprises, especially in light of these species being ones with enormous continental distributions and having population increases over large parts of their range.  This view sent conservation colleagues into veritable interplanetary orbit from which many have still not come down (although, interestingly, other papers making similar arguments have since appeared, and the wording of local conservation documents has subtly shifted to acknowledge that the previous paradigm of attempting to save everything was as full of holes as swiss cheese).  In short, although the penalty for initially doing so was great, at some point someone needed to point our that the emperor really didn't have any clothes.  Otherwise, the emperor was going to have a long, cold winter.

At the regional level we do tend to become myopic toward larger issues because we see only what is around us.  I myself have been as guilty of this as any of my colleagues.  Indeed, back in the 1970s I was one of the architects of the "lets save everything approach" in Connecticut which has since, unfortunately, evolved into an entrenched, inertia-ridden bureaucracy of naive thinking.  It took an extended tour of duty in the remote tropical Pacific, where there were real endangered species to deal with and real ecological triage decisions to make, for me to see the world through new, more realistic eyes.

So, back to our friend the Red-eyed Vireo: it is presently a species with immense regional populations because its core habitat is extensive and improving in character.  It is also one with a distinct Achilles' heel- it lives in a globally endangered biome and really prefers only a portion of this biome- interiors of mature, oak-dominated forests.  It may be thought of as a keystone species whose well being reflects the larger well being of the ecosystem as a whole. 

In southern New England, we are presently blessed with a landscape that is 2/3 forested, and the forests are assuming an old growth character not seen here since Pilgrim times.  This convergence of characteristics has made our forest bird community immensely viable, as is evidenced by the local success of the Red-eyed Vireo.  How long this confluence of abundance and quality can persist is another matter, however.  Our region has now passed the point at which forest is expanding in distribution.  In fact, it is now in decline and the remaining cover is becoming increasingly fragmented as urbanization proceeds.  This means that if we are to preserve sufficiently extensive examples of unbroken forest to ensure the long term viability of its forest bird communities, the time to act is now.

But, in case you haven't noticed, land in southern New England is very expensive.  The amount of capital required to purchase substantial acerages is enormous.  Several years back, I did battle with the forces of darkness (i.e. big developers) in Old Saybrook, CT, arguing that a tract of 1000 acres of coastal forest was disproportionately valuable to New England bird communities because of the high density of winter residents it supported.  Although my arguments contributed to slowing the tract's development, to my knowledge the vast sum of money required to wrestle the land from the hands of developers has yet to be raised.

The question then becomes what are we doing squandering our limited time and resources blowing out a few candles in the kitchen when the whole house is starting to burn down?  If we are to preserve our premier ecosystem in southern New England, we must keep our eyes clearly focused on our goal.

R.J. Craig

Friday, November 4, 2011


One of the very few advantages of aging is that as we do, we gain a larger perspective on events in the natural world.  When I first began performing systematic surveys of Connecticut birds nearly 40 years ago, the Yellow-bellied Sapsucker was a decidedly uncommon and local species of the highest elevations of northwestern Connecticut.  I found it in places like open swamps at the summit of Canaan Mt.  Observers from a century previous reported its occurrence in much the same manner.  It was primaily a boreal species that straggled south to enter the coolest locations of southern New England.

Then, something happened- something that made this northern bird expand dramatically at its southern range limit in opposition to the larger (and likely global warming-related) trend of southern species expanding north.

From its few residences in Connecticut in the early 1970s, it began expanding through the northwest hills.  During the initial year of the Forest Bird Survey of Southern New England (2001), it also appeared in the northeast CT town of Stafford.  Much as it had been in northwestern CT, it was associated with open swamps.  Initial population estimates there were 126.  Upon resurvey of the region in 2004, the number grew to an estimated 502.  However, even this number paled in comparison to the 21,587 estimated for northwestern CT in 2007.  Indeed, the sapsucker had become the commonest woodpecker of that region.  Surveys of southwestern CT also showed a substantial population of 891.  Furthermore, the species had expanded habitat use to virtually every kind of mature forest present.

So, what happened?  What could have released this boreal species to expand into mature upland forests and inhabit them at high densities? 

Within their principal range, sapusckers are characteristically birds of earlier successional, open deciduous forests, often those near water and with dead trees.  This characterization matches well with the swamp habitats in which birds first appeared in both northeastern and northwestern CT, and it suggests that this type of environment represents core habitat.  But what of this new, huge population now inhabiting mature upland forest?  How did that fit?

One can erect a number of alternate hypotheses to investigate- ones that would make interesting thesis projects for enterprising students.  One strong possibility is that southern New England forests have become increasingly interspersed with open swamps due to growth in beaver populations over the period of sapsucker expansion.  The landscape as a whole may, therefore, now be much more attractive to the species even though much of the forest has mature trees not generally favored by sapsuckers.

Another possibility is evolution at the range limit.  Although range limit populations tend to have reduced genetic variability, some key genetic change might have occurred that permitted our local population to expand into a new habitat.  There are strong genetic arguments that can be made against this being very likely, but still, it's an intriguing thought.

Still a third hypothesis is that population pressure from the heart of the sapsuckers' range is driving peripheral populations into marginal habitats.  This sort of thing has been documented for other bird species.  In this case the possibility can be explored by examining long term population trends recorded by the Breeding Bird Survey.  However, when one does this, population declines appear to have occurred across maritime Canada during the past 40 years, which argues against this hypothesis.  Moreover, the sheer abundance of sapsuckers in northwest (and now northeast) CT forests suggests that the species is doing very nicely there.

So, does anyone else have any speculations about what's going on here?  This is a prime example of the dynamic rather than static nature of natural systems.

R.J. Craig

Monday, October 24, 2011


Although many AP Environmental Science classes across the country do some version of a lab investigation on stream organisms, they tend to be overly simplistic in terms of  the questions asked and the type of analyses required.  We have developed an investigation that retains a simple sampling protocol and data gathering phase, but which tests hypotheses that are relevant in terms of current thinking on community ecology.  Hence, it goes beyond simply examining what kinds of organisms are present by examining community parameters such as species richness, species evenness and species diversity.  It also considers the importance of sampling design to hypothesis testing and introduces statistical concepts like variance, standard deviation and coefficient of variation.  The investigation compares the aquatic macrofaunal communities of two streams of distinctly different water quality and requires the use of spreadsheets to analyze the data.  It uses data from previous investigations of water chemistry, and requires that macrofaunal community data be analyzed in light of the physical and chemical nature of the streams.  The lab may be downloaded from

Sunday, October 16, 2011


This winter-plumaged American Goldfinch was one of the species examined in the first paper published through the newly updated Bird Conservation Research, Inc. Contributions series.

Open access publications on the internet are available to all readers at no cost. No journal subscription is required and access is 24 hours a day, seven days a week.

Although subscriber-only print journals have long dominated the field of science, the practical needs of researchers have already resulted in freely available internet sources of information surpassing traditional outlets in actual viewership. The free publications of Bird Conservation Research, Inc., for example, have a much larger annual viewership than many subscriber-based journals.

In the emerging field of online science journals, the frequent subscription requirement and extremely high author’s publication costs have continued to limit viewership and the variety of papers offered. These factors further drive researchers to seek new internet sources for information.

Bias has also emerged as an issue in scientific publication. Scholars have demonstrated that researchers from less prestigious institutions who do not have “celebrity” co-authors, as well as women researchers, can be at a significant disadvantage in having papers accepted by traditional outlets.

To address shortcomings like these in scientific publication, BCR has revised its Contributions series to make it not only open-access, but also available for author submissions in an unbiased and low cost environment. BCR uses a double blind peer review process that makes authors and their institutions invisible to reviewers. See author instructions.

Wednesday, September 28, 2011


This mount of an Eskimo Curlew taken in Connecticut is one of the only specimens known from this state.

The admittedly Quixotic search for surviving Eskimo Curlews that might still migrate through outermost coastal Massachusetts entered its second year this August.  Bird Conservation Research, Inc. continues to search habitats not traditionally scanned by bird watchers seeking migrating shorebirds. One tantalizing possibility is that bird move through each fall undetected because the birdwatching community has forgotten how to look for the species. We focus our efforts on sand flats and dune hollows where 19th century gunners once hunted curlews. We avoid searching the tidal mudflats used by the majority of other shorebird species.

Although the probability of success in finding a curlew is low, previous similar efforts by BCR Director Robert Craig have ultimately yielded success. In searching for the supposedly extinct Aguiguan Reed-warbler, knowing how to look for the species made all the difference in successfully relocating it. Similarly, learning how to look for the Black Rail led to its rediscovery as a Connecticut summer resident after 14 years of searching.

Thursday, September 15, 2011


In the VCP technique, observers measure the horizontal distance from a survey point to the location of individual animals.

Forest bird surveys conducted by Bird Conservation Research, Inc. use the variable circular plot technique (VCP)- a very sophisticated and powerful tool used in field surveys of animal populations. It is the method employed by most technically advanced wildlife studies, and is the field method underlying the BCR bird population estimates (see previous BLOG) made in Connecticut and Rhode Island.  During the eight years of this study, high school science students were involved in computerizing and analyzing the enormous volumes of data gathered (50,172 bird observations and 19,980 habitat measurements!).

To educate students into the utility of VCP, we have developed a lab investigation for AP Environmental Science that leads them through the steps of gathering field data.  This investigation moves from collection of field data to computing population densities using DISTANCE software.  Data gathered from VCP are entered into DISTANCE and turned into densities of species per square km.

Friday, September 9, 2011


The Northern Waterthrush has larger regional populations than preliminary estimates
The Connecticut State of the Birds 2011 report contained a contribution from Bird Conservation Research, Inc. entitled State of the Forest Birds.  In this report we projected, based on our findings for eastern Connecticut and Rhode Island, the total populations of certain species.  Since the appearance of our contribution, we have completed computerization of all of our data and are now able to estimate with accuracy populations present during the years of our study.  Although our analyses are undergoing further refinement by using new features available in Distance 6.0, our population estimation software, we can now make the following comparisons between preliminary Connecticut projections and actual estimates based on all data.
 Yellow-bellied Sapsucker: projected– most common woodpecker in northwestern CT, actual– 20,968 vs. 11,958 for the Downy Woodpecker.

Acadian Flycatcher: projected– 10,000, actual– 10,413.

Least Flycatcher: projected– 3,000, actual– 4,281.

Black-billed Cuckoo: projected– 1,000, actual– 411–815 in differing years.

Yellow-billed Cuckoo: projected– 1,000, actual- 2,156-4,354 in differing years.

Common Raven: projected– 400, actual– 395.

Cerulean Warbler: projected- 6,000, actual– 6,105.

Black-throated Blue Warbler: projected– 10,000+, actual– 41,712.

Pine Warbler: projected– 20,000+, actual– 74,431.

Louisiana Waterthrush: projected– 20,000, actual– 23,337.

Northern Waterthrush: projected– 3,000, actual– 4,190.

Eastern Towhee: projected– approaching 100,000, actual– 64,196.

Differences between initially projected and actual numbers are largely attributable to eastern Connecticut not always serving as a good model for populations in the rest of the state. This was particularly true in the case of the towhee, which had high densities in eastern Connecticut and Rhode Island but low densities elsewhere.

Thursday, September 1, 2011


This figure shows the detection probability of Black-throated Blue Warblers encountered at increasing distances from an observer.  It was generated using program DISTANCE.

Population estimation as traditionally taught in AP Environmental Science is simplistic and does not take into account species detectability or population changes due to immigration, emigration, mortality and natality.  Lab activities also make little use of free software available for estimating populations.  The software DISTANCE and MARK are are used for estimating populations, respectively, from large scale surveys and from mark-recapture projects.

To introduce teachers and students to more sophisticated methods of population estimation that go beyond the simple but unrealistic Lincoln-Peterson method, we have developed two laboratory investigations that use the 1) Jolly-Seber method, that computes populations by taking into account population changes that occur over time, and 2) Distance sampling (using the variable circular plot technique), that computes populations by accounting for differences in species detectability encountered during field surveys.  Both investigations may be found under the Educators tab of the Bird Conservation Research, Inc. web site.

Wednesday, July 20, 2011


The advanced placement or AP Environmental Science program for public schools has a nationally established curriculum with rigorous, college-level standards. The curriculum is a broad one that encompasses not only ecological topics but also Earth science and conservation issues.

Each year, a comprehensive examination is given throughout the USA to evaluate the knowledge of students in the field of environmental science. However, at present few educational resources are freely available for AP educators to use in their classrooms.  To begin providing these resources, Bird Conservation Research, Inc. has produced a series of Powerpoint presentations that cover the national curriculum standards. In addition, they go beyond present standards to present timely topics that are at the forefront of ecological and conservation thinking. 

To date, we have produced four presentations: Earth Systems and Resources, The Living World, Population and Environmental Pollution. These are available for free download via the BCR web site.

Tuesday, July 19, 2011


The Brown-headed Cowbird above was a fairly common invader of forest environments throughout the region.
Southern New England conservation groups have identified the development of a high level, quantitative database on wildlife resources as key to developing substantive regional open space acquisition strategies.  To meet this need, Bird Conservation Research, Inc. is developing such a database for the forest birds of southern New England- by far the most diverse wildlife group of the region. 

The first two stages in our investigation were to survey quantitatively the forest birds of eastern Connecticut and Rhode Island.  These studies are now complete, and have produced the land-planning atlas, Forest Birds of the Last Green Valley, as well as scholarly papers.  This atlas was distributed at no cost to every town conservation commission in the research area, and its distribution has been followed up with explanatory presentations (with the assistance of the Connecticut Cooperative Extension System) to town conservation commissions, regional watershed councils and conservation groups.  To date, the major findings of our work have been discovery of: 1) broad patterns of winter and summer bird diversity across the region, 2) major winter-summer shifts in densities of permanent resident species, 3) environmental heterogeneity across the region exerting a major influence on species density distributions, 4) even the most common species having non-uniform distributions, 5) no one tract being sufficient for preserving regional forest bird diversity.  Such findings demonstrate, for example, that coastal forests are the principal reservoir for wintering species and have among the highest priorities for regional protection.  

The next phases of the investigation was to survey of forest birds of central, northwestern and southwestern Connecticut in both summer and winter.  State of the art field, analytical, and Geographic Information Systems mapping procedures are being used to make the database accessible to conservationists. 

The analysis of forest bird survey data continues, with much larger samples from eight years of study leading to more accurate and precise measurements of regional populations Measurement of populations relies on construction of detectability curves– graphical representations of how easily a species is detected at increasingly greater distances from a point. More data lead to increasingly realistic detectability curves.  Although computations for the over 100 species detected during surveys will be continuing for some time, some population estimates completed to date follow:

Black-throated Blue Warbler:
Northeast CT: 5091
Southeast CT: 0
Central CT: 925
Northwest CT: 34,798
Southwest CT: 898
Rhode Island: 1333

Red-shouldered Hawk:
Northeast CT: 171
Southeast CT: 261
Central CT: 33
Northwest CT: 98
Southwest CT: 174
Rhode Island: 219
Northeast CT: 47
Southeast CT: 31
Central CT: 16
Northwest CT: 33
Southwest CT: 63
Rhode Island: 31
Northeast CT: 17,447
Wood Thrush:
Southeast CT: 23,401
Central CT: 37,816
Northwest CT: 21,374
Southwest CT: 20,076
Rhode Island: 11,474

Pileated Woodpecker:
Northeast CT: 321
Southeast CT: 369
Central CT: 363
Northwest CT: 980
Southwest CT: 379
Rhode Island: 80

Winter:Northeast CT: 424
Southeast CT: 367
Central CT: 194
Northwest CT: 195
Southwest CT: 699
Rhode Island: 27