North on the Wing

Home > Other > North on the Wing > Page 3
North on the Wing Page 3

by Bruce M. Beehler


  Various groups and institutions, both large and small, are battling the threats that have led to the decline of migratory birds. In addition, research teams at scores of colleges and universities are studying these impacts and seeking solutions, while federal, state, and local governments work to preserve and restore strategically important natural habitats. Many nongovernmental organizations aid the effort as well. Every group involved understands the need to address the threats that birds face at different stages of their life cycle, ensuring that no vulnerability in the bird’s life is overlooked.

  Because conservation is an essential facet of the spring migration story, it was one of my primary interests during my spring songbird quest. I visited and spoke with many people who work on programs to conserve migratory birds and the migratory system as a whole. They are the heroes of bird conservation in North America, and my reports on their work, scattered throughout the book, share both the lessons they have learned and the hope for the migrants’ future that their stories have given me.

  As I planned my journey, I knew I would travel in very different circumstances than the Teales did on their 1947 journey. While they had stayed in roadside motels, I camped in a tent every night. This not only saved a bundle of money but also kept me connected to the birds. For a hundred days and nights, I was surrounded by the sights and sounds of nature. I carried a two-burner gas stove and a cooler stocked with food, only occasionally eating in restaurants and instead dining al fresco at my campsite. The Teales had traveled through a world without interstate highways, Walmarts, Japanese SUVs, or East Coast suburban sprawl. I experienced phenomena unknown to the Teales as I traveled through a world of declining rural towns in the Deep South and Rust Belt cities in the North, bringing along my high-tech handheld devices. But we were both in search of the same thing: spring’s annual march through wild America.

  The Teales had traveled in a black Buick without seatbelts, stereo, or GPS. I drove a second-hand Nissan SUV carrying a kayak and a bike. The car got me to desirable campsites in various patches of woods, from which my bike, kayak, and foot power took me to interesting natural locations nearby. The SUV held storage containers stocked with maps, field guides, and local travel guides, as well as classic accounts of similar trips—North with the Spring, of course, as well as Wild America and On the Road with John James Audubon. Duffle bags held sleeping pads and bags, pillows, field clothing, and rain gear. I had digital camera equipment and two cell phones, a portable Wi-Fi hotspot, a GoPro video camera, and a laptop computer, which I used to record the journey and upload twice-weekly posts to a travel blog on the American Bird Conservancy website so people could follow the trip online. As I prepared, it became clear that even though I’d camp out and make meals in circumstances more rustic than those the Teales encountered in 1947, mine would be an utterly twenty-first-century operation.

  Now let’s take to the road. Let’s get away from the cities and head south. Let’s fill the ice chest with food and drink, get onto Interstate 81, set the cruise control, and make our way to Texas.

  TWO

  The Texas Gulf Coast

  FIRST LANDFALL

  Early April 2015

  Common Yellowthroat

  To the green mist of the cypresses and the moving clouds of the swallows we could add the movement of the stars as a sign of the sure approach of the spring. With all the galaxies and planets and stars, the solar system was setting the stage.

  —EDWIN WAY TEALE, North with the Spring

  The first port of call for many trans-Gulf migrant songbirds in spring is the coast of southeastern Texas, and that is where I plan to meet them. At the end of March, I depart Maryland and over three long days drive to Mad Island, on the Texas coast between Freeport and Corpus Christi. About sixty miles south of Houston, Mad Island is where I’ll meet the vanguard of the northbound migrants. My declared field trip focus is songbirds, but spring on the coast of southeastern Texas is the height of distraction for the curious naturalist because of its abundance of waterbirds, wildflowers, snakes, lizards, and more. At Mad Island, in fact, I’ll learn more about coastal prairie than I will about songbirds, which have been slow in arriving stateside this particular spring. Every turn of my trip, as I will learn, will introduce me to both little-known ecosystems and the people working to conserve these habitats and the diverse bird communities that depend upon them. I have come for songbirds, but I’ll encounter much more.

  Mad Island is not, despite its name, an island at all but a tract of prairie overlooking Matagorda Bay and protected from the Gulf by the long, narrow barrier island that runs nearly unbroken southwestward to the border with Mexico. The Clive Runnells Family Mad Island Marsh Preserve, a seven-thousand-acre tract of remnant coastal prairie under the management of the Texas chapter of the Nature Conservancy, lies just southwest of the hamlet of Wadsworth. This preserve, the southernmost point on my circuit, was the official starting point of my journey north to Canada. Here I visited a bird-banding project of the Smithsonian’s Migratory Bird Center (SMBC). Directed by Peter Marra, a Smithsonian colleague, the SMBC studies the ecology and conservation biology of migratory birds in the Western Hemisphere. A SMBC field team seasonally stations itself at the preserve, which generously allows the researchers to do their work here. Marra had advised me to visit the bird-banding projects at Mad Island and at Grand Chenier, Louisiana, to discover what’s being done on the ground to investigate the spring migration of songbirds across the Gulf of Mexico.

  With the Colorado River estuary to the east, Matagorda Bay to the south, and Tres Palacios Bay to the west, the Nature Conservancy reserve at Mad Island is a bird heaven year-round. During one recent winter, in fact, the wetlands and rice fields of the Mad Island region supported more than two hundred thousand waterfowl, and Mad Island is known for periodically generating the largest species list of any regional Christmas Bird Count (CBC) in America. The annual CBC, sponsored by the National Audubon Society, is the nation’s longest-running citizen-led annual birding event. Groups of volunteer birdwatchers all over the country count the birds they see and hear within a designated “count circle” fifteen miles in diameter on a designated day between mid-December and early January; more than two thousand such counts are conducted across the continent each year. CBC records on the changing status of winter bird populations have been compiled into a century’s worth of field data for researchers. Whereas the CBC censuses wintering bird populations, my arrival at Mad Island was timed to match the return flight north of many birds that wintered to the south. In preparation for the coming returnees, I set up my tent on the edge of a broad expanse of coastal prairie, just a short drive down a sandy track from the banding station at the coast.

  MIGRANT WINTER LIFE IN THE TROPICS

  When I showed up, the bulk of Neotropical migrants that the Mad Island banding crew awaited were still down in the Tropics, and I wondered whether the evolutionary origins of these migratory species were northern or southern. Although the details are messy, recent evolutionary analyses indicate that most Neotropical migrant lineages evolved in the temperate zone and subsequently adapted to spend the winter in the Tropics. Wood warblers are a mixed bag: most sublineages of the group evolved in North America, not the Tropics. So some birds evolved in the temperate zone and shifted their wintering habitat south, while others evolved in the Tropics and shifted their breeding habitat north, as originally sedentary lineages began to migrate to follow the seasonal availability of resources. The most likely factor motivating their transition to the migratory habit was the cycle of glacial advances and retreats over the past 2.5 million years. Glaciation’s major impact was the enforced relocation of breeding ranges for virtually all the North American songbirds, and it also had major influences on rainfall and seasonal drought in the Tropics.

  What is life in the Tropics like for Neotropical migrants? The migrants have to share their tropical habitat with the abundant local resident birdlife, and they are in this habitat during the dry seas
on, when resource levels are depressed. The migrants likely struggle to establish a winter foraging space, and they must work hard to prepare their bodies for the return flight north and the demands of nesting and raising offspring.

  Many migrants seem to select a winter habitat in the Tropics that comes as close as possible to their habitat in the north temperate and boreal zones of North America, but there are limits to their ability to match habitats between the temperate and tropical zones. The tropical flora is different, the days are shorter, the weather typically is warmer, and the food resources differ, offering substantially more fruit and nectar than the northern habitat does. The differences probably outweigh any similarities the birds are likely to find, and accordingly some species change their feeding behavior between summer and winter—those that are specialized insect eaters in the north shift, in some instances, to consuming nectar or small fruits in the Tropics.

  Migrants’ social behavior also can differ between north and south. Most Neotropical migrants establish breeding territories in the north, with a male-female pair occupying each territory and raising their young there. In the Tropics, by contrast, some species establish solitary (one-bird) territories for much of the winter, whereas others establish no territory at all, instead roaming about and joining other foraging species of migrants and nonmigrants. Some migrants, for example, take up with multispecies flocks of birds that regularly follow army ants, pursuing the invertebrates that flee before the moving ant columns. Other species become seasonally sociable with their own kind, foraging in monospecific flocks on flowering and nectar resources. And some species are reported to roost in single-species groups at night.

  Those that join mixed-species foraging flocks in the Tropics are part of a phenomenon that seems widespread in the Tropics of both the New and the Old World, one that may be driven, at least in part, by the threat of predation by bird-eating raptors, snakes, and mammals. The mixed flock occupies a large foraging territory each day, and as the group moves about, individual birds or bird pairs join up or drop out of the flock as it passes through their home ranges. A wintering thrush might forage alone on its solitary home range but join a flock to forage with the group as it passes.

  Studies of Kirtland’s Warbler have demonstrated that winter conditions influence subsequent breeding productivity. Sarah Rockwell and her collaborators, for example, have shown that ample rainfall in March on the Kirtland’s wintering ground in the Bahamas led to both earlier arrival by males back in Michigan and the production of more fledglings per male. So winter conditions for a warbler in one location can have measurable carryover effects on breeding results in a distant locale. Similar results have been shown for American Redstarts that winter in the Caribbean and breed in eastern North America.

  During the long and lean season in the Tropics, each migrant songbird has two clear objectives: to avoid being eaten by a predator and to maintain (or improve) its physical condition in readiness for the migration north and the breeding season to come.

  THE COASTAL PRAIRIE ECOSYSTEM

  Mad Island lies at the heart of the coastal prairie ecosystem that once dominated southwestern Louisiana and coastal Texas. The ecosystem, decimated by development, once included extensive tallgrass prairie, wetlands, and patches of gallery forest. Today, it is one of the most endangered natural habitats in all of North America. This narrow band of prairie lies just back from the coastal marshes and once stretched in an arc that paralleled the coast of the Gulf of Mexico. But less than 1 percent remains in its original state. Most has been degraded by cattle grazing, rice farming, sugar cane monoculture, oil and gas development, and creeping urbanization. Public and private restoration of prairie lands is crucial to the future of this critically endangered ecosystem, which is home to an array of birds—including the Mottled Duck, Attwater’s Prairie-Chicken, White-tailed Hawk, Crested Caracara, Scissor-tailed Flycatcher, Painted Bunting, and Dickcissel. Prior to western settlement, this unique tallgrass prairie evolved through the influences of seasonal rainfall, periodic fire, and grazing by Bison and other wild ungulates; native grasses thrive under conditions in which they can outcompete woody plants. In the intact areas of the coastal prairie, native grasses such as Big Bluestem, Indiangrass, Eastern Gamagrass, and many native wildflowers dominate. These species cannot tolerate heavy year-round grazing by domestic cattle, which encourages the invasion of exotic annual grasses such as Vasey Grass, from South America, and Johnson Grass, from the Mediterranean. Natural prairie is dominated by long-lived perennials, and, with a few exceptions, annuals are rare in undisturbed prairie sod.

  Most Neotropical migrant songbirds that cross the United States–Mexico border in spring have destinations far north of Texas. That said, a few do settle to breed in the Texas coastal prairie, including the Scissor-tailed Flycatcher, Common Yellowthroat, Painted Bunting, and Dickcissel. The first of these, with its long swallow tails and apricot underwings, is among the loveliest birds in America. It winters in open country in southern Mexico and Central America and breeds in the southern Great Plains all the way down to Mad Island, where it is one of the most familiar birds along rural roadsides. The Common Yellowthroat—one of my quest group, the eastern wood warblers—winters as far south as Panama. The only wood warbler to breed right along the Gulf Coast, this diminutive, bandit-masked songster is North America’s most widespread wood warbler. The yellowthroat is unusual in breeding in marshes and shrublands, places that often lack woods or trees. Most breeding wood warblers prefer forest as a nesting habitat, and Mad Island’s patches of low coastal scrub woodland are inadequate for that purpose.

  Male Painted Buntings show a splashy patchwork of bright colors, whereas the females are plain pale green. The species winters in Mexico, Central America, south Florida, and the Caribbean, hiding in scrub woodland and seen most frequently when visiting a backyard feeder. For birders, the male Painted Bunting, which sings a musical song reminiscent of that of the more widespread Indigo Bunting, is one of the most sought-after species of the Deep South. The Dickcissel, with its black bib, yellow breast, and chestnut wings, is one of the omnipresent songbirds of the prairie, giving its staccato six-note song from atop shrubs all day long. For birders visiting southeastern Texas for the first time, the sight of these four species out in the open prairie habitat is entrancing.

  BIRD-BANDING

  Called ringing in Europe, bird-banding has been a means to study bird movements since 1803, when John James Audubon tied silver wire around the legs of nestling Eastern Phoebes in Pennsylvania and found two of them back on their nesting site the following spring. Today, 6,500 bird-banders are active in North America. All operate under the auspices of the Bird-Banding Lab at the Patuxent Wildlife Research Center, in Laurel, Maryland. The lab, founded in 1920, is a collaboration between the U.S. Geological Survey and the Canadian Wildlife Service, and it provides the bird-banders with free numbered aluminum bands that they affix to the leg of each captured bird for permanent identification. Each year the lab provides bird-banders with fresh supplies of bands, and in return it receives detailed records of the birds banded over the previous year. The lab keeps computer files on the more than one million birds banded each year and the seventy million birds that have been banded over the past nine decades. Of greatest interest are the “recoveries”—banded birds that have been retrapped at a subsequent time or place. It is recovery data that yields information on the movements of birds across the continent.

  Traditionally, banding studies focus on productive “migrant traps”—areas where migrating songbirds concentrate because of optimal geography and habitat. For instance, from 1958 to 1969, Chandler S. Robbins, James Baird, and Aaron M. Bagg situated their Operation Recovery program at several coastal locales in the East to track songbird migration, and these initiatives led to the establishment of permanent bird observatories, such as Manomet, in coastal Massachusetts, and Point Blue, near Point Reyes, California. Several universities carry out seasonal bird-banding operations as well
, but most birds are banded by hobbyist banders in neighborhood woodlots and regional parklands. Collectively, these groups and individuals have generated invaluable information on species longevity and how large-scale weather patterns influence the seasonal movement of songbirds, both of which help scientists to understand the evolution of successful migratory strategies.

  Dickcissel

  Mad Island’s isolated patches of coastal woodland are ideal for spring bird-banding because they draw in arriving migrants, which find the vast expanses of prairie grassland and marshland unsuitable as stopover habitat.

  Over dinner on the night of my arrival, I meet the Smithsonian project team: Emily Cohen, a postdoctoral researcher; Tim Guida, project technical officer; and four assistants. These dedicated bird researchers use a spacious, modern hunting lodge donated, along with the reserve land itself, by the Runnells family in 1989. The compound includes a house with a large, elevated porch, a barn, and outbuildings. Floor-to-ceiling picture windows look out across Mad Island Lake, a birdy coastal estuary, and surrounding coastal saltmarshes. Vistas of prairie and marshland spread out before us.

  Mad Island is the southernmost banding project for Neotropical migrants approaching the United States. By the time of my arrival, the Mad Island team had documented a couple dozen species of Neotropical songbird migrants—including Indigo Bunting, Blue Grosbeak, and Hooded, Kentucky, and Swainson’s Warblers—dropping into its little patches of coastal scrub woodland. Amid the scrub, the researchers had strung mist nets: forty-foot-long black nylon nets spread tautly between two poles. Mist nets act like large spider webs, harmlessly entangling unsuspecting birds who fly into them without ever seeing them. Thus far, only a few birds were arriving daily, and the team was still waiting for the first big wave of migrants.

 

‹ Prev