|Lying in Wait for Partners In
Flight: Some Experiences Monitoring Birds in Southeastern Bottomlands
Paul B. Hamel 1, Norman L. Brunswig 2, Michael R. Dawson 2, and Mike Staten 3
Our charge is to examine the state of the art in monitoring. This usually is construed to mean the most advanced possible practice of the art; our purpose here is to examine the state of the art as it currently is practiced by managers in the Southeast, in the hope of demonstrating activities that are within the grasp of every manager. We believe that the success of the North American Bird Conservation Strategy depends more upon the activities of all managers than on those of practitioners on the so-called cutting edge.
We describe three separate monitoring activities. First, we outline the strategy adopted for monitoring by the Southeast Management Working Group, and its realization in the monitoring plans of the National Forest System and the U. S. Fish and Wildlife Service in the South. Second, we outline the experience of the managers of a private, nonprofit organization, the National Audubon Society, in monitoring birds on the Francis Beidler Forest. We here address monitoring the unintended consequences of management by examining bird response to a catastrophic natural disturbance. Third, we describe the experience of a private, industrial timber company, Anderson Tully Company, in monitoring birds on their extensive holdings in the Mississippi Alluvial Valley. This permits us to examine the unintended outcomes of monitoring.
Case 1. Southeast Management Working Group, Partners in Flight: Monitoring strategy for the region
The Southeast Management Working Group has produced a strategy and set of standards for conducting monitoring of birds in the South (Hamel et al. 1996). Based primarily on experience developed in the Mississippi Alluvial Valley (Smith et al. 1993), and consistent with the suggestions of Ralph et al. (1993), the strategy indicates to individual managers how they might establish a monitoring program using point counts. The strategy provides managers with step-by-step guidelines to determine the intent of the monitoring program, the level of detail sufficient to meet the intent of the program, and the sample sizes of counts sufficient to answer particular questions.
Our experience in the Southeast suggests that basic bird lists are unavailable for most properties. We believe that such lists are the keys to successful monitoring, inventory, and public education involving birds; hence we encourage managers to develop bird lists.
For properties with bird lists, we suggest that managers determine what sort of information is necessary for them to carry out their management tasks successfully. The step-by-step process outlined in the strategy guides the manager to assess whether information on presence-absence, relative abundance, or demography is the intent of the monitoring activity. Detailed instructions allow the manager to determine the number of point count samples required to address two basic questions related to differences in numbers of birds. First, we address the question "have bird communities on my property changed over the years?" Second, we address the question "have bird communities responded to management treatments that I have applied?" Our "Quick-and-Dirty" estimation of sample size starts with these questions, proceeds through the development of an estimate of variability derived from a pilot sample of counts, describes the estimation of the power available from a given sample of counts, and details how to determine a sufficient number of point counts to address a given question with a particular defined precision. Unfortunately, no magic number of counts can be named to satisfy every need.
Application of this process has resulted in the deployment of broad-scale monitoring schemes on U. S. Fish and Wildlife Service properties (R. Coon, pers. comm.) and on U. S. Forest Service properties (Gaines and Morris 1996) in the Southeast. Importantly, the schemes developed by these two public agencies are within reasonable reach of what managers across the region can endorse and carry out. An unintended benefit of this process has been the development of a commercially available software package that supports the data standards developed for the Southeast (Guddanti 1994).
All of these activities have taken place in the 1990s, with growing recognition of the impetus and value of becoming a Partner in Flight. Yet monitoring of birds is not a new idea in the South, although it has been a fairly low priority in many quarters. Nevertheless, some managers truly have been lying in wait for Partners in Flight. We present two additional case histories illustrating how the development of a monitoring program that is responsive to the information needs of the land manager can provide unintended benefits to the manager, including scientific information, public education, and revenue.
Case 2. The Francis Beidler Forest: Monitoring response to natural catastrophe
The Francis Beidler Forest, a sanctuary jointly owned by the National Audubon Society and The Nature Conservancy, is managed by the National Audubon Society primarily as a place where 1000-year-old baldcypress (Taxodium distichum) trees can grow in peace. At purchase in 1973 the sanctuary consisted of about 3600 acres (1500 ha) nearly evenly divided between old-growth bottomland hardwood forest, especially cypress-tupelo (Nyssa aquatica) stands, and similar stands that had been selectively logged during the 1960s.
To learn more about their land, NAS put a single person in the field for 10 mornings per year, and established two objectives for a bird monitoring program. First, they wanted to track fluctuations in populations of breeding birds on the property as one measure of forest health and to develop a bird species list for forest visitors. Second, they decided to compare populations on the old-growth with those on the cutover portion of the tract. To address these two objectives, Beidler Forest staff set up two 8-ha spot-mapping plots on sites of matching topography, with one in the old-growth portion, and the other in the cutover portion of the tract. After initial sampling of each tract in 1979, an alternating schedule was adopted in which one of the plots was sampled each year.
A total of 16 censuses have been conducted, eight on each of the plots. Fifty-seven species have been recorded on at least one of the censuses, 50 on the cutover plot and 49 on the old-growth plot. Seven species occur at different densities on the two plots, five of which are Neotropical migratory birds (Appendix). Four of these species have significantly greater densities on the cutover and three on the old-growth plot, indicating that neither habitat on the Beidler Forest is better for all the species of the sanctuary. Among the species more abundant on the cutover plot is the Swainson's Warbler (Limnothlypis swainsonii) whose population peaked on the cutover plot in 1979, 12 years after the initial harvesting of much of the canopy of the stand.
These data have been useful to Beidler Forest managers in their own right, but are not the major story of the bird monitoring program there. On the night of 21 September 1989, Hurricane Hugo visited the Low Country of South Carolina, bringing 150 mph (240 kph) winds and extensive destruction to the forests on the sanctuary. Among the very few positive remnants after the storm was the opportunity to compare the populations of bird species on the sample plots before and after the storm. These comparisons provided Beidler Forest staff with quantitative information to demonstrate to their visitors the actual effects of the storm on the birds of the Beidler Forest. Densities of five species were shown to be different on the Beidler Forest after the storm by Wilcoxon 2-sample tests; each is a Neotropical migratory bird. Densities of Yellow-billed Cuckoos (Coccyzus americanus; Z=-2.11, P<0.05), Chimney Swifts (Chaetura pelagica; Z=-1.97, P<0.05), Wood Thrushes (Hylocichla mustelina; Z=-2.86, P<0.01), and Northern Parulas (Parula americana; Z=-2.95, P<0.01) declined, while those of Blue-gray Gnatcatcher (Polioptila caerulea; Z=2.7, P<0.01) increased. Among species whose densities did not differ significantly after the storm were the Yellow-throated Warbler (Dendroica dominica) and Northern Cardinal (Cardinalis cardinalis). Because baldcypress trees were particularly resistant and hence relatively unchanged by the winds of the storm (N. Brunswig, unpubl. data), we expected that birds associated with cypress stands, such as Yellow-throated Warblers, would be affected less than others by the storm. Northern Cardinals, on the other hand, increased on the old-growth plot to numbers typical of their use of the cutover plot before the storm. These birds are associated with shrubby understory in the forest, which increased dramatically after the storm.
The effects of Hurricane Hugo on the bird community of the Beidler Forest could not be evaluated were it not for the foresight of managers to institute a monitoring program. The techniques used were put into practice long before a larger scale program like Partners in Flight, and so do not use the same standards as currently employed in the Southeast. Nevertheless, they serve the needs of the manager and have provided some valuable information on bird community response to a very infrequent natural perturbation that was, otherwise, a disastrous event for the sanctuary.
Case 3. Anderson Tully Co.: Monitoring provides unexpected benefit of information transfer
Anderson Tully Company is a private timber company engaged in production of large sawtimber trees (Staten and Hodges 1997). The company, founded in 1889, owns lands in the Mississippi Alluvial Valley from Illinois to Louisiana. It maintains a set of permanent 0.2-acre (0.08-ha) plots on which are measured timber inventory variables for their timber management activities. Plots, classified into one of 11 forest site types, are sampled every five years.
Because a large proportion of their land is leased to hunt clubs, over time the companys wildlife management staff bolstered the list of inventory variables by adding measures of presence of four browse plants important for white-tailed deer (Odocoileus virginianus): dewberry (Rubus sp.), honeysuckle (Lonicera sp.), greenbrier (Smilax sp.), and trumpetcreeper (Campsis radicans), as well as measurements of live and dead cavity trees.
During a 1990 field trip for the Arkansas Chapter of The Wildlife Society to company lands in eastern Arkansas, the land manager noticed that a considerable store of information about bird occurrence existed among his wildlife staff. Excited about the variety of birds on company lands, he decided to add information about bird occurrence to the standard inventory. Concern also existed about potential conflicts between bird populations and certain management activities.
The companys wildlife staff developed protocols for adding a continuous habitat inventory to the timber inventory. The approach represented a hybrid of available effort and desired coverage. A randomly selected subset of 65 of the permanent plots was chosen for wildlife habitat monitoring, comprising five plots in each of nine forest site types on company lands, and ten plots in each of the two most common types ("Riverfront hardwoods - Dry site," and "Upland Hardwoods- Good site"). Within each permanent wildlife monitoring plot, extent of downed woody vegetation, both long wood and top materials, and stumps are measured. A set of 20 1-yd2 (0.84-m2) subplots also is sampled for understory vegetation. Each of these 65 plots became the center for a star of five bird point count stations, spaced 200 m apart.
The first sampling of these 325 point count stations was conducted in 1994. A standard 10-min count was conducted at each of the points. Vegetation sampling on bird point count stations involved a visual determination of the actual forest site type, as well as estimation of percent coverage by the vegetation and the two most frequent species in vegetation layers. General layers involved ground cover, midstory, and overstory. Within each of these layers, field crews determined whether to separate coverage into three or fewer sublayers. Field effort for this monitoring activity involved a four-person crew for six field weeks.
Company plans call for resampling on the same 5-year interval as the timber inventory plots. Point count data will be prepared and summarized to the standards of the Southeast Management Working Group.
To train their staff to conduct this monitoring activity, Anderson Tully Company wildlife biologists realized that no simple field guide, complete with a set of tapes of songs of breeding birds, existed for the birds of bottomland hardwoods. Furthermore, existing larger treatments, such as Hamel (1992), did not present relationships of birds to standard habitat structures in a quickly understood visual manner. Consequently, company biologists and foresters compiled a book, complete with species accounts, identification, habitat and other details of species biology, and a narrative and visual summary of proposed silvicultural and structural habitat use for each of 71 species (Staten 1994). An audiotape or compact disk of songs of these species also was prepared. The book and tape are available from the company, and proceeds from the sale will be used to support the companys research activities.
The story, however, does not stop here. In the process of developing this guide, establishing monitoring activities, and training company biologists for this work, an interesting thing happened. The consciousness of not only company biologists, but of the companys entire forest management field staff, has expanded to include knowledge of birds as a part of the store of knowledge about company lands. Field foresters bring questions of bird identification to biologists; through this interaction and the combined understanding of forest structure and bird response on company lands, a practical understanding of silviculture for songbirds on company lands is emerging. At periodic department managers' meetings, the land manager now conducts "field tests" in which field managers make point counts of birds, and compare their counts to those of company biologists to improve the quality of the general knowledge of birds among all company field personnel. We propose that the experience of this one timber company can be instructive to others interested in, or apprehensive about, adding specific songbird activities to their land management program.
We have attempted to show, through three different examples, how monitoring activity must meet the recognized needs of the land manager of a property. Our examples indicate that well-designed monitoring activities can meet efficiently the objectives of information gathering necessary to the land management activities of a particular agency or property owner. They further show that monitoring designed for a specific purpose can incidentally provide outstanding opportunities for responding to natural or planned disturbances, for training staff, for educating clientele among the general public, and for increasing income.
Several additional points can be made. First, in monitoring birds on a given property, benefits will accrue to the managers who invest time in working with their neighbors to develop and implement a monitoring scheme. Our current understanding of the effects of landscape factors on bird distribution clearly indicates this common-sense notion. Second, successful monitoring activities can be developed through a step-by-step, adaptive process in which it is not necessary to try to "run before one can walk." Third, monitoring activities that are integrated into the ongoing staff training, public information, and decision-making processes of the agency or institution will satisfy not only the intended objectives. They also will put the enterprise in a position of preparedness, of "lying in wait" for a future promising unexpected opportunities.
Were it not for the diligence, enthusiasm, and guidance of others we would have little to present. In particular, the patient guidance of C. J. Ralph and the members of the Monitoring Working Group, Partners in Flight, has been of inestimable value. The work of Chuck Hunter, David Pashley, and the monitoring committee of the Southeast Management Working Group, especially Winston Smith, Dan Twedt, Jim Woehr, Eddie Morris, Bob Cooper, and Bob Hamilton, made possible the development of the Southeast Management Working Group strategy. Glen Gaines spearheaded implementation of the strategy in the U. S. Forest Service Southern Region, and Richard Coon oversaw its application in the U. S. Fish and Wildlife Service in the Southeast. Bob Cooper was a supportive editor and stimulator of this paper. He and Mel Warren served as reviewers for it. The use of trade or firm names in this publication is for reader information and does not imply endorsement by the U. S. Department of Agriculture of any product or service.
Appendix. Summary of bird occurrence on two 8-ha plots on the Francis Beidler Forest, South Carolina, 1979-1995.
Density ± S.D., in pairs/100 haa
a Partial territories and Visitor species quantified as by Hamel (1992).
b Species densities differ between the plots by Wilcoxon 2-sample test at P=0.05.
c Transient species recorded only as visitors during migration.
1 USDA Forest Service
2 Francis Beidler
3 Anderson Tully