General Technical Report
hroughout the New World attention is now being focused on the status of populations of landbirds, which are the many species of smaller birds, sometimes referred to as non-game birds. Landbirds have not usually been the focus of management activities except in a few cases of threatened or endangered species, such as the Kirtlands Warbler. Recent evidence suggests that some landbird species are declining in abundance, fueling much speculation upon the causes of these declines, the species involved, and their habitat preferences. Hypotheses about the causes of these declines are varied, ranging from tropical deforestation to nest parasitism by the cowbird. However, part of the difficulty in determining the status of landbirds results from problems in monitoring these small birds, as compared to larger, more easily-studied species. To determine population changes, and to hypothesize possible causes of these changes, more basic information needs to be gathered.
Much of the evidence for these population declines in the New World has come from the results of the Breeding Bird Survey coordinated by the U.S. Fish and Wildlife Service and the Canadian Wildlife Service (Robbins and others 1986, 1989). These roadside counts provide excellent baseline data. However, they do little to identify the factors contributing to changes in landbird populations, and are limited to areas with major roads.
The use of population size as a measure of health of a species has been a common tool of biologists for many years (Hutchinson 1978; Lack 1954, 1966). Methods for surveying population size have been detailed by Ralph and Scott (1981), in the excellent compendium by Cooperrider and others (1986), and the manual by Koskimies and Vaisanen (1991). Population size, however, is only a retrospective tool. It tells only after the fact that a species has enjoyed an increase or suffered a decline. To ponder causes of changes, the biologist must couple information on population size with data on the internal composition of a populationits demographics (Temple and Wiens 1989). For example, data on sex ratio, age distribution, nesting success, survivorship, average weight, and population movements can all give valuable cues to factors or events regulating a population. Moreover, such primary population characters can provide early warning signals of population problems prior to actual declines. Many studies have used data such as these to describe the dynamics of various populations (DeSante and Geupel 1987, Hutchinson 1978).
Several other efforts have been under way to document changes in adult populations and in productivity. For example, in the late 1970s, the Germans and Austrians started the Mettnau-Reit-Illmitz-Programm (Berthold and Scherner 1975). Since 1981, the British Trust for Ornithology has conducted their Constant Effort Sites (CES) Scheme (Baillie and others 1986; Baillie and Holden 1988; Peach and Baillie 1991; Peach and others 1991). Martin has started a program involving nest searches (Martin and Geupel> in press). DeSante (1991, 1992a,b) has started a cooperative mist-netting program in North America known as Monitoring Avian Productivity and Survivorship (MAPS) along these same lines. The Point Reyes Bird Observatory has been monitoring landbird populations in coastal California for more than 25 years (Ralph 1967, Geupel and DeSante 1990b).
In this handbook we outline the steps that might be followed in monitoring many species of landbirds. We cover methods used in monitoring of population size, productivity, age and sex ratios, survivorship, habitat relationships, and other parameters. We provide details of four methods that estimate population size, two methods that measure demographic factors, and two suggestions for conducting habitat assessment. We have tried to give the land manager, biologist, and others complete information on basic requirements, tools, resources, and methods to carry out a successful landbird monitoring program. Depending on funding and staffing, any combination of the techniques we describe is applicable to virtually any site and budget. This handbook does not provide the objectives of each study that might be conducted, or what analyses can be conducted. These both must be examined carefully before monitoring begins. We hope that this handbook will generate interest in monitoring programs using methods that can give insight into causes, as well as the facts, of population changes.
Objectives of a Monitoring Program
A monitoring program ideally should provide three types of data. One is an estimate of the population size and trends for various species of birds. The second is an estimate of the demographic parameters for at least some of those populations. The third is habitat data to link the density and demographic parameters of bird populations to habitat characteristics. Ideally a monitoring program should take a community approach and monitor all avian species in the area.
We have recently seen a marked increase in interest in monitoring, far outstripping available personnel, training, and resources. Indeed, this increase is the impetus for this handbook. While this has been gratifying, we think that it is essential that people first determine why they might want to establish a census, mist netting, or nest searching program. Not everyone requires a monitoring program to meet their goals. We have sometimes seen the establishment of a monitoring program first, followed by an attempt to decide what type of information can be obtained. We very strongly suggest that, before a monitoring program is put in place, the following steps be carried out: (1) decide the objectives and goals desired; (2) determine whether monitoring is the way to accomplish these; (3) with the goals firmly in mind, write down the questions being asked, clearly and objectively; (4) determine which monitoring methods most directly answer the questions posed; (5) review the types of data that can be obtained from these methods, and outline exactly how these data will answer the questions; (6) outline the analytical methods that can be employed; (7) determine the cost, logistics, availability of personnel, and probable length of commitment to the project; and (8) write a study plan and have it reviewed by a person competent in research and statistics. This procedure is vital, because accumulation of a data base does not itself lead to meaningful analyses later.
Participants in a monitoring program can include private, state, provincial, and federal groups. Our premise is that the basic entity for this exercise is an administrative unit, such as a Forest Service District or a State Park. Not all such units will want or need such a program. Each unit should outline its needs and goals before starting, suggest monitoring programs to meet those needs, and have them reviewed by a competent biostatistician. We do believe, however, that our recommendations below have generality among many types of administrative units. These units can be very heterogeneous, and thus a variety of methods may be needed.
Landbirds: the general term used for the generally smaller birds (usually exclusive of raptors and upland game birds) not usually associated with aquatic habitats. By contrast, waterbirds include seabirds and other aquatic species.
Region: an area of several thousand acres, often including several drainages, that the biologist wishes to sample. Here, extensive point counts are conducted on roads to monitor overall population sizes and their changes.
Administrative unit: the basic entity that conducts monitoring. Examples are a Forest Service District or Forest, a State Park, a National Wildlife Refuge, a private nature center, or a commercial forest.
Monitoring station: an area of usually less than about 50 ha (125 acres) within a region. Here, intensive censuses, nest searching, and mist netting are conducted.
Capture array: the generally rectangular or circular configuration of mist net locations at a station.
Nest search or census plot: an area of, preferably, a single habitat type where spot mapping or area searches are conducted.
Census grid: the arrangement of intensive point counts overlaying a demographic mist net array or nest search plot.
Census point: the place where a single point count census is taken.
Net location: the place where a single net is placed.
Nest site: the place where a single nest is found.
A 10-day time: interval is the basis for most monitoring and analyses.