General Technical Report
General Monitoring Procedures
Species To Be Covered
Although many species will be censused at a single station, fewer will be captured, and still fewer species will have their nests found. However, biologists at a single station should get a good sample of the population size of perhaps 30 species and some indication of demographics on about 10 species. In a region with perhaps six stations, more species will be monitored. Over a wide geographic area, these data can be combined to produce patterns of the population sizes and demographics of many species.
The period of study for the breeding season differs, depending upon the individual species, latitude, rainfall pattern, temperature, elevation, or even year. Therefore, each region should establish its own monitoring period on the basis of the local breeding season and the criteria described below.
Demographic monitoring, by mist nets or nest searches, should span the entire breeding season. Censusing, by contrast, is usually conducted only during approximately the first half of the breeding season, when birds are most active, paired, on territories, and vocal.
For all monitoring, we recommend the use of the sampling interval time period of 10 days, as used in the British CES project, for several reasons. This interval allows at least one weekend for making up for inclement weather, and divides the month into three approximately equal portions. It also provides a basis for direct comparison between stations.
Operation of the demographic monitoring station by mist netting or nest searching should begin no sooner than the 10-day interval when virtually all of the breeding birds have established territories, but before many have begun laying eggs. For most lower elevation areas in temperate North America this will be about May 1 or May 11. The date, however, should be adjusted to conform with the local situation. For example, in the more northern parts of the United States, the first period can begin May 21 or May 31. In Alaska or northern Canada, or at high altitudes, the first period may begin as late as June 10. In the southwestern United States or coastal southern California, where 90 percent of the species have begun nesting activities by mid to late March, the starting date could be April 1 or 10. In Mexico, it could be even earlier, and in much of Latin America it could be much earlier. It is considered important by some investigators to avoid netting before migrant individuals of breeding species have finished moving through. Early netting might result in later net avoidance during the breeding season, thus biasing a few of the demographic estimates. However, some adjustment for this factor can be made during analysis and many stations do this with good results.
A good measure of the establishment of territories is increased singing. Also, captured males will show a pronounced cloacal protuberance. Individuals carrying nesting material is another excellent indication that the breeding season is under way. The best measure of the start of the breeding season is the beginning of egg laying. Females normally develop a brood patch when the first egg is laid.
The termination of the demographic monitoring should be no earlier than when the local population begins to be augmented by fall migrants, or by an increase of dispersing individuals known to have not bred in the local area. In most of temperate North America, this will usually be about the second or third 10-day interval of August.
For uniformity, May 1 of each year should be considered the first 10-day period. If a season in a region begins earlier, it should be backdated from May 1. In fact, the season of monitoring for most areas in temperate North America will begin May 1 and continue for a maximum of twelve 10-day intervals until August 28. If a station begins before May 1, it should continue until late August, unless a pilot projects data indicate that all breeding individuals and their young have left earlier.
For most of temperate North America, we recommend, therefore, the following periods: May 1-10, May 11-20, May 21-30, May 31-June 9, June 10-19, June 20-29, June 30-July 9, July 10-19, July 20-29, July 30-August 8, August 9-18, and August 19-28.
Censuses conducted on demographic study stations, such as mist netting stations, need be done only in the first five 10-day intervals when birds are on territory and actively singing. In temperate North America, this will be usually from May 1 through June 19. In northern latitudes or higher elevations, the period could be as late as June 1 to July 9. Point censuses, and also area searches, should be done once on each plot in each of the five 10-day intervals, and preferably about the mid-point of the interval.
Operating a monitoring station in the spring or fall is an option in many areas. Spring and fall migration data from mist nets and censuses are confounded by many factors, particularly local weather, and the questions migration data can answer are different from those netting during the breeding season can answer. The data can provide interesting and insightful information about the timing, composition, and extent of migration (e.g., Ralph 1978, 1981a; Robbins and others 1959). The fall migration in particular gives a measure, derived from many source areas, of the overall productivity of a species. As mentioned previously, if a mist net program is operated in the spring in the same area as a breeding mist net array, a few demographic measures may be altered somewhat.
The value of winter studies is quite high. Winter is a time when populations are resident and relatively stable, thus providing excellent data on survivorship and mortality. It is very likely that habitat associations, for example, are much more defined in the winter than in the summer (e.g., Huff and others 1991, Manuwal and Huff 1987). The methods outlined here have full applicability in the non-breeding seasons, both in North America and Latin America. In the tropics, mist netting throughout the year at the same site would clarify many questions about molt, skull, and plumage patterns.
Maintenance of a Study Plot
Plots should be permanently marked with stakes, markers, or flagging that will survive over at least one winter. Rebar (steel reinforcing bar), rock cairns, or tags driven into landmarks all work well. Tags are available from biological supply companies. In general, markers should be laid out along a compass direction, be placed at regular intervals, and be visible at any point between the markers. Each marker should correspond to a numbered grid point on a map. In colder areas, be aware that in years with heavy snowfall, plot markers can still be buried in the spring. Net and census points should also be permanently marked. Be sure to record in your journal net height and angle of placement (use stakes or give a compass direction).
Customized maps of the study area should be traced from a large scale map or from aerial photographs. Landmarks, grid points, and net and census points should also be sketched. Blank maps can be used for spot mapping censuses, vegetation mapping, and other figures.
Each monitoring station ideally should be operated indefinitely. Although objectives will vary, we suggest that at a minimum, capture arrays of nets and nest searches should be operated for four consecutive years, and census plots for three years.
Journal keeping is an essential tool of all field biologists. The importance of regular, accurate journal keeping cannot be overemphasized. It is not uncommon for journals to be subpoenaed in court. The Grinnell method (Herman 1989) is the most widely used by vertebrate ecologists and is extremely detailed. Here we provide guidelines for basic information that may be useful for monitoring landbird populations.
As a minimum we recommend recording the following on a daily basis:
- Netting information: (a) the number and location of each net operated; (b) the exact hours of each net operated; and (c) the total capture and recapture rate for each species at each monitoring array.
- Censuses and nest searches: the number, location, and timing of each census conducted and the hours of nest searches.
- Personnel information: list the activities of each biologist conducting field work, including areas censused, net locations operated, and other activities.
- List of all birds seen or heard: basically presence/absence data; provide any interesting notes on potential breeding or other behavior of note.
- Weather data: in addition to the basic weather data that should be taken (see below), a general one- or two-sentence statement on the days weather is also helpful.
- Plant phenology: a list of what is blooming or in seed may help interpret changes in bird distribution.
- Interesting observations of mammals, herptiles, insects, and other natural history observations should also be included.
Training and Numbers of Personnel
Training is extremely important because the level of training and experience will greatly affect the reliability of the data collected. Training must be something that is continuous throughout the field season. It is necessary to transmit expectations early and often in data taking or responsibilities for certain tasks.
The length of time to train personnel will vary greatly depending upon the quality and interest of recruits. For many census procedures, the mechanical aspects can be taught in two or three 2-hour sessions. However, for a person who has minimal skill in identification of plant or animal taxa, it can take a week or longer, depending upon the taxon, and the persons previous experience. The suggestions for censusers in Kepler and Scott (1981) are especially relevant. For a completely untrained person to learn to remove birds from mist nets takes at least 2-3 weeks of intensive training. This training should include at least 3-4 hours of removing birds from nets each day. Training for nest searching requires a similar time commitment.
Probably the most important aspect of training is the testing of the observer. This should be done regularly in the field by the most experienced personnel available to make sure that data are accurate, and of high quality. This can also be accomplished by regularly checking data sheets as they come in from the field. Any delay prevents feedback to the field crews.
The number of persons required to operate a monitoring station depends upon several factors. If nets are the method of choice, we suggest a minimum of two people, one of whom is well-trained in removing birds from mist nets, and one of whom is well-trained in identification of birds by sight, song, and call. The less skilled person can be of great assistance, and with proper training can contribute much to the monitoring. Censuses and netting are both morning activities, and under some circumstances they can be conducted concurrently if the censusers position is known to the netter and he or she can be called upon for help if capture rate is moderately high. The health of the birds is of paramount importance, and all efforts to prevent injury must be taken. Nest searches can be conducted throughout the day, although it is most productive in the morning.
When conducting censuses, it is best to rotate observers, if at all possible, so that no observer censuses any given point more than the others.
Syllabi for training in the methods contained in this handbook have been prepared. These are for the use of persons experienced in the methods, so that they can efficiently pass on the methods to others. The syllabi are available from the senior author.
Data To Be Taken
Below we outline several types of data to be taken and provide sample forms for each. We have also prepared data entry programs using IBM compatible computers for these forms. Clean forms for reproduction and these programs can be obtained by contacting the senior author. These programs can use either standard entry systems such as dBASE or simple BASIC compilers.
For each point count census point, mist net location, and nest site, we suggest that the Location and Vegetation Form be filled out. It is described in detail in the Habitat Assessment section and contains important location information for data base files. At the minimum, for all monitoring programs, this location data should be taken.
All the data forms have constant the following information, to help relate between data bases:
- State or provinceThe 2-column code for each.
- RegionAn 8-column code, designated by the investigator. Often, the name of the USGS quad, a prominent landmark, or a nearby town will provide the best code name.
- Station/LocationThis is a 6-column unique identifier designated by the investigator to separate, within each region, the location of the various data points. We recommend that the station be a 4-letter code. The net location, point count census point, or nest number will be a 2-number code.