---

Table of Contents

• Introduction
• Abundance
• Seasonal Occurrence
• Elevation
• Habitat
• Header Information
• Area Covered
• Other Information
• Maps
• Detailed vs. Condensed Checklists
• References Cited

---

While researching material for the upcoming book Colorado Birds (Andrews and Righter 1992), we (BA, BR) collected and examined about 30 local bird checklists from Colorado. These checklists were produced by local bird clubs or Audubon Society chapters, government agencies, and private individuals. They covered counties, river valleys, national parks, monuments, grasslands, forests, and wildlife refuges, state parks and wildlife areas, half (Davis 1969) or all of the state (Holt and Lane 1987). Our objective was to get a better idea of how the abundance of species varied in different geographic areas, at different elevations, and in different seasons. For our purposes, some checklists were useless, while others were quite useful, such as Lambeth and Armstrong (1985), Jasper and Collins (1987), and others.

The checklists varied widely in the information included and how it was presented, ranging from simple lists of bird species with no additional information, to those with detailed abundance and habitat information. However, most checklists had limited information on arrival and departure dates and habitats, almost none provided altitudinal information, and several systems describing abundance were used. Many checklists attempted to cover so large and diverse an area that usefulness was limited. There was such variation that making meaningful comparisons between the areas covered was nearly impossible; it was like comparing apples and oranges. Additionally, many interesting (and some frequently birded) areas were not covered by any checklist: Cottonwood Canyon, the Tamarack Ranch-Jumbo Reservoir area, South Park, and the Yampa River Valley to mention a few.

We propose that a standardized format for Colorado checklists be established, and that checklists be published in the C.F.O. Journal. We suggest that the abundance categories, and perhaps even the types of graphs used in the forthcoming Colorado Birds be used in these checklists. Our intention is not to dictate to others how to make a checklist. But by standardizing the types of information included in checklists, and how that information is presented, the ability of birders, authors, and researchers to make meaningful comparisons in the abundance, altitudinal range, and habitat use of species in different regions of the state and in different seasons would be greatly enhanced. All checklists would be maintained on computer at a regional center, allowing easy updating. The following is a brief outline of our proposed format.

Abundance.  The single most aggravating variable in the checklists we examined was in how abundance of species was presented. Most checklists included abundance information, but so many definitions of abundance categories were used that it was usually very difficult, if not impossible, to determine how a given species varied in abundance between the different areas covered. In Colorado Birds, we have used a system of abundance categories based on the number of individuals likely to be seen in a single day by a typical observer (averaged over many visits in several years) (see Table 1). This system is similar to those used by Dunn 1981, Unitt 1984), and Bull (1985).

Table 1. Abundance Categories

Abundant:	> 100/day in appropriate season and habitat
Common:	25-100/day
Fairly common:	10-25/day
Uncommon:	1-10/day; usually seen daily
Rare:	1-5/day and 1-10/season; usually not seen daily
Very rare:	10-40 records (for the state as a whole, or within certain areas or seasons)
Casual:	4-10 records
Accidental:	1-3 records

It is important to note that these numbers are intended as a general guide as to the number of individuals that can possibly be seen rather than a specific prediction of how many will be seen. Actual numbers seen on any given day will vary from day to day, year to year, area to area, and from observer to observer. These categories are most useful when used flexibly rather than rigidly. Ideally, they should be based upon many field trips in at least several years.

The most common alternative system is to indicate the probability of encountering a given species. Probability may be expressed using either percentages or phrases. Thus, a species may be almost certain to be seen (> 95%), could possibly be seen (50%), or is very unlikely to be seen (10%). This is a useful system, and we do not suggest that it is incorrect. But we believe that the system of describing actual numbers is a more useful service to both observers and researchers. An observer that is told how many individuals can possibly be seen can discern for him/herself what the probability is of seeing that species. On the other hand, an observer that knows only the probability of seeing a species is still uncertain of how many individuals can reasonably be expected.

This can be illustrated using an example from the Barr Lake Christmas Bird Count. The Bald Eagle and Horned Lark have both been seen on all Barr Lake counts since 1981; hence, they would both have a high probability of being seen (> 95 %). However, there have never been more than six Bald Eagles seen (uncommon in our system), while Horned Larks are typically seen in hundreds or even thousands (abundant in our system). The probability of seeing these two species is similar, but their true abundances are vastly different. This difference is real, easily perceived by birders, and should be reflected on a local checklist. The probability system fails to reflect this difference.

Seasonal Occurrence.  Most checklists presented abundance information by season (e.g., common in spring, rare in fall). Although this is useful, information is lost because seasons are different for different species. For example, fall is a different time period for Rufous Hummingbirds than for Sabine's Gulls (mostly July and August for the former and September and October for the latter). In Colorado Birds, relative abundance in different seasons, average arrival and departure dates, and extreme dates are shown in graphic form. When appropriate, this information is presented for different geographic areas or elevations. Although the same information can be presented using dates and words (fairly common from 20 Mar to 5 Nov), the graphic form enables the pattern to be much more quickly and easily seen and understood.

To show how seasonal occurrence graphs can be made, we have provided the following examples. The simplest type of graph shows how numerous the species is, when it arrives, and when it leaves (see Fig. 1a). This shows a species that is fairly common from 20 Mar to 5 Nov. Even if the checklist includes only this most basic information, that would be very useful.

If information is available, and the author of the checklist desires, additional information can be added that would be valuable. For example, if the same species occurs in both low valleys and in the mountains, an extra level could be added (see Fig. 1b). This shows that the same species is also fairly common in the mountains from 25 May to 20 Aug.

Additional information such as peak migration periods and extreme dates can also be added (see Fig. 1c). In our example, the species is common from 15 Apr to 10 May and from 20 Sep to 20 Oct. There are also a few early and late records from 5 Mar and until 25 Nov, and there are extreme outlier records on 12 Feb and 28 Dec.

Many people may be intimated by the process of generating graphs on the computer. Of course, checklists don't have to include graphs, and the same information could be presented in another format. However, because graphs, are such an excellent visual tool, we propose that authors would be able to forward the information to a computer graphics specialist working with the Colorado Bird Observatory, and that person would make the graphs.

Elevation.  In Colorado, elevation plays a profound role in determining or influencing the distribution and abundance of birds, and their arrival and departure. However, checklists that covered a geographic area with a wide elevational range never presented abundance or arrival and departure dates for different elevations. Of course, birders don't cover all altitudes equally in all seasons, and so this information is not always available or complete. Nevertheless, presenting what is known can be useful, and the absence of what is not known focuses attention on what is yet to be learned. In Colorado Birds, elevation information is presented in a graphic form similar to seasonal occurrence (see Figure 2). The sample shows elevational distribution in spring and fall. To save space in the checklists, elevation graphs would either have to be much smaller or aligned horizontally rather than vertically.

Habitat.  Most geographic areas encompass more than one habitat, most species are not equally abundant in all habitats, and some species show geographic and seasonal variation in habitats used. Therefore, it is important to include habitat information. Many checklists provided a list of habitats present in the area, with abbreviations and codes that were listed under each species to indicate which habitats are used. This is useful, but we suggest that the relative abundance of species in each habitat also be included. One way to do this would be to list habitats according to their relative importance. In the example below, habitats are designated for summer and winter by codes (PP is ponderosa pine, DF is Douglas-fir, etc.). The major habitats are listed before the slash, and minor habitats after the slash. Habitats in which breeding is confirmed are underlined.

S     PP, DF/SF, PJ
W     PJ/PP, DF, SF

Header Information.  We suggest that checklists always include a space for observers to enter the following information: date, habitat, number of observers, foot miles, car miles, time at start and end, temperature (real or estimated) at start and end, wind (real or Beaufort estimates) at start and end, sky (clear, overcast, partly cloudy, rain, snow etc.) at start and end, snow cover, ice cover, etc. This information is especially valuable if individual daily checklists are collected and analyzed for population trends, as the Colorado Bird Observatory is doing.

Area Covered.  Careful thought should be given to the boundaries of the area covered. Natural geographic and ecological units are best, but political units (counties, state parks, etc.) can also work. In all cases, but especially with political units, it is important not to cover too large or diverse a geographic area and to adequately address varying abundance, arrival and departure, etc. in different areas (especially at different elevations) within the area covered.

Other information.  All checklists should provide a code indicating species confirmed or suspected to breed in the area. If the area covers different habitats or elevations, or a wide geographic area through which the species is not uniformly distributed, codes should be provided for different areas. One way to present this information is to use asterisks or underlining with habitats and elevations where breeding has occurred.

Another useful piece of information would be in what seasons, areas, or elevations unusual species should be documented by written details or photographs. Many observers think of documentation only with respect to species recorded only a few times in the state. But there are other types of records that should be documented. For example, House Wrens are so common in summer that documentation is obviously unnecessary; however, they are so rare in winter that all observations should be very carefully documented. Likewise, Calliope Hummingbirds are regular enough to fall in the mountains that they need not be documented, but one at Bonny Reservoir should be. These cases should be indicated on checklists to alert observers when to be extra careful about identifications and when to obtain supporting documentation.

Most checklists did not provide separate information for the forms of recently lumped species such as the juncos, orioles, and rosy flinches. This results in a loss of information. Those forms are still identifiable in the field and each form usually has a different status, and some forms that were lumped could conceivably be split into full species again. Therefore, checklists should include all field-identifiable subspecies and color morphs.

Although most checklists do use the AOU order for listing species, we have found some that do not. The AOU order should always be used.

Maps.  The most useful way to portray distributional information is with a map; Colorado Birds will have about 900 maps (with up to three maps per species for different seasons). Because they consume a lot of space, it is not practical to include maps on most checklists. However, observers who have the interest should consider creating local maps, especially for species of special interest and concern. Publishing local maps, using the system of shadings and symbols in Colorado Birds, could be very interesting and useful (see Fig. 3 for a sample map).

Detailed vs. condensed checklists.  All of the information and graphs suggested in our format can fit on the small cards on which most checklists are published. Dauphin et al. (1989) have done a good job of presenting a comparable amount of information on a small card and could be used as a model. The cards could be included to the C.F.O. Journal as an insert. But if it is not possible to put it all on the field card, then a detailed version of checklists could be published in the C.F.O. Journal, and a condensed version of the same checklist could then be prepared for small cards to be used in the field. On the field cards, Coen Dexter has suggested that the graphic information on seasonal occurrence could be summarized using numbers to indicate months and letters to indicate weeks (e.g., the third week of March would appear as 3c). Although the checklists published in the C.F.O. Journal could be in this condensed version as well, we encourage publishing checklists with the graphs if possible because they are such a good way to portray information.

There are several benefits that published standardized local checklists would provide. More detailed and specific information on local abundance, seasonal occurrence, and habitat use would be available for researchers interested in particular species or in the birds of a particular area, and for authors of publications (including future editions of Colorado Birds). These checklists would help both local and out-of-state birders to know more precisely where to look for species of interest and to determine whether their observations are unusual. Like the latilong and atlas, these checklists would focus attention on species for which more information is needed to fully understand their status in Colorado.

---

References Cited

---

NPWRC Home�|�Site�Map�|�About Us�|�Staff�|�Search�|�Contact�|�Web�Help�|�Copyright