Protocols for Conducting Forays in the Upper Cumberland Region of Kentucky and Tennessee
A foray is a bird-monitoring survey, usually lasting 3 to 7 days but sometimes as long as 10 days, designed to collect data about the breeding status, distribution, and abundance of all breeding species present within a single county.
Protocol for Forays Conducted in 1985 and Later
Typically, the county being surveyed during a foray is divided into discrete units, usually based on 7.5' USGS quadrangle maps. The county is first divided into units that each represent a "quad." The quads are then further divided into subunits, usually six of them. These subunits, representing 1/6 of a quad and referred to as "blocks," become the units surveyed during the foray. A typical county may have 40–50 blocks within its boundaries. During a foray, usually conducted in late May or early to mid-June, each block is visited by a birder or a team of birders who intensively searches the block (about 12 square miles in extent ) for all bird species present and for evidence of breeding for as many species as possible. When a block has been searched long enough for about 60 species to be found within it (and with approximately 25–50% in confirmed categories [see below]), it is deemed to have been surveyed well enough for most of its breeding species to be found. Usually a full day of field work, including night work, is sufficient to obtain this result, but this result may be obtained in some blocks with lesser effort.
To obtain information about the abundance of birds breeding in the county, a miniature Breeding Bird Survey (BBS) route was conducted in some or all blocks within a county. Typically a "mini-route" has 10–15 stops located at intervals of 0.8 km (0.5 mi) along roads within a block. At each stop along the mini-route, one observer conducts a point count for a set amount of time (which varies from 3 to 10 minutes), classifying all birds seen and heard into one of several time intervals (usually 0–3 min, 3–5 min, and 5–10 min) and into one of several distance intervals (usually within 25 m, 25–50 m, 50–100 m, and beyond 100 m). Data for each stop are recorded on a standard point count data sheet. These data are then used to generate abundance classifications for the species that are registered during the mini-route.
An alternative to the mini-route is to have all block workers spend a standardized amount of time carefully counting birds throughout as much of each block as they can get to in that amount of time; they may then spend any additional time they wished to spend in the block, but they would not be specifically counting individuals of all species during any extra time in a block, just noting breeding evidence for the species encountered, adding new species, etc. Two hours of counting in the early morning or three hours of counting in the late morning and afternoon appear to provide time-standardized data that can reasonably be substituted for mini-route data, which many birders do not feel as comfortable collecting as they do collecting data by simply counting any and all birds seen or heard during a specific period of time. The resulting numerical data are perhaps more prone to being skewed by differences in observer skill than would mini-route data (presuming that mini-route data are collected only by persons experienced with this mode of data collection), and the data are probably skewed by non-uniform coverage of roads and other areas from block to block, but having such time-standardized data from each block (instead of just the blocks where mini-routes can be run by experienced observers) makes collecting numerical data in this manner worthwhile, and this is indeed the method now (2008 and thereafter) used to collect abundance data during UCR forays. Results from the DeKalb County Foray suggest that abundance data collected in this way correlate well with abundance data obtained by other collection methods in terms of depicting relative abundances of breeding species and probably in other ways.
Codes for breeding evidence follow (a date is also attached to codes for probable and confirmed breeding):
Possible Breeders:
X = present during the foray
Probable Breeders:
[T = territorial--present at same site on dates at least a week apart; not
usually possible
during a foray]
P = pair
A = agitated
behavior (presumably near nest or young)
C = courtship
or copulation
Confirmed Breeders:
DD = distraction display
CN = adult
carrying nest material
NB = adult
building nest
UN = used
nest
FY = adult carrying
food for young
BG = begging
young
FL = fledged
young
FS = fecal sac removed from nest by
adult
ON = adult on
nest incubating
NE = nest
with eggs
NY = nest
with young
A field card used for forays in the Upper Cumberland Region may be viewed by clicking on the following link:
A set of instructions used for the White County Foray in May 2007 and a set for all later forays may be viewed by clicking on one of the following links:
Protocol for Forays Conducted Prior to 1985
Forays conducted prior to the advent of the Kentucky Breeding Bird Atlas project (1985–1991) and the Tennessee Breeding Bird Atlas project (1986–1991) were not conducted using the protocol outlined above; instead, counties in which forays were conducted prior to 1985 were divided into parcels (usually as roadways in the county allowed); then each parcel was visited by a team of birders who counted individuals of all bird species within that parcel during all the time that they spent in the block. The totals for all the parcels in the county were then summed for each species, giving county-wide totals for all species recorded during the foray. In addition, breeding bird surveys were undertaken during some forays; these were conducted according to the protocol for North American BBS routes; results entailed presenting the sum of individuals and the sum of stops at which each species was recorded. Sometimes breeding evidence, such as the presence of fledged young, was noted during these early forays.
When forays are conducted in UCR counties that were originally the subject of forays prior to 1985, it might be beneficial, albeit a bit difficult, to collect data using each of the general protocols noted above. When future forays are conducted in counties where forays were conducted in 1985 and thereafter, it would be simplest to use only the first of the protocols noted above.
Some Additional Thoughts on the Validity of Foray "Counting Period" Data
During the past four decades, an effort has been made to use data collected from continental Breeding Bird Surveys (BBS) to track trends in most of the breeding species of North America, employing the fairly rigorous point count method to collect the basic data.
During the past decade or two, an effort has also been made to corroborate BBS trend data using various other kinds of data. The chief dataset for doing that has proved to be Christmas Bird Count (CBC) data, which are collected using a very different, not very rigorous, collection method. Yet CBC data apparently provide a means of validating BBS trend data for the species that appear in both datasets; i.e., analysis of CBC data leads to trends in populations of many species that are similar to trends calculated using the more rigorously collected BBS data. Thus, the trend data coming from CBCs, while not collected according to a standard protocol that is very rigorous, are still good enough to provide a valuable confirmation of BBS trend data.
Interestingly, CBC data and foray data appear to share some important features.
The basic geographic unit for collecting CBC data is the "territory" (a variably sized portion of the 177-square-mile CBC circle where members of a "party"—sometimes one person and sometimes more than one person—count all birds seen and heard for a specific, but variable, number of hours, following no specific protocol for counting within the territory, resulting in each party collecting data in a slightly different manner from most or all other parties working within the circle—i.e., some use tapes to call in birds, and some don't; some use spishing, and some don't; some walk a lot, and some don't; some cover all roads within the territory, and some don't, etc.).
The basic geographic unit for collecting foray data is the "block" (an area that is theoretically uniform in size, but which actually varies quite a bit, depending on whether the entire block is within the boundaries of one county or not); either one person or a group collects data for 2 or 3 hours, usually covering a given block in a slightly to greatly different manner from the way other persons or groups cover other blocks.
The geographical unit of coverage for collecting CBC and Foray data is quite variable, unlike the standard "point" used for point counts, and the temporal unit of coverage for collecting CBC and foray data exhibits less variability but is still much more variable than the standard 3-, 5-, or 10-minutes used for point counts. If, despite these collection differences from point counts, the CBC data still yield reliable, corroboratory trend data for many species of breeding birds, then it is possible that foray data may also be reliable enough to provide useful abundance data for breeding birds (and, if collected over time, the foray data might yield useful trend data also, provided that enough forays can be conducted to achieve a fairly large dataset).
Another set of bird data that is collected in roughly the same way as CBC and foray data (i.e., with considerable variability in the geographic unit covered and moderate variability in the temporal unit covered) is the Great Backyard Bird Count set of bird data, now in its fourteenth year of collection. Some initial efforts to analyze this gigantic dataset are pointing to its data being a rough mirror of CBC data for many species that overlap the two datasets.
A possible way to determine the reliability of foray counting period data vis-a-vis point count data involves the ratios between pairs of commonly occurring species, such as Downy Woodpecker/Hairy Woodpecker, Red-eyed Vireo/Yellow-throated Vireo, and Indigo Bunting/Blue Grosbeak. If the ratio for such pairs from foray data is similar to the ratio of enough such pairs for point count data, then this comparison might indicate whether foray counting period data are reliable enough to be useful in determining the abundance of species.
All this suggests we may have become too dependent on point counts for the gathering of reliable bird data and are not open enough to other, less rigorous, methods of collecting bird data that can help us to understand basic issues relating to bird populations, such as trends and abundance.
Quick links to other sections of this website: