A Comprehensive Annotated List of the Butterflies
Chase Lake Butterfly Survey
Study Area and Methods of Sampling
The site of this study was lands within contiguous boundaries of the Chase Lake National Wildlife Refuge Complex in Stutsman County in central North Dakota. In 1995 two different sampling procedures were employed in an effort to confirm occurrence of as many species as possible. The two methods are referred to respectively as "Pollard transect" (or simply "transect") and "checklist" methods. In order to ensure adequate replication, only one Pollard transect (CL001) was employed in 1995. In 1996, the checklist method was dropped and a second transect route (CL002) was added. The two methods (1995) or transects (1996) were surveyed in an alternating pattern so that data could be compared statistically.
Both transects were rectangular (100m x 300m), totaling 800 meters in length, 5 meters in breadth, and 5 meters in height, and each was designed to traverse ecologically similar circumstances representative of the majority of terrain at the site. Both were located on WPA land, just north of the refuge proper in Section 4, T141N R69W. Each transect was marked off in flagged 10-meter segments and, where possible, with more permanent 50-meter reference posts (inverted metal electric fence posts).
For the checklist sampling method employed in 1995, a hypothetical list of regularly breeding resident species was first compiled for the site. The hypothetical list was derived from formal records for all included and adjacent counties. Excluded were occasional non-overwintering immigrant species. Native species for which it seemed likely that the absence of records represented incompleteness of information rather than actual absence were included in the hypothetical list and likewise in both editions of this list on the assumption that they will likely be confirmed at some time in the future. A separate reporting form was used for each checklist walk.
Each transect was represented schematically on the reporting form by a series of vertical linear scales, each divided into 10 segments corresponding to the 10-meter flagged segments on the ground. These scaled sections (0-100 meters) allow one to use the form to "walk" mentally through the transect and spatially envision each recorded encounter. Date, name of reporter, start time, finish time, temperature, estimated sunshine (10% increments), wind direction (eight points) and wind speed (Beaufort Scale) were all recorded at the finish of each transect walk. A separate reporting form was used for each transect walk.
Counting procedures for the two methods were identical in all respects except one, namely that in the transect method the recorder walked only along the designated transect route, whereas in the checklist method the recorder was free to wander at will. Procedures for 1996 transect counts were identical to those employed in 1995. Protocols for counting and recording in all situations generally followed Pollard and Yates (1993), Monitoring Butterflies for Ecology and Conservation:
- The sampling day extended from approximately 9 am until 5 pm.
- At the beginning of the day, a coin was flipped in order to determine which method or transect was to be employed first; methods or transects were thereafter alternated throughout the day.
- Butterflies were recorded with encoded four-letter acronyms (see appendix) later used in database management.
- No voucher specimens were collected within 100 meters of any point on a transect.
- The walking pace was steady and regular at approximately 35 meters per minute.
- Only butterflies actually seen within an estimated 2.5 meters on either side of the recorder, within 5 meters above the ground, and within 5 meters to the front were counted; butterflies behind the recorder or otherwise outside these limits were never counted.
- No active pursuit was undertaken, and no effort was made to find or to see butterflies hidden from view or perched out of sight of the recorder.
- A net was carried, and when problem individuals could be netted without undue difficulty or active pursuit, they were netted and released after identification had been made (in such cases, counting was stopped until the walk was resumed).
- Where distinction between two alternative species was not possible, the commoner of the two options (for that period of the season and the site in question) was recorded.
- For each case involving more than two such alternatives, identification was made to the lowest confident taxonomic level practicable.
- When individuals flew ahead of the recorder, when the recorder was certain that they had already been counted, they were ignored; if the recorder was uncertain, they were counted.
Table I. Summary of butterfly counts at Chase Lake National Wildlife Refuge, 1995 and 1996
Date Transect Counts Checklist Counts
06-Jun-95 5 6
14-Jun-95 4 4
27-Jun-95 3 3
03-Jul-95 4 4
02-Aug-95 4 4
15-Aug-95 4 4
Date Transect CL001 Transect CL002
14-Jun-96 2 1
26-Jun-96 3 2
04-Jul-96 3 3
15-Jul-96 4 4
02-Aug-96 3 3
09-Aug-96 4 4
22-Aug-96 4 4
Summary
Total 1995: 49
Total 1996: 44
Both years: 93
Return to Contents
Next Section -- Format of Listings
