USGS - science for a changing world

Northern Prairie Wildlife Research Center

  Home About NPWRC Our Science Staff Employment Contacts Common Questions About the Site

Canvasback Mortality from Illegal Hunting on the Upper Mississippi River

Results


Canvasback population

In 1991, canvasbacks did not begin moving into Lake Onalaska in large concentrations until 10 October when more than 3,000 were observed. A peak population of 17,330 canvasbacks was estimated in Pool 7 on 30 October (Table 1). In the aftermath of a severe winter storm on 1 November 1991, most canvasbacks left the area.

In 1992, canvasbacks arrived in Lake Onalaska during the first week of October and remained through the hunting season until ice covered the lake in early December. A peak population of nearly 50,000 canvasbacks was estimated on 26 October (Table 1). For the first time in 4 years, most canvasbacks remained in the open lake area (closed to waterfowl hunting) of Lake Onalaska and out of the shallow marsh habitat. In comparison, a peak population of about 107,500 canvasbacks was estimated on Pool 7 in 1978 (C. E. Korschgen, Natl. Biol. Serv., unpubl. data).

Table 1. Weekly canvasback numbers from aerial surveys of Pool 7 of the Upper Mississippi River, 1991-1992.
   
Number of canvasbacks in areas of
Lake Onalaska
Year
Survey date
Closed to
hunting
Open to
hunting
Total
1991
 
  
 
3 - 6 Oct
13 - 15 Oct
19 - 21 Oct
26 - 30 Oct
8 - 10 Nov
0
10,520
10,120
17,265
60
0
655
58
65
FROZEN
0
11,175
10,178
17,330
60
1992
30 Sep
3 - 6 Oct
13 - 15 Oct
19 -21 Oct
26 - 30 Oct
3 Nov
8 - 10 Nov
16 Nov
24 Nov
4 Dec
0
500
28,935
29,625
49,560
46,670
28,595
32,475
19,095
1,200
0
0
10
0
15
100
0
0
0
FROZEN
0
500
28,945
29,625
49,575
46,770
28,595
32,475
19,095
1,200

Hunters and hunter information

During the 17-day observation in 1991, 576 parties entered the Lake Onalaska study area from boat landings where vehicle counts were used to estimate hunter numbers. During the 26-day period in 1992, an estimated 1,005 hunting parties entered Lake Onalaska from these boat landings.

During 1991, 2,030 hunters in 1,087 boats entered the study area with an average of 1.9 hunters/boat (SD = 0.8). The average length of hunt of the parties was 3.2 hours (SD = 1.5). These parties contributed hunting effort of 6,605 hunter-hours in the Lake Onalaska study area. During 1992, 2,800 hunters in 1,504 boats entered the study area with an average of 1.9 hunters/boat (SD = 0.8). The average length of hunt of the parties was 3.3 hours (SD = 1.6), and 9,357 hunter-hours were estimated.

Interviewers contacted 52 parties with 111 hunters during 16 October - 1 November 1991 and 157 parties with 316 hunters during 14 October - 8 November 1992. Length of hunt of waterfowlers interviewed averaged 3.6 hours (SD = 1.8) in 1991 and 3.5 (SD = 1.5) in 1992. Most hunters (97% in 1991, 100% in 1992) indicated they were aware that canvasbacks were on the river and vulnerable to hunters.

Observations by law-enforcement personnel

Law-enforcement officers observed 102 hunting parties for 162 hours in 1991 and 156 parties for 257 hours in 1992. During the 1991 waterfowl hunting season, 119 canvasback flocks flew within shooting distance of 43 of the 102 (41%) hunting parties. During 1992, 88 canvasback flocks flew within shooting distance of 51 (33%) of the 156 parties. Canvasbacks were usually encountered by hunters as singles (49.8% frequency of occurrence). Number of canvasbacks/encounter was higher (t = -2.97, 96 df, P = 0.004) in 1992 (GIF -- Mean symbol = 3.98, SE = 0.68, median = 2) than in 1991 (GIF -- Mean symbol = 1.91, SE = 0.14, median = 1).

Twenty of 43 (47%) parties that encountered canvasbacks in the Lake Onalaska study area shot at the birds in 1991, and 21 of 51 (41%) parties with encounters shot at birds in 1992. Hunting parties shot at canvasbacks during 29 of 119 (24%) encounters in 1991 and during 27 of 88 (31%) encounters during 1992 (Table 2). Five canvasbacks were observed killed or crippled during 1991 and 6 during 1992.

The mean rate of canvasback encounters/hour was higher (t = 2.73, 133 df, P = 0.007) in 1991 ( GIF -- Mean symbol = 1.03, SE = 0.20) than in 1992 (GIF -- Mean symbol = 0.45, SE = 0.07); attempts to shoot at canvasbacks averaged 0.42/hour (SE = 0.14) in 1991 and 0.15 (SE = 0.05) in 1992 (t = 1.84, 126 df, P = 0.068). The average rate canvasbacks were shot at and downed was 0.05/hour (SE = 0.03) in 1991 and 1992 and did not differ (t = 0.80, 124 df, P = 0.423) between years.

Table 2. Numbers of canvasback encountered, shot at, and killed or crippled by hunters as observed by law-enforcement personnel in the Lake Onalaska study area, Navigation Pool 7 of the Upper Mississippi River, 1991-1992.
 
 
 
Year
 
 
Parties
observed
 
 
Total hours of
observation
 
 
Canvasback
encountered
 
Attempts to shoot
at canvasbacks
(% of encounters)
Canvasback
killed or
crippled
(% of attempts)
1991 102 162 119 29 (24) 5 (17)
1992 156 257 88 27 (31) 6 (22)
Overall 258 419 207 56 (27) 11 (20)

Attempts by hunters to shoot canvasbacks

The total number of parties (Ni) present in each Lake Onalaska study area stratum was based on both the number of vehicles present at access points and boat observation data (Table 3). We assumed 1.38 vehicles were used/ party during the first season segment of 1991, 1.09 vehicles/party during the second season segment, and 1.18 vehicles/party during 1992 based on hunter survey data. Estimation of hours hunted in each stratum (Hi) from boat observation data was direct, Hi from vehicle count data was based on an average hunt length of 3.57 hours for 1991 and 3.45 hours in 1992. The estimated number of attempts to shoot at canvasbacks was 790 (95% CI = 376) during 1991 and 837 (95% CI = 390)during 1992.

The estimated total attempts to shoot canvasbacks was lower in 1991 because a severe winter storm on 1 November terminated the season. The number of shots at canvasbacks and subsequent killed or crippled canvasbacks might have been much higher in 1991 if hunters had the opportunity to shoot at canvasbacks through the remainder of the season at a rate similar to that observed at the end of October. Had canvasback numbers and attempt rates continued at similar levels during the remaining 9 days of the season, we extrapolated that the 1991 estimate of shots at canvasbacks would have been 1,134.

Applying the ratio estimator model to number of canvasbacks shot and downed, we estimated 128 (95% CI = 128) canvasbacks were killed or crippled during the 1991 hunting season and 166 (95% CI = 148) killed or crippled in 1992. During 1991 observations, 5 of 29 (17%) canvasbacks shot at were downed in Lake Onalaska. Six of 27 (22%) canvasbacks were killed or crippled during 1992.

Table 3. Values used in ratio estimator of attempts to shoot at canvasbacks in the Lake Onalaska study area on Navigation Pool 7 of the Upper Mississippi River during the 1991 and 1992 hunting seasons.
Year
Season
Time of day
Na
nb
Hc
xd
ye
1991
 
 
 
16 - 20 Oct
 
21 Oct - 1 Nov
 
Morning
Afternoon
Morning
Afternoon
401.12
304.14
437.47
520.24
18
11
22
51
1,449.51
941.25
1,616.81
1,555.12
24.62
23.73
46.58
66.80
1
0
9
18
1992
 
14 Oct - 8 Nov
 
Morning
Afternoon
1,361.54
1,147.49
70
86
5,029.69
3,377.66
121.47
135.85
10
17

    aParties in stratum i.
    bParties under observation in stratum i.
    cHours hunted by all parties in stratum i.
    dHours parties were observed hunting in stratum i.
    eAttempts to shoot canvasbacks by parties under observation in stratum i.

Previous Section -- Methods
Return to Contents
Next Section -- Discussion

Accessibility FOIA Privacy Policies and Notices

Take Pride in America logo USA.gov logo U.S. Department of the Interior | U.S. Geological Survey
URL: http://www.npwrc.usgs.gov/resource/birds/canbacks/results.htm
Page Contact Information: Webmaster
Page Last Modified: Friday, 01-Feb-2013 18:36:20 EST
Sioux Falls, SD [sdww55]