Policies
The Twin Cities Canada goose program policy requiring local units of government to implement urban goose management programs has remain unchanged for 15 years. While legal challenges occurred in 1982, 1993 (MNDNR 1994), and 1996, none have stopped or modified the program. The requirement that LGUs with adequate open space for hunting permit shooting as part of a goose management plan has resulted in 3 cities and 2 county park system opening or keeping open portions of their jurisdictions to goose hunting.
Population Growth
Hawkins (1970) reported 480 geese in the Metropolitan Twin Cities in 1968, Cooper and Sayler (1974) found 1,500 birds, and Cooper (unpubl. data) estimated the population at 14,000 in 1984. An exponential model fit (Figure 1) to these data was highly significant (P<.001). Similar growth of a re-introduced Canada goose population in Ontario was reported by Ankney 1996. Determination of the carrying capacities of the Ontario and Twin Cities populations would require unrestricted growth in the future, which is unlikely given the conflicts with human activities at current levels. However, if nest habitat is the limiting factor--we believe it is--then an approximation of the Twin Cities maximum population can be made using the wetland area available for nesting, typical Canada goose nest densities, and the ratio of total geese to nesting pairs. Canada geese use type 3, 4, and 5 wetlands plus the lake shorelines for nesting (Cooper 1978). Assuming that the Metro Area geese will nest at densities similar to those recorded for B. c. maxima and B. c. moffitti breeding the Midwest
and western U.S. and Canada, where the average nest density approaches 3 nests/ha (Cooper 1978). With
39,343 ha of Type 3, 4, and 5 wetlands in Twin Cites, we estimated that the area
could support 118,000 breeding pairs, or a summer population of 472,000 geese
(236,000 adults plus 236,000 young), assuming a conservative production of 2 goslings
per nesting attempt (Sayler, 1977). If the birds were to nest colonially, as they do
on islands (Ewaschuk and Boag 1972), then, the maximum population would be higher.
Moreover, because Canada geese do not breed until 2 or 3 years of age (Cooper 1978)
and these non-breeding birds make a molt migration to northern areas in summer
(Zicus 1981, Fashingbauer 1993), the fall population when the molt migrants return
are typically twice that of summer. Thus, if the summer population were to approach
one-half million geese, the fall population would likely be >1 million.
Geese were found on 107 of 254 (42%) wetlands randomly
surveyed wetlands in 1994. The amount of open shoreline, estimated to the nearest
25%, ranged from none (36%), 1-25 (30%), 26-50 (11%), 51-75 (19%), and 76-100 (4%).
Wetlands with geese had a significantly higher (P<.05) amount of shoreline mowed or
pastured than those without geese (Figure 2). To assess the relationship between the
a length of shoreline habitat and geese using the wetland, we computed shoreline
length as circumference of a circle with an area equal to that of the wetland.
Shoreline grass-pasture length was then calculated multiplying circumference by the
percent of shoreline in mowed or pastured grass. This yielded a minimum shoreline
estimate for the respective wetlands inasmuch as few of the water areas were round.
We found a strong relationship between the amount of grass shoreline and the number
of geese at site (Figure 3).
During the 1994 survey 2,313 geese were counted on the 254
wetlands sampled. The expanded population estimate was 25,000 birds; less than 25%
of the level predicted by the exponential model (Figure 1). Geese were found most
frequently on type 5 wetlands (47%), lakes (42%), and least often on type 4 marshes
(28%); group sizes averaged 25 birds on type 4, 28 in type 5, and 34 on lakes, and
were not significantly different (P>.05). Goose densities per km2 of wetland and
lake were significantly higher (P<.05) in the cities and towns closed to hunting
(61/km2) than those open to goose shooting 22/km2).
Goose Complaint Levels
Goose complaints were received for 391 locations during the 1982-96 period and 176 of these sites visited at the time of the complaint. Summer brood-rearing period problems were most common (94%). The 23 fall complaints came from primarily from golf courses (11), athletic fields (3), and airports (3). Most complaints came from residential sites (52%), followed by parks (17%), golf courses (16%), swimming beaches (10%), and corporate grounds (6%). The geese causing a complaint was highly variable, ranging from 4 to 456 and averaged 70 birds, SD=62. Among complaint classes, commercial sites had the lowest tolerance threshold (59) and beach areas the highest (87); ANOVA showed no
significant differences (P>.05) between class means. Regression of threshold by year indicated no relationship (P>0.05), suggesting that the public has become neither more or less tolerant over time. Complaints per year were cyclic with peaks in 1988 and 1993, caused we believe by increased media coverage in the late 1980s and the during the legal suit in 1993.
Goose Redistribution Techniques
Short-term, goose redistribution methods were recorded as low (no or little effect), moderate (worked but the geese returned), and high (birds were displaced and stayed away). Of the 12 techniques used, 8 were rated low, 2 moderate, and 2 high. Fences, both permanent and temporary, blocking access during the brood-rearing period in June and July were most effective (Table 2). Harassment with dogs during brood-rearing was moderately successful, whereas dogs, particular border collies, were highly
effective on flying birds in fall. Geese avoided areas sprayed with methyl anthranilate but the effect lasted <3 weeks and the material was expensive ($396/ha/treatment).
Productivity Reduction
The effectiveness and costs of egg removal as population control procedure were measured during the 1990-96 period. Nest searching by 2-person teams was done on 3 type-5 wetlands totaling 278 ha. Searching was done by walking, wading, and canoeing. When a nest was found, the clutch size was recorded, and the stage of incubation determined from egg laying or by floatation. To prevent renesting (Cooper 1978), egg removal was delayed until at least the 18th day of incubation. We located 342 nests and removed 1,745 eggs over the 7-year period (Table 3). Two of the 3 sites were fenced, thus the proportion of nests not found could be estimated from the broods observed. Over the period, 8 broods were observed. Assuming a 75% nest success (Sayler 1978), we estimated that 97% nests were discovered.
Table 2. Ranking of effectiveness of short-term goose redistribution methods used in the Metropolitan Twin Cities, 1982-1996.
_________________________________________ Method N Rankinga _________________________________________ Decoy-dead goose 1 L Decoy-swan 156 L Feeding Ban 94 L Fencing-electric 4 H Fencing-permanent 15 H Fencing-Temporary 182 M Habitat modification 15 L Harassing-dog 8 M Harassing-people 310 L Harassing-vehicle 93 L Live swan 7 L Methyl anthranilateb 3 L _________________________________________ aL= low effectiveness, M= moderately effective, and H= highly effective. bRejects-It.TM
Table 3. Canada goose nests and eggs found and time spent searching by two-person teams, Gun Club Lake, Mother Lake, and Wood Lake, Twin Cities, Minnesota 1990-1996.
_________________________________________________ Year Nests Eggs Hours Eggs/Hour _________________________________________________ 1990 39 202 96 2.1 1991 43 218 119 1.8 1992 38 195 125 1.6 1993 45 216 175 1.2 1994 42 162 144 1.1 1995 110 626 171 3.6 1996 25 126 176 0.7 _________________________________________________ Total/Mean 342 1,745 1,006 1.7 _________________________________________________Trap and Relocate
A total of 40,515 flightless geese was trapped and removed
from the Twin Cities between 1982 and 1996 (Table 1). Trapping efficiency averaged
98%, range 96 to 99%, of all flightless geese present. Cities undertaking relocation
control increased from 1 in 1982 to 36 in 1996 and trapping sites increased from 1 to
104 (Figure 5). Capturing and relocating geese reduced populations significantly.
The procedure was most effective for urban concentrations and least for rural
populations. The geese found at a capture site decreased rapidly during the first 5
years, then more slowly in subsequent years. Overall, after 5 years of continuous
removal, the population was typically 60% lower, and after 10 years, an 80% reduction
was attained (Figure 6). Relocation was most effective at urban sites, where a 75%
decline occurred in 5 years, whereas suburban and rural populations declined 40%.
In 1982, 195 immatures were banded and relocated from Lake
of The Isles to Carver Park Reserve 32 km SW of the trap site, and 265 adults banded
and sent to Oklahoma. In 1984, 230 immatures and 439 adults were banded and shipped
to Oklahoma, and in 1985, 257 adults were banded sent to Oklahoma. Other immatures
were released at Minnesota sites 80+ km from the capture site in 1985 and subsequent
years. Legbands were placed on 256, 200, and 489 of these birds in 1984, 1985, and
1986, respectively.
The rate of return of relocated geese to the capture sites was lowest for immatures and highest for adults. Of the 195 flightless goslings relocated to Carver Park, 8 of 195 (4%) were observed or trapped at Lake of the Isles. Nine of 256 (4%) of the immatures released in Oklahoma in 1984 returned to the capture sites, and none of the 945 immatures released in Minnesota from 1984 to 1986 were retrapped or observed in the Twin Cites to date. In contrast, 57 (22%), 186 (42%), and 62 (24%) of relocated adults were observed, retrapped, or recovered in the Twin Cities. Based on a 71% annual survival rate (unpubl. data, Oklahoma Cons. Dept.) and assuming that all relocated geese returning to the Twin Cities were encountered, the proportion of relocated adults returning was estimated to be 42%, 80%, and 42% for 1982, 1984, and 1985, respectively.
Trap and Process
The processing of Twin Cities Canada geese for human food
was approved by the U.S. Fish and Wildlife Service in 1995. A total of 325 adult
geese were trapped and slaughtered in 1995, and 1,770 in 1996, including 154
immatures. St. Paul and Minneapolis foodshelf distribution centers agreed to
distribute the frozen goose products, provided the geese were processed in either a
USDA or MNDA inspected plant. The distribution centers also agreed to hand out a
postage-paid client questionnaire. The questionnaire asked the client whether he/she
had had goose in the past 10 years, how the client ranked the product taste (0-lowest
to 10-highest), and if he/she would use the product again. Distribution center
operators indicated that they had difficulty obtain donations of high protein
products and that the centers would take "all" the goose products we could supply.
The client surveys was small, 17 in 1995 and 36 in 1996, due to the confidentiality
maintained at food shelves, but the results clear. Most (88%) clients had not had
goose in the past 10 years, they rated the taste at 8.4 (range 6-10) in 1995 and 8
(5-10) in 1996. Sixteen 1995 clients (94%) and 32 (89%) 1996 clients indicated
that they would eat the goose product again.
In 1995, the 125 geese held in a 1.5 ha pen consumed 8 kg
food/month and required 8 hr/month care, or about $6/goose to hold until September.
Geese pasture rental and labor from July until November 1996 cost $6/bird . The
processing plant costs ranged from $6 to $8/goose, thus, the total costs per goose
for holding and/or processing was from $6 (process in July), to $12 (feed and
process in September), to $14 (pasture and process in November). Yields varied by
the product produced. The birds butchered in July 1995 were made into ground breast
and boneless breast. These products averaged 0.7 kg/bird. The 103 birds processed
in September 1995 were packaged as whole breasts and legs because abundant pin
feathers precluded an acceptable whole carcass product. This group yielded 1.6
kg/bird of meat. The 125 geese butchered on 11 January 1996 were packaged "in the
whole" with giblets and weighed an average 4.1 kg. The geese processed in the summer
and fall of 1996 were whole carcass products; the July products averaged 2.7 kg,
while the November geese yielded 3.2 kg./bird.
A literature search for potential contaminants was conducted
prior to the 1995 processing and indicated that, except for those from heavy
industrial areas, the geese posed no human health risk (Keefe 1996). The Minnesota
Department of Health concurred with this conclusion. Because an industrialized area
was added in 1996, the tissues of 3 brood-patch females trapped at this location were
analyzed by MNDA. Breast muscles and their livers were tested for PCBs, mercury, and
organochlorine pesticides. None were found at detection levels of 0.025 ppm for PCBs
and pesticides, or at 0.125 ppm for mercury.
Habitat Reduction
Based on the relationship between the number geese using a wetland and the shoreline in mowed-grass or pasture (Figure 3), it was clear that a modification of the shoreline habitat would reduced the number of geese at a site. Furthermore, if sufficient shoreline was converted from grass, the population could be limited. Using areas of the 3,081 Meto wetlands, we estimated that Twin Cities has a minimum of 3,550 km of shoreline. Based on estimates of grass shoreline made at 227 wetlands in 1994, one quarter (25.1%) of the shoreline is in mowed grass or pasture. This means that about 888 km of shoreline would have to be converted to alternative vegetation. Because we observed geese leading broods through up to 70 m of cattail to graze on grass beyond the wetland edge, we assume that at least this width of shoreline vegetation would have to be converted to non-grass plantings.
Hunting
Kill data from the MNDNR surveys indicate that goose harvest during the early September special season ranged from 2,782 to 9,979, and late season harvest from 376 to 895 (Table 4). Because the latter was incorporated into the regular goose season extending from on or near 1 October to mid-December, there were no data after 1992. However, if the harvest has been similar to that of the 1978-91 period, from 9,000 to 10,000 additional geese were killed during the special hunts. Combining with removals of 6,000 geese/year in the past 2 years (Table 1), the impact on the Metro goose population has been significant.
Table 4. Canada goose hunting harvest during the special Metro September and December seasons, Twin Cities, Minnesota 1987-1994. MNDNR data.
_______________________________________________________ Year September December Total _______________________________________________________ 1987 4,443 783 5,226 1988 2,818 491 3,309 1989 2,782 376 3,158 1990 4,673 392 5,065 1991 5,497 895 6,392 1992 8,325 a a 1993 9,532 a a 1994 9,979 a a _______________________________________________________ Totals 48,049 2,937 50,986 _______________________________________________________ aLate season incorporated into the regular season framework in 1992 and thereafter.