Effects of Human Activity
Colony-nesting birds can be readily exploited, and cormorant bones occur widely in middens. In historical times, cormorant eggs were used for making soap (Van Tets 1959) and as food for humans and animals, the skins for clothing, and carcasses for bait (Hatch 1995a). This cormorant has been widely perceived as a competitor by commercial and recreational fishermen (see Conflicts, below) and subject to extensive persecution, particularly by destruction of nests, eggs, and young, but also shooting of adults. Such actions probably account for widespread decline of this species in the 19th century (see Distribution: historical changes, above). Regional differences in wariness of nesting birds has been thought to reflect differences in this persecution (Bent 1922). Human disturbance, including research disturbance, can lower productivity of cormorant colonies (Kury and Gochfeld 1975, Ellison and Cleary 1978, DesGranges and Reed 1981, Carney and Sydeman 1999, Strombrog et al. 2012).
Shooting And Trapping
With changes in policy and regulation in the U.S. allowing lethal control (shooting and egg-oiling) of cormorants, information on effects of control has increased (Dorr et al. 2012c). Modeling efforts indicate that lethal culling of adults would result in the most rapid reduction in numbers versus egg-oiling alone (Blackwell et al. 2002). Lethal control by shooting on breeding colonies does result in reduced numbers of local breeding cormorants (Bedard et al. 1999, Strickland et al. 2011, Dorr et al. 2010a, Dorr et al. 2012c). Shooting at nesting colonies may also have substantial additional effects through subsequent year emigration to alternative breeding sites, male biased culling, disturbance and resulting mortality of young (Ewins and Weseloh 1994) (Bédard et al. 1997c, Bedard et al. 1999, Scherr et al. 2010, Strickland et al. 2011, Dorr et al. 2012c, in press; see Management, below). This is particularly the case when gulls and other species that prey on cormorant young may be present (Kury and Gochfeld 1975, Ellison and Cleary 1978, Duerr et al. 2006).
With respect to trapping, historically native Americans caught sleeping cormorants (probably Double-crested) by hand in the 17th century (Hatch 1995a). Trapping of cormorants is typically done for research purposes, usually with softcatch leg-hold traps or with nets (King and Tobin 2000), which impacts relatively few birds directly, although colony disturbance due to trapping, as with other human disturbance, can be significant. Trapping and handling for radio and satellite tag marking has been shown to have at times substantial effects on individual colony fidelity and movements (Scherr et al. 2010).
Pesticides And Other Contaminants/Toxics
Extensive literature includes field, aviary, and lab studies of levels and effects of contaminants on this species (Elliott et al. 1989b, Bishop et al. 1992, Powell et al. 1997b, Rattner et al. 1999). Cormorants acquire contaminants from the fish they eat. Influence of contaminants increased during the 1960s and 1970s, when cormorant populations declined dramatically and various other biological effects (see below) were commonplace. Breeding populations that accumulated the greatest burdens and showed the most severe impacts were in the Great Lakes (Weseloh et al. 1995b), Gulf of St. Lawrence (Pearce et al. 1989), and along the Pacific Coast of s. California (Gress et al. 1973). In the Great Lakes, the species was extirpated from Lakes Michigan and Superior, probably because of chemical contamination; in 1970–1972, contaminant levels (µg/g wet weight) in eggs were: for DDE, 18.56; for PCBs (Aroclor 1260), 27.25; for dieldrin, 0.47; for hexachlorobenzene, 0.18 (Postupalsky 1978b, Ryckman et al. 1998). Cormorants breeding in Maine and on Canadian prairies had lower levels (Kury 1969, Vermeer and Reynolds 1970). Cormorants wintering in the Houston Ship Channel, TX, accumulated polychlorinated styrenes from Nov to Feb (King et al. 1987).
Effects of contaminants on cormorants have been studied most intensively on the Great Lakes; they include eggshell-thinning (Anderson and Hickey 1972, Postupalsky 1978b), elevated embryonic mortality (Gilbertson et al. 1991), reproductive failure and population declines (Weseloh et al. Weseloh et al. 1983, Weseloh et al. 1995b; see also Gress et al. 1973 for California), increased adult mortality (Greichus and Hannon 1973), increased embryonic abnormalities and crossed bills (Figure 9; Fox et al. 1991, Yamashita et al. 1993, Ludwig et al. 1996), egg mortality (Tillitt et al. 1992), brain asymmetry (Henschel et al. 1997), and induction of cytochrome P–450 1A1 (Sanderson et al. Sanderson et al. 1994). Crossed bills were also observed in Massachusetts (JJH).
Contaminant levels began to decrease in the 1970s and continued to do so through the 1990s, with improvement in most associated biological parameters. By 1995 in the Great Lakes, most contaminant levels in cormorant eggs had declined by 83–94% and were as follows (for same sites as above in early 1970s): DDE, 2.80; PCBs (Aroclor 1260), 4.74; dieldrin, 0.06; hexachlorobenzene, 0.01 (Ryckman et al. 1998). However, contaminant levels in Great Lakes cormorants remain much higher than in most other areas of North America (cf. Somers et al. 1993, Sanderson et al. 1994), and biological effects persisted into the 1990s (Fox et al. 1991). Prevalence of deformities and of hatching failure were higher at more contaminated sites (bill deformities to 52/10,000 in Lake Michigan). Various polyhalogenated aromatic hydrocarbons (including DDE, PCB, and dioxins) were suspected as causes, but links have not been firmly established. Populations in Washington State, which showed low contaminant levels but continued low productivity, had been disturbed by humans (Henny et al. 1989). Dioxin equivalents in eggs from the Great Lakes ranged from 155 to 382 pg/g wet weight and showed a 31.3 mean bio-magnification factor from forage fish to cormorant eggs (Jones et al. 1994, Williams et al. 1995b).
Little work has been done with metals in cormorants, and no effects have been identified in the wild. Mercury is most often reported: mean levels in eggs 0.11–0.83 µg/g wet weight (Heinz et al. 1985, Noble and Elliott 1986, Henny et al. 1989). In New Brunswick, total mercury concentrations in tissues of Double-crested Cormorants were highest of 9 seabird species (Braune 1987a). For other locations, tissues, and elements, see Mora and Anderson 1995 (Baja California), Larson et al. 1996 (Lake Michigan), and Sepulveda et al. 1998 (Florida).
Very susceptible to oiling, like other surface-swimming and diving seabirds, but cormorants are only a small fraction of birds reported killed in oiling accidents and no major kills reported (Clapp et al. 1982a).
Ingestion Of Plastics, Lead, Etc.
Not reported in pellets (see Food habits: diet, above, for sources).
Between 1990 and 2011 double-crested cormorants were reported in 90 incidences of strikes with aircraft with damage estimated at circa $2.99 million (Dolbeer et al. 2012). Although the total number of reported strikes for cormorants is comparatively low, in a relative hazard score ranking of 77 wildlife species, cormorants ranked 9th, largely due to their large size and the amount of damage caused in each strike (Devault et al. 2011). Collision incidences are also occasionally reported at power lines.
In marine areas, known to be caught on hooks (especially live-baited hooks) and in gill-nets, lobster traps, and trawls. In freshwater areas cormorants are occasionally caught on trot lines (BSD) and sometimes in commercial nets, particularly in the Great Lakes but no published data on incidence are available for the inshore waters frequented by most individuals. In some areas locally significant fishery-related mortality has been through direct killing by anglers (see Conflicts, below). These incidences are generally isolated, particularly in the last 40 yr, and are unlikely to have population level effects.
Alteration Of Habitat
Clearing of forested wetlands reduced available inland habitat for nesting and foraging, especially in the South. Extensive new foraging habitat has been created by dams, and aquacultural ventures, particularly in the southeastern U.S. (Jackson and Jackson 1995). However, primarily due to market and economic factors, the aquaculture industry in the Southeast has been in steep decline (Bastola and Engle 2012, Dorr et al. 2012b); the effect of this decline on cormorant numbers is unknown. Increases in food are not limited to such relatively new feeding areas but include large changes in fish communities following introductions of new species and other changes (see Christie et al. 1987).
The value of impoundments to cormorants is greatly enhanced by intentional and systematic stocking with sport and forage fish to promote recreational fisheries. Human-made impoundments also provide additional habitat for native and introduced forage fish such as gizzard shad and carp. Alterations in fish communities through unintentional introductions of non-native species such as alewife and more recently round goby have been shown to be important in cormorant population resurgence and diet (Weseloh and Ewins 1994, Belyea et al. 1999, Somers et al. 2003, Seefelt and Gillingham 2006, Seefelt and Gillingham 2008, Johnson et al. 2010, Coleman et al. 2012, Debruyne et al. 2012, DeBruyne et al. 2013; see Food Habits).
Cormorants have been shown to readily adapt foraging strategies with perturbation in fish communities (See Diet Section). Impoundments without islands or trees are unsuitable for breeding. Degradation of nesting sites results chiefly from the birds' own actions and is well documented; they pluck twigs for nest material, and their feces kill trees in a few years largely due to changes in soil chemistry (Lemmon et al. 1994) Weseloh and Ewins 1994, Shieldcastle and Martin 1999, Cuthbert et al. 2002, Hebert et al. 2005, Somers et al. 2007, Boutin et al. 2011, Koh et al. 2012).
Disturbance At Nest And Roost Sites
Very susceptible to disturbance at mixed ground colonies; where nesting with gulls, cormorants are first to leave and last to return, providing ample opportunity for the gulls to eat regurgitated fish, the cormorants' eggs, and newly hatched young, which they do readily, and often in that order (Kury and Gochfeld 1975, JJH). Hasty departures caused by sudden disturbances lead to eggs being tossed from nests because the incubating parent holds eggs on its feet. However, the species is well adapted to such losses and relays readily and rapidly.
The synergistic effects of disturbance by Bald Eagles and humans in enabling predation by crows on Mandarte I., British Columbia, were described by Verbeek (Verbeek 1982). Predation by Fish Crows (Corvus ossifragus) is facilitated by disturbance; may restrict coastal nesting in the Carolinas and favor nesting inland (Post 1988c). Frequent human visits caused gull predation and nest abandonment, and discouraged settlement by late-nesting cormorants in Quebec (Ellison and Cleary 1978). Disturbance by people or Bald Eagles may lead to slow shrinking of a colony, or abrupt shift to a new site. Predation on young and eggs by egg-oiling caused higher emigration to alternative breeding colonies in subsequent years on Lake Champlain (Duerr et al. 2006). When, gull predation was largely eliminated by egg-oiling at night, colony abandonment in subsequent years was substantially reduced (Duerr et al. 2006). Gull predation itself may have a greater impact on colony fidelity of cormorants than human disturbance alone. Long term management (> 5 yr), including both lethal culling and egg-oiling to reduce breeding cormorant numbers have not caused complete colony abandonment at locations in Lake Ontario (Farquhar et al. 2012) and Lake Huron (Dorr et al. 2010a), despite the presence of gulls.
Young cormorants are very sensitive to disturbance, especially during the first 2 wk of life, before they can thermoregulate (see Breeding: parental care, above). Flushing of adults from their nests at this time may lead to significant mortality of young because of exposure to sun (Weseloh et al. 1995b). Small chicks may die in 11 min (mean 22.7 min ± 11.6, n = 38) from such exposure, at deep body temperature 45.7°C ± 1.7, but shorter times are sufficient to cause larger chicks to move and thus fall from nests. When exposed to cold, nestlings became comatose at 16–19°C but recovered upon warming; cooling to 11.5°C was fatal (Van Scheik 1985). In ground colonies, young older than about 21 d leave their nests and may enter the water; they are thought to return to nests later, but effects of such disturbance have not been examined systematically.
Nocturnal visits to arboreal roosts in Mississippi for the purpose of capturing and radio-tagging appeared to have little effect on use of those sites by the birds (King et al. 1995a) but systematic analyses have not been conducted. However, extended harassment at nocturnal roosts with pyrotechnics reduced both numbers of cormorants using the roost and those feeding at nearby catfish farms (Mott et al. 1998): use of pyrotechnics to harass birds from night roosts in aquaculture-producing areas has been for many years a recognized control method (Mott et al. 1998, Glahn 2000, Glahn et al. 2000).
Since modification of the Aquaculture Depredation Order in 2003 (USFWS 2003, Barras and Tobin 2003, Godwin et al. 2003, Werner and Hanisch 2003, Dorr et al. 2012c), some lethal control has been allowed in night roost locations in aquaculture producing areas. Research is limited but Glahn (2000) has shown pyrotechnics and lethal control appear comparable with respect to effectiveness in night roost dispersal (Glahn 2000).
Direct Human/Research Impacts
Cormorants have long been viewed with antipathy, and there is a long but poorly documented history of persecution (Duffy 1995b). Conclusions of early studies (e.g., Taverner 1915, Lewis 1929, Mendall 1936a)
were that cormorants have only small effects on openwater fisheries.
More recent work has shown that cormorants can impact fisheries but
measuring their impact is difficult and interpretations can be ambiguous
(see below). Increases in numbers since 1980 have led to increased
conflicts on 3 main fronts: (1) potential impacts to sport fisheries in
marine and freshwater areas, (2) depredations at commercial aquaculture
facilities (primarily catfish) and (3) alteration of vegetation and nest
trees, resulting in vegetation and habitat damage, lowering of property
values and impacts on other colonial waterbirds. Less attention has
been given to complaints of fouling vessels or buildings, of
eutrophication or lowering water quality, and of possible transmission
of fish diseases or parasites (except in aquaculture).
Open Water Fish. Compelling evidence that cormorants seriously damage fisheries is
limited but has been increasing (Dorr et al. 2012c, Dorr and Somers
2012). Comprehensive studies evaluating cormorant impacts to fisheries
are likely limited owing to cost, complexity, and duration necessary
to draw reliable conclusions -- and even then results may be ambiguous.
Diet alone is not a measure of impact. It is often assumed that small
percentages of fish of commercial or sporting interest in the diet
indicate no impact. However, many diet studies are conducted during
periods when sportfish are not typically consumed and their populations
are at low levels historically, contributing to low consumption rates
(Trapp et al. 1999, Johnson et al. 2002, Diana et al. 2006, Dorr and
Somers 2012, Van Guilder and Seefelt 2013). Diet studies are often
conducted during incubation and chick rearing when cormorants are tied
to nesting areas and bolus's and regurgitate are easily sampled (Dorr
and Somers 2012, Van Guilder and Seefelt 2013); missing important
periods of sportfish consumption during spring and fall (Johnson et al.
2002, Diana et al. 2006, Dorr et al. 2010b, Dorr and Somers 2012; e.g.
87% of consumption of Yellow Perch and Northern Pike occurred prior to
May 15 and after September 1, in Lake Huron, Michigan.)
Cormorant consumption of sportfish is often compared to total number
or biomass of fish populations within a system, which can be misleading.
Cormorants do not randomly consume fish spatially across systems or by
size and age classes within a population. Typically cormorants eat
specific size and age classes and may consume a large percentage of
these age classes, limiting recruitment, even when sportfish consumption
is a small percentage of total diet (Ross and Johnson 1995, Ross and
Johnson 1999, Burnett et al. 2002, Lantry et al. 2002, Rudstam et al.
2004, Fielder 2008, Fielder 2010, Dorr et al. 2010c; see Food Habits, above).
To show that cormorant predation is additive rather than compensatory
to other losses requires studies that are difficult and expensive; such
studies have rarely been performed. It can also be difficult to
estimate cormorant numbers, foraging effort and therefore consumption,
although estimation methods are being developed and evaluated (Pearse et
al. 2006, Dorr et al. 2008, Ridgway 2010b). Even where cormorants do
eat substantial numbers of commercial or sport fish, it can be difficult
to demonstrate damage indisputably (Draulans 1987, Duffy and Schneider 1994, Dorr and Somers 2012; see Demography and Populations: population regulation, above).
There have been some cases were cormorant control has been
implemented and outcomes with respect to open water sport fisheries have
been monitored (Dorr et al. 2010b, Fielder 2010, Dorr et al. 2012c,
Farquhar et al. 2012). In these cases fishery improvements have been
observed (Dorr et al. 2010b, Fielder 2010, Dorr et al. 2012c, Farquhar
et al. 2012). These cases independently provide evidence that cormorants
were impacting local fisheries and management can improve fisheries.
However, the strength of evidence varies for each location, and in some
cases is complicated by other contributing factors (Dorr et al. 2012c;
See below for details). Cormorants (and other bird species) can serve
as a potential vector of Myxobolus cerebralis (Whirling disease) in
trout (Koel et al. 2010), although their role in disease epidemiology in
general is poorly understood.
On Lake Ontario, annual consumption of fish by cormorants was
calculated as 2.88, 1.64, and 0.90 × 106 kg, respectively, for 1991,
1993, and 1994 (Weseloh and Casselman 1992, Ross and Johnson 1995).
Total forage base of fish was estimated as 418 × 106 kg, so cormorants
consumed <1% of available fish, compared to 13.3% taken by salmonids (Weseloh and Casselman 1992).
In Lake Erie, cormorants consumed <2% of biomass of forage fish
taken by walleyes, the major commercial species, and ate fewer fish than
did Red-breasted Mergansers (Mergus serrator), Herring Gulls, or Ring-billed Gulls (Madenjian and Gabrey 1995).
These and other studies conclude that cormorants usually have only
small impacts on fish populations (Trap et al. 1999; see also Food Habits: diet, above). However, such estimates may be at an inappropriate
scale—lakewide rather than local; impacts may differ at smaller scales
(especially for sedentary species; Birt et al. 1987).
For example, smallmouth bass constitute <2% of total prey taken in
the Eastern Basin of Lake Ontario (see Food Habits: diet, above), but this amount may represent 21–35% of all bass available and
could be a substantial impact on that fishery (Schneider et al. 1998).
In the eastern basin of Lake Ontario, Johnson et al. (2002) found
that biomass of smallmouth bass and yellow perch consumed annually by
cormorants exceeded what is taken by sport (bass and yellow perch) and
commercial perch fishermen. The majority of perch consumed by cormorants
were age-1-3 (48%, 20% and 20% respectively; see Burnett et al. 2002).
Even at high perch density estimates cormorants were capable of
consuming 29% of the age-3 perch stock and could reduce recruitment to
the fishery (Burnett et al. 2002).
In 1993-1994, Lantry et al. (2002) found age-specific cormorant
consumption of smallmouth bass in the Eastern Basin of Lake Ontario
yielded a 23% reduction in bass ages 3 to 5. In 1998 a 36% reduction in
age-3 to -5 bass resulted in a 78% decline in recruitment to fishable
stock (Lantry et al. 2002). The above studies indicate that successive
years of cormorant predation on a small number of age cohorts can
potentially cause a recruitment bottleneck to older age classes and
fisheries. This recruitment bottleneck scenario was also observed in the
walleye fishery, in Brevoort Lake, MI (Dorr et al. 2010b).
Cormorants and anglers were reported to have only a small impact on
yellow perch in the Les Cheneaux Islands region (LCI) of Lake Huron in
the mid-1990's (Belyea et al. 1997).
Total annual mortality of yellow perch in the mid-1990's from all
sources was considered low (Diana et al. 2006). Later studies suggested
cormorants were a primary limiting mortality factor and contributed to
collapse of the yellow perch fishery in the early 2000's (Fielder 2004,
Fielder 2008, Dorr et al. 2010c, Fielder 2010).
In marine areas, perceived conflicts have been most severe in the
vicinity of rivers in e. Canada and New England where restoration of
Atlantic salmon is ongoing (Krohn et al. 1995), and at other salmon rivers in the Northwest (Bayer 1989c), but cormorants have been blamed for declines of other species as well.
Cormorants respond rapidly to concentrations of fish or to fish made
vulnerable by human activities. For example, on the Penobscot River, ME,
after release of hatchery-reared salmon smolts, cormorants forage
disproportionately at the dams, where the naive fish are disoriented and
easily caught (Blackwell and Krohn 1997).
Spawning fish are similarly concentrated and vulnerable, which accounts
for some of the seasonal patterns (see Food Habits, above). Cormorant
predation on spring spawning fish has been documented in several studies
(Diana et al. 2006, Dorr et al. 2010b, Dorr et al. 2012c).
Aquaculture. By concentrating desirable prey in places where they can be easily
caught, fish farms may attract many cormorants. In Florida, cormorants
exploited shallow ponds teeming with easy-to-capture prey by becoming
resident in the area and breeding nearby (Schramm et al. 1984). Injuries to fish in pound-nets was described by Craven and Lev (Craven and Lev 1987).
Cormorant depredations at catfish farms, particularly in the
southeastern U.S., have been relatively extensively studied (Glahn et
al. 2000a, Werner 2000, Barras and Tobin 2003, Taylor and Dorr 2003,
Werner and Hanisch 2003, Glahn and King 2004). Increasing conflicts
occurred as aquaculture acreage and cormorant numbers increased through
the 1980's and 1990's (Stickley 1989, Stickley et al. 1992, Mott and
Brunson 1997, Glahn and Stickley 1995, Glahn et al. 1999, Glahn et al.
2000b, Taylor and Dorr 2003, Glahn and King 2004). A large part of the
cormorant's diet in the aquaculture-producing regions was commercially
raised catfish (Stickley at el. 1992, Glahn and Brugger 1995, Glahn et
al. 1995, Glahn et al. 1998, Glahn et al. 1999). Male cormorants appear
to depredate catfish aquaculture in higher proportion than female
cormorants, possibly due to their larger size and ability to consume
large spiny catfish (Glahn et al. 1995, Glahn et al. 1998). Larger males may consume more catfish although this has
not been factored in estimates of loss.
Wintering cormorants may take 4% of the standing crop of stocker-size
catfish (~16 cm) in the Delta region of central Mississippi (Glahn and Brugger 1995).
Captive trials that included non-catfish buffer prey species indicated
that catfish losses at harvest were 30% by number and 23% by biomass per
pond (Glahn and Dorr 2002). Mortality was primarily additive, but compensatory growth of catfish mitigated losses to some degree (Glahn
and Dorr 2002). Glahn et al. (2002) modeled a 20% production loss at
harvest based on average 30 cormorants feeding at a 6-ha catfish pond
for 100 d that resulted in a 111% loss of profits. Higher initial
stocking rates may ameliorate losses to some extent (Glahn et al. 2002).
The above predation rates have been observed in the field but the
extent ponds are impacted at this level in the industry is not known.
Early estimates of loss to the catfish aquaculture industry were
estimated at between (U.S. circa 2000) $5-$25 million annually (Glahn
and Dorr 2002, Glahn et al. 2002). Recent research indicates that
cormorants avoid brood fish ponds and feed on fingerling and foodfish
ponds proportional to their abundance (Dorr et al. 2012a).
Based on foraging distribution and other factors cormorant impacts to
catfish aquaculture can vary greatly, ranging from (U.S.) $5.6 - $12
million during 2000-2004 (Dorr et al 2012b). Additional comprehensive
studies that address pond, farm, and industry level costs and benefits
of cormorant impacts and their management are needed (Jackson and Jackson 1995, Glahn et al. 2000, Dorr et al. 2012b). The catfish aquaculture industry
has been in steep decline in recent years due primarily to increased
energy and feed costs and competition from foreign markets (Dorr et al.
2012b, Bastola and Engle 2012). It is unclear what effect this may have
on cormorant impacts to catfish aquaculture and how the loss of this
heavily used resource may impact the cormorant population (Dorr et al.
Terrestrial Habitats. Cormorants have caused extensive damage to vegetation where they nest
due to excessive guano and physical destruction (Lemmon et al. 1994,
Shieldcastle and Martin 1999, Hebert et al. 2005, Boutin et al. 2011,
Craig et al. 2012). Conflicts can occur, particularly when trees are
rare species, or aesthetically valued. Accumulation of cormorant
droppings, which contribute excessive ammonium nitrogen, phosphorous,
metals and reduce pH and stripping leaves for nest material may kill
trees within 3–10 yr (Lemmon et al. 1994, Boutin et al. 2011, Craig et al. 2012). Changes in soil have been
shown to affect plant species composition resulting in reduced number of
species and opportunity for exotic, invasive plants (Boutin et al.
2011, Craig et al. 2012).
As woody vegetation dies, cormorants will continue to nest on downed
trees or the ground (Weseloh and Ewins 1994), leading to open areas free
of any vegetation (Hebert et al. 2005). Adverse impacts on other
breeding birds may result from alterations to vegetation (e.g.,
Black-crowned Night-Herons [Nycticorax nycticorax] and Common Eiders [Somateria mollissima] in the St. Lawrence River; Bédard et al. 1997c), or occupation of limited nesting space (e.g., Common Terns [Sterna hirundo],
Oneida Lake, NY). Positive impacts may occur to species that nest in
open areas (e.g. American White Pelican, Herring Gull). Avian community
level impacts, whether positive or negative, due to cormorant induced
habitat changes are poorly understood.
From being a Species of Concern in 1970 because of low numbers in many jurisdictions (e.g., Meier 1981), numerical increases have led to widespread perception of Double-crested Cormorant as a pest (Hatch 1995a, Nettleship and Duffy 1995; see also Demography and Populations, above). The species was not included in the Migratory Bird Convention (in 1916) between U.S. and Canada, and consequently in Canada cormorants continue to be under the jurisdiction of provinces (Keith 1995). The species has been protected under federal law in U.S. since 1972, when cormorants and other birds were added to list for U.S.–Mexican Convention (Trapp et al. 1995). Major changes in regulatory status occurred beginning in the late 1990's (USFWS 2009, Dorr et al. 2012c, Weseloh et al. 2012).
The most significant regulatory changes were the establishment of the aquaculture depredation order (AQDO) in 1998 and the public resource depredation order (PRDO) in 2003 (USFWS 2009). These rule changes allowed for significantly greater cormorant management, including egg-oiling and culling by state, federal, and tribal entities in the U.S. to protect commercial and natural resources (USFWS 2009). Cormorant management efforts have also been conducted on their breeding grounds in Canada, but management and policy vary widely on a province-by-province basis (Keith 1995, Bedard et al. 1999, Weseloh et al. 2012).
Measures Proposed And Taken
Shooting and destruction of nests, eggs, and young have long occurred (Lewis 1929, Bayer 1989c). In some areas, such activities continue on a large scale (notably in Manitoba, where flamethrowers have been used in colonies; see Sheppard 1994). In response to complaints from fishermen, large-scale egg-oiling projects, intended to reduce numbers, were implemented in New England (1944–1952; Krohn et al. 1995) and the Great Lakes (1948–1963; Ontario Ministry of Natural Resources unpubl.). In New England, eggs were sprayed with emulsions of oil and formalin. On the West Coast, shooting and destruction of eggs and nests have been the main methods of reducing numbers (Bayer 1989c). Egg-oiling with vegetable or mineral-based oils are now the typical method for inhibiting reproduction and are very effective (Shonk et al. 2004, Duerr et al. 2006, Farquhar 2012), and the only legal method in the U.S.
On the St. Lawrence River, alterations to forested island habitats by cormorant feces led to a 5-yr management plan, based on a population model, intended to reduce numbers from 17,000 to 10,000 nesting pairs (Bédard et al. 1995a). In ground colonies, eggs were sprayed with nontoxic mineral oil, and in tree colonies adults were shot, starting in 1989. Based on evidence for damage to the smallmouth bass fishery in e. Lake Ontario, control measures of oiling eggs and shooting adults were proposed for 1999–2000 by New York State Department of Environmental Conservation.
In recent years, U.S. Fish and Wildlife Service [USFWS] has issued increasing numbers of depredation permits for control of fish-eating birds at aquaculture facilities (Trapp 1998). In 1993–1994, 251 permits were issued, and 8,239 cormorants were reported taken (57% of all birds reported under such permits). In Mar 1998, USFWS established an Aquaculture Depredation Order (AQDO) that allows people engaged in commercial aquaculture to shoot cormorants without a federal permit at freshwater aquaculture premises or state-operated hatcheries (Trapp 1998 Glahn et al. 2000a). This order applies to 13 states where cormorant depredations have been recognized as potentially significant: primarily southern states with substantial production of channel catfish for human consumption or bait fish (Minnesota). This shooting may occur only in conjunction with a certified nonlethal harassment program, and the order requires that records be maintained of all cormorants killed. In other states, depredation permits are issued case by case. Glahn et al. (2000b) reported a take of at least 9,557 birds by catfish farmers in Mississippi alone. However, this take had no apparent impacts on wintering numbers in the region during 1998-99. The USFWS (2009a) reported the total take from 2004 through 2007 under the AQDO in the U.S. was 104,207 birds. Band analyses have indicated management activities on the wintering grounds may be affecting survival of cormorants at the Spider I. colony, Lake Michigan (Stromborg et al. 2012).
In 2003, the USFWS established the Public Resource Depredation Order (PRDO: USFWS 2003, USFWS 2009, Dorr et al. 2012c). This regulation authorizes officials of Federal and State wildlife agencies and Native American tribes to control cormorants to protect fish, wildlife, plants, and their habitat in 24 U.S. states (USFWS 2003, USFWS 2009). The PRDO allows for harassment, egg-oiling and lethal control of cormorants primarily on their breeding grounds. The USFWS has regulatory authority over any actions under the PRDO and must be notified all actions. Actions that would reduce local populations by more than 10% require approval from USFWS and may require a separate Environmental Impact Assessment (EIA) under the National Environmental Policy Act (NEPA; USFWS 2003, USFWS 2009, Dorr et al. 2012c). Numerous management actions have been taken under the PRDO. The USFWS (2009a) reported the total lethal take from 2004 through 2007 under the PRDO in the U.S. was 56,167 birds.
The average annual lethal take for both the AQDO and PRDO during 2004-2007 of 40,094 is likely < 4% of the estimated Interior and Atlantic population of cormorants (Tyson et al. 1999, USFWS 2009a). Lethal take does not account for egg-oiling activities and effects on productivity of young. Band analyses suggest that survival of sub-adult cormorants has declined although it is not clear how much of this is due to management or density dependent population effects (Seamans et al. 2012).
To manage fish-eating birds at aquaculture ponds in the s. U.S., Brugger (Brugger 1995) suggested 5 options: frightening devices, aerial barriers, altering of aquacultural practices, changing of cormorant behavior, and reduction in cormorant numbers by shooting. Cormorants habituate to frightening devices, and nets and other barriers are deemed economically impractical at present, especially where ponds are large. Methods currently used near Mississippi catfish farms include harassment at feeding ponds with vehicles and other frightening devices (reinforced by shooting), and harassment at nocturnal roosts using pyrotechnics (Mott et al. 1998, Glahn et al. 2000a). Regional harassment of cormorants to push them from night roosts in aquaculture producing areas to roosts along the Mississippi River, where there is more natural foraging habitat, have shown some success (Mott et al. 1998, Glahn et al. 2000a, Glahn et al. 2000b, Dorr et al. 2012a, Dorr et al. 2012b). Laser guns are a new technology that shows promise for use at roosts (Glahn et al. 2001). Cormorants soon become wary and difficult to kill, but lethal control is thought to be essential for continued effectiveness of harassment (Glahn 2000). Modification of underwater habitat to impede predation by cormorants has received little attention but may be limited due to interference with production practices.
Nonlethal methods for reducing the impact of cormorants on newly stocked fish include delaying releases (e.g., until cormorants are nesting), releasing prey unpredictably, or at night, and making them more difficult to catch by altering environment or behavior of the fish. New “modular” catfish production systems may help reduce impacts by reducing the area available to birds and making control efforts more efficient (Dorr et al. 2012c).
To reduce the impact of human disturbance on cormorant colonies, Kury and Gochfeld (Kury and Gochfeld 1975) recommended planning visits toward end of the nesting period (when young are less vulnerable) and regulating the number and duration of visits because most predation by gulls occurs when human intruders leave the colony rather than when they are present. In some studies, biologists checked nests and banded chicks after dark to avoid predation by gulls (Ainley and Boekelheide 1990, K. Stromberg pers. comm.). For timing visits to colonies, avoiding midday sun is an important general rule.
Effectiveness Of Measures
In New England, >180,000 eggs were sprayed (1944–1952), and an unknown number of nests were destroyed, but the programs were deemed ineffective at reducing the population because the birds re-layed or moved to new nest sites (Krohn et al. 1995); in retrospect, however, these efforts may have stabilized numbers (Drury 1973). A less extensive program on Canadian Great Lakes (1948–1963) yielded similar results and conclusions (DVW). During the culling program in Quebec, 25,095 nests were sprayed and 7,917 adults shot. This program reduced the population from >17,000 nests in 1989 to <10,000 nests in 1993, faster than predicted by the model (Bédard et al. 1997c). A possible basis for this faster drop in numbers is that males were more vulnerable to shooting.
In the Delta region of central Mississippi, where several strategies were attempted (Mott and Boyd 1995), modification of cormorant behavior through harassment at nocturnal roosts, making them relocate to other areas, appeared to have the greatest impact on reducing the number of cormorants visiting catfish ponds nearby (Mott et al. 1998, Glahn et al. 2000a). Dorr et al. (2012b) estimated the industry could reduce pond-side losses by ~U.S. $1 million annually by shifting 10% of wintering cormorants to Mississippi River roosts.
In the Les Cheneaux Islands (LCI) area of Michigan, cormorant management using egg-oiling to limit reproduction and lethal control of adults on breeding colonies was implemented in 2004, to reduce cormorant numbers and foraging (Dorr et al. 2010a). Fishery data suggested that cormorants were a limiting mortality factor on the yellow perch fishery in the Les Cheneaux Is., Michigan (Fielder 2004, Fielder 2008). Management of nesting cormorants by egg oiling and lethal culling resulted in a large and rapid decline (~75%, and near 90% by 2010) in nesting numbers and foraging was greatly reduced over surveys from the mid 1990's (Dorr et al. 2012a). The fishery and fish population had improved and short-term management goals had been met subsequent to management Fielder (2010). Continued monitoring has indicated that fish population improvements have been largely sustained (BSD).
Nonlethal harassment supplemented by limited lethal take of spring migrating cormorants was implemented on Brevoort Lake and in Potagannissing Bay Michigan, in 2004 and 2005 respectively. These programs were implemented due to declining walleye and yellow perch numbers attributed to cormorant predation (Dorr et al. 2010b, Dorr et al. 2012c). Harassment deterred on average ~90% of cormorant foraging attempts and lethal take was <6% of the cormorants migrating through both sites (Dorr et al. 2010b, Dorr et al. 2012c). Walleye survival increased and abundance at age 3 reached near record levels subsequent to management at Brevoort Lake (Dorr et al. 2010b, Dorr et al. 2012c). A significant increase in yellow perch and walleye abundance occurred for Potagannissing Bay subsequent to management (Dorr et al. 2010b, Dorr et al. 2012c). However, for Potagannissing Bay it was unclear whether there are more walleye and yellow perch because of better reproductive success, lower mortality, or a combination of these factors (Dorr et al. 2010b, Dorr et al. 2012c).
In Lake Ontario, a cormorant management plan, on Little Galloo Island (LGI), began in 1999 (Farquhar et al. 2012). Management included egg-oiling, nest destruction, and prevention of new colonies on neighboring islands and after 2004 limited lethal culling. Cormorant numbers declined from 5,861 breeding pairs on Little Galloo Island in 1999 to 2,730 in 2006. Diversity and numbers of co-occurring waterbirds either increased or was not negatively impacted by management (Farquhar et al. 2012). There was a reduction in estimated fish consumption and a 2.5-fold increase in the abundance of Smallmouth Bass and improved recruitment to the fishery (Farquhar et al 2012). The program shifted in 2007 from a population reduction focus towards a less intensive program to prevent population resurgence (Farquhar et al. 2012).
Ridgway et al. (2011) found the response to egg oiling in Georgian Bay, Lake Huron was as expected with a decline in nest abundance attributable to egg oiling. Conversely, in the North Channel, Lake Huron nest abundance did not decline because of egg oiling but increased. Retention of nesting adults or recruitment to colonies may have been due to escapement from nearby fish pen culture facilities (Ridgway et al. 2011).
The above management outcomes reflect situations where long-term fishery data indicated cormorant predation was an issue; expertise and institutional commitment also supported multiyear management, research, and monitoring programs (Dorr et al. 2012c). Management situations like the above are often not the case and cannot be generalized to all situations. Cormorants are a significant influence on aquatic food webs (Rudstam et al. 2004, Ridgway et al. 2006a) and their role as a sentinel species with regard to contaminants and ecosystem health is well documented (Weseloh et al. 1995, Ryckman et al. 1998). Management should be evaluated on a case basis and in the context that cormorants are an important part of the ecosystems in which they exist.
Little has been done to evaluate cost effectiveness of management. McGregor and Davis (2012) found that egg oiling may be more cost-effective than lethal control for management of ground-nesting cormorants to limit fish consumption. Further effort in evaluation of novel non-lethal methods should also be pursued. Quinn et al. (2012) used a tethered raptor with some success to disperse nesting cormorants from a colony site. Evaluating management in terms of ecosystem response also needs further investigation. Weise et al. (2008) found that non-lethal harassment of cormorants would have a greater impact on salmon smolt survival in the Columbia River than lethal harassment because surviving cormorants also consume pike minnow, which are a predator on salmon smolts.
Impacts of cormorant management to cormorants at meta-population and population scales are also limited (Seefelt 2012). Recent information suggests that density dependent regulatory processes are more influential than management for the interior population of cormorants, but management has reduced colony numbers and growth rates, particularly at local and regional scales (Guillaumet et al. 2014). Management of cormorants in the U.S. to reduce impacts to fisheries, habitat, and co-nesting species are relatively recent and long-term, large-scale efforts to evaluate outcomes are limited.