SPECIES: Dendroica kirtlandii
Table of Contents

 

INTRODUCTORY


  Joel Trick, USFWS

AUTHORSHIP AND CITATION:
Meyer, Rachelle. 2010. Dendroica kirtlandii. In: Fire Effects Information System, [Online]. U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station, Fire Sciences Laboratory (Producer). Available: http://www.fs.fed.us/database/feis/ [ ].

FEIS ABBREVIATION:
DEKI

COMMON NAMES:
Kirtland's warbler
jack pine warbler

TAXONOMY:
The scientific name of Kirtland's warbler is Dendroica kirtlandii (Parulidae) [4]. A possible hybrid with Blackburnian warbler (D. fusca) was observed in the Dominican Republic [64].

SYNONYMS:
None

ORDER:
Passeriformes

CLASS:
Bird

DISTRIBUTION AND OCCURRENCE

SPECIES: Dendroica kirtlandii
GENERAL DISTRIBUTION:
The core of the Kirtland's warbler's breeding range is in the northern portion of Lower Michigan from Presque Isle County south to Clare County [3,81] and west to Wexford County [3]. The expansion of the breeding range into upper Michigan, Wisconsin, and Ontario began in the late 1990s [93]. In Upper Michigan, 95 male Kirtland's warblers were observed in 8 locations from 1994 to 2000 compared to only 3 observations of male Kirtland's warblers on 2 sites from 1978 to 1993. Breeding was documented in upper Michigan from 1995 to 2000 [93] and, according to a Michigan Department of Natural Resources press release [69], likely occurred in 2008. Kirtland's warblers were observed in Wisconsin during the breeding seasons of the late 1970s, 1980, and 1988 [45]. Breeding in Wisconsin was first documented in Adams County, Wisconsin, in 2007 [123] and 2008 [40,123]. In 2009, breeding was observed in both Adams and Marinette counties in Wisconsin [124]. Breeding was also documented in Renfrew County, Ontario, in 2007 [27] and in an unnamed location in Ontario in 2008 [69]. Information on nesting and occurrence of Kirtland's warbler in Ontario in 2009 and Upper Michigan from 2001 to 2007 and in 2009 was not available in late 2009, although Kirk and others [27] suggest that Kirtland's warblers were breeding in upper Michigan from 1995 through 2007.

Kirtland's warblers winter throughout The Bahamas from Grand Bahama [43], Abaco [1], and Eleuthera islands [28] in the north to Crooked [101], Great Inagua [3], and Caicos islands [1] in the south. NatureServe provides a map of Kirtland's warbler's distribution in North America.

PLANT COMMUNITIES:
Breeding grounds: Characteristic Kirtland's warbler breeding habitat is in jack pine (Pinus banksiana)-dominated stands [81,87,130,137]. Red pine (P. resinosa) [81,112,130], eastern white pine (P. strobus) [81,130], quaking aspen (Populus tremuloides) [81,112,130], bigtooth aspen (P. grandidentata) [112,130], pin cherry (Prunus pensylvanica) [81,130], and oaks (Quercus spp.) [112,137] such as northern pin oak (Q. ellipsoidalis) [46,81,130,137] and scarlet oak (Q. coccinea) [81] may occur in these stands, and in some cases species other than jack pine are numerous. Oak canopy coverage in areas occupied by Kirtland's warbler colonies in the late 1970s reached 16% in Ogemaw County, Michigan [109] and 21.75% in a comparatively high-elevation portion of the Bald Hill Burn in Crawford County, Michigan [54]. Sites where jack pine codominates with oaks or other deciduous trees do not provide suitable Kirtland's warbler habitat [81,112]. Common shrubs include blueberries (Vaccinium spp.) [46,81,87,128,130], especially low sweet blueberry (V. angustifolium) [19,130,137], sand cherry (P. pumila) [46,81,87,130], kinnikinnick (Arctostaphylos uva-ursi) [46,81,87,128,130,137], sweetfern (Comptonia peregrina) [19,81,87,128,130], and serviceberry (Amelanchier sp.) [81,130]. Species comprising the ground cover include bluestems (Andropogoneae) [19,87,137], especially little bluestem (Schizachyrium scoparium var. scoparium) and big bluestem (Andropogon gerardii) [46,81,130], oatgrasses (Danthonia spp.) [19], especially poverty oatgrass (D. spicata) [46,81,130], sedges (Carex spp.) [19,46,81,130], western bracken fern (Pteridium aquilinum) [130], mosses [19,81,95,137], and lichens [19,95]. Kirtland's warblers also breed in red pine plantations with appropriate characteristics (see Cover requirements) [81,87,92,112]. However, the Kirtland's warbler recovery plan notes that Kirtland's warbler use of red pine stands typically occurs in areas adjacent to jack pine habitat [21], and Huber and others [47] state that use of these areas by Kirtland's warblers is rare and occurs for short durations.

A hatchling-year Kirtland's warbler was observed 3 miles (5 km) from the nearest breeding habitat during the postbreeding period in mixed coniferous-deciduous secondary growth surrounded by mature conifer forest. The overstory was dominated by black cherry (Prunus serotina), quaking aspen, northern pin oak, and jack pine. Average height of this stratum was 15 feet (4.5 m). Mean overstory coverage was 70%, and trees were patchily distributed. Ground cover was 100%; it was dominated by little bluestem and sweetfern [12].

Migratory habitat: Little information is available on the species composition of areas used by Kirtland's warbler during migration. A male Kirtland's warbler was mistnetted in a hawthorn-sweet crab apple (Crataegus spp.-Malus coronaria) thicket in southwestern Pennsylvania 3 times from 21 September to 2 October 1971 [26]. In Indiana in the late 1800s, Kirtland's warbler specimens were collected in "thickets", including a "plum thicket". In 1981, a Kirtland's warbler was photographed in a Scots pine (Pinus sylvestris) planting in Porter County, Indiana [84]. Kirtland's warbler was listed as a species that would use maritime forest/shrub-scrub in the southeastern coastal plain, a community dominated by live oak (Quercus virginiana), loblolly pine (P. taeda), and cabbage palmetto (Sabal palmetto), with thickets of understory shrubs including wax-myrtle (Myrica cerifera) and yaupon (Ilex vomitoria) [132].

Wintering grounds: Kirtland's warbler wintering habitat appears less specific than its breeding habitat. Kirtland's warblers have been observed in saline-upland transition communities [101,117], in natural and second-growth broadleaf scrub communities [30,73,115,117], and in Caribbean pine (P. caribaea) stands [43] with a shrubby understory [22,28,117]. On Eleuthera Island, wild sage (Lantana involucrata) and cinnecord (Acacia choriophylla) codominated the secondary shrub-scrub and low coppice habitats where Kirtland's warbler was observed. On Grand Turk Island, Kirtland's warbler was observed on sites with dominant and/or codominant species including darling plum (Reynosia septentrionalis), seagrape (Coccoloba uvifera), varnish leaf (Dodonaea viscosa), and broom brush (Baccharis dioica). Seashore dropseed (Sporobolus virginicus) was frequently the dominant ground cover on these sites [117]. On a Crooked Island site where a male Kirtland's warbler was observed, button mangrove (Conocarpus erectus) dominated the shrub layer, and black mangrove (Avicennia germinans) occurred in lesser amounts. Bermuda grass (Cynodon dactylon) dominated the ground cover of this area [101]. A Kirtland's warbler was detected in tall coppice habitat 130 feet (40 m) from the edge of a short coppice habitat on Andros Island. It is possible that the playing of recorded Kirtland's warbler songs attracted it from the nearby short coppice habitat. The vegetation on the site was broadleaved woodland over 15 feet (4.6 m) tall [29]. Other studies did not document Kirtland's warbler in tall coppice communities [74,117].

BIOLOGICAL DATA AND HABITAT REQUIREMENTS

SPECIES: Dendroica kirtlandii
LIFE HISTORY:
Kirtland's warblers are ground-nesting neotropical migrants. They leave the Great Lakes Region for wintering grounds in The Bahamas in August and September and return in May. During the breeding season male Kirtland's warblers defend territories that are clustered into colonies [80,91,137]. Dispersal to new colonies is most common when individuals are settling on their first territory [130]. Based on birds banded at ≥1 year old, the average lifespan is 2.76 years for males and 1.74 years for females. The estimate of female lifespan is likely biased low due difficulty in detecting females [9,130]. The maximum recorded age of a Kirtland's warbler is 9 years [60], although a review cites unpublished data for an individual that was 10 years old [116]. Kirtland's warblers breed in May and June [81,130] and generally fledge one brood each season [100,130]. They have potential for high reproductive output [87,130], although reproduction was greatly reduced by brown-headed cowbird (Molothrus ater) parasitism before implementation of brown-headed cowbird control in the early 1970s [131]. Unless otherwise noted, data reported below are from breeding sites in the core range of the Kirtland's warbler in northern Lower Michigan.

Migration: Kirtland's warblers arrive on the breeding grounds from 10 to 25 May. Males return a few days before females [76,130]. Kirtland's warblers depart the breeding grounds in August and September [76,118]. Hatching-year birds generally leave earlier than older birds. On 5 sites in Lower Michigan, hatching-year birds comprised 60% of the Kirtland's warblers netted from 16 to 31 August, while none of the Kirtland's warblers netted from 16 to 30 September were hatching-year birds [118]. The latest observation of a Kirtland's warbler on the nesting grounds following the breeding season was a female that was banded in the Mack Lake Burn on 1 October 1988 [119]. See reliance on plantations for details of that fire.

Available data do not provide definitive evidence for the Kirtland's warbler migration route(s) [26,89]. A direct south-southeast path has been suggested [76,89], including a nearly nonstop direct route from The Bahamas to Michigan in the spring [76]. Spring migration records support this hypothesis. A majority of spring migration records are from areas near the breeding grounds. States and provinces with 10 or more spring migration records, in decreasing abundance of records, are Michigan, Ohio, Ontario, Indiana, and Illinois. Spring migration records in coastal states are less common; as of 1989 there were 6 in South Carolina, 5 in Georgia, and 3 in Florida. [20]. Observations of fall migrants are also often near the breeding grounds. As of 1973, most fall sightings were in Ohio. Fall observations from coastal states included 3 records from South Carolina, 3 from Florida, and 1 each from North Carolina and Virginia [26]. However, multiple migration routes are possible [26] given the locations of observations [18,89] and the tendency of Kirtland's warblers to migrate alone [118] or with other species [89]. Hunter and others [49] suggest that the Atlantic and Florida coastlines are important for Kirtland's warblers during migration. Walkinshaw [130] provides an extensive list of spring and fall sightings through the early 1980s.

Weather patterns may influence the route [89] and timing [26] of Kirtland's warbler migration. For instance, arrival and departure of a male Kirtland's warbler in southwestern Pennsylvania in late September coincided with wind from the northwest [26]. Strong winds on 28 September 1991 may have blown a Kirtland's warbler to Ocracoke Island, North Carolina, where it was observed with several palm warblers (Dendroica palmarum) [89].

A Kirtland's warbler that spent at least 11 days in southwestern Pennsylvania gained nearly 2 grams during that period [26]. Netting Kirtland's warblers in August and September on the breeding grounds did not provide evidence of fat accumulation prior to migration [118].

Territories and colonies: Territories are arranged in clusters called colonies, which grow and decline as habitat ages. Territories are typically close enough that singing males can be heard on neighboring territories [80]. From 1986 to 1988, the average distance between territories within the Mack Lake Burn colony varied from 883 to 1,713 feet (269-522 m), depending on habitat characteristics [137]. Colonies tend to grow quickly during formation [91,106] and reach peak population size between 7 and 10 years [99]. Walkinshaw [130] reported that the area burned in the Artillery Range South Fire of 1955 was occupied by Kirtland's warblers for 17 summers, with peak colony size occurring 10 years after the first males were found. Baker [6] provides a graph showing the formation, peak, and decline of several colonies from 1971 to 1988. Although older birds continue to return to the same territory year after year [106,130], 1-year-old birds tend to nest in stands younger than the ones in which they hatched [130] (see Dispersal). Therefore, colony growth likely depends on recruitment of 1-year-old Kirtland's warblers from other colonies [106].

On the breeding grounds Kirtland's warblers typically have large territories [130]. Territories observed in the 1950s averaged 8.4 acres (3.4 ha) and ranged from 1.5 to 16.5 acres (0.6-6.7 ha) [81]. Territory sizes ranged from 10 to 21 acres (4.12-8.48 ha) on several sites during the late 1960s and 1970s. Territories were larger at low colony sizes and smaller at large colony sizes [130]. The average territory size of 12 males observed in the late 1970s was 6 acres (2.4 ha) and ranged from 2.5 to 13.8 acres (1.0-5.6 ha) [109].

Once a territory is established, adult Kirtland's warblers generally return to the same territory [9,81,106,130], with males more likely to return than females [9,81]. Walkinshaw [130] reports that only 3 of 27 males established new territories in their 2nd or 3rd breeding season; of 9 females, 7 returned to their original territory. Berger and Radabaugh [9] observed 18 of 28 males return to the territory used previously and 9 of 28 establish new territories within previously occupied colonies. Only 1 male moved to a new colony. Average distance between individual male territories in consecutive years was 1,176 feet (358 m). Of 41 females, 12 returned to the same territory, 23 returned to the same colony, and 5 moved to new colonies. Average distance between territories occupied by individual females in consecutive years was 1,560 feet (476 m) [9]. Mayfield [81] also found greater territory fidelity for males than females. Less than 25% of males moved more than the width of 2 territories between seasons, with an average movement of 660 feet (200 m). Over 50% of females moved more than the width of 2 territories, with an average movement of 1,400 feet (426 m).

Kirtland's warbler densities vary with habitat quality and stand age. In 1989 the density of Kirtland's warblers in Mack Lake Burn varied with jack pine density, with sparse stands having less than 1 male Kirtland's warbler/100 ha and stands with substantial cover having about 1 male Kirtland's warbler/23 ha [85]. Kirtland's warblers density in 8- to 20-year old wildfire stands averaged about 1 male/19 ha, while density in 13- to15-year old wildfire stands averaged 1 male/14 ha [98]. Probst and Weinrich [99] reported densities of about 1 male/14 ha in some plantation habitats. Mayfield [82] described densities of 1 male/33 ha as typical but observed Kirtland's warblers at densities as high as 1 male/4 ha in southeastern Oscoda County and as low as 1 male/120 ha in northwestern Montmorency and southwestern Presque Isla counties.

Limited data suggest that Kirtland's warblers use territories of about 20 acres (8 ha) on the wintering grounds. Based on 3 months' observation of 2 Kirtland's warblers on Eleuthera Island, territory size was estimated as 20.5 acres (8.3 ha) [117]. Based on observations over 2 days, Radabaugh [101] guessed that a Kirtland's warbler on Crooked Island had a territory of at least 17 acres (7 ha).

Dispersal: It is common for year-old Kirtland's warblers to establish a territory outside of their hatching colony. Of 27 males, 13 dispersed an average distance of 20.4 miles (33 km) from where they hatched, 12 returned to within 1 mile (1.6 km) of where they hatched, and 2 were found over 300 miles (500 km) from where they hatched [130]. Of 5 females banded as nestlings, 2 were found within 1 mile (1.6 km) of the banding site, and 3 others were found from 19 to 45 miles (30.6-73 km) from the banding site [102]. Only 8 of 296 Kirtland's warblers banded as nestlings or fledglings from 1933 to 1965 were observed as adults. Of these 8, 5 returned to their hatching colony. A lack of sampling effort away from hatching colonies may explain the low return rate observed [9].

Kirtland's warblers generally establish their first territories in habitat younger than the colony where they hatched. The average age of the stand where 25 males hatched in the late 1960s and 1970s was 14.9 years (range: 9-22 years). Average age of stands where these males established their first territories was 9.8 years (range: 6-15 years) [130].

Kirtland's warblers began dispersing into Upper Michigan, Wisconsin, and Ontario in the late 1990s [93]. The 2 birds Walkinshaw [130] reported as dispersing over 340 miles (550 km) to Ontario and Wisconsin were males hatched the previous breeding season. Several males that established territories in these areas returned in subsequent years [1,45,124]. Kirtland's warblers were observed moving between Lower and Upper Michigan both between and within breeding seasons. Dispersal distances from Lower Michigan to peripheral habitat were up to 220 miles (350 km) [93]. Although the decline of habitat suitability with time suggests that dispersal of young birds is adaptive [90], at low population levels dispersal into peripheral areas may reduce reproductive output because patches of suitable habitat and mates may be difficult to find [81,90,92].

Survival: Although preliminary and in some cases speculative, survival estimates for Kirtland's warbler adults are greater than those of younger birds. Estimates of annual adult survival range from about 60% [81] up to 75% [91]. Preliminary analysis of unpublished banding data (Keplar and Sykes cited in [58]) collected from 1988 to 1991 led to an annual adult survival estimate of 70%. Bergland [10] estimated annual adult male survival was greater than female survival (66.1% vs. 45.8%). However, the female survival estimate was probably underestimated due to difficulty in detecting females. Estimates of Kirtland's warbler survival in their first year range from 27% to 35%. Preliminary analysis of unpublished banding data (Keplar and Sykes cited in [58]) led to an estimate of 27% for annual yearling Kirtland's warbler survival. Bergland [10] hypothesized that annual survival rate of yearling birds was about 30% [10]. Ryel [106] backcalculated survival of yearling birds as 35%. Survival during the approximately 2- to 3-month period from fledging to autumn migration was 65% to 82%, based on an estimate that was derived from available literature [91].

Estimates of rate of return of yearling birds from the wintering grounds apparently vary with duration of parental care. Based on the available literature, Probst [91] estimated that 28% of yearling birds return to breed. First-brood Kirtland's warblers that were cared for by their parents for 38 to 40 days had greater survival rates than those nurtured for shorter periods. Young from 2nd broods apparently have very low return rates. Only 1 of 69 2nd-brood nestlings had been resighted through the early 1980s compared to 40 of 377 1st-brood nestlings [130].

  Reproduction:
Ron Austing, USDA Forest Service Find-a-Photo

Kirtland's warblers form pairs from 20 to 30 May [130]. Clutches of 4 or 5 eggs [87,128,130] are laid in nests buried 1.1 to 1.6 inches (2.9-4.1 cm) deep in duff [112]. A detailed description of nest dimensions and materials is provided by Southern [112]; nest-building activities of the female are described by Van Tyne [128]. Eggs are incubated for 13 to 16 days. Nestlings fledge 8 to 12 days after hatching and are typically fed for 29 to 44 days [81,130]. Kirtland's warblers observed in the 1930s and 1940s typically hatched from 11 June to 20 June [128]. Males assist with all stages of the nesting cycle, including feeding the female during incubation and feeding the nestlings and fledglings [81,130]. Pairs occasionally mate in 2 successive summers [9,130]. Kirtland's warblers renest 5 to 6 days following failed nesting attempts [81]. Based on Mayfield's [81] and Berger and Radabaugh's [9] observations, the average interval between successive nesting attempts was 6.75 days. Kirtland's warblers typically reach reproductive maturity in the breeding season following hatching [9,116,130]. For information on breeding behaviors such as territory defense and parental care, see Walkinshaw [130] and/or Mayfield [81]. A 1992 review [78] summarizes these and many other Kirtland's warbler behaviors.

Several authors have observed Kirtland's warbler pairs raising 2 broods in a summer [9,87,100,130]. Of 163 pairs observed in the late 1960s and 1970s, 12 (7.4%) raised 2 broods in a breeding season [130]. Radabaugh [100] observed 7 double-brood attempts from 1963 to 1971 in Oscoda County. Although most females did not attempt a 2nd brood, he suggested that double brooding was most likely when the 1st brood fledged before the end of June and males fed the 1st-brood fledglings often [100]. Of 3 double-brood attempts observed by Orr [87], 1 was successful, with 35 days between successive fledging dates. Of 163 pairs, no female laid more than 2 clutches in one summer [130].

Males occasionally pair with 2 females [97,130]. Of 163 pairs observed in the late 1960s and 1970s, 4 males mated with 2 females each. Six of the 8 females successfully reproduced and none attempted a 2nd brood [130]. Bocetti [13] observed polyterritorial males and spatially and temporally widespread polygyny. Probst [97] suggested that females may choose a paired male in high-quality habitat over a solitary male in marginal habitat. However, there was no apparent association between polygyny and availability of suitable habitat based on only 5 years of data with relatively small changes in available habitat [13].

Few data are available on Kirtland's warbler adult sex ratios [97,106]. Data from 5 breeding-ground sites from 16 August to 10 September of 1984 to 1987 showed a male-biased sex ratio of 1.4 males to 1 female, although this was not statistically significant. Differences in capture rates between the sexes may have influenced the results [118]. Mayfield [81] noted a surplus of males was possible based on observations of unpaired males in the 1950s. In contrast, Ryel [106] suggested cases of polygyny may indicate a female-biased sex ratio. An unequal sex ratio could have substantial impacts on reproductive output [106,118].

Reproductive success: In the absence of brown-headed cowbirds, Kirtland's warblers may have high reproductive rates. Mayfield [81] estimated Kirtland's warbler pairs would fledge an average of 2.2 young each season without brown-headed cowbirds. From 1972 to 1974, 198 Kirtland's warbler pairs fledged 615 nestlings, an average of 3.11 nestlings/pair [130]. Kirtland's warbler pairs raised an average of 3 to 4.75 fledglings on 3 sites in 1974 [87]. From 1973 to 1981 Kirtland's warblers produced an average of 2.76 young/nest/year [57]. Probability of a Kirtland's warbler egg producing a fledgling in the absence of brown-headed cowbirds was estimated at 32% by Mayfield [81] and 33% by Bergland [10], who used a pooled data set that incorporated Mayfield's data.

Reproductive output per male may be greater in suitable habitat than in marginal habitat. Based on data and observations during the 1980s, Probst and Weinrich [98] suggest a reduction in per male productivity of about 25% to 35% in marginal habitat compared to suitable habitat due to reduced pairing success, delayed settlement time, and territory abandonment. In marginal habitat less than 60% of males paired, significantly (P<0.01) fewer than the 95% pairing success of males in suitable habitat [97]. In 1986, 93% of territories in suitable habitat were established by 5 June, while only 52% of territories in marginal habitat were established by then. Territory abandonment by the end of June was 8% in suitable habitat and 19% in marginal habitat. Forty-four percent of the males in marginal habitat were subadults compared with 23% of males in suitable habitat [92]. Although pairing success and fledglings produced per male were lower in plantation than wildfire stands in the early 1990s (see Demography and use), pairing success was not impacted by the stage of stand occupation or the isolation of the stand [13]. A study in the late 1960s and 1970s found high pairing success across several locations within the breeding grounds, with only 4 of 149 males unpaired [130]. Most areas studied were likely suitable Kirtland's warbler habitat.

Female age apparently does not influence clutch size [130] or reproductive output [5]. Based on data from 1931 to 1975, neither the age of the colony, the number of researcher visits to the nest during laying, nor incubation significantly influenced reproductive output [5].

Brown-headed cowbird parasitism: In the absence of brown-headed cowbird control, parasitism on Kirtland's warbler nests is high and reduces Kirtland's warbler reproductive success. From data collected in the 1940s and 1950s, Mayfield [81] estimated that brown-headed cowbirds parasitized about 55% of nests and reduced fledgling production by 60%. From 1966 to 1971, 69% of Kirtland's warbler nests at one site were parasitized. In the first 3 years following brown-headed cowbird control, from 1972 to 1974, only 7% nests at this site were parasitized [131]. From 1975 to 1981 parasitism rates averaged 3.4% and ranged from 0% to 9% [57]. Kirtland's warbler clutch sizes were smaller before brown-headed cowbird control was implemented than after, with an average clutch size of 2.34 eggs from 1966 to 1971 and an average clutch size of 4.36 eggs from 1972 to 1974 [131]. The table below shows increased fledgling production on sites with cowbird control compared to those without cowbird control. From 1931 to 1975 nests with 2 or more brown-headed cowbird eggs produced an average of <0.5 Kirtland's warbler fledgling/nest [5]. Orr [87] suggested that 2nd broods may be more susceptible to parasitism, because fewer nests are available and brown-headed cowbirds still occurring in breeding areas are likely trap shy. See Brown-headed cowbird control for management-related information.

Fledgling production on sites with and without brown-headed cowbird control*
With control Without control Significance level
2.46 0.8 not tested [131]
2.28 1.11 P<0.001 [5]
2.55 1.20 P<0.05 [10]
*All 3 studies used shared data.

DISEASES AND SOURCES OF MORTALITY:
Diseases: No information is available on this topic.

Sources of Mortality: Predation is likely a substantial cause of Kirtland's warbler nest failure. Of 21 nesting attempts on 3 Lower Michigan sites in the 1974 nesting season, 4 (19%) failed due to nest predation [87]. Bergland [10] suggested that the low percentage of eggs that fledge may be due to high predation rates. However, Mayfield suggested that overall, Kirtland's warblers are not more susceptible to predation than similar species and described predators in the habitats occupied by Kirtland's warblers as "rare" [77].

Predators: Kirtland's warbler nest predators include blue jays (Cyanocitta cristata) [75,87,130], common crows (Corvus brachyrhynchos) [81,87], thirteen-lined ground squirrels (Spermophilus tridecemlineatus), red squirrels (Tamiasciurus hudsonicus) [81,87,130], and garter snakes (Thamnophis sp.) [130]. In addition to these species, Van Tyne [128] includes red foxes (Vulpes vulpes), masked shrews (Sorex cinereus), hawks such as sharp-shinned hawks (Accipiter striatus) and Cooper's hawks (A. cooperi), and northern harriers (Circus cyaneus) as potential Kirtland's warbler predators. Biting ants (Formicidae) may be pests of nestlings. At one location in June of 1935, ants in a Kirtland's warbler nest were identified as Crematogaster lineolata [128]. However, Walkinshaw [130] notes that there is no evidence to suggest ants are harmful to Kirtland's warblers.

There is evidence that house cats (Felis catus) have preyed upon adult Kirtland's warblers [117,130].

Other sources of mortality: Hurricanes or other major storms have been suggested as a possible cause of Kirtland's warbler mortality [122,130]. The most extensive investigation of this possibility did not show any general association with storms. Periods of population increase did not correspond to lulls in storm activity, and periods of population decline did not correspond to increased storm activity. However, only 2 of 19 storms occurred during the fall migration period, and both of these occurred during the population decline between 1961 and 1971. Based on these findings, the large area of migratory and wintering habitat, and the extended fall migratory period, the risk of a single storm affecting a substantial portion of the Kirtland's warbler population appears minimal [107].

There is some evidence that winter or spring drought in The Bahamas may increase Kirtland's warbler mortality. From 1971 to 1980, population size on the breeding grounds in June was positively correlated (P<0.05) with rainfall in The Bahamas from November to April. During the population decline between 1961 and 1971, rainfall on the wintering grounds was "generally low" [107]. High mortality of many songbirds was noted in The Bahamas during the spring of 1971, and drought during this period was suggested as the cause [101]. Rainfall may influence overwintering mortality of Kirtland's warblers due to its impact on food availability [107,117]. Although Mayfield [79] acknowledges the potential threat of consecutive dry winters, he suggests that the association between rainfall and breeding population size is weak. Sykes and Clench [117] suggest that drought impacts on wintering Kirtland's warblers would be localized and unlikely to impact the entire wintering population, which encompasses the whole archipelago.

PREFERRED HABITAT:
Stand age and density, soil permeability, large patch size, landscape level diversity, and presence of low shrub and herbaceous cover are all important, interrelated factors in providing Kirtland's warbler breeding habitat. Wintering habitats are more generalized than breeding habitats and are typically shrubby. Climate: Climatic conditions in areas occupied by Kirtland's warblers are typically moderate [81,125]. Temperatures on the wintering grounds from December to April and on the breeding grounds from May to August usually range from 55 °F to 88 °F (13-31 °C) [125]. The average July temperature in northern Lower Michigan is about 68 °F (20 °C) [81]. The frost- free period in northern Lower Michigan ranges from 80 to 130 days, although there is substantial risk of frost in the growing season [2,81]. Monthly rainfall from December to August in Kirtland's warbler-occupied areas ranges from 1 to 3 inches (25-75 mm) [125]. Mayfield [81] noted that the average annual rainfall in northern Lower Michigan was 30 inches (76 cm). In September, October, and November, monthly rainfall in The Bahamas is typically about 7 to 8 inches (170-200 mm) [125]. Given the extended duration of fall migration—with reviews noting arrivals in The Bahamas as early as August [21,78] and migrants present in the southeastern United States into early November [26]—the extent of Kirtland's warbler exposure to this level of rainfall is likely varied.

Breeding habitat: Kirtland's warbler breeding habitat is highly specialized and consists of large areas of interspersed dense and open patches of young to intermediate-aged jack pine on permeable soil. Trees must be dense enough in some areas that branches interlace, as well as have enough open spaces to allow light to maintain the low jack pine branches that reach the short ground cover [81]. Both interlacing branches and branches low enough to touch ground cover may be required for Kirtland's warbler nesting cover. A review suggests that foliage volume is a major determinant of habitat suitability, with dense pine foliage associated with early stand occupancy and high Kirtland's warbler densities, pairing rates, and territory fidelity [92]. Cover for the nest [81], fledglings [81,92], and foraging adult females [99], as well as food availability [92], have been suggested as reasons for the importance of high foliage volume. Because of the specificity of Kirtland's warblers' stand density and tree size requirements, young stands grow into suitable habitat and then age to the point that they are no longer suitable.

The Grayling Outwash Plain landscape occurs from 900 to 1,580 feet (274-482 m) [2] and comprises much of the core Kirtland's warbler breeding range in northern Lower Michigan. Kirtland's warbler occupied an area of the Bald Hill Burn that ranged from about 1,100 to 1,170 feet (335-357 m) [54]. An arbitrary cutoff of 1,200 feet (372 m) was used to separate high- and low-elevation habitats in the Mack Lake Burn; Kirtland's warblers occupied both habitats at different times following the Mack Lake Fire [7].

Jack pine age and size: Jack pine stands used by nesting Kirtland's warblers are 5 to 20 feet (1.5-6 m) tall and from 6 [81,130] to 22 years old [82]. Stands are often colonized when trees are 5 to 8 feet (1.4-2.3 m) tall, reach peak occupancy when trees are 8 to 12.5 feet (2.4-3.8 m) tall, and are vacated when trees are ≥12.8 feet (3.9-5.6 m) tall [99]. Trees over 10 feet (3 m) were significantly (P<0.05) more abundant in an area with a declining colony and in an area that had been abandoned by Kirtland's warblers compared with a recently formed colony and a colony at its peak [109]. Average age of occupied jack pine stands in Lower Michigan from 1979 to 2004 ranged from 9.4 to 14.9 years. Stands from 12 to 14 years old are considered optimal due to their high density of male Kirtland's warblers. Stand age was consistently a significant (P<0.001) predictor of male Kirtland's warbler density [35]. At these sizes and ages, jack pines are large enough to form dense thickets with interlacing branches but have not become so large that the lower branches become shaded and die [81].

Tall trees or snags on a site may reduce Kirtland's warbler nesting success. Based on nesting data from 1931 to 1975, snags or large trees in breeding stands were negatively associated (P<0.02) with breeding success [5]. Bergland [10] also found a negative association between snags and breeding success. This association may have been due to increased detection of nests by brown-headed cowbirds and/or predators [5,10]. During the 1974 breeding season, the greatest predation rates and all parasitism occurred in a stand where nests were near tall (x =10.8 feet (3.3 m)) trees, while no predation or parasitism occurred in a stand where nests where near the shortest trees (x =4 feet (1.2 m)) [87]. However, based on pooled nesting data collected from 1931 to 1975, Kirtland's warbler nest success was not influenced by variation in tree heights typically occurring in suitable habitat [5]. Despite the potential impacts on breeding success, a review notes that scattered snags and residual trees comprising less than 5% canopy cover provide song perches and may be valuable in areas where brown-headed cowbirds are controlled [92].

Stand density: Nesting Kirtland's warblers require patches of dense jack pine of at least 2,000 stems/ha. According to a review, stands with canopy coverage less than 20% are rarely used for breeding, and optimal habitat generally has 35% to 65% canopy cover [92]. Although ranges of canopy cover on sites of varying age overlapped, recently occupied habitat had an average canopy cover of 27%; areas with established populations had an average canopy cover of 43%; and areas with declining Kirtland's warbler populations had an average canopy cover of 61% [99]. In May and June of 1989, Kirtland's warblers selected jack pine stands with >5,000 stems/ha and >35% canopy cover more than expected based on uniform use (P<0.001) [85]. The most important factors discriminating a site with a recently established colony from a similar area that was not used by Kirtland's warblers were greater total jack pine density, 3- to 7-foot (1-2 m) jack pine density, and 7- to 10-foot (2-3 m) jack pine density in the recently established colony [109]. Kirtland's warbler density increased with increasing jack pine stem density and canopy cover, with less than 1 bird/100 ha in stands with <2,000 stems/ha or <35% canopy cover; 2 birds/100 ha in stands with >5,000 stems/ha; and over 4 birds/100 ha in stands with >35% canopy cover [85]. More males were paired in stands that had been occupied ≥3 years and had 2,500 jack pine stems/ha than in less dense stands or stands that had not been occupied for 3 years [97]. A recently formed colony occurred in an area with 10,600 jack pine stems/ha, and an abandoned area had 1,600 jack pine stems/ha. Oak density was 10 stems/100 m² in an unused stand compared to 0.4 stem/100 m² in the recently established colony [109]. Kirtland's warblers used a stand with 48,260 jack pine stems/ha and 260 oak stems/ha to a greater extent than a stand with 4,210 jack pine stems/ha and 1,620 oak stems/ha. Greater tree density was associated with greater height of lower live limbs, less short (<20 cm) ground vegetation, higher coverage of dead wood and bare ground, and lower coverage of mosses and lichens, grasses, and prostrate shrubs [19]. For stand characteristics specific to nest sites, see Cover requirements. For information on the importance of heterogeneity in tree density, see Landscape factors. For information on the influence of tree density on the timing of Kirtland's warbler colonization and length of occurrence in a stand, see Factors influencing the duration of occupancy.

Soil: Kirtland's warbler breeding colonies occur in areas with sandy, acidic soils that are permeable. Nests often occur on sites with Grayling sand [54,81,130] that has an 0.8- to 1.6-inch (2-4 cm) thick humus layer [82], is acidic [81,130], and dry, loose, and permeable to 3 to 7 feet (1-3 m) below ground [82]. Nests have also been found in Rubicon sand [82], Graycalm sand [54,137], and Montcalm coarse loam [137]. Occurrence of Kirtland's warblers in Wisconsin may be most likely on similar soils, such as Vilas, Omega, and Hiawatha sands and Plainfield loamy sands [45]. The Mack Lake ecosystem with the greatest abundance of Kirtland's warblers in 1987 and 1988 had Grayling and Rubicon soils with a pH of 5.0. During this period, Kirtland's warbler abundance was lower in areas with more finely-textured soils or pH of 6.1 or higher [137]. Soil pH was 4.1 and 4.4 on 2 sites in habitat that developed after the Bald Hill Fire [54]. Kirtland's warblers are commonly associated with glacial outwash plains [2,7,45,54]. Soils that do not hold surface water are critical for nests placed just below ground level [81].

Landscape factors: Kirtland's warblers require large areas of habitat, are influenced by habitat patchiness and interspersion of grassy openings, and may have greater reproductive rates in landscapes with flat compared to rolling topography. Kirtland's warbler distribution is patchy at the landscape level, although the spatial distribution of management areas contributes to this pattern [34].

Kirtland's warblers generally require large patches of habitat for nesting [79,137]. Of the male Kirtland's warblers censused from 1979 to 1989, 77% occurred in areas larger than 200 acres (80 ha) [99]. In a pooled dataset from 1931 to 1975, only 37 of 424 nests occurred in stands smaller than 200 acres (80 ha) [5]. Habitat patches smaller than 80 acres (32 ha) are rarely used by Kirtland's warblers [80,81,92,130]. Kirtland's warblers generally occur at greater densities and for longer durations in habitat patches of 1,000 acres (400 ha) or larger [47]. Site variability may influence the stand size required for Kirtland's warbler occupancy. In diverse landscapes Kirtland’s warblers may occupy smaller stands than those occupied in homogeneous landscapes [54,55]. For information on the importance of landscape heterogeneity, see Factors influencing duration of occupancy. A review notes that stand colonization by Kirtland's warblers may be influenced by stand size [92], presumably with larger stands colonized more quickly and/or more frequently. Kirtland's warbler's association with large patches may be due to fewer predators and competitors in adjacent habitats encroaching into large Kirtland's warbler habitat patches compared to encroachment into small patches [81].

Distance between habitat patches may influence stand colonization and, in some circumstances, male Kirtland's warbler density. During a period with increasing Kirtland's warbler population size and a stable and relatively high level of habitat availability, male Kirtland's warbler density was greater in patches that were closer to other occupied patches than in more isolated patches [35]. According to a review, the colonization of stands by Kirtland's warblers is influenced by the distance to an occupied breeding area [92], presumably with less isolated stands being colonized more quickly and/or more frequently.

Suitable Kirtland's warbler habitat includes dense thickets of jack pine (see Stand density) interspersed with grassy open patches [112,130]. On average these sparse areas comprised 16% to 27% of Kirtland's warbler territories observed in May and June of 1989 [85]. Kirtland's warbler territory size was significantly (P<0.01) related to interspersion, with territories with more changes between jack pine and open ground cover in the 1.6- to 6.6-foot (0.5-2 m) height zone generally smaller than those with more homogeneous vegetation. This results in stands with many small openings having higher Kirtland's warbler densities than less patchy stands. In this study open cover comprised about 55% of territories [109]. Heterogeneity in jack pine stand density may be important in maintaining adequate ground cover [19] and increasing the period of Kirtland's warbler occupancy [7,85]. During the 1970s, Kirtland's warbler used jack pine stands with fewer, generally smaller, openings more than a jack pine stand with more, generally larger, openings. Factors such as much greater density of jack pine, lower density of oaks, and taller jack pines likely influenced greater Kirtland's warbler use of the area with fewer openings [19].

Although Kirtland's warblers occur in stands with both rolling and level topography [13,81,137], reproductive success may be greater in flat regions than hilly ones. Nesting data from 1931 to 1975 [5], as well as an analysis incorporating data collected by Walkinshaw [10], suggested greater rates of brown-headed cowbird parasitism and lower nest success on hilly compared to flat terrain. Nests in flat areas had significantly (P0.01) fewer brown-headed cowbird eggs regardless of brown-headed cowbird control. In areas where brown-headed cowbirds were controlled, the average Kirtland's warbler fledglings per nest was significantly (P=0.033) greater on flat terrain than on hilly terrain [5]. Mayfield [81] notes that there are few waterlogged lowlands in the Kirtland's warbler's breeding range.

Factors influencing duration of occupancy: Stand density, variation in stand density, and landscape heterogeneity are interacting factors that influence how long Kirtland's warblers occupy a stand.

Jack pine density influences the age at which a stand is colonized by Kirtland's warblers and the length of time a stand provides breeding habitat [19,81,85]. Mayfield [81] noted that a dense stand was occupied by Kirtland's warblers when jack pine was about 3 feet (1 m) tall. Sparse stands become suitable and are occupied by Kirtland's warblers at older ages than dense stands [19,81,85]. Suitability of sparse stands increases as tree crowns broaden and canopy cover increases [85]. Buech [19] estimated that characteristics associated with good Kirtland's warbler habitat occurred at 10 years of age in stands with 6,000 trees/ha, 15 years of age in stands with 2,500 tree/ha, and 22 years of age in stands with 1,500 trees/ha. Because of this, stands with patchy tree densities may mature into suitable habitat at staggered intervals and provide habitat over a longer period than those with uniform density [7,81,85]. Areas of low density allow sunlight to reach lower limbs of many jack pines, increasing the persistence of lower branches and potentially increasing the length of Kirtland's warbler occupancy [81].

Variation at the landscape scale, such as different microclimates, soil conditions, and/or physiography, promotes increased length of Kirtland's warbler occupancy [7,54,55]. Productive sites are colonized relatively soon after fire but are only suitable for Kirtland's warblers for short periods [55], while sites where jack pine growth is slow are occupied for a longer period once they become suitable [55,130]. From 1986 to 1988, 6 to 8 years after the Mack Lake Fire, Kirtland's warbler occupied areas with trees taller than those in unoccupied portions of the burn. The site with the tallest average height of northern pin oak had the greatest density of Kirtland's warbler [137]. In 1986, 71% of 14 males occupying the Mack Lake Burn occurred in a portion of the burn with warmer temperatures, better soil, and taller trees compared to other portions of the burn. By 1995, 83% of males occupied the portion of the burn with comparatively cool temperatures, poor soils, and slower jack pine growth [7]. Differing habitats within the burn allowed Kirtland's warbler to occupy the area longer than would have been possible in a homogenous landscape [129]. A similar trend was observed following the 1975 Bald Hill Fire. From the 1982 to 1986, over 70% of Kirtland's warblers in the burned area occurred in a high-elevation region with significantly (P<0.001) warmer mean weekly minimum temperatures, significantly faster (P=0.002) and more dense (P=0.041) jack pine growth, and significantly greater coverage (P=0.002) of taller (P=0.001) northern pin oak compared to the low-elevation area. From 1992 to 1997, 60% to 100% of Kirtland's warblers occurred in low-elevation habitat. The decrease in average elevation with length of Kirtland's warbler occupation of the burn was significant (P<0.001). In the case of the Bald Hill Burn, the diverse growing conditions within the burn extended the occupation of the site for 4 to 6 years [54]. Although the effects of differing site characteristics on jack pine growth rate have often been noted [7,54], effects on ground cover, stand characteristics, tree characteristics, and their interactions may also be important in Kirtland's warbler occupancy [55].

Cover requirements: Nests are located on the ground in dense cover under low jack pine branches. They typically occur near jack pine trees [87] in dense stands [13,87]. Based on data from the early 1990s, density at nest sites averaged 5,800 jack pine/ha, which was significantly (P=0.02) greater than the average 4,800 jack pine/ha at random sites. Significantly (P=0.005) fewer nests than random points occurred in openings, and nests occurred significantly (P=0.005) closer to trees than random points. The average distance from a nest to a tree was 7.6 feet (2.33 m), while the average distance from random points to the nearest tree was 11.7 feet (3.55 m) [13]. Nests on 3 breeding sites during the 1974 season averaged 15.5 inches (39.4 cm) from the base of either a jack or red pine [87]. Jack pine density (P=0.02) and cover category (P<0.001) were significantly greater on nest plots than random plots sampled in the early 1990s. The 5 cover categories were: absent; 1% to 10%; 11% to 25%; 26% to 50%; 51% to 75%; and 76% to 100%. These categories are referred to in the ground cover section below. The mean relative frequency of jack pine was 48% in nest plots, significantly (P=0.001) greater than the 26% relative frequency on random plots [13].

Live lower limbs: The association of Kirtland's warbler colony decline with increasing height of lower live jack pine branches (see table below) has led to suggestions that lower live branches provide cover needed for fledglings and/or foraging habitat near the nest [92,99].

Average height and range of lower live limbs in habitats at varying stages of Kirtland's warbler use [99]
Habitat Stage Average height of lower live limbs Range of heights of lower live limbs
Habitat with new colony 12 inches (30 cm) 4 to 35 inches (10-90 cm)
Habitat with established colony 24 inches (60 cm) 8 to 43 inches (20-110 cm)
Habitat with declining colony 48 inches (120 cm) 24 to 60 inches (70-150 cm)

The height to the first live branch on trees near nests was significantly (P=0.04) shorter than the height to the first live branch on trees near random sites, although the height difference was only 1.5 inch (3.8 cm) [13]. Kirtland's warblers occurred in areas with lower live limbs below 3 feet (1 m) above ground in Oscoda, County. However, an area within this site where lower live limbs were generally less than 20 inches (50 cm) above ground had relatively sparse tree density and was used by Kirtland's warbler to a lesser extent than an area with lower live limbs generally from 20 to 36 inches (50-100 cm) and relatively dense tree cover [19]. Generally, colonies begin to decline when there is no live pine foliage below about 3 feet [92]. Mayfield [81] suggested that maximum cover for Kirtland's warblers is provided when there is no gap between the ground vegetation and lower pine branches. However, Walkinshaw [130] suggested that changes in groundlayer vegetation as jack pine stands age were more important to Kirtland's warbler occupancy than the loss of lower live branches. Lower pine branches persist with access to sunlight [81], and tree density influences the height of lower live limbs [19]. Thus, many openings in a stand may increase the duration that lower live limbs persist, thereby increasing the duration of Kirtland's warbler occupancy [81].

Ground cover: Kirtland's warbler nest sites have short ground cover. Height of vegetation near nesting sites is often less than 12 inches (30 cm) [19,81]. Most vegetation on a breeding site studied in 1976 was from 8 to 12 inches (20-30 cm) tall [19]. Mayfield [81] and Van Tyne [128] described the vegetation around nests as thick or fairly thick. Vegetation is often arched over nests, allowing entrance from only one direction [112,128]. However, ground cover density and nest concealment were not significantly associated with nesting success based on data collected from 1931 to 1975 [5]. Southern [112] suggests that the role of grasses in providing cover and nesting material could limit Kirtland's warbler to areas of jack pine forests with adequate coverage of grasses and sedges. However, in a review, Probst [92] suggests that ground cover required by Kirtland's warbler is readily available and does not limit habitat selection.

Specific groundlayer species do not appear necessary for Kirtland's warbler nest sites [19,95,112], although a combination of grasses, sedges, blueberries and other low shrubs [19,87,95], lichens, mosses, and bare ground [19,95] is common. Mayfield [81] found that bluestem grasses (77%) and/or blueberries (61%) covered the majority of nests. During the 1974 breeding season, low sweet blueberry provided an average of 28% cover over 85.7% of nests [87]. Based on data collected in the early 1990s, low sweet blueberry cover category was significantly (P<0.001) greater on nest sites than random sites [13]. Grasses, predominantly Andropogon spp., provided an average of 26.5% cover over 80.9% of nests, and kinnikinnick provided an average 14% cover over 42.8% of nests observed in 1974. Species that provided less than 10% cover over less than 30% of nests included sand cherry, sweetfern, and wintergreen (Gaultheria procumbens) [87]. On burned (3 wildfires, 1 prescribed burn) and unburned (3 plantations and 2 naturally regenerated timber harvests) sites from 7 to 23 years old, lichen and moss cover ranged from 3.7% to 39.8%, sedge and grass cover ranged from 2.2% to 11.6%, and bare ground and/or litter cover averaged 5.6%. Blueberry cover ranged from 0.3% to 17.4% and increased with stand age [95]. Birdfoot violet (Viola pedata) had significantly (P=0.004) lower relative frequency and a slightly, but significantly (P=0.004), lower cover category on nest sites than random sites [13].

Amount of coarse woody debris is variable in Kirtland's warbler nesting habitat. Frequency of dead wood in the Muskrat Lake colony was 100%, with coverage ranging from 8.1% to 17.9% [19]. Based on data from the early 1990s, coarse woody debris cover category was slightly but significantly (P=0.007) lower on nest plots than on randomly selected plots. Considering the possibility of higher productivity per male in habitat regenerated following wildfire [13] (see Demography and use) and the greater cover of coarse woody debris in burned areas (see Differences in habitat characteristics), coarse woody debris may provide other benefits to Kirtland's warbler, perhaps related to foraging substrate (see Food Habits).

Wintering habitat: Wintering Kirtland's warblers occupy short, dense, shrubby vegetation, occasionally with a Caribbean pine overstory [22,28,117]. On a Crooked Island site where a Kirtland's warbler was observed, scattered trees reached about 20 feet (6 m) [101]. The height of shrubby vegetation typically ranges from 3 to 10 feet (1-3 m) [22,101,115,117]. Mayfield [73] suggested wintering habitat was comprised of broadleaved scrub less than 15 feet (4.6 m) tall. In a low coppice community occupied by Kirtland's warbler on Eleuthera Island, average tree height was 11 feet (3.3 m). Average shrub height across occupied habitats of Eleuthera and Grand Turk islands ranged from 3.3 to 3.6 feet (1.0-1.1 m). The vegetation stratum with the greatest stem density was ≤12 inches (30 cm). Other structural characteristics of habitats occupied by Kirtland's warbler on these islands are provided in the table below. All occupied habitats occurred on gentle slopes [117]. Elevations of these sites range from near sea level [117] to 3,389 feet (1,033 m) [22]. For information on the only tall coppice site where Kirtland's warbler has been observed, see Currie and others [29]. For details of this record, a general discussion of community types occupied by Kirtland's warblers in winter, and examples of plant species comprising these communities, see Plant Communities.

Habitat characteristics of winter territories occupied by 4 Kirtland's warblers at 3 sites on Eleuthera and Grand Turk islands [117]
Habitat types Tree density
(trees/ha)
Average shrub density
(stems/ha)
Ground cover
(%)
Leaf litter cover
(%)
Bare soil cover
(%)
Low coppice
(n=3 plots)
2,248 31,246 25% 77% 7%
Shrub-scrub
(n=5 plots)
0 8,601 11% 55% 43%
Second growth shrub-scrub
(n=1 plot)
25 54,711 14% 68% 18%
Saline-upland ecotone
(n=5 plots)
0 10,611 45% 19% 77%

FOOD HABITS:
Kirtland's warblers generally forage for insects and fruits near ground level by gleaning. Few data address the relationship between food availability and Kirtland's warbler habitat selection.

Kirtland's warblers eat a variety of insects and fruits. Details on the insect proportion of the Kirtland's warbler diet come primarily from the breeding grounds [32,81,130]. Frequent insect food items in the Kirtland's warbler diet from June to September of 1995 to 1997 were ants and wasps (Hymenoptera; 45%), beetles (Coleoptera; 25%), moth larvae (Lepidoptera; 22%), spittlebugs (Cercopidae; 36%) and aphids (Aphididae; 21%). In this study low sweet blueberry occurred in 42% of samples [32]. Adult Kirtland's warblers were observed eating small beetles, cicadas (Hemiptera) [130], adult ant lions (Myrmeleontidae), damsel flies (Zygoptera), beelike insects (Hymenoptera) [81], moths, flies (Diptera), and caterpillars (Lepidoptera) and other larvae [81,130]. According to a personal observation reported in a review, caterpillars comprised a large proportion of forage brought to young Kirtland's warblers [116]. Adults were observed feeding on blueberries while the fruits were available [81,130]. A migrating Kirtland's warbler in Indiana was observed on the ground or in low vegetation foraging on insects [20]. Kirtland's warblers have been observed eating insects on the wintering grounds; details on the types of insects eaten were not available as of early 2010 [101,115]. Fruits may comprise a large proportion of the diet of wintering Kirtland's warblers. Of 448 foraging observations of 2 wintering Kirtland's warblers, 59% were of small fruit, over 80% of these from a single species [115]. According to unpublished data (Sykes cited in [117]), wild sage produces abundant fruit that is heavily used by Kirtland's warblers in the northern Bahamas during winter, and Carey and others [22] noted the use of black torch (Erithalis fruticosa) fruit by Kirtland's warblers on Eleuthera Island.

Data collected during migration and on the wintering grounds suggest Kirtland's warblers often forage near the ground in shrubs and herbaceous vegetation. On the wintering grounds, 98% of foraging observations of 2 Kirtland's warblers were within 10 feet (3 m) of the ground. The remaining observations were from 10 to 20 feet (3-6 m). The 2 Kirtland's warblers were in wild sage in 76%, twining soldierbush (Tournefortia volubilis) in 8%, West Indian milkberry (Chiococca alba) in 4%, cinnecord in 3.5%, and black torch in 3.3% of foraging observations [117]. Another Kirtland's warbler on the wintering grounds spent 70% of the time it was observed on the ground searching for food [101]. Kirtland's warblers observed in Indiana during fall migration foraged primarily within 3 feet (1 m) of the ground in "shrubs and goldenrod" [18], while a Kirtland's warbler observed during spring in Indiana foraged entirely within 2 feet (0.6 m) of the ground in herbaceous vegetation [115].

On the breeding grounds about 80% of Kirtland's warbler foraging observations were on jack pines. Slightly less than 10% were on the ground, and less than 5% were on northern pin oak. Variability in the use of oak suggested that use was influenced by availability [39]. Data on the proportion of time spent at varying heights from the ground while foraging on the breeding grounds are not available. However, Fussman [39] provides information on the proportional use of jack pine at increments of a quarter of tree height. Nearly 25% of Kirtland's warbler foraging observations were in the top quarter of jack pine trees in young (6-13 years old) and old (14-21 years old) nesting habitat [39]. Mayfield [81] noted foraging on the ground and in high limbs of large nearby trees. In a review, Probst [92] notes extensive use of logging slash and coarse woody debris by foraging Kirtland's warblers, while Fussman [39] found that coarse woody debris was used in less than 5% of foraging observations. It is not known if use of downed wood is facultative or if downed wood improves habitat quality [92].

The little information available on feeding methods suggests that most food is gleaned from vegetation. Mayfield [81] noted food was gathered from the ground and trees on the breeding grounds. On the wintering grounds 75% of the foraging of 2 Kirtland's warblers was gleaning. Other Kirtland's warbler foraging behaviors on the wintering grounds included probing (13%), "hovering-gleaning" (7%), and other techniques (5%) [115]. Hovering [20,115], flycatching [81,84], or both have been noted on the breeding grounds [81], during migration [20,84], and on the wintering grounds [115].

Few data address the possible influence of food availability on habitat selection. Stand age had a marginal (P=0.063) influence on the vertical distribution of arthropod biomass, with old nesting habitat (14-21 years old) having greater arthropod biomass in the lowest quarter of tree height than abandoned, mature (>21 years old) stands. This trend was especially noticeable for larvae, which were not observed in the lower quarter of trees in mature stands. However, only 12% of Kirtland's warblers foraging observations were in the lower quarter of tree height. Different population segments used different jack pine strata. Females used the top quarter of the tree significantly (P<0.001) less than males, and females carrying food back to the nest foraged lower than females not carrying food (P=0.001). The author concluded that the benefit provided to food-carrying females by foraging lower in trees compared to either males or females foraging for themselves seemed strong, since food availability was greater in the upper halves of trees than the lower halves of trees over all age classes [39]. Probst [99] suggested the volume of foliage required to meet Kirtland's warblers foraging needs may influence habitat selection.

FEDERAL LEGAL STATUS:
Endangered [126]

OTHER STATUS:
Information on state- and province-level protection status of animals in the United States and Canada is available at NatureServe, although recent changes in status may not be included.

An extensive history of Kirtland's warbler population fluctuations is available. Relative ease of collecting Kirtland's warblers on their wintering grounds in the late 1800s [75,80,101], more collections of migrants [74], and greater habitat availability during that period [56,74,81] suggest they were comparatively common at that time. In 1953 the first census estimated the male Kirtland's warbler population at about 432 [80]. The population declined from an estimated 502 males in 1961 [82] to 201 males in 1971 [72]. From 1971 to 1987, breeding-season estimates of Kirtland's warbler's population size stabilized at an average of 206 males [99]. The population began rebounding in the early 1990s [94,99,133]. The population reached the Recovery Team target of 1,000 singing male Kirtland's warblers [58] in 2001 [27]. According to a Michigan Department of Natural Resources press release, in 2008 there were an estimated 1,791 males in Upper and Lower Michigan [69], and in 2009 the census documented 1,795 singing males (Huber 2009 personal communication [48]). The number of Kirtland's warblers counted on the wintering grounds started increasing in about 2002 [28]. See DeCapita [31] and Solomon [110] for histories of Kirtland's warbler population trends through the mid-1990s, and Probst and others [94] for a comparison of census methods. For a discussion of causes of the population decline in the 1960s and increase in the 1990s, see Factors influencing population size.

MANAGEMENT CONSIDERATIONS:
Cooperation between federal and state agencies has been a predominant aspect of Kirtland's warbler management for decades [50,58,121]. A survey of local attitudes regarding the Kirtland's warbler in 1998 found generally positive responses [111]. The Kirtland's warbler is beneficial to the local area in some respects, including as a source of tourism [110]. Some aspects of the management program have been unpopular, including area and road closures and the large clearcuts created for Kirtland's warbler habitat [61].

Preliminary information regarding methods for reintroducing Kirtland's warblers to unoccupied areas of their range is available [11,17]. See DeCapita [31] and Solomon [110] for histories of Kirtland's warbler management through the mid-1990s. For a discussion of the history of Kirtland's warbler habitat management, see Trauger and Bocetti [121] or Huber and others [47].

Factors influencing population size: It is likely that declines in available habitat [21,34,79,83,91,107] and large reductions in reproductive success due to brown-headed cowbird parasitism [21,81,92,107,130] contributed to the decline of the Kirtland's warbler population between 1961 and 1971. Reproductive rates increased with brown-headed cowbird control (see brown-headed cowbird parasitism), beginning in 1972, with control likely averting further population decline [31,72]. However, it has been hypothesized that habitat limitation may have been the major cause of the decline [105]. Population size began to increase in the early 1990s, with the maturation of habitat burned in the late 1970s and 1980 [27,34,79,93,99]. Availability of suitable habitat increased from a bit less than 25,000 acres (10,000 ha) in the early 1980s to about 50,000 acres (20,000 ha) in the late 1990s and early 2000s. From 1988 to 1994 the Kirtland's warbler population more than tripled, and from 1994 to 2004 it increased another 74.6% [34]. Inference of a causal relationship between increased habitat and increased population size is based primarily on the timing of habitat increase and population growth. The strength of this inference was ranked as intermediate by James and McCulloch [52].

Logging operations in The Bahamas may have contributed to Kirtland's warbler decline [43,101]. However, several abundant habitat types apparently meet Kirtland's warbler wintering habitat requirements [74,117] suggesting that it has not been a major influence on Kirtland's warbler population trends.

See Other sources of mortality for discussions of the possible impacts of large storms and drought in The Bahamas on Kirtland's warbler population size.

Threats: A major hurdle to Kirtland's warbler recovery is the financial cost of continued management. Given the limited period that breeding habitat is suitable and the ongoing threat from brown-headed cowbirds, extensive habitat management and brown-headed cowbird control are integral and permanent aspects of Kirtland's warbler management [31]. Although some of the cost is mitigated by harvesting jack pine, this revenue did not cover costs under 1989 market conditions [61]. In 2009, cost of reforestation on the Huron-Manistee National Forest was estimated at about $500 per acre (Huber 2009 personal communication [48]). This was offset to some extent by income from timber sales [88]. According to reviews, a lack of funding for management is a recurrent concern [58,116], and a newsletter notes that this is an impediment to delisting the Kirtland's warbler [88]. Strategies for addressing this issue include increasing cost effectiveness of [58,66,70,71] and establishing an endowment to assist with expenses of habitat management [88].

There are little data available addressing the potential impact of climate change on Kirtland's warblers. In 1991, modeling of jack pine in northern Lower Michigan suggested it is sensitive to increases in temperature and will decline rapidly with climate change [15]. However, as of 2007, these predictions had not been tested [14]. Recent range expansions (see General Distribution) suggest that if jack pine were to decline in the southern part of its range, Kirtland's warbler could move north. However, the ability of northern portions of the Kirtland's warbler distribution to support an adequate population size under current or altered climate scenarios has not been investigated. In addition, the impacts of any climate-induced change to the fire regime on Kirtland's warblers and their habitat have not been addressed. A summary of these and other potential impacts of climate change on Kirtland's warbler will likely be completed by late 2010 (Huber 2009 personal communication [48]).

Disturbance of Kirtland's warblers during the breeding and postbreeding seasons is a concern [107] and has led to area closures during the breeding season [58,61]. Due to Kirtland's warbler occurrence on the breeding grounds well into September, Sykes and others [118] recommended closing areas with colonies of more than 10 males through mid-September. According the 2001 review, areas with high and increasing singing male abundance were closed to public entry from 1 May to 10 September, while areas with low and declining abundance of singing males were closed from 1 May to 15 August [47].

The risk of a natural disaster having a major negative impact on the entire Kirtland's warbler population [6] may be mitigated to some extent if Kirtland's warblers maintain colonies in Upper Michigan and Wisconsin.

 

Plantation management:

Kirtland's warbler management area, showing "cutting pattern".
Photo by Linda Haugen, USDA Forest Service, Bugwood.org

Limitations in the use of prescribed burning have resulted in reliance on plantations to create Kirtland's warbler habitat. There are significant differences between habitat characteristics in areas following wildfire and in plantations, and reproductive output per Kirtland's warbler male suggests that stands originating after wildfire provide slightly higher quality habitat than plantations (see Demography and use). The following information on plantation management focuses on stand-level details such as rotation length, block size, planting density, and snags as well as larger-scale considerations such as temporal and spatial arrangement and selection of locations to meet management objectives. For detailed summaries of habitat management, see the strategy for Kirtland's warbler habitat management [47], the Kirtland's warbler Recovery Plan [21], Radtke and others' [103] 1989 review, or Kepler and others' [58] 1996 review.

Stand-level details: Plantations managed for Kirtland's warblers in Michigan have a rotation of about 50 years [21,27,47,58,61]. Blocks have patchiness incorporated into the design [46,47,92] and are often more than 300 acres (100 ha) in size [27]. In a 2008 review of Kirtland's warbler management, rotation length is listed as 55 years in Lower Michigan, 45 years in Upper Michigan, and 75 years in Ontario [27]. One to 5 openings per acre [27,47] are incorporated into jack pine plantations [58], with 20% [31] to 25% [27,46,47] of the site left unplanted. In the 1980s, treatment blocks ranged from 100 [130] to 370 acres (40.5-150 ha), but were most commonly 200 to 300 acres (80-120 ha) [61]. A 1996 review notes management areas were slightly larger, at least 320 acres (130 acres) [58]. Huber and others [47] recommend treatment blocks of more than 300 acres (120 ha) and over 1,000 acres (400 ha) where possible [27,47]. In the late 2000s many plantations created on federal land were 400 to 500 acres (160-200 ha), and those on state lands were as large as 1,000 to 2,000 acres (400-800 ha) (Huber 2009 personal communication [48]).

Planting jack pine seedlings has resulted in better success than use of natural regeneration or direct seeding, and use of a V plow and careful handling of planting stock has improved the consistency of results [58]. Planting seedlings 4 feet (1.2 m) apart in rows that are 6 feet (1.8 m) apart is typical spacing in Kirtland's warbler plantations [27,130]. This gives a density of 4,510 trees/ha [27]. In a review, Probst [92] notes that 2,800 stems/ha provides adequate density for Kirtland's warblers in plantations, because jack pine is less clumped than in naturally regenerating stands. He recommends experimenting with specific alternate densities [92]. Huber and others [47] recommend evaluating stocking densities 1 and 3 years after regeneration attempts. For specific recommendations including harvest, site preparation, and regeneration techniques, see the 1988 Probst [92] review, the 1996 Kepler [58] review, the 2001 Huber and others [47] review, or the 2008 Kirk and others [27] review. For information on plantation methods including the use of prescribed burning, see Fire Management Considerations.

The amount of snags, residual trees, and coarse woody debris needed for Kirtland's warbler is unclear. In a 1988 review, Probst [92] recommends avoiding the creation of narrow stands with long borders adjacent to tall stands, since predation rates could be high in these areas. Concern has been raised over large residual trees and snags within stands leading to increased predation or parasitism [5,10] (see Jack pine size/age). However, areas with less than 5% coverage of snags and/or residual tall trees, where brown-headed cowbirds are controlled, may be beneficial to Kirtland's warblers [92]. Huber and others [47] considered snags important habitat components of Kirtland's warbler habitat and the jack pine ecosystem in general. Substantial differences in coarse woody debris and snags between wildfire and plantation habitats (see Differences in habitat characteristics) led Spaulding and Rothstein [113] to suggest increasing snags, coarse woody debris, and potentially girdling groups of trees in plantations. Probst [92] recommended more direct investigation of the importance of coarse woody debris in Kirtland's warbler habitat.

Spatial and temporal considerations: Plantation occupancy may be increased by clustering plantations and scheduling rotations so that the periods of habitat suitability in neighboring plantations overlap to some extent. To facilitate Kirtland's warbler occupancy, a1996 review recommended clumping management areas so that 5 or more were within 2 miles (3 km) of one another and planting closely arranged plantations in sequence. It is especially important that small plantations be located near other, preferably large, plantations [58]. A 2001 review also recommended scheduling habitat development so habitats of similar age are closely spaced [47]. Some dispersal data suggest that spatial continuity is not required for new habitat in peripheral areas [93]. More frequent establishment of new plantations, such as cutting appropriately aged stands every 5 years instead of every 10, could allow for overlap in the timing of stand suitability and minimize problems of biogeographic dispersal [92]. Management planning and decision making should incorporate information on the spatial and temporal availability of habitats of various qualities, Kirtland's warbler population size, and recent population trends [34].

Landscape considerations help inform decisions on the location, size, and tree density of plantations. Plantation locations may be selected to better achieve management objectives. For instance, if habitat is needed quickly, a plantation can be located in an area where jack pine growth is relatively fast. In contrast, creating habitat that will be occupied for a long period can be achieved by planting in areas where jack pine growth is slow or in diverse areas [55,129]. Size of a plantation may be influenced by landscape diversity, with larger plantations needed in homogeneous landscapes and smaller plantations able to support Kirtland's warblers in more diverse landscapes. In areas with homogeneous landscapes, staggered planting schedules [54,55] or varied planting densities [92] may provide at least some of the diversity that would occur in heterogeneous landscapes. Variation in the density of jack pine may also increase habitat diversity [85].

Brown-headed cowbird control: Kirtland's warbler management includes extensive annual trapping of brown-headed cowbirds. From 1972 to 1978, over 17,500 brown-headed cowbirds were trapped in Kirtland's warbler breeding areas [75]. By 2000, over 98,000 brown-headed cowbirds had been removed from breeding areas in northern Michigan. Despite extensive annual trapping, brown-headed cowbirds return in large numbers each year [31]. A population viability analysis estimated that nesting failure of 35% from predation, parasitism, and abandonment combined would reduce the Kirtland's warbler population. Based on this estimate, either parasitism rates without brown-headed cowbird control (see brown-headed cowbird parasitism) or parasitism expected at half the control effort of the 1990s are likely to lead Kirtland's warbler to extinction [16]. Given the continued threat, brown-headed cowbird control will likely be a permanent component of Kirtland's warbler management [31,34,107]. For information on parasitism dynamics, see McGreen [83]; for limitations of brown-headed cowbird control efforts, see Rothstein and Cook [105]; for characteristics of trapped brown-headed cowbirds, see DeCapita [31]; and for information on traps and trapping methods, see Kepler and others [58], Mayfield [75] and/or Trick and others [123].

FIRE EFFECTS AND MANAGEMENT

SPECIES: Dendroica kirtlandii
DIRECT FIRE EFFECTS:
Adult Kirtland's warblers are unlikely to suffer directly from fire. It is generally accepted that large, fast-moving fires may result in mortality, but adult birds typically have the mobility to avoid fire [24,33,68].

Kirtland's warbler nests are likely vulnerable to fires occurring from May to mid-July [128]. Although there were no data directly investigating Kirtland's warbler nest mortality due to fire as of 2010, literature reviews have used fire characteristics and life history of species to speculate on possible effects of fire on nesting success and bird populations [68,104]. Since Kirtland's warblers nest on the ground, low-severity surface fires during the breeding season could result in considerable nest and/or fledgling mortality. Renesting following failed nesting attempts [9,81] (see Reproduction) may mitigate the direct effects of fire on Kirtland's warbler recruitment [68,104]. However, since Kirtland's warblers typically rear only 1 brood per year, fires in the mid- to late-breeding season may have a larger detrimental effect on mortality of Kirtland's warbler nestlings and/or fledglings. Most large jack pine wildfires in northern Lower Michigan occur in late April and early May, before Kirtland's warblers arrive on the breeding grounds (Huber 2009 personal communication [48]). For information regarding seasonality of fires, see Fire Regimes.

INDIRECT FIRE EFFECTS:
Kirtland's warbler habitat is fire dependent, but fire effects vary with fire and site characteristics. Burned and unburned habitats occupied by Kirtland's warblers exhibit some important differences.

There is considerable research on fire, Kirtland's warblers, and their habitat, but readers should be aware of its limitations. As noted by a review [38] summarizing songbird responses to fire in southwestern ponderosa pine (Pinus ponderosa) forests, studies addressing bird response to fire are generally opportunistic, anecdotal, and/or restricted in spatial or temporal scale. They may also have small sample sizes and/or include confounding factors [38]. Although several studies compare Kirtland's warbler demography between habitats of varying origins, it is not always clear how the sites differed. For instance, comparisons between wildfire and plantation habitat do not always describe details of management, such as type of harvesting or prescribing burning. The presence and potential impact of confounding factors such as comparing large patches of wildfire habitat to smaller plantations were not always addressed. Also, there is potential for overlapping data between studies.

Fire-dependant nature of Kirtland's warbler habitat: Kirtland's warblers occupy extensive stands of young to intermediate-aged jack pine forests (see Jack pine age/size). Under natural conditions, these stands result from stand-replacement wildfires [81,128]. Although a review notes that jack pine may establish without fire [130], fires of the appropriate severity open jack pine's serotinous cones and release numerous seeds simultaneously [23,81,135] into an environment with optimal conditions for germination [23,135] and growth [23,81,135], including exposed mineral soil [23] and reduced interference from other plant species [23,81]. In the laboratory, seeds in jack pine cones retained viability following exposure to temperatures of 900 °F (480 °C) for 30 seconds and 700 °F (370 °C) for 1 to 2 minutes, and exposed jack pine seeds were viable after exposure to 700 °F for 10 to 15 seconds [8]. See the FEIS review of jack pine for more details. According to the 1985 recovery plan, nearly all occupied habitat where Kirtland's warblers bred successfully developed following fire [21]. However, stand-replacement fire does not guarantee the development of suitable Kirtland's warbler habitat. Several factors may interfere with the regeneration of jack pine following fire [23], including fire severity, postfire weather [23,130], seed availability, and site characteristics [23].

If fire is excluded, jack pine may convert to red or eastern white pine [127] or to hardwoods [56], which have less flammable fuel beds [86]. Since these cover types are not suitable for Kirtland's warbler, fire exclusion generally leads to loss of habitat and declines in Kirtland's warbler populations [21,128] (see Factors influencing population size). In contrast, fires in mixed red and eastern white pine stands may convert them to jack pine stands [23,128].

Potential detrimental impacts of fire on Kirtland's warbler habitat: Fires may threaten existing and developing Kirtland's warbler habitat. Fires in occupied areas reduce the amount of currently available habitat [131,135]. A fire of unreported severity burned through existing habitat within the Bald Hill Burn in May of 1990. During the 1990 breeding season few Kirtland's warbler males were documented in the area, but the following year—16 years after the Bald Hill Fire—the Kirtland's warbler colony in this habitat reached its peak occupancy of 36 males. The fire also expedited the shift of Kirtland's warblers from high- to low-elevation habitat [54]. In 1967, a May fire burned a portion of an area burned 7 years earlier in the Artillery Range Fire. Some of the area within the 2nd burn had poor jack pine regeneration [131]. Since jack pine seed production is considerable by 6 to 8 years of age, the impacts of fires within stands older than this are likely smaller than impacts of fires within younger habitat [37]. In some instances, fires in suitable habitat with appropriate timing may lengthen the duration of Kirtland's warbler occupancy [131]. Although multiple and/or large fires burning through unoccupied Kirtland's warbler management areas over a short time would likely create Kirtland's warbler habitat in the intermediate term, they could reduce amounts of jack pine becoming suitable for Kirtland's warbler in the long term [135]. Huber and others [47] recommend adjusting the habitat development schedule following wildfires to "ensure a sustained supply of occupiable habitat over the long term".

Comparison of habitats of varying origins:
Demography and use: Kirtland's warbler densities are likely similar between plantation and wildfire habitat and are greater in those habitats than in stands that regenerate naturally following timber harvest. Based on data from 1980 to 1992, predicted average Kirtland's warbler densities were 0.21 male/40 acres in plantations, 0.19 male/40 acres in wildfire regenerated habitat, and 0.02 male/40 acres in stands that regenerated naturally following harvest [13]. In 1993, Probst and Weinrich [99] reported densities of about 2.8 males/100 acres in burned and unburned plantation habitat. According to the 1985 recovery plan, Kirtland's warbler densities were about twice as high in wildfire-regenerated habitat than in stands established without fire [21]. At low population levels and low habitat availability, average male Kirtland's warbler density was greater in habitat with wildfire origins than in plantations or naturally regenerated, unburned stands [35].

There is evidence that during a period of relatively low population sizes and low availability of suitable habitat (1979-1991) [35], Kirtland's warblers selected wildfire habitat over unburned or plantation habitat. During the 1980s, burned areas were used substantially more than management plantings [58,130]. From 1987 to 1994, approximately 75% of male Kirtland's warblers occurred in wildfire habitat, which comprised about 33% of the suitable habitat available [34]. In 1984, 30% of male Kirtland's warblers occupied unburned, unplanted areas, which comprised 54% of the available habitat; 55% of males occupied wildfire habitat, which comprised 41% of available habitat; and 4% of males occupied plantation habitat, which comprised 5% of the available habitat. In 1989, no Kirtland's warblers occupied unburned, unplanted areas despite those areas comprising 46% of available habitat. Seventy-six percent of males occupied wildfire habitat, which comprised 36% of available habitat, and 24% of males occupied plantation habitat, which comprised 19% of available habitat. The avoidance of unburned, unplanted areas was significant (P<0.001) in both years, and selection for wildfire habitat was significant (P<0.001) in 1989 [99]. In general, the use of unburned, naturally regenerated habitat declined as burned and plantation habitat became available [34,99]. These results led Probst and Weinrich [99] to conclude that use of unburned, unplanted areas before 1987 was due to the Kirtland's warbler population reaching the carrying capacity of the more suitable types. As burns aged and the Kirtland's warbler population increased, use of plantation habitat became more common [34,93]. After 2004, over 85% of males occupied plantation habitat [27,34]. Plantations were important in the 1970s, with an average of 42% of singing males occupying plantations (Ryel 1979 cited in [58]). However, the relative availability of plantation habitat during this period was not provided [58].

According to sources published through the early 1980s, Kirtland's warblers occupied plantation habitat for shorter periods than wildfire habitat, possibly because plantations have low tree density and thus become suitable later than stands established after wildfire. In addition, plantations have fewer openings, so low branches die earlier resulting in earlier abandonment of plantation habitat compared to wildfire habitat [19,81,130]. Unburned stands that have not been planted for Kirtland's warblers are generally not dense enough to provide suitable habitat [97,99]. Kirtland's warblers occasionally occupy these areas, but for short durations [92] due to significantly later colonization than on plantations (P<0.01) or in stands established following wildfire (P<0.001) [99]. Although not directly investigated, changes in plantation management may have increased occupancy lengths and overall habitat quality in plantations. These changes include incorporating landscape characteristics such as soils and elevation into selection of plantation locations [47,129], adding openings into plantation design [113], and increasing tree density [113,121], plantation size (see Stand-level details), and overall amounts of managed habitat [34,103]. The amount of habitat managed for Kirtland's warbler increased from about 11,400 acres (4,620 ha) in the early 1960s [58] to 153,000 acres (62,000 ha) by the early 2000s [47]

Increased reproductive success and population growth may be associated with wildfire habitat. Based on data on banded Kirtland's warblers collected from May to August from 1990 to 1992 and preliminary data from 1988 and 1989, colonies in plantations had positive growth and produced a similar number of young/nest as wildfire habitat. However, on average 28% of males were unmated in plantations, while only 8% were unmated in wildfire habitat. In addition the occurrence of polygyny was 22% in wildfire habitat compared to 6% in plantation habitat. This results in fewer young produced per male in plantation habitat [13]. The wildfire sites investigated were generally larger than the plantations (Donner 2007 cited in [34]), which may have influenced differences between Kirtland's warbler demography in the 2 habitat types [34]. See Landscape factors for importance of patch size. There is some evidence that prior to widespread brown-headed cowbird control, nesting success was higher in plantation habitat than in wildfire habitat. Following brown-headed cowbird control, nesting success in the 2 habitats was similar [10].

Increased population growth may also be associated with wildfire habitat. One explanation for faster population growth from 1988 to 1994 than following 1994 is lower productivity in plantations. Kirtland's warblers began moving out of aging burns and into plantations in about 1994 [34]. For more detail on the relationship between population size and habitat availability, see Factors influencing population size.

Insect abundance, diversity, and distribution near the edge of jack pine openings was similar (P>0.05) in stands planted for Kirtland's warbler and stands that established after wildfire. Differences at other scales were not investigated [39].

Differences in habitat characteristics: Stand-level features of Kirtland's warbler breeding habitat vary with stand origin. In young (3-6 years old) and older (12-17 years old) jack pine stands, stem densities, patchiness, and snag density were less in plantations that were probably not burned under prescription than in wildfire habitat. In young plantations, jack pine stem density averaged 2,300 stems/ha compared to 11,000 stems/ha in stands regenerated following wildfire. Snags were significantly (P=0.002) less dense in plantation habitat than in wildfire habitat. Snag density averaged 3 snags/ha in young plantations and 83 snags/ha in older plantations compared to 252 snags/ha in young wildfire habitat and 156 snags/ha in older wildfire habitat [113]. A study from the early 1990s reported similar results. Jack pine density was significantly (P=0.001) less in plantations than wildfire habitat, and there were marginally (P=0.075) fewer openings in plantations than wildfire habitat. Jack pine cover category (P=0.002) and relative frequency (P=0.006) were significantly lower in plantations compared to wildfire habitat [13]. Jack pine density was greater on sites burned by the Mack Lake Fire than plantations that were burned under prescription and then planted after a 1- or 3-year delay [46]. Mayfield [81] noted lower occurrence of fire-killed snags and cavity-nesting birds in plantations than wildfire areas.

Ground cover characteristics important to Kirtland's warblers also vary with stand origin (see Breeding habitat). In young (3-6 years old) and older (12-17 years old) jack pine stands, coarse woody debris and forest floor mass were less in plantations that were probably not burned under prescription than in wildfire habitat. Coarse woody debris volume averaged 12 m³/ha in young plantation habitat, significantly (P=0.002) less than the 49 m³/ha average volume in young wildfire habitat. Volume of coarse woody debris increased with age of plantation habitat and declined with age of wildfire habitat [113]. Based on data from the early 1990s, coarse woody debris cover category was significantly (P<0.001) lower in plantations than wildfire habitat. Kinnikinnick cover category (P=0.006) and relative frequency (P=0.002) were significantly lower in plantations compared to wildfire habitat. In contrast, yellow sedge (Carex pensylvanica) cover category was significantly (P<0.001) higher in plantations than in wildfire habitat [13]. Coverage of low sweet blueberry was greater (P<0.003) on sites that were burned and planted with jack pine 3 years later than on unburned plots with either 1- or 3-year planting delays [46]. Based on data collected from 1977 to 1986, burned sites, including one site burned under prescription, had significantly (P<0.005) less cover of bare ground and/or litter, "coarse grass" (which included bluegrasses and brome grass (Brome sp.)), and "sedge/grass" (which included yellow sedge, oatgrasses, wavy hairgrass (Deschampsia flexuosa), prairie Junegrass (Koeleria macrantha), and ricegrass (Oryzopsis sp.)) than 3 unburned plantations and 2 naturally regenerated unburned areas. These burned areas had significantly (P<0.001) more cover of lichen and/or moss, dead wood, and sweetfern than the unburned areas. However, use of habitats with a wide range of ground cover characteristics suggested that Kirtland's warblers were not selecting habitats with specific ground cover composition [95]. The naturally regenerated areas may have provided low-quality habitat [99]. Within Kirtland's warbler habitat site differences influence the response of vegetation to fire. For instance, tree coverage before the Mack Lake Fire was positively associated with postfire coverage of both kinnikinnick (P=0.01) and sweetfern (P<0.01) [95].

FIRE REGIMES:
Breeding habitat: Jack pine forests experience varying fire severities, with stand-replacement fires occurring about every 30 to 60 years during spring, summer, or fall. Before European settlement of the northern Great Lakes Region, large, stand-replacement fires occurred at variable frequencies, and low-severity surface fires occurred at high frequencies [56]. The fire season in the northern Great Lakes region lasts from April to October. From 1960 to 1969, fires on the Huron-Manistee National Forest peaked from April to June [41]. Fire season within the core of the Kirtland's warbler range in northern Lower Michigan peaks in late April to early May (Huber 2009 personal communication [48]). Fires in northeastern Wisconsin typically occurred from May to October [67]. Risk of fire in jack pine is especially high during dry periods [81,118], which are relatively common in northeastern Wisconsin [67].

Although jack pine stands often experience severe, stand-replacement fires [23,51], several factors influence fire severity, including weather [56,127], site characteristics [56], and fuel conditions [114,127]. An immature jack pine stand with an average height of 27 feet (8.2 m) "readily sustained" crown fires due to the proximity of surface and canopy fuels [114]. Johnson [53] also notes the horizontal and vertical fuel continuity in immature jack pine stands. Hutto and others [51] suggest that Kirtland's warbler dependence on large areas of young jack pine trees is evidence that stand-replacement fires are an important component of jack pine fire regimes at high, mid-, and to some extent, low elevations. However, surface fires do occur in jack pine forests. In mature jack pine the base height of the live crown is well above the ground. This and the lack of surface and ladder fuels [53] contribute to a low incidence of crown fire in mature stands [114].

Fires generally occur about every 30 to 60 years in jack pine forests in Michigan [25,108], although fire has become infrequent with fire exclusion. Based on General Land Office data collected in Michigan from 1836 to 1858, the fire-return interval in the 836,200-acre (338,402 ha) jack pine association was 59 years [25]. An estimate of mean fire frequency of 28 years for the Mack Lake area was based on historical data from 1820 to 1980 [108]. There were multiple large wildfires during the late 1890s and early 1900s within the Kirtland's warbler range, including at least 3 individual fires over 90,000 acres (247,000 ha) [56]. These fires may have provided abundant habitat and resulted in an increased Kirtland's warbler population (see Other status). Effective fire suppression followed, with a coincident drop in the acres burned annually in Michigan from 573,000 acres (232,000 ha) from 1910 to 1925 to 27,000 acres (11,000 ha) from 1939 to 1948 (Mitchell and Robson 1950 cited in [56]). Based on data from 1985 to 2000, the mean fire-return interval in the jack pine association had increased from 58 to 787 years [25]. Mayfield [81] speculated that before European settlement fires were less frequent but larger in extent and created more Kirtland's warbler wildfire habitat than was available in the mid-1900s.

The Fire Regime Table summarizes characteristics of fire regimes for vegetation communities in which Kirtland's warblers are likely to occur during the breeding season. Follow the links in the table to documents that provide more detailed information on these fire regimes.

Wintering habitat: The relative importance of various habitats used by Kirtland's warblers on the wintering grounds is not well understood. However, most habitats likely experience fairly frequent disturbance. The Caribbean pine forests occupied by Kirtland's warblers in The Bahamas experience low-severity surface fires about every 2 to 5 years [29,43]. Mayfield [73] reports fire scars at ground level on Caribbean pines on Grand Bahamas, although he did not consider these habitats widely used. Disturbances such as hurricanes and agricultural practices may play a larger role than fire in broadleaved habitats. There was not enough information to associate Kirtland's warbler with a fire-return interval category in The Bahamas [29].

FIRE MANAGEMENT CONSIDERATIONS:
The use of prescribed burning for Kirtland's warbler began slowly and is still limited. In the early 1970s use of prescribed burning was uncommon and experimental [23]. From the early 1970s to the late 1980s, about 4,500 acres (1,820 ha) were burned under prescription, less than the management goal. The average cost of prescribed burning over this period was a little more than $18.00/acre [134,135]. Acres managed for Kirtland's warbler increased substantially in the late 1980s and 1990s [34]. As of 2009, prescribed burning is rarely used in developing Kirtland's warbler habitat in plantations or elsewhere (Huber 2009 personal communication [48]). A 1988 review includes a list for prioritizing sites for prescribed burning [92] (see Prescribed burning implementation). High priority for burning is assigned to sites that were not burned in the previous rotation, suggesting that a considerable proportion of available, developing, and/or planned plantation habitat during the late 1980s and 1990s was not burned under prescription. Although the cut-burn-plant method of creating plantations is commonly discussed [21,58,92], its benefits for creating Kirtland's warbler habitat and/or mimicking wildfire habitat are not. An exception is Wilson's [135] discussion of the benefits of prescribed burning including an apparent improvement in planting efficiency by removing slash that interferes with planting machinery following conventional cutting, a reduction in "competition" between planted seedlings and residual trees, a reduction of risk of jack pine budworm (Choristoneura pinus) infestation by reducing overstory, a reduction in fuel—which aids fire suppression if needed to prevent fire spread into adjacent occupied habitat—and an opportunity to provide training for suppression activities in occupied habitat [135]. In contrast, other articles mention difficulties in implementing prescribed fire [42,58,58,135], limiting its use. The development of fire surrogates was recommended for promoting coverage of low shrubs [13] and in situations where prescribed burning is not feasible [13,58] (see Reliance on plantations). As of 2009, whole-tree harvest followed by trenching and hand-planting or machine planting is the typical method for development of Kirtland's warbler habitat [47] and provides similar fuel reduction benefits as those of prescribed fire mentioned by Wilson [135]. For details of management and recommendations regarding plantation establishment and design, see Plantation management. Cayford [23] provides several site preparation methods for regeneration of jack pine for timber production purposes.

Fire is excluded from occupied areas to prevent losses of currently available habitat [21]. Huber and others [47] recommend that occupied Kirtland's warbler habitat and developing habitat (1- to 21-year old jack pine stands) be given "very high priority" in fire prevention and suppression plans. However, successful control of severe fire in occupied jack pine stands is largely dependent on weather conditions and fuel characteristics (Huber 2009 personal communication [48]). Thus, suppression of wildfires in occupied habitat may not be possible under severe burning conditions. Details of prescribed fire implementation: The following information is primarily from 2 reviews, the Kirtland's warbler Recovery Plan from 1985 [21] and a review by Probst [92] from 1988. Apparently the most common use for prescribed burning in the 1980s was to prepare sites for planting following timber harvest [21,92]. According to the 1985 Recovery Plan, this method was costly but was the most successful and possibly the fastest way to generate dense jack pine stands [21]. Prescribed burning following seed tree harvests often fails to regenerate jack pine [47,92], and subsequent planting has been needed following use of this method [21,92]. Shelterwood harvests followed by burning have also not resulted in satisfactory jack pine regeneration, and windthrow in these stands may be a problem [92]. Seed tree and shelterwood harvests may have greater success in cooler, moister portions of the Kirtland's warbler's range [47]. In his 1988 review, Probst [92] recommends leaving clumps of mature trees during timber harvest to promote dense patches of regeneration [92]. Sites prioritized for prescribed burning include those with residual jack pines, abundant slash, dense broadleaf undergrowth, and/or were not burned in the last rotation [92]. Based on differences in wildfire and plantation habitat, Bocetti [13] suggested that plantations would better mimic wildfire habitats if logging slash were left on site to fuel more severe prescribed fires. This could lead to greater jack pine and low shrub regeneration. However, in a 2001 review, Huber and others [47] note that burning slash following timber harvest has not produced Kirtland's warbler habitat and suggest that burning standing jack pine may provide Kirtland's warbler habitat. Bocetti [13] also recommended increasing the number of openings in plantations and investigating the potential of alternative scarification techniques to increase abundance of shrubby groundlayer species. Based primarily on the response of low sweet blueberry to experimental planting, Houseman and Anderson [46] recommend a planting delay of 3 years following prescribed burning and 1 year if prescribed burning is not possible. In contrast, Probst and Donnerwright [95] suggest that prescribed burning should focus on jack pine regeneration (see Indirect habitat effects) and creation of diverse groundlayer vegetation, not regeneration of specific ground cover species. Since Kirtland's warblers use habitats with a wide range of groundlayer characteristics and planting breaks up sedge (Carex spp.) mats, which could reduce groundlayer species diversity, they suggest that prescribed burning to maintain ground cover for the Kirtland's warbler is not necessary after every harvest [95]. Ideas presented in the Kirtland's warbler Recovery Plan for future use of fire include reducing the use of fire suppression in habitat suitable for Kirtland's warblers and incorporating firebreaks to contain fires and reduce risk of unwanted fire spread [21]. On poor sites where jack pine is not merchantable, prescribed fire is recommended as the primary tool for generating dense jack pine stands, along with other noncommercial treatments [21]. For various combinations of harvest, site preparation, and regeneration methods used in attempting to create Kirtland's warbler habitat, see the 1988 review by Probst [92]. More details of plantation characteristics and selection are available in Plantation management. A 1971 article provides considerable information on techniques used when prescribed fire was first being implemented in jack pine forests [23].

Management in recently burned areas: In addition to adjusting the habitat development schedule following wildfires [47], recommendations for recently burned areas range from implementing measures to improve Kirtland's warbler habitat quality [21,92] to performing no management activities [113]. Reviews of Kirtland's warbler management suggest in-fill planting in poorly-stocked stands 5 or 6 years after prescribed fire or wildfires [21,92], when stands are less than 5 feet (1.5 m) in height [21]. Improvements to habitat may be especially beneficial in occupied habitat or developing habitat near occupied areas. In addition to in-fill planting in poorly-stocked areas, small openings may be created to increase patchiness, and spot burning may improve suitability of groundlayer vegetation [21]. Although preliminary observations suggested that Kirtland's warbler abundance was similar in areas that were not salvage logged and those that were (Taylor 2009 personal communication [120]), Huber and others [47] do not recommend salvage logging in most cases. Due to the differences between plantations and wildfire habitats, Spaulding and Rothstein [113] suggest that providing structure associated with young burned areas is best achieved by leaving wildfire-burned areas alone, including no salvage logging. Increases in the amounts of mature habitat on the landscape were also recommended to provide a "more natural distribution of stand ages" [113], although historic proportions of varying age classes of jack pine within the Kirtland's warbler range were not addressed. See differences in habitat characteristics and reliance on plantations for discussions of some negative consequences of widespread use of plantations to provide Kirtland's warbler habitat.

Limitations of prescribed burning: Obstacles to prescribed burning include a narrow window of acceptable burning conditions [58,135] that occur only about 20 days out of the year [135], land use conflicts, and air quality concerns (Smith 2002 cited in [42]), which add an additional requirement of appropriate wind direction near areas with a substantial human presence [135]. In addition to the difficulties of using prescribed fire, it is not clear if prescribed fire creates conditions that mimic wildfire habitat. The degree to which use of prescribed fire in plantation establishment could minimize differences between plantations and wildfire has not been directly addressed.

Mitigation of the impacts of disturbance: Habitat management activities are best performed outside the breeding season, mostly in fall and winter, to reduce impacts on Kirtland's warblers [21]. In addition to reducing direct impact of humans on Kirtland's warblers [107], area closures and limited use of mechanized equipment during periods with high fire danger reduce risk of human-caused fire during the breeding and postbreeding seasons [118].

Reliance on plantations: The amount of plantation habitat suitable for Kirtland's warblers increased from about 12,400 acres (5,000 ha) in the early 1980s to over 49,400 acres (20,000 ha) as of 2004 [34]. The relative importance of plantations to Kirtland's warblers has also increased, with less than 5,000 acres (2,000 ha) of plantation habitat occupied by Kirtland's warblers in each year of the 1980s and over 22,200 acres (9,000 ha) of plantation habitat occupied by 85% of male Kirtland's warblers in 2004 [34] (see Comparision of habitats of varying origins).

Managers rely on plantations because of the risks of using fire to provide Kirtland's warbler habitat. First, not all the area burned in a wildfire becomes suitable Kirtland's warbler habitat [61,79]. According to a review by Kline [61], none of the wildfires that burned within the breeding range from 1981 to 1989 provided suitable Kirtland's warbler nesting habitat. Second, the stand-replacement fires that historically produced Kirtland's warbler habitat in jack pine forests (see Indirect Fire Effects) are logistically difficult to mimic and pose high liability risks [61] in the Kirtland's warbler range that has an increasing human presence [42]. Use of large fires in this context is not feasible socially or politically [113]. As of 2009, it is not clear from the literature if prescribed burning provides any substantial benefits over other site preparation methods either in quality of Kirtland's warbler habitat or in greater similarities to wildfire habitat. In addition, the difficulties associated with prescribed surface fires used to prepare sites for jack pine planting are many (see Limitations of prescribed burnin). The Mack Lake Fire started as a prescribed surface fire that escaped control [36]. It ultimately burned about 25,000 acres (10,000 ha) [36,108], destroyed 44 structures [58], and killed Forest Service technician Jim Swiderski [36,61]. Discussions of the circumstances behind [36,108], costs of [36,58,61,108], and lessons learned [59,108] from the Mack Lake Fire are available.

Lingering impacts of the Mack Lake Fire include a possible decrease in the use of prescribed burning, at least through the 1980s, and an increased focus on other site preparation methods, including planting following whole tree harvest, burning slash piles following harvest, and/or mechanical scalping [58]. Although not discussed in the literature, it is likely that the Mack Lake Fire is at least partially responsible for the dependence on the whole tree harvest, plowing, and planting methodology that was most used for creation of Kirtland's warbler habitat as of 2009.  
Machine planting of jack pine for Kirtland's warbler habitat.
Phil Huber, USDA Forest Service Find-a-Photo

Although the focus of plantations is providing habitat for Kirtland's warblers, managers stress that these areas meet multiple-use objectives and are part of broader management of the jack pine ecosystem. Probst and Ennis [96] note that Kirtland's warbler plantations provide timber and recreation and meet habitat needs of several species that occur in mature jack pine communities. In addition, different management in adjacent lands meets the needs of species with different requirements [96]. Although the primary reason for a 50-year harvest rotation is to maximize the commercial value of the timber, extended rotation lengths also allow for accommodation of other wildlife needs [47,58]. For an analysis of incorporating multiple objectives, including Kirtland's warbler habitat, into ecosystem management planning, see Zollner and others [136].

There is concern regarding the consequences of replacing the natural disturbance of wildfire with an artificial one with a primary goal of providing habitat for a single species. Differences in nitrification between wildfire-burned and harvested stands may ultimately degrade sites because of leaching [65]. Houseman and Anderson [46] recommend an investigation of the impacts of furrows created during jack pine planting on soil systems, seed banks, and other groundlevel ecosystem components, and suggest seeding openings in plantations with native species to restore and maintain understory species of jack pine barrens [46]. Structural differences between burned and harvested areas (see Differences in habitat characteristics) may impact other wildlife species in jack pine habitat, especially those that require snags and/or coarse woody debris [113]. Spaulding and Rothstein [113] recommend leaving burned areas alone to protect the features of recently burned habitat (see Management in recently burned areas) that are not represented in plantations (see Differences in habitat characteristics). Hutto and others [51] also note that using alternatives to the natural disturbance processes for single-species management prevents large-scale management that conforms to evolutionary disturbance processes.

APPENDIX: FIRE REGIME TABLE

SPECIES: Dendroica kirtlandii
The following table provides fire regime information that may be relevant to Kirtland's warbler habitats. Follow the links in the table to documents that provide more detailed information on these fire regimes.

Fire regime information on vegetation communities in which Kirtland's warbler is mostly likely to occur. This information is taken from the LANDFIRE Rapid Assessment Vegetation Models [63], which were developed by local experts using available literature and expert opinion. This table summarizes fire regime characteristics for each plant community listed. The PDF file linked from each plant community name describes the model and synthesizes the knowledge available on vegetation composition, structure, and dynamics in that community. Cells are blank where information is not available in the Rapid Assessment Vegetation Model.
Great Lakes
Vegetation Community (Potential Natural Vegetation Group) Fire severity* Fire regime characteristics
Percent of fires Mean interval
(years)
Minimum interval
(years)
Maximum interval
(years)
Great Lakes Woodland
Great Lakes pine barrens Replacement 8% 41 10 80
Mixed 9% 36 10 80
Surface or low 83% 4 1 20
Great Lakes Forested
Great Lakes pine forest, jack pine Replacement 67% 50    
Mixed 23% 143    
Surface or low 10% 333    
*Fire Severities
Replacement: Any fire that causes greater than 75% top removal of a vegetation-fuel type, resulting in general replacement of existing vegetation; may or may not cause a lethal effect on the plants.
Mixed: Any fire burning more than 5% of an area that does not qualify as a replacement, surface, or low-severity fire; includes mosaic and other fires that are intermediate in effects.
Surface or low: Any fire that causes less than 25% upper layer replacement and/or removal in a vegetation-fuel class but burns 5% or more of the area [44,62].

REFERENCES

SPECIES: Dendroica kirtlandii
1. Aird, Paul. 1989. The dispersal of the Kirtland's warbler: myths and reality. In: Ennis, K. R., ed. At the crossroads--extinction or survival: Proceedings, Kirtland's warbler symposium; 1989 February 9-11; Lansing, MI. Cadillac, MI: U.S. Department of Agriculture, Forest Service, Huron-Manistee National Forests: 32-37. [19204]
2. Albert, Dennis A. 1995. Regional landscape ecosystems of Michigan, Minnesota, and Wisconsin: a working map classification--4th revision: July 1994. Gen. Tech. Rep. NC-178. St. Paul, MN: U.S. Department of Agriculture, Forest Service, North Central Forest Experiment Station. 250 p. [27980]
3. American Ornithologists' Union. 1957. Checklist of North American birds. 5th ed. Baltimore, MD: The Lord Baltimore Press, Inc. 691 p. [21235]
4. American Ornithologists' Union. 2010. The A.O.U. check-list of North American birds, 7th ed., [Online]. American Ornithologists' Union (Producer). Available: http://www.aou.org/checklist/north/index.php. [50863]
5. Anderson, Walter L.; Storer, Robert W. 1976. Factors influencing Kirtland's warbler nesting success. The Jack-Pine Warbler. 54(3): 105-115. [75667]
6. Baker, Richard J. 1989. Population viability and the Kirtland's warbler. In: Ennis, K. R., ed. At the crossroads--extinction or survival: Proceedings, Kirtland's warbler symposium; 1989 February 9-11; Lansing, MI. Cadillac, MI: U.S. Department of Agriculture, Forest Service, Huron-Manistee National Forests: 12-26. [51031]
7. Barnes, Burton V. 1996. Silviculture, landscape ecosystems, and the Iron Law of the Site. Forstarchiv. 67(3): 226-235. [75765]
8. Beaufait, William R. 1960. Some effects of high temperatures on the cones and seeds of jack pine. Forest Science. 6(3): 194-199. [12407]
9. Berger, Andrew J.; Radabaugh, Bruce E. 1968. Returns of Kirtland's warblers to the breeding grounds. Bird-Banding. 39(3): 161-186. [75668]
10. Bergland, Mark. 1983. Factors influencing nesting success of Kirtland's warbler. In: Walkinshaw, Lawrence H. Kirtland's warbler: The natural history of an endangered species. Bulletin 58. Bloomfield Hills, MI: Cranbrook Insititute of Science: 161-183. [76756]
11. Bocetti, Carol I. 1989. Kirtland's warbler reintroduction study. In: Ennis, K. R., ed. At the crossroads--extinction or survival: Proceedings, Kirtland's warbler symposium; 1989 February 9-11; Lansing, MI. Cadillac, MI: U.S. Department of Agriculture, Forest Service, Huron-Manistee National Forests: 63-68. [51057]
12. Bocetti, Carol I. 1993. Hatching year Kirtland's warbler captured in unusual habitat. The Wilson Bulletin. 105(3): 532-533. [75669]
13. Bocetti, Carol I. 1994. Density, demography, and mating success of Kirtland's warblers in managed and natural habitats. Columbus, OH: Ohio State University. 140 p. Dissertation. [76482]
14. Botkin, Daniel B.; Saxe, Henrik; Araujo, Miguel B.; Betts, Richard; Bradshaw, Richard H. W.; Cedhagen, Tomas; Chesson, Peter; Dawson, Terry P.; Etterson, Julie R.; Faith, Daniel P.; Ferrier, Simon; Guisan, Antonine; Hansen, Anja Skjoldbor; [and others]. 2007. Forecasting the effects of global warming on biodiversity. BioScience. 57(3): 227-236. [75670]
15. Botkin, Daniel B.; Woodby, Douglas A.; Nisbet, Robert A. 1991. Kirtland's warbler habitats: a possible early indicator of climatic warming. Biological Conservation. 56(1): 63-78. [14415]
16. Brewer, Bruce; Keplar, Cameron; Moen, Sharon; Bocetti, Carol; Sykes, Paul. 1996. Population biology, life history characteristics, and simulation modeling. In Seal, U. S., ed. Kirtland's warbler (Dendroica kirtlandii) population and habitat viability assessment. Workshop report--Conservation Breeding Specialist Group (IUCN/SSC) and the U.S. Fish and Wildlife Service; 1992 January 7-9; Bloomington, MN. Apple Valley, MN: International Union of Conservation Networks, Species Survival Commission, Conservation Breeding Specialist Group: 13-28. [In partial fulfillment of USFWS Contract # 14-160003-91-995]. [78247]
17. Brewer, Richard; Morris, Kenneth. 1984. Cross-fostering as a management tool for the Kirtland's warbler. The Journal of Wildlife Management. 48(3): 1041-1045. [76990]
18. Brock, Kenneth J. 1995. Kirtland's warbler: Indiana's first fall record. Indiana Audubon Quarterly. 73(1): 1-2. [75768]
19. Buech, Richard R. 1980. Vegetation of a Kirtland's warbler breeding area and 10 nest sites. The Jack-Pine Warbler. 58(2): 59-72. [75671]
20. Bull, James N. 1989. Spring migration in the Kirtland's warbler. In: Ennis, K. R., ed. At the crossroads--extinction or survival: Proceedings, Kirtland's warbler symposium; 1989 February 9-11; Lansing, MI. Cadillac, MI: U.S. Department of Agriculture, Forest Service, Huron-Manistee National Forests: 40-50. [51037]
21. Byelich, John; DeCapita, Michael E.; Irvine, George; Radtke, Robert E.; Johnson, Nels I.; Jones, Wesley, R.; Mayfield, Harold; Mahalak, William J. 1985. Kirtland's warbler...recovery plan. Twin Cities, MN: U.S. Fish and Wildlife Service. 39 p. [+ appendices]. [Prepared by the Kirtland's Warbler Recovery Team]. [77007]
22. Carey, Eric; Wunderle, Joseph M., Jr.; Ewert, David N. 2004. A research and training program for conservation of wintering Kirtland's warbler and associated species in the Bahamas: the first field season. Journal of Caribbean Ornithology. 17(Special Issue Honoring Nedra Klein): 81-85. [75672]
23. Cayford, J. H. 1971. The role of fire in the ecology and silviculture of jack pine. In: Proceedings, annual Tall Timbers fire ecology conference; 1970 August 20-21; Fredericton, NB. No. 10. Tallahassee, FL: Tall Timbers Research Station: 221-244. [18941]
24. Chandler, Craig; Cheney, Phillip; Thomas, Philip; Trabaud, Louis; Williams, Dave. 1983. Fire in forestry: Vol. I. Forest fire behavior and effects. New York: John Wiley & Sons. 450 p. [12241]
25. Cleland, David T.; Crow, Thomas R.; Saunders, Sari C.; Dickmann, Donald I.; Maclean, Ann L.; Jordan, James K.; Watson, Richard L.; Sloan, Alyssa M.; Brosofske, Kimberley D. 2004. Characterizing historical and modern fire regimes in Michigan (USA): a landscape ecosystem approach. Landscape Ecology. 19: 311-325. [54326]
26. Clench, Mary Heimerdinger. 1973. The fall migration route of Kirtland's warbler. The Wilson Bulletin. 85(4): 417-428. [75674]
27. Committe on the Status of Endangered Wildlife in Canada. 2008. COSEWIC assessment and update status report on the Kirtland's warbler (Dendroica kirtlandii), in Canada--Endangered 2008, [Online]. In: Species at risk public registry--COSEWIC status reports. Ottawa: Environment Canada, Committee on the Status of Endangered Wildlife in Canada (Producer). Available: http://www.sararegistry.gc.ca/virtual_sara/files/cosewic/sr%5Fkirtland%5Fwarbler%5F0808%5Fe%2Epdf [2010, January 4]. [75949]
28. Currie, Dave; Wunderle, Joe; Ewert, Dave; Carey, Eric. 2003. The most elusive bird in the Bahamas? Getting to grips with wintering Kirtland's warblers. World Birdwatch. 25(4): 13-15. [75675]
29. Currie, Dave; Wunderle, Joseph M. Jr.; Ewert, David N.; Anderson, Matthew R.; Davis, Ancilleno; Turner, Jasmine. 2005. Habitat distribution of birds wintering in central Andros, The Bahamas: Implications for management. Caribbean Journal of Science. 41(1): 75-87. [75676]
30. Currie, Dave; Wunderle, Joseph M., Jr.; Ewert, David N.; Davis, Ancilleno; McKenzie, Zeko. 2005. Winter avian distribution and relative abundance in six terrestrial habitats on southern Eleuthera, The Bahamas. Caribbean Journal of Science. 41(1): 88-100. [75677]
31. DeCapita, Michael E. 2000. Brown-headed cowbird control on Kirtland's warbler nesting areas in Michigan, 1972-1995. In: Smith, James N.; Cook, Terry L.; Rothstein, Stephen I.; Robinson, Scott K.; Sealy, Spencer G., eds. Ecology and management of cowbirds and their hosts. Studies in the Conservation of North American Passerine Birds. Austin, TX: The University of Texas Press: 333-341. [75678]
32. Deloria-Sheffield, Christie M.; Millenbah, Kelly F.; Bocetti, Carol I.; Sykes, Paul W., Jr.; Kepler, Cameron B. 2001. Kirtland's warbler diet as determined through fecal analysis. The Wilson Bulletin. 113(4): 384-387. [75680]
33. Dickson, James G. 2002. Fire and bird communities in the South. In: Ford, W. Mark; Russell, Kevin R.; Moorman, Christopher E., eds. The role of fire in nongame wildlife management and community restoration: traditional uses and new directions: Proceedings of a special workshop; 2000 December 15; Nashville, TN. Gen. Tech. Rep. NE-288. Newtown Square, PA: U.S. Department of Agriculture, Forest Service, Northeastern Research Station: 52-57. [41557]
34. Donner, Deahn M.; Probst, John R.; Ribic, Christine A. 2008. Influence of habitat amount, arrangement, and use on population trend estimates of male Kirtland's warblers. Landscape Ecology. 23(4): 467-480. [75681]
35. Donner, Deahn M.; Ribic, Christine A.; Probst, John R. 2009. Male Kirtland's warblers' patch-level response to landscape structure during periods of varying population size and habitat amounts. Forest Ecology and Management. 258: 1093-1101. [75906]
36. Doyle, Nancy. 1980. Forest Service fire claims Michigan man. Northern Logger and Timber Processor. 29(2): 10-11. [19239]
37. Eyre, F. H.; LeBarron, Russell K. 1944. Management of jack pine stands in the Lake States. Tech. Bull. No. 863. Washington, DC: U.S. Department of Agriculture, Forest Service. 66 p. [11643]
38. Finch, Deborah M.; Ganey, Joseph L.; Yong, Wang; Kimball, Rebecca T.; Sallabanks, Rex. 1997. Effects and interactions of fire, logging, and grazing. In: Block, William M.; Finch, Deborah M., tech. eds. Songbird ecology in southwestern ponderosa pine forests: a literature review. Gen. Tech. Rep. RM-GTR-292. Fort Collins, CO: U.S. Department of Agriculture, Forest Service, Rocky Mountain Forest and Range Experiment Station: 103-136. [27990]
39. Fussman, Jennifer Lynn. 1997. Foraging ecology of Kirtland's warbler in managed and natural breeding habitat. Columbus, OH: The Ohio State University. 73 p. Thesis. [75867]
40. Grveles, Kim. 2009. The first annual census of the Kirtland's warbler (Dendroica kirtlandii) in Wisconsin. The Passenger Pigeon. 71(2): 123-130. [75683]
41. Haines, Donald A.; Johnson, Von J.; Main, William A. 1975. Wildfire atlas of the Northeastern and North Central States. Gen. Tech. Rep. NC-16. St. Paul, MN: U.S. Department of Agriculture, Forest Service, North Central Forest Experiment Station. 25 p. [25732]
42. Hamel, Paul B.; Rosenberg, Kenneth V.; Buehler, David A. 2005. Is management for golden-winged warblers and cerulean warblers compatible? In: Ralph, C. John; Rich, Terrell D., eds. Bird conservation implementation and integration in the Americas: proceedings of the 3rd international Partners in Flight conference. Volume 1; 2002 March 20-24; Asilomar, CA. Gen. Tech. Rep. PSW-GTR-191. Albany, CA: U.S. Department of Agriculture, Forest Service, Pacific Southwest Research Station: 322-331. [63721]
43. Haney, J. Christopher; Lee, David S.; Walsh-McGehee, Martha. 1998. A quantitative analysis of winter distribution and habitats of Kirtland's warblers in the Bahamas. The Condor. 100(2): 201-217. [75862]
44. Hann, Wendel; Havlina, Doug; Shlisky, Ayn; [and others]. 2008. Interagency fire regime condition class guidebook. Version 1.3, [Online]. In: Interagency fire regime condition class website. U.S. Department of Agriculture, Forest Service; U.S. Department of the Interior; The Nature Conservancy; Systems for Environmental Management (Producer). 119 p. Available: http://frames.nbii.gov/frcc/documents/FRCC_Guidebook_2008.07.10.pdf [2008, September 03]. [70966]
45. Hoffman, Randy. 1989. History of Kirtland's warbler found in Wisconsin. In: Ennis, K. R., ed. At the crossroads--extinction or survival: Proceedings, Kirtland's warbler symposium; 1989 February 9-11; Lansing, MI. Cadillac, MI: U.S. Department of Agriculture, Forest Service, Huron-Manistee National Forests: 29-31. [19205]
46. Houseman, Gregory R.; Anderson, Roger C. 2002. Effects of jack pine plantation management on barrens flora and potential Kirtland's warbler nest habitat. Restoration Ecology. 10(1): 27-36. [75691]
47. Huber, Philip W.; Weinrich, Jerry A.; Carlson, Elaine S. 2001. Strategy for Kirtland's warbler habitat management. Lansing, MI: Michigan Department of Natural Resources; Milwaukee, WI: U.S. Department of Agriculture, Forest Service, Eastern Region; Fort Snelling, MN: U.S. Department of the Interior, Fish and Wildlife Service, Midwest Region. 27 p. [78002]
48. Huber, Philip. 2009. [Personal communication]. December 2. Regarding Kirtland's warbler habitat management and fire regime within Kirtland's warblers range. Mio, MI: U.S. Department of Agriculture, Forest Service, Huron-Manistee National Forest, Mio Ranger District. On file at: U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station, Fire Sciences Laboratory, Missoula, MT; FEIS files. [77450]
49. Hunter, William C.; Pashley, David N.; Escano, Ronald E. F. 1993. Neotropical migratory landbird species and their habitats of special concern within the Southeast Region. In: Finch, Deborah M.; Stangel, Peter W., eds. Status and management of neotropical migratory birds: Proceedings; 1992 September 21-25; Estes Park, CO. Gen. Tech. Rep. RM-229. Fort Collins, CO: U.S. Department of Agriculture, Forest Service, Rocky Mountain Forest and Range Experiment Station: 159-171. [65648]
50. Huntington, Greg. 1989. Military cooperation in the Kirtland's warbler recovery effort. In: Ennis, K. R., ed. At the crossroads--extinction or survival: Proceedings, Kirtland's warbler symposium; 1989 February 9-11; Lansing, MI. Cadillac, MI: U.S. Department of Agriculture, Forest Service, Huron-Manistee National Forests: 54. [51049]
51. Hutto, Richard L.; Conway, Courtney J.; Saab, Victoria A.; Walters, Jeffrey R. 2009. What constitutes a natural fire regime? Insight from the ecology and distribution of coniferous forest birds in North America. Fire Ecology Special Issue. 4(2): 115-132. [74275]
52. James, Frances C.; McCulloch, Charles E. 1995. The strength of inferences about causes of trends in populations. In: Martin, Thomas E.; Finch, Deborah M., eds. Ecology and management of neotropical migratory birds: A synthesis and review of critical issues. New York: Oxford University Press: 40-51. [76385]
53. Johnson, Edward A. 1992. Fire and vegetation dynamics: Studies from the North American boreal forest. Cambridge Studies in Ecology. Cambridge, UK: Cambridge University Press. 129 p. [19950]
54. Kashian, Daniel M.; Barnes, Burton V. 2000. Landscape influence on the spatial and temporal distribution of the Kirtland's warbler at the Bald Hill burn, northern Lower Michigan, U.S.A. Canadian Journal of Forest Research. 30: 1895-1904. [38800]
55. Kashian, Daniel M.; Barnes, Burton V.; Walker, Wayne S. 2003. Landscape ecosystems of northern Lower Michigan and the occurrence and management of the Kirtland's warbler. Forest Science. 49(1): 140-159. [43862]
56. Keane, Robert E.; Agee, James K.; Fule, Peter; Keeley, Jon E.; Key, Carl; Kitchen, Stanley G.; Miller, Richard; Schulte, Lisa A. 2008. Ecological effects of large fires on US landscapes: benefit or catastrophe? International Journal of Wildland Fire. 17: 696-712. [73387]
57. Kelly, Sean T.; DeCapita, Michael E. 1982. Cowbird control and its effect on Kirtland's warbler reproductive success. The Wilson Bulletin. 94(3): 363-365. [27702]
58. Kepler, Cameron B.; Irvine, G. William; DeCapita, Michael E.; Weinrich, Jerry. 1996. The conservation management of Kirtland's warbler, Dendroica kirtlandii. Bird Conservation International. 6(1): 11-22. [75698]
59. Kilgore, Bruce M. 1982. Fire management programs in national parks and wilderness. In: Lotan, James E., ed. Fire--its field effects: Proceedings of the symposium; 1982 October 19-21; Jackson, WY. Missoula, MT: The Intermountain Fire Council; Pierre, SD: The Rocky Mountain Fire Council: 61-90. [30691]
60. Klimkiewicz, M. Kathleen; Clapp, Roger B.; Futcher, Anthony G. 1983. Longevity records of North American birds: Remizidae through Parulinae. Journal of Field Ornithology. 54(3): 287-294. [75701]
61. Kline, David. 1989. A manager's perspective. In: Ennis, K. R., ed. At the crossroads--extinction or survival: Proceedings, Kirtland's warbler symposium; 1989 February 9-11; Lansing, MI. Cadillac, MI: U.S. Department of Agriculture, Forest Service, Huron-Manistee National Forests: 55-56. [51050]
62. LANDFIRE Rapid Assessment. 2005. Reference condition modeling manual (Version 2.1), [Online]. In: LANDFIRE. Cooperative Agreement 04-CA-11132543-189. Boulder, CO: The Nature Conservancy; U.S. Department of Agriculture, Forest Service; U.S. Department of the Interior (Producers). 72 p. Available: http://www.landfire.gov/downloadfile.php?file=RA_Modeling_Manual_v2_1.pdf [2007, May 24]. [66741]
63. LANDFIRE Rapid Assessment. 2007. Rapid assessment reference condition models, [Online]. In: LANDFIRE. U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station, Fire Sciences Lab; U.S. Geological Survey; The Nature Conservancy (Producers). Available: http://www.landfire.gov/models_EW.php [2008, April 18] [66533]
64. Latta, Steven C.; Parkes, Kenneth C. 2001. A possible Dendroica kirtlandii hybrid from Hispaniola. The Wilson Bulletin. 113(4): 378-383. [75863]
65. Leduc, Stephen D.; Rothstein, David E. 2007. Initial recovery of soil carbon and nitrogen pools and dynamics following disturbance in jack pine forests: a comparison of wildfire and clearcut harvesting. Soil Biology & Biochemistry. 39(11): 2865-2876. [69767]
66. Leefers, Larry; Mwangi, Albert. 1989. Economics and the preservation of Kirtland's warbler. In: Ennis, K. R., ed. At the crossroads--extinction or survival: Proceedings, Kirtland's warbler symposium; 1989 February 9-11; Lansing, MI. Cadillac, MI: U.S. Department of Agriculture, Forest Service, Huron-Manistee National Forests: 85-93. [51062]
67. Lorimer, Craig G.; Gough, William R. 1988. Frequency of drought and severe fire weather in north-eastern Wisconsin. Journal of Environmental Management. 26: 203-219. [43946]
68. Lyon, L. Jack; Telfer, Edmund S.; Schreiner, David Scott. 2000. Direct effects of fire and animal responses. In: Smith, Jane Kapler, ed. Wildland fire in ecosystems: Effects of fire on fauna. Gen. Tech. Rep. RMRS-GTR-42-vol. 1. Ogden, UT: U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station: 17-23. [44435]
69. MacKinnon, Sherry. 2008. Michigan's 2008 Kirtland's warbler population reaches another record high, [Online]. In: Press Releases--September 29, 2008. Lansing, MI: Michigan Department of Natural Resources (Producer). Available: http://www.michigan.gov/dnr/0,1607,7-153-10371_10402-200899--,00.html [2009, November 11]. [76984]
70. Marshall, Elizabeth; Haight, Robert; Homans, Frances. 1998. Incorporating environmental uncertainty into species management decisions: Kirtland's warbler habitat management as a case study. Conservation Biology. 12(5): 975-985. [76989]
71. Marshall, Elizabeth; Homans, Frances; Haight, Robert. 2000. Exploring strategies for improving the cost effectiveness of endangered species management: the Kirtland's warbler as a case study. Land Economics. 76(3): 462-473. [75781]
72. Mayfield, Harold F. 1972. Third decennial census of Kirtland's warbler. The Auk. 89(2): 263-268. [75710]
73. Mayfield, Harold F. 1972. Winter habitat of Kirtland's warbler. The Wilson Bulletin. 84(3): 347-349. [75712]
74. Mayfield, Harold F. 1975. The numbers of Kirtlands's warblers. The Jack-Pine Warbler. 53(2): 39-47. [24989]
75. Mayfield, Harold F. 1978. Brood parasitism: Reducing interactions between Kirtland's warblers and brown-headed cowbirds. In: Temple, Stanley A., ed. Endangered birds: management techniques for preserving threatened species. Madison, WI: University of Wisconsin Press: 85-91. [75782]
76. Mayfield, Harold F. 1988. Do Kirtland's warblers migrate in one hop? The Auk. 105(1): 204-205. [75705]
77. Mayfield, Harold F. 1989. Perspective on the Kirtland's warbler. In: Ennis, K. R., ed. At the crossroads--extinction or survival: Proceedings, Kirtland's warbler symposium; 1989 February 9-11; Lansing, MI. Cadillac, MI: U.S. Department of Agriculture, Forest Service, Huron-Manistee National Forests: 1-7. [51028]
78. Mayfield, Harold F. 1992. Kirtland's warbler (Dendroica kirtlandii). In: Poole, A.; Stettenheim, P.; Gill, G., eds. The birds of North America. No. 19. Philadelphia, PA: The Academy of Natural Sciences; Washington, DC: The American Ornithologists' Union. 16 p. [78003]
79. Mayfield, Harold F. 1993. Kirtland's warblers benefit from large forest tracts. The Wilson Bulletin. 105(2): 351-353. [22270]
80. Mayfield, Harold. 1953. A census of the Kirtland's warbler. The Auk. 70(1): 17-20. [75714]
81. Mayfield, Harold. 1960. The Kirtland's warbler. Bulletin 40. Bloomfield Hills, MI: Cranbrook Institute of Science. 242 p. [16778]
82. Mayfield, Harold. 1962. 1961 decennial census of the Kirtland's warbler. The Auk. 79(2): 173-182. [75713]
83. McGeen, Daniel S. 1989. A review of the predator-prey interactions versus habitat considerations for the Kirtland's warbler. In: Ennis, K. R., ed. At the crossroads--extinction or survival: Proceedings, Kirtland's warbler symposium; 1989 February 9-11; Lansing, MI. Cadillac, MI: U.S. Department of Agriculture, Forest Service, Huron-Manistee National Forests: 70-84. [51061]
84. Mumford, Russell E.; Keller, Charles E. 1984. The birds of Indiana. Bloomington, IN: Indiana University Press. 376 p. [60761]
85. Nelson, Mark D.; Buech, Richard R. 1996. A test of 3 models of Kirtland's warbler habitat suitability. Wildlife Society Bulletin. 24(1): 89-97. [75911]
86. Nowacki, Gregory J.; Abrams, Marc D. 2008. The demise of fire and "mesophication" of forests in the eastern United States. BioScience. 58(2): 123-138. [70112]
87. Orr, Craig D. 1975. 1974 breeding success of the Kirtland's warbler. The Jack-Pine Warbler. 53(2): 59-66. [75717]
88. Payne, Amy L. 2009. Kirtland's warbler on verge of overcoming endangered status, but habitat problems remain, [Online]. January 14. In: The Bay City Times--Saginaw Bay Watershed Watch: Environmental issues surrounding the Saginaw Bay area. mlive.com (Producer). Available: http://blog.mlive.com/watershedwatch/2009/01/kirtlands_warbler_on_verge_of.html [2009, November 3]. [76959]
89. Potter, Eloise F.; Radovsky, Frank J. 1992. Kirtland's warbler on the North Carolina coast in fall: its occurrence and possible significance. The Chat. 56(2): 21-29. [75718]
90. Probst, John R. 1985. Summer records and management implications of Kirtland's warbler in Michigan's Upper Peninsula. The Jack-Pine Warbler. 63(1): 9-16. [75721]
91. Probst, John R. 1986. A review of factors limiting the Kirtland's warbler on its breeding grounds. The American Midland Naturalist. 116(1): 87-100. [75719]
92. Probst, John R. 1988. Kirtland's warbler breeding biology and habitat management. In: Hoekstra, Thomas W.; Capp, Jack, compilers. Integrating forest management for wildlife and fish. Gen. Tech. Rep. NC-122. St. Paul, MN: U.S. Department of Agriculture, Forest Service, North Central Forest Experiment Station: 28-35. [76264]
93. Probst, John R.; Donner, Deahn M.; Bocetti, Carol I.; Sjogren, Steve. 2003. Population increase in Kirtland's warbler and summer range expansion to Wisconsin and Michigan's Upper Peninsula, USA. Oryx. 37(3): 365-373. [75722]
94. Probst, John R.; Donner, Deahn M.; Worland, Mike; Weinrich, Jerry; Huber, Phillip; Ennis, Kenneth R. 2005. Comparing census methods for the endangered Kirtland's warbler. Journal of Field Ornithology. 76(1): 50-60. [75865]
95. Probst, John R.; Donnerwright, Deahn. 2003. Fire and shade effects on ground cover structure in Kirtland's warbler habitat. The American Midland Naturalist. 149(2): 320-334. [43954]
96. Probst, John R.; Ennis, Kenneth Rex. 1989. Multi-resource values of Kirtland's warbler habitat. In: Ennis, K. R., ed. At the crossroads--extinction or survival: Proceedings, Kirtland's warbler symposium; 1989 February 9-11; Lansing, MI. Cadillac, MI: U.S. Department of Agriculture, Forest Service, Huron-Manistee National Forests: 69. [51059]
97. Probst, John R.; Hayes, Jack P. 1987. Pairing success of Kirtland's warblers in marginal vs. suitable habitat. The Auk. 104(2): 234-241. [75720]
98. Probst, John R.; Weinrich, Jerry. 1989. Predicting Kirtland's warbler populations by habitat conditions. In: Ennis, K. R., ed. At the crossroads--extinction or survival: Proceedings, Kirtland's warbler symposium; 1989 February 9-11; Lansing, MI. Cadillac, MI: U.S. Department of Agriculture, Forest Service, Huron-Manistee National Forests: 61-62. [51054]
99. Probst, John R.; Weinrich, Jerry. 1993. Relating Kirtland's warbler population to changing landscape composition and structure. Landscape Ecology. 8(4): 257-271. [23629]
100. Radabaugh, Bruce E. 1972. Double-broodedness in the Kirtland's warbler. Bird-Banding. 43(1): 55-55. [75723]
101. Radabaugh, Bruce E. 1974. Kirtland's warbler and its Bahama wintering grounds. The Wilson Bulletin. 86(4): 374-383. [75724]
102. Radabaugh, Bruce E.; Radabaugh, Floyd E.; Radabaugh, Clarice A. 1966. Returns of Kirtland's warblers banded as nestlings. The Wilson Bulletin. 78(3): 322. [75725]
103. Radtke, Robert; Irvine, G. William; Byelich, John D. 1989. Kirtland's warbler management. In: Ennis, K. R., ed. At the crossroads--extinction or survival: Proceedings, Kirtland's warbler symposium; 1989 February 9-11; Lansing, MI. Cadillac, MI: U.S. Department of Agriculture, Forest Service, Huron-Manistee National Forests: 8-11. [19206]
104. Robbins, Louise E.; Myers, Ronald L. 1992. Seasonal effects of prescribed burning in Florida: a review. Misc. Publ. No. 8. Tallahassee, FL: Tall Timbers Research, Inc. 96 p. [21094]
105. Rothstein, Stephen I.; Cook, Terry L. 2000. Introduction: Cowbird management, host population regulation and efforts to save endangered species. In: Smith, James N. M.; Cook, Terry L.; Rothstein, Stephen I.; Robinson, Scott K.; Sealy, Spencer G., eds. Ecology and management of cowbirds and their hosts. Studies in the Conservation of North American Passerine Birds. Austin, TX: University of Texas Press: 323–332. [75732]
106. Ryel, L. A. 1979. On the population dynamics of Kirtland's warbler. The Jack-Pine Warbler. 57(2): 76-83. [75731]
107. Ryel, Lawrence A. 1981. Population change in the Kirtland's warbler. The Jack-Pine Warbler. 59(3): 76-91. [75733]
108. Simard, Albert J.; Haines, Donald A.; Blank, Richard W.; Frost, John S. 1983. The Mack Lake fire. Gen. Tech. Rep. NC-83. East Lansing, MI: U.S. Department of Agriculture, Forest Service, North Central Forest Experiment Station. 36 p. [29662]
109. Smith, Elaine Louise. 1979. Analysis of Kirtland’s warbler breeding habitat in Ogemaw and Roscommon Counties, Michigan. East Lansing, MI: Michigan State University. 42 p. Thesis. [75738]
110. Solomon, Barry D. 1998. Impending recovery of Kirtland's warbler: case study of the effectiveness of the Endangered Species Act. Environmental Management. 22(1): 9-17. [75739]
111. Solomon, Barry D. 1998. Public support for endangered species recovery: an exploratory study of the Kirtland's warbler. Human Dimensions of Wildlife. 3(3): 62-74. DOI: 10.1080/10871209809359132. [75740]
112. Southern, William E. 1961. A botanical analysis of Kirtland's warbler nests. The Wilson Bulletin. 73(2): 148-154. [75741]
113. Spaulding, Susan E.; Rothstein, David E. 2009. How well does Kirtland's warbler management emulate the effects of natural disturbance on stand structure in Michigan jack pine forests? Forest Ecology and Management. 258(11): 2609-2618. [76565]
114. Stocks, Brian J.; Alexander, Martin E. 1980. Forest fire behaviour and effects research in northern Ontario: a field oriented program. In: Martin, Robert E.; Edmonds, Robert L.; Faulkner, Donald A.; Harrington, James B.; Fuquay, Donald M.; Stocks, Brian J.; Barr, Sumner, eds. Proceedings, 6th conference on fire and forest meteorology; 1980 April 22-24; Seattle, WA. Washington, DC: Society of American Foresters: 18-24. [10291]
115. Sykes, Paul W., Jr. 1989. Kirtland's warblers on their wintering grounds in the Bahamas Archipelago--a preliminary report. In: Ennis, K. R., ed. At the crossroads--extinction or survival: Proceedings, Kirtland's warbler symposium; 1989 February 9-11; Lansing, MI. Cadillac, MI: U.S. Department of Agriculture, Forest Service, Huron-Manistee National Forests: 28. [19207]
116. Sykes, Paul W., Jr. 1997. Rare, local, little-known, and declining North American breeders--a closer look: Kirtland's warbler. Birding. 29(3): 220-227. [75744]
117. Sykes, Paul W., Jr.; Clench, Mary H. 1998. Winter habitat of Kirtland's warbler: an endangered nearctic/neotropical migrant. The Wilson Bulletin. 110(2): 244-261. [75743]
118. Sykes, Paul W., Jr.; Kepler, Cameron B.; Jett, David A.; DeCapita, Michael E. 1989. Kirtland's warblers on the nesting grounds during the post-breeding period. The Wilson Bulletin. 101(4): 545-558. [75745]
119. Sykes, Paul W., Jr.; Munson, Douglas J. 1989. Late record of Kirtland's warbler on the breeding grounds. The Jack-Pine Warbler. 67(3): 101. [75746]
120. Taylor, Sylvia M. 2009. [Email to Rachelle Meyer]. October 28. Regarding 1993 Kirtland's warbler article. Ann Arbor, MI: University of Michigan, School of Natural Resources and Environment. On file with: U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station, Fire Sciences Laboratory, Missoula, MT; FEIS files. [76991]
121. Trauger, David L.; Bocetti, Carol I. 1993. Kirtland's Warbler Recovery Team effectively coordinates interagency research and management. Endangered Species Technical Bulletin. 18(2): 14, 16-17. [22063]
122. Trautman, Milton B. 1979. Experiences and thoughts relative to the Kirtland's warbler. The Jack-Pine Warbler. 57(3): 135-140. [75749]
123. Trick, Joel A.; Grveles, Kim; Goyette, Jennifer L. 2009. The 2008 nesting season: first documented successful nesting of Kirtland's warbler (Dendroica kirtlandii) in Wisconsin. The Passenger Pigeon. 71(2): 100-114. [75750]
124. Trick, Joel; Grveles, Kim. 2009. Wisconsin Kirtland's warbler updates: Wisconsin Kirtland's warbler 2009 season summary, [Online]. In: Endangered species in the Upper Midwest--Green Bay Ecological Services Office. Green Bay, WI: U.S. Fish and Wildlife Service, Wisconsin Ecological Services Office, Midwest Region (Producer). Available: http://www.fws.gov/midwest/GreenBay/kiwa/2009Summary.html [2009, October 23]. [76738]
125. Twomey, Arthur C. 1936. Climographic studies of certain introduced and migratory birds. Ecology. 17(1): 122-132. [75919]
126. U.S. Department of the Interior, Fish and Wildlife Service. 2013. Endangered Species Program, [Online]. Available: http://www.fws.gov/endangered/. [86564]
127. Van Lear, D. H.; Harlow, R. F. 2002. Fire in the eastern United States: influence on wildlife habitat. In: Ford, W. Mark; Russell, Kevin R.; Moorman, Christopher E., eds. The role of fire in nongame wildlife management and community restoration: traditional uses and new directions: Proceedings of a special workshop; 2000 December 15; Nashville, TN. Gen. Tech. Rep. NE-288. Newtown Square, PA: U.S. Department of Agriculture, Forest Service, Northeastern Research Station: 2-10. [41533]
128. Van Tyne, J. 1953. Kirtland's warbler. In: Bent, Arthur Cleveland. Life histories of North American wood warblers. United States National Museum: Bulletin 203. Washington, DC: Smithsonian Institution: 417-428. [75866]
129. Walker, Wayne S.; Barnes, Burton V.; Kashian, Daniel M. 2003. Landscape ecosystems of the Mack Lake burn, northern Lower Michigan, and the occurrence of the Kirtland's warbler. Forest Science. 49: 119-139. [43728]
130. Walkinshaw, Lawrence H. 1983. Kirtland's warbler: the natural history of an endangered species. Bulletin 58. Bloomfield Hills, MI: Cranbrook Insititute of Science. 207 p. [19203]
131. Walkinshaw, Lawrence H.; Faust, Warren R. 1975. 1974 Kirtland's warbler nesting success in northern Crawford County, Michigan. The Jack-Pine Warbler. 53(2): 54-58. [75754]
132. Watson, Craig; Hayes, Chuck; McCauley, Joseph; Milliken, Andrew. 2005. The South Atlantic Migratory Bird Initiative--an integrated approach to conservation of "all birds across all habitats". In: Ralph, C. John; Rich, Terrell D., eds. Bird conservation implementation and integration in the Americas: proceedings of the 3rd international Partners in Flight conference. Volume 1; 2002 March 20-24; Asilomar, CA. Gen. Tech. Rep. PSW-GTR-191. Albany, CA: U.S. Department of Agriculture, Forest Service, Pacific Southwest Research Station: 266-276. [63712]
133. Weise, Thomas. 1991. Around the states: Michigan. Ecology USA. 20(15): 147. [16688]
134. Wilson, Ronald L. 1987. Prescribed burning in Michigan. Fire Management Notes. 48(4): 12-14. [21167]
135. Wilson, Ronald L. 1989. Fire and fire effects--its impact on forest vegetation for Kirtland's warbler habitat. In: Ennis, K. R., ed. At the crossroads--extinction or survival: Proceedings, Kirtland's warbler symposium; 1989 February 9-11; Lansing, MI. Cadillac, MI: U.S. Department of Agriculture, Forest Service, Huron-Manistee National Forests: 57-60. [19208]
136. Zollner, Patrick A.; Roberts, L. Jay; Gustafson, Eric J.; He, Hong S.; Radeloff, Volker. 2008. Influence of forest planning alternatives on landscape pattern and ecosystem processes in northern Wisconsin, USA. Forest Ecology & Management. 254(3): 429-444. [75758]
137. Zou, Xiaoming; Theiss, Corinna; Barnes, Burton V. 1992. Pattern of Kirtland's warbler occurrence in relation to the landscape structure of its summer habitat in northern Lower Michigan. Landscape Ecology. 6(4): 221-231. [19211]

FEIS Home Page