General Technical Report RMRS-GTR-22
The Northern Goshawk in Utah: Habitat Assessment and Management
Recommendations
General Technical Report RMRS-GTR-22
The Northern Goshawk in Utah: Habitat Assessment and Management Recommendations
Status and Distribution of Habitat in Utah
Habitat Assessment
The goshawk has been located in a variety of forest cover types throughout Utah. As interpreted from Geospatial Analysis Processes (GAP, USGS 1995), they range from the subalpine environments typified by Engelmann spruce and subalpine fir to pinyon/juniper woodlands bordering the grass and shrub lands. In general, the elevations of these forests range from 4,000 to 11,000 ft, with rugged and broken topography. In addition to major mountain ranges, such as the Uinta and Wasatch, Utah contains major plateaus such as the Markagunt and Tavaputs. The climate, and therefore the vegetation of Utah, is highly influenced by elevation and latitude. In general, precipitation on forest lands ranges from 12 to 42 inches annually. With this variation in topography, climate, soils, and geology a wide range of forest compositions and structures are typical.
The forests of Utah are and have been occupied by humans for centuries. Native Americans used and settled in many forests. European settlement of the valleys started in earnest during the mid 1800's. Along with the disturbances caused by human presence (harvesting, burning), natural disturbances from flood, wind, fire, snow, and ice all shaped successional pathways and current vegetation. Since the early-to-mid 1900's, effective fire exclusion has prevented fire from playing its historical role of maintaining and regenerating many of these forests. Also, with European settlement, forests in and along valley bottoms were extensively harvested to supply materials for construction, mining, and the railroad industries. Timber harvesting, mining, domestic livestock grazing, recreation, and fire exclusion continue to impact Utah forests. The historic and current uses and diversity of Utah forests, necessitated the development of a carefully structured assessment for goshawk habitat.
Effective assessments of natural resources require a balance between data resolution and geographic extent. Both are referred to as scale often leading to confusion. Resolution refers to the intensity or graininess of the data. Geographic extent refers to the area (spatial scale) to be addressed (Graham and others
in press; Haynes and others 1996). Both must be appropriate to the questions being asked and the issue being addressed. By definition, this assessment will address goshawk habitat throughout Utah.Another important component of assessments is to conduct them across two or more spatial and temporal scales (Graham and others in press). By conducting assessments at multiple scales, the large geographic extents can set context for the small. Similarly, the long time frames provide context for the short. Conducting assessments at two scales assures processes identifiable at one scale will not be overlooked if they are not observable or easily addressed at the other. For example, the migration or dispersal of the goshawk could not easily be addressed by examining only nest stands, and nest stand characteristics could not easily be addressed using only landscape or subregional assessments. Population trends also cannot be easily addressed using nesting success for a single year, but multiple year observations allow for meaningful descriptions of population trends.
Utah was the largest geographic area used for assessing goshawk habitat. It would have been useful to look at a regional scale to set the Utah assessment in context to explore how the habitat in Utah is related to habitat in adjacent states. But, time, budget, and personnel constraints, did not permit the wider analysis. Only recommendations and inferences on the status of goshawk habitat within Utah were requested by the involved and cooperating parties (appendix A). This assessment is designed to provide general habitat trends for the next 25 years for Utah.
To fully understand influences of goshawk habitat in Utah, and to disclose immediate threats and goshawk habitat risks to the State, a smaller geographic area was required. Counties, watersheds, National Forests, or other political or geographic delineations could have been applied. An important component of choice was using a geographic unit that would provide interpretive power and be spatially explicit across the forests and woodlands of the State, and be independent of ownership or administrative boundaries. The most useful geographic unit was potential vegetation. This classification system integrates a variety of physical and biological components including climate, soil, geology and vegetation. Potential vegetation types are identified by species indicative of similar conditions. For example, pinyon/juniper indicates a warmer and drier environment than ponderosa pine. Due to growth, mortality, and disturbance, many other kinds of vegetation can occur on this type through time (fig. 2). In some cases the indicator species may not be present, due to disturbance. Pinyon/juniper is simply a vegetative indicator, and a name, for a physical and biological environment stratification system useful for predicting response to disturbance (Hann and others 1997). For this assessment of goshawk habitat, we defined eight potential vegetation types. These potential vegetation types cover the most representative and important environments in Utah, from the warm and dry pinyon/juniper type to moist and cool subalpine fir type. We will use potential vegetation type to describe components such as different seral stages, the range of cover types, successional pathways, disease and insect relations, and fire regimes on a given geographic unit. Seral stages are different vegetative communities that occur through time and in response to different disturbances. These stages should not be confused with vegetative structural stages (young, mid-aged, mature, etc.) as defined by Reynolds and others (1992).
Potential Vegetation Type Descriptions
White Fir -- Forest lands with white fir as the potential dominant vegetation species are probably too dry and warm for subalpine fir (Abies lasiocarpa) or Engelmann spruce (Picea engelmannii) to be dominant. This potential vegetation type occupies about 7 percent of the forested land in the State in montane sites between 5,000 and 9,200 ft, depending on the location within the State (table 5). This potential vegetation type usually occurs on gravelly soils that are well drained and derived from a variety of parent materials. In the north, this type is often associated with colluvium and in the south it is often associated with limestone and sandstone. Including white fir, a variety of other species can also occur in this type such as juniper, pinyon pine (Pinus edulis), quaking aspen, ponderosa pine, Douglas-fir, limber pine (Pinus flexilis), bristle cone pine (Pinus longaeva), blue spruce (Picea pungens), Engelmann spruce, subalpine fir, and Gambel oak. Historically, fire intervals were indeterminate for the type (Bradley and others 1992), but likely ranged between 8 and 18 years. In addition, large volumes of ponderosa pine and Douglas-fir were removed from this type in the early 1800's. Root diseases such as Fomes annosus and insects such as spruce budworm (Choristoneura occidentalis) are only two of the insects and pathogens common in this type, especially in the tension zones.
Table 5 -- Proportion of Utah forested land in each potential vegetation type.
Subalpine Fir -- At the higher elevations throughout the State, the forest landscape is dominated by the subalpine fir potential vegetation type. It covers approximately 17 percent of forest land in Utah and it occurs between 6,000 and 11,000 ft, depending upon the location (table 5). Pure stands of subalpine fir are rarely found, but mixed conifer stands with Douglas-fir and Engelmann spruce are much more common. Quaking aspen, along with lodgepole pine, are common seral species with various amounts of ponderosa pine, limber pine, blue spruce, and Gambel oak. Because this type can occur on all parent materials found in the State, soils range from coarse to fine. The climate can be characterized as cool, with frequent summer frosts and deep snow packs (Lawton 1979). Fire intervals have been estimated as low as 40 years in some areas, to over 300 years in others (Bradley and others 1992). In combination with harvesting, fire has created many stands dominated by quaking aspen (Mauk and Henderson 1984). Spruce beetle, Armilaria root disease, and balsam bark beetle (Dendroctonus confusus) are common disturbance agents as are snow, ice and wind. All of these agents, singly and in concert, can create a variety of different successional pathways producing a variety of stand structures and compositions.
Lodgepole Pine -- This is the only potential vegetation type where lodgepole pine persists with no evidence that another species is the potential climax (Pfister and others 1977). This type occurs between 7,600 and 10,300 ft in the Uinta Mountains and represents about 1 percent of the forest lands in the State (table 5). Exposures are relatively warm and usually quite arid with well drained soils (Mauk and Henderson 1984). Soils are primarily derived from quartzite with a wide range of depths. Lodgepole pine in this type has both endemic and epidemic levels of dwarf mistletoe, depending on the location. Also Comandra blister rust (Cronartium commandrae), and western gall rust (Endocronartium harknessii) occasionally kill trees. Root and stem decays can kill older trees and are often more damaging than rusts (Krebill 1975). Within this type, there are small amounts of quaking aspen, Engelmann spruce, subalpine fir, Douglas-fir, white fir, and common juniper. For the most part, fires are either the smoldering type where the surface fuels are slowly burned at 22 year intervals (Arno 1976), or severe stand replacing events occurring at intervals as great as 300 years (Romme 1982). Stand development is a combination of insect and disease mortality, fuel accumulation, and fire (Brown 1975). Dwarf mistletoe and bark beetles (Dendroctonus ponderosae) can work singly or in concert to kill trees. Forests thinned by surface fires are susceptible to mistletoe because well spaced trees allowing easier dispersment of mistletoe seeds on residual trees and new seedlings (Parmeter 1978). Mountain pine beetles have caused extensive mortality in the type (Hutchinson and others 1965).
Engelmann Spruce -- The Engelmann spruce potential vegetation type occurs at elevations from 9,000 to over 11,000 ft and occupies around 1 percent of the forest and woodlands of Utah (table 5). In addition to Engelmann spruce, which often lives 300 years, Douglas-fir, blue spruce, lodgepole pine and subalpine fir are found in the type. At lower elevations, lodgepole pine and quaking aspen are major seral species (Mauk and Henderson 1984). Soils of this type are usually gravelly and derived from quartzite in the northern Utah, or weathered andesitic flows in the southern Utah (Youngblood and Mauk 1985). Extensive winds on these sites diminish the snow pack and tip over trees. Fire is frequent at low elevations, but its effect may be severe at higher elevations where infrequent stand replacing events occur.
Ponderosa Pine -- In northern Utah, the ponderosa pine potential vegetation type is mostly limited to the eastern and southern Uinta Mountains, but it is found throughout the southern part of the State. This type occurs at elevations ranging from 6,800 to 9,000 ft. It represents about 5 percent of the forests and woodlands of Utah (table 5). Important seral trees include quaking aspen and Gambel oak. Other species include pinyon pine, limber pine, Rocky Mountain juniper (Juniperus scopulorum), or Utah juniper. Soils in the north are usually well drained, gravelly and shallow when over bedrock. In southern Utah, parent materials can include basalt, andesetic flows, intrusive granitoids and others (Youngblood and Mauk 1985). In general, this type, especially in southern Utah, receives considerable precipitation during the summer with total amounts near 15.6 inches. Historically, fire was a frequent disturbance in the type, often occurring every other year in some areas, but with 48 year intervals in others (Dieterich 1980a). Historically, throughout Utah the ponderosa pine potential vegetation type burned at a frequency of less than 20 years (Bradley and others 1992). The intensities of these fires were low, but they thinned and cleaned stands of regeneration and surface fuels. Without fire, this type is prone to both live and dead fuel accumulations, increasing the potential for stand replacing fires. Historically, lightning, beetles and diseases constantly killed large trees which were burned, creating a site well suited for regeneration. Domestic livestock grazing, along with fire exclusion, has disrupted fire cycles and created conditions that were rare or non-existent in primeval forests (Bradley and others 1992). Along with these structural changes, Armillaria, bark beetles, dwarf mistletoe, and other pathogens and insects attack, stress or kill trees.
Douglas-fir -- The Douglas-fir potential vegetation type covers a small portion of the south, while in the north it is well represented. This type occurs from 5,000 to 8,800 ft in the north and up to 9,700 ft in the south (Mauk and Henderson 1984, Youngblood and Mauk 1985) and covers approximately 9 percent of the forest and woodlands of Utah (table 5). This type occurs on a variety of soils and parent materials but in the Uintas it is restricted to calcareous substrates and areas where the soils are at least weakly calcareous. Soils are well drained and all textures are possible for the surface, but most are loamy or finer. Forests on this type range from scattered to dense depending on exposure. Douglas-fir is the most common conifer but ponderosa pine, quaking aspen, and lodgepole pine are frequent seral species. Also, on the edges of the type, white fir, juniper, limber pine and Engelmann spruce occur occasionally. Root diseases, spruce budworm, and Douglas-fir beetle are endemic in the type, with occasional epidemic eruptions on local levels. Historically, in drier portions of the type, stands were open allowing frequent (4 to 7 years) low intensity fires to promote the establishment and dominance of seral ponderosa pine. Higher severity and stand replacing fires are possible when Douglas-fir ladder fuels allow fires to reach the overstory crowns, especially at the longer return intervals (50 years plus). In the moist areas of the type, a more variable fire regime historically occurred, with both surface and stand-replacing fires common. As a result, a mosaic of stand structures and compositions existed. Fire return intervals were probably in the range of 15 to 30 years (Bradley and others 1992). This type was harvested heavily after European settlement, with extensive removal of both large Douglas-fir and ponderosa pine.
Quaking Aspen -- Quaking aspen similar to lodgepole pine is a seral species growing on other potential vegetation types. Only on the quaking aspen potential vegetation type does this tree appear to be long persistent or climax. Quaking aspen vigorously regenerates by root suckers following fire (Mueggler 1988) and it is able to dominate a site rapidly after a disturbance. The environmental conditions determining quaking aspen's role as a seral or climax species remain ill-defined (Mueggler 1989). An occasional subalpine fir, Douglas-fir, lodgepole pine, or Engelmann spruce might occur along with quaking aspen even when the latter is persistant. Quaking aspen occurs at elevations ranging from 5,500 to 10,500 ft on a variety of soils that are derived from sandstone, limestone, quartzite and granitics. It covers about 10 percent of the State (table 5). In southern Utah, it occurs primarily on volcanics. Historically, livestock grazing and fire were the primary disturbances. Even-aged stands of quaking aspen usually originate from fire and may occur on other potential vegetation types. In contrast, climax stands of this tree tend to be uneven-aged where regeneration is a gradual but continual process. Quaking aspen is sensitive to fire because of its thin bark, and fire scarred trees usually contain heart rot (Jones and Debyle 1985). Fire appears to stimulate suckering by killing most or all of the clonal stems (Brown and Debyle 1987). Quaking aspen is the dominant tree species in this type with most successional changes occurring in the forb, shrub, and grass layers, as they respond to different disturbances.
Pinyon/juniper -- The pinyon/juniper potential vegetation type occupies approximately 50 percent of the forest and woodlands in Utah (table 5). By far it is the driest potential vegetation type growing at elevations from 4,500 to 7,500 ft. It occurs on a variety of soils that are derived from granites, limestones, volcanics, and mixed alluvium (Evans 1988). Lower limit is set by available water and upper limit by unfavorable temperatures. Douglas-fir and ponderosa pine potential vegetation types border this type at high elevations and grass/shrub dominated communities border it at lower elevations. Pinyon pine usually dominate higher elevations while juniper occupy lower elevations. Fires open stands and create a mosaic of structures and compositions (Bradley and others 1992). Fire return intervals range from 8 and 50 years (Burkhardt and Tisdale 1976; Moir 1982). Domestic livestock grazing occurred on some sites in the Southwest for as long as 400 years (Tausch and others 1981). Succession after fire begins with annuals and continues with mixes of perennial grasses, forbs, and shrubs, culminating with pinyon/juniper (Evans 1988) (fig. 1). In mature stands, most trees are pinyon pine, and low-severity fires will remove understory. Because of the scarcity of fuels to carry a fire in closed stands, fires become rare (Bradley and others 1992) but when fires invade from adjacent types, especially if driven by wind, stand-replacing fires can occur.
Delineation of Potential Vegetation Types
No map of potential vegetation types for Utah was available. Using Geospatial Analysis Processes data to identify patches of vegetation, 1,112 vegetative polygons of forests and woodlands of Utah were identified (USGS 1995). To each of these polygons, a potential vegetation type was assigned using inventory plot data points located near or in the polygon, and by local knowledge supplied by resource managers familiar with the area. Inventory points were random plots sampled by the Forest Inventory and Analysis (FIA) group of the USDA Forest Service, Rocky Mountain Research Station. In addition to the potential vegetation type, data for each plot included slope, aspect, elevation, current vegetation, and other site-specific information. This auxiliary information helped identify the potential vegetation type within each polygon. Often, more than one point would be located in a vegetation polygon, further strengthening the potential vegetation type assignment. All polygons in Utah were assigned to one of eight potential vegetation types (map 1).
Current Habitat
Current vegetation was determined using a combination of potential vegetation type, Geospatial Analysis Processes, and FIA data, along with expert knowledge of resource managers working in Utah. Current forest cover type, structural stage (seedling, sapling, young forest, middle-aged forest, mature forest or old growth), stand size, and understory composition were identified for each potential vegetation type polygon. Coarse woody debris, snags, water, large trees, and multiple canopies within in each polygon were also determined. These variables described attributes believed to be important for one or more of the primary prey species (table 4). When necessary to describe current conditions accurately, some potential vegetation type polygons were divided into two or more subpolygons.
Potential vegetation types in Utah currently support various forest cover types ranging from pure ponderosa pine to complex mixtures of Douglas-fir, ponderosa pine, quaking aspen, and lodgepole pine (table 6). The pinyon/juniper cover type is most common, covering 51 percent of Utah's forests and woodlands (table 7). Spruce and fir cover types are common over the State's forested lands. However the ponderosa pine cover type represents only 4 percent of the forest and woodlands. The trend across the entire State is for late seral species to be better represented than early seral species (map 2).Table 6 -- Possible forest cover types for each of the potential vegetation types.
Table 7 -- The proportion of the current cover types found in Utah forests and woodlands.
Quaking aspen occupies 9 percent of Utah's forests and woodlands (table 7). It dominates 84 percent of the quaking aspen potential vegetation type (table 8) but is underrepresented in many potential vegetation types where it is a major seral species. For example, in the Douglas-fir, white fir and subalpine fir potential vegetation types it represents less than 6 percent (table 8). Most of these stands are located in central and north-central Utah, and in the mountains of southeastern Utah (map 2). Quaking aspen also occurs as a component of mixed species stands on 51 percent of the lodgepole potential vegetation type and 61 percent of the Douglas-fir potential vegetation type (table 8). Goshawk nests are often associated with mixed lodgepole pine and quaking aspen forests in northeastern Utah (table 1). Mixtures of Douglas-fir and quaking aspen are common in the Tavaputs Plateau of east-central Utah and in the Bear River Range east of Logan, UT. Few goshawk nests have been located in these forests and the importance as habitat is unclear (map 2).
Table 8 -- Proportion of each potential vegetation type currently in various forest cover types.
The cover types most often occupied by goshawks (based on sightings and nest locations) are Engelmann spruce, subalpine fir, lodgepole pine and quaking aspen, either in single or mixed species forests (table 1). Ponderosa pine can also be locally important, particularly in riparian areas where the species is mixed with quaking aspen. Subalpine fir and Engelmann spruce are late seral species dominating their respective potential vegetation types. Lodgepole pine or a lodgepole pine mix occurs on over 75 percent of the lodgepole pine potential vegetation type but occupy less than 20 percent of the subalpine fir and Engelmann spruce potential vegetation types. Subalpine fir, Engelmann spruce, quaking aspen, and lodgepole pine (either alone or in mixed species stands) dominate the tall forests in Utah and are all commonly used by goshawks for nesting.
Other cover types used by goshawks include ponderosa pine, white fir and Douglas-fir. Ponderosa pine, an important early seral species in many of the potential vegetation types, is currently under-represented in the white fir and Douglas-fir potential vegetation types (table 8). As a late seral species, it covers 84 percent of the ponderosa pine potential vegetation type but is rarely found on any other. White fir dominates the white fir potential vegetation type, even though ponderosa pine, quaking aspen, Douglas-fir, and lodgepole pine are major seral species that can exist on this potential vegetation type (Mauk and Henderson 1984). In all potential vegetation types, late seral species have greatest representation. With the exception of lodgepole pine, ponderosa pine, and quaking aspen potential vegetation types, these conditions are unstable (Harvey and others in press) (see section on "Trends and Risks to Habitat").
Habitat Valuation Process
Forest cover type alone does not determine habitat suitability for nesting goshawks or goshawk prey. Stand structure, patch size, landscape features, woody debris, snags, understory vegetation, openings, and interlocking crowns are some of the habitat attributes that are important for the goshawk and its prey (tables 1, 2, 4). Abundance of important prey species varies with potential vegetation type (tables 3, 9). Each vegetative polygon was evaluated by resource managers, as to its value for goshawk foraging and nesting based on the presence or absence of these prey and habitat characteristics. Each polygon was rated as high, medium, or low quality in four categories: goshawk nesting habitat, small to medium-sized mammal habitat, woodpecker habitat, and other medium-sized bird habitat.
Table 9 -- Occurrence of selected prey species in Utah potential vegetation types.
Nesting Habitat
Most high quality nesting habitat across all potential vegetation types occurs on the subalpine fir and quaking aspen potential vegetation types (table 10; fig. 3) . As would be expected, very little high quality nesting habitat occurs on the pinyon/juniper potential vegetation type. Within individual potential vegetation types, the subalpine fir, lodgepole pine, ponderosa pine and quaking aspen are primarily composed of high value nesting habitat (table 11). Although nesting habitat is found in most potential vegetation types, forests occurring on the subalpine fir and quaking aspen potential vegetation types appear to be particularly suited for nesting. As would be expected, the pinyon/juniper potential vegetation type was rated as low value for nesting. No nests are known to occur in pinyon/juniper habitats in Utah. The central mountains of the State have the greatest concentration of high rated nesting habitat interspersed with medium rated forest lands and bordered by the low rated pinyon/juniper woodlands (map 3).
Table 10 -- Proportion of high, medium, or low nesting habitat among the potential vegetation types.
Table 11 -- Proportion of each potential vegetation type in high, medium, or low nesting habitat.
Map 3 -- Distribution of high, medium, or low value goshawk nesting habitat in Utah.
Foraging Habitat
Forests occurring on ponderosa pine, subalpine fir, and quaking aspen potential vegetation types provide most of the high valued habitat for mammals (table 12). With the exception of the pinyon/juniper potential vegetation type, most of the forests and woodlands of Utah were rated high or medium for mammal habitat. For both woodpeckers and other birds, the quaking aspen potential vegetation type supports mostly high rated habitat with only small amounts of low rated habitat for any of the prey groups. The Engelmann spruce potential vegetation type was rated medium for woodpeckers and low for other birds. Fifty percent of the forests on the ponderosa pine potential vegetation type were rated high for woodpeckers and 60 percent had a medium rating for other birds. Forests growing on the white fir and Douglas-fir potential vegetation types rated mostly medium as habitat for each prey group.
High value mammal habitat is well distributed throughout the forests of the State especially through the central mountains (map 4). This habitat quality is likely related to the amount of cone producing trees available and the amount of woody debris that occurs in these forests. Forests and woodlands with a medium rating for mammal habitat are well distributed across the State. Minor amounts of low value habitat are located in east-central and southwestern Utah. Similar to mammal habitat, high value woodpecker habitat is also located in the central mountains (map 5). Woodpecker habitat is usually good when snag densities are high, when both endemic and epidemic occurrences of insects and diseases are high. In contrast to mammal habitat, there is much more low rated woodpecker habitat in pinyon/juniper woodlands. The largest concentration of medium valued woodpecker habitat was located in northeastern Utah. High value habitat for medium-sized birds (other than woodpeckers) is distributed north to south across the State (map 6). High value habitat is interspersed with both low and medium valued habitat throughout the State. A large block of medium habitat for other birds is located in east-central Utah. Forest grouse inhabit primarily Douglas-fir and quaking aspen forests throughout the State. Both quaking aspen and ponderosa pine forests contain high numbers of songbirds. Engelmann spruce, white fir, Douglas-fir, and lodgepole pine forests generally have lower numbers of songbirds except when quaking aspen is present.
Map 4 -- Distribution of high, medium, or low value small mammal habitat in Utah.
Map 5 -- Distribution of high, medium, or low value woodpecker habitat in Utah.
Combined Habitat Rankings
Nesting and foraging habitat ratings were used to produce a combined goshawk rating for each of the 1,112 vegetative polygons in the State. Optimal habitat represents areas in the State in which mammal, woodpecker, other bird, and nesting habitat were all rated high. Optimal habitat would be expected to consistently support breeding goshawks. If nests occurred in these areas, it is expected that they would likely fledge young, even when annual fluctuations in weather or prey reduce nest success in lower quality habitats.
Goshawks are opportunists and able to adapt their diet to take advantage of whatever prey species is abundant in a given habitat and year (Squires and Reynolds 1997). Therefore, to reflect this adaptability we rated areas as "high value" if they were rated high for nesting and also high for at least one prey group. This combined habitat rating includes those areas rated as "optimal" and areas where populations of one or two of the prey groups are expected to be abundant. While it seems intuitive that a variety of abundant prey would be optimal, variety may be less important than abundance. Therefore, habitats in which prey are abundant but not necessarily diverse can still support high densities of goshawks. Average nest productivity may vary more from year to year in high value habitats than in optimum habitats, since it includes some areas where the prey base is limited to one species group. However, long-term averages in goshawk densities and nest success in optimum and high value habitats may differ minimally.
Forty percent of the high value habitat and 25 percent of the optimum habitat is located on the subalpine fir potential vegetation type (table 13). The quaking aspen potential vegetation type also contains high proportions of high value and optimum habitat. In contrast, forests growing on the Engelmann spruce, white fir, pinyon/juniper, and lodgepole pine potential vegetation types all had minimal amounts of high and optimum habitat. The majority of the high value habitat is located in the central portion of the State (map 7).
Within the subalpine fir potential vegetation type, 54 percent is rated as high value habitat and 16 percent rated optimum (table 14). Similarly, large proportions of the quaking aspen and lodgepole potential vegetation types were rated as high or optimum habitat. The majority of optimum habitat occurs in the center portion of the State (map 8). These data show how important the quaking aspen and subalpine fir potential vegetation types are in providing quality goshawk habitat in Utah. Moreover, larger amounts of the subalpine fir potential vegetation type could be growing quaking aspen.
Habitat Connectivity
Habitat is only connected and available to goshawks if it is accessible from existing population centers. That is, if the spatial distribution of habitat allows each patch to be reached by individuals dispersing from one or more adjacent patches. If every patch can be reached and subsequently occupied, then all areas would be considered connected. Connectivity has positive implications for population viability because it allows juveniles to disperse from natal areas, and allows individuals to emigrate to new areas when their current habitat declines in value. Declines in habitat value could be caused by stand-replacing fires, timber harvesting, periodic lows in prey abundance, urban encroachment, or other disturbances. Connected habitat ensures that individuals will be available to re-colonize habitats or emigrate to new breeding territories throughout the State.
Distances involved in goshawk dispersal and habitat selection need to be determined prior to evaluating connectivity (Keitt and others 1997). Perhaps the best indication of connectivity is the distance goshawks move from natal areas to adult breeding territories. These distances can be determined by color-banding nestlings or fledglings and then relocating them as nesting adults. Several reports indicate that the dispersal distances of young goshawks range from 6 to 20 miles (Reynolds and Joy 1998; Woodbridge and Detrich 1994). Often, several years elapse between fledging and relocation of adult birds, so the distances may be the cumulative result of successive movements. In addition, one banded female who fledged on the north slope of the Uinta Mountains was found nesting 17 miles from her natal site (Ashley National Forest 1998b). Dispersal can also be estimated by observing adults who breed on two or more spatially distinct territories. Of the 19 reports of territory switching in the contiguous Western United States, the distance moved averaged less than 6 miles for both sexes (Reynolds and Joy 1998; Woodbridge and Detrich 1994; Younk and Bechard 1994b). In Alaska, female goshawks relocated to territories 27 miles apart (Iverson and others 1996).
Winter movements of goshawks are considerably longer. During the winter, female goshawks in the Uinta Mountains usually move about 60 miles from the last known nest site, but one female moved approximately 180 miles (Squires 1997). Adults in southern Wyoming were observed to winter in Colorado 116 miles from the nest sites (Squires and Ruggiero 1995). In Alaska, after the breeding season, males have been known to move 59 miles, females 34 miles, and juveniles up to 101 miles, with an average maximum distance moved by juveniles of 39 miles (Iverson and others 1996). These studies probably underestimate winter movements because not all birds were relocated and some may have traveled beyond the search radius.
Although goshawks are clearly capable of traveling long distances to find suitable habitat, the previous information indicates that 20 to 60 mile movements are typical. Therefore, a maximum distance of 60 miles between patches of high value habitat would represent a reasonable method of defining connectivity. We also restricted our definition to high value habitat patches, even though goshawks may use patches of low or medium quality as stepping stones when dispersing from one high value patch to another. This definition of habitat connectivity would conservatively ensure that goshawks will be able to disperse freely throughout the State, always finding at least high value habitat.
High value habitat in the State is well connected except for two portions in the southwestern corner of the State (map 9). Habitat in the LaSal and Abajo Mountains were connected to each other. However, they were separated by 66 miles from other high valued habitat. Our analysis did not consider potential habitat in western Colorado, northern Arizona, or northwestern New Mexico. It is likely these areas could be well connected to these adjacent regions.
Title: Status
and Distribution of Habitat in Utah:
RMRS-GTR-22 - The Northern Goshawk in Utah: Habitat Assessment
and Management Recommendations
Electronic Publish Date: May 26, 1999
Expires: Indefinite
Last Update: January
15, 2002