Endangered Species Recovery Program
Taxonomy.-- The kit fox, Vulpes macrotis, was described by C. Hart Merriam (1888). The area of the type locality, near Riverside in Southern California, is now highly urbanized. Eight subspecies were recognized historically (e.g., Hall 1981). V. m. mutica, the San Joaquin kit fox, wa first described by Merriam (1902). Today, only V. m. macrotis and V. m. mutica are recognized (Mercure et al. 1993). The type locality is near Tracy, San Joaquin County, California.
Several different taxonomies for the species and subspecies of small, North American foxes have been proposed over the last 110 years (historical literature summarized by Hall 1946, Hall and Kelson 1959, Rohwer and Kilgore 1973, Waithman and Roest 1977, Hall 1981). Two recent studies examined the evolutionary and taxonomic relationships among small, North American foxes (Dragoo et al. 1990, Mercure et al. 1993). Dragoo et al. (1990) concluded that all North American arid-land foxes belonged to the species V. velox (swift fox). The subspecific statuses of the taxa historically regarded as subspecies of V. macrotis also were challenged by Dragoo et al. (1990), who recommended that all be synonymized under V. velox macrotis. Genetic work by Mercure et al. (1993) led them to conclude that, though there was evidence of hybridization between kit and swift foxes over a limited geographic area, they should be considered separate species. Further, Mercure et al. concluded that of the traditional subspecies of the kit fox, the San Joaquin Valley population is the most distinct and should be considered a subspecies (1993, p. 1323). Their data recognize the swift fox as a separate monotypic species, and two subspecies of kit foxes: V. macrotis macrotis, found throughout the remaining habitat within the historical range of the species, except the San Joaquin kit fox range; and V. macrotis mutica, the San Joaquin kit fox.
Description.-- The kit fox is the smallest canid species in North America and the San Joaquin kit fox is the largest subspecies in skeletal measurements, body size, and weight. Grinnell et al. (1937) found a difference in body size between males and females: males averaged 80.5 centimeters (31.7 inches) in total length, and 29.5 centimeters (11.6 inches) in tail length; females averaged 76.9 centimeters (30.3 inches) in total length, and 28.4 centimeters (11.2 inches) in tail length. Kit foxes have long slender legs and are about 30 centimeters (12 inches) high at the shoulder. The average weight of adult males is 2.3 kilograms (5 pounds), and of adult females is 2.1 kilograms (4.6 pounds) (Morrell 1972).
General physical characteristics of kit foxes include a small, slim body, relatively large ears set close together, narrow nose, and a long, bushy tail tapering slightly toward the tip (Figure 50). The tail is typically carried low and straight.
Color and texture of the fur coat of kit foxes varies geographically and seasonally. The most commonly described colorations are buff, tan, grizzled, or yellowish-gray dorsal coats (McGrew 1979). The guard hairs on the back are black tipped, which accounts for the grizzled appearance (Bell 1994). Two distinctive coats develop each year: a tan summer coat and a silver-gray winter coat (Morrell 1972). The undersides vary from light buff to white (Grinnell et al. 1937), with the shoulders, lower sides, flanks and chest varying from buff to a rust color. The ear pinna (external ear flap) is dark on the back side, with a thick border of white hairs on the forward-inner edge and inner base. The tail is distinctly black-tipped.
Identification.-- The foot pads of kit foxes are small by comparison with other canids. A sample of 21 tracks from throughout the San Joaquin Valley had an average length of 3.1 centimeters (1.2 inches) and an average width of 2.6 centimeters (1 inch) (Orloff et al. 1993). Other characteristics such as the degree to which the feet are furred and the size, shape, and configuration of the pads distinguish kit fox tracks from those of co-occurring canids and domestic cats (Orloff et al. 1993).
Because all three fox species that occur in the San Joaquin Valley are primarily nocturnal, identification of free-living, and often fast-moving, animals can be a challenge. The black-tipped tail and coat color differences usually distinguish kit foxes from red foxes (V. vulpes). At 4 to 5 kilograms (8 to 11 pounds), the red fox also is much heaver than the kit fox. Gray foxes (Urocyon cinereoargenteus) however are sometimes misidentified as kit foxes, especially in winter when the kit fox coat is thicker and has more gray. Both species have a black tail tip but gray foxes also have a distinctive black stripe running along the top of the tail. Gray foxes are more robust than kit foxes; they are heavier with an average body weight of about 3.6 kilograms (8 pounds) (Grinnell et al. 1937). However, San Joaquin kit foxes have longer ears, averaging 8.6 centimeters (3.4 inches) compared with 7.8 centimeters (3 inches) for gray foxes (Grinnell et al. 1937).
Figure 50. Illustration of a kit fox by Jodi Sears (© D.F. Williams)
Historical Distribution.-- The historical range was first defined by Grinnell et al. (1937). Prior to 1930, kit foxes inhabited most of the San Joaquin Valley from southern Kern County north to Tracy, San Joaquin County, on the west side, and near La Grange, Stanislaus County, on the east side. These authors believed that by 1930 the kit fox range had been reduced by more than half, with the largest portion of the range remaining in the southern and western parts of the Valley (Figure 51), though they provided no indication for why they believed foxes had been eliminated from most of the east side and Valley floor.
Current Distribution.-- Although the San Joaquin kit fox has been listed as endangered for over 30 years, there has never been a comprehensive survey of its entire historical range. And, despite the loss of habitat and apparent decline in numbers since the early 1970s, there has been no new survey of habitat that was then thought to be occupied (Morrell 1975).
Despite the lack of a comprehensive survey, local surveys, research projects and incidental sightings indicate that kit foxes currently inhabit some areas of suitable habitat on the San Joaquin Valley floor and in the surrounding foothills of the coastal ranges, Sierra Nevada, and Tehachapi Mountains, from southern Kern County north to Contra Costa, Alameda, and San Joaquin Counties on the west, and near La Grange, Stanislaus County on the east side of the Valley (Williams in litt. 1990), and some of the larger scattered islands of natural land on the Valley floor in Kern, Tulare, Kings, Fresno, Madera, and Merced Counties (Figure 51). Kit foxes also occur westward into the interior coastal ranges in Monterey, San Benito, and Santa Clara Counties (Pajaro River watershed), in the Salinas River watershed, Monterey and San Luis Obispo Counties, and in the upper Cuyama River watershed in northern Ventura and Santa Barbara Counties and southeastern San Luis Obispo County. Kit foxes are also known to live within the city limits of the city of Bakersfield in Kern County (Laughrin 1970, Jensen 1972, Morrell 1975, USFWS 1983, Swick 1973, Waithman 1974a, Endangered Species Recovery Program unpubl. data).
Some researchers have suggested that as San Joaquin Valley natural lands were cultivated or otherwise developed, displaced kit foxes colonized nearby valleys and foothills (Laughrin 1970, Jensen 1972); however, there is no concrete evidence to support this assertion. As early as 1925, Grinnell et al. reported kit fox specimens from the Panoche Creek area in the foothills of western Fresno County, and east of Rose Station (Fort Tejon) in southern Kern County at an elevation of 363 meters (1,200 feet) (Grinnell et al. 1937, USFWS 1983). Therefore, it is more probable that kit foxes have always occurred in these areas, possibly at low density.
The largest extant populations of kit foxes are in western Kern County on and around the Elk Hills and Buena Vista Valley, Kern County, and in the Carrizo Plain Natural Area, San Luis Obispo County. The kit fox populations of Elk Hills and the City of Bakersfield, Kern County (B.L. Cypher pers. comm.), Carrizo Plain Natural Area, San Luis Obispo County (White and Ralls 1993, Ralls and White 1995), Ciervo-Panoche Natural Area, Fresno and San Benito Counties (Endangered Species Recovery Program), Fort Hunter Liggett, Monterey County (V. Getz pers. comm.), and Camp Roberts, Monterey and San Luis Obispo Counties (W Berry pers. comm.) have been recently, or are currently, the focus of various research projects. Though monitoring has not been continuous in the central and northern portions of the range, populations were recorded in the late 1980s at San Luis Reservoir, Merced County (Briden et al. 1987), North Grasslands and Kesterson National Wildlife Refuge area on the Valley floor, Merced County (Paveglio and Clifton 1988), and in the Los Vaqueros watershed, Contra Costa County in the early 1990s (V. Getz pers. comm.). Smaller populations and isolated sightings of kit foxes are also known from other parts of the San Joaquin Valley floor, including Madera County and eastern Stanislaus County (Williams 1990).
Food and Foraging.-- Diet of kit foxes varies geographically, seasonally, and annually, based on variation in abundance of potential prey. In the southern portion of their range, kangaroo rats, pocket mice, white-footed mice (Peromyscus spp.), and other nocturnal rodents comprise about one-third or more of their diets. Kit foxes there also prey on California ground squirrels, black-tailed hares, San Joaquin antelope squirrels, desert cottontails, ground-nesting birds, and insects (Scrivner et al. 1987a). Vegetation and insects occur frequently in feces. Grass is the most commonly ingested plant material (Morrell 1971, C.A. Vanderbilt-White pers. comm.). In the central portion of their geographic range, defined here as Kings, Tulare, Fresno, Madera, San Benito, Merced, Stanislaus, and Monterey Counties, known prey species include white-footed mice, insects, California ground squirrels, kangaroo rats, San Joaquin antelope squirrels, black-tailed hares, and chukar (Alectoris chukar) (Jensen 1972, Archon 1992), listed in approximate proportion of occurrence in fecal samples. In the northern part of their range, defined here as San Joaquin, Alameda and Contra Costa Counties, kit foxes most frequently consume California ground squirrels (Orloff et al. 1986). Cottontails, black-tailed hares, pocket mice, and kangaroo rats also are eaten (Hall 1983, D.F. Williams unpubl. data). Though ground squirrels are diurnal and kit foxes are predominantly nocturnal, kit foxes are commonly seen during the day during late spring and early summer (Orloff et al. 1986).
Reproduction and Demography.-- Kit foxes can breed when 1 year old, but may not breed their first year of adulthood (Morrell 1972). Adult pairs remain together all year, sharing the home range but not necessarily the same den (K. Ralls pers. comm.). During September and October, adult females begin to clean and enlarge natal or pupping dens (they select dens with multiple openings; Morrell 1972). Mating and conception take place between late December and March (Egoscue 1956, Morrell 1972, Zoellick et al. 1987a, Spiegel et al. in press). The median gestation period is estimated to range from 48 to 52 days (Spiegel et al. in press). Litters of from two to six pups are born sometime between February and late March (Egoscue 1962, Morrell 1972, Zoellick et al. 1987a, Spiegel et al. in press).
The female is rarely seen hunting during the time she is lactating. During this period the male provides most of the food for her and the pups. The pups emerge above ground at slightly more than 1 month of age. After 4 to 5 months, usually in August or September, the family bonds begin to dissolve and the young begin dispersing. Occasionally a juvenile female will remain with the adult female for several more months (ONeal et al. 1992, Spiegel et al. in press). Offspring of both sexes sometimes remain with their parents through the following year and help raise a subsequent litter (White and Ralls 1993, Spiegel et al. in press, B.L. Cypher pers. comm.).
Reproductive success of kit foxes is correlated with abundance of their prey (Egoscue 1975). Success decreases when the density of prey species drops because of drought, too much rainfall, or other circumstances (White and Ralls 1993, Spiegel et al. in press, B.L. Cypher pers. comm., White and Garrott 1998).
During a 6-yer study at the Elk Hills Naval Petroleum Reserves in California, pups dispersed an average of 8 1.4 kilometers (5.0 0.9 mile; Scrivner et al. 1987b). Maximum reported distances can vary considerably (Hall 1983). One individual traveled a minimum of 40 kilometers (25 miles) from its whelping den (V. Getz pers. comm.), and a prime adult male dispersed from Camp Roberts to the Carrizo Plain in 1989 (P.J. White pers. comm.). Adult and juvenile kit foxes radio-collared at the Elk Hills Naval Petroleum Reserves in California dispersed through disturbed habitats, including agricultural fields, oil fields, rangelands, and across highways and aqueducts. One pup crossed the Temblor Range into the Carrizo Plain (Scrivner et al. 1987b).
The average age of kit foxes in a Utah population was about 2 years (Egoscue 1975). One fox in another Utah study was estimated to be at least 7 years old (Egoscue 1962). Kit foxes at Camp Roberts are reported to be over 8 years old (P.J. White pers. comm.). Kit foxes on Naval Petroleum Reserve-1 in California are known to live as long as 8 years but such longevity is rare; animals less than 1 year old outnumber older foxes by 2.8:1 (Berry et al. 1987a). Annual survival rates of juvenile foxes have ranged from 0.26 on Naval Petroleum Reserve-1 in California (Berry et al. 1987a) to 0.21 to 0.41 on the Carrizo Plain (Ralls and White 1995). In captivity, kit foxes have lived up to 10 years (McGrew 1979, M. Johnson pers. comm.).
An annual adult mortality rate of approximately 50 percent has been reported (Morrell 1972, Egoscue 1975, Berry et al. 1987a, Ralls and White 1995, Standley et al. 1992). The annual mortality rate for juvenile kit foxes may be closer to 70 percent (Berry et al. 1987a). Predation by larger carnivores (e.g., coyotes) accounts for the majority of San Joaquin kit fox mortality. The effects of disease, parasites and accidental death are largely unknown, but were thought to account for only a small portion of mortality (Berry et al. 1987a). Drought plays a role in low reproductive success (i.e., pups are born but do not survive to weaning). Adults can maintain weight and body condition and females can give birth, but pairs apparently cannot catch enough prey to support pups (White and Ralls 1993, Spiegel et al. in press).
San Joaquin kit fox densities on the west side of the San Joaquin Valley were estimated to be 0.4 per square kilometer (1.04 per square mile) prior to 1925, based on fur trapping efforts (Grinnell et al. 1937). In 1969, Laughrin (1970) estimated that range-wide kit fox densities were 0.2 to 0.4 per square kilometer (0.52 to 1.04 per square mile). Morrell (1975) estimated densities of 1.2 per square kilometer (3.11 per square mile) in optimal habitats in "good" years. In the 1983 recovery plan (USFWS 1983), Morrells data was corrected for habitat loss and an estimate of 0.5 per square kilometer (1.30 per square mile) was obtained. The estimated mean density of trappable adult kit foxes was from 0.8 to 1.1 per square kilometer (2 to 2.8 per square mile) between 1980 and 1982 on the Naval Petroleum Reserves in California (OFarrell 1984). More recently, kit fox densities at the Naval Petroleum Reserves were determined from annual live-trapping efforts (Enterprise Advisory Services, Inc., unpubl. data). On Naval Petroleum Reserve-1 in California, the mean density from 1981 to 1993 was 0.12 per square kilometer (0.31 per square mile) in winter, but varied from 0.72 per square kilometer (1.86 per square mile) in 1981 to 0.01 per square kilometer (0.03 per square mile) in 1991. On Naval Petroleum Reserve-2 in California, mean density from 1983 to 1993 was 0.38 per square kilometer (0.98 per square mile), and varied from 0.72 per square kilometer (1.86 per square mile) in summer 1983 to 0.1 per square kilometer (0.30 per square mile) in winter 1991. On the nearby Carrizo Plain Natural Area, kit fox densities were estimated to be 0.15 to 0.24 per square kilometer (0.39 to 0.62 per square mile) (White and Ralls 1993).
In the 1983 recovery plan (USFWS 1983) it was estimated that the population range-wide of adult kit foxes prior to 1930 may have been between 8,667 and 12,34 assuming an occupied range of 22,447 square kilometers (8,667 square miles) and densities of 0.4 to 0.6 per square kilometer (1.04 to 1.55 per square mile). The kit fox population in San Luis Obispo, Santa Barbara, Kings, Tulare and Kern Counties was estimated to be about 11,000 animals in the early 1970s based on limited aerial surveys of pupping dens and amount of historic habitat, but without correction for cultivated and urbanized lands (Waithman 1974b). Laughrin (1970) reported an estimated total population size of 1,000 to 3,000 foxes in 1969. Morrell (1975) conducted a more thorough investigation of kit fox abundance in 14 counties in which kit foxes were known to occur and estimated the total population at 14,832. In the 1983 recovery plan (USFWS 1983), Morrells data was adjusted and a corrected estimate of 6,961 foxes in 1975 was obtained. When compared to the pre-1930 estimate, this represents a possible population decline of 20 to 43 percent. Approximately 85 percent of the fox population in 1975 was found in only six counties (Kern, Tulare, Kings, San Luis Obispo, Fresno, and Monterey), and over half the population occurred in two of those counties: Kern (41 percent) and San Luis Obispo (10 percent) (Morrell 1975).
Behavior and Species Interactions.-- San Joaquin kit foxes use dens for temperature regulation, shelter from adverse environmental conditions, reproduction, and escape from predators. Though kit foxes are reputed to be poor diggers (Jensen 1972, Morrell 1972), the complexity and depth of their dens do not support this assessment (USFWS 1983). Kit foxes also modify and use dens constructed by other animals, such as ground squirrels, badgers, and coyotes (Jensen 1972, Morrell 1972, Hall 1983, Berry et al. 1987b), and human-made structures (culverts, abandoned pipelines, and banks in sumps or roadbeds) (Spiegel et al. in press, B.L. Cypher pers. comm.).
Den characteristics vary across the San Joaquin kit foxs geographic range. In the southernmost portion, dens with two entrances are most frequently found. Natal and pupping dens, in which pups are born and raised, tend to be larger with more entrances (2 to 18) (Morrell 1972, OFarrell and Gilbertson 1979, OFarrell et al. 1980, OFarrell and McCue 1981, Berry et al. 1987b). Entrances are usually from 20 to 25 centimeters (8 to 10 inches) in diameter and normally are higher than wide. Ramp-shaped mounds of dirt from 1 to 2 meters (3 to 6 feet) long are deposited at some den entrances (Morrell 1972). Most hillsides where kit fox dens are found (95 percent) have a slope of less than 40 degrees (Reese et al. 1992). Natal and pupping dens are found on flatter ground with slopes of about 6 degrees (OFarrell and McCue 1981, OFarrell et al. 1980). The entrances of pupping dens show more evidence of use, such as fox scat, prey remains, and matted vegetation. In the central portion of their geographic range, dens also have several openings; however, instead of a mound of dirt in front of the opening, the dirt is more often scattered into a long tailing ramp, generally with a runway down the middle. In areas of tall grass, matted grass in front of the entrance is obvious. In western Merced County, most dens are found on slopes of less than 10 degrees, but a few are found on slopes of up to 55 degrees (Archon 1992). In the northern portion of the kit fox range, dens appeared to be placed higher than most surrounding ground compared to areas farther south, perhaps reflecting the topography of the area. Dens most often are located on the lower section of the slope (Orloff et al. 1986), yet foxes are sometimes seen entering dens on the upper part of a slope (Bell 1992). Most dens lack the ramp or runway characteristic of dens in the southern and central portions of the Valley. No evidence has been found to indicate that kit foxes in this area construct their own dens (Hall 1983). Kit foxes probably enlarge California ground squirrel burrows (Orloff et al. 1986), but they also may construct their own dens.
Kit foxes often change dens and numerous dens may be used throughout the year. However, evidence that a den is in use may be absent (V. Getz pers. comm.). Reese et al (192) found that 64 percent of the dens used by radio-collared kit foxes at Camp Roberts during 1988-1991 exhibited no sign of kit foxes. Foxes change dens four or five times during the summer months, and change natal dens one or two times per month (Morrell 1972). One family of 7 kit foxes used 43 dens; the maximum number used by 1 individual was 70 (Hall 1983). Foxes on the Carrizo Plain Natural Area changed dens much more frequently than indicated by Morrells study (White and Ralls 1993). Radiotelemetry studies indicate that foxes use individual dens for a median of 2 days (mean of 3.5 days) before moving to a different den. One fox was tracked to 70 different dens during a two year study (K. Ralls pers. comm.). Den changes have been attributed to depletion of prey in the vicinity of the den or to increases in external parasites such as fleas (Egoscue 1956). Avoidance of coyotes is a more probable reason for frequently changing dens because kit foxes can easily search their home range in one night for prey, and parasites are unlikely to build to intolerable levels in 2 or 3 days (K. Ralls pers. comm.)
Nightly movements on the Elk Hills Naval Petroleum Reserves in California averaged 15.4 kilometers (9.6 miles) during the breeding season and were significantly longer than the average nightly movements of 10.2 kilometers (6.3 miles) during the pup-rearing season. Movements during the breeding season also were significantly longer than those made during the pup-dispersal season (10.4 kilometers, 6.5 miles) (Zoellick et al. 1987b).
Home ranges of from less than 2.6 square kilometers (1 square mile) up to approximately 31 square kilometers (12 square miles) have been reported by several researchers (Morrell 1972, Knapp 1978, Zoellick et al. 1987b, Spiegel and Bradbury 1992, White and Ralls 1993, Paveglio and Clifton 1988). The maintenance of large and relatively non-overlapping home ranges, as noted on the Carrizo Plain, may be an adaptation to drought-induced periods of prey scarcity that are episodic and temporary on the Carrizo Plain (White and Ralls 1993). Differences in home range size among study sites tend to be related to prey abundance (White and Ralls 1993, White and Garrott 1998).
Kit foxes are subject to predation or competitive exclusion by other species, such as the coyote, nonnative red foxes, domestic dog (Canis familiaris), bobcat (Felis rufus), and large raptors (Hall 1983, Berry et al. 1987a, OFarrell et al. 1987b, White et al. 1994, Ralls and White 1995, CDFG 1987). Coyotes are known to kill kit foxes, though an experimental coyote-control program at the Elk Hills Naval Petroleum Reserves in California did not result in an increase in survival rate for kit foxes, nor did coyote-induced mortality decrease (Cypher and Scrivner 1992, Scrivner and Harris 1986, Scrivner 1987). The extent to which gray and kit foxes compete for resources is unknown. The need for similar den sites and prey species probably place nonnative red foxes in direct competition with the much smaller kit fox. Nonnative red foxes are expanding their geographic range in central California (Orloff et al. 1986, Lewis et al. 1993), and competition with or predation on kit foxes may be a factor in the apparent decline of kit foxes in the Santa Clara Valley (T. Rado pers. comm.), and perhaps elsewhere in the northwestern segment of their range. Coyotes aggressively dominate encounters with red foxes and will pursue and kill both red and gray foxes (Sargeant and Allen 1989), as well as kit foxes. Coyotes may reduce the negative impacts of red foxes on kit foxes by limiting red fox abundance and distribution, but details of interactions between the two species and the extent to which coyotes might slow or prevent the invasion of red foxes into kit fox habitats are unknown (White et al. 1994, Ralls and White 1995).
Activity Cycle.-- San Joaquin kit foxes are primarily active at night (i.e., nocturnal), and active throughout the year (Grinnell et al. 1937, Morrell 1972). Adults and pups sometimes rest and play near the den entrance in the afternoons, but most above-ground activities begin near sunset and continue sporadically throughout the night. Morrel (1972) reported that hunting occurred only at night. Yet predation on ground squirrels, which are active during the day (i.e., diurnal), by some populations indicates that kit foxes are not strictly nocturnal, adapting to the activities of available prey (Balestreri 1981, Hall 1983, Orloff et al. 1986, OFarrell et al. 1987b, Hansen in litt. 1988).
Habitat and Community Associations.-- Kit foxes prefer loose-textured soils (Grinnell et al. 1937, Hall 1946, Egoscue 1962, Morrell 1972), but are found on virtually every soil type. Dens appear to be scarce in areas with shallow soils because of the proximity to bedrock (OFarrell and Gilbertson 1979, OFarrell et al. 1980), high water tables (McCue et al. 1981), or impenetrable hardpan layers (Morrell 1972). However, kit foxes will occupy soils with a high clay content, such as in the Altamont Pass area in Alameda County, where they modify burrows dug by other animals (Orloff et al. 1986).
Historically, San Joaquin kit foxes occurred in several native plant communities of the San Joaquin Valley. Because of extensive land conversions and intensive land use, some of these communities only are represented by small, degraded remnants today. Other habitats in which kit foxes are currently found have been extensively modified by humans. These include grasslands and scrublands with active oil fields, wind turbines, and an agricultural matrix of row crops, irrigated pasture, orchards, vineyards, and grazed annual grasslands (nonirrigated pasture). Other plant communities in the San Joaquin Valley providing kit fox habitat include Northern Hardpan Vernal Pool, Northern Claypan Vernal Pool, Alkali Meadow, and Alkali Playa. These are found as relatively small patches in scattered locations. In general, they do not provide good denning habitat for kit foxes because all have moist or waterlogged clay or clay-like soils. However, where they are interspersed with more suitable kit fox habitats they provide food and cover.
In the southernmost portion of the range, the kit fox is commonly associated with Valley Sink Scrub, Valley Saltbush Scrub, Upper Sonoran Subshrub Scrub, and Annual Grassland. Kit foxes also inhabit grazed grasslands, petroleum fields (Morrell 1971, OFarrell 1980), urban areas (B. Cypher pers. comm.), and survive adjacent to tilled or fallow fields (Jensen 1972, Ralls and White 1991). In the central portion of the range, the kit fox is associated with Valley Sink Scrub, Interior Coast Range Saltbush Scrub, Upper Sonoran Subshrub Scrub, Annual Grassland and the remaining native grasslands. Agriculture dominates this region where kit foxes mostly inhabit grazed, nonirrigated grasslands, but also live next to and forage in tilled or fallow fields, irrigated row crops, orchards, and vineyards. In the northern portion of their range, kit foxes commonly are associated with annual grassland (Hall 1983) and Valley Oak Woodland (Bell 1994). Kit foxes inhabit grazed grasslands, grasslands with wind turbines, and also live adjacent to and forage in tilled and fallow fields, and irrigated row crops (Bell 1994).
Kit foxes use some types of agricultural land where uncultivated land is maintained, allowing for denning sites and a suitable prey base (Jensen 1972, Knapp 1978, Hansen 1988). Kit foxes also den on small parcels of native habitat surrounded by intensively maintained agricultural lands (Knapp 1978), and adjacent to dryland farms (Jensen 1972, Kato 1986, Orloff et al. 1986).
Reasons for Decline.-- Numerous causes of kit fox mortality have been identified, though these have probably varied considerably in relative importance over time. Researchers since the early 1970s have implicated predation, starvation, flooding, disease, and drought as natural mortality factors. Shooting, trapping, poisoning, electrocution, road kills, and suffocation have been recognized as human-induced mortality factors (Grinnell et al. 1937, Morrell 1972, Egoscue 1975, Berry et al. 1987a, Ralls and White 1991, Ralls and White 1995, Standley et al. 1992).
y the 1950s the principal factors in the decline of the San Joaquin kit fox were loss, degradation, and fragmentation of habitats associated with agricultural, industrial, and urban developments in the San Joaquin Valley (Laughrin 1970, Jensen 1972, Morrell 1975, Knapp 1978). Extensive land conversions in the San Joaquin Valley began as early as the mid-1800s with the Arkansas Reclamation Act, and by 1958 an estimated 50 percent of the Valleys original natural communities had been lost (USFWS 1980a). In recent decades this rate of loss has accelerated rapidly with completion of the Central Valley Project and the State Water Project, which diverted and imported new water supplies for irrigated agriculture (USFWS in litt. 1995a). From 1959 to 1969 alone, an estimated 34 percent of natural lands were lost within the then-known kit fox range (Laughrin 1970). By 1979, only about 6.7 percent of the San Joaquin Valley floors original wildlands south of Stanislaus County remained untilled and undeveloped (USFWS 1980a). Such land conversions contribute to kit fox declines through displacement, direct and indirect mortalities, and reduction of prey populations.
Threats to Survival.-- Loss and degradation of habitat by agricultural and industrial developments and urbanization continue, decreasing carrying capacity of remaining habitat and threatening kit foxes. Livestock grazing is not thought to be detrimental to kit foxes (Morrell 1975, Orloff et al. 1986), but may alter the numbers of different prey species, depending on the intensity of the grazing. Livestock grazing may benefit kit foxes in some areas (Laughrin 1970, Balestreri 1981), but grazing that destroys shrub cover and reduces prey abundance may be detrimental (OFarrell et al. 1980, OFarrell and McCue 1981, USFWS 1983, Kato 1986).
Petroleum field development in the southern half of the San Joaquin Valley affects kit foxes by habitat loss due to grading and construction for roads, well pads, tank settings, pipelines, and settling ponds. Habitat degradation derives from increased noise, ground vibrations, venting of toxic and noxious gases, and release of petroleum products and waste waters. Traffic-related mortality is also a factor for kit foxes living in oil fields. The cumulative and long-term effects of these activities on kit fox populations are not fully known, but recent studies indicate that areas of moderate oil development may provide good habitat for kit foxes, as long as suitable mitigation policies are observed (OFarrell et al. 1980, Spiegel et al. in press). The impacts of oil activities at the Elk Hills Naval Petroleum Reserves in California on kit fox population density, reproduction, dispersal, and mortality appeared to be similar in developed and undeveloped areas of the Reserve (Berry et al. 1987a). The most significant impact on kit fox abundance in developed oil fields appears to be mediated through habitat loss. However, the relationship between habitat loss and population size in western Kern County is unclear: the Midway-Sunset oil field is highly developed with about 70 percent ground disturbance yet fox abundance is about 50 percent that of the undeveloped Lokern area (Spiegel et al. in press).
Other developments within the kit foxs range include cities and towns, aqueducts, irrigation canals, surface mining, road networks, non-petroleum industrial projects, power lines, and wind farms. These developments negatively impact kit fox habitat, but kit foxes may survive within or adjacent to them given adequate prey base and den sites. Kit foxes have been documented denning along canals and in levees (Jones and Stokes 1981, Hansen 1988), adjacent to highways (ESA Planning and Environmental Services 1986b, Hansen 1988), near wind farms (Hall 1983, Orloff et al. 1986), along power line corridors (Swick 1973), and at sanitary land fills (R. Faubion pers. comm.). Kit foxes also are known to live in and adjacent to towns such as Tulare (G. Presley pers comm.), Visalia (Zikratch pers. comm.), Porterville (Hansen 1988), Maricopa, Taft, and McKittrick (J.M. Sheppard pers. comm.) and the City of Bakersfield (Jones and Stokes 1981, B.L. Cypher pers. comm.). Bakersfieldfoxes (living in the Kern River Parkway) are reported to behave differently from animals in more remote populations: they often scavenge food from parking lots and dumpsters, have small foraging ranges, often are diurnal, and are relatively tame. This may be an expression of their ecological plasticity (e.g., Grinnell et al. 1937, p. 411, T. Murphy pers. comm., B.L. Cypher pers. comm.).
All these influences combine to compress and constrict the kit fox into fragmented areas, varying in size and habitat quality. The fragmentation of these areas coupled with the suspected high mortality during dispersal may limit movement to and habitat of these lands. As the human population of California continues to grow, the amount and quality of habitat suitable for kit foxes will inevitably decrease. Continued habitat fragmentation is a serious threat to the survival of kit fox populations.
The use of pesticides and rodenticides also pose threats to kit foxes. Pest control practices have impacted kit foxes in the past, either directly, secondarily, or indirectly by reducing prey. In 1925, near Buena Vista Lake, Kern County, seven kit foxes were found dead within a distance of 1 mile, having been killed by strychnine-poisoned baits put out for coyotes. It was suspected that hundreds of kit foxes were similarly destroyed in a single season (Grinnell et al. 1937). In 1975 in Contra Costa County (where the main prey item of kit foxes is the California ground squirrel), the ground squirrel was thought to have been eliminated county wide after extensive rodent eradication programs (Bell et al. 1994). In 1992, two kit foxes at Camp Roberts died as a result of secondary poisoning from rodenticides (Berry et al. 1992, Standley et al. 1992). The Federal government began controlling the use of rodenticides in 1972 with a ban of Compound 1080 on Federal lands pursuant to Executive Order. Above-ground application of strychnine within the geographic ranges of listed species was prohibited in 1988. Efforts have been underway to greatly reduce the risk of rodenticides to kit foxes (USFWS in litt. 1993).
Invasion and occupation of historical and potential kit fox habitats by nonnative red foxes may limit opportunities for kit foxes. Exclusion of kit foxes by competing red foxes, direct mortality, and potential for disease and parasite transmission all are issues that have not yet been researched. Therefore, we know neither the historical impacts to the kit fox, nor to what extent the continuing expansion of the range of nonnative red foxes will have on kit foxes.
Accidents and disease, though not well documented, are thought to play a minor role in kit fox mortality (USFWS 1983), however, at Camp Roberts rabies accounted for 6.3 percent of deaths of radio-collared kit foxes (Standley et al. 1992) and there is concern that rabies may be a contributing factor in the recent decline of kit foxes at Camp Roberts (P.J. White pers. comm.). Random catastrophic events such as drought or flooding present a significant threat. Drought, with a corresponding decline in prey availability, results in a decrease in kit fox reproductive success (White and Ralls 1993, Spiegel et al. in press). How extended periods of drought may affect kit fox populations is unclear, but local extinctions are likely in some isolated areas. Recently, small mammal populations have declined rapidly and severely, apparently due to the above average rainfall in the 1994-1995 precipitation year. In the Elk Hills region, relatively few pupping dens were found in 1995, and only a small proportion of kit fox pairs apparently raised pups (B.L. Cypher pers. comm., L.K. Spiegel pers. comm.).
The San Joaquin kit fox was listed as endangered by the U.S. Department of the Interior in 1967 (USFWS 1967) and by the State of California in 1971 (Table 1). A recovery plan approved in 1983 proposed interim objectives of halting the decline of the San Joaquin kit fox and increasing population sizes above 1981 levels (USFWS 1983).
Conservation efforts subsequent to the 1983 recovery plan have included habitat acquisition by USBLM, CDF, California Energy Commission, Bureau of Reclamation, USFWS, and The Nature Conservancy. Purchases most significant to conservation efforts were the acquisitions in the Carrizo Plain, Ciervo-Panoche Natural Area, and the Lokern Natural Area. A multi-agency acquisition is underway which would secure 60,000 acres straddling western Merced, Stanislaus, and eastern Santa Clara Counties. Other lands have been acquired as mitigation for land conversions, both temporary and permanent (Table 2). Mitigation in the form of management and research was granted to the California Energy Commission, U.S. Department of Energy (Naval Petroleum Reserves in California), Army National Guard (Camp Roberts), and Department of Defense (Fort Hunter Liggett). Most of the current research literature arises from these sources and The Smithsonian/Nature Conservancy-sponsored research on the Carrizo Plain Natural Area (White and Ralls 1993, White et al. 1994, Ralls and White 1995, White et al. 1996).
For over 15 years EG&G Energy Measurements has conducted research into the ecology of the kit fox population on the Naval Petroleum Reserves in California, Kern County. Reports have covered such topics as dispersal (Scrivner et al. 1987b), mortality (Berry et al. 1987a), and movements and home range (Zoellick et al. 1987b). Additionally, they have evaluated habitat enhancement, kit fox relocation, supplemental feeding (EG&G Energy Measurements 1992), and coyote control (Cypher and Scrivner 1992) as means of enhancing recovery. Other life history information has come from studies sponsored in whole or in part by CDFG, California Department of Water Resources, USFWS, Smithsonian Institution, Department of the Army and Air Force, California Energy Commission, and The Nature Conservancy (Hall 1983, Archon 1992, Spiegel and Bradbury 1992, White and Ralls 1993, White et al. 1994, 1996). Following the 1983 recovery plan, only three surveys for distribution have been conducted, two in the northern range of the fox (Orloff et al. 1986, Bell et al. 1994), and one in western Madera County (Williams 1990).
Large-scale habitat surveys have been conducted on the Carrizo Plain (Kato 1986, Kakiba-Russell et al. 1991) and the southern San Joaquin Valley (Anderson et al. 1991). A preliminary aerial survey for potential habitat was conducted along the east side of the Valley (Bell et al. 1994). There also have been numerous smaller-scale preproject surveys as part of the section 7 and 10(a) permit process of the Endangered Species Act, National Environmental Protection Act, and California Environmental Quality Act laws and regulations.
A population viability analysis was prepared for USFWS using RAMAS/a, a Monte Carlo simulation of the dynamics of age-structured populations (Buechner 1989). Since this analysis, deficiencies in the database have been identified and a metapopulation analysis has been completed (Kelly et al. 1995). This analysis, however, is preliminary and will be updated as new information is collected.
The U.S. Environmental Protection Agency County Bulletins governing use of rodenticides have greatly reduced the risk of direct mortality to San Joaquin kit fox populations by State and county rodent-control activities. The California Environmental Protection Agency, California Department of Food and Agriculture, county agricultural departments, CDFG, and U.S. Environmental Protection Agency collaborated with the USFWS in the development of County Bulletins that are both efficacious and acceptable to land owners (R.A. Marovich pers. comm.).
Though the kit fox has been listed for over 30 years, its status throughout much of its current range is poorly known. This is partly because so much of its historical range in the San Joaquin Valley is in private ownership. Similar gaps in information are common to many of the other listed and candidate species being addressed in this recovery plan. However, recovery actions for the kit fox are also considered critical to the recovery of many of these other species in the San Joaquin Valley. The kit foxs occurrence in the same naturalcommunities as most other species featured in this plan and its requirement for relatively large areas of habitat mean its conservation will provide an umbrella of protection for many of those other species that require less habitat. Therefore, a conservative recovery strategy is appropriate for this species and the following regional (or ecosystem level) recovery actions should be given high priority.
Given the importance and urgency of the situation, the recovery strategy for the kit fox needs to operate on two distinct but equally important levels: the continuation and expansion of recovery actions initiated subsequent to the original recovery plan using existing information; and, the development of new information in concert with expansion of existing information, which is currently inadequate for some aspects of recovery management.
Level A Strategy.-- The goal of this strategy is to work toward the establishment of a viable complex of kit fox populations (i.e., a viable metapopulation) on private and public lands throughout its geographic range. Although the exact dimensions of a viable kit fox metapopulation cannot be predicted in advance, there are general principles from conservation biology that can and must be applied for recovery of the San Joaquin kit fox (with due consideration to the current, inadequate knowledge about the animals life history, distribution, and status). Because kit foxes require large areas of habitat and have dramatic, short-term population fluctuations, one cannot rely on a single population to achieve recovery. Preliminary population viability analyses suggest that the Carrizo Plain population, the largest remaining, is not viable by itself nor is it viable in combination with populations in western Kern County and the Salinas Valley.
Conserving a number of populations, some much more significant than others because of their large sizes or strategic locations, therefore, will be a necessary foundation for recovery. The areas these populations inhabit need to encompass as much of the environmental variability of the historical range as possible. This will ensure that maximal genetic diversity is conserved in the kit fox metapopulation to respond to varying environmental conditions, and that one environmental event does not negatively impact to the same extent all existing populations. Also, connections need to be established, maintained, and promoted between populations to counteract negative consequences of inbreeding, random catastrophic events (e.g., droughts) and demographic factors.
A sound, conservative strategy hinges on the enhanced protection and management of three geographically-distinct core populations, which will anchor the spine of the metapopulation. A number of smaller satellite populations (number and location yet to be determined, probably 9 to 12 or more) will be fostered in remaining fragmented landscapes through habitat management on public land and conservation agreements with private land owners.
The three core populations are:
These three core populations each are distinct. The western Kern County and Carrizo Plain populations, although geographically close, are separated by the Temblor Range. Although both locations have high fox densities from time to time, they also have differet environmental conditions, which are reflected in the fact that their population dynamics are not always synchronous (B.L. Cypher pers. comm., Endangered Species Recovery Program unpubl. observ.). These differences amongst the core populations are important considerations in conservation planning. Also, preliminary population viability analyses indicate that extinction probabilities increase dramatically if either the Carrizo Plain or western Kern County population is eliminated. Finally, both of these locations have large amounts of land in public ownership, lowering the burden on private land owners to assist in recovery of the kit fox. The Carrizo Plain and western Kern County populations are important for kit fox recovery.
The Ciervo-Panoche Natural Area population is located more than 160 kilometers (100 miles) northwest of the other two core populations. As with the other core populations, it has significant numbers of foxes, at least it had historically and it still may from time to time, and large expanses of land are in public ownership. It also experiences a different environmental regime from the other two. Finally, preliminary metapopulation viability analyses indicate that recovery probabilities increase if a population is established or maintained in this area, apparently because of its different environmental regime.
In addition to basing the choice of these three core populations on the above criteria, this particular metapopulation configuration has an additional important advantage over combinations of other fox populations. These three populations are more or less connected to each other by grazing lands, although they are steep and rugged in many places. Kit foxes occur at varying densities in the areas between the core populations (e.g., Kettleman Hills), providing linkages between core populations, and also probably with smaller, more isolated populations in adjacent valleys.
Important kit fox populations in the Salinas-Pajaro Region (herein defined as the area of the Salinas River and Pajaro River watersheds with habitat for kit foxes; Figs. 1 and 51) are located at Camp Roberts and Fort Hunter Liggett in the Salinas River Watershed. Though there are natural connections between the Salinas-Pajaro Region, the Carrizo Plain Natural Area, and the San Joaquin Valley, the amount of movement of kit foxes between the Salinas-Pajaro Region and these areas is unknown, though one fox is known to have moved from Camp Roberts to the Carrizo Plain (K. Ralls pers. comm.).
Other lands in the San Joaquin Valley that have kit foxes, or the potential to have them, include refuges and other lands managed by the CDFG, California Department of Water Resources, Center for Natural Lands Management, Lemoore Naval Air Station, Bureau of Reclamation, and USFWS, as well as those on private lands in western Madera County, central, western, and eastern Merced County, eastern Stanislaus County, northern Kings County, around Pixley National Wildlife Refuge and Allensworth Ecological Reserve in Tulare County, Semitropic Ridge Natural Area and around the Bakersfield metropolitan area of Kern County (Figure 51).
Many of these more isolated natural lands exhibit symptoms of ecosystem fragmentation such as degradation of natural communities and loss of biodiversity. Nevertheless, some fragments have resident kit foxes by virtue of their proximity to other populations, and others serve as important corridors between kit fox populations. For example, the California Department of Water Resourcess Kern Fan Element provides an important linkage between kit foxes along the Kern River Parkway in Bakersfield and the western Kern County core population.
Yet, many of these areas, despite having suitable habitat, have become so degraded over time, reduced in size, and isolated from extant kit fox populations that they rarely have kit foxes today. When they do, these small, isolated populations are very susceptible to local extinction. It is likely that the degree of isolation from larger, more stable kit fox populations is the primary reason for absence or very low densities of kit foxes on some of the larger parcels of natural land remaining on the Valley floor (e.g., central Merced Count, western Madera County, and the Mendota area, Fresno County; Williams 1990).
Connecting larger blocks of isolated natural land to core and other populations, thus, is an important element of recovery of kit foxes. Connecting large blocks will help reduce the harmful effects of habitat loss and fragmentation. To enhance these connections, conservation lands on the Valley floor could be increased in size through acquisition of title or conservation easements, or a combination of both.
Another complementary approach is to reduce the level of isolation by promoting conservation of kit foxes on agricultural lands through "safe harbor" and other initiatives. New procedures and regulations must ensure that farmers are not penalized and farming not disrupted by enhancing use of farmland by kit foxes. The goal should be specific incentive programs to encourage farmers to maintain, enhance, or create habitat conditions for kit foxes. The ideal situation would be to establish a small number of breeding kit foxes in farm lands. A proposal to address habitat fragmentation in this way has already been developed by the American Farmland Trust (Scott-Graham 1994). Those lands could then serve as bridges between the more isolated refuges and reserves and the larger populations along the spine of the metapopulation, on the west side of the San Joaquin Valley.
Concurrently, strategic retirement of agricultural lands that have serious drainage problems will help reduce the effects of widespread habitat fragmentation of populations. Land retirement for reducing or eliminating drainage problems has been authorized by both State and Federal governments. In particular, the Central Valley Project Improvement Act of 1992 has provisions and funding for such land retirement. If land retirement proves not to pose a contaminant issue, the program can greatly boost recovery of kit foxes and other listed species and species of concern in the San Joaquin Valley. If large blocks (ideally, no less than 2,023 to 2,428 hectares [5,000 to 6,000 acres]) of drainage-problem lands are retired from irrigated agriculture, the retired farmland can be converted to habitat for kit foxes, kangaroo rats, blunt-nosed leopard lizards, and other listed and sensitive species. Those land blocks can provide more than just habitat. They can also reduce isolation and its detrimental effects. If strategically located, they can provide "stepping stones" for movement of kit foxes between Valley floor and west side populations. Strategic irrigated land retirement and subsequent establishment as habitat conservation areas is the most cost effective and rapid route to recovery of kit foxes.
Level B Strategy.-- While land retirement and habitat restoration and management get under way, other urgent recovery needs, which are primarily research-related or informational in nature, must be addressed. The acquisition of new and better information will permit refinement of the viability models and land-use optimization models that are under development for the kit fox. In turn these models will assist in management of kit fox populations.
Needed is information on distribution and status throughout most of its current and historical range. Much better information on the distribution, status and movements of kit foxes is needed, particularly in the Salinas-Pajaro Region and the northern and eastern San Joaquin Valley.
Good data also are needed on the use of agricultural lands by kit foxes. Better demographic information is needed for kit foxes living in natural, agricultural, residential, and industrial lands throughout their range. Most of the existing data are for the southern part of the Valley where the environmental regime is more arid, and destruction of former fox habitat has been much more recent. Better data on the relationship between prey populations and kit fox population dynamics also are needed. A better understanding is needed of how kit foxes interact with red foxes, the indirect impacts of rodenticide use, and the influence of predator control activities.
Recovery Actions.-- Recognizing that recovery requires a dual track with simultaneous actions, recovery actions ar ordered in two lists, each of approximately equal priority to the other: a) habitat protection and population interchange, and, b) population ecology and management. Habitat protection and enhancement requires appropriate land use and management. To do so often requires purchase of title or conservation easement, or another mechanism of controlling land use. However, until needed research is completed, if listed species occur on an acquired parcel, the general rule of thumb should be that no dramatic changes in land use be made until appropriate management prescriptions have been determined. Many elements of management must first be determined by scientific research; thus the concept of adaptive management (monitoring and evaluating outcomes, then readjusting management directions accordingly) is operative here. A high priority therefore is the research required to determine appropriate habitat management and other recovery actions.