Adult and Fawn Mortality of Sonoran Pronghorn Author(S): Jill L

Adult and fawn mortality of Sonoran pronghorn Author(s): Jill L. Bright and John J. Hervert Source: Wildlife Society Bulletin, 33(1):43-50. Published By: The Wildlife Society DOI: http://dx.doi.org/10.2193/0091-7648(2005)33[43:AAFMOS]2.0.CO;2 URL: http://www.bioone.org/doi/full/10.2193/0091-7648%282005%2933%5B43%3AAAFMOS%5D2.0.CO%3B2

BioOne (www.bioone.org) is a nonprofit, online aggregation of core research in the biological, ecological, and environmental sciences. BioOne provides a sustainable online platform for over 170 journals and books published by nonprofit societies, associations, museums, institutions, and presses. Your use of this PDF, the BioOne Web site, and all posted and associated content indicates your acceptance of BioOne’s Terms of Use, available at www.bioone.org/page/terms_of_use. Usage of BioOne content is strictly limited to personal, educational, and non-commercial use. Commercial inquiries or rights and permissions requests should be directed to the individual publisher as copyright holder.

BioOne sees sustainable scholarly publishing as an inherently collaborative enterprise connecting authors, nonprofit publishers, academic institutions, research libraries, and research funders in the common goal of maximizing access to critical research. 06 SC (Bright and Hervert).qxp 6/24/2005 12:28 PM Page 43

SPECIAL COVERAGE 43

Adult and fawn mortality of Sonoran pronghorn by Jill L. Bright and John J. Hervert

Abstract We documented adult mortality and fawn recruitment of Sonoran pronghorn (Antilocapra americana sonoriensis) in Arizona. This population is endangered (N<30) and is decreasing due to low fawn recruitment and adult mortality. We radiocollared Sonoran pronghorn to monitor survival and recruitment from 1995–2002. We relocated each radiocollared pronghorn during weekly telemetry flights from a fixed-wing airplane and recorded group composition to determine recruitment. Mortalities detected during flights were investigated as soon as possi- ble (i.e., <48 hours) to document adult mortality. Adult mortality rates varied from 11–83%/year. Adult pronghorn were killed by coyotes (Canis latrans), bobcats (Felis rufus), mountain lions (Puma concolor), capturing efforts, drought, and unknown causes. Fawn mortality varied from year to year and was correlated with the amount and timing of rainfall. Drought may be a major factor in the survival of adults and fawns. A lack of nutritious forage and water, caused by dry conditions, led to high fawn mortality and caused adult mortality during a particularly severe drought. Drought also may indirectly affect adult mortality by causing animals to use areas where predators are more successful. Disease may affect mortality but remains largely uninvestigated. Management applications aimed at increasing fawn recruitment and reducing adult mortality should increase the chances of survival of this species. Providing sources of highly nutritious forage during early spring and summer when fawns are susceptible to poor nutrition may increase their chances of survival. Providing water sources and nutritious forage in areas where predators are less successful may increase both adult and fawn survival. Predator control may be useful in limited situations but likely would be prohibitively expensive, with lit- tle chance of making a difference over the entire range of Sonoran pronghorn. Key Words Antilocapra americana sonoriensis, drought, fawn recruitment, mortality, preda- tion, Sonoran pronghorn

ronghorn (Antilocapra americana) occur only in North FR 4001). Sonoran pronghorn range includes the plains America, and 5 subspecies are recognized throughout the of west-central Sonora, Mexico, north to southwestern continent. The Sonoran pronghorn subspecies (A. a. Arizona (Wright and deVos 1986). Historically, prong- sonoriensis) was classified in 1945 (Goldman 1945). It horn were found in every open valley along the United Pis the smallest of the subspecies and has lighter pelage, States border with Mexico from Nogales to Yuma (Carr and distinctive cranial features (Wright and deVos 1986). 1971). The United States population of Sonoran prong- The Sonoran pronghorn was listed as endangered in horn was estimated to be 105 in 1924, 60 (excluding 1967 by the United States Fish and Wildlife Service (32 Organ Pipe Cactus National Monument) in 1941, <100 in

Authors' address: Arizona Game and Fish Department, 9140 E. 28th Street, Yuma, AZ 85364, USA; e-mail for Bright: [email protected].

Wildlife Society Bulletin 2005, 33(1):43–50 Peer refereed 06 SC (Bright and Hervert).qxp 6/24/2005 12:28 PM Page 44

44 Wildlife Society Bulletin 2005, 33(1):43–50

1956, <50 in 1968, 50–150 during 1968–1974, and adults and fawns. We hypothesized that mortality rates 85–100 from 1983–1985 (Wright and deVos 1986). would be higher in dry years and that mortality rates More recent population survey estimates were 246 in would vary among the different vegetation associations 1992, 184 in 1994 (Snow 1994), 130 in 1996, 142 in pronghorn occupy. To assess these hypotheses, we exam- 1998 (Bright et al. 1999), and 99 in 2000 (Bright et al. ined correlations between mortality and vegetation asso- 2001). Following a drought, 80% of the population died, ciation use-patterns and rainfall. and the Arizona population was estimated at 21 animals in 2002 (Bright and Hervert 2003). A population viability model done in 1996 indicated Study area that Sonoran pronghorn have a 23% probability of The study area encompassed the current range of extinction within the next 100 years; if the population Sonoran pronghorn in the United States. It was bounded by Interstate 8 to the north, Mexico to Predator control targeting fawn survival would be the south, Arizona Highway 85 to the east, and the Copper and Cabeza Prieta successful only when adequate forage is available to Mountains to the west (Figure 1). meet the nutritional needs of pronghorn fawns. Broad alluvial valleys separated by block fault mountains characterized the area. Elevations ranged from 100 m in falls below 100 individuals, the probability of extinction the San Cristobal Valley to >550 m near Ajo, Arizona. increases markedly. In addition, the model suggested Pronghorn primarily occupied the valleys and bajadas, that the Sonoran pronghorn population is most sensitive with occasional use of the foothills and lower drainages to fawn survival rates. Adult survival rates also were of mountain ranges. Water sources were not common strongly correlated with the likelihood of species sur- within the range of Sonoran pronghorn. vival, and even the loss of 1 animal/year could be a sig- The study area included the Lower Colorado River nificant threat (Hosack et al. 2002). Valley and the Arizona Upland subdivisions of the Lower Understanding the effects of predation and other mor- Sonoran Desert Life Zone (Brown 1982). The Lower tality factors on Sonoran pronghorn is critical to the long- Colorado River Valley subdivision was dominated by cre- term survival of the species. Management actions that osote bush (Larrea tridentata) and white bursage result in a decrease in mortal- ity rates for adults and fawns would be expected to provide benefits to Sonoran prong- horn. The United States Fish and Wildlife Service (USFWS) recognized the lack of survival data and the importance of increasing adult and fawn survival rates. The Sonoran pronghorn recovery plan (USFWS 1998) recommended 1) investigat- ing the effects of predation on Sonoran pronghorn, 2) enhancing Sonoran prong- horn numbers through fawn recruitment, and 3) increasing adult and fawn survival through habitat enhancement techniques. Our objectives were to estimate mortality rates and causes in Sonoran pronghorn Figure 1. Distribution of Sonoran pronghorn in the United States, 2004. 06 SC (Bright and Hervert).qxp 6/24/2005 12:28 PM Page 45

Mortality in Sonoran pronghorn • Bright and Hervert 45

(Ambrosia dumosa). The Arizona Upland subdivision plotted locations of pronghorn mortalities on a map was characterized by palo verdes (Cercidium spp.), showing the distribution of palo verde–chain fruit cholla saguaro (Carnegiea gigantea), chollas (Opuntia spp.), associations (PVC), which are selected by pronghorn ocotillo (Fouquieria splendens), and ironwood (Olneya (Hervert et al. 2000). We compared numbers of mortali- tesota). We divided the study area into 3 vegetation asso- ties between PVC availability versus non-PVC associa- ciations. Creosote–bursage was dominated by creosote tions availability (creosote–bursage and palo and bursage and generally was on flat terrain. Palo verde–mixed cacti) using chi-square tests. We also com- verde–mixed cacti was more diverse and included cre- pared mortalities to known vegetation association use- osote, palo verde, ironwood, saguaro, many types of patterns (Hervert et al. 2005). To examine the effect of chollas, and other cacti. Palo verde–chain fruit cholla (O. drought on pronghorn mortality, we compared mortality fulgida) was similar to palo verde–mixed cacti but rates in dry years (>5.0 cm below normal rainfall) to included chain fruit cholla. mortality rates in wet years using t-tests. Extreme aridity and heat characterized the Sonoran Desert. The average annual maximum temperature was Fawn recruitment and mortality 32oC and the minimum was 12oC. Temperatures rarely We estimated fawn recruitment and mortality from dropped below 0oC and summer highs could exceed group composition recorded on weekly telemetry flights. 49oC, with surface temperatures approaching 82oC We recorded the number of fawns and females in groups (Dimmitt 2000). Average annual precipitation ranged associated with radiocollared animals and calculated from 11 cm at Tacna near the western edge of Sonoran number of fawns/100 females to use as an estimate of pronghorn range to 25 cm in the east at Organ Pipe recruitment. We used Pearson’s correlation to determine Cactus National Monument. whether recruitment was correlated with winter rainfall, The Sonoran desert was characterized by a bimodal summer rainfall, total rainfall, or the number of days rainfall pattern. From December to March, frontal between the last winter rain and the first summer rain. storms originating in the north Pacific brought wide- An estimate of recruitment was obtained each week and spread gentle rain. Heavy localized thunderstorms graphed to determine when mortality was occurring. occurred from July through September associated with warm moist air from the Gulfs of Mexico and California. These rainfall periods were separated by intervals of low Results rainfall in autumn and spring. May and June typically Twenty-two animals (4 M, 18 F) were captured in were hot, with very low humidity and no rain in most November and December 1994, 9 (3 M, 6 F) in years (Dimmitt 2000). December 1997 and January 1998, and 4 (4 F) in December 2000. We tracked these animals until they died or their radiocollars ceased functioning. We tracked Methods the last animal until November 2002, when its radiocollar Adult mortality failed. We captured Sonoran pronghorn using a net gun fired from a helicopter. We placed VHF radiocollars (Telonics, Adult mortality Mesa, Ariz.) equipped with a mortality sensor on each We documented mortalities of 28 radiocollared prong- animal. We monitored radiocollared animals from an air- horn. In addition, we found 4 mortalities of uncollared plane weekly and recorded location and group composi- pronghorn. Of these 32 mortalities; 6 were from coyote tion. When we detected a mortality signal, the site was (Canis latrans) predation, 3 from bobcat (Felis rufus) investigated as soon as possible, usually the same day. In predation, 2 from mountain lion (Puma concolor) preda- addition, we observed and investigated mortalities of tion, 1 from unknown predation, 5 from capture myopa- some uncollared pronghorn during the course of other thy, 4 from drought-related factors (i.e., malnutrition, fieldwork. Cause of death was determined from sign at starvation, or dehydration), and 11 from unknown causes. the carcass (i.e., tracks, location of bite marks, signs of Of the 11 unknowns, 4 cases revealed no signs of injury struggle, length of chase) and condition of the carcass. or predation. One unknown might have died from a We calculated adult mortality rates using the radiocol- vehicle or train collision near Interstate 8. lared sample of pronghorn that was available to be moni- Since 1995 adult mortality averaged 28%/year; how- tored each year. Uncollared pronghorn mortalities were ever, mortality was 83% in 2002, which was the driest not used in calculation of mortality rates. year on record in southern Arizona (Table 1). Prior to To examine mortalities by vegetation association, we 2002, mortality averaged 20%/year. Mortality rates in 06 SC (Bright and Hervert).qxp 6/24/2005 12:28 PM Page 46

46 Wildlife Society Bulletin 2005, 33(1):43–50

Table 1. Mortalities of collared Sonoran pronghorn in Arizona, slight evidence that fawn recruitment was positively cor- 1995–2002. related with the amount of winter rain (r=0.69, P=0.06; Figure 2). No significant correlations were found Year No. mortalities No. working collars between recruitment and summer rainfall (r=–0.007, P= a 1995 316 0.99) or total rainfall (r=0.525, P=0.18). 1996a 513 1997 1 8 1998 3 15 1999 1 0 Discussion 2000a 39Adult mortality 2001 1 9 We documented adult mortality from various causes. 2002a 56From the original capture in 1994, 6 pronghorn died within 2 months. We attributed 5 of these directly to a Years classified as dry (>5.0 cm below normal rainfall). capture myopathy. The pronghorn captured in 1994 like- ly were pre-stressed by drought conditions, possibly pre- dry years averaged 43%/year compared to 13%/year in disposing them to capture myopathy. The other animal wetter years; however, there was no significant difference showed evidence of coyote predation, but it might have between the wet and dry years (t6=–2.14, P=0.076). We been stressed from the capture. There were no capture- suspect we could not detect a statistical difference related deaths from the second capture in 1997, when because of the large variation in mortality rates caused by range conditions were better and intravenous fluids (i.e., 83% mortality in 2002 (highest prior to this was lactated ringers) were given to each animal. During the 33%/year). If this year was removed from the analysis, third capture in 2000, 1 animal died from a broken neck; mortality in dry years averaged 30%/year versus 13% in however, none of the successfully collared animals suf- wet years, which was statistically significant (t5=3.06, P fered from capture myopathy. =0.03). In 2002 the most severe drought on record occurred in Mortalities from predation were more common in southern Arizona. Rain gauges near the study area PVC than would be expected based on availability of the recorded only 3.3 cm of rain, 17.3 cm below the long- associations (χ2=16.2, P<0.001). Most predation mor- term normal amount (National Oceanic and Atmospheric talities from coyotes (83%) and mountain lions (100%) Administration [NOAA] 2002). During summer, occurred in PVC. One of 3 bobcat predations occurred between 1 June and 13 August, 4 out of 5 (80%) of col- in PVC. lared pronghorn died. These animals were found intact, with no evidence of predation or scavenging. Some Fawn recruitment and mortality pronghorn rumens were almost completely full of chain Estimates of fawn recruitment/100 females from fruit cholla fruits, which, although high in water content, December telemetry varied from 0–78 fawns/100 females provide little nutrition (Fox 1997). Cause of death was (Table 2). Number of fawns recruited was inversely cor- believed to be malnutrition, starvation, or dehydration, related to the number of days between the last winter rain attributed to drought. During this drought we observed and the first summer rain (r=–0.78, P=0.02). There was dead and dying perennial shrubs and trees and a complete lack of annual plants. Table 2. Rainfall and recruitment of Sonoran pronghorn fawns in Arizona, 1995–2002. Nutritious forage was largely unavailable, and forage that Amount of rainfall Fawn recruitment did exist was dry. Known Year Winter (cm) Summer (cm) Total (cm) No. days a Fawnsb alive to end of December sources of water were limited 1995 11.4 3.6 15.0 90 12 to 2 catchments on the 1996 3.0 7.6 10.6 123 0 Cabeza Prieta National 1997 3.7 6.2 9.9 108 0 Wildlife Refuge, both of 1998 18.8 9.4 28.2 66 33 which were utilized by 1999 5.9 12.4 18.3 93 0 pronghorn. We suggest that 2000 4.5 3.5 8.0 107 14 during this drought the lack 2001 14.0 6.4 20.4 61 78 of quality forage most likely 2002 1.8 1.5 3.3 127 8 was responsible for prong- a No. days between last winter rain and first summer rain. horn mortalities. b No. fawns/100 females. A less severe drought 06 SC (Bright and Hervert).qxp 6/24/2005 12:28 PM Page 47

Mortality in Sonoran pronghorn • Bright and Hervert 47

and denser than in the sur- rounding creosote–bursage- dominated valleys, which may make pronghorn more vulnerable to predators. We suspect that predator densities may be greater and visibility lower, and that more rugged terrain may favor the predator in these areas. Pronghorn mortalities from predation were more common in PVC (9 of 12) than non-PVC asso- ciations. Pronghorn used PVC 48% of the time, and 75% of the predation mortali- ties occurred there. In con- trast, pronghorn used non- PVC associations 52% of the time where 25% of the mor- talities occurred. These Figure 2. Number of Sonoran pronghorn fawns/100 females observed on December telemetry flights, compared to amount of winter rainfall (October–May), 1994-2002. movements to bajadas and chain fruit cholla areas dur- ing dry periods may influ- occurred between June 1995 and August 1997 during ence pronghorn mortality by increasing predation. which 23 of 27 months had below-average rainfall Reducing pronghorn dependence on these areas by creat- (NOAA 1995, 1996, 1997). Nine of 16 collared prong- ing water and forage sources in more open, less rugged horn (56%) died during this time. Range conditions from habitats may decrease mortality rates. October 1998 through September 2000 were dry, Although one might expect the hot summer months in although there was good rainfall in July and August the Sonoran Desert to be hardest on Sonoran pronghorn, 1999. During these dry periods, we observed that forage most of the mortalities (83% of predation and 50% of un- was available in some areas, and we did not document knowns) occurred during the winter months. Of the mor- any adult deaths from starvation or malnutrition related talities known to have been caused by predation, all those to the drought. However, 11 of 12 mortalities (92%) from coyotes and bobcats and half from mountain lions oc- attributed to predation occurred during these dry periods. curred in winter months. Coyotes breed in winter and coy- Sonoran pronghorn move to bajadas during dry condi- ote group sizes are at their highest then, making them more tions (Hervert et al. 2000, 2005). Additional moisture is successful at taking large prey. Pack-hunting behavior available to plants due to runoff from nearby mountain most likely is necessary to successfully take large animals slopes and the water-shedding nature of the desert pave- such as adult pronghorn (Gese et al. 1988). This combina- ment commonly found on bajadas (McAuliffe 2000). tion of pronghorn preference for bajadas during dry periods This leads to a greater diversity of forage, and the con- and coyote life history during the winter months may make centration of water allows plants to remain green and a winter drought especially harmful for adult pronghorn. succulent longer than plants on flat terrain. In addition, Bobcats may prey on Sonoran pronghorn during the winter chain fruit cholla commonly grows on bajadas and pro- when prey availability is more limited. vides a source of water for pronghorn that consume the Predation is a natural process that typically causes no fruits. The fruit of this plant contains up to 85% water long-term ill effects for a large, healthy prey population. by weight and retains high moisture content even during However, a population that is depressed such as Sonoran the hot, dry summer (Hughes and Smith 1990, Fox pronghorn may no longer be able to withstand such pres- 1997). Sonoran pronghorn prefer PVC associations, and sures. Population fluctuations due to natural causes such their use of this association increases in dry conditions as predation may help drive a small population to a (Hervert et al. 2000). threshold below which the probability of survival and However, vegetation in these areas generally is taller recovery becomes extremely low (Hecht and Nickerson 06 SC (Bright and Hervert).qxp 6/24/2005 12:28 PM Page 48

48 Wildlife Society Bulletin 2005, 33(1):43–50

1999). In cases such as these, predator management may mid April in the Sonoran Desert. Approximately 75% of become necessary in the short term. the total cost of reproduction in ungulates occurs during In 3 cases listed as unknown (June, July, and Septem- lactation (Byers and Moodie 1990). Rainfall is least like- ber 1998), pronghorn died during a wet summer that fol- ly to occur during May and June (Dimmitt 2000), and lowed an especially wet winter, when range conditions with low quality and quantity of forage available, Sono- were favorable throughout pronghorn range. We found ran pronghorn may be unable to produce sufficient milk no evidence of injury or predation in these cases, perhaps for the entire period of lactation (expected to be >12 indicating disease as a factor. Disease-related mortality weeks based on observations of American pronghorn [A. could not be adequately evaluated under the field condi- a. americana], Byers 1997). As a consequence of poor tions we experienced. Bloodwork analyzed from Sono- milk production, some fawns probably died or were ran pronghorn captures has shown positive titers for blue forced to forage sooner and to a greater extent to meet di- tongue and epizootic hemorrhagic disease (J. Hervert, etary needs. In years with poor winter rainfall, the nutri- unpublished data). These diseases likely would be more tional quality of forage may be insufficient to keep fawns prevalent during warm, wet periods (Hoff and Trainer alive. Nutrition was acknowledged as a factor in fawn 1981). Disease-related mortality could be a significant survival, given that mortality was less pronounced when mortality factor that remains largely uninvestigated. forage with high moisture content was more abundant during spring and summer (Beale and Smith 1966). Fawn mortality We did not find any statistical correlations between Fawn mortality peaked at 2 times, depending on the fawn mortality and summer rainfall. We suspect that year, between birth and their first winter (Figure 3). relationships with summer rainfall are hard to detect for Winter storms are the primary stimulant of plant produc- several reasons. Summer rains generally are localized tivity in the Sonoran Desert (Patten 1978). Fawn mortal- and may not have the same widespread effect across ity is directly related to the amount of winter rain and to pronghorn range as winter rains. Due to the localized the amount of time between winter and summer rains. In nature of summer rains, rainfall amounts at recording sta- years with low winter rainfall such as 1996, 1997, 1999, tions and on pronghorn range may have differed signifi- and 2000, we saw an increase in fawn mortality during cantly. In addition, pronghorn are highly mobile and May and June. Fawns typically are born late February– most likely move to areas where rain has fallen. Finally, in dry winter years such as 1996 and 1997, few fawns survived until summer rains arrived; therefore the amount of summer rain was of little consequence. In years with late or mini- mal summer rains, a second period of high fawn mortality occurred during July and August. This occurred in every year except 1999, which had above-average and early summer rainfall. This most likely was due to increasing temperatures, desiccation of forage, and increased water needs of pronghorn. The availability of quality forage for pronghorn is a primary factor in fawn survival (Ellis 1970). The highest fawn mor- tality occurred during or fol-

Figure 3. Number of Sonoran pronghorn fawns/100 females based on weekly telemetry flights in south- lowing drought periods when western Arizona from April–December, 1995–2002. forage was in poor condition 06 SC (Bright and Hervert).qxp 6/24/2005 12:28 PM Page 49

Mortality in Sonoran pronghorn • Bright and Hervert 49

in a study by Beale and Smith (1966). did not document any adult pronghorn mortalities attribut- In addition, freestanding water may be important to able to drought during these less severe droughts. How- Sonoran pronghorn during hot, dry conditions. American ever, chain fruit cholla provides little nutrition, and this pronghorn could not live without water during hot weath- diet probably is not sufficient for growing fawns. We er, even if forage succulence was above average, and documented fawn mortality during drought periods during fawns were most affected by a lack of water (Beale and all years of the study. Drought also may also affect adult Holmgren 1975). Summer forb production and free- mortality by causing pronghorn to use areas where forage standing water availability may determine the ultimate diversity and succulence is greater, but predators may be level of fawn recruitment observed in this population more numerous and successful. Predation appears to be from year to year. especially high in winter, and during winter droughts. Coyote predation has been widely reported as a main Management actions such as providing water sources cause of pronghorn fawn mortality during the first 3 and increasing nutritious forage in habitats where preda- weeks after birth (Trainer et al. 1983, Ockenfels et al. tors are less successful may increase adult and fawn sur- 1992, Canon 1993). We found fawn survival high for >4 vival. Providing additional sources of highly nutritious weeks during all years except 1996. This would indicate forage during the early spring and summer, when fawns that coyote predation probably was not a major factor. are most susceptible to poor nutrition, may also enhance Instead, most fawn mortality occurred between 3–5 their chances of survival. months of age. Although we could not document the Limited, localized coyote control, such as in areas cause of death, we do not believe coyote predation was where newborn fawns exist or in forage-rich areas where the primary factor after fawns had reached this age and adults are known to forage during dry winters, may help were less vulnerable. reduce pronghorn mortalities in the short term. However, Bobcat predation on pronghorn fawns has been report- due to the large areas and scarcity of pronghorn, range- ed to be significant: 27 of 29 fawns killed by predators wide coyote control programs likely would be prohibi- were killed by bobcats over a 5-year period in Utah tively expensive and have little chance to make a differ- (Beale and Smith 1973). In addition, bobcats were suc- ence. Our data suggested that large numbers of fawns cessful in taking older fawns, up to 104 days after birth. are likely to die in most years due to a lack of adequate Beale and Smith found that dry conditions caused prong- nutrition. Predator control targeting fawn survival would horn to use washes, which also were the preferred habitat be successful only when adequate forage is available to of bobcats. Although we did not document any fawn meet the nutritional needs of pronghorn fawns. In addi- deaths to bobcats, the 3 adult pronghorn killed by bob- tion, bobcats move into coyote habitat when coyotes are cats occurred near washes. Beale and Smith (1973) sug- removed (Robinson 1961). Removing coyotes may have gested placing watering devices in open areas away from a negative effect, because bobcats may be more success- habitats preferred by bobcats to achieve separation in ful than coyotes for longer periods of time at reducing habitat used by the 2 species. However, if Sonoran pronghorn fawn numbers. Furthermore, nearly complete pronghorn are using habitats preferred by bobcats to removal of bobcats would be required to significantly access better-quality forage, both water and forage may reduce predation (Beale and Smith 1973). Relative den- need to be placed in open areas to achieve separation. sities of bobcats and their habitat-use patterns in the Bobcats are common in the Sonoran Desert and probably Sonoran desert are not well documented and should be are a factor in Sonoran pronghorn fawn mortalities. investigated further. Further research also is needed on predation of Sonoran pronghorn fawns.

Management implications Acknowledgments. We thank M. Brown, B. Henry, Adult and fawn mortality rates strongly influence and L. Piest for assistance with data collection. We are Sonoran pronghorn population size. Management actions especially grateful to S. Sunde, who gave up his Sunday that reduce mortality should greatly increase the chances mornings to fly nearly every weekly telemetry flight of survival of this subspecies. Our data suggested that since the study began in 1995. We appreciate P. R. drought is a factor in pronghorn mortality. Drought indi- Krausman, B. Steidl, and an anonymous reviewer for rectly influences mortality by decreasing the quality and providing constructive reviews on earlier versions of this quantity of forage and water available to pronghorn. Dur- manuscript. This study was funded by the Marine Corp ing all but the most severe droughts, adult pronghorn Air Station Yuma, the United States Air Force 56th seem to be able to survive on less nutritious forage and Range Management Office, and the United States Fish chain fruit cholla as a source of preformed water, as we and Wildlife Service. 06 SC (Bright and Hervert).qxp 6/24/2005 12:28 PM Page 50

50 Wildlife Society Bulletin 2005, 33(1):43–50

Literature cited southwest Arizona. Final Report 14–16–009–1564 RWO#6. Coopera- tive Fish and Wildlife Resources Unit,Tucson,Arizona, USA. BEALE,D.M.,AND R. C. HOLMGREN. 1975. Water requirements for antelope MCAULIFFE, J. R. 2000. Desert soils. Pages 87–104 in S. J. Phillips, and P.W. fawn survival and growth. Utah Department of Natural Resources Comus, editors. A natural history of the Sonoran Desert. Arizona- Wildlife Federal Aid Project W-65-D23. Salt Lake City, USA. Sonora Desert Museum,Tucson, USA. BEALE,D.M.,AND A. D. SMITH. 1966. Antelope fawn production and sur- NATIONAL OCEANIC AND ATMOSPHERIC ADMINISTRATION. 1995. Climatological vival on semi-desert range in western Utah. Proceedings Antelope data annual summary,Arizona, 1995. 100 (13). States Workshop 2:66–74. NATIONAL OCEANIC AND ATMOSPHERIC ADMINISTRATION. 1996. Climatological BEALE,D.M.,AND A. D. SMITH. 1973. Mortality of pronghorn antelope data annual summary,Arizona, 1996. 101 (13). fawns in western Utah. Journal of Wildlife Management 37:343–352. NATIONAL OCEANIC AND ATMOSPHERIC ADMINISTRATION. 1997. Climatological BRIGHT,J.L.AND J. J. HERVERT. 2003. Sonoran pronghorn 2002 aerial survey data annual summary,Arizona, 1997. 102 (13). summary. Arizona Game and Fish Department Nongame and Endan- NATIONAL OCEANIC AND ATMOSPHERIC ADMINISTRATION. 2002. Climatological gered Wildlife Program Technical Report 236. Phoenix, USA. data annual summary,Arizona, 2002. 106 (13). BRIGHT,J.L.,J.J.HERVERT,AND M.T.BROWN. 2001. Sonoran pronghorn 2000 OCKENFELS, R.A., C. L. DOROTHY,AND J. D. KIRKLAND. 1992. Mortality and aerial survey summary. Arizona Game and Fish Department home range of pronghorn fawns in central Arizona. Proceedings Nongame and Endangered Wildlife Program Technical Report 180. Pronghorn Antelope Workshop 15:78–92. Phoenix, USA. PATTEN, D.T. 1978. Productivity and production efficiency of an upper BRIGHT,J.L.,J.J.HERVERT,L.A.PIEST,R.S.HENRY,AND M.T.BROWN. 1999. Sonoran desert ephemeral community. American Journal of Botany Sonoran pronghorn 1998 aerial survey summary. Arizona Game and 65:891–895. Fish Department Nongame and Endangered Wildlife Program Techni- ROBINSON,W.B. 1961. Population changes of carnivores in some coyote cal Report 152. Phoenix, USA. control areas. Journal of Mammalogy 42:510–515 BROWN, D. E. 1982. Biotic communities of the American Southwest: Unit- SNOW,T.K. 1994. Sonoran pronghorn aerial survey summary, ed States and Mexico. Desert Plants 4:1–342. 1992–1994. Arizona Game and Fish Department Nongame and BYERS, J.A. 1997. American pronghorn: social adaptations and the ghosts Endangered Wildlife Program Technical Report 51. Phoenix, USA. of predators past. The University of Chicago Press, Chicago, Illinois, TRAINER, C. E., M. J.WILLIS,G.P.KEISTER,JR.,AND D. P. SHEELY. 1983. Fawn USA. mortality and habitat use among pronghorn during spring and sum- BYERS, J.A.,AND J. D. MOODIE. 1990. Sex-specific maternal investment in mer in southeastern Oregon, 1981–1982. Oregon Department of Fish pronghorn, and the question of a limit on differential provisioning in and Wildlife Wildlife Resources Report 12. Portland, USA. ungulates. Behavioral Ecology and Sociobiology 26:157–164. UNITED STATES FISH AND WILDLIFE SERVICE. 1998. Final Revised Sonoran CANON, K. S. 1993. Fawn survival and bed-site characteristics of Trans- Pronghorn Recovery Plan. Albuquerque, New Mexico, USA. Pecos pronghorn. Dissertation,Texas Tech University, Lubbock, USA. WRIGHT,R.L.,AND J. C. DEVOS JR. 1986. Final report on Sonoran prong- CARR, J. N. 1971. Endangered species investigation, Sonoran pronghorn. horn status in Arizona. Arizona Game and Fish Department, Phoenix, Arizona Game and Fish Department Project W-53–21,Work Plan 8, USA. Job 1. Phoenix, USA. DIMMITT, M.A. 2000. Biomes and communities of the Sonoran Desert region. Pages 3–18 in S. J. Phillips, and P.W.Comus, editors. A natural history of the Sonoran Desert. Arizona-Sonora Desert Museum,Tuc- son, USA. ELLIS, J. E. 1970. A computer analysis of fawn survival in the pronghorn antelope. Dissertation, University of California, Davis, USA. FOX, L. M. 1997. Nutritional content of forage in Sonoran pronghorn habitat,Arizona. Thesis, University of Arizona,Tucson, USA. GESE,E.M.,O. J.RONGSTAD,AND W. R . M YTTON. 1988. Relationship between coyote group size and diet in southeastern Colorado. Jour- nal of Wildlife Management 52:647–653. GOLDMAN, E.A. 1945. A new pronghorn antelope from Sonora. Proceed- ings of the Biological Society of Washington 58:3–4. HECHT,A.,AND P. R N ICKERSON. 1999. The need for predator management in conservation of some vulnerable species. Endangered Species Update 16:114–118. HERVERT,J.J.,J.L.BRIGHT,M.T.BROWN,L.A.PIEST,AND R. S. HENRY. 2000. Sonoran pronghorn population monitoring:1994–1998. Arizona Game and Fish Department Nongame and Endangered Wildlife Pro- Jill L. Bright is the Sonoran pronghorn projects coordinator for the gram Technical Report 162. Phoenix, USA. Yuma Region, Arizona Game and Fish Department. She received a HERVERT,J.J.,J.L.BRIGHT,R.S.HENRY,L.A.PIEST,AND M.T.BROWN. 2005. B.S. in wildlife management from Utah State University and an M.S. Home ranges and habitat use patterns of Sonoran pronghorn in from Northern Arizona University. Currently she specializes in Sono- Arizona. Wildlife Society Bulletin:8–15. ran pronghorn management, and has worked on other endangered species such as black-footed ferrets and spotted owls. John J. Hervert HOFF,D.L.,AND D. O.TRAINER. 1981. Hemorrhagic diseases of wild rumi- is a wildlife program manager for the Arizona Game and Fish Depart- nants. Pages 45–53 in J.W.Davis, L. H. Karstad, and D. O.Trainer, edi- ment in Yuma. He received his B.S. and M.S. from the University of tors. Infectious diseases of wild animals. Iowa State University Press, Arizona. His research focus is ungulate ecology in desert environ- Ames, USA. ments (25 years) and population estimation. HOSACK, D.A., P.S. MILLER,J.J.HERVERT,AND R. C. LACY. 2002. A population viability analysis for the endangered Sonoran pronghorn, Antilocapra americana sonoriensis. Mammalia 66:207–229. HUGHES, K. S.,AND N. S. SMITH. 1990. Sonoran pronghorn use of habitat in Special Section Associate Editor: Krausman