James C. Devos

Evaluation of Factors Potentially Influencing a Desert Bighorn Sheep Population

Abstract We studied a desert bighorn sheep (Ovis canadensis) population in the Mazatzal Mountains (primary study area) in central Arizona and population indices on reference areas between 1989 and 2003. We evaluated disease exposure and nutritional status of desert bighorn sheep, vegetation parameters, predator diets, and mountain lion (Puma concolor) harvest and abundance (1999–2003) and mountain lion predation (1995–2003) as factors potentially affecting desert bighorn sheep and population parameters. We measured rainfall monthly, monitored demography and relative abundance of desert bighorn sheep using aerial surveys, captured and placed radio collars on desert bighorn sheep, and collected samples of blood, parasites, and other pathogenic agents from captured animals. We measured mineral content, relative use, and structural composition of vegetation and determined diets of desert bighorn sheep adults and lambs, dietary intakes of nitrogen (FN), 2,6-diaminopimelic acid (FDAPA), neutral detergent fiber, and minerals using fecal analyses. We incorporated mountain lion reductions as an experimental element, monitored harvest, and used track surveys as an index of relative abundance of the predator and monitored radio-collared desert bighorn sheep to determine mortalities and causes of death. We determined diets of bobcats (Lynx rufus), coyotes (Canis latrans), and mountain lions using fecal analyses. Drought conditions occurred during summer (July–September) and winter (November–April) during 4 and 3 years, respectively, between 1999 and 2003. Annual surveys indicated that the Mazatzal Mountains population declined during drought between 1994 and 1997, experienced low growth and lamb production coincident with above-normal rainfall in 1998 and drought in 1999, and exhibited higher growth, production, and productivity during 2000–2003 despite persistent drought conditions during this period. We observed no clinical symptoms of disease in radio-collared desert bighorn sheep, and hematological and other evidence of exposure to disease agents was unremarkable. Population indices on the primary study and reference areas were positively correlated with winter (November–April) rainfall. We found no evidence of forage overutilization on the primary study area. Rainfall on Mazatzal Mountains was associated with differences in primary production, particularly of forbs, forage mineral concentrations, and diets, nutritional status, and demographic attributes of desert bighorn sheep between 1999 and 2003. Higher winter rainfall was associated with higher forb growth, and higher rainfall was associated with higher concentrations of P and Se but lower levels of Fe in browse; higher concentrations of Ca, P, and Zn in forbs; and higher levels of P, Se, and Zn in grasses. Narrower mean Ca:P ratios of browse and forbs were associated with higher rainfall. Diets of desert bighorn sheep adults and lambs generally were similar, particularly near summer, and forbs tended to predominate in diets during wetter and drier years. Higher winter rainfall was associated in adult feces with more prolonged winter-to-spring increases in FN and FDAPA concentrations, higher fecal phosphorus, lower fecal Ca levels, and narrower fecal Ca:P and Na:K ratios, but levels of fecal Na increased during the driest year. Higher winter rainfall corresponded in lamb feces with higher levels of FN, FDAPA, and fecal P; lower concentrations of fecal Ca; and narrower fecal Ca:P ratios. Thus, we hypothesized that diets and nutritional status of desert bighorn sheep adults and lambs tended to correspond with rainfall patterns and associated differences in relative abundance and mineral content of forages. We found no evidence that bobcats or coyotes preyed on or scavenged desert bighorn sheep. Decline of desert bighorn sheep abundance during 1994–1997 was greater than declines on reference areas lacking mountain lions despite continually higher, and a lesser decline in, winter rainfall on the primary study area. In comparison, population indices on a reference area and on Mazatzal Mountains increased between 1999 and 2003 in association with predator reductions and lower abundance of mountain lions and predation of radio-collared animals despite continued occurrences of drought during this period. We thus identified 2 proximate factors that likely acted to influence demographic trends of the Mazatzal Mountains desert bighorn sheep population: nutritional status (higher rainfall [ultimate factor] was associated with higher availability and differences in mineral content of forages and improved indices of desert bighorn sheep nutritional status) and predation by mountain lions. We hypothesize that nutritional status and mountain lion predation during a period of drought influenced desert bighorn sheep population parameters in Mazatzal Mountains and that short-term removal of mountain lions by lethal harvest contributed to higher growth and productivity of the small, isolated population, even during periods of drought.

Habitat use and survival of Sonoran pronghorn in years with above-average rainfall

Abstract Sonoran pronghorn (Antilocapra americana sonoriensis) occur in an arid desert in northwestern Sonora, Mexico and southwestern Arizona; their numbers have declined since 1920. Causes of the decline are largely speculative but include habitat alteration by humans, excessive herbivory by domestic livestock, illegal harvest, and exposure to domestic livestock diseases. We initiated this study to develop information on habitat use and survival to aid in developing recovery strategies for Sonoran pronghorn. We captured 19 Sonoran pronghorn and monitored their movements and habitat use patterns from 1983 through 1991 and compared these locations to random locations to assess habitat use. We also calculated survival. Sonoran pronghorn used the creosote bush (Larrea tridentata)–white bursage (Ambrosia dumosa)–paloverde (Parkinsoniaspp.)–mixed cacti association greater than expected. Locations of Sonoran pronghorn close to water occurred more frequently than random locations, and areas farthest from waters were used less than expected. Sonoran pronghorn used areas within 1 km of roads less than random locations and used locations >1 km from roads as expected or greater than expected as distance from roads increased. Mean annual estimated survival for females and males was 0.96±0.04 and 0.92±0.04, respectively. Sonoran pronghorn had very low mortality rates and the population expanded to levels that likely have not occurred since the 1920s.

Mountain Lion Predation of Translocated Desert Bighorn Sheep in Arizona

Abstract We analyzed data for 422 unmarked and 369 radiocollared desert bighorn sheep (Ovis canadensis) translocated into vacant historical habitats in 12 Arizona locations between 1979 and 1995. We evaluated factors potentially influencing predation of radiocollared desert bighorn sheep by mountain lions (Puma concolor) by determining relationships between predation rates, number released, size of releases, escape terrain, available terrain (escape terrain as a percentage of area with slopes ≥40%), habitat quality associated with release locations, and mule deer (Odocoileus hemionus) and predator abundance. We hypothesized that numbers of radiocollared animals released, quality of habitat and available terrain associated with release locations, and relative abundance of mule deer influenced predation of translocated desert bighorn sheep by mountain lions.

POPULATION GENETIC ANALYSIS OF SONORAN PRONGHORN (ANTILOCAPRA AMERICANA SONORIENSIS)

Abstract The endangered Sonoran pronghorn (Antilocapra americana sonoriensis) consists of only 2 small populations, 1 in Arizona and 1 in Mexico. Mitochondrial DNA sequence data and 5 microsatellite loci were used to compare levels of genetic diversity and differentiation between the 2 Sonoran pronghorn populations and between this subspecies and other selected pronghorn populations, both proximate and distant to the Sonoran pronghorns current range. The data support a history of recent isolation of Sonoran pronghorn populations from those in Arizona, New Mexico, and Texas. Loss of genetic diversity due to bottlenecks and drift has been severe in both populations of the Sonoran pronghorn, but is most pronounced in the Arizona population. These results are discussed in terms of conservation management strategies.

Bighorn Sheep Selection of Landscape Features in an Active Copper Mine

Abstract Bighorn sheep (Ovis canadensis) and Dalls sheep (O. dalli) use mining areas despite high human activity. We studied bighorn sheep selection of landscape features within a mine in a desert environment to determine those important for bighorn sheep and to enhance reclamation efforts of mines in desert environments. We collared and observed 8 male and 6 female bighorn sheep from December 2003 to January 2005 in the Silver Bell Mountains, Arizona, USA. We classified 13 unique features within a mine landscape based on topography, hydrology, and vegetation. Subadult male, adult male, and female bighorn sheep used desert islands (54, 76, and 54%, respectively) followed by highwalls (14, <10, and 11%, respectively). Subadult male, adult male, and female bighorn sheep selected for desert islands, and subadult males and females also selected for highwalls, whereas adult males did not. We did not observe bighorn sheep using leach ponds, pit bottoms, or tailings dumps. Bighorn sheep behavior while on desert islands was similar to behavior recorded when outside the mine perimeter. Subadult male and female bighorn sheep fed and were alert less and socially interacted more on highwalls than they did while outside the mine perimeter. Bighorn sheep in mines select areas similar to conspecifics outside of mined areas. In areas where mining and bighorn sheep are in proximity, mining engineers and wildlife biologists should work together to design reclamation plans that benefit bighorn sheep. In places where revegetation is difficult (i.e., deserts), mine engineers should design infrastructure (i.e., roadways, waste dumps, buildings) to minimize the unnecessary destruction of native slopes and vegetation.

Population Dynamics and Behavior of Bighorn Sheep with Infectious Keratoconjunctivitis

Abstract Introduced disease is a major mortality factor in some populations of bighorn sheep (Ovis canadensis). Epizootics of infectious keratoconjunctivitis (IKC) and contagious ecthyma occurred in bighorn sheep in the Silver Bell Mountains of south-central Arizona, USA, from 1 December 2003 to 31 March 2004. Our objectives were to 1) investigate the influence of the epizootic on abundance and demographics and 2) examine how IKC affected the mortality, behavior, and movements of clinically affected animals. Morbidity was 39%, and all sex and age classes were affected. The population declined 23%, with most mortality in the adult female (1 M, 11 F) segment of the population. Of the diseased animals that were marked (n = 27), 44% recovered and 44% died. Predation (50%) and starvation (33%) were the primary causes of mortality of diseased bighorn sheep. Bighorn sheep that were infected spent less time feeding and moved less than noninfected animals during the epizootic. Managers might be able to minimize losses of infected animals through predator control. To minimize losses to starvation, managers should refrain from any activity that disturbs infected animals (including treatment) because disturbances increase energy expenditures and expose infected animals to injury.

Probability of detecting band-tailed pigeons during call-broadcast versus auditory surveys

Abstract Estimates of population trend for the interior subspecies of band-tailed pigeon (Patagioenas fasciata fasciata) are not available because no standardized survey method exists for monitoring the interior subspecies. We evaluated 2 potential band-tailed pigeon survey methods (auditory and call-broadcast surveys) from 2002 to 2004 in 5 mountain ranges in southern Arizona, USA, and in mixed-conifer forest throughout the state. Both auditory and call-broadcast surveys produced low numbers of cooing pigeons detected per survey route (x̄ ≤ 0.67) and had relatively high temporal variance in average number of cooing pigeons detected during replicate surveys (CV ≥ 161%). However, compared to auditory surveys, use of call-broadcast increased 1) the percentage of replicate surveys on which ≥1 cooing pigeon was detected by an average of 16%, and 2) the number of cooing pigeons detected per survey route by an average of 29%, with this difference being greatest during the first 45 minutes of the morning survey period. Moreover, probability of detecting a cooing pigeon was 27% greater during call-broadcast (0.80) versus auditory (0.63) surveys. We found that cooing pigeons were most common in mixed-conifer forest in southern Arizona and density of male pigeons in mixed-conifer forest throughout the state averaged 0.004 (SE = 0.001) pigeons/ha. Our results are the first to show that call-broadcast increases the probability of detecting band-tailed pigeons (or any species of Columbidae) during surveys. Call-broadcast surveys may provide a useful method for monitoring populations of the interior subspecies of band-tailed pigeon in areas where other survey methods are inappropriate.

Managing wildlife by ballot initiative: The arizona experience

Abstract Ballot initiatives are being used frequently to change wildlife management practices. Arizona has experienced three initiatives in the last decade: two related to anti‐trapping and one a funding initiative for wildlife and parks management. In 1991, a public initiative was started to prohibit trapping on public lands. The wording was vague and potentially provided a basis for a legal challenge to hunting and fishing. Efforts to defeat the initiative focused on the vague language. Upon defeat, the proponents clarified the language to prohibit lethal trapping methods and to permit hunting and angling, and the initiative passed. Based largely on data from a human dimensions survey, the Arizona Game and Fish Department did not oppose this initiative. Recognizing that funding was insufficient for agency requirements, the department developed plans to provide alternative funding. Most attempts failed in legislative hearings. The department, in concert with consumptive and non‐consumptive nongovernmenta...

Influence of wildfire and coyote presence on habitat use by collared peccaries

Abstract Collared peccaries (Pecari tajacu) occur in the southwestern United States where fires can severely damage slow-growing, drought-resistant vegetation. Collared peccaries are herbivores and depend upon vegetation for food and cover. Coyotes (Canis latrans) are sympatric with collared peccaries and, in some areas, are significant predators of collared peccaries. Coyotes can affect prey populations directly by killing individuals and indirectly by influencing behavior, habitat use, and activity patterns. We evaluated the habitat use of collared peccaries in burned and unburned areas without coyotes present in 1998 and both with and without coyotes present in 2000. Collared peccaries used vegetation associations in the enclosure nonrandomly (P<0.10), with less use of burned areas than unburned areas. However, coyotes did not appear to influence habitat use of collared peccaries in the enclosure.

Diurnal Bed-Site Selection of Urban-Dwelling Javelina in Prescott, Arizona

Javelina (Tayassu tajacu) are an economically important big game species in the southwestern United States, but are considered nuisance wildlife in urban areas. We radiotracked individuals from 6 herds to identify diurnal bed sites of urban-dwelling javelina in 1992-93. We compared microsite habitat characteristics at 46 diurnal javelina bed sites with the same characteristics at 49 random plots. Characteristics of bed sites included south- and east-facing sites with >20% slope, >50% shrub cover, and chaparral habitat with >75 cm dbh trees, javelina also selected sites where visibility was completely obstructed by dense vegetation and-or rocks within 3-4 m.