With Anglo settlement of southern Arizona in the late 1800s, fire regimes were altered dramatically. Currently, desert scrub has experienced a dramatic rise in frequency and size of wildfire which may remove fire-intolerant perennials that define the community structure. On the contrary, woodlands and forests now contain excessive fuel build up and require prescribed burning to restore the system's balance. These opposing attitudes reflect the conundrum of how to manage naturally occurring fires, or where and when to initiate prescribed burning to recreate each vegetation community's "natural" fire regime. To resolve this dilemma, there is need to discern natural fire histories by collecting specific baseline data such as where, when and how the landscape has burned. This research has been focused on the design and development of a relational database, and the incorporation of a geographic information system to develop a system that can be used to fulfill this need.

Little is known about habitat selection by elf owls (Micrathene whitneyi) and western screech-owls (Otus kennicottii). From 1994 to 1996 in the Sonoran Desert, I used point counts and nest searches to examine habitat selection by both species at multiple spatial scales. The abundance of both species had a positive association with percent cover of washes and mesquite (Prosopis spp.) at the scale of the study area. At both the scale of the study area and the nesting area, elf owls selected areas with high densities of mature saguaros (Carnegiea gigantea) and saguaro cavities. Elf owls nested only in woodpecker cavities in saguaros, whereas western screech-owls nested in both saguaro cavities and in natural cavities in mesquite. Western screech-owls nested nearly exclusively in gilded flicker (Colaptes chrysoides) cavities when they nested in saguaros. Patterns of nest cavity selection by elf owls suggest they may choose cavities that provide thermoregulatory advantages. I give management recommendations based on my findings.

I studied bat use of 3 isolated water developments on the Cabeza Prieta National Wildlife Refuge in southwestern Arizona between May 1995 and August 1997. I recorded bat echolocation calls to measure overall activity, videotaped bat behavior and identified aquatic insects to determine whether bats were feeding or drinking, mist netted to assess visitation patterns with respect to season, sex, and reproductive condition for each of 4 species captured, studied movements of banded individuals, and monitored roosts of the California leaf-nosed bat (Macrotus californicus) in nearby mines. Echolocation activity was much higher at water than in nearby dry desert washes and was greater in dry washes than at random sites away from water. Bats visited water in all seasons, visiting primarily to drink, not to feed on insects. The California leaf-nosed bat constituted 41% of more than 1,000 captures. I concluded that in my study area this species made extensive use of water developments for drinking, particularly during lactation.

On the Colorado Plateau, environments occupied by Mexican spotted owls (Strix occidentalis lucida) receive intense levels of recreational activity that could adversely effect their behavior and reproductive success. Any spatial restrictions on recreational activities in these narrow canyons would likely eliminate all recreational activity within a given canyon. We assessed changes to activity budgets and nest attendance caused by hikers (Chapter 1) and examined factors that influenced flush responses of roosting owls to hikers (Chapter 2). We conclude that high levels of recreational activity in nesting habitat may be detrimental to Mexican spotted owls and recommend a 205-m radius buffer zone around occupied nests. Based on responses of roosting owls to hikers, we established response thresholds as a basis for exploring management options that will maximize protection of owls and minimize restrictions to recreationists.

The effects of fire on vegetation in the upper encinal woodlands, Santa Catalina Mountains, southeastern Arizona, were studied following a wildfire in July 1995. Similar unburned and burned sites were compared to determine differences (alpha = .05) for frequency, density, and percent cover of herbs, shrubs, and trees through observations made during early June of 1997 and 1998 (two and three years postfire). For herbs, which showed their opportunistic characters, differences existed for frequency and percent cover each year and density in the third year. For shrubs, due to the ability of the dominant species, Arctostaphylos pungens (pointleaf manzanita), to rapidly reseed itself, no differences existed for frequency or density each year; but total mortality of mature A. pungens resulted in differences for percent cover each year. For trees, due to the ability of Quercus spp. to vigorously stump sprout, no differences existed for frequency, density, or percent cover each year.

Tree-ring records were used to reconstruct the spatial and temporal patterns of western spruce budworm (Choristoneura occidentalis Freeman) outbreaks in the Rio Grande National Forest (RGNF) of southern Colorado. Reconstructions at 11 host stands showed a synchronous pattern of outbreaks with a peak in the number of trees recording outbreaks over the entire RGNF on average every 24 years. These synchronous periods of outbreaks coincided with periods of increased moisture as indicated by an independently reconstructed summer Palmer Drought Severity Index, while relatively few trees recorded outbreaks during dry periods. The reconstruction on the RGNF does not support the hypothesis that human land use has significantly altered outbreak patterns. Tree response to outbreaks in the RGNF was different from prior studies as reductions in growth were typically detectable only when growth was compared to that of nonhost tree species.

This study is the first comprehensive survey of a frog assemblage in the central Amazon. Forty-three species of frog were found in a 2,000 hectare tract of primary forest near Manaus, Brazil. Breeding habitat affiliation, breeding phenology, spatial distribution, and relative abundance were determined for most species principally by counting calling males along variable-width strip transects. Frogs were also censused with visual detection methods and in litter plots. However, these techniques were not as powerful as audio transects for assessing species composition, habitat occupancy, and abundances throughout a large area. Anuran species and generic richnesses are similar throughout the Amazon except for Eleutherodactylus which is depauperate in the central and lower Amazon. As many species were abundant in the study area as were rare. The distribution of breeding habitat abundance parallels the distribution of species-abundance and so breeding habitat availability seems the best candidate to explain the relative abundances of species in the study area. / There is pervasive correlation between life history variables, particularly reproductive mode and habitat affiliation, and phylogeny in tropical forest anurans. This association supports a hypothesis that, at a regional scale, the history of an area's colonization and speciation rates of the colonists influenced the distribution of habitat use by frogs in forest assemblages more than selection. / Litter plot sampling was biased in favour of conspicuously behaving litter species. Since the proportion of the litter fauna that is conspicuously behaving is much lower in southeast Asia than the Neotropics, there is no evidence that litter frogs per se are more abundant in the Neotropics. However, abundances of conspicuously behaving species are higher in the Neotropics. Practically all abundant species at Manaus undergo terrestrial development. It appears that terrestrial reproduction releases frog populations from dependence on limited aquatic habitat. Since two prominent Neotropical families of litter frog reproduce terrestrially whereas terrestrial development is not associated with any southeast Asian taxon, the historical colonization of these regions by different lineages explains inter-regional abundance differences most parsimoniously. / Source: Dissertation Abstracts International, Volume: 52-04, Section: B, page: 1904. / Major Professor: Daniel Simberloff. / Thesis (Ph.D.)--The Florida State University, 1991.

The tropical forest of Guatemala's 21,130 square kilometer Maya Biosphere Reserve and buffer zone is being impacted by deforestation due to an increase of the local population and establishment of over 200 new settlements over the last 20 years. Existing geographic information system databases and remote sensing data were used to determine how much of the observed deforestation could be explained by three factors: roads, human settlements, and soil quality. Each factor was analyzed separately using spatial and statistical analysis methods. These factors were then combined to create a final deforestation risk model. The deterministic model enables policy makers, as well as managers, to create scenarios that assess the impact of their actions on the forest on a regional scale.

Although twin-spotted rattlesnakes (Crotalus pricei ) were first identified more than a century ago, little has been published about the ecology of this species and no information has been available regarding population characteristics. During 1997--99, I captured, measured, and marked 117 C. pricei in the Chiricahua Mountains to address this need for information. Lizards constituted the bulk of C. pricei prey, but their diet also included mammals, birds, and snakes. Mating and parturition were concentrated in August and early September. Movement patterns varied dramatically from year to year, as males moved substantially farther during the 1998 monsoon season (July--September) than in 1997. Crotalus pricei are sometimes collected illegally for the pet trade. About 90 people may have hunted for C. pricei at a well-known site during 1997--98. On average, snakes at this site were 38.1 mm smaller than snakes at other sites, an observation that could be attributed to collecting pressure.

The southernmost known North American stand of corkbark fir ( Abies lasiocarpa var. arizonica (Merriam) Lemm.) is found in the Santa Catalina Mountains just north of Tucson, Arizona. Climate response, age distribution, and fire history were studied in this small corkbark fir stand to provide baseline information for future management. Response function analysis indicated April--June precipitation from the current growing season, April--June temperature from the current growing season, November--March precipitation prior to the growing season, and August--October precipitation from the previous growing season as the most highly correlated factor with ring-width variance. Age distribution appeared to be steady state. A fire chronology developed for the corkbark fir site was used to test synchroneity of fire events with previously developed chronologies from nearby sites. Chi-squared analyses indicated significant association of fire years for all sites but not spread of fire from one site to another.