Effects of Grazing on Lizard Abundance and Diversity in Western Arizona

Abstract

-Lizard populations sampled on heavily grazed chaparral, desert grassland, mixed riparian scrub, and cottonwood-willow vegetative communities were characterized by lower relative abundance and species diversity indices than those of similar, lightly grazed sites. Lower relative abundance and diversity at heavily grazed sites were a result of vegetative structures that favored lizards who forage while roosting on trees, downed tree limbs, or under trees. Lightly grazed sites were characterized by vegetative structures that favored a wide variety of foraging styles. There were no differences in lizard abundance and diversity between lightly and heavily grazed Sonoran Desertscrub. Similarities in lizard abundance and diversity of lightly and heavily grazed Sonoran Desertscrub were attributed to similarities in vegetative structure. Many natural factors limit species distribution-competition, topography, climate, and vegetation to name a few. Recently we have become concerned with the way in which human activities, such as off-road vehicle use, road construction, and livestock grazing may affect species distribution. Few studies have addressed these effects on lizards. Bury and Busack (1974) found an inverse relationship between sheep grazing and lizard population density: an ungrazed study plot had twice the number of lizards and three times the biomass of a grazed plot. They related reduction in biomass on the grazed plot to the loss of cover, loss of social structure, invertebrate faunal degradation, and direct casualties. Although Bury and Busack (1974) related grazing-caused vegetative degradation to declines in lizard biomass, they made no attempt to relate the decline in numbers of individual species to changes in vegetative structure. Pianka (1966) discussed the importance of vegetative structure in determining the diversity of lizards in a given vegetative community. In most cases he found vegetative communities with increased plant structures supported more lizard species than those with fewer plant structures. He correlated this finding to increases in both structure of vegetative communities and foraging niches. Based on vegetative structural and foraging requirements, Pianka placed lizards into foraging styles: widely foraging species; sit-and-wait species who forage while sitting on rocks, trees and downed tree limbs; sit-and-wait species who forage in open spaces between shrubs; herbivorous species; and nocturnal species. Studies on the effects of livestock grazing on vegetation have been related primarily to changes in plant species composition. Ellison (1960), Laycock (1967), Potter and Krenetsky (1967), Brown and Schuster (1969), Turner (1971), and Blydenstein et al. (1957) indicated that heavy livestock use reduced biomass and diversity of annual forbs and grasses and changed composition of shrub species. However, there have been no studies relating plant compositional changes, due to livestock grazing, with vegetative structure. T E SO T ESTERN NATURALIST 26(2):107-115 MAY 21, 981 This content downloaded from 157.55.39.186 on Sun, 09 Oct 2016 06:27:18 UTC All use subject to http://about.jstor.org/terms The Southwestern Naturalist The objective of this study was to document how vegetative structural changes observed under lightly grazed as compared to heavily grazed situations affect lizard species abundance and diversity. Concern of various interest groups on the effects of human activities on vertebrate species has led to legislation dictating evaluation of these activities and their effect on wildlife (ie., National Environmental Policy Act). Due to these mandates, and other factors, the Bureau of Land Management, Phoenix District, conducted inventories in western Arizona to determine the effects of grazing on lizards. METHODS.-Field work was conducted in three planning areas of the Bureau of Land Management's Phoenix District: Black Canyon, Hualapai-Aquarius, and Lower Gila North (Fig. 1), encompassing approximately three million acres of federal, state, and patented lands. Fourteen sample sites, seven heavily and seven lightly grazed, were established in each of five vegetative communities: chaparral, desert grassland, mixed riparian scrub, cottonwood-willow riparian, and Sonoran Desertscrub (Brown and Lowe, 1974). Chaparral and desert grassland occurred at elevations from 1143 to 1571 m, with the former occurring on broken, hilly topography and the latter on mesa tops where slopes were less than 10°. Precipitation ranged from 0.31 to 0.41 m annually at chaparral and desert grassland sites. Sonoran Desertscrub occurred at elevations from 343 to 657 m where precipitation ranged from 0.18 to 0.26 m. Topography of Sonoran Desertscrub communities ranged from slopes in excess of 45° on mountain terrain to valleys of less than 10°. Mixed riparian scrub and cottonwood-willow riparian communities occurred in sand bottom drainages with the former at elevations from 314 to 629 m, annual precipitation from 0.18 to 0.26 m, and no surface water, and the latter at elevations from 571 to 1429 m, annual precipitation ranging from 0.18 to 0.41 m, and water at or near the surface. While mixed riparian scrub communities were found exclusively in open, broken topography, many of the cottonwood-willow sites were in deep, narrow canyons. Heavily grazed sites were characterized by existence of cattle trails, presence of livestock, and poor range condition (Davis, 1976). Lightly grazed sites were characterized by absence of livestock and good to excellent range condition (Davis, 1976). Deer and other wild ungulates were in low abundance throughout the study area and had little effect on range condition. Heavily and lightly grazed sites of a given vegetative community possessed similar topography, rainfall, and elevation to minimize the effects of natural environmental factors on vegetative structure. With the exception of Sonoran Desertscrub, lightly grazed sites were structurally more diverse than heavily grazed sites. Lightly grazed sites also possessed greater percentages of structure at heights of less than 1 m. Percent live vegetative cover also was greater at lightly grazed sites with the exception of chaparral and Sonoran Desertscrub communities. Dead plant cover in the form of tree limbs was greater at all heavily grazed sites with the exception of Sonoran Desertscrub. Relative lizard abundance and species diversity of lizards were estimated by use of an Array trapping method during the periods March-June and September-November 1978, and MarchOctober 1979. The Array trapping scheme (Campbell and Christman, 1977) consisted of a series of 18.3 1 buckets buried in the ground and connected by aluminum fences. Campbell and Christman (1977) used eight can traps with an open-center configuration. A four-bucket scheme was used in this study to reduce material cost and time constructing sampling equipment. Each Array consisted of a center bucket and three evenly dispersed (120°) peripheral buckets, 7.14 m from the center. Buckets were connected by 0.41 m high aluminum drift fence anchored by stakes and wire. Seventy Arrays were constructed, seven at each heavily and lightly grazed site in each vegetative community. Arrays were checked every 3 days during the study period. Lizards trapped were measured (mm), toe-clipped, sexed, and released not less than 3 m from the nearest bucket. Relative abundance was calculated as the number of lizards caught in 24 hours by one Array (lizards/Array/night). Mean relative abundance ± one standard deviation was calculated for each species on heavily and lightly grazed study sites. Total relative abundance for each heavily and lightly grazed vegetative community was compiled ± one standard deviation and tested (student's t) for difference at p < 0.05. Species diversity indices were calculated for heavily and lightly grazed sites by a modified 108 vol. 26, no. 2 This content downloaded from 157.55.39.186 on Sun, 09 Oct 2016 06:27:18 UTC All use subject to http://about.jstor.org/terms Jones-Effects of Grazing on Lizards