Papers of the Applied Geography Conferences (2009) 32: 322-331 322 EXAMINATION OF VEGETATION INTERCEPT-TRANSECT SAMPLING IN AN ARID ENVIRONMENT Anna Klimaszewski-Patterson Department of Geography New Mexico State University Las Cruces, NM 88003-8001 1. INTRODUCTION Traditionally, rapid assessments of vegetative boundaries based on geomorphology rely on interpretation of remotely sensed imagery (e.g., Landsat TM), and are performed within a matter of days. These assessments require a complementary rapid field verification to determine if plant boundaries are properly located. Determining plant boundaries requires an appropriate sampling method, with variations of intercept-transect sampling (ITS) (line, strip, point, and Parker transects) being the most common. Pioneering ITS literature focused on general vegetative quantification of densities and coverage (Canfield, 1941; Bauer, 1943; Heady et al., 1959; Gates, 1969). However, ITS can also be appropriate for field verification of vegetative boundaries and can be implemented with relatively low cost and effort (Anderson et al., 1979). ITS methods have become some of the de facto standards in vegetation sampling, with only minor implementation variations made by individual researchers (Carter et al., 1994; Fortin et al., 2000; Gregoire and Valentine, 2003; Gordon et al., 2004; Affleck et al., 2005; Alford et al., 2005; NBII, 2005; Caratti, 2006; Galbrand et al., 2007). Regarding transects for establishment of vegetative boundaries, most literature is concerned with demarcating wetland and upland regions based on distinct flora, and variously employ line transects, line-intercept sampling (LIS), and strip transects (Carter et al., 1994; Fortin et al., 2000; Galbrand et al., 2007). These methods employ frameworks configured for areas where vegetation is the dominant visible land feature. Such approaches are not appropriate for certain ecoregions, such as the Chihuahuan Desert of southern New Mexico, where barren soil is the predominant visible feature and vegetation is scarce. This article examines best practices and implementations of ITS methods for field verification of plant boundaries created using a rapid assessment methodology in an arid environment. To critically analyze ITS methods, I performed a preliminary micro-assessment of each method to strengthen my hypothesis concerning their relative applicability to vegetative boundary identification in an arid environment. Recommendations are made here for sampling rules, sampling implementation, and terminology standardizations. 2. AREA OF INTEREST The Chihuahuan Desert (Figure 1) is sparsely vegetated and has elevations ranging from 600 m to 1675 m. In southern New Mexico, the Chihuahuan Desert has grass-shrub ecotones with very clumped and patchy vegetation (Chopping et al., 2004). Grasses include black grama (Bouteloua eriopoda), blue grama (Bouteloua gracilis), mesa dropseed (Sporobolus flexuosus), red threeawn (Aristida purpurea), tobosa (Hilaria mutica), and burro grass (Scleropogon longisetus). Shrubs include mesquite (various), creosote bush (Larrea tridentata), tarbush (Flourensia cernua), and alkali sacaton (Sporobolus airoides) (DeMers et al., 2003; Chopping et al., 2004).