Tropical montane cloud forest is one of the most threatened ecosystems, in central Veracruz, Mexico. Within this ecosystem, terrestrial orchids are strongly dependent on forest conditions and may be sensitive to environmental change. We applied field surveys of abiotic and biotic factors associated with the presence of the terrestrial orchid Cyclopogon luteo-albus and combined this with correlative niche modeling approaches to evaluate its potential distribution under two different environmental sets and two different extents. Layers of environmental information were obtained from Landsat imagery and interpolated bioclimatic and soil property layers. Existing species records were used as training data by sampling five forest fragments in central Veracruz, utilizing herbarium and global biodiversity information facility databases. The resulting predictions were tested by sampling at 15 sites of potential distribution. The model predicted the presence of C. luteo-albus with a reliability of 80%. The most important variables of models derived from interpolated layers coincide with site-relevant parameters suggesting the utility of these tools to further explore species suitabilities at larger geographic extents. Potential distribution mapping is an important tool to identify key areas for conservation and priority areas for future studies of species and partially resolves the lack of records of many orchid species.

To obtain vouchered records of terrestrial orchids collected in central Veracruz, the database of the XAL Herbarium (Instituto de Ecología, A.C. at Xalapa, Veracruz, Mexico) was consulted. Species records that were correctly georeferenced within the altitudinal range and collected in TMCF in the region of the municipalities of Banderilla-Xalapa-Coatepec were selected, resulting in 15 locations for C. luteo-albus. Due to the limited number of collections in the herbarium, we also downloaded presence data from the public databases global biodiversity information facility (GBIF) and Instituto de Biología (UNAM). This search resulted in 25 additional unique observational records. These records were used to generate a model of the environmental suitability of C. luteo-albus across two regions of interest, one close to the area of central Veracruz where field surveys were also carried out and another for the entire extent of occurrence for the species in Mexico. Spatial analysis was performed in ArcView ver. 3.2 (ESRI, Redlands, CA)

After review of the XAL herbarium and observational records, five TMCF fragments were identified in which C. luteo-albus has been previously collected. We visited each of these sites (all sites were visited within the same week) to register richness and abundance of C. luteo-albus, following the method of Peet et al. (1998). Sampling in the study sites was timed to correspond with the flowering peak (June–September). To assess possible effects of site conditions on the presence of C. luteo-albus, a suite of site-level variables was assessed in each of five randomly distributed plots (10 × 10 m), starting with percentage of canopy openness (measured with a convex spherical densiometer, Forestry Suppliers, Inc. ®), weed cover (%), diameter at breast height (DBH) of trees > 5 cm (diameter measuring tape, Forestry Suppliers, Inc. ®).

Microsoft Excel