PARENT SESSION
Poster Session # 21: Biogeography, Biodiversity, Populations, and Genetics.
Thursday, August 7 Presentation from 5:00 PM to 6:30 PM. SITCC Exhibit Hall B.
Local distributions of Bouteloua curtipendula varieties within their zone of sympatry.
Tomas, Andrea*,1, Gelwick, Fran1, Archer, Steve2, 1 Texas A & M University, College Station, TX2 University of Arizona, Tucson, AZ
ABSTRACT- Bouteloua curtipendula, an important component of native prairie ecosystems, consists of two varieties: var. caespitosa, a bunchgrass growth-form lacking rhizomes that reproduces apomictically; and var. curtipendula, which reproduces sexually and from spreading rhizomes. At the regional scale, var. curtipendula dominates the relatively mesic (eastern) portions of the Great Plains, with var. caespitosa occupying the more xeric (western) portions, extending into southwestern desert grasslands. We hypothesized that if the differences in these varieties reflects natural selection for aridity and/or disturbance, then within their zone of sympatry, the two growth forms would be locally segregated in accordance with variables indicative of landscape moisture/temperature, fire history or grazing gradients. To test this hypothesis, we inventoried 321 - 0.5 x 1 m plots in pastures with contrasting grazing/prescribed fire histories in an Edwards Plateau Juniper-Oak savanna site in central Texas. For each plot we recorded the % cover of each variety and 11 explanatory variables including disturbance history (fire and grazing), abiotic setting (topographic position, limestone type, soil depth, % bare ground, rock size and rock cover) and proximity to the nearest tree or shrub (source of radiant energy amelioration). Data were analyzed using canonical correspondence analysis (CCA) with forward selection of explanatory variables. Significance of variables was tested using Monte Carlo permutations. Local segregation of B. curtipendula varieties was evident, in that only 1.9% of the plots sampled contained both varieties. The rhizomatous variety was associated with Buda limestone, small rocks, no grazing and winter fire, whereas the caespitose growth form was related to Edwards limestone sites with large rocks. However, environmental and disturbance variables we measured accounted for only 18.6 % of the variation, suggesting factors other than those measured are affecting local distributions of the two varieties.
Key words: caespitose/rhizomatous varieties, Bouteloua curtipendula, local distribution, canonical correspondence analysis