SELF-ORGANIZED PINNACLE PATTERNS IN ANTARCTIC LAKES: UNDERSTANDING MICROBIALITE MORPHOLOGY

Ecologic communities spatially self-organize in response to environmental feedbacks, such as nutrient availability. Microbial communities are no exception. Current stromatolite growth models explain stromatolite morphologies as a relationship between growth rate and resource availability. Models with these assumptions result in evenly spaced pinnacles. In contrast, spatial ecology studies show that positive and negative feedbacks at different spatial scales can also explain pattern formation in multiple ecosystems. Long-scale resource competition causes regular spacing of features, and short-scale mutualistic behaviors result in clustering.

Lake Vanda, a perennially ice-covered lake, McMurdo Dry Valleys, Antarctica, provides an opportunity to examine spatial self-organization in benthic microbial communities. Vanda hosts microbial mats with pinnacles that grow across well characterized gradients in irradiance, nutrients, sedimentation, and colonization age. Preliminary analyses suggest changes in pinnacle morphology and spatial patterning due to both environmental gradients and age. The spatial patterning of pinnacles from 11 sites ranging from 10.06 to 47.9 meters in depth have been analyzed using standard spatial ecology metrics. Across these sites, the average number of neighbors per pinnacle determined by Voronoi analysis was 5.95, indicating efficient packing. Spatial patterning ranges from clustered (Clarks-Evans R = 0.32, p < 0.01) to randomly spaced (Clarks-Evans R = 1.05, p < 0.01), and the three sites with the highest degree of clustering (Clarks-Evans R < 0.5) all occur near the boundaries of the mixed and convecting zones. 26 more models will be analyzed, and Bayesian modelling will be used to infer causal relationships between spatial patterning and environmental gradients, allowing for hypotheses about what short-distance feedbacks cause the observed clustering behavior.