Title

Bacterial diversity in soil as a function of soil texture

Date of Completion

January 2009

Keywords

Hydrology|Biology, Microbiology|Engineering, Environmental

Degree

Ph.D.

Abstract

Bacterial diversity in soil exceeds by orders of magnitude that found in oceans and others compartments of the biosphere. Strong evidence suggests that spatial isolation imparted by fragmented aquatic microhabitats in unsaturated soil plays a large part in creating this diversity. Furthermore, since soil bacteria depend on water for hydration and diffusion of nutrients, examination of the hydrologic conditions at relevant spatial scales in soil is important. I evaluate the role of soil texture, which determines the extent and connectivity of microhabitats, in constraining the development of soil bacterial communities. A range of soil samples of varying textures was collected from sixteen locations across Connecticut and Massachusetts. Soil particle size distributions were measured, and samples were tested for chemical characteristics (e.g., pH, %C, %N) that might influence diversity. T-RFLP analysis was performed to evaluate the richness, diversity and composition of bacterial communities in the samples. Bacterial species richness significantly increases (p = 0.037) with the coarseness of the soil, quantified as % sand. No trend in Shannon's diversity index H' was observed; all sample communities were highly diverse. The increase in species richness in coarser soils is likely due to the increased number of isolated water films in soils with larger pores, suggesting that pore-scale hydrologic regime constrains bacterial richness in soil. Both clustering and matrix regression failed to show an effect of soil texture on bacterial community composition, as measured by Jaccard distance between communities. The lack of effect of texture shows that pore-scale hydrologic regime has little influence on the identities of the bacteria that inhabit the soils. Only copper content was found to significantly influence community composition, an effect that is corroborated by other studies. Finally, a lattice Boltzmann modeling study was conducted to elucidate the mechanisms by which soil texture controls water-phase fragmentation in soils of varying textures. Results confirm differential mechanisms controlling fragmentation in fine and coarse soils, but scale of model (1 mm3 106 nodes) was too coarse to resolve fragmentation in water films. Estimates of necessary spatial resolution and required computing resources are given. The overall findings of the three studies indicate that texture-induced fragmentation of the water phase is a significant factor shaping the richness but no composition of bacterial communities in soil. ^