COS 57-2 - Landsliding and rhizobiota of tree-ferns link the short- and long-term carbon cycles through silicate rock weathering

Wednesday, August 14, 2019: 1:50 PM
M111, Kentucky International Convention Center
Yakshi N. Ortiz-Maldonado, Biology, University of Puerto Rico- Rio Piedras campus, San Juan, PR, Filipa Godoy-Vitorino, Microbiology & Medical Zoology, University of Puerto Rico, San Juan, PR and Carla Restrepo, Department of Biology, University of Puerto Rico Rio Piedras, San Juan, PR
Background/Question/Methods

Landslides set in motion biogeochemical processes that have not been conceptually linked. On the one hand, landslide activity may promote chemical weathering that releases mineral-bound nutrients. On the other, rhizobiomes of plants colonizing landslides may accelerate rock weathering. Of importance is the weathering of (Ca,Mg)-bearing silicate rocks given their critical role in the short- and long-term carbon cycles. The goals of this study are two-fold: 1) characterize the rhizobiomes of tree-ferns growing in landslide and forest underlain by (Ca,Mg)-bearing silicate rocks and 2) assess their contribution to silicates weathering. In Central Puerto Rico, we located landslide and forests underlain by granodiorite, a (Ca,Mg)-silicate rock, and within each, four tree-ferns (Cyathea arborea-landslides; Cyathea pungens-forest). We collected soil from rhizosphere and bulk soil microhabitats from each individual, extracted DNA, and sequenced the 16S-V4 region. We also collected soil samples for elemental characterization of elements in the soil around each individual. Using the same sampling design, we established a long-term in-situ incubation experiment in which granodiorite tiles and pressed-powder pellets were buried to examine the role of root exclusion and root proximity on rock weathering. For this, we measured the elements concentration of the rock tiles and pellets using SEM-EDX and WDXRF spectroscopy, respectively.

Results/Conclusions

After analyzing the soil samples through WDXRF and characterizing their elemental composition, we discovered a signature of elements distinct for each habitat. After quality filtering and rarefying to even-depth, our 48 rhizosphere and bulk soil samples yielded 11,171 Operational Taxonomic Units assigned to 42 phyla. We found that the microbial species richness was greater in the forest that in the landslides. We also found that in both habitats, the rhizosphere had greater species richness than their respective bulk soils. An analysis between communities (beta-diversity) shows a clear differentiation between the microbial communities of the different habitats and microhabitats, following a similar trend to that of the soil elemental composition. After 7 months on the field, the rock tiles had a weigh difference of 0.004g in average. However, we were able to detect slight, non-significative, differences in elements concentrations’ total loss using the Ruxton ratio. Interestingly, the element loss appears to be higher on the rock tiles where we excluded root growth. Altogether, these results suggest that mycorrhizal fungi or microorganisms with a longer reach from the rhizosphere are the ones driving the highest silicates weathering rates, and thus promoting the Ca and Mg release.