The microbial mats in the streams regulate the fluvial flux of nutrients, dissolved organic matter (DOM) and coarse particulate matter (CPOM) from the landscape to the lakes (Gooseff et al 2004, Cullis et al. 2014). Thus, processes that control the persistence of these mats over varying flow influence connectivity across the landscape. We have found that the microbial mats in the streams have high diversity of bacteria and diatoms compared to other aquatic habitats and that these mats grow very slowly; the regrowth of biomass from scraped rocks takes about five summers (Kohler et al., 2015). The streamflow intermittency for each habitat can be ranked by accounting for annual and historical discharge records thereby creating a harshness index (Fritz and Dodds 2005). Streams with a higher harshness index (i.e. longer streams) host diatom communities dominated by endemic Antarctic taxa with a significant relationship between harshness and diatom endemism (Esposito et al., 2006). Long streams tend to be more intermittent and have more geochemical weathering and biogeochemical cycling. Short streams, which tend to have lower harshness indexes, are characterized by consistent hydrology, dilute solutes and high nutrient uptake, and host more widespread diatom taxa.

We are conducting a multi-year transplant experiment to explore how physical and chemical regime controls mat communities and movement of biomass and nutrients. Our two hypotheses were 1) competitive interactions will eventually result in turnover of diatom and bacterial communities, and 2) local adaptations of communities influence the adhesion of mats to substrata and their interannual persistence. The experiment involved moving microbial mat patches between short streams (<1 km; Green Creek and Bowles Creek) and long streams (>10 km; Delta Stream and Von Guerard Stream) in Lake Fryxell Basin. We moved intact mat patches from one stream to another and moved mats within the same stream as controls.

Experimental approach: In late January 2018, near the end of the austral summer flow season, 96 mat patches were transplanted among four streams, with the mats carefully secured during the transit by helicopter. Mat patches were transplanted from different growth habitats: half were ‘epipelic’ and attached to the streambed sediment surface and half were ‘epilithic’ and growing on rocks. Each stream received 4 treatments with 6 replicates where transplanted mats came from 3 other streams plus the relocation of mats within that stream. Using a total station in combination with a LIDAR survey, we surveyed each stream reach to locate each transplanted match patch. The transplant mat patches were sampled for nitrogen and carbon contents, chlorophyll-a, and bacteria and diatom community composition.  A year later mats were sampled at the beginning of the summer, December 2018, and again in January 2019. The mats will be sampled again in January 2020. Bacterial and eukaryal community composition analyses are underway.

Initial findings: One year later, we observed significant differences in the persistence of the transplanted mat patches among the four streams. The persistence of one mat patch transplanted from Delta Stream to Bowles Creek is shown in Fig. 43. Mat patches from long streams (Delta and Von Guerard) retained high coverage, averaging about 60 to 70% of the initial coverage of the transplanted rock or sediment patch, regardless of whether they were moved within the same stream or to another long or short stream. In contrast, there was low persistence (only about 5% or less) of mats from Green and Bowles Creeks moved within the same short stream (Fig. 44). By the end of summer, mat patches transplanted to the two long streams retained their coverage, and mats transplanted from Green Creek to a short stream exhibited regrowth.