Dispersal of microbes from grassland fire smoke to soils

Abstract

Wildland fire is increasingly recognized as a driver of bioaerosol emissions, but the effects that smoke-emitted microbes have on the diversity and community assembly patterns of the habitats where they are deposited remain unknown. In this study, we examined whether microbes aerosolized by biomass burning smoke detectably impact the composition and function of soil sinks using lab-based mesocosm experiments. Soils either containing the native microbial community or presterilized by γ-irradiation were inundated with various doses of smoke from native tallgrass prairie grasses. Smoke-inundated, γ-irradiated soils exhibited significantly higher respiration rates than both smoke-inundated, native soils and γ-irradiated soils exposed to ambient air only. Microbial communities in γ-irradiated soils were significantly different between smoke-treated and control soils, which supports the hypothesis that wildland fire smoke can act as a dispersal agent. Community compositions differed based on smoke dose, incubation time, and soil type. Concentrations of phosphate and microbial biomass carbon and nitrogen together with pH were significant predictors of community composition. Source tracking analysis attributed smoke as contributing nearly 30% of the taxa found in smoke-inundated, γ-irradiated soils, suggesting smoke may play a role in the recovery of microbial communities in similar damaged soils. Our findings demonstrate that short-distance microbial dispersal by biomass burning smoke can influence the assembly processes of microbial communities in soils and has implications for a broad range of subjects including agriculture, restoration, plant disease, and biodiversity.

Keywords: biodiversity; fire ecology; microbial ecology; prairie; smoke transport; soil respiration; wildfire.

Figure 1

Effect of smoke inundation on soil respiration. (A) Conceptual diagram depicting how the experimental design reflects shifts in the relative influence of soil community assembly processes in lab-isolated mesocosms, comparing native soils exposed to ambient conditions (i.e., no smoke); native soils inoculated with smoke; and pre-sterilized soils with smoke. Terms as defined in Vellend et al. (2010). (B) Schematic depiction of soil inoculation procedure and subsequent analyses. See sections 2.2–2.5 for details. (C) Soil respiration rates measured as Δ[CO2] (ppm s^-1 g soil^-1) for the first 8 days of incubation. Grey area around each line represents 95% confidence interval. (D) Sum of the respiration rates for the first 8 days of incubation. Boxes represent the interquartile range, the middle horizontal line is the median, and the whiskers represent minimum and maximum values. Outliers are 1.5 interquartile ranges below the first quartile and above the third quartile. Asterisks represent significance (^* = P < 0.05; ^**** = P < 0.0001; Tukey’s post-test).

Figure 2

Effect of smoke inundation on the chemical composition and pH of soils. Measurements of (A) biomass C, (B) biomass N, and (C) phosphate concentrations for samples taken 10 days post-inundation and (D) pH for samples taken 35 days post-inundation. Boxes represent the interquartile range, the middle horizontal line is the median, and the whiskers represent minimum and maximum values. Outliers are 1.5 interquartile ranges below the first quartile and above the third quartile. Asterisks represent significance (^* = P < 0.05; ^** = P < 0.01, ^**** = P < 0.0001; Dunn’s test [A-C]; Tukey’s post-test [D]).

Figure 3

Alpha diversity and composition of soil mesocosms. (A) Observed ASVs, Shannon, and inverse Simpson diversity indices of taxa and (B) NMDS ordination of community composition by sample type for bacterial and archaeal (16S rRNA genes) communities in soil mesocosms.

Figure 4

Microbial species composition and drivers of community assembly. (A) CAP ordination of community composition for bacterial and archaeal (16S rRNA genes) communities in soil mesocosms 10 days post-inundation constrained by soil environmental parameters. (B) Relative abundance of bacterial and archaeal (16S rRNA genes) phyla in soil mesocosms. Timepoint: 0 = time-zero controls; 1 = 10 days post-inundation; 2 = 35 days post-inundation.

Figure 5

Shared taxa between environmental compartments and source contribution to sink communities. (A) Venn diagram of shared ASVs among fuel, smoke, ambient air, and soil samples. (B) Percent of ambient air/smoke sources contribution to native and γ-irradiated soil sinks 10 days post-inundation and (C) 35 days post-inundation based on FEAST analysis. “Unknown” source represents the portion of the community that could not be attributed to either of the known sources.