Impact of inundation on soil microbiology

Introduction

Based on 16S rRNA gene sequencing and other molecular methods, as well as culture-based techniques, diverse aerobic and anaerobic microbes are known to inhabit wetland ecosystems.^1–5 They contribute to essential metabolic and biogeochemical processes, such as the mineralisation of organic compounds and the cycling of carbon, nitrogen, phosphorus, sulfur, iron and other elements.^2,3,5–7 In inland floodplain wetlands that receive water primarily from riverine flows and flooding, including those in Australia, soil microbial communities may be typically dominated by members of the phyla Proteobacteria,^2,3,8,9 Acidobacteria,^2,3,9,10 Actinobacteria,^2,3,8–10 Bacteroidetes,² Verrucomicrobia² and Firmicutes.⁸ Proteobacteria is also known to be the most abundant in natural as well as constructed wetlands.¹¹ Floodplain soil microbes are subject to varying wet and dry conditions.^4,9 Observational and experimental studies have been conducted to understand the association of the microbial community structure and function with varying wet and dry conditions in floodplain wetlands.^2,6,8,12,13 Hydrological conditions (including a gradient of hydrological connectivity, water depth fluctuations and inundation frequency) are considered critically important in determining the microbial community structure and function in floodplain wetlands.^6,8,9,12 In this article, we briefly review the recent studies on the relationships of soil bacteria with hydrological conditions in floodplain wetlands including those in Australia, with a focus on the bacterial relative abundances, taxonomic diversity and functional diversity.

Relative abundances

There are studies that explicitly report the relationships between relative abundances of soil bacteria and hydrological conditions in floodplain wetlands^2,8 including those in semi-arid Australia.⁹ Although these studies were conducted in different geographical regions and have dealt with different aspects of hydrological conditions, the soil bacteria show consistent taxon-specific patterns in terms of how their relative abundances change with increasing wet (or dry) conditions. They can be categorised into positive, negative and neutral (independent) groups in terms of their relative abundances in communities increasing with increasing wet conditions.⁹ For example, the positive-response group of bacteria includes Bacteroidetes,⁸ Chloroflexi,⁸ Euryarchaeota,⁸ Firmicutes⁸ and Proteobacteria.^8,9 The negative-response group of bacteria includes Acidobacteria,^2,8 Actinobacteria^2,3,8,9 and Planctomycetes.⁸ The neutral-response group of bacteria includes Cyanobacteria.⁹

Increased wet conditions are likely to lead to more anaerobic conditions than aerobic conditions in floodplain soils, favouring anaerobic taxa over aerobic taxa in microbial communities.^2,8 In the positive-response group, Bacteroides is obligately anaerobic.¹⁴ Many members of the Proteobacteria and Chloroflexi are obligately or facultatively anaerobic,^15,16 although the members of Euryarchaeota and Firmicutes may be more heterogeneous in terms of oxygen preference.^17,18 In the negative-response group, many members of Actinobacteria, Acidobacteria and Planctomycetes are aerobic.^19–21 In the neutral-response group, Cyanobacteria are capable of anoxic, anoxygenic photosynthesis^22,23 and are likely to grow even in increased wet conditions. In addition, Cyanobacteria can exist in the form of dormant cells, such as akinetes, during dry times.^24,25 Such physiological adaptations by Cyanobacteria may render the presence of Cyanobacteria independent of hydrological conditions (e.g. inundation frequency) in floodplain soils, relative to other members of the microbes (Fig. 1).⁹

Fig. 1.

Examples of dry aquatic environments in floodplain wetlands of the Macquarie Marshes, Australia (a), with associated elevation (b), inundation frequency (c) and logit-transformed relative abundances of the floodplain Proteobacteria (d), Actinobacteria (e) and Cyanobacteria (f) against inundation frequency from the period 1979–2006.³⁷ Fitted regression lines for Proteobacteria and Actinobacteria (d, e) are significant at P = 0.05.⁹

Because floodplain wetlands are comprised of a suite of landforms that are geomorphologically dynamic and underpin a mosaic of aquatic habitats,^26,27 small, spatial heterogeneity in hydrological conditions (e.g. due to elevational differences within <1-km distance) can produce taxon-specific effects on their relative abundances of soil bacteria in floodplain wetlands.^4,9

Taxonomic diversity

The fluctuations in hydrological conditions are known to affect the diversity of floodplain soil bacteria. For example, Peralta et al.²⁸ have shown that bacterial communities originating from fluctuating hydrological conditions were more diverse and capable of greater changes in community composition in response to changes in soil moisture than bacterial communities shaped by stable, less dynamic, hydrological conditions (i.e. more permanently wet or dry conditions). Shen et al.⁴ also reported that the diversity of soil bacteria was high at middle elevation sites (corresponding to sites of intermediate flooding duration), relative to high and low elevation sites. They stated that the observed curvilinear relationship between the soil bacterial diversity and flooding disturbance can be explained by Connell’s ‘intermediate disturbance hypothesis’.²⁹ Flooding or inundation is thought to be a reasonably common and important disturbance type in floodplain wetlands including those in semi-arid Australia.^11,30,31 Interestingly, Rayburg et al.³¹ have shown that semi-arid floodplain sites subject to an intermediate level of flood disturbance also have a greater level of diversity in soil properties than those sites subject to frequent flood disturbances.

Functional diversity

Changes in hydrological conditions also influence the functional diversity of floodplain soil bacteria, by affecting soil redox, nutrient and carbon environments as well as grain size distributions in topsoil.^32,33 For example, they may allow the coexistence of nitrifying and denitrifying bacteria in floodplains following inundation, with additional functional pathways of nitrogen cycling.^32,34,35 In terms of carbon cycling, Limpert et al.³ reported the changes in the rates of carbon dioxide (CO₂) and methane (CH₄) emissions in a semi-arid Australian floodplain wetland before, during and after inundation. Overall, the total carbon emissions in the form of CO₂ and CH₄ are reduced during and after wetting due to increasing vegetative growth and subsequent CO₂ sequestration.

By using Biolog EcoPlates (Biolog, Inc., Hayward, CA, USA), Wilson et al.¹² measured the diversity of carbon substrate use by Australian floodplain soil bacteria. They reported that initially following flooding, a diverse range of carbon substrates was used by floodplain soil bacteria. However, after 24 days there was no significant difference between the carbon substrates used in flooded and non-flooded samples. They suggest that the initial rapid increase in substrate utilisation is due to opportunistic growth of r-selected species responding to a change in their immediate environmental condition, which was followed by growth of K-selected species. Similarly, using Ecoplates, Rana et al.¹³ investigated the intensity and diversity of carbon substrate use by soil bacterial communities in areas exposed to different historical inundation frequencies (moderate, low and very low areas) in semi-arid Australia. They detected the highest microbial activity from the soil bacteria community in a very low inundation frequency area. The diversity of carbon substrates used also differed among communities initially following inundation but became similar diversity levels after ~1 week of inundation.

Conclusion

The structure and function of soil bacteria are intrinsically related to hydrological conditions in floodplain wetlands. There are consistent response patterns among soil bacterial communities with changing hydrological conditions in floodplain wetlands, despite their complex topography, landscape and aquatic habitats. Considering hydrological changes as a form of disturbance, Connell’s ‘intermediate disturbance hypothesis’ has been used to explain the observed bell-shaped response of soil microbial communities with varying hydrological conditions.²⁹ There is inherent difficulty in clearly defining what is meant by an ‘intermediate’ disturbance due to ecosystem-specific reference conditions such as average, minimum and maximum values.³⁶ Nevertheless, further application and testing of general ecological theories and hypotheses may help advance our understanding and predictive modelling capability for the dynamics of floodplain soil bacterial communities with changing hydrological conditions.

Data availability

Data sharing is not applicable as no new data were generated or analysed during this study.

Conflicts of interest

The authors declare that they have no conflicts of interest.

Declaration of funding

This study did not receive any specific funding.