Microb Ecol DOI 10.1007/s00248-014-0436-z
Short-Term Precipitation Exclusion Alters Microbial Responses to Soil Moisture in a Wet Tropical Forest Bonnie G. Waring & Christine V. Hawkes
Received: 2 February 2014 / Accepted: 12 May 2014 # Springer Science+Business Media New York 2014
Abstract Many wet tropical forests, which contain a quarter of global terrestrial biomass carbon stocks, will experience changes in precipitation regime over the next century. Soil microbial responses to altered rainfall are likely to be an important feedback on ecosystem carbon cycling, but the ecological mechanisms underpinning these responses are poorly understood. We examined how reduced rainfall affected soil microbial abundance, activity, and community composition using a 6-month precipitation exclusion experiment at La Selva Biological Station, Costa Rica. Thereafter, we addressed the persistent effects of field moisture treatments by exposing soils to a controlled soil moisture gradient in the lab for 4 weeks. In the field, compositional and functional responses to reduced rainfall were dependent on initial conditions, consistent with a large degree of spatial heterogeneity in tropical forests. However, the precipitation manipulation significantly altered microbial functional responses to soil moisture. Communities with prior drought exposure exhibited higher respiration rates per unit microbial biomass under all conditions and respired significantly more CO2 than control soils at low soil moisture. These functional patterns suggest that changes in microbial physiology may drive positive feedbacks to rising atmospheric CO2 concentrations if wet tropical forests experience longer or more intense dry seasons in the future. Submitted to Microbial Ecology as an Original Research Article on February 2, 2014. Associated sequences deposited in the GenBank Sequence Read Archive under Accession number SRP034845. Electronic supplementary material The online version of this article (doi:10.1007/s00248-014-0436-z) contains supplementary material, which is available to authorized users. B. G. Waring (*) : C. V. Hawkes Section of Integrative Biology, University of Texas at Austin, Austin, TX, USA e-mail: [email protected]
Introduction Because soil microbes regulate the global carbon (C) cycle via their role in the formation of soil organic matter, the specific mechanisms by which microbial communities respond to climate change may determine important ecosystem feedbacks to rising atmospheric CO2 concentrations. However, these mechanisms are difficult to isolate because microorganisms have short generation times, are phylogenetically diverse, and live in highly heterogeneous environments . Bacterial and fungal communities may respond to altered temperature or precipitation regimes via changes in microbial physiology  or community composition . Each of these responses is constrained by different biotic factors; for instance, the magnitude of physiological shifts is largely dictated by phenotypic plasticity, whereas changes in community structure are primarily limited by the diversity of organisms in the local species pool and dispersal rates from other environments. Both model simulations  and climate manipulation experiments [4, 5] predict that microbial physiological responses to climate change may alter ecosystem C cycling even in the absence of community shifts. Changes in parameters such as growth efficiency, enzyme production, or nutrient demand may all influence rates of soil C cycling. For example, elevated temperature can impact respiration rates by altering mass-specific respiration rates [2, 6]. Microbial communities can also change allocation to the production of extracellular enzymes that depolymerize soil organic matter . Changes in enzyme production or shifts in their relative abundance may alter element cycling rates and nutrient availability to plants and microbes . Such physiological adjustments not only may take place within a single microbial generation, but can also arise due to evolutionary shifts within microbial populations .
B. G. Waring, C. V. Hawkes
Changes in belowground C cycling in response to altered climate can also reflect shifts in microbial community composition. For example, shifts in the relative abundance of soil fungi and bacteria can alter biomass-specific respiration rates and nutrient mineralization . Finer-scale community changes may also impact element cycling, as evidenced by compositional differences among decomposer communities that can account for up to 20 % of the variance in C mineralization from leaf litter . If ecological trade-offs among microbial taxa are linked to traits that affect C cycling, changes in community structure will have the potential to alter soil C storage. Trade-offs between C and N acquisition  and drought tolerance and growth rate  have been identified in bacterial communities. There is some evidence that such trade-offs affect biogeochemical cycling at the ecosystem level; for example, stress imposed by drying-rewetting cycles alters community composition, which in turn influences rates of C mineralization independently of moisture [14–16]. Examining phylogenetic signals within community assemblages may provide insight into the relationship between functional traits and species identity within microbial communities . If species traits related to habitat use are phylogenetically conserved, habitat filtering can cause community members to be more clustered than expected by chance alone . In microbial communities, phylogenetically clustered microbial assemblages exhibit differential responses to variation in water availability  and soil nutrient concentrations . However, trait conservation may also lead to increased resource competition among closely related taxa, yielding communities that are phylogenetically over-dispersed . Phylogenetic over-dispersion can also result from convergent evolution of traits in distantly related taxa . Therefore, although diverse mechanisms underlie such patterns, non-random phylogenetic structure suggests variation in response and effect traits among microbial taxa. To examine both microbial community responses to climate change and identify potential mechanisms underlying those responses, we manipulated soil moisture regime in a wet tropical forest at La Selva Biological Station, Costa Rica. Although the site currently receives over 4,000 mm of precipitation a year, climate models predict that rainfall will decrease throughout Central America over the next century . In this forest, tree growth decreases strongly during drier years , but belowground responses are less certain. Therefore, we designed our experiments to address two main research goals. First, we aimed to characterize shifts in microbial abundance, community structure, and aggregate community function following experimental reduction of precipitation inputs in the field. Second, we asked whether drought-induced changes in microbial physiology and/or community structure would subsequently impact functional responses to a gradient of soil moisture.
Materials and Methods Experimental Overview To test compositional and functional responses of microbial communities to changes in water availability, we paired a field precipitation exclusion experiment with a laboratory soil moisture manipulation. In the field experiment, water availability was manipulated in situ by excluding precipitation for 6 months with a combination of shelters and trenching. The exclusion took place from September 2011 to February 2012. Although La Selva is a wet tropical forest which receives at least 100 mm of precipitation per month, historically, the period from January to April is a drier time of year (