A major challenge in ecology is to understand how different species interact to determine ecosystem function, particularly in communities with large numbers of co-occurring species. We use a trait-based model of microbial litter decomposition to quantify how different taxa impact ecosystem function. Further, we build a novel framework that highlights the interplay between taxon traits and environmental conditions, focusing on their combined influence on community interactions and ecosystem function. Our results suggest that the impact of a taxon is driven by its resource acquisition traits and the community functional capacity, but that physiological stress amplifies the impact of both positive and negative interactions. Further, net positive impacts on ecosystem function can arise even as microbes have negative pairwise interactions with other taxa. As communities shift in response to global climate change, our findings reveal the potential to predict the biogeochemical functioning of communities from taxon traits and interactions.

Understanding microbial adaptation is crucial for predicting how soil carbon dynamics and global biogeochemical cycles will respond to climate change. This study employs the DEMENT model of microbial decomposition, along with empirical mutation and dispersal rates, to explore the roles of mutation and dispersal in adaptation of soil microbial populations to shifts in litter chemistry, changes that are anticipated with climate-driven vegetation dynamics. Following a change in litter chemistry, mutation generally allows for a higher rate of litter decomposition than dispersal, especially when dispersal predominantly introduces genotypes already present in the population. These findings challenge the common idea that mutation rates are too low to affect ecosystem processes on ecological timescales. These results demonstrate that evolutionary processes, such as mutation, can help maintain ecosystem functioning as the climate changes.

Urban ecosystems, although highly altered by humans, host diverse microbiomes that support vital ecosystem processes. While microbial ecologists are beginning to understand the drivers of microbial assembly and the link between community structure and function in many ecosystems, few of these advances have been applied to urban ecosystems. In this synthesis, we review research on the urban soil microbiome and develop a framework to integrate soil microbial communities with urban ecosystem function. We identify disturbance, altered resources, and heterogeneity as key drivers through which urbanization affects soils and soil microorganisms. Steep environmental gradients in many urban systems present a unique opportunity to address fundamental questions in microbial ecology, such as how microbes respond to stress and how biogeochemical rates relate to microbial diversity and composition. Answering such questions will help develop practical and equitable strategies for managing ecosystem benefits in cities where billions of people live.

Soil biogeochemical models (SBMs) simulate element transfer processes between organic soil pools. These models can be used to specify falsifiable quantitative assertions about soil system dynamics and their responses to global surface temperature warming. To determine whether SBMs are useful for representing and forecasting data-generating processes in soils, it is important to conduct data assimilation and fitting of SBMs conditioned on soil pool and flux measurements to validate model predictive accuracy. SBM data assimilation has previously been carried out in approaches ranging from visual qualitative tuning of model output against data to more statistically rigorous Bayesian inferences that estimate posterior parameter distributions with Markov chain Monte Carlo (MCMC) methods. MCMC inference is better able to account for data and parameter uncertainty, but the computational inefficiency of MCMC methods limits their ability to scale assimilations to larger data sets. With formulation of efficient and statistically rigorous SBM inference frameworks remaining an open problem, we demonstrate the novel application of a variational inference framework that uses a method called normalizing flows to approximate SBMs that have been discretized into state space models. We fit the approximated SBMs to synthetic data sourced from known data-generating processes to identify discrepancies between the inference results and true parameter values and ensure functionality of our method. Our approach trades estimation accuracy for algorithmic efficiency gains that make SBM data assimilation more tractable and achievable under computational time and resource limitations.