Annual Review of Environment and Resources

Soil Microbiomes Under

Climate Change and
Implications for
Carbon Cycling
Dan Naylor,1 Natalie Sadler,1 Arunima Bhattacharjee,2
Emily B. Graham,1,3 Christopher R. Anderton,2
Ryan McClure,1 Mary Lipton,2 Kirsten S. Hofmockel,1,4
and Janet K. Jansson1
1
Biological Sciences Division, Earth and Biological Sciences Directorate, Pacific Northwest
National Laboratory, Richland, Washington 99352, USA; email: [email protected],
[email protected], [email protected], [email protected],
[email protected], [email protected]
2
Environmental Molecular Sciences Division, Earth and Biological Sciences Directorate,
Pacific Northwest National Laboratory, Richland, Washington 99352, USA;
email: [email protected], [email protected], [email protected]
3
School of Biological Sciences, Washington State University, Pullman, Washington 99163, USA
4
Department of Ecology, Evolution and Organismal Biology, Iowa State University, Ames,
Iowa 50011, USA

Annu. Rev. Environ. Resour. 2020. 45:29–59 Keywords

First published as a Review in Advance on
soil microbiome, carbon cycling, climate change, carbon sequestration, soil
June 2, 2020
carbon
The Annual Review of Environment and Resources is
online at environ.annualreviews.org Abstract
https://doi.org/10.1146/annurev-environ-012320-
Communities of soil microorganisms (soil microbiomes) play a major role
082720
in biogeochemical cycles and support of plant growth. Here we focus pri-
Copyright © 2020 by Annual Reviews.
marily on the roles that the soil microbiome plays in cycling soil organic
This work is licensed under a Creative Commons
Attribution 4.0 International License, which permits carbon and the impact of climate change on the soil carbon cycle. We
unrestricted use, distribution, and reproduction in first discuss current challenges in understanding the roles carried out by
any medium, provided the original author and
highly diverse and heterogeneous soil microbiomes and review existing
source are credited. See credit lines of images or
other third-party material in this article for license knowledge gaps in understanding how climate change will impact soil car-
information bon cycling by the soil microbiome. Because soil microbiome stability is a
key metric to understand as the climate changes, we discuss different as-
pects of stability, including resistance, resilience, and functional redundancy.

29
 We then review recent research pertaining to the impact of major climate perturbations on the soil
microbiome and the functions that they carry out. Finally, we review new experimental method-
ologies and modeling approaches under development that should facilitate our understanding of
the complex nature of the soil microbiome to better predict its future responses to climate change.

Contents
1. INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
2. CURRENT CHALLENGES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
3. STABILITY METRICS OF THE SOIL MICROBIOME . . . . . . . . . . . . . . . . . . . . . . . 32
3.1. Resistance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
3.2. Resilience . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
3.3. Functional Redundancy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
4. CLIMATE CHANGE IMPACTS ON THE SOIL MICROBIOME . . . . . . . . . . . . . 35
4.1. Soil Warming . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
4.2. Elevated Carbon Dioxide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
4.3. Combinatorial and Indirect Effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
4.4. Microbial Biochemical Pathways and Climate Change . . . . . . . . . . . . . . . . . . . . . . . 38
5. INTERKINGDOM INTERACTIONS AND SOIL CARBON . . . . . . . . . . . . . . . . . 39
5.1. Interkingdom Interactions in the Rhizosphere . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
5.2. Interkingdom Interactions in Biological Soil Crusts . . . . . . . . . . . . . . . . . . . . . . . . . 39
5.3. Fungal Loops Link Rhizospheres and Biocrusts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
5.4. Interactions Across Microbial Trophic Levels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
6. MINERAL WEATHERING AND SOIL CARBON . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
7. ADVANCES IN APPROACHES TO STUDY SOIL MICROBIOMES . . . . . . . . . . 42
7.1. Stable Isotope Probing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
8. MODELING CLIMATE CHANGE IMPACTS ON SOIL CARBON . . . . . . . . . . 45
8.1. Metabolic and Network Modeling of Soil Microbes and the Carbon Cycle . . . 46
8.2. Importance of Cross-Scale Experimentation and Data-Model Integration . . . . 47
9. CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49

1. INTRODUCTION
The soil microbiome contributes to ecosystem health in a variety of ways, including biogeochem-
ical cycling, bioremediation, plant growth, and primary productivity (1, 2). Its role in greenhouse
gas emissions and mediating soil organic carbon (SOC) is of particular interest in light of future
climate predictions. Climate change and shifts in land management practices can adversely affect
soil fertility and SOC (3, 4), which in turn impacts the soil microbiome and its net influence on
soil carbon sequestration. Here, we review the state of knowledge of the soil microbiome and
how its transformation of SOC is influenced by climate change factors. In particular, we address
challenges and metrics associated with studying the soil microbiome; we review effects of climate
change disturbances on taxonomic and functional profiles, as well as biochemical pathways; we
discuss the importance of interkingdom interactions in multiple soil systems; and we describe
recent advances in technical and modeling approaches.

30 Naylor et al.
2. CURRENT CHALLENGES
Soil ecosystems are highly complex and subject to different landscape-scale perturbations that
govern whether soil carbon is retained or released to the atmosphere (5). The ultimate fate of
SOC is a function of the combined activities of plants and belowground organisms, including
soil microbes. Although soil microorganisms are known to support a plethora of biogeochemical
functions related to carbon cycling (6), the vast majority of the soil microbiome remains uncul-
tivated and has largely cryptic functions (7). Only a mere fraction of soil microbial life has been
catalogued to date, although new soil microbes (7) and viruses are increasingly being discovered
(8). This lack of knowledge results in uncertainty of the contribution of soil microorganisms to
SOC cycling and hinders construction of accurate predictive models for global carbon flux under
climate change (9). Therefore, we are constantly refining our understanding of the biochemical
potential of the soil microbiome and the metabolic fate of SOC.
The lack of information concerning the soil microbiome metabolic potential makes it particu-
larly challenging to accurately account for the shifts in microbial activities that occur in response
to environmental change. For example, plant-derived carbon inputs can prime microbial activ-
ity to decompose existing SOC at rates higher than model expectations, resulting in error within
predictive models of carbon fluxes (10). To account for this, a conceptual model known as the
microbial carbon pump has been developed to define how soil microorganisms transform and sta-
bilize soil organic matter (11). In this model, microbial metabolic activities for carbon turnover are
segregated into two categories: ex vivo modification, referring to transformation of plant-derived
carbon by extracellular enzymes, and in vivo turnover, for intracellular carbon used in microbial
biomass turnover or deposited as dead microbial biomass, referred to as necromass. The con-
trasting impacts of catabolic activities that release SOC as carbon dioxide (CO2 ), versus anabolic
pathways that produce stable carbon compounds, control net carbon retention rates. In particular,
microbial carbon sequestration represents an underrepresented aspect of soil carbon flux that the
microbial carbon pump model attempts to address (11). A related area of uncertainty is how the
type of plant-derived carbon enhances microbial SOC storage or alternatively accelerates SOC
decomposition (12). For example, leaf litter and needle litter serve as sources of carbon for mi-
crobial growth in forest soils, but litter chemistry and pH varies by vegetation type [e.g., between
root and foliar litter (13) or between deciduous and coniferous forest litter (14)]. In turn, these
biochemical differences influence SOC levels through changing decomposition dynamics (14).
Also, increased diversity of plant communities increases rates of rhizodeposition, stimulating mi-
crobial activity and SOC storage (15), although soils eventually reach a saturation point beyond
which they cannot store additional carbon (16). Taking local environmental and biogeochemical
variables into account will help address our lack of understanding about net carbon flux in soils.
Quiescence also impacts microbial metabolic rates. Many soil microorganisms are transiently
active, alternating between dormant and active states (17). Even during dormancy, some soil mi-
croorganisms are capable of utilizing their energy reserves to metabolize SOC and contribute
to soil biomass turnover, albeit at slower rates (17, 18). Nevertheless, active members of the soil
community contribute the most to biogeochemical transformations, and a new paradigm is to shift
analyses away from taxonomic profiles and toward microbiome functional pathways and pheno-
types (19). However, current sequencing technologies for community composition also measure
dormant microorganisms and even exogenous DNA (20, 21) and are thus biased against active
functioning members of the community. Refining approaches to focus on function is therefore
anticipated to aid model construction through more accurate assessment of real-world processes.
Another challenge is accounting for the chemistry and physical structure of soils themselves,
both of which influence SOC decomposition. Traditionally, slow rates of carbon turnover were

www.annualreviews.org • Soil Microbiomes and Climate Change 31

 thought to be attributable to physical protection of carbon molecules in microaggregates or
mineral associations (22), or to their chemical recalcitrance to biodegradation (23). The current
paradigm expands on how mineral associations occur, namely through soil particles’ sorption of
Mineralization:
conversion of an biopolymers from microbial and plant necromass (24, 25)—indeed, deep soil organic matter is
element (e.g., carbon mainly comprised of microbe-derived products (26). Additionally, the spatiotemporal structure of
or nitrogen) from an soils is heterogeneous and dynamic, with “hot spots” or “hot moments” of microbial activity (27).
organic to an inorganic For instance, water availability is typically uneven, so carbon cycling is limited to areas with suf-
form, for example
ficient water, or to microbes capable of dealing with desiccation stress [e.g., through production
when organic carbon is
converted to carbon of extracellular polymeric substances to maintain a hydrated microenvironment (28)]. In addi-
dioxide and inorganic tion, other factors influencing SOC mineralization include presence of anaerobic versus aerobic
nutrients through microsites (anaerobic respiration of carbon being less energetically favorable than aerobic), avail-
microbial metabolism ability of electron acceptors, and redox status of the soil (29). Carbon profiles also matter: For
Resistance: example, carbon compounds recalcitrant to anaerobic degradation (e.g., lipids) are preferentially
the capacity of a retained in anaerobic aggregates (29). Therefore, comprehensive comparisons of SOC decom-
microbiome to retain a position profiles across different soil types and structures are necessary to better understand soil
consistent community
microbial metabolism and carbon flux.
and functional
composition in the
face of an external
perturbation 3. STABILITY METRICS OF THE SOIL MICROBIOME
Resilience: A major concern of climate change is its impact on soil microbiome stability and function and
the capacity of a by extension ecosystem sustainability (30–32). Meta-analyses have demonstrated that in approx-
microbiome to recover imately 80% of published studies, soil disturbances evoked measurable effects on microbiome
to an initial stable state stability (30, 33). Community stability is typically qualified with respect to one or more of three
after cessation of an
main metrics: resistance (remaining unchanged during disturbance), resilience (recovery to a sta-
external perturbation
ble state), and functional redundancy (functional profiles are maintained despite taxonomic shifts)
(32). Ideally, all of these metrics would be incorporated into microbiome disturbance studies, but
limitations in sampling time and effort often preclude this possibility. In particular, the degree of
resistance is often measurable during and immediately after a disturbance, but resilience trends
may only be visible years later (34). As climate disturbances increase in severity or frequency, un-
derstanding microbiome reaction patterns will improve prediction of future responses. Therefore,
these metrics represent an important consideration to take into account when designing distur-
bance experiments, and each is reviewed in detail below.

3.1. Resistance
The majority of disturbance studies have focused on resistance rather than resilience due to its
comparative ease of quantification. Resistance is commonly measured as shifts in community or
functional profiles under stress. For example, soil water limitation adversely affects members of
the Proteobacteria phylum and increases relative abundances of members of Actinobacteria and/or
Firmicutes phyla (35). Through their effects on phylogenetic profiles, disturbances will in turn af-
fect ecosystem functioning. For example, soil drying altered the abundance of guilds for microor-
ganisms involved in methane oxidation (36), whereas soil warming or elevated carbon dioxide
(eCO2 ) affected ammonia-oxidizing microbes (37). Anthropogenic nitrogen deposition (through
excessive fertilizer addition) can enrich for nitrogen-cycling processes, including urea decomposi-
tion and tricarboxylate transport (38, 39). Some environmental stresses may hinder carbon cycling
through decreasing metabolic diversity of a community (40) or by limiting microbial uptake of car-
bon through decreased diffusion rates (41). For example, enzymatic activity rates, including that
of carbon cycling enzymes (beta-glucosidase, aminopeptidase) or other nutrient cycling enzymes

32 Naylor et al.
(acid phosphatase, arylsulfatase), have been shown to be suppressed under drought and following
soil burning (42, 43). As a consequence, predictions of how stress affects biogeochemical processes
for carbon and nitrogen mineralization need to account for microbial responses.
Rhizosphere:
Microbial life strategies are closely tied to resistance, namely ratios of K- to r-selected or- the region of soil
ganisms. (K-selected microbes maximize survival by being slow growing and resource efficient, surrounding and
whereas r-selected organisms are energy and resource inefficient but maximize survival through directly influenced by
rapid rates of growth and reproduction.) In one study, communities with higher ratios of Gram- plant roots, including
their related secretions
positive (normally K-selected) to Gram-negative (normally r-selected) bacteria were more resis-
and exudates,
tant to eCO2 (44). K-selected organisms are associated with slower growth, higher enzyme sub- harboring microbes,
strate affinities, and usage of more recalcitrant forms of carbon (45), traits tied to stress resistance. and microbial
By contrast, r-selected organisms are normally more dependent on labile carbon compounds for processes essential for
growth, such as those released into the rhizosphere through plant root exudates. Because some plant health
endemic plant species decrease rhizodeposition into soil under drought stress to maintain a car-
bon supply for their own survival, there is a depletion of labile SOC stocks into the surrounding
soil. As the principal remaining carbon sources are recalcitrant carbon molecules, K-strategists are
favored over r-strategists (35).
Physiological adaptation is a resource-intensive but effective means of conferring stress resis-
tance. Some soil microbes have adopted thicker cell walls to tolerate desiccation stress (35), and/or
membrane adaptations to tolerate exposure to toxic metals (34). Previous exposure to a stress con-
dition (34) can prime a community to resist future stresses with a similar mode of action, for
example, through upregulation of and/or increased dispersion of resistance genes (32, 46). How-
ever, investment into a novel resistance mechanism often has the trade-off of losing a previous
one, and microbes may become susceptible to a stress that they were previously resistant to (34).
These trends have been observed for numerous (nonclimate change–related) ecological distur-
bances: For example, long-term copper stress hindered the soil microbiome’s capacity to respond
to fluctuating environmental conditions (47). Similarly, chronically trampled dryland soils were
less able to respond to rewetting than non-trampled ones (48). The most resistant communities
often display functional plasticity and shift metabolic profiles as a function of environmental con-
ditions, enhancing their survivability if a particular niche is destroyed (33). However, it remains to
be seen whether physiological adaptations and/or functional plasticity will be widespread enough
under climate change disturbances to ensure the survival of soil ecosystems.

3.2. Resilience
The phenomenon of soil microbiome resilience is arguably underreported, as studies incorporat-
ing a long-enough time course to track full recovery are uncommon (32). Even when explicitly
measured, pre-disturbance profiles may take years to re-establish (49), and in some cases puta-
tively irreversible changes occur (30, 50); these trends emphasize the importance of long-term
studies incorporating decadal timescales to track microbial responses to disturbance (51–53). In a
meta-analysis of short- and long-term disturbances, recovery was generally observed in less than
half of the studies (33). As disturbances increase in frequency and duration, such as during climate
change, it is imperative to understand how, if at all, microbiomes are able to recover.
Similar to resistance, microbiome resilience may be assessed based on taxonomic and/or func-
tional profiles. One approach for measuring resilience is through clustering taxa based on recovery
patterns—for example, taxa that increase under stress before subsequently decreasing during re-
covery would form one cluster, whereas taxa that show the opposite trend would form another
(54). Resilience can also vary by rate of recovery. For example, members of the Planctomycetes,
Crenarchaea, and Acidobacteria phyla recovered faster after a soil warming treatment than did

www.annualreviews.org • Soil Microbiomes and Climate Change 33

 Actinobacteria or Verrucomicrobia (55). However, not all members of a given phylum respond
universally in the same manner. For example, specific classes within the Acidobacteria and Pro-
teobacteria phyla were shown to differ in their resiliencies to drought stress (35). Distinct resilience
trends by phyla have implications for the carbon cycling processes they mediate, as individual
taxa have characteristic growth and carbon assimilation patterns (56). For example, Actinobacte-
ria abundance was negatively associated with carbon mineralization, whereas Bacteroidetes and
Proteobacteria were positively associated (57). Therefore, rates of soil carbon cycling will largely
depend on how fast members from these phyla recover to a given stress. Similarly, for functional
profiles, resilience depends on the function in question and the phylogenetic resolution that is be-
ing examined. For example, nitrification is less resilient than denitrification (32, 58), likely because
it is mediated by a narrower guild of microbes. Therefore, functions based on broadly distributed
enzymes generally have more resistance but lower resilience, whereas those with narrowly dis-
tributed enzymes, such as complex polysaccharide degradation, have less resistance but higher
resilience (59). Another discrepancy between resistance and resilience is the influence of prior
stress—previous exposure to a stress often decreases rates of resilience to a new one, whereas
resistance is generally strengthened (43).
Several factors contribute to microbial resilience. One is prevalence: Highly abundant and/or
widely dispersed organisms are less likely to be obliterated by a stress. Another strategy for resilient
microbes is to enter dormancy, forming what is known as the microbial seed bank (60). In both
scenarios, surviving microbes are better poised to reseed the soil microbiome upon stress ame-
lioration (33). Rapid ribosome synthesis and shorter generation times are advantageous traits, as
they increase the speed of recovery; however, fast-growing taxa (e.g., r-strategists) are often highly
resource dependent and therefore more susceptible to stress (45). Overall community resilience
is also aided by stress resistance mechanisms, as they may be passed from tolerant to susceptible
individuals via gene flow to aid recovery (61). Alternatively, tolerant but less altruistic organisms
may keep resistance mechanisms to themselves, growing rapidly under a given stress condition
while susceptible organisms die off (60). In extreme cases, opportunistic individuals have been
shown to adapt their metabolic pathways to incorporate an otherwise stressful toxic compound
as a carbon/nitrogen source (62). Even outright antagonism against other recovering groups may
aid resilience, which was posited as the reason behind increased survivability for bacteria relative
to fungi after soil heating (63).

3.3. Functional Redundancy

Functional profiles may be resistant or resilient under stress even if taxonomic profiles change.
Desiccation-rewetting cycles have been shown to alter community composition, but often func-
tions (extracellular enzyme activity, methane oxidation, carbon cycling) remain resilient (36, 42,
48). In such cases, the relative abundances of members of a functional guild might shift depend-
ing on their inherent stress susceptibilities, where the more resistant members predominate after
stress is lifted (60). Functional redundancy can also be tied to ubiquity. For example, common pro-
cesses such as respiration are conserved in all but the most extreme disturbances, whereas more
specific processes such as nitrification may not be (30).
The disconnect between function and phylogeny is exacerbated by limitations of current meth-
ods (20). To track members of a functional guild, a typical approach is to perform amplicon se-
quencing of a gene involved in the function. However, if gene variants or functional orthologs with
dissimilar sequences exist, a significant proportion of the functional guild will remain uncaptured
through this approach (32). Resolving this discrepancy will reconcile how a function remains con-
stant in the face of taxonomic shifts (e.g., confirming that a guild member increases in abundance to

34 Naylor et al.
compensate for loss of another). The greater functional redundancy a soil community displays, the
more resilient it will be following disturbance (59), highlighting the importance of understanding
functional redundancy in the context of other microbiome stability metrics. Accordingly, certain
aspects of stability metrics are now being incorporated into new trait-based microbial models in
soils (64).

4. CLIMATE CHANGE IMPACTS ON THE SOIL MICROBIOME

Climate change–associated disturbances can significantly alter soil microbial community and func-
tional profiles (5). If soil carbon and/or nitrogen cycling are affected, this can in turn affect climate
change either through positive feedbacks to the atmosphere (e.g., greenhouse gas emissions) or
negative feedbacks (e.g., carbon immobilization into microbial or plant biomass) (12). Better un-
derstanding of how soil microorganisms respond to climate change will therefore ultimately im-
prove climate models. However, climate change can invoke several distinct perturbations or even
compounding disturbances, which can exert contrasting effects on the soil microbiome (5). Given
the uncertainty as to the interplay between different climate change factors, recent studies have
begun to incorporate multiple factors in combination (37, 65–67). Here, we specifically review soil
microbiome responses to soil warming and eCO2 , and how these factors interact with one another
directly and indirectly to effect change in soil community and functional profiles.

4.1. Soil Warming

Current climate models predict a global temperature rise of roughly 3.7°C by 2100 (68). Consid-
ering that soil microbial communities are demonstrably affected by warming (5), this represents
an unavoidable impact of climate change on the soil microbiome. Soil warming is thought to im-
pact resident microbial communities in a stepwise fashion. First, organic carbon decomposition
rates are enhanced over the short term, increasing microbial biomass. One study found that the
soil microbial population size increased by 40–150% under soil warming (55). Next, microbial
respiration has been shown to decline over time as labile carbon is depleted (69). After years of
exposure, changes have been observed in microbial physiologies, community composition, and
functional profiles, both as microbes adapt to warming, and as their metabolism shifts to utilize
the remaining recalcitrant carbon sources (70). The nuances behind these steps are outlined below.
Warming has been observed to increase microbiome community diversity and richness (55,
71, 72), as well as to enrich for members of the Acidobacteria and Actinobacteria phyla and
class Alphaproteobacteria (55, 69, 73). These taxonomic shifts overlap with functional profiles:
Oligotrophic taxa (i.e., slow-growing microbes capable of surviving in nutrient-poor conditions,
e.g., Actinobacteria) are promoted over copiotrophic taxa (i.e., fast-growing microbes optimized
for nutrient-rich environments, e.g., Bacteroidetes), possibly as a response to changing soil carbon
composition (74). For example, warming treatments lasting 5 to 8 years were shown to favor more
recalcitrant carbon-degrading taxa from the Actinobacteria or Acidobacteria, despite few overall
measurable responses in community composition (52). Measurable differences in functional guilds
responsible for ammonia oxidation (37) or diazotrophy (72) have also been observed following soil
warming.
Microbial function can be impacted by warming both directly (e.g., through acceleration of en-
zymatic rates) or indirectly (stimulating plant growth and rhizodeposition and altering soil prop-
erties). For example, cycling of phosphorus and sulfur has been shown to be stimulated under
warming (70, 75), but making inferences for carbon and nitrogen cycling is more difficult. Warm-
ing has been demonstrated to raise rates of nitrogen cycling processes, including denitrification,

www.annualreviews.org • Soil Microbiomes and Climate Change 35

 nitrogen fixation, nitrification, and nitrogen mineralization (75), although its exact effects depend
on the gene/process under study (70). For example, in some cases warming suppressed certain
nitrogen cycling functions (65, 72). One explanation is negative feedback: Warming increases soil
Copiotrophs:
fast-growing inorganic nitrogen and plant nitrogen pool sizes (66), ultimately depressing rates of microbial de-
microorganisms that composition and nitrogen cycling (76, 77). Therefore, it is possible that nitrogen cycling can shift
are prevalent in over time as a function of the duration/magnitude of warming and nitrogen availability.
nutrient-rich By contrast, carbon cycling has been shown to be initially accelerated by warming (73, 74) if
environments with
carbon bioavailability is sufficient. The temperature optima of extracellular enzymes for carbon
labile sources of
carbon degradation are such that warming can act as a stimulus (69). Over long periods of warming, studies
have observed decreased numbers of genes involved in labile carbon degradation, with increases in
Oligotrophs:
those for recalcitrant carbon metabolism (65, 70, 74) and a higher diversity of the responsible func-
slow-growing
microorganisms that tional guild (73). These findings may be at least partly attributable to water loss from evaporation
are prevalent in during heating. When soil moisture is controlled, labile carbon degradation can remain stimulated
low-nutrient while degradation of recalcitrant carbon is unchanged (75). Carbon cycling shifts also vary by soil
environments layer, where organic and mineral horizons have different responses in carbohydrate-degradation
potential after decadal timescales of warming (52). Analyzing soil warming as a single factor thus
represents a suboptimal approach, as warming is likely to be coupled with other climate change
factors that also influence carbon cycling, not only depletion of soil moisture but also eCO2 .

4.2. Elevated Carbon Dioxide

As with warming, eCO2 has both direct and indirect effects on the soil microbiome (Figure 1). In
the short term, eCO2 increases respiration rates, microbial biomass, and genetic signals for carbon
cycling processes (78). It also stimulates plant production and rhizodeposition, in turn priming
copiotrophs in the rhizosphere to break down labile and (later) recalcitrant carbon (65, 79, 80).
Nevertheless, taxonomic trends for soil microbiomes under eCO2 are by no means consistent.
One study analyzing trends of eCO2 across soil ecosystems found that the only common response
was depletion of Acidobacteria Groups 1 and 2 (81). Similar to warming, however, over a long
timescale eCO2 is predicted to enrich for oligotrophs. After 14 years of eCO2 in a California
grassland, decreases in copiotrophic (r-selected) Bacteroidetes were observed, along with increases
in microbes with lower rRNA copy numbers, a common trait of oligotrophs (K-selected) (82).
Under warming, enrichment of oligotrophic microorganisms is expected, due to decreased soil
moisture and depletion of labile carbon. By contrast, eCO2 is predicted to stimulate plant and
microbial growth, which depletes soil nitrogen. As a result, soil carbon cycling is predicted to
decline. Indeed, over longer timescales of eCO2 treatment, there was a reported marked decrease
in soil carbon cycling, with little to no change in carbon degradation (82). Such conditions will
thus favor slower-growing, resource-efficient oligotrophic microorganisms.
Under eCO2 , enzymatic activities for phosphorus cycling tend to increase (65, 78, 82, 83),
but nitrogen cycling is more variable. Increases in plant net primary production, microbial im-
mobilization of soil nitrogen, and microbial denitrification rates will all deplete soil mineral ni-
trogen (66, 82, 84). As a consequence, maintaining soil nitrogen availability (and by extension
plant/microbial growth rates) necessitates an increase in relative rates of nitrogen cycling and
mineralization. Enhanced nitrogen cycling under eCO2 has been observed (37, 53, 78, 79, 85, 86),
although actual enzymatic rates are often unchanged or decline (82). This discrepancy may be
attributable to higher abundances of nitrogen fixers (e.g., Rhizobiales) or ammonia oxidizers (37,
85), although this is not a universal trend (44, 77). Differing results for nitrogen cycling are occa-
sionally observed across eCO2 studies and may be influenced by variability in confounding factors
such as soil moisture availability, proximity to root exudates, soil depth, and degree of nitrogen

36 Naylor et al.
 C C CO2 CO2

Soil microbial
respiration

Rhizodeposition

Microbes Carbon

Short term
• Increase plant productivity
• Increase rhizodeposition
Long term
• Deplete soil C reserves
(CO2 to atmosphere)

Figure 1
Predicted effects of elevated carbon dioxide (eCO2 ) on soil carbon reserves. In the short term, plant growth is stimulated by eCO2 ,
resulting in increased rhizodeposition, priming microbes to mineralize soil organic carbon (SOC) and adding CO2 to the atmosphere
through respiration. But the net impact on greenhouse gas emissions will be reduced by the increased uptake of CO2 from the
atmosphere by increased plant growth. However, over the long term, soil reserves of easily decomposed carbon will be depleted by the
increase in microbial activity, resulting in increased catabolism of SOC reservoirs, thus increasing atmospheric CO2 concentrations
beyond what is taken up by plants. This is predicted to be a particular problem in thawing permafrost that contains large reserves of
SOC that are becoming increasingly susceptible to microbial degradation as the permafrost thaws (195, 196).

limitation (44, 79). In addition, the ecosystem in question, e.g., agroecosystems, may have different
results from uncultivated forests (81).

4.3. Combinatorial and Indirect Effects

Considering any climate change factor in isolation fails to address the interplay between them that
is likely to impact soils in real-world scenarios. To account for this knowledge gap, many recent
studies have incorporated multifactorial designs, whether with eCO2 and warming (37, 65–67),
eCO2 and elevated ozone (79, 85, 87), eCO2 and nitrogen addition (44), or other combinations.
Often, differing results are found for combinations compared to single-factor treatments, high-
lighting the importance of this approach. For instance, in one experiment modeling the effects of
warming and/or eCO2 on field soils in a cotton agroecosystem, the combination of warming with
eCO2 provoked shifts in ammonia-oxidizing microbial communities and increases in soil nitrifica-
tion rates, whereas few significant effects were seen for warming alone (37). Often, a combination
of perturbations results in one factor attenuating the effects of the other. With respect to eCO2 and
warming, often eCO2 counteracts warming-induced decreases in soil moisture or promotes plant

www.annualreviews.org • Soil Microbiomes and Climate Change 37

 rhizodeposition to maintain carbon cycling and heterotrophic respiration as carbon is depleted
under warming (65). It is likely that the relative importance of the two factors varies by environ-
ment. For example, different trends might be seen in grasslands compared to forests, or agroe-
cosystems compared to arctic biomes (5). For instance, in a dryland community study, warming
prevailed over eCO2 (67), while in a grassland study the opposite trend was observed (65). In the
latter study, the combination of eCO2 and warming had similar effects to eCO2 alone—warming
decreased signals for carbon cycling, ammonia oxidation, and production, whereas eCO2 and the
combination had the opposite trend. Notably, a subset of eCO2 -stimulated genes for nitrogen cy-
cling and carbon degradation were no longer enriched under the combination, including genes
for recalcitrant carbon degradation (65), which may be a result of increased rhizodeposition of
labile carbon precluding the necessity of such genes.
A complicating factor for studying disturbance responses in the soil microbiome is disentan-
gling direct from indirect effects. As discussed above, eCO2 indirectly impacts soil communities
through increased plant rhizodeposition, soil nitrogen limitation, and higher soil moisture con-
tent (eCO2 induces plant water conservation through reduced stomatal conductance) (65, 83, 88),
as well as through root exudate profiles, soil structure, or leaf litter chemistry (85). Conversely,
warming stimulates plant growth but lowers soil moisture through evaporation, and such shifts
in water availability may have a greater impact on the soil microbiome than warming alone (55,
89). Specifically, enrichment for oligotrophs under warming may be at least in part due to their
higher enzyme substrate affinities representing an advantage as diffusion decreases under water
limitation (74). Other complicating factors include treatment duration (73, 79), seasonality (83),
and soil depth or horizon (52, 81, 90). Such discrepancies highlight the importance of accounting
for confounding parameters during soil perturbation studies.

4.4. Microbial Biochemical Pathways and Climate Change

Although the response of the soil microbiome is often studied at a high level, such as community-
wide taxonomic shifts, another important aspect of climate change response is how specific bio-
chemical pathways are affected. A recent study on warmed soils from Arctic and Antarctic envi-
ronments found numerous common metabolic responses (84). For example, methane production
and metabolism of acetate and di- and mono-methylamine increased as temperatures were raised
from 1°C to 30°C, whereas decreases were seen in propionate and acetate oxidation as well as
metabolism of H2 and formate (91). Furthermore, as temperatures were raised above 7°C, the
rate-limiting step for methane production shifted from propionate oxidation to polysaccharide
hydrolysis. Similarly, the drying of Puerto Rican soils increased signals for carbon metabolism en-
zymes including beta-glucosidase, cellobiohydrolase, N-acetylglucosaminidase, and xylanase (92).
However, this effect was reduced through pretreatment of soils with simulated drought, suggest-
ing long-term changes to soil functioning in response to a disturbance may ameliorate effects of
future stresses.
Microbial biochemical pathways are also indirectly affected through climate change impacts on
plant-microbe interactions. A recent study found that microbes take up less plant-derived carbon
under both heat and drought stress (93). Additionally, if environmental stresses cause exotic or
invasive species to thrive, root exudate profiles (representing a major source of bioavailable carbon
for microbes) will shift. A study contrasting native with exotic plants found that soil microbiomes in
the presence of the latter shifted to osmotic stress conditions, but had an increased supply of labile
carbon inputs (94). Furthermore, climate change may alter plant cover (88) or plant community
profiles, e.g., through plant migration to colder climes (77) or increased ratios of C4 :C3 plant
types (75). Responses in rhizodeposition under stress can also vary by plant species or cultivar

38 Naylor et al.
(85). For example, wild-type plants were shown to have higher rates of root exudation under eCO2
than cultivated varieties (88), as did C4 grasses relative to C3 plants (70, 75). As a consequence of
changing type and quantity of plant-derived carbon inputs to the soil, different microbial pathways
for carbon uptake and metabolism will be stimulated.

5. INTERKINGDOM INTERACTIONS AND SOIL CARBON

Biotic and abiotic factors, including climate change disturbances, can affect carbon cycling through
altering interspecies and interkingdom interactions. The functional relationships between or-
ganisms capable of fixing atmospheric CO2 (e.g., plants and autotrophic microorganisms) and
heterotrophic organisms who obtain carbon byproducts through mutualistic, commensalistic, or
parasitic interactions continue to be explored. Active areas of investigation include elucidating
keystone species or symbiotic interaction trends of soil communities, and how these interactions
inform carbon cycling shifts under disturbances. However, trends may be particular to a given en-
vironment, so here we review interkingdom interactions across multiple soil ecosystems, including
rhizospheres and biological soil crusts.

5.1. Interkingdom Interactions in the Rhizosphere

Plant-microbe interactions are among the most prominent interkingdom interactions in soil. For
example, plants can stimulate microbial growth and activity in the rhizosphere through root ex-
udate secretion. A higher aboveground plant diversity has been shown to be beneficial for soil
microbial community diversity and nutrient cycling (15). Higher plant species richness can in-
crease microbial biomass and necromass and accelerate microbial growth and turnover, although
microbial carbon-use efficiency and respiration were not shown to be affected (95). In mesocosm
experiments, soil warming selected for larger and more productive plant species, but a less diverse
plant community (96). However, bacterial and protist abundances increased, leading to greater soil
multifunctionality (96). Together these findings demonstrate the importance of plant-microbe
interactions for regulating ecosystem responses to environmental change and the potential for
managing plant-microbe interactions to mitigate environmental impacts.
Soil interkingdom interactions are also influential in carbon flow. In a recent study tracking
plant-derived carbon in soil through stable-isotope probing, rhizodeposits from Zea mays moved
through bacteria, unicellular fungi, and protists, likely using arbuscular mycorrhizal fungi (AMF)
as a shunt for fresh root exudate supply to microbes (97). The researchers also observed trophic
webs involving different flagellates, amoeba, and ciliates in the rhizosphere and surrounding soil,
although isotopic labeling of filamentous saprotrophic fungi was not apparent (97). AMF were
also observed to act as a carbon shunt from plants to soils in a study tracking 13 C passage (98),
with subsequent direct carbon transfer from AMF to bacteria and fungi, and indirect transfer
to protozoa and nematodes. The authors hypothesized that eCO2 levels changed the nature of
these interkingdom interactions, not only by increasing rhizodeposition but also by provoking
emergence of distinct AMF and soil microbial communities, which in turn altered soil carbon
cycling profiles (98). This study signifies the importance of studying carbon flow in the context of
climate change factors, thereby discovering which additional factors affect carbon flow and what
their specific effects might be.

5.2. Interkingdom Interactions in Biological Soil Crusts

Another soil ecosystem that has been extensively studied with respect to interkingdom interac-
tions is biological soil crusts, or biocrusts. Biocrusts are stratified communities of photoautotrophs

www.annualreviews.org • Soil Microbiomes and Climate Change 39

 (cyanobacteria, algae, lichens) and heterotrophs (fungi, bacteria, archaea). They cover ∼12% of
Earth’s landmass and contribute to carbon and nitrogen cycling, soil stability, and water reten-
tion in arid environments (99–101). Biocrust successional development can occur over years to
decades and is hypersensitive to environmental and physical disturbances (101). Over the course
of succession, photoautotrophic organisms facilitate assembly of specific microbial communities,
establishing a network of interactions (99). For example, a putative mutualism exists between
keystone cyanobacteria Microcoleus vaginatus and copiotrophic diazotrophs, where both carbon-
and nitrogen-fixation activities are covered through the interaction. This consortium has been
posited as the “true” pioneer community enabling biocrust colonization of nitrogen-poor soils
(102). Other interactions are not so beneficial: Later in succession, production of the alkaloid
sunscreen pigment scytonemin by heat-tolerant cyanobacteria M. steenstrupii results in biocrust
surface warming by up to 10°C, allowing them to replace the susceptible M. vaginatus (103).
Climate stresses are predicted to alter the nature of interactions in biocrusts. Rewetting cycles
are common disturbances in these systems and have significant impacts on the interactions therein.
One recent study found hydration pulses result in large blooms of members of the bacterial or-
der, Bacillales, accompanied by collapse of pioneer cyanobacteria within the Oscillatoriales order
(101). Mortality of cyanobacteria in the above manner may have important implications for car-
bon balance in semiarid ecosystems. For example, if hydration pulses cannot sufficiently stimulate
autotrophs (i.e., cyanobacteria) to the extent where photosynthesis compensates for respiration
(104), biocrusts will experience autotroph mortality and net carbon loss (101).

5.3. Fungal Loops Link Rhizospheres and Biocrusts

In arid ecosystems lacking large reserves of soil organic matter, biotic retention of nutrients is
critical to primary production. Spatiotemporal resource partitioning and a variety of species in-
teractions can aid ecosystem stability in the face of an increasingly arid environment (as is pre-
dicted to occur with climate change) by retaining resources within a biotic loop. For example,
although autotrophic bacteria within biocrusts are directly responsible for carbon inputs, the fun-
gal loop hypothesis proposes that fungal hyphae function as reservoirs of carbon and as conduits
to translocate resources between soils and host plants (105, 106). At lower water potentials, fungi
can take up nutrients produced by biocrusts, reserving them in the biotic pool until larger water
pulses allow plants to take them up. In return, plants can contribute excess photosynthate to their
fungal symbionts. This example illustrates how fungal association can connect the spatially and
temporally distinct activities of biocrusts and plants (106).

5.4. Interactions Across Microbial Trophic Levels

In soils, filamentous fungi cope with heterogeneous environments and resource distributions via
growth of extensive hyphae that penetrate air-water interfaces and traverse air-filled pores. Such
“fungal highways” may mobilize bacteria, connect microbial microhabitats, and disperse nutrients
(107). Bacteria are impacted by presence of mycelia in several ways: For example, mycelia pro-
mote horizontal gene transfer between bacteria (108). In another study, mycelia enabled contact
between the bacterial predator Bdellovibrio bacteriovorus 109J and its prey Pseudomonas fluorescens
LP6a, allowing for foraging and shaping of prey populations in a manner not seen in the absence
of mycelia (107). Low soil pH was shown to limit survival and dispersal of Paraburkholderia terrae
BS001; however, this effect was attenuated by a fungal hyphae-mediated increase in soil pH, sug-
gesting fungi exert protective effects on bacterial cells (109). Together, these studies demonstrate
fungi occupy a myriad of roles that can influence bacterial survival.

40 Naylor et al.
 Apart from plants and fungi, soil bacteria and archaea also display significant associations with
protists and viruses. Some protists have been shown to promote plant-beneficial functions, accel-
erate nutrient cycling, and regulate population growth of specific bacterial species within the soil
microbiome (110). They can occupy central hubs in soil microbial networks, thus linking diverse
bacterial and fungal populations (110). An increased fungal:bacterial ratio was also shown to be
negatively correlated with phagotrophic protists (111). This may be attributable to phagotrophic
activity, as protists graze on bacteria of certain morphologies, thereby regulating which bacterial
morphotypes persist in a soil community (112).
Viruses can also directly influence their microbial hosts and play a key role in controlling host
abundances. Recently, viruses have been proposed to participate in regulation of the structure
of local soil microbial communities through transfer of resistance or metabolic genes (113). In
peatland soils under a permafrost thaw gradient, as well as in mangrove soils, viruses have been
implicated in contributing to carbon metabolism by carrying auxiliary metabolic genes, such as
glycoside hydrolase genes (e.g., endomannanase), that can complement activities carried by their
bacterial hosts (114, 115). As a result, viruses can potentially affect local and global biogeochemical
cycles through supplementing the soil microbiome with genes involved in different metabolic
pathways. Although viruses have been extensively studied in aquatic systems, the ecological role
of viruses is a pressing frontier in terrestrial ecosystem science. New methods are emerging to
address this knowledge gap, including analyzing viral sequences within bulk-soil metagenomes
or optimizing viral recovery from soil through new viral resuspension protocols (113). Current
research frontiers include determining the rates of viral turnover in soil, infectivity rates of phages
toward soil bacterial and fungal hosts, and potential functional roles of soil viruses (116).

6. MINERAL WEATHERING AND SOIL CARBON

Soil microorganisms also interact during weathering of soil mineral interfaces, which in turn af-
fects the global carbon cycle through sequestration of atmospheric carbon in weathered prod-
ucts. Mineral weathering is enhanced by plants and microbes—for example, lichens and microbial
biofilms play a role in chemical weathering of rocks (117). Plant-associated arbuscular mycorrhizal
and ectomycorrhizal fungi have been shown to accelerate weathering of carbonate rocks, where
the rates of weathering depend on the precise nature of the plant-fungi symbiosis (118). In turn,
soil weathering accelerates development of soils and drawdown of large amounts of CO2 from the
atmosphere. Under climate change, weathering is projected to be subject to negative feedback,
where weathering rates initially go up with higher concentrations of atmospheric CO2 , thereby
increasing CO2 drawdown and reducing weathering (119, 120).
Mineral weathering also yields highly reactive secondary clay minerals that alter soil biogeo-
chemical properties. As community composition is a function of the soil mineral profile, and car-
bon breakdown and mineralization are in turn a function of community composition (121), weath-
ering therefore has an influence over microbial carbon cycling. Quantifying microbial soil carbon
turnover in the presence of clay minerals is therefore essential in the context of shifts in the soil
matrix, such as those that may occur under climate change disturbances. One relevant influence
that clay minerals have is for carbon persistence, where carbon molecules’ interactions with miner-
als physically protects them from degradation (122). Soil organo-mineral interactions thus create
microenvironments of distinct carbon and mineral profiles, primed for microbial growth and de-
velopment. Characteristics of these microenvironments (mineral content, soil texture, aggregate
size) influence the stability of soil organic matter, microbial growth, and community composition
(123, 124). Depending on how community profiles shift, there may be further implications for
carbon mineralization and storage. For example, a higher fungal:bacterial ratio favors soil carbon

www.annualreviews.org • Soil Microbiomes and Climate Change 41

 storage, as fungi are primary contributors to soil carbon decomposition, whereas bacteria typically
assimilate carbon substrates into biomass or respire them back into the atmosphere (125).
Although the importance of soil and associated secondary minerals in stabilizing carbon is rec-
Metatranscriptomics:
comprehensive ognized, the links between processes governing land-atmosphere carbon exchange are not well
analysis of gene understood. Recent studies have demonstrated that over time mineral weathering first increases
expression at the then decreases nutrient availability and soil carbon stabilization, due to reductions in nutrient- and
transcriptional level carbon-stabilizing capacity, and less carbon inputs from plants (126). Additionally, drastic changes
within a complex
such as land conversion and erosion can lead to removal of weathered and reactive soil materials.
community/
environmental sample These changes affect both the soil matrix and carbon storage capacity (127). They also negatively
impact both soil nutrient quality and can result in declining forest ecosystem sustainability. Un-
Metaproteomics:
der these conditions, rhizosphere bacteria and fungi can promote carbon and nutrient cycling to
comprehensive
analysis of proteins maintain nutrient supplies to associated plants (128). Ultimately, the influence of geochemical fac-
produced within a tors needs to be better integrated into soil microbial ecology as they represent a complex, integral
complex community/ part of global carbon cycling.
environmental sample,
as a means to quantify
gene expression at the 7. ADVANCES IN APPROACHES TO STUDY SOIL MICROBIOMES
translation level,
following As the soil microbiome is very heterogeneous and complex, the direct application of omics tech-
post-transcriptional niques to discern microbial community function, not to mention synthesis of multiple omics types
modifications into coherent ecological information, is very difficult. Even the comparatively simple analysis of
amplicon sequencing for community composition can yield different results as a consequence of
variations in sequencing depth, primer bias, choice of processing pipeline, or DNA extraction
protocol (129, 130). When moving to metagenomics analysis, computational demand becomes an
issue, due to the processing power required to assemble metagenomes containing tens of thou-
sands of species (131). Although metatranscriptomics has the benefit over amplicon sequencing
or metagenomics approaches, in that one can determine which species are active and which genes
are expressed, it is not without its drawbacks (132). Extraneous ribosomal RNA represents the
predominant fraction of extracted RNA, requiring an additional rRNA depletion step and/or ne-
cessitating greater sequencing depth to make this analysis feasible (133). Ultimately, metapro-
teomics provides valuable information about gene expression and subsequent translation patterns
(134), although proteomics methods are currently too low-throughput for widespread adoption.
Similarly, soil metabolomics are useful for informing biogeochemical cycling through obtaining
high-resolution snapshots of soil nutrient profiles and metabolite flux patterns, although these
techniques may be limited by underannotated reference databases and discerning metabolite ori-
gins (e.g., between microbes, plants, or other sources) (135). As each omics technique becomes
more inexpensive and high-throughput, a desirable avenue will be incorporating multiple omics
datasets into a combined data stream.
Rapid advances in technology development are helping to overcome current limitations and
enabling multi-omics data streams that are well-suited for determining taxonomy, phenotypic fea-
tures, and metabolic functions of soil communities. For example, a combination of metagenomics,
metatranscriptomics, and metaproteomics was used to determine the response of permafrost soil
microbial communities to thaw (136). This study revealed that functional inferences from multi-
omics approaches corresponded better with process rates in thawed soils than those for frozen
soils, while also discovering novel strategies for microbes to maintain activity and survive under
frozen conditions (136). The application of multi-omics (amplicon sequencing, metatranscrip-
tomics, and metabolomics) to determine how soil microbial communities responded to soil wet-
ting or soil desiccation demonstrated consistent metabolic responses to wet-dry cycles, including
production of sugars and osmolytes as a drying response (137). Ultimately, the combination of

42 Naylor et al.
multiple data streams will benefit soil analyses by providing a more complete, multilevel picture
of soil processes than would be achievable through any technique in isolation.
Another approach to overcome the complexity of the soil habitat is to obtain model, simpli-
fied soil communities and study them in highly controlled soil-emulating environments. A va-
riety of simulated soil environments are under development that range from microscale- (e.g.,
microfluidic devices) to mesoscale-simulated soil ecosystems. Two recent studies characterized
reduced-complexity microbial consortia from parent soil that were cultured either in liquid media
(138, 139) and/or sterilized soil (139). In the former study, it was concluded that community-
level function and high-level taxonomy were predictable, governed by nutrients, and ultimately
a generalizable feature of community assembly. In the latter, vastly different consortia resulted
from soil versus liquid microcosms, indicating a physical selective pressure exists for community
assembly. One benefit in using a simplified microbial community is its utility for determining
key mechanisms governing soil microbial community ecology (139), specifically by enriching for
the subcommunity implicated in a process of interest. In doing so, experimentally tractable sub-
sets can be obtained that maintain native interactions while allowing for the opportunity to study
these subcomponents of the parent microbiome at high resolution.

7.1. Stable Isotope Probing

The evaluation and tracking of stable isotopes represent a powerful approach for determining
the fate of organic carbon in soils. The use of stable isotopes (most commonly 13 C or H2 18 O)
can disentangle carbon cycling processes by determining the turnover rate(s) of specific carbon
compounds or pools, and by parsing active from dormant microbial populations. For example, the
methods by which carbon enters the soil may also be distinguished by tracking different distri-
bution profiles of naturally occurring isotopes—for example, photosynthate from C3 plants has
a 13 C:12 C ratio of ∼27%, whereas that of C4 plants is closer to 12% (140). As a result, measur-
ing these ratios can specify the metabolic pathways by which carbon is cycled through plants and
microbes, and subsequently whether it is entombed as microbial necromass or respired as CO2
(Figure 2).
Stable isotope tracers have been used to characterize the mechanisms driving soil carbon
turnover in soils, as well as informing their representations in predictive models. In one tech-
nique, pulse-chase labeling, communities are successively supplied with a heavy isotope-labeled
compound, then an unlabeled version of the same compound, allowing real-time tracking of the
label over time. For instance, 13 C isotopic tracers have demonstrated microbial biomass contribu-
tions to stabilized organic matter (141), as well as revealed bacterial versus fungal contributions to
soil organic matter formation (142). Relative bioavailability (turnover rate, carbon use efficiency)
of different carbon molecules will impact tracer rates. Residual microbial necromass is of partic-
ular interest, given its role in belowground carbon storage. To investigate this area, stable isotope
probing has been used to track how carbon assimilated as microbial biomass is subsequently min-
eralized and recycled into other organisms after cell death (143, 144). Stable isotope probing can
also be used to conduct real-time analyses of how soil carbon flux is affected by perturbations
(145), in particular with respect to climate disturbances.
A new methodology, termed quantitative stable isotope probing (qSIP), can quantify taxon-
specific microbial contributions to biogeochemical processes (146, 147). One benefit of qSIP is
that it enables detection of phylogenetic traits that can be used in predictive frameworks of mi-
crobial ecology in complex systems. For example, by tracking the response of multiple popula-
tions concurrently, qSIP can measure relative growth rates or carbon assimilation rates across
all members of a community (148). This push beyond targeting isotopes in fatty acids that are

www.annualreviews.org • Soil Microbiomes and Climate Change 43

 CO2

Leaf and root

litter
3 Efflux

1 Complex compounds
Lignin, cellulose,
hemicellulose 1 Simple compounds
Plant and microbial
2 Microbial factory

Microbial
community

4 Stable carbon pool

Partly decomposed plant material
Extracellular metabolites
Microbial necromass
Entombing effect

Figure 2
Soil carbon (C) cycle through the microbial loop. Carbon dioxide (CO2 ) in the atmosphere is fixed by plants
(or autotrophic microorganisms) and added to soil through processes such as ● 1 root exudation of

low-molecular weight simple carbon compounds, or deposition of leaf and root litter leading to accumulation
of complex plant polysaccharides. ● 2 Through these processes, carbon is made bioavailable to the microbial

metabolic “factory” and subsequently is either ●3 respired to the atmosphere or ●4 enters the stable carbon

pool as microbial necromass. The exact balance of carbon efflux versus persistence is a function of several
factors, including aboveground plant community composition and root exudate profiles, environmental
variables, and collective microbial phenotypes [i.e., the metaphenome (19)]. Figure inspired by 197.

coarse biomarkers or using nucleic acid probes (Chip-SIP) provides an unprecedented opportunity
for comprehending how specific microbial clades impact ecosystem-level biogeochemistry. As an
example, one recent study used 13 C SIP to predict that rhizosphere-associated Saccharibacteria fer-
ment root exudates and microbial necromass (149), providing a new conceptual model of above-
and belowground carbon cycle linkages.
From an ecosystem perspective, molecularly resolved stable isotopes help to explain environ-
mental responses to perturbation. Pulse-chase isotope experiments have been used to demonstrate
that wet-dry cycles impact soil organic matter stabilization through rhizosphere processes mod-
ulated by plant growth (150), that root-associated communities implicated in drought survival
have roles in regulating rhizodeposition and carbon accumulation in deep soils (151), and that
photosynthate transfer from plants to soil microbes is inhibited under drought (152). Such studies
will serve to upscale these microbial stress-response mechanisms into predictive ecosystem
models.

44 Naylor et al.
 Finally, advances in isotopically informed metabolomics have shed light on the function of
specific microbial/biochemical pathways for soil carbon fates. For example, natural abundance
13
C nuclear magnetic resonance (NMR) spectroscopy has identified the solutes accumulated by
bacteria under osmotic stress (153). The nondestructive nature of NMR spectroscopy is a clear
advantage for biological studies and may be used to identify organic solutes in intact cells or cell
extracts. Additionally, metabolites that are labeled at specific positions can be used to track ac-
tive processes in soil microbial communities. Through tracking production of isotopically labeled
carbon after addition of differentially labeled glucose and pyruvate, a study calculated carbon flux
rates through specific metabolic pathways (154). Carbon flux patterns can then be used to calcu-
late carbon-use efficiency, energy production and usage, substrate metabolism, nitrogen demand,
and oxygen consumption (155).
Advances in detecting isotopic signatures across molecular data types have advanced mecha-
nistic understanding of belowground carbon storage in highly dynamic systems. From an experi-
mental perspective, stable isotopes have the potential to link all aspects of environmental systems.
For example, 13 C-labeled photosynthate might be tracked through roots, mycorrhizal associa-
tions, bacteria, nematodes, plant roots, leaves, litter, and eventually saprotrophs. In total, these
approaches have the potential to not only disentangle belowground processes, but also increase
model accuracy.

8. MODELING CLIMATE CHANGE IMPACTS ON SOIL CARBON

An International Soil Carbon Network was recently established to identify gaps in SOC model-
ing (156). One of the biggest challenges identified was detection of changes in SOC, due to both
its spatiotemporal variation across soil ecosystems and an incomplete understanding of the pro-
cesses governing whether SOC is stabilized or decomposed. Ideally, models would be derived from
mechanistic understandings of SOC dynamics, but most are instead based on simulations, due to
challenges in obtaining empirical data and measuring SOC (156). Examples of current research
priorities include understanding of SOC dynamics in soil (micro)aggregate microenvironments
and how priming influences soil carbon turnover (156). Eventually, integration of mechanistic in-
sights from molecular data into climate models will better predict the fate of soil carbon under
climate change.
Another area that needs to be addressed is inclusion of climate-relevant microbial processes.
Most climate models assume that soil organic matter decomposition is a first-order decay pro-
cess between conceptual pools. In 2009 there were 33 SOC models represented within the Global
Change and Terrestrial Ecosystems Soil Organic Matter Network database, and 30 of those were
multicompartment, process-based models (157), in which decay rates are typically expressed as a
function of carbon concentration and a rate constant. Although global models incorporate infor-
mation about soil and climate properties (4), microbial processes may not be included in first-order
assumptions (157). Upon their inclusion, however, the predictive ability for SOC fate under cli-
mate is demonstrably improved (158). This has resulted in continued development of improved
Earth System Models (ESMs) that integrate microbial influences on SOC flux (4, 157), and new
models for linking decomposition to the size and activity of the soil microbiome (159). These
developments highlight the importance of second-order processes (microbial activities for SOC
transformation) for predicting SOC flux either as microbial biomass or as respiratory loss to the
atmosphere as CO2 .
Models aim to predict how climate warming will impact soil-sourced greenhouse gas emission
in the future, which necessitates empirical determinations of the extent of soil carbon feedbacks.
However, climate predictions may be based on outdated soil models that do not reflect the current

www.annualreviews.org • Soil Microbiomes and Climate Change 45

 scientific consensus on soil carbon formation and stabilization (160). For example, although SOC
is the result of net outputs (respiration) and inputs (carbon fixation) of plant-derived carbon, the
majority of empirical data have focused on outputs alone, failing to account for possible compen-
satory effects such as heightened soil carbon formation (160). A balance of carbon outputs and
inputs is captured by ESMs (158), but is not yet widely included in global predictions (161).
By contrast, models on SOC flux have begun to include aspects of the plant-soil ecosystem,
including plant types and mineral interactions, which may have variable effects on SOC flux de-
pending on specific circumstances. The CORPSE (Carbon, Organisms, Rhizosphere and Pro-
tection in the Soil Environment) model includes aspects of priming and soil protection, which
promote soil decomposition and carbon storage, respectively (12). However, upon obtaining em-
pirical data, they found contrasting results from the two soil warming experiments: At one site
(Oak Ridge, Tennessee), carbon stabilization in the soil exceeded SOC loss from priming under
warming, whereas at a separate site (FACE at Duke Forest, North Carolina) the opposite trend
was found, resulting in net SOC loss (12). These simulations demonstrated increased CO2 levels
stimulated priming to a greater extent than carbon storage, which will yield a net global carbon
loss under climate change. Other models have incorporated information on plant functional types
(e.g., C3 versus C4 grasses, broadleaf versus needleleaf) that in turn distinguish between plant soil
inputs (157). Recently a new model (MEMS v1.0) was proposed, linking soil organic matter chem-
istry with both microbial processing and interactions with soil minerals, to improve climate model
predictions (162).
On a related note, a modeling approach has recently been proposed that takes into account mi-
crobial life strategies (64). Although soil microbial life strategies have typically been assigned to
two categories—fast-growing r-strategists and slower-growing, energy-conserving K-strategists—
the new model splits life strategies into three categories: Y for growth yield, A for resource ac-
quisition, and S for stress tolerance. Each of these three categories would represent an advantage
under a different set of environmental conditions and availabilities, such that it would be unlikely
for a microbe to belong to more than one (64). Furthermore, as each category has a distinct pro-
file for carbon use, validating this framework will help to predict overall microbial carbon cycling
rates and active processes.
Further needs for improving SOC models include obtaining empirical data from across scales,
for example laboratory and field experiments (Figure 3). Currently, interpolation between systems
of widely varying spatial resolution is challenging, as information gleaned at one scale is often
erroneous when applied to others (163). For example, soil carbon fluxes cannot be measured at
scales larger than 1 m2 (164), which is a hurdle when attempting to extrapolate to landscape-scale
processes.

8.1. Metabolic and Network Modeling of Soil Microbes and the Carbon Cycle
Although the above SOC models focus solely on the biogeochemical turnover of bulk/coarse-
grained carbon pools, it is also possible to model the influence of microbe-microbe interactions
on carbon cycle metabolic processes and their response to climate change. Current experiments
can collect tens to hundreds of omics datasets to infer feature coexpression networks (137). Al-
though specific methods vary, all networks are based on linking pairs of features (e.g., species,
transcripts, proteins, metabolites) across variable environmental conditions, where linkages are
based on similarities in expression or abundance. Such networks can reveal characteristics of
the system in question, including how features are related to each other by expression or abun-
dance, which functional processes are coordinated, which features occupy central positions in the
growth and bioactivity of the system, which transcripts/proteins act in regulatory pathways, what

46 Naylor et al.
Transcriptomic data Coexpressed
transcript network Outcomes of networks Examples
Species A
Multiomic network of Identification of critical Red squares or blue circles with
transcripts/proteins transcripts/proteins multiple edges (gray or orange
lines)

Viewing how processes/ Identifying functional groups of

Species B pathways are related proteins/transcripts that are
linked

Where are the links Edges connecting transcripts

between species? from different species (varying
blue circles)

How does all of this respond Inference of multiple networks

as a function of climate reflecting control and climate
change? change data, comparison of
Coexpressed protein networks to identify differences
Proteomic data
network

Figure 3
Feature coexpression networks can be based on multiple types of omics data. Coexpression networks from different data types
(metatranscriptomics, metaproteomics, etc.) allow for a more refined understanding of how functions respond across expression levels.
Questions that may be answered through this approach (as described in the box on the right) include which features (proteins,
transcripts, metabolites, etc.) are coexpressed, the identity of keystone features central for system functioning, and relatedness among
species or metabolic pathways. Furthermore, by gathering omics data from a system under different sets of conditions (e.g., climate
change perturbations), one can visualize which aspects of the network shift and therefore how overall functioning will be impacted as a
consequence of disturbance.

putative functions of uncharacterized genes may be, and more. Co-occurrence networks for micro-
bial species abundance have been inferred for many different systems, including soil microbiomes
under climate stresses (165–167), often finding that factors such as similarities in pH, aridity, and
nutrient preferences lead to interspecies interactions. Thus, co-occurrence analyses can identify
groups able to withstand climate change by linking the species with known stress resistance profiles
(84).
Co-occurrence networks may also be used to identify keystone members central to soil func-
tion. In this analysis, microbial species are represented as nodes within a network, and centrality
measures are determined for each, where centrality is usually estimated by degree (number of
edges a node shares with other nodes) or betweenness (how important a node is for connect-
ing two other nodes) (168). Nodes of high centrality measures may be thought of as keystone
nodes, as their removal would have deleterious effects on network structure and by extension soil
functioning. This analysis has been successfully applied to soil microbiome networks to identify
keystone taxa and factors influencing community structure (169, 170). New, emerging methods
take advantage of multiple omics data types to link microbes and metabolites, or to link genes
between microbial species to identify coordination of metabolism across species (171). Others ap-
ply a network approach to transcripts and proteins, determining which processes and pathways
might be coexpressed under disturbances (168) (Figure 3). Such network approaches will inform
how interkingdom interactions, and by extension microbiome stability, are impacted by climate
change–related disturbances.

8.2. Importance of Cross-Scale Experimentation and Data-Model Integration

Chemical transformations that occur in soil are greatly influenced or even controlled by processes
occurring at the micrometer to nanometer scale (microscale hereafter). In a physical context,

www.annualreviews.org • Soil Microbiomes and Climate Change 47

 transformations in soil aggregate pore spaces, down to the very mineral interfaces composing
said aggregates, strongly govern larger soil processes (172). Beyond this, evidence has shown
how the microscale size and position of aggregates influence their resident microbial community
composition (173, 174). The role of geography on soil microbial diversity has been confirmed,
whether across countries (175, 176), a continent (177), or the globe (178, 179)—all while revealing
the role that local microscale constraints have on microbial processes.
The biogeochemical reactions occurring at the microscale are orders of magnitude smaller
than their corresponding ecosystem-scale effect size. This fact is more striking when we con-
sider that soil biogeochemistry is largely regulated by microbial metabolism that is, in turn, fu-
eled by exchange of electrons. Microbial metabolic interactions reflect how the hyperdiverse soil
microbiome responds to its physical and chemical landscape: Historical conditions influence mi-
crobial responses to novel conditions created by compounding disturbances [e.g., hydrological
moments of wetting and aqueous phase connectedness (180)]. A central challenge in soil micro-
biology is understanding the emergent, nonadditive effects of microscale microbial metabolism
at the ecosystem scale or larger. Complicating matters further, these processes not only have a
physical space dependence but a temporal one as well. Seasonal and subseasonal variability in soil
microbial properties continues to be resolved (181), as well as multiyear trends (182). Combined,
the spatiotemporal result of microscale reactions has a cumulative outcome disproportionate to
its effective size, as evidenced by the microbial influence on Earth’s atmosphere (183). However, it
remains a significant challenge to incorporate microscale processes into models that are relevant
at larger scales (Figure 4).
In the field context, measuring microscale biochemical reactions in the soil remains unattain-
able. As a result, our understanding of soil ecology is based on research conducted along a con-
tinuum of scales, from in vitro studies of pure microbial cultures (184) to ecosystem experiments
replicated around the globe (185). One approach undertaken by multiple groups is development
of mesocosms or microfabricated devices that both emulate the soil microenvironment as well
as directly visualize microbial processes (186–189). The use of rhizobox-like and universal meso-
cosms (190), in conjunction with transparent material mimicking soil structure and gas exchange
properties (189, 191), will help bridge this continuum of scales. Ultimately, the frontier of soil
microbial ecology is to connect the knowledge and understanding across these scales (Figure 4),
for example corroborating lab-based and field-based studies. Furthermore, it will be beneficial to
incorporate novel analytical methods such as SIP, reduced-complexity consortia, and other new
omics tools (192, 193) to elucidate the mathematical relationships governing microbial population
dynamics.
Beyond conducting studies at a variety of scales and resolution, coupling empirical with com-
putational efforts will ultimately improve confidence in predictive models relating the soil micro-
biome with ecosystem function (5, 19). Recently, to understand the influence of soil moisture at the
cellular scale, a tandem empirical-computational strategy was used to demonstrate how the dis-
tribution of water in unsaturated soil systems impacts the range and frequency of microbial inter-
actions (189). This concept has been further supported by microscale imaging that demonstrates
fungal hyphae prefer to grow in air-filled soil pores rather than saturated habitats (188). Such con-
trolled model-experimental studies can be linked to similar soil incubation and field studies. For
example, individual-based models for microbial community activity in soil aggregates demonstrate
the importance of hydration dynamics in regulating carbon use and greenhouse gas emissions un-
der field conditions (194). This approach elegantly scales the microbial habitat (soil pores) to soil
structure (aggregates) to simulate and validate soil biogeochemical processes (ecosystem scale).
Together, these studies demonstrate the power of designing experiments where empirical data are
generated to test and parameterize computational models.

48 Naylor et al.
 a Landscape models b Soil carbon models c Agent-based models d Molecular pathways/
metabolic maps

L a n d sca pe Soil core M i c ro sc o pi c Omics

Hydrology

Soil type

Climate

Climate Gas flux Reduced-complexity Biochemical/

variables measurements consortia genomes

Figure 4
A central challenge in predicting climate change effects is integrating predictions derived from models across different scales of
resolution. (a) Landscape-scale models couple hydrologic, climatic, and soil-ecosystem models (198). (b) Soil carbon models (e.g., 158)
integrate simulations of overall microbial activity and greenhouse gas flux from aboveground, surface, and subsurface soil horizons.
(c) Agent-based models, which detail processes occurring at the microbial population scale, allow for both elucidation of what
properties of a system are attributable to individual groups of microbes and what are emergent (169, 199, 200). (d) Molecular
pathways/metabolic maps (e.g., metagenomics, metatranscriptomics, metaproteomics, metabolomics) can be used to model soil
microbial phylogenetic and phenotypic responses to change, including elucidating which metabolic pathways are represented in a soil
system and the rates at which they occur.

9. CONCLUSIONS
A recent call to action emphasized the importance of understanding environmental microorgan-
isms in the face of climate change (2). Abundant evidence reveals that soil microbes are affected
by climate change–associated disturbances with important feedbacks to ecosystem health and cli-
mate forcing. Under these disturbances, shifts in microbial community composition and function
will in turn have repercussions for interkingdom interactions, biogeochemical cycling, and carbon
flow, in ways that may exacerbate or attenuate climate change. As we begin to fully understand key
roles carried out by microorganisms inhabiting soil ecosystems, this knowledge may be used to
predict how critical metabolic processes are impacted by environmental change, and furthermore
may be leveraged for mitigation of negative aspects of climate change (5).

SUMMARY POINTS
1. Soil microbiome stability in the face of external perturbations is a combination of resis-
tance, resilience, and functional redundancy. Incorporating all three metrics into studies
will provide more comprehensive analyses of microbiome disturbances.

www.annualreviews.org • Soil Microbiomes and Climate Change 49

 2. Historical conditions influence microbial responses to novel conditions created by com-
pounding disturbances.
3. Interkingdom interactions influence population dynamics and metabolic exchanges with
repercussions for soil carbon and nitrogen cycling and persistence.
4. Chemical weathering of rocks alters soil mineral profiles, and the resultant microbial
community shifts will have impacts on soil carbon mineralization and storage.
5. Studying soil responses to disturbance will benefit from integrating disparate data
streams through multi-omics approaches.
6. Stable isotope probing yields quantitative insights into biogeochemical transformations
and population dynamics within the soil microbiome.
7. Acquiring sufficient empirical data on carbon cycling–relevant soil microbial processes
and networks, across a range of scales, is necessary to improve predictive climate models.

FUTURE ISSUES
1. What are the factors that influence the balance between soil carbon storage and
respiration?
2. How does the combination of various climate change–related factors (eCO2, soil warm-
ing, drought, etc.) interact to affect soil microbiomes, both directly and indirectly?
3. In what way are interkingdom interactions and their associated carbon flow impacted by
climate change factors?
4. To what extent does functional redundancy play a role in resistance and resilience to
climate change?
5. To what extent are organo-mineral interactions and the associated mineral carbon sta-
bilization impacted by the shifting microbial communities under climate change?
6. How can we integrate the multiple data streams from different omics approaches to-
gether, in a cost-effective and computationally efficient manner, for higher-resolution
study of microbiomes?
7. How will we be able to identify the correct parameters, build appropriate databases, and
integrate omics data to improve climate models?

DISCLOSURE STATEMENT
The authors are not aware of any affiliations, memberships, funding, or financial holdings that
might be perceived as affecting the objectivity of this review.

LITERATURE CITED
1. Amundson R, Berhe AA, Hopmans JW, Olson C, Sztein AE, Sparks DL. 2015. Soil and human security
in the 21st century. Science 348(6235):1261071
2. Cavicchioli R, Ripple WJ, Timmis KN, Azam F, Bakken LR, et al. 2019. Scientists’ warning to humanity:
microorganisms and climate change. Nat. Rev. Microbiol. 17(9):569–86

50 Naylor et al.
 3. Lal R. 2004. Soil carbon sequestration impacts on global climate change and food security. Science
304(5677):1623–27
4. Campbell EE, Paustian K. 2015. Current developments in soil organic matter modeling and the expan-
sion of model applications: a review. Environ. Res. Lett. 10(12):123004
5. Jansson JK, Hofmockel KS. 2020. Soil microbiomes and climate change. Nat. Rev. Microbiol. 18:35–46
6. Thompson LR, Sanders JG, McDonald D, Amir A, Ladau J, et al. 2017. A communal catalogue reveals
Earth’s multiscale microbial diversity. Nature 551:457–63
7. Goel R, Kumar V, Suyal DC, Narayan Soni R. 2018. Toward the unculturable microbes for sustainable
agricultural production. In Role of Rhizospheric Microbes in Soil, ed. VS Meena, pp. 107–23. Singapore:
Springer Singapore
8. Schulz F, Alteio L, Goudeau D, Ryan EM, Yu FB, et al. 2018. Hidden diversity of soil giant viruses. Nat.
Commun. 9(1):4881
9. Friedlingstein P, Cox P, Betts R, Bopp L, von Bloh W, et al. 2006. Climate-carbon cycle feedback anal-
ysis: results from the C4 MIP Model Intercomparison. J. Clim. 19(14):3337–53
10. Kuzyakov Y, Horwath WR, Dorodnikov M, Blagodatskaya E. 2019. Review and synthesis of the effects
of elevated atmospheric CO2 on soil processes: no changes in pools, but increased fluxes and accelerated
cycles. Soil Biol. Biochem. 128:66–78
11. Liang C, Schimel JP, Jastrow JD. 2017. The importance of anabolism in microbial control over soil
carbon storage. Nat. Microbiol. 2(8):17105
12. Sulman BN, Phillips RP, Oishi AC, Shevliakova E, Pacala SW. 2014. Microbe-driven turnover offsets
mineral-mediated storage of soil carbon under elevated CO2 . Nat. Clim. Change 4(12):1099–102
13. Harmon ME, Silver WL, Fasth B, Chen H, Burke IC, et al. 2009. Long-term patterns of mass loss during
the decomposition of leaf and fine root litter: an intersite comparison. Glob. Change Biol. 15(5):1320–
38
14. Qualls RG. 2016. Long-term (13 years) decomposition rates of forest floor organic matter on paired
coniferous and deciduous watersheds with contrasting temperature regimes. Forests 7(10):231
15. Lange M, Eisenhauer N, Sierra CA, Bessler H, Engels C, et al. 2015. Plant diversity increases soil mi-
crobial activity and soil carbon storage. Nat. Commun. 6(1):6707
16. Stewart CE, Paustian K, Conant RT, Plante AF, Six J. 2007. Soil carbon saturation: concept, evidence
and evaluation. Biogeochemistry 86(1):19–31
17. Joergensen RG, Wichern F. 2018. Alive and kicking: why dormant soil microorganisms matter. Soil Biol.
Biochem. 116:419–30
18. Jones SE, Lennon JT. 2010. Dormancy contributes to the maintenance of microbial diversity. PNAS
107(13):5881–86
19. Jansson JK, Hofmockel KS. 2018. The soil microbiome—from metagenomics to metaphenomics. Curr.
Opin. Microbiol. 43:162–68
20. Graham EB, Knelman JE, Schindlbacher A, Siciliano S, Breulmann M, et al. 2016. Microbes as engines
of ecosystem function: When does community structure enhance predictions of ecosystem processes?
Front. Microbiol. 7:214
21. Carini P, Marsden PJ, Leff JW, Morgan EE, Strickland MS, Fierer N. 2017. Relic DNA is abundant in
soil and obscures estimates of soil microbial diversity. Nat. Microbiol. 2(3):16242
22. Balesdent J, Chenu C, Balabane M. 2000. Relationship of soil organic matter dynamics to physical pro-
tection and tillage. Soil Tillage Res. 53(3–4):215–30
23. Marschner B, Brodowski S, Dreves A, Gleixner G, Gude A, et al. 2008. How relevant is recalcitrance for
the stabilization of organic matter in soils? J. Plant Nutr. Soil Sci. 171(1):91–110
24. Cotrufo MF, Wallenstein MD, Boot CM, Denef K, Paul E. 2013. The Microbial Efficiency-Matrix
Stabilization (MEMS) framework integrates plant litter decomposition with soil organic matter stabi-
lization: Do labile plant inputs form stable soil organic matter? Glob. Change Biol. 19(4):988–95
25. Schmidt MWI, Torn MS, Abiven S, Dittmar T, Guggenberger G, et al. 2011. Persistence of soil organic
matter as an ecosystem property. Nature 478(7367):49–56
26. Erich MS, Plante AF, Fernández JM, Mallory EB, Ohno T. 2012. Effects of profile depth and manage-
ment on the composition of labile and total soil organic matter. Soil Sci. Soc. Am. J. 76(2):408–19

www.annualreviews.org • Soil Microbiomes and Climate Change 51

 27. Bach EM, Hofmockel KS. 2016. A time for every season: soil aggregate turnover stimulates decompo-
sition and reduces carbon loss in grasslands managed for bioenergy. GCB Bioenergy 8(3):588–99
28. Vurukonda SSKP, Vardharajula S, Shrivastava M, SkZ A. 2016. Enhancement of drought stress tolerance
in crops by plant growth promoting rhizobacteria. Microbiol. Res. 184:13–24
29. Keiluweit M, Wanzek T, Kleber M, Nico P, Fendorf S. 2017. Anaerobic microsites have an unaccounted
role in soil carbon stabilization. Nat. Commun. 8(1):1771
30. Allison SD, Martiny JBH. 2008. Resistance, resilience, and redundancy in microbial communities. PNAS
105(Supplement 1):11512–19
31. Orwin KH, Wardle DA. 2005. Plant species composition effects on belowground properties and the
resistance and resilience of the soil microflora to a drying disturbance. Plant Soil 278(1–2):205–21
32. Griffiths BS, Philippot L. 2013. Insights into the resistance and resilience of the soil microbial commu-
nity. FEMS Microbiol. Rev. 37(2):112–29
33. Shade A, Peter H, Allison SD, Baho DL, Berga M, et al. 2012. Fundamentals of microbial community
resistance and resilience. Front. Microbiol. 3:417
34. Azarbad H, van Gestel C, Niklińska M, Laskowski R, Röling W, van Straalen N. 2016. Resilience of soil
microbial communities to metals and additional stressors: DNA-based approaches for assessing “stress-
on-stress” responses. Int. J. Mol. Sci. 17(6):933
35. Naylor D, Coleman-Derr D. 2018. Drought stress and root-associated bacterial communities. Front.
Plant Sci. 8:2223
36. van Kruistum H, Bodelier PLE, Ho A, Meima-Franke M, Veraart AJ. 2018. Resistance and recovery
of methane-oxidizing communities depends on stress regime and history; a microcosm study. Front.
Microbiol. 9:1714
37. Nguyen LTT, Broughton K, Osanai Y, Anderson IC, Bange MP, et al. 2019. Effects of elevated tem-
perature and elevated CO2 on soil nitrification and ammonia-oxidizing microbial communities in field-
grown crop. Sci. Total Environ. 675:81–89
38. Fierer N, Lauber CL, Ramirez KS, Zaneveld J, Bradford MA, Knight R. 2012. Comparative metage-
nomic, phylogenetic and physiological analyses of soil microbial communities across nitrogen gradients.
ISME J. 6(5):1007–17
39. Leff JW, Jones SE, Prober SM, Barberán A, Borer ET, et al. 2015. Consistent responses of soil microbial
communities to elevated nutrient inputs in grasslands across the globe. PNAS 112(35):10967–72
40. Hueso S, García C, Hernández T. 2012. Severe drought conditions modify the microbial community
structure, size and activity in amended and unamended soils. Soil Biol. Biochem. 50:167–73
41. Schaeffer SM, Homyak PM, Boot CM, Roux-Michollet D, Schimel JP. 2017. Soil carbon and nitrogen
dynamics throughout the summer drought in a California annual grassland. Soil Biol. Biochem. 115:54–
62
42. McKew BA, Taylor JD, McGenity TJ, Underwood GJC. 2011. Resistance and resilience of ben-
thic biofilm communities from a temperate saltmarsh to desiccation and rewetting. ISME J. 5(1):30–
41
43. Hinojosa MB, Parra A, Laudicina VA, Moreno JM. 2016. Post-fire soil functionality and microbial com-
munity structure in a Mediterranean shrubland subjected to experimental drought. Sci. Total Environ.
573:1178–89
44. Simonin M, Nunan N, Bloor JMG, Pouteau V, Niboyet A. 2017. Short-term responses and resistance
of soil microbial community structure to elevated CO2 and N addition in grassland mesocosms. FEMS
Microbiol. Lett. 364(9):fnx077
45. Orwin KH, Dickie IA, Wood JR, Bonner KI, Holdaway RJ. 2016. Soil microbial community structure
explains the resistance of respiration to a dry-rewet cycle, but not soil functioning under static conditions.
Funct. Ecol. 30(8):1430–39
46. Jacquiod S, Nunes I, Brejnrod A, Hansen MA, Holm PE, et al. 2018. Long-term soil metal exposure
impaired temporal variation in microbial metatranscriptomes and enriched active phages. Microbiome
6(1):223
47. Jacquiod S, Franqueville L, Cécillon S, Vogel TM, Simonet P. 2013. Soil bacterial community shifts
after chitin enrichment: an integrative metagenomic approach. PLOS ONE 8(11):e79699

52 Naylor et al.
48. Steven B, Belnap J, Kuske CR. 2018. Chronic physical disturbance substantially alters the response
of biological soil crusts to a wetting pulse, as characterized by metatranscriptomic sequencing. Front.
Microbiol. 9:2382
49. Lourenço KS, Suleiman AKA, Pijl A, van Veen JA, Cantarella H, Kuramae EE. 2018. Resilience of the
resident soil microbiome to organic and inorganic amendment disturbances and to temporary bacterial
invasion. Microbiome 6(1):142
50. Knelman J, Schmidt S, Garayburu-Caruso V, Kumar S, Graham E. 2019. Multiple, compounding distur-
bances in a forest ecosystem: Fire increases susceptibility of soil edaphic properties, bacterial community
structure, and function to change with extreme precipitation event. Soil Syst. 3(2):40
51. Melillo JM, Frey SD, DeAngelis KM, Werner WJ, Bernard MJ, et al. 2017. Long-term pattern and
magnitude of soil carbon feedback to the climate system in a warming world. Science 358(6359):101–
5
52. Pold G, Billings AF, Blanchard JL, Burkhardt DB, Frey SD, et al. 2016. Long-term warming alters
carbohydrate degradation potential in temperate forest soils. Appl. Environ. Microbiol. 82(22):6518–
30
53. Drake JE, Gallet-Budynek A, Hofmockel KS, Bernhardt ES, Billings SA, et al. 2011. Increases in the
flux of carbon belowground stimulate nitrogen uptake and sustain the long-term enhancement of forest
productivity under elevated CO2 : C fluxes belowground and long-term FACE productivity. Ecol. Lett.
14(4):349–57
54. Jiao S, Chen W, Wei G. 2019. Resilience and assemblage of soil microbiome in response to chemical
contamination combined with plant growth. Appl. Environ. Microbiol. 85(6):e02523-18
55. Sheik CS, Beasley WH, Elshahed MS, Zhou X, Luo Y, Krumholz LR. 2011. Effect of warming and
drought on grassland microbial communities. ISME J. 5(10):1692–1700
56. Morrissey EM, Mau RL, Hayer M, Liu X-JA, Schwartz E, et al. 2019. Evolutionary history constrains
microbial traits across environmental variation. Nat. Ecol. Evol. 3(7):1064–69
57. Fierer N, Bradford MA, Jackson RB. 2007. Toward an ecological classification of soil bacteria. Ecology
88(6):1354–64
58. Wertz S, Degrange V, Prosser JI, Poly F, Commeaux C, et al. 2007. Decline of soil microbial diversity
does not influence the resistance and resilience of key soil microbial functional groups following a model
disturbance. Environ. Microbiol. 9(9):2211–19
59. Chaer G, Fernandes M, Myrold D, Bottomley P. 2009. Comparative resistance and resilience of soil
microbial communities and enzyme activities in adjacent native forest and agricultural soils. Microb. Ecol.
58(2):414–24
60. Ho A, Lüke C, Reim A, Frenzel P. 2016. Resilience of (seed bank) aerobic methanotrophs and methan-
otrophic activity to desiccation and heat stress. Soil Biol. Biochem. 101:130–38
61. Puglisi E, Hamon R, Vasileiadis S, Coppolecchia D, Trevisan M. 2012. Adaptation of soil microorganisms
to trace element contamination: a review of mechanisms, methodologies, and consequences for risk
assessment and remediation. Crit. Rev. Environ. Sci. Technol. 42(22):2435–70
62. Udiković-Kolić N, Devers-Lamrani M, Petrić I, Hršak D, Martin-Laurent F. 2011. Evidence for tax-
onomic and functional drift of an atrazine-degrading culture in response to high atrazine input. Appl.
Microbiol. Biotechnol. 90(4):1547–54
63. Bárcenas-Moreno G, Bååth E. 2009. Bacterial and fungal growth in soil heated at different temperatures
to simulate a range of fire intensities. Soil Biol. Biochem. 41(12):2517–26
64. Malik AA, Martiny JBH, Brodie EL, Martiny AC, Treseder KK, Allison SD. 2020. Defining trait-based
microbial strategies with consequences for soil carbon cycling under climate change. ISME J. 14:1–9
65. Yu H, Deng Y, He Z, Van Nostrand JD, Wang S, et al. 2018. Elevated CO2 and warming altered grassland
microbial communities in soil top-layers. Front. Microbiol. 9:1790
66. Dijkstra FA, Blumenthal D, Morgan JA, Pendall E, Carrillo Y, Follett RF. 2010. Contrasting effects of
elevated CO2 and warming on nitrogen cycling in a semiarid grassland. New Phytol. 187(2):426–37
67. Hu H-W, Macdonald CA, Trivedi P, Anderson IC, Zheng Y, et al. 2016. Effects of climate warming and
elevated CO2 on autotrophic nitrification and nitrifiers in dryland ecosystems. Soil Biol. Biochem. 92:1–15

www.annualreviews.org • Soil Microbiomes and Climate Change 53

 68. Solomon S, Intergov. Panel Clim. Change, eds. 2007. Climate Change 2007: The Physical Science Basis:
Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate
Change. New York: United Nations Environ. Progr.
69. Pold G, Melillo JM, DeAngelis KM. 2015. Two decades of warming increases diversity of a potentially
lignolytic bacterial community. Front. Microbiol. 6:480
70. Xue K, Xie J, Zhou A, Liu F, Li D, et al. 2016. Warming alters expressions of microbial functional genes
important to ecosystem functioning. Front. Microbiol. 7:668
71. Rocca JD, Simonin M, Blaszczak JR, Ernakovich JG, Gibbons SM, et al. 2019. The Microbiome Stress
Project: toward a global meta-analysis of environmental stressors and their effects on microbial com-
munities. Front. Microbiol. 9:3272
72. Carrell AA, Kolton M, Glass JB, Pelletier DA, Warren MJ, et al. 2019. Experimental warming alters the
community composition, diversity, and N2 fixation activity of peat moss (Sphagnum fallax) microbiomes.
Glob. Change Biol. 25:2993–3004
73. DeAngelis KM, Pold G, Topçuoğlu BD, van Diepen LTA, Varney RM, et al. 2015. Long-term forest soil
warming alters microbial communities in temperate forest soils. Front. Microbiol. 6:104
74. Li Y, Lv W, Jiang L, Zhang L, Wang S, et al. 2019. Microbial community responses reduce soil carbon
loss in Tibetan alpine grasslands under short-term warming. Glob. Change Biol. 25:3438–49
75. Zhou J, Xue K, Xie J, Deng Y, Wu L, et al. 2012. Microbial mediation of carbon-cycle feedbacks to
climate warming. Nat. Clim. Change 2(2):106–10
76. Frey SD, Ollinger S, Nadelhoffer K, Bowden R, Brzostek E, et al. 2014. Chronic nitrogen additions
suppress decomposition and sequester soil carbon in temperate forests. Biogeochemistry 121(2):305–
16
77. Lladó S, López-Mondéjar R, Baldrian P. 2017. Forest soil bacteria: diversity, involvement in ecosystem
processes, and response to global change. Microbiol. Mol. Biol. Rev. 81(2):e00063-16
78. Xiong J, He Z, Shi S, Kent A, Deng Y, et al. 2015. Elevated CO2 shifts the functional structure and
metabolic potentials of soil microbial communities in a C4 agroecosystem. Sci. Rep. 5(1):9316
79. Cheng L, Booker FL, Burkey KO, Tu C, Shew HD, et al. 2011. Soil microbial responses to elevated
CO2 and O3 in a nitrogen-aggrading agroecosystem. PLOS ONE 6(6):e2137
80. Chen YP, Liu Q, Liu YJ, Jia FA, He XH. 2015. Responses of soil microbial activity to cadmium pollution
and elevated CO2 . Sci. Rep. 4(1):4287
81. Dunbar J, Eichorst SA, Gallegos-Graves LV, Silva S, Xie G, et al. 2012. Common bacterial responses in
six ecosystems exposed to 10 years of elevated atmospheric carbon dioxide: soil bacterial response in six
ecosystems. Environ. Microbiol. 14(5):1145–58
82. Yang S, Zheng Q, Yuan M, Shi Z, Chiariello NR, et al. 2019. Long-term elevated CO2 shifts composition
of soil microbial communities in a Californian annual grassland, reducing growth and N utilization
potentials. Sci. Total Environ. 652:1474–81
83. Ebersberger D, Niklaus PA, Kandeler E. 2003. Long term CO2 enrichment stimulates N-mineralisation
and enzyme activities in calcareous grassland. Soil Biol. Biochem. 35(7):965–72
84. Delgado-Baquerizo M, Maestre FT, Escolar C, Gallardo A, Ochoa V, et al. 2014. Direct and indirect
impacts of climate change on microbial and biocrust communities alter the resistance of the N cycle in
a semiarid grassland. J. Ecol. 102(6):1592–1605
85. Wang P, Marsh EL, Ainsworth EA, Leakey ADB, Sheflin AM, Schachtman DP. 2017. Shifts in microbial
communities in soil, rhizosphere and roots of two major crop systems under elevated CO2 and O3 . Sci.
Rep. 7(1):15019
86. Hofmockel KS, Gallet-Budynek A, McCarthy HR, Currie WS, Jackson RB, Finzi A. 2011. Sources of
increased N uptake in forest trees growing under elevated CO2 : results of a large-scale 15 N study. Glob.
Change Biol. 17(11):3338–50
87. Dunbar J, Gallegos-Graves LV, Steven B, Mueller R, Hesse C, et al. 2014. Surface soil fungal and bacte-
rial communities in aspen stands are resilient to eleven years of elevated CO2 and O3 . Soil Biol. Biochem.
76:227–34
88. Formánek P, Rejšek K, Vranová V. 2014. Effect of elevated CO2 , O3 , and UV radiation on soils. Sci.
World J. 2014:730149

54 Naylor et al.
 89. Kardol P, Cregger MA, Campany CE, Classen AT. 2010. Soil ecosystem functioning under climate
change: plant species and community effects. Ecology 91(3):767–81
90. Hayden HL, Mele PM, Bougoure DS, Allan CY, Norng S, et al. 2012. Changes in the microbial commu-
nity structure of bacteria, archaea and fungi in response to elevated CO2 and warming in an Australian
native grassland soil: climate change effects on microbial communities. Environ. Microbiol. 14(12):3081–
96
91. Tveit AT, Urich T, Frenzel P, Svenning MM. 2015. Metabolic and trophic interactions modulate
methane production by Arctic peat microbiota in response to warming. PNAS 112(19):E2507–16
92. Bouskill NJ, Wood TE, Baran R, Ye Z, Bowen BP, et al. 2016. Belowground response to drought in a
tropical forest soil. I. Changes in microbial functional potential and metabolism. Front. Microbiol. 7:525
93. von Rein I, Gessler A, Premke K, Keitel C, Ulrich A, Kayler ZE. 2016. Forest understory plant and
soil microbial response to an experimentally induced drought and heat-pulse event: the importance of
maintaining the continuum. Glob. Change Biol. 22(8):2861–74
94. Pérez Castro S, Cleland EE, Wagner R, Sawad RA, Lipson DA. 2019. Soil microbial responses to drought
and exotic plants shift carbon metabolism. ISME J. 13(7):1776–87
95. Prommer J, Walker TWN, Wanek W, Braun J, Zezula D, et al. 2019. Increased microbial growth,
biomass, and turnover drive soil organic carbon accumulation at higher plant diversity. Glob. Change
Biol. 26:669–81
96. Valencia E, Gross N, Quero JL, Carmona CP, Ochoa V, et al. 2018. Cascading effects from plants to
soil microorganisms explain how plant species richness and simulated climate change affect soil multi-
functionality. Glob. Change Biol. 24(12):5642–54
97. Hünninghaus M, Dibbern D, Kramer S, Koller R, Pausch J, et al. 2019. Disentangling carbon flow across
microbial kingdoms in the rhizosphere of maize. Soil Biol. Biochem. 134:122–30
98. Drigo B, Pijl AS, Duyts H, Kielak AM, Gamper HA, et al. 2010. Shifting carbon flow from roots into
associated microbial communities in response to elevated atmospheric CO2 . PNAS 107(24):10938–
42
99. Maier S, Tamm A, Wu D, Caesar J, Grube M, Weber B. 2018. Photoautotrophic organisms con-
trol microbial abundance, diversity, and physiology in different types of biological soil crusts. ISME
J. 12(4):1032–46
100. Couradeau E, Karaoz U, Lim HC, Nunes da Rocha U, Northen T, et al. 2016. Bacteria increase arid-
land soil surface temperature through the production of sunscreens. Nat. Commun. 7(1):10373
101. Karaoz U, Couradeau E, Nunes da Rocha U, Lim H-C, Northen T, et al. 2018. Large blooms of Bacillales
(Firmicutes) underlie the response to wetting of cyanobacterial biocrusts at various stages of maturity.
mBio 9(2):e01366-16
102. Couradeau E, Giraldo-Silva A, De Martini F, Garcia-Pichel F. 2019. Spatial segregation of the biological
soil crust microbiome around its foundational cyanobacterium, Microcoleus vaginatus, and the formation
of a nitrogen-fixing cyanosphere. Microbiome 7(1):55
103. Rodriguez-Caballero E, Belnap J, Büdel B, Crutzen PJ, Andreae MO, et al. 2018. Dryland photoau-
totrophic soil surface communities endangered by global change. Nat. Geosci. 11(3):185–89
104. Johnson SL, Kuske CR, Carney TD, Housman DC, Gallegos-Graves LV, Belnap J. 2012. Increased tem-
perature and altered summer precipitation have differential effects on biological soil crusts in a dryland
ecosystem. Glob. Change Biol. 18(8):2583–93
105. Rudgers JA, Dettweiler-Robinson E, Belnap J, Green LE, Sinsabaugh RL, et al. 2018. Are fungal net-
works key to dryland primary production? Am. J. Bot. 105(11):1783–87
106. Aanderud ZT, Smart TB, Wu N, Taylor AS, Zhang Y, Belnap J. 2018. Fungal loop transfer of nitrogen
depends on biocrust constituents and nitrogen form. Biogeosciences 15(12):3831–40
107. Otto S, Bruni EP, Harms H, Wick LY. 2017. Catch me if you can: dispersal and foraging of Bdellovibrio
bacteriovorus 109J along mycelia. ISME J. 11(2):386–93
108. Berthold T, Centler F, Hübschmann T, Remer R, Thullner M, et al. 2016. Mycelia as a focal point for
horizontal gene transfer among soil bacteria. Sci. Rep. 6(1):36390
109. Yang P, Oliveira da Rocha Calixto R, van Elsas JD. 2018. Migration of Paraburkholderia terrae BS001
along old fungal hyphae in soil at various pH levels. Microb. Ecol. 76(2):443–52

www.annualreviews.org • Soil Microbiomes and Climate Change 55

 110. Xiong W, Jousset A, Guo S, Karlsson I, Zhao Q, et al. 2018. Soil protist communities form a dynamic
hub in the soil microbiome. ISME J. 12(2):634–38
111. Xiong W, Li R, Guo S, Karlsson I, Jiao Z, et al. 2019. Microbial amendments alter protist communities
within the soil microbiome. Soil Biol. Biochem. 135:379–82
112. Murase J. 2017. Quest of soil protists in a new era. Microbes Environ. 32(2):99–102
113. Trubl G, Jang HB, Roux S, Emergon JB, Solonenko N, et al. 2018. Soil viruses are underexplored players
in ecosystem carbon processing. mSystems 3(5):e00076-18
114. Emerson JB, Roux S, Brum JR, Bolduc B, Woodcroft BJ, et al. 2018. Host-linked soil viral ecology along
a permafrost thaw gradient. Nat. Microbiol. 3(8):870–80
115. Jin M, Guo X, Zhang R, Qu W, Gao B, Zeng R. 2019. Diversities and potential biogeochemical impacts
of mangrove soil viruses. Microbiome 7(1):58
116. Emerson JB. 2019. Soil viruses: a new hope. mSystems 4(3):e00120-19
117. Goudie AS, Viles HA. 2012. Weathering and the global carbon cycle: geomorphological perspectives.
Earth-Sci. Rev. 113(1–2):59–71
118. Thorley RMS, Taylor LL, Banwart SA, Leake JR, Beerling DJ. 2015. The role of forest trees and their
mycorrhizal fungi in carbonate rock weathering and its significance for global carbon cycling: tree-fungal
carbonate rock weathering. Plant Cell Environ. 38(9):1947–61
119. Goll DS, Moosdorf N, Hartmann J, Brovkin V. 2014. Climate-driven changes in chemical weathering
and associated phosphorus release since 1850: implications for the land carbon balance. Geophys. Res.
Lett. 41(10):3553–58
120. Beaulieu E, Goddéris Y, Donnadieu Y, Labat D, Roelandt C. 2012. High sensitivity of the continental-
weathering carbon dioxide sink to future climate change. Nat. Clim. Change 2(5):346–49
121. Don A, Böhme IH, Dohrmann AB, Poeplau C, Tebbe CC. 2017. Microbial community composition
affects soil organic carbon turnover in mineral soils. Biol. Fertil. Soils 53(4):445–56
122. Dwivedi D, Tang J, Bouskill N, Georgiou K, Chacon SS, Riley WJ. 2019. Abiotic and biotic controls
on soil organo-mineral interactions: developing model structures to analyze why soil organic matter
persists. Rev. Mineral. Geochem. 85(1):329–48
123. Long P, Sui P, Gao W, Wang B, Huang J, et al. 2015. Aggregate stability and associated C and N in a
silty loam soil as affected by organic material inputs. J. Integr. Agric. 14(4):774–87
124. Li L, Xu M, Eyakub Ali M, Zhang W, Duan Y, Li D. 2018. Factors affecting soil microbial biomass
and functional diversity with the application of organic amendments in three contrasting cropland soils
during a field experiment. PLOS ONE 13(9):e0203812
125. Fabian J, Zlatanovic S, Mutz M, Premke K. 2017. Fungal-bacterial dynamics and their contribution to
terrigenous carbon turnover in relation to organic matter quality. ISME J. 11(2):415–25
126. Doetterl S, Berhe AA, Arnold C, Bodé S, Fiener P, et al. 2018. Links among warming, carbon and
microbial dynamics mediated by soil mineral weathering. Nat. Geosci. 11(8):589–93
127. Doetterl S, Stevens A, Six J, Merckx R, Van Oost K, et al. 2015. Soil carbon storage controlled by inter-
actions between geochemistry and climate. Nat. Geosci. 8(10):780–83
128. Uroz S, Oger P, Lepleux C, Collignon C, Frey-Klett P, Turpault M-P. 2011. Bacterial weathering and
its contribution to nutrient cycling in temperate forest ecosystems. Res. Microbiol. 162(9):820–31
129. Zaheer R, Noyes N, Ortega Polo R, Cook SR, Marinier E, et al. 2018. Impact of sequencing depth on
the characterization of the microbiome and resistome. Sci. Rep. 8(1):5890
130. Xie K, Deng Y, Zhang X, Wang X, Kang G, et al. 2018. Biases in prokaryotic community amplicon
sequencing affected by DNA extraction methods in both saline and non-saline soil. Front. Microbiol.
9:1796
131. Wooley JC, Godzik A, Friedberg I. 2010. A primer on metagenomics. PLOS Comput. Biol. 6(2):e1000667
132. Shakya M, Lo C-C, Chain PSG. 2019. Advances and challenges in metatranscriptomic analysis. Front.
Genet. 10:904
133. Wolf JBW. 2013. Principles of transcriptome analysis and gene expression quantification: an RNA-seq
tutorial. Mol. Ecol. Resour. 13(4):559–72
134. Callister SJ, Fillmore TL, Nicora CD, Shaw JB, Purvine SO, et al. 2018. Addressing the challenge of
soil metaproteome complexity by improving metaproteome depth of coverage through two-dimensional
liquid chromatography. Soil Biol. Biochem. 125:290–99

56 Naylor et al.
135. Heaven MW, Benheim D. 2016. Soil microbial metabolomics. In Microbial Metabolomics: Applications
in Clinical, Environmental, and Industrial Microbiology, ed. DJ Beale, KA Kouremenos, EA Palombo,
pp. 147–98. Cham: Springer Int. Publ.
136. Hultman J, Waldrop MP, Mackelprang R, David MM, McFarland J, et al. 2015. Multi-omics of per-
mafrost, active layer and thermokarst bog soil microbiomes. Nature 521(7551):208–12
137. Roy Chowdhury T, Lee J-Y, Bottos EM, Brislawn CJ, White RA, et al. 2019. Metaphenomic responses
of a native prairie soil microbiome to moisture perturbations. mSystems 4(4):e00061-19
138. Goldford JE, Lu N, Bajic D, Estrela S, Tikhonov M, et al. 2018. Emergent simplicity in microbial com-
munity assembly. Science 361(6401):469–74
139. Zegeye EK, Brislawn CJ, Farris Y, Fansler SJ, Hofmockel KS, et al. 2019. Selection, succession, and
stabilization of soil microbial consortia. mSystems 4(4):e00055-19
140. Smith BN, Epsten S. 1971. Two categories of 13 C/12 C ratios for higher plants. Plant Physiol. 47:380–
84
141. Schweigert M, Herrmann S, Miltner A, Fester T, Kästner M. 2015. Fate of ectomycorrhizal fungal
biomass in a soil bioreactor system and its contribution to soil organic matter formation. Soil Biol.
Biochem. 88:120–27
142. Throckmorton HM, Bird JA, Dane L, Firestone MK, Horwath WR. 2012. The source of microbial C
has little impact on soil organic matter stabilisation in forest ecosystems. Ecol. Lett. 15(11):1257–65
143. Kindler R, Miltner A, Thullner M, Richnow H-H, Kästner M. 2009. Fate of bacterial biomass derived
fatty acids in soil and their contribution to soil organic matter. Org. Geochem. 40(1):29–37
144. Miltner A, Bombach P, Schmidt-Brücken B, Kästner M. 2012. SOM genesis: microbial biomass as a
significant source. Biogeochemistry 111(1–3):41–55
145. Liao H, Li Y, Yao H. 2019. Biochar amendment stimulates utilization of plant-derived carbon by soil
bacteria in an intercropping system. Front. Microbiol. 10:1361
146. Hungate BA, Mau RL, Schwartz E, Caporaso JG, Dijkstra P, et al. 2015. Quantitative microbial ecology
through stable isotope probing. Appl. Environ. Microbiol. 81(21):7570–81
147. Papp K, Hungate BA, Schwartz E. 2020. Glucose triggers strong taxon-specific responses in microbial
growth and activity: insights from DNA and RNA qSIP. Ecology 101:e02887
148. Papp K, Mau RL, Hayer M, Koch BJ, Hungate BA, Schwartz E. 2018. Quantitative stable isotope
probing with H2 18 O reveals that most bacterial taxa in soil synthesize new ribosomal RNA. ISME J.
12(12):3043–45
149. Starr EP, Shi S, Blazewicz SJ, Probst AJ, Herman DJ, et al. 2018. Stable isotope informed genome-
resolved metagenomics reveals that Saccharibacteria utilize microbially-processed plant-derived carbon.
Microbiome 6(1):122
150. Zhu W, Lu H, Hill J, Guo X, Wang H, Wu W. 2014. 13 C pulse-chase labeling comparative assessment of
the active methanogenic archaeal community composition in the transgenic and nontransgenic parental
rice rhizospheres. FEMS Microbiol. Ecol. 87(3):746–56
151. Stewart CE, Roosendaal D, Denef K, Pruessner E, Comas LH, et al. 2017. Seasonal switchgrass eco-
type contributions to soil organic carbon, deep soil microbial community composition and rhizodeposit
uptake during an extreme drought. Soil Biol. Biochem. 112:191–203
152. Chomel M, Lavallee JM, Alvarez-Segura N, Castro F, Rhymes JM, et al. 2019. Drought decreases incor-
poration of recent plant photosynthate into soil food webs regardless of their trophic complexity. Glob.
Change Biol. 25(10):3549–61
153. Welsh DT, Herbert RA. 1993. Identification of organic solutes accumulated by purple and green sulphur
bacteria during osmotic stress using natural abundance 13 C nuclear magnetic resonance spectroscopy.
FEMS Microbiol. Ecol. 13(2):145–49
154. Dijkstra P, Dalder JJ, Selmants PC, Hart SC, Koch GW, et al. 2011. Modeling soil metabolic pro-
cesses using isotopologue pairs of position-specific 13 C-labeled glucose and pyruvate. Soil Biol. Biochem.
43(9):1848–57
155. Bore EK, Apostel C, Halicki S, Kuzyakov Y, Dippold MA. 2017. Soil microorganisms can overcome
respiration inhibition by coupling intra- and extracellular metabolism: 13 C metabolic tracing reveals the
mechanisms. ISME J. 11(6):1423–33

www.annualreviews.org • Soil Microbiomes and Climate Change 57

 156. Harden JW, Hugelius G, Ahlström A, Blankinship JC, Bond-Lamberty B, et al. 2018. Networking our
science to characterize the state, vulnerabilities, and management opportunities of soil organic matter.
Glob. Change Biol. 24(2):e705–18
157. Ostle NJ, Smith P, Fisher R, Woodward FI, Fisher JB, et al. 2009. Integrating plant-soil interactions
into global carbon cycle models. J. Ecol. 97(5):851–63
158. Wieder WR, Bonan GB, Allison SD. 2013. Global soil carbon projections are improved by modelling
microbial processes. Nat. Clim. Change 3(10):909–12
159. Todd-Brown KEO, Randerson JT, Post WM, Hoffman FM, Tarnocai C, et al. 2012. Causes of variation
in soil carbon predictions from CMIP5 Earth system models and comparison with observations. Biogeosci.
Discuss. 9(10):14437–73
160. Bradford MA, Wieder WR, Bonan GB, Fierer N, Raymond PA, Crowther TW. 2016. Managing uncer-
tainty in soil carbon feedbacks to climate change. Nat. Clim. Change 6(8):751–58
161. Conant RT, Ryan MG, Ågren GI, Birge HE, Davidson EA, et al. 2011. Temperature and soil organic
matter decomposition rates—synthesis of current knowledge and a way forward. Glob. Change Biol.
17(11):3392–404
162. Robertson AD, Paustian K, Ogle S, Wallenstein MD, Lugato E, Cotrufo MF. 2019. Unifying soil organic
matter formation and persistence frameworks: the MEMS model. Biogeosciences 16(6):1225–48
163. Manzoni S, Porporato A. 2009. Soil carbon and nitrogen mineralization: theory and models across scales.
Soil Biol. Biochem. 41(7):1355–79
164. Bond-Lamberty B, Epron D, Harden J, Harmon ME, Hoffman F, et al. 2016. Estimating heterotrophic
respiration at large scales: challenges, approaches, and next steps. Ecosphere 7(6):e01380
165. Cobo-Díaz JF, Baroncelli R, Le Floch G, Picot A. 2019. Combined metabarcoding and co-occurrence
network analysis to profile the bacterial, fungal and Fusarium communities and their interactions in
maize stalks. Front. Microbiol. 10:261
166. Li S, Wu F. 2018. Diversity and co-occurrence patterns of soil bacterial and fungal communities in seven
intercropping systems. Front. Microbiol. 9:1521
167. de Vries FT, Griffiths RI, Bailey M, Craig H, Girlanda M, et al. 2018. Soil bacterial networks are less
stable under drought than fungal networks. Nat. Commun. 9(1):3033
168. McClure RS. 2019. Toward a better understanding of species interactions through network biology.
mSystems 4(3):e00114-19
169. Röttjers L, Faust K. 2018. From hairballs to hypotheses—biological insights from microbial networks.
FEMS Microbiol. Rev. 42(6):761–80
170. Wang H, Wei Z, Mei L, Gu J, Yin S, et al. 2017. Combined use of network inference tools identifies
ecologically meaningful bacterial associations in a paddy soil. Soil Biol. Biochem. 105:227–35
171. McClure RS, Overall CC, Hill EA, Song H-S, Charania M, et al. 2018. Species-specific transcriptomic
network inference of interspecies interactions. ISME J. 12(8):2011–23
172. Cordero OX, Datta MS. 2016. Microbial interactions and community assembly at microscales. Curr.
Opin. Microbiol. 31:227–34
173. Fox A, Ikoyi I, Torres-Sallan G, Lanigan G, Schmalenberger A, et al. 2018. The influence of aggregate
size fraction and horizon position on microbial community composition. Appl. Soil Ecol. 127:19–29
174. Zhao X-F, Hao Y-Q, Zhang D-Y, Zhang Q-G. 2019. Local biotic interactions drive species-specific
divergence in soil bacterial communities. ISME J. 13(11):2846–55
175. Karimi B, Terrat S, Dequiedt S, Saby NPA, Horrigue W, et al. 2018. Biogeography of soil bacteria and
archaea across France. Sci. Adv. 4(7):eaat1808
176. Fillinger L, Hug K, Griebler C. 2019. Selection imposed by local environmental conditions drives differ-
ences in microbial community composition across geographically distinct groundwater aquifers. FEMS
Microbiol. Ecol. 95(11):fiz160
177. Feeser KL, Van Horn DJ, Buelow HN, Colman DR, McHugh TA, et al. 2018. Local and regional scale
heterogeneity drive bacterial community diversity and composition in a polar desert. Front. Microbiol.
9:1928
178. O’Brien SL, Gibbons SM, Owens SM, Hampton-Marcell J, Johnston ER, et al. 2016. Spatial scale drives
patterns in soil bacterial diversity. Environ. Microbiol. 18(6):2039–51

58 Naylor et al.
179. Ramirez KS, Knight CG, de Hollander M, Brearley FQ, Constantinides B, et al. 2018. Detecting macro-
ecological patterns in bacterial communities across independent studies of global soils. Nat. Microbiol.
3(2):189–96
180. Tecon R, Or D. 2017. Biophysical processes supporting the diversity of microbial life in soil. FEMS
Microbiol. Rev. 41(5):599–623
181. Piotrowska-Długosz A, Breza-Boruta B, Długosz J. 2019. Spatial and temporal variability of the soil
microbiological properties in two soils with a different pedogenesis cropped to winter rape (Brassica
napus L.). Geoderma 340:313–24
182. Barber NA, Chantos-Davidson KM, Amel Peralta R, Sherwood JP, Swingley WD. 2017. Soil micro-
bial community composition in tallgrass prairie restorations converge with remnants across a 27-year
chronosequence. Environ. Microbiol. 19(8):3118–31
183. Jansson JK, Fredrickson JK. 2010. Stewards of a changing planet: commentaries from ISME13 Plenary
Lecturers. ISME J. 4(9):1079–80
184. Zengler K. 2009. Central role of the cell in microbial ecology. Microbiol. Mol. Biol. Rev. 73(4):712–29
185. Borer ET, Harpole WS, Adler PB, Lind EM, Orrock JL, et al. 2014. Finding generality in ecology: a
model for globally distributed experiments. Methods Ecol. Evol. 5(1):65–73
186. Guo Y-S, Furrer JM, Kadilak AL, Hinestroza HF, Gage DJ, et al. 2018. Bacterial extracellular polymeric
substances amplify water content variability at the pore scale. Front. Environ. Sci. 6:93
187. Huang X, Li Y, Liu B, Guggenberger G, Shibistova O, et al. 2017. SoilChip-XPS integrated technique
to study formation of soil biogeochemical interfaces. Soil Biol. Biochem. 113:71–79
188. Soufan R, Delaunay Y, Gonod LV, Shor LM, Garnier P, et al. 2018. Pore-scale monitoring of the effect
of microarchitecture on fungal growth in a two-dimensional soil-like micromodel. Front. Environ. Sci.
6:68
189. Tecon R, Ebrahimi A, Kleyer H, Erev Levi S, Or D. 2018. Cell-to-cell bacterial interactions promoted
by drier conditions on soil surfaces. PNAS 115(39):9791–96
190. Zengler K, Hofmockel K, Baliga NS, Behie SW, Bernstein HC, et al. 2019. EcoFABs: advancing micro-
biome science through standardized fabricated ecosystems. Nat. Methods 16(7):567–71
191. Downie H, Holden N, Otten W, Spiers AJ, Valentine TA, Dupuy LX. 2012. Transparent soil for imaging
the rhizosphere. PLOS ONE 7(9):e44276
192. Fiorentino N, Sánchez-Monedero MA, Lehmann J, Enders A, Fagnano M, Cayuela ML. 2019. Inter-
active priming of soil N transformations from combining biochar and urea inputs: a 15 N isotope tracer
study. Soil Biol. Biochem. 131:166–75
193. Hobbie EA, Chen J, Hanson PJ, Iversen CM, McFarlane KJ, et al. 2017. Long-term carbon and nitrogen
dynamics at SPRUCE revealed through stable isotopes in peat profiles. Biogeosciences 14(9):2481–94
194. Ebrahimi A, Or D. 2016. Microbial community dynamics in soil aggregates shape biogeochemical gas
fluxes from soil profiles—upscaling an aggregate biophysical model. Glob. Change Biol. 22(9):3141–56
195. Jansson JK, Taş N. 2014. The microbial ecology of permafrost. Nat. Rev. Microbiol. 12(6):414–25
196. Mackelprang R, Saleska SR, Jacobsen CS, Jansson JK, Taş N. 2016. Permafrost meta-omics and climate
change. Annu. Rev. Earth Planet. Sci. 44:439–62
197. Bonkowski M. 2004. Protozoa and plant growth: the microbial loop in soil revisited. New Phytol.
162(3):617–31
198. Vereecken H, Schnepf A, Hopmans JW, Javaux M, Or D, et al. 2016. Modeling soil processes: review,
key challenges, and new perspectives. Vadose Zone J. 15(5):1–57
199. Patt A, Siebenhüner B. 2005. Agent based modeling and adaptation to climate change. Vierteljahrsh. Zur.
Wirtsch. 74(2):310–20
200. Song H-S, Nelson WC, Lee J-Y, Taylor RC, Henry CS, et al. 2018. Metabolic network modeling for
computer-aided design of microbial interactions. In Emerging Areas in Bioengineering, ed. HN Chang,
pp. 793–801. Weinheim, Ger.: Wiley-VCH

www.annualreviews.org • Soil Microbiomes and Climate Change 59