Compounded Disturbance Chronology Modulates the Resilience of Soil Microbial Communities and N-Cycle Related Functions

Abstract

There is a growing interest of overcoming the uncertainty related to the cumulative impacts of multiple disturbances of different nature in all ecosystems. With global change leading to acute environmental disturbances, recent studies demonstrated a significant increase in the possible number of interactions between disturbances that can generate complex, non-additive effects on ecosystems functioning. However, how the chronology of disturbances can affect ecosystems functioning is unknown even though there is increasing evidence that community assembly history dictates ecosystems functioning. Here, we experimentally examined the importance of the disturbances chronology in modulating the resilience of soil microbial communities and N-cycle related functions. We studied the impact of 3-way combinations of global change related disturbances on total bacterial diversity and composition, on the abundance of N-cycle related guilds and on N-cycle related activities in soil microcosms. The model pulse disturbances, i.e., short-term ceasing disturbances studied were heat, freeze-thaw and anaerobic cycles. We determined that repeated disturbances of the same nature can either lead to the resilience or to shifts in N-cycle related functions concomitant with diversity loss. When considering disturbances of different nature, we demonstrated that the chronology of compounded disturbances impacting an ecosystem determines the aggregated impact on ecosystem properties and functions. Thus, after 3 weeks the impact of the 'anoxia/heat/freeze-thaw' sequence was almost two times stronger than that of the 'heat/anoxia/freeze-thaw' sequence. Finally, we showed that about 29% of the observed variance in ecosystem aggregated impact caused by series of disturbances could be attributed to changes in the microbial community composition measured by weighted UniFrac distances. This indicates that surveying changes in bacterial community composition can help predict the strength of the impact of compounded disturbances on N-related functions and properties.

Keywords: community composition; compounded disturbances; diversity; nitrogen cycling; resilience.

FIGURE 1

Aggregated impact of repeated disturbances on ecosystem properties and functions. The “Ecosystem Aggregated Impact” was calculated as the sum of the absolute value of Hedges’ g for the 26 studied variables. The corresponding variance is the sum of the variance of each variable Hedges’ g. 95% confidence intervals are represented for each treatment (A) corresponds to the Freeze-Thaw disturbances: F; Panel B to the Heat disturbance: H; and panel C to the Anoxia disturbance: A. In each panel, the corresponding treatment effects on ammonium and nitrate pools as well as on chosen ecosystem properties and functions is given. Note that the control treatment cannot be plotted on this figure because each ecosystem functions and properties (EAI) is calculated as an effect size of a given treatment relative to the control. Different letters above the bars indicate significant differences.

FIGURE 2

Aggregated impact of compounded disturbances with alternative chronologies on ecosystem properties and functions. The “Ecosystem Aggregated Impact” was calculated as the sum of the absolute value of Hedges’ g for all studied variables. The corresponding variance is the sum of the variance of each variable Hedges’ g. 95% confidence intervals are represented for each treatment. Panel A corresponds to T₃ and panel B to T₄, Disturbance sequences are encoded with F: Freeze-thaw, H: Heat and A: Anoxia. Note that the control treatment cannot be plotted on this figure because each EAI is calculated as an effect size of a given treatment relative to the control. Different letters above the bars indicate significant differences.

FIGURE 3

Ecosystem properties significantly impacted by the chronology of compounded disturbance at T₃. Different letters above the bars indicate significant differences according to Tukey’s test (p < 0.05). (A) qAOB corresponds to the abundance of ammonia-oxidizing bacteria and is expressed in gene copies x g^-1 DNA. (B) qnirK corresponds to the abundance of bacteria harboring the nirK nitrite reductase gene and is expressed in gene copies x g^-1 DNA. (C) qnosZII corresponds to the abundance of bacteria harboring the nosZ clade II N₂O reductase gene and is expressed in gene copies x g^-1 DNA. (D) Phylogenetic Diversity corresponds to estimates of the Faith’s phylogenetic diversity index.

FIGURE 4

Ecosystem properties significantly impacted by the chronology of compounded disturbance at T₄. Different letters above the bars indicate significant differences according to Tukey’s test (p < 0.05). (A) qnosZII corresponds to the abundance of bacteria harboring the nosZ clade II N₂O reductase gene and is expressed in gene copies x g^-1 DNA. (B) NH₄ ⁺ corresponds to soil NH₄ ⁺ pool sizes and is expressed in mg N x kg^-1 DNA. (C) Observed Species corresponds to the species richness observed in treated and control microcosms and is expressed in number of species.

FIGURE 5

Principal Coordinates Analyses of the weighted UniFrac distance matrix representing differences in community structure between control and treated microcosms at T₃ (A) and T₄ (B). Different colors correspond to the different treatments and to control microcosms. Closed symbols are used to show treatments that are significantly different from the control microcosms (pairwise-PERMANOVAs).