A multidimensional framework enhances interpretation of carbon (C) cycling stability following disturbance

Abstract:

Understanding how forests will respond to globally rising disturbances is critical for forecasting the stability of the carbon (C) cycle. Growing evidence suggests that C cycling stability following moderate disturbance can be highly variable among C fluxes, ecosystems and disturbance sources. While community ecologists have led the advancement of disturbance stability theory, biogeochemists have mostly neglected these frameworks, limiting our ability to quantify and directly compare C cycling stability across fluxes and ecosystems. Our objective is to demonstrate the utility of a multidimensional stability framework for interpreting and comparing forest C flux responses to disturbance. We have adapted a framework that quantifies and standardizes four interrelated but mathematically distinct dimensions of stability—resistance, resilience, temporal stability and recovery. Using long-term soil respiration (Rs) and net ecosystem production (NEP) data following an experimental disturbance, we illustrate how this framework produces a quantitative inter-comparison of disturbance recovery cycles between fluxes that cannot be fully captured with conventional time-series analyses. We found that while both Rs and NEP lacked complete resistance to the disturbance, Rs showed high resilience and temporal stability, resulting in a full recovery, whereas NEP showed lower resilience with very low temporal stability, resulting in an unclear recovery pattern. Using this framework, we were able to compare and contrast the stability of these two fluxes following disturbance despite differences in measurement methodology and units. Additionally, using a published time-series synthesis of NEP responses to a breath of North American disturbances (Amiro et al. 2010), we show how this framework reveals large differences in resistance across disturbance types, showing quantitative support for variability in C cycling response to different disturbances. We conclude that the application of a standardized stability framework will not only allow for more direct comparisons of C flux stability following disturbance, but through a more developed understanding of relationships between stability dimensions, may aid in forecasting long-term C cycling disturbance recovery cycles.