Advanced

The response of heterotrophic CO2-flux to soil warming

Eliasson, Peter LU ; McMurtrie, Ross E.; Pepper, David A.; Strömgren, Monika; Linder, Sune and Ågren, Göran I. (2005) In Global Change Biology 11(1). p.167-181
Abstract
In a forest ecosystem at steady state, net carbon (C) assimilation by plants and C loss through soil and litter decomposition by heterotrophic organisms are balanced. However, a perturbation to the system, such as increased mean soil temperature, will lead to faster decay, enhancing CO2 release from decomposers, and thus upsetting the balance. Recent in situ experiments have indicated that the stimulation of soil respiration following a step increase in annual average soil temperature declines over time. One possible explanation for this decline may be changes in substrate availability. This hypothesis is examined by using the ecosystem model G'DAY, which simulates C and nitrogen (N) dynamics in plants and soil. We applied the model to... (More)
In a forest ecosystem at steady state, net carbon (C) assimilation by plants and C loss through soil and litter decomposition by heterotrophic organisms are balanced. However, a perturbation to the system, such as increased mean soil temperature, will lead to faster decay, enhancing CO2 release from decomposers, and thus upsetting the balance. Recent in situ experiments have indicated that the stimulation of soil respiration following a step increase in annual average soil temperature declines over time. One possible explanation for this decline may be changes in substrate availability. This hypothesis is examined by using the ecosystem model G'DAY, which simulates C and nitrogen (N) dynamics in plants and soil. We applied the model to observations from a soil-warming experiment in a Norway spruce (Picea abies (L.) Karst.) stand by simulating a step increase of soil temperature. The model provided a good qualitative reproduction of the observed reduction of heterotrophic respiration (R-h) under sustained warming. The simulations showed how the combined effects of faster turnover and reduced substrate availability lead to a transient increase of R-h. The simulated annual increase in R-h from soil was 60% in the first year after perturbation but decreased to 30% after a decade. One conclusion from the analysis of the simulations is that R-h can decrease even though the temperature response function for decomposition remains unchanged. G'DAY suggests that acclimation of R-h to soil warming is partly an effect of substrate depletion of labile C pools during the first decade of warming as a result of accelerated rates of mineralization. The response is attributed mainly to changing levels of C in pools with short time constants, reflecting the importance of high-quality soil C fractions. Changes of the structure or physiology of the decomposer community were not invoked. Therefore, it becomes a question of definition whether the simulated dynamics of the declining response of CO2 release to the warming should be named acclimation or seen as a natural part of the system dynamics. (Less)
Please use this url to cite or link to this publication:
@article{14102a9e-d760-4811-ac6c-f3d05b9b1d24,
  abstract     = {In a forest ecosystem at steady state, net carbon (C) assimilation by plants and C loss through soil and litter decomposition by heterotrophic organisms are balanced. However, a perturbation to the system, such as increased mean soil temperature, will lead to faster decay, enhancing CO2 release from decomposers, and thus upsetting the balance. Recent in situ experiments have indicated that the stimulation of soil respiration following a step increase in annual average soil temperature declines over time. One possible explanation for this decline may be changes in substrate availability. This hypothesis is examined by using the ecosystem model G'DAY, which simulates C and nitrogen (N) dynamics in plants and soil. We applied the model to observations from a soil-warming experiment in a Norway spruce (Picea abies (L.) Karst.) stand by simulating a step increase of soil temperature. The model provided a good qualitative reproduction of the observed reduction of heterotrophic respiration (R-h) under sustained warming. The simulations showed how the combined effects of faster turnover and reduced substrate availability lead to a transient increase of R-h. The simulated annual increase in R-h from soil was 60% in the first year after perturbation but decreased to 30% after a decade. One conclusion from the analysis of the simulations is that R-h can decrease even though the temperature response function for decomposition remains unchanged. G'DAY suggests that acclimation of R-h to soil warming is partly an effect of substrate depletion of labile C pools during the first decade of warming as a result of accelerated rates of mineralization. The response is attributed mainly to changing levels of C in pools with short time constants, reflecting the importance of high-quality soil C fractions. Changes of the structure or physiology of the decomposer community were not invoked. Therefore, it becomes a question of definition whether the simulated dynamics of the declining response of CO2 release to the warming should be named acclimation or seen as a natural part of the system dynamics.},
  author       = {Eliasson, Peter and McMurtrie, Ross E. and Pepper, David A. and Strömgren, Monika and Linder, Sune and Ågren, Göran I.},
  issn         = {1354-1013},
  keyword      = {acclimation,carbon storage,century,ecosystem model,feedback,g'day,global warming,q(10),soil respiration,soil warming,carbon-cycle feedbacks,long-term response,organic-matter,norway spruce,elevated co2,temperature-dependence,litter decomposition,modeling analysis,root respiration,atmospheric co2},
  language     = {eng},
  number       = {1},
  pages        = {167--181},
  publisher    = {Wiley-Blackwell},
  series       = {Global Change Biology},
  title        = {The response of heterotrophic CO2-flux to soil warming},
  url          = {http://dx.doi.org/10.1111/j.1365-2486.2004.00878.x},
  volume       = {11},
  year         = {2005},
}