Login / Signup

Litter accumulation and fire risks show direct and indirect climate-dependence at continental scale.

Mark A AdamsMathias Neumann
Published in: Nature communications (2023)
Litter decomposition / accumulation are rate limiting steps in soil formation, carbon sequestration, nutrient cycling and fire risk in temperate forests, highlighting the importance of robust predictive models at all geographic scales. Using a data set for the Australian continent, we show that among a range of models, >60% of the variance in litter mass over a 40-year time span can be accounted for by a parsimonious model with elapsed time, and indices of aridity and litter quality, as independent drivers. Aridity is an important driver of variation across large geographic and climatic ranges while litter quality shows emergent properties of climate-dependence. Up to 90% of variance in litter mass for individual forest types can be explained using models of identical structure. Results provide guidance for future decomposition studies. Algorithms reported here can significantly improve accuracy and reliability of predictions of carbon and nutrient dynamics and fire risk.
Keyphrases
  • climate change
  • machine learning
  • quality improvement
  • electronic health record
  • high intensity
  • big data
  • artificial intelligence