Assessing the climate impacts of land cover change under present and future environmental conditions

Changes in land cover deeply affect the surface properties and therefore the direction and magnitude of the energy, water and carbon fluxes between the atmosphere and the land, ultimately impacting the local and global climate. The processes underlying biophysical and biogeochemical vegetation properties are themselves strongly influenced by the background climate and therefore affected by climate change in a complex and circular manner. For these reasons, deepening our understanding and prediction capacity of the ongoing changes in the land-climate nexus is of paramount importance to developing land-based climate mitigation strategies and policies that are robust and achievable.

The high complexity of the land-based climate interaction is driven by its bi-directional nature that involves multiple positive and negative feedbacks, and by the co-occurrence of processes with opposite climate impacts (i.e. climate cooling and warming) both for biophysical and biogeochemical processes (e.g. radiative vs. non-radiative effects, respiration versus photosynthesis). These specific features of the land system lead to an extremely high sensitivity of the net climate impact of land cover change to management practices and environmental drivers.

Because of the inherent complexities of the land-climate systems, simulations performed with land-atmosphere coupled models proved to be rather uncertain and strongly affected by knowledge gaps, weak assumptions and oversimplistic parameterization. On the other hand, disentangling the signals of the different processes from Earth observations is particularly complex and leads to uncertain attribution of causality. Even more challenging is using experimental signals derived under present climate conditions to project the future direction and magnitude of biophysical and biogeochemical impacts of land cover change on climate. To this scope, among the key limitations of the widely used “space for time” substitution we can list the role played by unaccounted factors (e.g CO₂ fertilization), the speed and span of ecosystem adaptation, the assumption of steady state and the temporal and spatial dependence of the processes. Given the importance of this research topic in the current fight to mitigate climate warming, new approaches and methodological advances are required to benefit from the increasing computational capacity and by the expanding observation of the Earth system. To address the issue, in this presentation I will review the most recent progress of data-driven and hybrid analyses, and report on a recent attempt to investigate the impact of land transformation on the climate trajectory under future climatic conditions. Additional discussion points will deal with the emerging research needs related to non-linearity in the system and tipping points (e.g. related to plant mortality rates), and the possible way forward on the ingestion of knowledge derived from Earth observation in process-oriented modelling frameworks.