Better understanding and quantification of the terrestrial carbon cycle remains a pressing issue for climate science. The ability of the land surface to continue acting as a sink of anthropogenic emissions is unknown and, should it weaken or become a net source, there are significant implications for the rate of increase of atmospheric carbon dioxide and hence the rate of climate change.

There are an increasing number of Earth Observation data streams that provide information beyond traditional spectral vegetation indices. Examples include Solar Induced Fluorescence from chlorophyll (SIF), Vegetation Optical Depth (VOD), LIDAR, as well as surface level flux inversions from observations of atmospheric trace gas concentrations (e.g. carbon dioxide and carbonyl sulphide). At the same time there are increasingly sophisticated methods for constraining models of the land surface with observations, including Machine Learning techniques and novel Data Assimilation algorithms.

This session will take stock of the diverse observations and model-data fusion methods for constraining models of the terrestrial biosphere. We invite abstracts on this topic in the broadest sense, from empirical top-down models through to detailed bottom up models using any type of algorithm to implement observational constraints or provide model validation. We are especially interested in submissions that use multiple data streams to constrain or validate models.