The availability of data through in-situ and remote sensing measurements is ever increasing, providing us with more opportunities than ever before to understand systems in both space and time. However, the potential to maximise the extraction and use of information held within these datasets is not fully realized and much information can remain hidden within existing datasets requiring combined analysis with auxiliary data and simultaneous consideration of (bio-/geo-) physical processes. Coupled with the sheer size and amount of data, there is also a need to handle any processing in a manageable and deployable way whilst not limiting the quality of outputs. The techniques and outcomes of such analyses have the potential to benefit researchers from multi-disciplinary backgrounds which this session intends to foster.

We welcome submissions which consider various data (continuous and/or categorical such as from remote sensing, in situ data, phenocams, etc.) and apply new or existing methods (e.g. moving window approaches, automated recognition, Gaussian Process models, Machine Learning, AI) with applications which could cover a range of topics from land use/land cover dynamics, phenology, trait evolution, biogeochemical cycles, geological processes and landscape evolution, through to natural hazard susceptibility and prediction.