Data Analytics for Sustainability and Environmental Risk Workshop Report

Published on 01 December 2016

Challenge: There exists a significant information gap between the sources of climate data and the kinds of local, actionable knowledge required for mitigation and adaptation policy and planning decisions. The causes of this gap are discontinuous and heterogeneous data sources, a plethora of loosely connected models of varying complexity, and limitations of expertise transfer across domains of climate science, environmental science, modelling and big data, policy, legal and financial decision making, and planning and civil engineering. The ultimate consequence is that it is difficult to make meaningful decisions and plans on the local and regional level in response to climate change.

Solution: Explore the use of big data and machine learning to harmonise the data sources and connect models and data at different scales. An example of what could be achieved is a system that continuously integrates diverse streams of live climate and environmental data and produces a set of key planetary vital signs. Another example is an interactive knowledge base for decision-makers that takes into account local and regional exposure to various climate-related risks. The systems would ideally have a fully transparent pathway from data to knowledge that can be investigated, challenged and improved upon by domain experts, showcasing the data sources, the methods underlying them and their reliability levels, the global/regional/local climate and environmental models and datasets and associated confidence/uncertainty, and the regional/local risk exposure assessments using decision-relevant metrics.

Opportunity: The challenge outlined cannot be tackled by climate scientists or environmental scientists alone. New capabilities in big data analytics, sensor networks and Earth observation have resulted in an opportunity to develop a much more sophisticated set of indicators of climate change and resulting impact. Cambridge seems well-placed to draw on expertise from these various disciplines.

If successful, having traceable connections from the processes affecting planetary vital signs through to the impacts relevant to society (e.g. food and water security, flooding, species loss), business (e.g. correlated extreme events, threshold exceedance) and decision-makers at all levels (e.g. risk of abrupt or irreversible change) would be valuable evidence-base for developing local, regional and global policy and planning.

Read report

Subscribe to our mailing list to get our latest updates