Future Sustainable Ecosystems: Complexity, Risk, Uncertainty

Presenters: Dr. Nathaniel Newlands and Prof. Alexander Brenning

Overview

Sustainability and climate change present grand, urgent challenges for science and society for the 21st Century. Sustaining our marine and terrestrial ecosystems to deliver what people need and value, under future multi-scale and multi-type risks/threats/hazards will require a complex set of mitigation and adaptation strategies and actions. Such complex decision-making, to be reliable and effective, will require a broad array of different science-based indicators, metrics, frameworks and modeling systems that will enhance our capability to make more informed decisions.

Knowing how we make decisions is as important as knowing which decisions to make - and we must confront better the complex interplay involved in translating scientific evidence into real-world operational or actionable solutions. Real-world uncertainties may be exceedingly difficult to quantify, because it is often highly fragmented in space and highly dynamic in time. Often, their dynamic behaviour precludes any simple (e.g. deterministic) identification of a fixed set of driving factors or causal linkages. Sustainability assessment pathways will, increasingly, be facilitated by predictive systems (e.g. those that integrate artificial intelligence (AI) and machine-learning algorithms) capable of rapidly integrating and synthesizing mixed-type data obtained across a broad range of observational and monitoring platforms (e.g. fixed/mobile sensor arrays and ground stations, unmanned aerial vehicles (UAVs)/drones, Earth Observational Satellites).

The course will focus on sustainability, the behaviour of complex dynamic systems and forecasting, statistics of predictive modeling using geospatial data, geospatial and machine-learning. Several lectures (40 min.) will be delivered followed by supplementary practical exercises with case-study examples and data (i.e. lab classes) using the R software.

A wireless internet connection will be available for participants. Lecture slides and code samples will be shared, and participants will need to bring their own laptop with R installed.

Presenter profiles

Nathaniel Newlands conducts interdisciplinary R&D aimed at developing effective broad, integrative solutions to real-world sustainability challenges and contentious societal issues within the food-water-energy nexus. He has collaborated strongly with both natural and social scientists who work at the interface of science and policy development (e.g., climate-change, land, water, energy resources). He is an Adjunct Professor in Statistics, University of British Columbia (UBC-Vancouver), Adjunct Associate Professor in Geography, University of Victoria (UVic), Canada, an Editor for Frontiers Environmental Science/Interdisciplinary Climate Change Journal/Nature Publishing Group, and a member of the Statistical Society of Canada (SSC). He has conducted innovative research in modeling farm greenhouse gas emissions and regional to national-scale daily-scale climate interpolation/validation and crop yield forecasting for Canada, bioenergy and climate-crop adaptation (e.g., alternative oilseeds), soil moisture wireless sensor networks and remote-sensing monitoring.

He is currently pursuing several lines of inquiry: 1) developing a geospatial model to predict crop disease/epidemics with machine learning, 2) devising a novel index-based integrated assessment model (environmental and socio-economic indices) to better assess the potential risks and future impacts of extreme weather on Canadian agriculture, and, 3) exploring the use of machine-learning/intelligence for improved decision support and the rapid appraisal of disasters and associated environmental and human health risks.

Prof. Alexander Brenning is a quantitative geographer and geographic information scientist whose research focuses on spatial statistical and computational tools and their application in a variety of contexts, in particular mountain geomorphology and environmental remote sensing.

Alexander currently holds the Chair of Geographic Information Science at the University of Jena, Germany, where he is a member of the Michael Stifel Center Jena for Data-Driven and Simulation Science. He is also an Adjunct Associate Professor of Geography at the University of Waterloo, Canada, where he was a faculty member from 2007 to 2015.

Alexander has had visiting appointments at the University of Heidelberg as a Humboldt research fellow, at the University of Vienna and at Catholic University of Chile. Alexander's spatial approach to the assessment of classification and regression methods has been instrumental in a variety of applications and international collaborations mainly related to landslide susceptibility, mountain permafrost distribution and remotely-sensed land cover classification.

Recent and ongoing research has been funded through NSERC Discovery Grants, the Carl Zeiss Foundation and the European Union's LIFE program, among others.