Ecosystem modelling for the ocean decade - facing the challenges

The worlds’marine ecosystems are degrading under wide ranges of ever intensifying, diversifying and co-occurring human pressures. Ecosystem-based management (EBM) approaches have emerged as an alternative to ineffective single species and single sector management, veering away from siloed top-down approaches towards science-based, participatory processes that recognise connections across the system and seek to balance economic benefits with sustainably harvested and healthy ecosystems. To galvanize a global push towards EBM, the United Nations declaration of the Decade of Ocean Science for Sustainable Development (Ocean Decade) has given the oceanographic community a unique imperative to transform marine sciences into holistic, participatory, transparent and inclusive forms that involve and serve society. 
Transforming actual ocean sciences is easier said than done. Marine Ecosystem Models or MEMs are powerful mathematical tools for understanding past marine ecosystem and their dynamics under cumulative pressures, and have utility for predicting how ecosystems may continue to develop under scenarios of change. MEMs are widely used in science, and have significant utility to advice decision making and policy. However, despite decades of scientific progress, and despite an abundance of scientific recipes in the literature that can be deployed towards the aims of the Ocean Decade, the actual uptake of MEMs in management remains low. 
 
This dissertation explores why this is, and argues that the actual uptake of MEMs in policy and society is in part hampered by a factor largely ignored by the marine sciences: technical issues, institutionalized by the current competitive and 
achievement-driven academic funding model. The dissertation is based on four manuscripts, which explore the specific challenges raised by the Ocean Decade, and define and implement working prototypes to demonstrate that the gap between theory and practice can be bridged. 
 
The first challenge, enabling decision processes to use MEMs, is addressed in manuscript 1 where a MEM is integrated into a decision support tool for marine spatial planning, beyond the operational control of marine scientists. The second challenge, related to meaningfully communicating MEM output to outside audiences, is addressed in manuscript 2 where a MEM is interconnected with a 3D gaming engine to empathically visualize environmental change. The last challenge, making sure that MEM output is robust, is discussed in manuscripts 3 and 4. Of these, the first manuscript explores the reasons behind lack of systematic MEM assessments and puts forth a potential framework to overcome this 30-year old limitation. Manuscript 4 introduces a working and open-source prototype of that framework. 
 
Overall, these studies show that relatively simple software engineering can empower the use of MEMs towards the aims of the Ocean Decade, EBM, and beyond. This dissertation underscores that scientific and technical developments must go hand in hand, but also suggests that the status quo may not change unless long-term tool development and support become academic funding priorities. Last, although the prototypes developed in this dissertation should be taken as ideas that need further maturing in future research, the ideas throughout irrevocably demonstrate that the field of marine ecosystem modelling with relatively simple means can be made operational for the Ocean Decade. If anything, this dissertation is a rallying cry to the global marine ecosystem modelling community to rethink and reshape how we build, validate, calibrate and deploy our tools, with the aim to reach and involve the audiences that need marine science advice but do not have the means to generate it.