Alistair Hobday and Jason Hartog, CSIRO Australia; Claire Spillman, Bureau of Meteorology Australia; Mark Payne and Brian MacKenzie, DTU-Aqua, Denmark; and Desiree Tommasi, Atmospheric and Ocean Sciences Program at Princeton University.
Marine scientists have always known the ocean environment is different to the terrestrial domain. A combination of planetary motion, gravitational pull from celestial bodies, heat from the sun, and atmospheric winds drive constant motions in the fluid we call the ocean. Continents and the shape of the sea floor constrain and direct ocean currents and the vertical movement of water masses. This constant motion – evident in beautiful flows, eddies, and filaments – means ocean habitats and the species that occur in these habitats are constantly changing position.
Marine resource management, however, has historically paralleled management in terrestrial systems, with fixed spatial management approaches such as protected areas, special closed areas, shipping channels, and fishery zones. Unfortunately, oceanographic features and habitats that both animals and resource users follow, like fronts and eddies, move without respecting these arbitrary lines on a map.
Marine resource management is a complex business, as the world's oceans are also under an unprecedented level of pressure from resource use and commercial activities — for example, fisheries, shipping, aquaculture, and mineral, natural gas, and oil extraction often seek to use the same regions. Balancing sustainable ecological and economic objectives is a continuing challenge for resource managers, and continuing to think of the ocean as a set of static areas is not fruitful.
This has led to calls for wider use of Dynamic Ocean Management (DOM), which uses near real-time data to guide the spatial management of a wide range of activities. Among other advantages, DOM can provide a balance between ocean resource use and conservation and meet multiple objectives—for example, managing target quotas, bycatch reduction, and reducing interactions with species of conservation concern. The technical challenges of delivering DOM have largely been addressed, although some legal issues need additional attention.
Dynamic ocean management has been advanced by the development of habitat nowcasts – the predicted distribution of a species of interest in real-time. Early applications of nowcasts were designed to help fishers avoid unwanted tuna bycatch in Australia or sea turtles in the Pacific Ocean north of the Hawaiian Islands.
Once marine managers, policy-makers, and end users acknowledge that management will need to be dynamic – for example, with the location of a particular fishing zone – information about the possible future location of such a zone becomes valuable to all parties. Armed with information about future ocean conditions, proactive decision-making becomes possible. Depending on the context, and the decisions that are impacted by the changing location of a particular ocean "habitat" or management zone, information on a range of time scales becomes useful to many different users.
For example, fishers planning to catch tuna may need information about probable location of the species several weeks ahead of time in order for vessels to reach fishing grounds. Aquaculture operators may choose to plan harvest dates before critical warm temperatures are experienced and require several months to prepare processing and delivery contracts. Scientists planning monitoring surveys need to allocate their resources to ensure full coverage of the stock in the most cost-effective manner. Policy-makers planning catch allocation between adjacent jurisdictions may be interested in the time of the season that migration may be triggered and need several weeks to implement sharing arrangements.
Information about future ocean conditions is even more important in the context of climate change – past experience and static management are less useful as the future will be different. Changes due to non-stationary climate only reinforce the need for dynamic approaches and information about the future.
There has been an increase in the amount of oceanographic observations available, and in particular, the parallel development of ocean models that produce environmental forecasts that have the ability to predict conditions on seasonal time scales. Marine resource systems are benefiting from these advances, with marine scientists now building forecast systems for a range of focal species.
The first wave of marine ecological forecast products to be routinely delivered are strongly biased toward seasonal prediction of spatial distributions, with fewer examples of abundance, productivity, or phenological forecasts. Seasonal forecasts of the dynamic spatial distribution of southern bluefin tuna in Australia with the aim of avoiding bycatch was one of the first such operational examples. Recently developed forecast systems project the spatial distribution of tuna in southern Australia, and sardines and blue whales in the California Current ecosystem. A second wave of forecasts is now building, with the goal of developing multi-year, or decadal forecasts, although no operational examples for marine resource management are known as yet.
To advance development and management relevance of these dynamic forecasting approaches, the ICES Annual Science Conference (ASC) in 2016 featured a session titled "Seasonal-to-decadal prediction of marine ecosystems: opportunities, approaches, and applications". More than twenty presentations explored marine ecosystem prediction at seasonal to decadal scales. Such forecasts are potentially of great value to the ICES community, as they mirror the timescales on which ICES generates scientific advice, and plans monitoring programmes. Manuscripts presented in this theme session are also being submitted to a "research topic" (i.e. a special issue) of the journal Frontiers in Marine Science.
Last year's ASC theme session highlighted examples and potential in forecasting biological systems and to the management of marine ecosystems: the challenge now is to accelerate the process of developing forecasting products and to incorporate them into policy, management and end-user strategies. While North America and Australia are clearly leading the world in terms of applying this type of knowledge, it is still only used in a few stock assessments (North America) or fisheries (Australia). In these examples, it is clear that engaging end users in forecast development has been critical to successful uptake.
Progress can be accelerated by connecting researchers around the world and developing a community-of-practice. Three initiatives by fisheries and oceans organizations, the ICES Working Group on Seasonal-to-Decadal Prediction of Marine Ecosystems (WGS2D), the PICES/CLIVAR Study Group on Climate and Ecosystem Predictability (SG-CEP), and the IMBER/CLIOTOP Task Team all focus on the common task of predicting ecosystems and ecological responses. These groups have begun to form a wider global network where ideas and experiences can be exchanged and developed - and a joint session at the Fourth International Symposium on the Effects of Climate Change on the World Oceans in 2018 is in the planning stages. These and other initiatives indicate an exciting time in the development of improved marine management via development of seasonal and decadal views of the future ocean and the organisms that live there.
Seasonal forecast models have been developed for species such as southern bluefin tuna; photo: Alistair Hobday, CSIRO
Read the research
Hazen, E. L., Palacios, D. M., Forney, K. A., Howell, E.A., Becker, E., Hoover, A.L., Irvine, L., et al. 2016. WhaleWatch: a dynamic management tool for predicting blue whale density in the California. Journal of Applied Ecology: doi: 10.1111/1365-2664.12820.
Hobday, A. J. and Hartog, J. R. 2014. Dynamic Ocean Features for use in Ocean Management. Oceanography 27(4): 134–145.
Hobday, A. J. and Hartmann, K. 2006. Near real-time spatial management based on habitat predictions for a longline bycatch species. Fisheries Management & Ecology 13(6): 365-380.
Hobday, A. J., Spillman, C. M., Eveson, J. P., and Hartog, J. R. 2016. Seasonal forecasting for decision support in marine fisheries and aquaculture. Fisheries Oceanography 25(S1): 45-56.
Hobday, A. J., Maxwell, S. M., Forgie, J., McDonald, J., Darby, M., Seto, K., Bailey, H. et al. 2014. Dynamic ocean management: Integrating scientific and technological capacity with law, policy and management. Stanford Environmental Law Journal 33(2): 125-165.
Kaplan, I. C., Williams, G. D., Bond, N. D., Hermann, A. J., and Siedlecki, S. 2016. Cloudy with a chance of sardines: forecasting sardine distributions using regional climate models. Fisheries Oceanography 25(1): 15-27.
Maxwell, S. M., Hazen, E. L., Lewison, R., Dunn, D., Bailey, H., Bograd, S. J., Briscoe, D. K., et al. 2015. Dynamic ocean management: Defining and conceptualizing real-time management of the ocean. Marine Policy 58: 42–50.
Salinger, J., Hobday, A. J., Matear, R. J., O'Kane, T. J., Risbey, J. S., Dunstan, P. K., Eveson, J. P., et al. 2016. Decadal-scale forecasting of climate drivers for marine applications. Advances in Marine Biology 74: 1-68. doi:10.1016/bs.amb.2016.04.002.