A Framework for Combining Seasonal Forecasts and Climate Projections to Aid Risk Management for Fisheries and Aquaculture

Abstract

A changing climate, in particular a warming ocean, is likely to impact marine industries in a variety of ways. For example, aquaculture businesses may not be able to maintain production in their current location into the future, or area-restricted fisheries may need to follow the fish as they change distribution. Preparation for these potential climate impacts can be improved with information about the future. Such information can support a risk-based management strategy for industries exposed to both short-term environmental variability and long-term change. In southern Australia, adverse climate impacts on valuable seafood industries have occurred, and they are now seeking advice about future environmental conditions. We introduce a decision tree to explain the potential use of long-term climate projections and seasonal forecasts by these industries. Climate projections provide insight into the likely time in the future when current locations will no longer be suitable for growing or catching particular species. Until this time, seasonal forecasting is beneficial in helping industries plan ahead to reduce impacts in poor years and maximize opportunities in good years. Use of seasonal forecasting can extend the period of time in which industries can cope in a location as environmental suitability declines due to climate change. While a range of short-term forecasting approaches exist, including persistence and climatological forecasts, only dynamic model forecasts provide a viable option for managing environmental risk for marine industries in regions where climate change is reducing environmental suitability and creating novel conditions.

Climate change and aquaculture: considering adaptation potential

“…Aquaculture sectors can clearly benefit from longterm climate predictions of environmental variables. For example, templates of Tasmanian sea surface temperature projections can be applied by managers to identify onset years when some regions will become unsuitable for the culture of certain species (Hobday et al 2018). However, one of the most pressing predictive needs is arguably predicting disease outbreak.…”

Spatio‐temporal models of intermediate complexity for ecosystem assessments: A new tool for spatial fisheries management

“…This illustration allows comparison of true simulated density (1st and 3rd rows) versus estimated density (2nd and 4th rows) when simulating a new data set conditional on fixed effects estimated from real-world data. As such, future short-term forecasts and long-term projections of many fish stocks will likely require models that include climate-driven changes to spatial distributions and species interactions (Deyle, May, Munch, & Sugihara, 2016;Hobday, Cochrane, et al, 2016b;Hobday et al, 2018;. For visual clarity, we do not show simulated density for the two species that are not then included in the estimation model for each simulation replicate Table 2 for true values).…”

“…The visualization of biological reference points (b) similarly shows a histogram of estimates and the true value, and again lists the mean and standard deviation of estimates drivers of mortality, selectivity and growth (Skern-Mauritzen et al, 2016;Pinsky et al, 2018). As such, future short-term forecasts and long-term projections of many fish stocks will likely require models that include climate-driven changes to spatial distributions and species interactions (Deyle, May, Munch, & Sugihara, 2016;Hobday, Cochrane, et al, 2016b;Hobday et al, 2018;. Spatially explicit MICE models such as the one presented here represent a potential tool for managing fisheries under changing conditions, as they can be used to forecast changes in spatial distribution while accounting for species interactions (Howell & Filin, 2014).…”

“…Forecasting climate impacts is a growing concern for identifying suitable management strategies Miller et al, 2010). Covariates explaining historical variation in density will improve predictive skill for shortterm forecasts in some cases and not others (Hobday et al, 2018;Thorson, 2019Thorson, , 2019a, so we encourage future skill testing to measure the potential benefits of incorporating multispecies interactions, density covariates and species associations for forecasting distribution shifts over short (<3 year) or longer time horizons.…”

Climate change and aquaculture: considering biological response and resources

“…Capture fisheries and aquaculture provide 3 billion people with 20% of their average per capita intake of animal protein and a further 1.5 billion people with about 15% of their animal protein (HLPE 2014). Seafood is the most highly traded food (Smith et al 2010), and demand is only expected to increase (World Bank 2013).…”