Prediction puzzles

As you have hopefully gathered by now, predicting how climate change will impact our oceans is a pretty complex job.  Hence, the star paper in this blog is attempting to do exactly that! Griffith et al. 2012 aim to  quantify the impact that fishing, ocean warming and ocean acidification have on biodiversity and are investigating this using the southeastern Australian marine ecosystem up to 2050. 

The model used integrates the biophysical, economic and social parts of marine ecosystems. The impact on biomass was calculated for all possible stressor combinations. The effect size was calculated using Hedge's d, which is similar to a measure of variance. The figure below shows the resulting effect size of all the scenarios.

It was found that ocean acidification on its own decreased biomass. When all three stressors were joined together, it was also found that they had a greater effect together on biomass than would have occurred had they been run separately, also known as a 'synergistic effect'. Fishing and ocean acidification combined together also produced a synergistic effect on biomass change.

For higher trophic organisms such as predators  the response to all three stressors was greater. Whereas primary producers had an additive response to ocean warming and ocean acidification, and the addition of fishing reduced the antagonistic response.

Overall, the model found that reduced fishing in the SE Australian marine ecosystem may mitigate the effects of ocean warming and ocean acidification, which is the most positive outcome because this can be very efficiently managed regionally, unlike ocean warming and acidification. However it was also found that ocean acidification was the main driver when all three stressors interacted together.

This approach may help us to understand complex interactions however, it is scaled up to find the impacts on communities and ecosystems. While this provide good preliminary analysis, in the real world, there are many more stressors such as habitat loss which are not included in this model. Having said this, the model gives a specific regional prediction, which provides a more effective analysis than global prediction which would be very generalised.