Sources of uncertainties in 21st century projections of potential ocean ecosystem stressors

Abstract

AbstractFuture projections of potential ocean ecosystem stressors, such as acidification, warming, deoxygenation, and changes in ocean productivity, are uncertain due to incomplete understanding of fundamental processes, internal climate variability, and divergent carbon emission scenarios. This complicates climate change impact assessments. We evaluate the relative importance of these uncertainty sources in projections of potential stressors as a function of projection lead time and spatial scale. Internally generated climate variability is the dominant source of uncertainty in middle‐to‐low latitudes and in most coastal large marine ecosystems over the next few decades, suggesting irreducible uncertainty inherent in these short projections. Uncertainty in projections of century‐scale global sea surface temperature (SST), global thermocline oxygen, and regional surface pH is dominated by scenario uncertainty, highlighting the critical importance of policy decisions on carbon emissions. In contrast, uncertainty in century‐scale projections of net primary productivity, low‐oxygen waters, and Southern Ocean SST is dominated by model uncertainty, underscoring that the importance of overcoming deficiencies in scientific understanding and improved process representation in Earth system models are critical for making more robust projections these potential stressors. We also show that changes in the combined potential stressors emerge from the noise in 39% (34–44%) of the ocean by 2016–2035 relative to the 1986–2005 reference period and in 54% (50–60%) of the ocean by 2076–2095 following a high‐carbon emission scenario. Projected large changes in surface pH and SST can be reduced substantially and rapidly with aggressive carbon emission mitigation but only marginally for oxygen. The regional importance of model uncertainty and internal variability underscores the need for expanded and improved multimodel and large initial condition ensemble projections with Earth system models for evaluating regional marine resource impacts.