Chair: James Orr
Rodgers, Keith B. (1), Coronado, Maricela (2), Schlunegger, Sarah (1), Frölicher, Thomas L. (3), Frenger, Ivy (1), Ishii, Masao (4), Sasano, Daisuke (4)
1 Princeton University, AOS Program Princeton, NJ, 08544, USA
2 Princeton University, Princeton, NJ, 08540, USA
3 Environmental Physics, Institute of Biogeochemistry and Poluttant Dynamics, ETH, Zürich, Switzerland
4 Oceanography and Geochemistry Research Department, Meteorological Research Institute, Tsukuba, Japan
Potential marine ecosystem stressors, such as acidification and de-oxygenation, are expected to impact ocean biology over the course of the 21 st century. Detection of acidification and de-oxygenation is complicated by elevated natural background variability of the climate system. This presents a challenge for inferring secular trends from repeat hydrographic measurements.
We consider a large initial-condition ensemble suite of simulations with GFDL’s Earth system model ESM2M over 1950-2100. The ensemble approach provides a means to deconvolve natural variability and the forced secular trend, and this is approached in a water mass framework (Frölicher et al., 2009). The initial analysis is applied to interpret the repeat hydrographic section along 165°E reported by Sasano et al. (2015) for the emergence of oxygen and acidification trends. Emergence in this framework is defined as when the anthropogenic signal exceeds the noise level of natural variability.
The emergence of anthropogenic trends in acidification ( arag) along 165°E between 0°N and 50°N emerge sooner and with greater confidence than do trends in ocean interior O2 over a 30-year time frame across the major thermocline water masses of the North Pacific. The prior emergence of arag relative to O2 within the ocean interior over the historical period largely reflects the difference in the atmospheric boundary conditions for these two fields.
Detection of trends in ocean interior acidification and de-oxygenation is considered in a water mass framework for the North Pacific. The results indicate that acidification trends emerge prior to de-oxygenation, while simultaneously supporting the arguments of Sasano et al. (2015) that natural decadal variability is substantial and complicates identification of trends on decadal timescales. Trend detection is particularly complicated near gyre boundaries. The results for 165°E are shown to be more generally representative of the North Pacific for the ESM output.