Is deoxygenation detectable before warming in the thermocline?
Multiple lines of evidence from observation- and model-based studies show that anthropogenic greenhouse gas emissions cause ocean warming and oxygen depletion, with adverse impacts on marine organisms and ecosystems. Temperature is considered as one of the main indicators of climate change, but, in...
| Main Authors: | , , , |
|---|---|
| Format: | Text |
| Language: | English |
| Published: |
2019
|
| Subjects: | |
| Online Access: | https://doi.org/10.5194/bg-2019-339 https://www.biogeosciences-discuss.net/bg-2019-339/ |
| Summary: | Multiple lines of evidence from observation- and model-based studies show that anthropogenic greenhouse gas emissions cause ocean warming and oxygen depletion, with adverse impacts on marine organisms and ecosystems. Temperature is considered as one of the main indicators of climate change, but, in the thermocline, anthropogenic changes in biogeochemical tracers such as oxygen may emerge from the bounds of natural variability before changes in temperature. Here, we compare the local time of emergence (ToE) of anthropogenic temperature and oxygen changes in the thermocline within an ensemble of Earth system model simulations from the fifth phase of the Coupled Model Intercomparison Project (CMIP5). Anthropogenic deoxygenation emerges from natural internal variability before warming in 35 ± 11 % of the global thermocline. Earlier emergence of oxygen than temperature change is simulated by all models in parts of the subtropical gyres of the Pacific and the Southern Ocean. Earlier detectable changes in oxygen than temperature are typically related to decreasing trends in ventilation. The supply of oxygen-rich surface waters to the thermocline is reduced as evidenced by an increase in apparent oxygen utilisation over the simulations. Concomitantly, the propagation of the warming signal is hindered by slowing ventilation, which delays the warming in the thermocline. As the magnitudes of internal variability and simulated temperature and oxygen changes, which determine ToE, vary considerably among models, we compute the local ToE relative to the global mean ToE within each model. This reduces the inter-model spread in the relative ToE compared to the traditionally evaluated absolute ToE. Our results underline the importance of an ocean biogeochemical observing system and that the detection of anthropogenic impacts becomes more likely when using multi-tracer observations. |
|---|