It is accepted that volcanic eruptions can have a major impact on short term climate. A new study in Nature Geoscience uses instrument records, proxy data and climate modeling to show that multidecadal variability is a dominant feature of North Atlantic sea-surface temperature (SST), which, in turn, impacts regional climate. It turns out that the timing of multidecadal SST fluctuations in the North Atlantic over the past 600 years has, to a large degree, been governed by changes in external solar and volcanic forcings. Solar influence is not surprising but the fact that volcanoes cause climate change lasting decades has some significant implications for those trying to model climate over the next century.

When a volcano erupts it spews large quantities of ash, water vapor, sulfur dioxide and even some carbon dioxide into the atmosphere. Sulfur dioxide reacts with water to form sulfuric acid droplets (aerosol particles), which are highly reflective and reduce the amount of incoming sunlight, leading to what some refer to as a “volcanic winter.” An eruption large enough to to depress global temperatures by 1°C (1.8°F) and trigger widespread crop failures for several years afterwards should occur about once every 200-300 years.

Volcanic aerosols are also injected directly into the stratosphere, where they modify both short-wave and long-wave radiation transfer. This can cause strong heating of the lower tropical stratosphere by absorption of terrestrial and solar near-infrared radiation. The strengthened polar vortex that follows traps the wave energy of the tropospheric circulation, and the North Atlantic oscillation (NAO) dominates winter circulation, producing winter warming over large parts of the Northern Hemisphere. Evidently volcanoes first cause cooling and then longer-term warming.

A volcano modifying the climate.

In “External forcing as a metronome for Atlantic multidecadal variability,” Odd Helge Otterå et al. examine the driving forces behind the Atlantic multidecadal oscillation (AMO), a basin-wide variation marked by alternation of warm and cold sea surface temperature (SST) anomalies in the North Atlantic with a period of about 60–80 years. Their analysis of multiple proxies indicates that AMO variability has existed for several centuries.

It has been suggested, on the basis of climate model simulations, that these variations are internally driven and related to multidecadal fluctuations in the Atlantic meridional overturning circulation (AMOC). Otterå et al, find that AMO is not solely driven by the changes in the AMOC. Instead, external forcings such as total solar irradiance (TSI) variations and volcanic eruptions are important drivers. As stated in the article abstract:

We find that volcanoes play a particularly important part in the phasing of the multidecadal variability through their direct influence on tropical sea-surface temperatures, on the leading mode of northern-hemisphere atmosphere circulation and on the Atlantic thermohaline circulation. We suggest that the implications of our findings for decadal climate prediction are twofold: because volcanic eruptions cannot be predicted a decade in advance, longer-term climate predictability may prove challenging, whereas the systematic post-eruption changes in ocean and atmosphere may hold promise for shorter-term climate prediction.

That longer-term climate predictions “may prove challenging” is science speak for “probably never work.” The researchers used a fully coupled climate model, the Bergen Climate Model (BCM), to demonstrate that external forcing has been instrumental in pacing multidecadal variability in the Atlantic region over the past 600 years. A total of seven simulations were carried out, the results of which are shown in the figures below.

a, Simulated standardized indices of AMO (black), AMOC (purple), global SST PC1 (grey) and PC3 (pink) together with reconstructed standardized AMO indices based on multiple proxies (dark green) and tree-ring data (light green). Correlations (α<0.1) and root mean square errors between EXT600 and reconstructions are also shown. b, Regression of global SST in EXT600 on PC1. c, The same as b, but for PC3. d, Cross-correlations of the simulated AMO, PC1, PC3 and AMOC indices with the TSI forcing in EXT600. Positive lags mean that the forcing is leading. e, The same as d, but for correlations with the total (TSI+volcano) forcing. f, Cross-correlations of the simulated AMO, PC1 and PC3 indices with the AMOC index. Positive lags mean that the AMOC is leading. In d–f significance levels (α<0.05) are shown in grey shading.

Several earlier studies have suggested lagged relationships between low-frequency TSI variations and the NAO. The proposed mechanisms include atmospheric teleconnections from the Pacific Ocean as well as stratosphere–troposphere coupling. Otterå et al. reinforce the findings previously reported on this blog (see “Pacific Warming, Atlantic Hurricanes & Global Climate Non-Disruption”). “In EXT600, we find no significant correlation between the simulated NAO and the applied TSI forcing,” they state. “However, there is a significant negative correlation between the NAO and the total external forcing, suggesting a potential role for volcanoes.”

Positive or increasing NAO is typically associated with large tropical volcanic eruptions. It is known from both observations and other modeling studies that large tropical eruptions have a tendency to induce a positive NAO response, causing the well-known posteruption winter warming phenomenon over Northern Hemisphere land masses. However, climate models have only shown limited ability in simulating this robust, observation-based feature, possibly linked to inadequate treatment of stratosphere–troposphere dynamical interactions.

Simulated responses to volcanic forcing.

The authors present a number of possible caveats to the findings. “For example, it could be argued that the BCM underestimates the internal variability of the AMOC on multidecadal timescales,” they state. “This question is, however, difficult to adequately address in the absence of instrumental observations of the AMOC.” As usual, the models are untrustworthy and there is a lack of good hard empirical data. Nonetheless, Otterå et al. conclude:

Although the external forcing is clearly important for the AMO characteristics in the BCM, it cannot explain all of the simulated variability. In the model, and also probably in nature, there is an interplay between the intrinsic climate variability and the external forcing. Rather, we conclude that the external forcing acts as a metronome for the Atlantic multidecadal variability. In view of this, the frequency and intensity of external forcing need to be better understood and quantified to produce reliable near-term climate forecasts.

Volcanoes are among Earth's most destructive natural phenomena. Is it any wonder that, to be able to predict climate variation over multidecadal timescales, you need to be able to predict volcanic eruptions. The impact of a volcanic eruption is not a smooth, continuously varying function over time, like waxing and waning of solar intensity or the slow buildup of gases in the atmosphere. Volcanic eruption sends a sudden shock throughout the global environment, a pulse of change, a perturbation of the system.

Volcanoes cause decadal scale climate change.

Climate models have no way to model such phenomena since future eruptions cannot be predicted. Most models use a constant value to represent aerosol inputs averaged over time, which means they are always wrong: they overestimate levels when there have been no recent eruptions and underestimate levels when an eruption occurs. Since volcanoes are unpredictable in their timing, location and intensity, this is pretty much a show stopper for climate prediction models.

Albert Einstein once said that God does not play dice with the Universe. Here is proof that Einstein was wrong, at least about our little corner of the Universe, because it looks like nature does play dice with climate change. This, of course, has not stopped climate scientists from trying to play god.

Be safe, enjoy the interglacial and stay skeptical.