by Judith Curry
If Booth and colleagues’ results can be corroborated, then they suggest that multidecadal temperature fluctuations of the North Atlantic are dominated by human activity, with natural variability taking a secondary role. This has many implications. Foremost among them is that the AMO does not exist, in the sense that the temperature variations concerned are neither intrinsically oscillatory nor purely multidecadal.
Aerosols implicated as a prime driver of twentieth-century North Atlantic climate variability
Ben Booth, Nick Dunstone, Paul Halloran, Timothy Andrews Nicolas Bellouin
Systematic climate shifts have been linked to multidecadal variability in observed sea surface temperatures in the North Atlantic Ocean1. These links are extensive, influencing a range of climate processes such as hurricane activity and African Sahel and Amazonian droughts. The variability is distinct from historical global-mean temperature changes and is commonly attributed to natural ocean oscillations. A number of studies have provided evidence that aerosols can influence long-term changes in sea surface temperatures, but climate models have so far failed to reproduce these interactions and the role of aerosols in decadal variability remains unclear. Here we use a state-of-the-art Earth system climate model to show that aerosol emissions and periods of volcanic activity explain 76 per cent of the simulated multidecadal variance in detrended 1860–2005 North Atlantic sea surface temperatures. After 1950, simulated variability is within observational estimates; our estimates for 1910–1940 capture twice the warming of previous generation models but do not explain the entire observed trend. Other processes, such as ocean circulation, may also have contributed to variability in the early twentieth century. Mechanistically, we find that inclusion of aerosol–cloud microphysical effects, which were included in few previous multimodel ensembles, dominates the magnitude (80 per cent) and the spatial pattern of the total surface aerosol forcing in the North Atlantic. Our findings suggest that anthropogenic aerosol emissions influenced a range of societally important historical climate events such as peaks in hurricane activity and Sahel drought. Decadal-scale model predictions of regional Atlantic climate will probably be improved by incorporating aerosol–cloud microphysical interactions and estimates of future concentrations of aerosols, emissions of which are directly addressable by policy actions.
Nature (2012) doi:10.1038/nature10946 [link to abstract]
Over the past century, the surface of the North Atlantic Ocean has gone through warm and cool periods that are not observed in other ocean basins. This Atlantic multidecadal oscillation (AMO) is thought to affect climate processes ranging from the current high levels of Atlantic hurricane activity to the devastating sub-Saharan droughts of the early 1980s. Although the influence of the AMO on extreme weather events has long been recognized, the physical processes underlying these temperature changes are not understood. In a paper published on Nature’s website today, Booth et al. report their use of a state-of-the-art model of Earth’s climate to demonstrate that, at least over the past century, the AMO is largely the response of the upper ocean to changes in the concentration of pollution aerosols in the atmosphere. If correct, their results imply that the influence of human activity on the Atlantic regional climate is more pervasive than previously thought.
The AMO is best depicted as the difference between average ocean surface temperatures over the North Atlantic and those over the global oceans. It therefore reflects the deviation of the North Atlantic Ocean from global mean temperatures, which are dominated by the long-term warming that is forced by greenhouse gases. Conventional wisdom has held that the AMO is the natural result of internal processes in the Atlantic Ocean — most notably, fluctuations in deep-ocean circulation, as supported by multi-century climate-model studies.
Booth and colleagues’ evidence that the AMO is caused by changes in the regional abundance of aerosols is compelling, but their results are sensitive to model parameterizations of microphysical processes, particularly the interaction between cloud water droplets and aerosols, that are not well constrained by observations. In addition, their model was unable to reproduce observed multidecadal variability in outbreaks of African dust storms, which alter the tem- perature of the tropical Atlantic; this may explain why the model does a poorer job of simulating temperatures in the tropical North Atlantic Ocean than it does in the extratropical regions. Furthermore, the authors’ conclusion that internal variability of the Atlantic Ocean had a negligible role in shaping the AMO during the twentieth century is at odds with the findings of several previous studies. The reason for this discrepancy is not clear.
If Booth and colleagues’ results can be corroborated, then they suggest that multidecadal temperature fluctuations of the North Atlantic are dominated by human activity, with natural variability taking a secondary role. This has many implications. Foremost among them is that the AMO does not exist, in the sense that the temperature variations concerned are neither intrinsically oscillatory nor purely multidecadal.
Nature also has an editorial on this [here]. I’ll be nice and just ask who writes this stuff? It can’t be a scientist.
JC evaluation
The Jones et al. paper describing the simulations for HadGEM2-ES can be found [here]. The model has a fairly sophisticated atmospheric chemistry module. Of particular relevance to the aerosol forcing:
In HadGEM2-ES sea-salt and mineral dust aerosol emissions are computed interactively, whereas emission datasets drive schemes for sulphate, fossil-fuel black and organic carbon, and biomass aerosols. Unless otherwise stated, datasets are derived from the historical and RCP time series prepared for CMIP5.
Referring to Ch 9 of the IPCC AR4, particularly Fig 9.7, the CMIP3 version of the HADGEM1 has the spectrum of natural internal variability too low by about 40% for timescales beyond about 35 years (when compared with the spectra of uncertain observations).
The result of this paper is driven by the so-called aerosol indirect effect, whereby the aerosols change the physical and optical properties of clouds. The uncertainty in the aerosol indirect effect is estimated in the AR4 to be by far the most uncertain element of radiative forcing, and the estimates in AR4 neglect many of the modes of the aerosol indirect effect, notably those associated with ice clouds.
Without having time to dig up references, there is another important issue. Specifying aerosol characteristics (which is mostly done here, esp sulfate) and then allowing interactive cloud microphysics and optics results in an overestimate of the aerosol indirect effect, since compensating dynamics and precip don’t influence the aerosols. van den Heever and Stephens have a recent paper on this, but even more than 10 years ago Rostayn(sp?) was writing about this.
I’m trying to find what the CMIP5 is using for historical aerosol forcing. A quick google search doesn’t turn up much in the way of documentation, but I did spot this. Not an error bar in sight. Realistic error bars on current aerosol optical depth measurements are quite large; historical error bars must be huge. The fortuitous agreement of the aerosol optical depth with temperature variability is serendipitous climate magic, almost certainly with circular reasoning buried deeply or not so deeply in the aerosol estimates.
And finally, if this paper is correct and there is no AMO other than aerosol forcing, this is going to overthrow a very substantial body of work by oceanographers on the Atlantic Meridional Overturning Circulation . At best, the period in this paper covers 2 oscillations.
Color me unconvinced by this paper. I suspect that if this paper had been submitted to J. Geophysical Research or J. Climate, it would have been rejected. In any event, a much more lengthy manuscript would have been submitted with more details, allowing people to more critically assess this. By publishing this, Nature seems to be looking for headlines, rather than promoting good science.
A final reflection: recall my previous post on Trends, change points & hypotheses. This paper is squarely in the camp of hypothesis #1, where all climate variability (other than ENSO) is externally forced. I think this view is incorrect, but it seems to be ruling the IPCC’s mode of thinking.