Cooperative Institute for Research in Environmental Sciences

A winning proposal for the Innovative Research Program, 2006:

Possibility of Abrupt Climate Change in the Next few Decades

Investigator: Prashant D. Sardeshmukh, CIRES, NOAA Earth System Research Laboratory/Physical Sciences Division

Objective. We propose to perform several 100-year integrations of two global atmospheric general circulation models(GCMs) with prescribed ocean sea surface temperatures (SSTs) in the tropics and with coupling to a simple mixed layer ocean model elsewhere. The prescribed tropical SST fields will incorporate a hypothesized continuation of the warming trend observed over the past 50 years, including the continued warming of the tropical Indian, western Pacific, and Atlantic oceans. For reasons explained below, we suspect that the global climate system is close to a "tipping point" with regard to the continued warming of especially the Indian ocean.We believe that our GCM experiments, conducted with two different GCMs to generate confidence in the results, will provide a clear assessment of the likelihood of this tipping point being crossed in the next several decades, with potentially dramatic consequences for the climates of North America, Asia, and the Arctic.

Background. There is growing evidence that much of the global and regional climate changes over the next century associated with anthropogenic forcing will be mediated, or at least be strongly influenced, by tropical SST changes. As Figure 1 shows, the tropical oceans have already warmed substantially over the last 50 years. In a recently completed study (Sardeshmukh and Penland 2006), we demonstrated that this warming is outside the range of natural ENSO SST variability at almost all tropical locations. The warming of the Indian, western Pacific, and south Atlantic oceans is especially significant in this regard. At present it is not clear to what degree this warming is associated with anthropogenic forcing or with natural decadal and longer term variations of the

Figure 1. Observed warming of the Tropical Sea Surface Temperatures 1951-2000

climate system. The answer to this question has a large bearing on how this warming pattern will evolve over the next several decades and what its worldwide impacts will be. Most climate models involved in the IPCC Fourth Assessment of projected climate changes misrepresent important aspects of tropical climate dynamics, including ENSO, the long-term oceanic subsurface controls on ENSO, and the oceanic "dipole" dynamics of the Indian and Atlantic oceans. These deficiencies greatly limit our confidence in their ability to predict the future evolution of the warming pattern shown in Figure 1. This is unfortunate, since there is also growing evidence(Sardeshmukh, Barsugli and Shin 2006) that regional climate changes around the globe will be highly sensitive to the precise pattern of the tropical SST warming. This sensitivity arises mainly from the generally opposite global circulation responses to deep atmospheric convective heat sources over the west Pacific and Indian oceans (Ting and Sardeshmukh, J. Atmos. Sci. 1993; Barsugli and Sardeshmukh J. Climate 2002).

Importance. The opposite sensitivity of many aspects of the global climate to strong tropical Indian and west Pacific SST warming (strong enough that it leads to enhanced deep atmospheric convection in both basins) raises the possibility of a climatic "tipping point". The Indian ocean warming has thus far not been strong enough to enhance the deep convection there, but if it continues, then at some point it will certainly trigger atmospheric deep convection and rapidly force an opposite global response to that already being forced by the west Pacific warming (which, among other effects, has been implicated in the extended four-year 1998-2002 drought over the Southwestern United States). This rapid change will be felt most dramatically in the Pacific-North American (PNA) sector, but also in the high Arctic latitudes. We believe this to be a serious possibility, and would like to assess it quantitatively in a realistic modeling environment.

What is innovative about our approach? Given our current poor understanding of the precise causes of the warming trend shown in Figure 1, and also the inadequacies of current fully coupled atmosphere-ocean climate models in predicting its future evolution, we will take a completely different approach to assess the likelihood and impact of "awakening the sleeping Indian ocean giant": we will force state-of-the-art global atmospheric GCMs with prescribed evolving sea surface temperatures, simply assuming that the pattern of the tropical SST warming observed over the past several decades will also continue for the next several decades. This assumption might seem unrealistic to some, but may in fact not be so, considering that the amplification of the warming in Figure 1 has been quite coherent over the past 50 years (not shown). There is also no better alternative at present.

Interdisciplinary Aspects. The possibility of a "tipping-point" being crossed by the triggering of enhanced atmospheric deep convection over the Indian ocean has enormous implications not only for the climates of the nearby densely populated Asian continent but also, through planetary Rossby wave dispersion, the climates of remote North America, the Arctic and even parts of Europe. Through its associated surface temperature and precipitation changes, it will influence drought and energy and water resources management in these areas, the future evolution of the Asian"brown cloud" through more efficient precipitation scavenging, and perhaps even the melting of the Greenland ice sheet.

Research Plan. We will run two different state-of-the-art atmospheric GCMs, the NCAR/CAM3 and the NCEP/GFS model,with prescribed SSTs in the tropics and with coupling to a simple slab mixed-layer ocean model elsewhere. The prescribed tropical SST fields will be climatology plus the trend pattern of Figure 1, whose amplitude will be increased linearly at its past 50-yr rate over the next 100 years. We will also make control 100-yr runs with the tropical SST fields fixed at climatology. Both of the GCMs to be run are already available to us in-house and we have extensive experience with their use. Computing needs will be met using the ESRL high performance computing and storage at the ESRL/GSD high performance computing facility.

Expected Outcome and Impact. Our focus in the warming trend runs will be on the point in time at which the Indian ocean warming triggers significantly enhanced local deep atmospheric convection and impacts the global climate. We will document the differences between the simulated global climates before and after this triggering, and assess their implications for climate change mitigation and adaptation strategies around the globe.