GC44A-01 INVITED
Tipping Points
A climate tipping point, at least as I have used the phrase, refers to a situation in which a changing climate forcing has reached a point such that little additional forcing (or global temperature change) is needed to cause large, relatively rapid, climate change. Present examples include potential loss of all Arctic sea ice and instability of the West Antarctic and Greenland ice sheets. Tipping points are characterized by ready feedbacks that amplify the effect of forcings. The notion that these may be runaway feedbacks is a misconception. However, present "unrealized" global warming, due to the climate system's thermal inertia, exacerbates the difficulty of avoiding global warming tipping points. I argue that prompt efforts to slow CO2 emissions and absolutely reduce non-CO2 forcings are both essential if we are to avoid tipping points that would be disastrous for humanity and creation, the planet as civilization knows it.
GC44A-02 INVITED
Ice Sheets and Abrupt Change
The sea-level rise projections from the IPCC (AR4 SPM 2007) appeared as "Model-based range excluding future rapid dynamical changes in ice flow", and the text noted "…understanding…is too limited to provide a best estimate or an upper bound for sea level rise." Of the possible mechanisms of instability of ice sheets, attention especially focuses on loss of buttressing through ice-shelf removal, and increased lubrication through drainage of surface meltwater to the bed. Quantification of both is improving, but confident predictions are not possible. However, it does appear possible that warming over decades could reach levels that would produce large ice- sheet changes over centuries.
GC44A-03
Has the Arctic Perennial Ice Cover Reached the Tipping Point?
Analysis of satellite data from 1979 to the present has indicated that the Arctic perennial ice cover has been declining at a rapid rate of about 10 percent per decade. The yearly fluctuation was relatively large during the first 20 years but in the last decade, the extent and area of the perennial ice cover have been persistently low. The coverage was a record low in 2002, followed by a slight recovery in 2003 and 2004, another record low in 2005 and a mild recovery in 2006. During the summer of 2007, however, the rate of decline was phenomenal. The extent and area of the ice cover as of 4 September 2007 were 4.5 and 3.9 million square km, respectively, which are considerably less (by 18 to 22 percent) than those of 2005, the corresponding values of which are 5.5 and 5.0 million square km. The large decline suggests that the tipping point for the perennial ice has been reached and a recovery is no longer possible in the foreseeable future. Such hypothesis is supported by studies of the impact of ice-albedo feedback using satellite observed (and in situ) changes in SST (sea surface temperature) in conjunction with a thermodynamic model.
GC44A-04
Response of Thermohaline Circulation to Freshwater Forcing under Present Day and LGM Conditions
Responses of the thermohaline circulation (THC) to freshwater forcing (hosing) in the subpolar North Atlantic Ocean under present day and the last glacial maximum (LGM) conditions are investigated using the National Center for Atmospheric Research Community Climate System Model versions 2 and 3. Three sets of simulations are analyzed, with each set including a control run and a freshwater hosing run. The first two sets are under present day conditions with an open and closed Bering Strait. The third one is under LGM conditions, which has a closed Bering Strait. Results show that the THC nearly collapses in all three hosing runs when the freshwater forcing is turned on. The full recovery of the THC, however, is at least a century earlier in the open Bering Strait run than the closed Bering Strait and LGM runs. This is because the excessive freshwater is diverged almost equally towards north and south from the subpolar North Atlantic when the Bering Strait is open. A significant portion of the freshwater flowing northward into the Arctic exits into the North Pacific via a reversed Bering Strait throughflow, which accelerates the THC recovery. When the Bering Strait is closed, this Arctic to Pacific transport is absent and freshwater can only be removed through the southern end of the North Atlantic. Together with the surface freshwater excess due to precipitation, evaporation, river runoff, and melting ice in the closed Bering Strait experiments after the hosing, the removal of the excessive freshwater takes longer, and this slows the recovery of the THC. Although the background conditions are quite different between the present day closed Bering Strait run and the LGM run, the THC responds to the freshwater forcing added in the North Atlantic in a very similar manner.
GC44A-05
Possibility of Abrupt Climate Change in the Next Several Decades
The tropical oceans have warmed significantly over the last half century, especially in the already warm Indian and western Pacific ocean basins. We have performed extended integrations with the NCAR model to assess the global implications of this Indo-Pacific warming trend. Our integrations suggest that if this warming trend continues, it will gradually modify the atmospheric jet streams to the point where their ability to channel atmospheric disturbances will undergo a relatively sudden shift in the next century. This will cause relatively rapid shifts in the climates of North America and Europe.
GC44A-06
Variability and Expansion of the Tropical Ocean Warm Pool
The tropical warm pool plays a determining role in the global climate since it acts as a sorce of thermodynamic forcing for the atmospheric general circulation. The warm pools (SST>28°C) extend from the Indian Ocean, across the Indonesian Archipelago into the western Pacific with a secondary area crossing Central America into the Caribbean and the central Atlantic ocean. The heating in the atmosphere above the warm pool influences climate over wide ranges of the planet. As there are zonal asymmetries in the extent of the warm pool, and hence variations in the locations of total heating of the atmospheric column, the warm pools also create centers of diabatic heating along the equator which set up the position and strength of the east-west Circulations which play integral roles in the coupled ocean-atmosphere tropical climate. In fact, almost all of the global vertically integrated heating resides over waters >27°C. The tropical warm pool is characterized by large-scale variations of SST on time scales that range from intraseasonal to interdecadal, considerably altering the forcing to the atmosphere. In addition to the existence of the large variability of the tropical warm pool SST, there is an upward trend in the tropical warm pool area, which is evident in the Atlantic, Indian and Pacific oceans with the area encompassed by the 28C isotherm groewing by 67% since 1920. Changes in the zonal and meridional circulation associated with the variability and expansion of the warm pool are studied using NCEP-NCAR and ERA40 reanalsysis. It is found that the impacts extend around the tropics and are associated with a slowing down of the Asian monsoon circulation and modulation of the of the equatorial Walker cells. Analysis of the IPCC-CMIP3 models for the 20th century show similar changes in the warm pool extent suggesting that changes that occur under different future emission scenarios may poossess credence. With greenhouse warming it is found that the warm pool doubles in size from the observed values in the 21st century. We explore the changes in the column integrated heating in the different CMIP3 integration scenarios to see if the threshold between columnar heating and cooling changes to higher temperatures and whether there will be linear or nonlinear transitions. Results suggest the existence of important changes in the zonal and meridional circulation in the Atlantic, Pacific and Indian Ocean in association with the expansion of the tropical warm pool. We also show an example of how the expansion of the tropical warm pool not only has effects on the atmospheric general circulations but also in the dynamics and sustainability of the marine biodiversity. Corals for example are highly sensitive to temperature increase and could face extensive bleaching under the expansion of the tropical warm pool.
GC44A-07 INVITED
Changes in the Discharge of Major Rivers in the Ganges, the Brahmaputra and the Yangtze Rivers During the Next 100 years
Our goal is to estimate the behavior of future river discharge in the Brahmaputra, Ganges, and Yangtze Rivers, forced by increasing greenhouse gas concentrations. Such determinations are critical in order to determine risk from flooding or droughts in the context of rapidly increasing populations. We concentrate on the hydrological discharge in three major river basins: the Yangtze, the Ganges, and the Brahmaputra as these rivers are critical for survival of a large proportion of humankind, where there are extensive historical records of discharge and proven relationships with sea surface temperature variations in the present era. We aim to project the river discharge during the next 100 years under the full range of IPCC greenhouse gas scenarios. We use the suite of IPCC-AR4 CMIP3 climate models to determine future discharge. However, the projected precipitation from models have very large regional model-to-model variability, which limits the ability to quantify the regional precipitation trends. To overcome this problem, we use a basic statistical quantile-to-quantile technique to build a mapping index linking each modeled 20th century river basin precipitation and associated observed 20th century discharge. Model adequacy is stratified by their ability to simulate the observed annual cycle and a set of models chosen for each river basin. Using the "good" models so chosen, the future discharge of the Yangtze, the Ganges, and the Brahmaputra over the next 100 years are for different emission scenarios. The Yangtze, Ganges, Brahmaputra mean wet season river discharges are projected to increase up to 15-25% at the end of 21st century under high greenhouse gases concentration scenarios (SRESA1B and SRESA2). Besides greater river discharge, flooding risk determined by year-to-year variability also increases substantially. Scenarios with fixed greenhouse gas concentration suggest little discharge change on mean value, year-to-year variation, and model-to-model deviation. It is estimated that the Yangtze, Brahmaputra and Ganges river basins might have population doubling times of between 45 and 90 years. Based on these numbers, the amount of fresh water available in these river basins for agriculture, human consumption and industrial use in the present era are compared with that will be available in the future. Despite the anticipated increase in river discharge, the amount of fresh water available per capita is expected to decrease by 50-75%, suggesting the approach of a societal-climate tipping point.
GC44A-08
Changes in Tropical Cyclone Maximum Potential Intensity
We examine the relationship between changes in sea surface temperature (SST) and in tropical cyclone maximum potential intensity (MPI), using both observations and global climate model (GCM) projections. We show that local SST changes alone are inadequate for characterizing changes in tropical cyclone MPI, with the sign of MPI change not constrained by that of the SST change (warming can be associated with MPI decrease). However, long-term changes in MPI can be largely recovered by accounting for both local and remote SST changes. The remote effect of SST can be approximated by the tropical-mean SST change, such that regions that warm more (less) than the tropical mean are characterized by increased (reduced) MPI. We exploit this simple relationship to develop a simplified index for long-term changes in MPI, which we use to assess the implications of historical SST changes for long-term changes in MPI.
GC44A-09
Causes of Recent Changes of Rainfall Variabilities and Implications to the Future Climate in the Amazon Region
Observations show that the Standardized Precipitation Index (SPI) over the Amazon region has decreased by about 0.32 /decade since 1970, implying an increase in dry events. Simulations of current climate models indicate that such decrease of the SPI cannot be explained by natural variability alone. Using the pooled results from 23 different climate models that participated in the IPCC Fourth Assessment Report (AR4), the simulated pattern of anthropogenically-caused changes in the SPI is identified with statistical confidence in the observed data. Based on projections of the climate models, more dry events, as indicated by stronger negative SPI, likely will occur over the Amazon region in the future when carbon dioxide continues to increase
GC44A-10
Amazon collapse in the next century: exploring the sensitivity to climate and model formulation uncertainties
A number of recent studies have highlighted the risk of abrupt dieback of the Amazon Rain Forest as the result of climate changes over the next century. The recent 2005 Amazon drought brought wider acceptance of the idea that that climate drivers will play a significant role in future rain forest stability, yet that stability is still subject to considerable degree of uncertainty. We present a study which seeks to explore some of the underlying uncertainties both in the climate drivers of dieback and in the terrestrial land surface formulation used in GCMs. We adopt a perturbed physics approach which forms part of a wider project which is covered in an accompanying abstract submitted to the multi-model ensembles session. We first couple the same interactive land surface model to a number of different versions of the Hadley Centre atmosphere-ocean model that exhibit a wide range of different physical climate responses in the future. The rainforest extent is shown to collapse in all model cases but the timing of the collapse is dependent on the magnitude of the climate drivers. In the second part, we explore uncertainties in the terrestrial land surface model using the perturbed physics ensemble approach, perturbing uncertain parameters which have an important role in the vegetation and soil response. Contrasting the two approaches enables a greater understanding of the relative importance of climatic and land surface model uncertainties in Amazon dieback.