2010 Literature Review Archives - Climate Change Projections

Davis, S.J., K. Caldeira, and H.D. Matthews. 2010. Future CO2 emissions and climate change from existing energy infrastructure. Science 10 September, 2010 Vol 328 pp 1330-1333.

The long lifetime of existing transportation and energy infrastructure means that continued emissions of CO2 from these sources are likely for a number of decades. This ‘infrastructural inertia’ alone is projected to produce a warming commitment of 1.3°C above the pre-industrial era. This result emphasizes that extraordinary measures will be required to limit emissions from new energy and transportation sources if global temperature is to be stabilized below 2°C.

Climate modeling has demonstrated that even if atmospheric composition was fixed at current levels, continued warming of the climate would occur due to inertia in the climate system. This form of climate change commitment has become widely recognized. Davis et al. focus attention on inertia in human systems, by asking ‘what CO2 levels and global mean temperature would be attained if no additional CO2-emitting devices (e.g., power plants, motor vehicles) were built but all the existing CO2-emitting devices were allowed to live out their normal lifetimes?”. Barring widespread retrofitting or early decommissioning of existing infrastructure, these committed emissions represent ‘infrastructural inertia’. The authors developed scenarios of global CO2 emissions from existing infrastructure directly emitting CO2 to the atmosphere for the period 2010 to 2060 (with emissions approaching zero at the end of this time period) and used the University of Victoria Earth System Climate Model to project the resulting changes in atmospheric CO2 and global mean temperature. Projections with low, mid and high emissions scenarios led to projected global average warming of 1.3°C (1.1° to 1.4°C) above the pre-industrial era.  Since new sources of CO2 are bound to be built in the future in order to satisfy growing demands for energy and transportation, the committed warming from existing infrastructure makes clear that satisfying these demands and achieving the 2°C target of the Copenhagen Accord will be an enormous challenge.


Diffenbaugh, N.S. and M. Ashfaq. 2010. Intensification of hot extremes in the United States. Geophysical Research Letters, Vol 37, L15701, doi:10.1029/2010GL043888.

A recent modeling study suggests that increasing greenhouse gas concentrations may lead to a significant intensification of hot extremes in the United States within the next thirty years.

Diffenbaugh and Ashfaq use a suite of climate models that incorporate both large-scale atmospheric circulation and fine-scale surface-atmosphere interactions to evaluate potential changes in warm-season hot extremes across the continental United States over the coming decades.  Projections for the A1B scenario  - a mid-level emission scenario - are evaluated from ensembles derived from a high-resolution regional climate model (RegCM3) nested within a coupled General Circulation Model (GCM) (NCAR CCSM3) and, for comparison, from an ensemble of GCM output from 22 models (from the CMIP3 climate model archive).  Projected changes in heat extremes (hottest season, longest heat wave and maximum temperature extreme) are presented as maps displaying the number of exceedences from baseline conditions (1951-1999) for three time periods: 2010-2019, 2020-2029 and 2030-2039.  The simulations from the regional model suggest that the historical hottest season may be exceeded 3-4 times over much of the U.S. during the current decade (2010-2019). The GCM simulations produced similar results although the changes were slower to emerge than in the regional model.  Intensification of heat extremes increases each decade and is most pronounced in the west.  For example, projections for 2030-2039 suggest that much of the American west can expect at least 7 exceedences of the historical hottest season, 46 days above the baseline maximum temperature extreme and more than 5 heatwaves longer than the longest historical heatwave.  Further analyses associate the changes in hot extremes with warm-season drying over much of the U.S. which may amplify the effects of rising greenhouse gases in this region.  Projected global warming above pre-industrial conditions for the ensembles used here ranges from 1.8 to 2.5oC for the CMIP3 ensemble and from 1.9 to 2.1oC for the CCSM3 ensemble by 2040 (A1B scenario).  The authors therefore conclude that constraining global warming to 2oC above pre-industrial conditions may not be sufficient to prevent dangerous climate change (in this case an intensification of potentially dangerous hot extremes) in the continental United States. 


Fischer, E.M. and C. Schär, 2010. Consistent geographical patterns of changes in high-impact European heatwaves. Nature Geoscience, Vol 3, pp 398-403, doi: 10.1038/NGEO866.

A new study presents high resolution projections of future high-impact heatwaves across
Europe.  Health impacts are expected to be greatest in low-altitude river valleys and the Mediterranean coast including densely populated urban areas such as Athens, Bucharest, Marseille, Milan, Rome and Naples.

It is estimated that the summer heatwave of 2003 resulted in 40,000 heat-related deaths across Europe.  Fischer and Schär (2010) evaluate the future likelihood of high-impact European heatwaves using simulations from high-resolution Regional Climate Models (RCMs) from the ENSEMBLES multi-model scenario experiment.  The six RCMs employed in the study are driven by three different GCM runs forced with the emissions scenario IPCC SRES A1B.  Daily summer temperature statistics (mean and variance) as well as heat-wave characteristics (length and amplitude) and health indicators (based on humidity, diurnal temperature range and heatwave duration) are evaluated for three thirty-year time slices (1961-1990; 2021-2050 and 2071-2100).  The authors emphasize that there is a strong overall similarity in geographical patterns of change projected using the different RCMs.  Ensemble mean results suggest that the greatest increases in summer heatwave frequency and duration will occur in southernmost Europe.  For example, the number of heatwave days per summer in the Iberian peninsula and Mediterranean region is projected to increase from an average of two in 1961-1990 to 13 for the period 2021-2050 and as high as 40 for 2071-2100.  The most severe health impacts are projected for the densely populated low-altitude river basins in southern Europe and for the Mediterranean coasts.  The authors conclude that health risks in these regions may actually be greater than their projections because the models used do not include urban heat island effects.


Jevrejeva, S., J.C. Moore and A. Grinsted. 2010. How will sea level respond to changes in natural and anthropogenic forcings by 2100? GRL Vol 37, L07703, doi:10.1029/2010GL042947.

Projections of sea level rise by 2100 using a semi-empirical model are about twice as large as the estimates in the IPCC Fourth Assessment.


The estimates of sea level rise (SLR) for the 21st century in the IPCC Fourth Assessment were based primarily on climate model projections of thermal expansion (along with an off-line estimate of the additional contribution from land ice based on climate model results). An alternative approach is to use a statistical model based on semi-empirical relationships between past changes in a variety of parameters and tide-gauge observations of sea level. Jevrejeva et al. use one such model, in this case based on changes in radiative forcing in the past, to investigate how global sea level will respond to future changes in forcing and the relative influence of natural versus anthropogenic forcings. They find that 21st century sea level rise is clearly dominated by anthropogenic forcings even under scenarios of rather extreme changes in natural forcing (e.g. frequent large volcanic eruptions). Projected SLR ranged from 0.6 – 1.6m, similar to results from other recent studies using observationally constrained statistical models of SLR. This is roughly a factor of 2 higher than the estimates provided in the IPCC Fourth Assessment.


Pechony, O. and D.T. Shindell. 2010. Driving forces of global wildfires over the past millennium and the forthcoming century. PNAS, Vol 107, pp 19167-19170, DOI: 10.1073/pnas.1003669107.

A new study suggests that the main driver of global wildfire activity in the future will be increasing temperature, replacing direct human influence, which has been the dominant control for the last two centuries.


Despite advances in fire suppression and fire-fighting capacities, the incidence of large wildfires has increased worldwide.  This recent increase coincides with changes in the global climate and  human activities that affect fire ignition and suppression, raising concerns about the future impacts of these changes on global fire activity.  However, the relative importance of the main drivers of global fire trends (climate and direct anthropogenic influences) remain poorly documented through time.  A recent study by Pechony and Shindell (2010) explores the main drivers of global wildfires over the past millennium and provides projections of future global wildfire activity out to 2100 AD.  The authors developed a method for modeling past fire activity based on estimates of vegetation and population density, climate conditions, availability of ignition sources and fire suppression rates.  The estimates of past fire activity correspond well with independently reconstructed fire histories over the past millennium and therefore, the model is used to project future fire activity.  This is achieved using climate simulations from a general circulation model (AR4 version of GISS) and information on population and vegetation density from three of the IPCC SRES scenarios (A2, A1B and B1).  The results indicate that prior to the industrial revolution, the main driver of wildfire activity on a global scale, was the amount of precipitation whereas direct anthropogenic influences dominated through the 19th and 20th century.  The projections suggest the main driver in the future for global wildfires will be rising temperatures with a sharp increase in fire activity after ~2050 AD for all three scenarios.  Interestingly, the scenario with the highest temperature projections (A2) does not yield the highest estimates of future fire activity because of counteracting changes in population and vegetation density.  At the smaller regional scale, changes in the hydrological cycle are projected to be more important determinants of fire activity than temperature changes (which are projected to change much more uniformly).  The authors conclude that future fire management policies will likely have to adapt to a world in which climate plays a stronger role in driving fire trends than it has in the past two centuries, outweighing the direct human influence on fire.


Solomon, S., J.S. Daniel, T.J. Sanford, D.M. Murphy, G-K. Plattner, R. Knutti and P. Friedlingstein. 2010. Persistence of climate changes due to a range of greenhouse gases. Proceedings of the National Academy of Sciences 107 (43) 18354-18359; published ahead of print October 11, 2010, doi: 10.1073/pnas.1006282107.

Study shows that the global warming associated with anthropogenic emissions of non-CO2 GHGs, such as methane and nitrous oxide, persists far longer than would be expected based on their atmospheric lifetimes.  Therefore, while mitigation of shorter-lived agents will help reduce peak warming, there will still be a prolonged warming perturbation.

Reducing emissions of shorter-lived non-CO2 greenhouse gases can quickly reduce atmospheric concentrations of these substances. This makes these substances fundamentally different than CO2 of which atmospheric levels will continue to build even as emissions decline due to its very long atmospheric lifetime.  An outstanding question concerns how quickly declines in atmospheric levels of shorter-lived GHGs will bring about changes in climate. This paper by Solomon and colleagues investigates the processes that contribute to the prolonged global warming perturbation from shorter-lived greenhouse gases, in particular methane and nitrous oxide (their atmospheric lifetimes are ~10 and 100 yrs respectively). They used an Earth System Model of Intermediate Complexity (Bern 2.5CC EMIC) to model the surface warming response to increasing emissions of CO2, CH4 and N2O following the mid-range IPCC SRES A1B scenario followed by sudden cessation of emissions in 2050. Computed surface warming for the individual and combined forcings are presented. Similar to other studies, sustained elevated global warming is shown to persist for at least 1,000 years, driven largely by the elevated atmospheric CO2 levels. Of particular interest is that about 2/3 of the calculated future warming due to N20 is still present 114 years (one atmospheric lifetime) after emissions have ceased even though atmospheric concentration and associated radiative forcing has dropped to about 1/3 of peak levels. Similarly, about 3/4 of the future warming due to CH4 is still present a decade (one atmospheric lifetime) after emissions cease, with about 20% of the signal remaining even after 50 years. In exploring the contributions from the two main factors – optical depth effects and climate system inertia (due to deep ocean heat uptake) – they conclude that while both factors are at play in the prolonged warming perturbation from CO2 (and to some extent, CH4), the slow release of heat stored in the deep ocean is the main cause of the persistent warming perturbation to all three gases.  Since maintaining a forcing for a longer period of time allows more heat to be transferred from the atmosphere to the ocean, the sooner emissions of shorter-lived GHGs (as well as CO2) are reduced, the more effective they will be in avoiding transfer of heat to the deep ocean.


Zahn, M. and H. von Storch. 2010. Decreased frequency of North Atlantic polar lows associated with future climate warming. Nature, Vol 467, pp. 309-311, doi:10.1038/nature09388.

A recent downscaling study projects that a decrease in the occurrence of North Atlantic polar lows - severe mid-sized storms accompanied by strong winds and heavy precipitation - and a northward shift in their mean location will accompany future anthropogenic warming. 

North Atlantic polar lows are not well captured by coarse-resolution global simulations of future climates.  Zahn and von Storch use a regional downscaling model, capable of simulating polar lows over the North Atlantic, to project the frequency and spatial characteristics of their occurrence for three future warming scenarios (IPCC SRES B1, A1B and A2) simulated using the global climate model ECHAM5/MPI-OM.  The downscaled projections reveal a decrease in the occurrence of North Atlantic polar lows in the three scenarios when compared to a present-day simulation from the same climate model.  The ECHAM5/MPI-OM simulations showed that the average tropospheric temperature (500 hPa, Oct-Mar) increased more than surface temperatures (ice-free SSTs, Oct-Mar) over the study region.  A comparison with temperature simulations for the same three scenarios (B1, A1B and A2) from a suite of climate models (CMIP3) showed a similar vertical profile of warming.  The authors therefore attribute the decrease in polar lows to increased atmospheric stability over the region.  Projected spatial changes in polar low genesis include shifts in the main centres of action and an overall northward shift of roughly 2o latitude which is consistent with the projected poleward shift of the Arctic ice edge.  This study provides a rare example of an extreme weather event that is projected to decrease rather than increase in frequency in simulations of future anthropogenic climate change.


Zhang, J., M. Steele, and A. Schweiger. 2010. Arctic sea ice response to atmospheric forcing with varying levels of anthropogenic warming and climate variability. Geophysical Research Letters, Vol 37, L20505, doi:10.1029/2010GL044988.

A recent modeling study suggests that it is unlikely that the Arctic Ocean will be free of sea ice during summer much earlier than 2050 if arctic warming is <4OC.


The majority of global coupled climate models (GCMs) used for the Intergovernmental Panel on Climate Change Fourth Assessment Report (IPCC AR4) underestimate the decline in Arctic Ocean summer sea ice extent observed in recent years.  This has led to speculation that future declines may be more rapid than many of the GCM projections indicate.  Zhang and colleagues attempt to determine if the Arctic Ocean will become ice free in summer and, if so, when.  To address this question, the authors use an arctic ice-ocean model (which has replicated observed spatial and temporal patterns of ice concentration, extent and thickness) in a series of experiments over the period 2010-2050.  The experiments are conducted with two levels of possible future anthropogenic warming for the Arctic (2 or 4oC) and with two datasets representing historical climate variability (Arctic surface air temperature from 1948-2009 and 1989-2009; the former has a greater overall range) for a total of four future projections. The results indicate that even with the highest warming (4OC) and climate variability similar to that of the past two relatively warm decades, the Arctic Ocean would not be permanently ice free in summer until near 2050.  The experiments also indicated that less warming or an increase in the range of past climate variability decreased the likelihood that ice free summer conditions will occur in the Arctic Ocean by 2050.  Summer ice volume appears more sensitive to anthropogenic warming than summer ice extent with projected losses by 2050 ranging from 71-92% of the 1978-2009 September mean volume.  Annual trends in ice volume are projected to decrease strongly until ~2025 but this rate flattens out from 2030-2050 due to increased ice growth during winter (thinner ice has a higher growth rate) and reduced ice export from the Arctic.


Zhang, X. 2010. Sensitivity of summer sea ice coverage to global warming forcing: towards reducing uncertainty in arctic climate projections. Tellus 62A:220-227.

Climate change simulations with models that best replicate current sea ice conditions suggest summer Arctic sea ice cover could pass the 80% loss threshold as early as 2030.


In general, simulations with different climate models produce significantly different projections of how rapidly Arctic sea ice cover will respond to future warming.  This large disagreement is primarily due to differences in how sea ice mass balance and related feedbacks are computed, and in initial sea ice conditions used at the start of the simulations.  Furthermore, models have also substantially underestimated the recent rapid decline in sea ice cover.  In a new study, reported on in the journal Tellus, Alaskan researcher Xiangdong Zhang assessed future sea ice projections using only that subset of climate models that were best able to replicate past observations of the sensitivity of Arctic sea ice area to changes in surface air temperatures.  The range of uncertainties in projections of ice loss over the next century using this subset of models is much lower than that for the larger unconstrained ensemble of model results.  The subset of constrained scenarios also indicates that summer ice area in the Arctic Ocean could decrease by more than 80% as early as 2030, and that regional mean winter and summer surface air temperatures are likely to increase by 8.5°C and 3.7°C, respectively, by 2100.


Zickfeld, K., M.G. Morgan, D.J. Frame and D.W. Keith. Expert judgments about transient climate response to alternative future trajectories of radiative forcing. PNAS Early edition. Published online before print June 28, 2010, doi: 10.1073/pnas.0908906107.

The expert opinion of a sample of climate researchers was evaluated as a means of assessing the nature of uncertainty in the physical response of the climate system to changes in radiative forcing. Although there was considerable disagreement among experts about the amount of global warming to be expected under different forcing scenarios, there was strong agreement that a high forcing scenario (i.e., one consistent with high concentrations of greenhouse gases) has a good chance of triggering a shift to a fundamentally different global climate state
.

Although formal scientific assessment processes that evaluate the large body of published peer-reviewed literature are the most robust means of evaluating the state of scientific understanding on climate change, expert opinion can provide a useful, complimentary form of assessment. A paper published recently in the journal the Proceedings of the National Academies of Science, by a team of international authors led by Canadian scientist Kirsten Zickfeld (currently with Environment Canada), reports the results of using formal expert elicitation methods to assess the views of a small group of top scientists regarding uncertainties in the physical response of the climate system to three different scenarios of radiative forcing (low, medium, high). The 14 internationally recognized experts interviewed for this work included two from Canada. The experts were unanimous in identifying cloud feedbacks as the factor that contributes most to uncertainty about future climate change, irrespective of the forcing scenario. This is in agreement with conclusions in the last IPCC assessment published in 2007. There was disagreement about the relative contribution of other processes to uncertainty. Expert opinion was quite divided about how much global warming is likely to result from the different forcing scenarios, and about the amount of uncertainty around those projections. For example, the 90% confidence interval of global mean surface temperature change in the year 2050 relative to year 2000 (a time span during which the forcing from all three scenarios was fairly similar) ranged from 0.1 – 3.8°C; median estimates ranged from 0.8 – 1.8°C. However, there was strong agreement among experts (13/14) that the probability of triggering a ‘basic change in state’ in the climate system that would persist for at least several decades and would have global consequences was ³50% under the high forcing scenario. This study presents an interesting exploration of the magnitude and sources of uncertainty in the physical climate system in response to changing atmospheric compositions. In light of the additional result that progress in resolving these uncertainties is likely to be modest in the near term, this paper serves as a reminder that strategies for responding to climate change will need to acknowledge these uncertainties and the attendant risks.