The Future of Extreme Events

The future of extreme events

Understanding changes in climate extremes is vital for assessing impacts on human and natural systems. In this paper Tebaldi et al analyse the effectiveness of nine general circulation models (GCMs) at simulating climate extremes (ten indicators associated with climate extremes: five related to temperature, five to precipitation). The indicators, based on those developed by a previous study1, were chosen because they represent a wide spectrum of climatic variables, ranging from a heat wave duration index to maximum numbers of consecutive dry days. An additional criterion was that these indicators had to be ‘robust’ in terms of measurement and predictive uncertainty – it was necessary there were sufficient occurrences found in the historical observation period of 1961 to 1990. The authors acknowledge some of the problems and limitations of using these indicators, particularly in terms of the choice of thresholds. The indicators were assessed against simulated historical, future projected trends and in terms of geographical patterns.
Analysis from the nine model simulations agreed with observations from the historical record. This leads the authors to make a number of general statements regarding tendencies for climate extremes for the 21st Century:
1. a positive trend for the following indicators: growing season, heat waves and warm nights, and a negative trend for frost days and temperature range (difference between highest and lowest temperatures of the year);
2. over the next century all models show greater temperature extremes consistent with a warmer climate due mainly to increases in anthropogenic greenhouse gas emissions;
3. projected geographical changes indicate a number of ‘hot spots’ associated with particular indicators;
4. there is a global trend towards intensified precipitation, particularly in the northern hemisphere.

Potential effects of climate change on marine methane hydrates

Marine methane hydrate, found at the margins of Continental shelves, is a form of ice whose crystal structure contains a large amount of methane gas. As methane is a significant greenhouse gas, these hydrates are attracting increasing attention due to their potential role in climate change. Studies have produced different estimates for the mass of these hydrates, and estimates for ‘carbon weight’ range from 100 to 74,400 Gt (the atmosphere is currently estimated to hold around 750 Gt of carbon). This is a significant reservoir which may be sensitive to changes in the climate. Fyke and Weaver used seafloor temperature change (simulated under a series of climate sensitivity and climate change experiments) to explore the sensitivity of the marine gas hydrate stability zone (GHSZ) to elevated levels of atmospheric CO2. The results suggest that the volume of global GHSZ may be affected significantly by future atmospheric greenhouse gas increases, with the GHSZ response varying in timing and intensity as a function of regional seafloor temperature change.

Oceanic conditions and the European summer of 2003

Black and Sutton examine the influence of oceanic conditions on the development of the European summer of 2003. Understanding the causes of such an event may allow conclusions to be drawn about its likelihood and the possibility of prediction of future events months or seasons ahead. The Hadley Centre HadAM3 climate model was used for four different model simulations with only sea surface temperature (SST) conditions altered each time:
(i) observed SST data for 1970-1999
(ii) observed SST for January 2003 from around the globe
(iii) observed SST for January 2003 for everywhere except the Indian Ocean,
for which the 1970-1999 observed data was used
(iv) observed SST for January 2003 for everywhere except the Mediterranean Sea, for which the 1970-1999 observed data was used.
The model results suggest that oceanic conditions in both the Indian Ocean and Mediterranean Sea did play a part in the development of warm air temperatures through the summer of 2003, and that the warm and dry conditions are therefore potentially predictable given adequate observations or predictions of SST.

Climate driven trends in ocean productivity

Behrenfield et al report on changes in the net primary productivity (NPP) of ocean phytoplankton over the last decade as determined from satellite measurements of ocean
colour. Oceanic phytoplankton is a vital link in the cycling of carbon between organic and inorganic stocks. Changes in NPP are evaluated alongside a multivariate El Ninõ southern oscillation index (MEI), which encompasses changes to a range of environmental forcings, including: sea-level pressure, surface winds, sea surface temperature, surface air temperature and cloudiness. The results indicated a significant correspondence between the levels of NPP and MEI records. This suggests that MEI may provide a first order approximation for past changes in NPP.
This relationship has important implications regarding future NPP under a climate change. In the last decades trends in NPP have closely matched changes in SST, suggesting that warmer conditions in the future would result in lower and redistributed ocean carbon fixation relative to contemporary conditions. Consequentially this could lead to changes in the magnitude and distribution of air-sea CO2 exchange, basin-scale biological processes and fisheries yields.

Simulated climatic variability

A modelling exercise to determine the extent to which internal climatic variability, as opposed to natural external factors (such as volcanic eruptions and solar variability) and human factors (greenhouse gas emissions), can achieve variations in climate such as those associated with the Medieval Warm Period (MWP) and Little Ice Age (LIA). The authors use the CSIRO Mark 2 global climate model to simulate 10,000 years of 'present' climate conditions, using a constant CO2 concentration of 330 ppm. No external forcings were permitted during the model run. Analysis of the final 1000 years of the output revealed simulated naturally occurring climatic variability (in terms of, for example, temperature, precipitation and sea-ice) of the kind that would be associated with the concepts of the MWP and LIA. The most plausible explanation for these events occurring in the simulation is stochastic influences within the climate system. But the infrequency of these events within simulation implies that they are at the extreme limit of stochastic forcing. The authors conclude that the unforced climate system is unable to sustain the generation of long-term climatic anomalies.

UKCIP; www.eci.ox.ac.uk