Analyzing regional changes using observations and models

At regional scales, the temperature and precipitation anomalies are strongly influenced by the dominant mode of variability of the climate system, such as the North Atlantic Oscillation (NAO) or El Nino-Southern Oscillation (ENSO). Any change in these modes of variability, for instance, a shift from a positive to a negative phase, could be purely related to internal dynamics or caused by changes in the forcing. The analysis of proxy data could allow determination of the magnitude of the changes and the potential correlation with the reconstructions of past forcings. Alternatively, if a model is able to simulate reasonably well the observed changes, this could provide a reasonable hypothesis for the mechanism responsible for the change. Despite the complementary information from model and data, determining

Figure 7.8 Reconstructed (top) and simulated (bottom: GCM, general circulation model) annual (ANN) average surface temperature differences between 1660-80 and 1770-90. The reconstructed surface temperatures are based on a multi-proxy estimate using tree rings, ice cores, corals, and historical data. Model results are based on the sum of the response in two simulations: one incorporating reconstructed solar irradiance changes during this period and one using volcanic forcing scaled to changes over this time. (From Schmidt etal. 2004. Copyright Elsevier (2004).)

Proxy data reconstruction, 1660-80 vs 1770-90, ANN surface temperature

Proxy data reconstruction, 1660-80 vs 1770-90, ANN surface temperature

m

-1.5 -0.8 -0.6 -0.4 -0.2 0.0 0.2 0.4 0.6 0.8 GCM simulation, 1680 vs 1780 solar + volcano, ANN surface temperature

-1.5 -0.8 -0.6 -0.4 -0.2 0.0 0.2 0.4 0.6 0.8 GCM simulation, 1680 vs 1780 solar + volcano, ANN surface temperature

the cause of a change in one mode of variability or in the probability to have one phase of a mode is nearly impossible, currently, as internal variability alone could explain the observed changes in those modes.

The observed substantial cooling in large parts of Europe during the late 17th and early 18th centuries appears to be related to long-term variations in the NAO pattern of climate variability (Figure 7.8), which are associated with variations over time in the pattern of the Northern Hemisphere jet stream. In model simulations, these changes are associated with a large-scale dynamical response of the climate system to natural radiative forcing by explosive volcanic activity (Robock 2000; Schmidt et al. 2004; Shindell et al. 2004) and solar output (Shindell et al. 2001; Schmidt et al. 2004) that interacts with the NAO pattern of atmospheric circulation. This has led to the suggestion that the moderate apparent lowering of solar irradiance during the 17th century leads to only moderate decreases in hemispheric mean temperature, but a tendency for strong annual mean cooling in some regions, such as Europe.

Recent work suggests that the El Niño-Southern Oscillation may also be an important component of the response of the climate to forcing over the past 1000 years. Model simulations using simple models of the coupled tropical ocean-atmosphere to study the response of ENSO to solar and volcanic forcing (Mann et al. 2005a) indicate a counterintuitive tendency towards El Niño (warm eastern and central tropical Pacific) conditions in response to negative radiative forcing (past explosive tropical volcanic eruptions or decreases in solar irradiance) and a tendency for La Nina-like conditions in response to positive radiative forcing (i.e., increases in solar irradiance). This prediction matches available evidence from tropical Pacific coral records (Cobb et al. 2003). The response of ENSO to the combined effects of solar and volcanic forcing in past centuries provides a potential explanation for why the tropical Pacific appears to have been in a cold La Ninalike state during the so-called "Medieval Warm Period" and a warm El Nino-like state during the "Little Ice Age" (Adams et al. 2003).

Changes in oceanic circulation could also have an impact on regional changes during the past millennium. Unfortunately, there are not enough high-frequency observations in the ocean to derive a clear and comprehensive view of oceanic changes during this period. The available information is mainly derived from some local observations and model results. A thorough model-data comparison has not been performed yet but that could certainly be very useful and would help to fill some of the gaps in our knowledge. In particular, when driven by a reduction of the solar irradiance (or more generally in case of a moderate cooling), models tend to simulate an intensification of the meridional overturning circulation in the Atlantic (Cubasch et al. 1997; Goosse and Renssen 2004; Weber et al. 2004). This is due to an increase in the density of the surface water in the North Atlantic, caused by cooling and by changes in the freshwater budget, inducing more vigorous deep water formation there. The intensification of the oceanic circulation implies a larger heat transport towards the high latitudes and is thus associated with a negative feedback in those regions of the initial cooling. For a moderate warming, the opposite mechanism is noticed, presenting some similarities with what is expected for the 21st century (e.g. Gregory et al. 2005). A reduction of the solar forcing might also modify the probability of having years with significantly reduced inflow of warm Atlantic waters at high latitudes (Goosse and Renssen 2004). This leads to a higher probability of having very cold conditions in the Nordic Seas and over Scandinavia, resulting overall in a very strong positive feedback in the northern North Atlantic and surrounding regions during those cold periods. There is also evidence for the potentially important response of the Atlantic meridional overturning to low-frequency changes in the NAO (Delworth and Dixon 2000), which have in turn been simulated as a possible response to long-term solar and volcanic forcing in past centuries (Shindell et al. 2001, 2004; Schmidt et al. 2004).

0 0

Post a comment