Page 27

CIM32

N°32 I quarterly I January 2014 Focus | 27 this variability, but they are obtained indirectly and come with their own share of uncertainties. “Take the Sun, for example,” says Dufresne, “we know that its activity—measured by the number of dark spots on its surface—influences the Earth’s climate. But how can we translate past records of sunspots observed by astronomers into radiative flux or spectrum?” To take another example, volcanoes produce ash clouds that cool the atmosphere. If recent eruptions are well documented, records of past volcanic events are scarce. How much ash was ejected? How large were the particles? “We also lack perspective on extreme events like storms,” Le Treut notes. “In the past, those who were in the best position to collect data on storms, namely sailors and pilots, were of course doing their best to avoid them. And for some parts of the world, the only record available is oral tradition.” THE LIMIT S OF modeling The chaotic nature of the climate and its variability are not the only factors that make research difficult. Future conditions are projected using numerical models based on the laws of atmospheric and oceanic physics. These models are very similar to those used by meteorologists to forecast the weather. “Yet unlike weather forecasting models, climate models cannot use direct observation to adjust trajectories,” Dufresne points out. “For example, when low pressure is observed over the Atlantic, the weather models integrate the data and use it to simulate a depression and its evolution. If the model proves inaccurate, the trajectory of the depression can be corrected by subsequent observations, resulting in fairly reliable forecasts over a few days’ span.” In climatology, the future is a blank page. The model has to do it all: define the initial conditions and determine the evolution of the main atmospheric structures, such as depressions and anticyclones, over several decades and with no observations to correct the calculations. “If, for example, the depressions and associated structures are placed too far south, all climatic structures will be out of position,” Dufresne says. “There’s nothing to correct them.” This becomes even more difficult at the subcontinental scale. Regional climates are highly dependent on the global climate, which must be very accurately simulated for the models to make reliable regional projections. And, since every climatic action triggers a reaction, the local effects in turn influence the global circulation. “We know, for example, that the evolution of rainfall will vary greatly across the planet,” Dufresne adds. “It will increase along the equator and at high latitudes, decrease in the subtropical zones, and rise or decline in the mid- latitudes depending on the seasons. But the exact frontiers between these zones depend on the global circulation and its variations, which are still subject to many uncertainties.” A GEN UINE CLIMATI C DANGER According to Le Treut, this element of uncertainty raises a key problem for research on how to best adapt to climate change. “We can’t blindly trust the models,” he warns. “Just because they predict more drought in a given region doesn’t mean that it should be planted with drought-resistant species. Projections cannot be seen as indications of permanent risk. Rainfall could also increase locally, and over-specialized plants would not be able to survive.” Yet what is certain is that the situation is urgent, and climate risk is an unavoidable reality. “As climatologists, our role is to ring the alarm bells and debunk the myths on climate change,” Le Treut says. “We will probably never know what the future holds in store, but that shouldn’t keep us from taking action. We cannot know precisely when or where an earthquake will occur, but we know which areas are most at risk, and we can make buildings more resistant. Denying climate change is like saying that earthquakes don’t exist because they cannot be accurately predicted.” 01. L aboratoire de météorologie dynamique (CNRS / École Polytechnique / UPMC / ENS). Contact information: Jean-Louis Dufresne > jean-louis.dufresne@lmd.jussieu.fr Hervé Le Treut > letreut@lmd.jussieu.fr 16 Extreme weather phenomena, like this storm in Nice in 2011, increasingly threaten the planet’s populations. 17 The reduction of greenhouse gas emissions to limit climate change is a recognized necessity, but difficult to put into action. 16 17 © V. HACHE/AFP © jianan yu /REUTERS


CIM32
To see the actual publication please follow the link above