Can the observed climate changes be explained by natural variability, including changes in solar output?
Since our entire climate system is fundamentally driven by energy from the sun, it stands to reason that if the sun's energy output were to change, then so would the climate. Since the advent of space-borne measurements in the late 1970s, solar output has indeed been shown to vary. There appears to be confirmation of earlier suggestions of an 11 (and 22) year cycle of irradiance. With only 20 years of reliable measurements however, it is difficult to deduce a trend. But, from the short record we have so far, the trend in solar irradiance is estimated at ~0.09 W/m2 compared to 0.4 W/m2 from well-mixed greenhouse gases. There are many indications that the sun also has a longer-term variation which has potentially contributed to the century-scale forcing to a greater degree. There is though, a great deal of uncertainty in estimates of solar irradiance beyond what can be measured by satellites, and still the contribution of direct solar irradiance forcing is small compared to the greenhouse gas component.
However, our understanding of the indirect effects of changes in solar output and feedbacks in the climate system is minimal. There is much need to refine our understanding of key natural forcing mechanisms of the climate, including solar irradiance changes, in order to reduce uncertainty in our projections of future climate change.
In addition to changes in energy from the sun itself, the Earth's position and orientation relative to the sun (our orbit) also varies slightly, thereby bringing us closer and further away from the sun in predictable cycles (called Milankovitch cycles). Variations in these cycles are believed to be the cause of Earth's ice-ages (glacials). Particularly important for the development of glacials is the radiation receipt at high northern latitudes. Diminishing radiation at these latitudes during the summer months would have enabled winter snow and ice cover to persist throughout the year, eventually leading to a permanent snow- or icepack. While Milankovitch cycles have tremendous value as a theory to explain ice-ages and long-term changes in the climate, they are unlikely to have very much impact on the decade-century timescale. Over several centuries, it may be possible to observe the effect of these orbital parameters, however for the prediction of climate change in the 21st century, these changes will be far less important than radiative forcing from greenhouse gases.
Due to the enormous complexity of the atmosphere, the most useful tools for gauging future changes are 'climate models'. These are computer-based mathematical models which simulate, in three dimensions, the climate's behavior, its components and their interactions. Climate models are constantly improving based on both our understanding and the increase in computer power, though by definition, a computer model is a simplification and simulation of reality, meaning that it is an approximation of the climate system. The first step in any modeled projection of climate change is to first simulate the present climate and compare it to observations. If the model is considered to do a good job at representing modern climate, then certain parameters can be changed, such as the concentration of greenhouse gases, which helps us understand how the climate would change in response. Projections of future climate change therefore depend on how well the computer climate model simulates the climate and on our understanding of how forcing functions will change in the future.
The IPCC Special Report on Emission Scenarios determines the range of future possible greenhouse gas concentrations (and other forcings) based on considerations such as population growth, economic growth, energy efficiency and a host of other factors. This leads a wide range of possible forcing scenarios, and consequently a wide range of possible future climates.
According to the range of possible forcing scenarios, and taking into account uncertainty in climate model performance, the IPCC projects a global temperature increase of anywhere from 1.4 - 5.8°C from 1990-2100. However, this global average will integrate widely varying regional responses, such as the likelihood that land areas will warm much faster than ocean temperatures, particularly those land areas in northern high latitudes (and mostly in the cold season).
Snow extent and sea-ice are also projected to decrease further in the northern hemisphere, and glaciers and ice-caps are expected to continue to retreat.
Source: National Climatic Data Center, U.S. Department of Commerce