The presence of radiatively active gases in the Earth's atmosphere (water vapor, carbon dioxide, and ozone) raises its global mean surface temperature by 30 K, making our planet habitable by life as we know it. There has been an increase in carbon dioxide and other trace gases since the Industrial Revolution, largely as a result of man's activities, increasing the radiative heating of the troposphere and surface by about 2 W m−2. This heating is likely to be enhanced by resulting changes in water vapor, snow and sea ice, and cloud. The associated equilibrium temperature rise is estimated to be between 1 and 2 K, there being uncertainties in the strength of climate feedbacks, particularly those due to cloud. The large thermal inertia of the oceans will slow the rate of warming, so that the expected temperature rise will be smaller than the equilibrium rise. This increases the uncertainty in the expected warming to date, with estimates ranging from less than 0.5 K to over 1 K. The observed increase of 0.5 K since 1900 is consistent with the lower range of these estimates, but the variability in the observed record is such that one cannot necessarily conclude that the observed temperature change is due to increases in trace gases. The prediction of changes in temperature over the next 50 years depends on assumptions concerning future changes in trace gas concentrations, the sensitivity of climate, and the effective thermal inertia of the oceans. On the basis of our current understanding a further warming of at least 1 K seems likely. Numerical models of climate indicate that the changes will not be uniform, nor will they be confined to temperature. The simulated warming is largest in high latitudes in winter and smallest over sea ice in summer, with little seasonal variation in the tropics. Annual mean precipitation and runoff increase in high latitudes, and most simulations indicate a drier land surface in northern mid-latitudes in summer. The agreement between different models is much better for temperature than for changes in the hydrological cycle. Priorities for future research include developing an improved representation of cloud in numerical models, obtaining a better understanding of vertical mixing in the deep ocean, and determining the inherent variability of the ocean-atmosphere system. Progress in these areas should enable detection of a man-made “greenhouse” warming within the next two decades.