The large changes in global temperature and atmospheric CO2 from deep Earth history prove that natural changes in the climate should be expected over long time periods. Climate change on human timescales, season-to-season and year-to-year, also occurs for reasons that have nothing to do with human influence.

In gauging the human influence on climate, the strongest conclusions from climate scientists have been for the last half-century or so. Multiple studies find that most of the warming over that period can be caulked up to human influence, because solar and volcanic effects, and the biggest climate oscillations were mostly neutral over that period. Additionally, the pattern of temperature increase (global changes, larger near over land than ocean, concurrent with ocean warming) is inconsistent with the temperature patterns that accompany the known sources of variability on the scale of half centuries (high latitude oceans).

Some experts are critical of the conclusion that human activities are the primary cause of the most recent warming and argue that it discounts plausible alternative explanations. There are minority concerns that natural variability could account for a significant fraction of the observed temperature increase and mainstream climate science is overstating the role of human activities in recent climate change.

One set of concerns holds that some unknown external factor—excluded from computer experiments like those shown in Figure 4, and unaccounted for in studies of modern climate change—has contributed to the recent temperature increase. However, the major external factors that scientists think have affected climate over the late 21st century are already included in such simulations. Other potential influences have been proposed, but not proven to be important to climate over the late 20th century.


Another set of concerns holds that random oscillations in climate system could have caused a significant amount of the measured global temperature increase. Such large internal variability, moreover, is inconsistent with observations and scientific understanding to date. Internal variability in climate models and statistical analyses of historical global temperature is smaller than the warming observed in the late 20th century over the same time periods. The change in temperature over that period was about 0.6 °C. While is not much confidence that climate models have the right amount of internal variability, because of limited observational evidence, internal variability would have to be three times larger than in the models to disturb the conclusion that human influence was the leading climate driver over the last half-century. No mechanism that could provide such large variability, but would have escaped detection over the last century, has been articulated.