By averaging data over long time spans, typically 30 years or more, climatology datasets are developed. These datasets serve as reference points for comparing points in time to what would be expected. For example, is the temperature warmer or cooler this month than would be expected from the historical average? Temperature climatologies are extremely important as measurements, models, and reconstructions of Earth's temperature changes are analyzed.
The Global Historical Climatology Network monthly analysis from the National Climatic Data Center uses data from thousands of land and ocean temperature stations worldwide to compute global temperature averages and anomalies. The majority of these locations date back to the 1950's. This image plots the temperature anomaly for the past month. Blue areas are cooler than the historical average, red areas are warmer. Data in the high latitudes, around the Arctic and Antarctic circles, are not plotted as there are few weather monitoring stations in these regions, so data is too sparse to be included in the analysis.