Many people equate climate with the weather, but it is more than that. Climate scientists study the planet’s atmosphere and its interactions with water, forests, grasslands and other ecosystems and the creatures that depend on them. This interdisciplinary field of research, known as climatology, draws upon physics, math, chemistry and biology to understand and predict long-term climate trends as well as short-term extreme events.
Complex computer models, incorporating representations of a wide range of climate system processes, are a mainstay of modern climate research. These model simulations allow researchers to test hypotheses and make projections (i.e. predictions of future conditions based on external forcing scenarios). Attribution studies, which attempt to identify the causes of observed changes in climate, are also a common practice (Swainson et al. 2019). These include basic physical reasoning—considering whether the observed change is qualitatively consistent with a given potential cause—and empirical-statistical analyses, sometimes called “fingerprint” studies, in which GCM/ESMs are used to simulate what would happen over time if one causal factor changed but other factors stayed constant.
Climate science relies on a large amount of data, collected by an international network of monitoring stations that cover the globe. Thousands of these station records, comprising millions of individual measurement values, are merged, subjected to quality control procedures and homogenized, to construct global temperature datasets. These are then used in forecasts and projections. A key goal is to increase the confidence level of these predictions by considering a number of different modeling versions or model versions, in an approach known as ensemble forecasting.