To understand how climate has changed in the past, and more importantly, what will happen in the future, scientists must rely on models.

To look at the past, present and future weather and climate, observations including EO satellites, airborne and ground-observations are used within numerical models to simulate the Earth system. The output gives a synthetic estimate of the state of the system, from the ground up around the middle of the stratosphere. This scientific method is called reanalysis and can extend over several decades.

Over the long term, to before records began, scientists can use proxy data to reconstruct past climates from many millions of years ago, which is known as Paleoclimatology. This includes proxy data from ice cores taken from areas such as Antarctica, and the Greenland Ice Sheet, which can tell scientists about temperature, precipitation, atmospheric composition, volcanic activity, and even wind patterns from past climates. It is important to have this data of the past, so that we can understand current climate change and current weather extremes, as well as projections of future climate, by comparing rates of past climate change to the present.

Predicting the changes in gaseous composition of the atmosphere is a difficult exercise. How the gaseous composition of the atmosphere will change in the future depends a great deal on the impact of human activities and our ability to control emissions. In addition to changes brought about by humans, increasing temperatures as a result of an enhanced greenhouse effect will impact the concentration of various gases in earth's atmosphere.

To predict atmospheric composition in the near future for particular gasses and constituents, Atmosphere composition forecasts are produced. These products are generated using observations from satellites which have been used as input for numerical models that can predict the atmospheric composition for several days ahead. They include, ozone forecasts, GHG forecasts, and aerosol forecasts, such as black carbon and dust, which can be useful for predicting where pollutants will go, as well as the sources. Pictured below is an example of an ozone forecast centred on the Antarctic region, for 22 March, 2022, and the image above is an example of an aerosol optical depth forecast for Europe, for 25 March, 2022, both provided by CAMS, the Copernicus Atmosphere Monitoring Service.

‍

Scientists can also predict the far future, typically up to 2100, using projections generated by models. These are typically future emissions scenarios, showing what is likely to happen if the world does or does not reduce global greenhouse gas emissions.

While space agencies such as ESA and organisations like Copernicus do not have a direct role in decision-making, the data their satellites provide does. Satellite data has been used for the Conference of the Parties (COP), to provide accurate scientific data to policy makers so they can effectively respond to climate change challenges.

‍

‍

‍