What is a climate model and projections?

A climate model is a set of numerical simulations of important physical process involving the land, atmosphere, ocean, and ice-covered regions around the globe. Climate models are used to simulate the transfer of energy and materials through the climate system. Climate models are an essential tool for understanding Earth’s climate since they provide past climate cycles and projections for future climate. Climate model outputs are used by scientists to drive climate science and to understand how human activity can affect future climate.

Building and running a climate model is a complex process of identifying and quantifying Earth system processes, representing them with mathematical equations, setting variables to represent initial conditions and including subsequent changes in climate forcing (e.g. changes in albedo, changes in greenhouse gasses in the atmosphere, etc). The models are repeatedly solving equations and laws of the physical, chemical and biochemical mechanisms of our Earth system (figure below). Such massive calculations require the use of powerful supercomputers.

Currently, climate models use the “Representative Concentration Pathways” (RCPs), which are descriptions of the future (climate projection), based on different socio-economic scenarios of how global society could develop.

Usually we use the General Circulation Models, also known as Global Climate Models (GCM) which simulate the physics of the climate for the entire world. More sophisticated “coupled” models have linked together the atmosphere and ocean to provide a comprehensive representation of the climate system. The GCM have a coarse spatial resolution of approximately 2 degrees.

There are also the Regional Climate Models (RCM) with a similar functionality as the GCM but for a limited area over the Earth, with an average resolution of 0.5 degrees and higher. RCM are one way of taking information provided by a GCM or coarse-scale observations and applying it to a specific area or region (downscaling). 

Read more: Statistical and dynamical downscaling