Are climate models reliable?

CLAIM

Climate models are unreliable. 

FACT

Climate models are reliable and are successful in projecting future conditions of the Earth. 

WHAT THEY SAY

Climate models are not so reliable, especially in terms of predicting whether global warming will continue in the future. There’s too much uncertainty to trust these climate models which are projecting/predicting climate change. 

WHAT WE FOUND

Climate models are sets of mathematical equations representing the processes and interactions that drive the Earth’s climate. They cover the land, oceans, atmosphere, and ice-covered areas of the planet. Since the proper study of climate is not possible in a laboratory, these models are used to simulate various factors acting on the Earth’s climate to project how climate change will be taking place in the future. 

One can say that climate models are an extension of weather forecasting. While weather models involve predictions over short time spans of specific areas, climate models are used to predict long time spans and are broader in comparison to weather models. 

Some of the common usages of climate models in quantifying different climate change-related aspects include models to project the amount of human-induced greenhouses that will get emitted in the future, models to project the reaction of the Earth’s climate system to such emissions, and impact models for projecting the way these climate changes will have an effect on the inhabitants of this planet. 

DIFFERENT TYPES

The three common types of simple climate models are energy balance models, intermediate complexity models, and general circulation models. These models employ various mathematical calculations to emulate various factors like atmospheric mixing and ocean current that affect climate while making projections. 

Energy balance models are used to predict climate changes as a result of Earth’s energy budget. They take into consideration surface temperatures because of the incoming solar energy, reflectivity, and the natural cooling that takes place when the earth emits the heat back. 

Intermediate complexity models are different from the energy balance models in the sense that they also include Earth’s geographical structures like oceans, land, and ice for example. The introduction of geographical features allows the simulation of big-scale climate scenarios like ocean currents, glacial changes, and atmospheric composition over long time periods. 

General circulation models are regarded as the most exact and complex models for the prediction of climate change. These models take into account the type of land, carbon cycle, glacial composition, ocean circulation, atmospheric chemistry, etc of the particular area under consideration. They are more complex than the previous two types of models mentioned above but also require more computing time in comparison. 

HINDCASTING

All said and done, these climate models need to be tested to find out if they ‘actually work’ or in other words, whether their projections are reliable and can be trusted. Since waiting for another 20-30 years to find out if a model ‘works’ or not is feasible, models are tested against what we ‘reliably’ know to have happened i.e, against the past. It is only practical that a model that can correctly predict trends from some fixed ‘beginning point’ in the past, can be trusted to predict with ‘certainty’ something related to the future. This process of testing models is known as Hindcasting, also known as backtesting. For example, Hindcasting ensures that only models that can accurately calculate past climate changes are used to predict future global warming. 

GETTING BETTER AND BETTER 

Climate models are getting better and better over time as technology is evolving with every passing year. A study of climate models published between 1970 and 2007 found that almost all the models were quite good in their prediction of temperatures in the years ahead. The study found that “climate models published over the past five decades were skillful in predicting subsequent GMT (global mean surface temperature) changes, with most models examined showing warming consistent with observations, particularly when mismatches between model-projected and observationally estimated forcings were taken into account.”

TOO ALARMIST?

Many people, especially those who reject the scientific consensus on climate change use the word ‘alarmist’ as a form of criticism to speak about the people who consider the topic of ‘climate change’ objectively and scientifically. The ‘climate models’ which predict or project climate change are the main targets in this as it is often alleged that such models are too melodramatic or alarmist in their projections. But in reality, the ‘opposite’ is the reality, climate models are actually quite ‘conservative’ in the predictions they make. 

For example, this graph of sea-level rise shows that projections by climate models are more ‘conservative’ in comparison to satellite observations:

We see here that the models have actually ‘understated’ the problem. The observed sea level (satellite result) is tracking at the upper range of the climate model projections. Of course, all models have limits because of the complex systems but models are improving over time. With the increasing availability of real-world information such as satellites, the projections of models can go through refinement to enhance their capabilities and usage. 

Category: