To the Editor —

We agree with the point made in a recent Editorial in this journal1 that the assumptions behind models of all types, including integrated assessment models (IAMs), should be as transparent as possible. However, it is incorrect to imply that the IAM community is just “now emulating the efforts of climate researchers by instigating their own model inter-comparison projects.”

In fact, model comparisons for integrated assessment and climate models followed a remarkably similar trajectory. Early general circulation model (GCM) comparison efforts2 evolved to the first Atmospheric Model Inter-comparison Project (AMIP), which was initiated in the early 1990s3. Atmospheric models developed into coupled atmosphere–ocean models (AOGCMs) and results from the first Coupled Model Inter-Comparison Project (CMIP1) became available about a decade later4.

Results of first energy model comparison exercise, conducted under the auspices of the Stanford Energy Modeling Forum, were published in 19775. A summary of the first comparison focused on climate change was published in 19936. As energy models were coupled to simple economic and climate models to form IAMs, the first comparison exercise for IAMs (EMF 14; https://emf.stanford.edu/projects) was initiated in 1994, and IAM comparison exercises have been ongoing since this time7,8,9,10 — and were recently assessed in the latest IPCC report11 — including a publicly accessible database of scenarios (https://secure.iiasa.ac.at/web-apps/ene/AR5DB).