Applied Seminar - Projecting global mean surface warming over the 21st century, Phil Goodwin (UoS, Oceanography) Seminar

Event details

The United Nations Paris Agreement of 2015 set out a globally political target to restrict the global mean surface warming to less than 2.0 °C above preindustrial times, and preferably less than 1.5 °C. The preferred projections of global mean surface warming for different future forcing scenarios are undertaken using complex Earth system models that include 3-dimensional representations of the atmosphere and ocean systems. However, these Earth system models make very different future projections, leading to high uncertainty in future warming for a given scenario. The complex models are highly computationally expensive and difficult to build and run, so there are only ~ 20 such independent models to make future projections with. Also, many of the Earth system model simulations do not agree with observations of historic surface warming and ocean heat uptake, leading to uncertainty when using the same simulations to make future projections. This talk explores an alternative strategy for making future projections of global mean surface warming. Utilising a single equation that links global mean surface warming to carbon emissions, a highly computationally efficient Earth system model is discussed. This efficient model is used to generate 100-million simulations of historic climate change to year 2017 using a Monte Carlo approach. Each of these 100-million simulations is then tested against historic observations of the real climate system, and only the simulations that show good observational-agreement (~0.03%) are used to make future projections.