Professor Myles Allen CBE FRS

Abstract:

Peak human-induced warming is primarily determined by cumulative CO2 emissions up to the time they are reduced to zero1,2,3. In an idealized economically optimal scenario4,5, warming continues until the social cost of carbon, which increases with both temperature and consumption because of greater willingness to pay for climate change avoidance in a prosperous world, exceeds the marginal cost of abatement at zero emissions, which is the cost of preventing, or recapturing, the last net tonne of CO2 emissions. Here I show that, under these conditions, peak warming is primarily determined by two quantities that are directly affected by near-term policy: the cost of ‘backstop’ mitigation measures available as temperatures approach their peak (those whose cost per tonne abated does not increase as emissions fall to zero); and the average carbon intensity of growth (the ratio between average emissions and the average rate of economic growth) between now and the time of peak warming. Backstop costs are particularly important at low peak warming levels. This highlights the importance of maintaining economic growth in a carbon-constrained world and reducing the cost of backstop measures, such as large-scale CO2 removal, in any ambitious consumption-maximizing strategy to limit peak warming.

Abstract:

A succession of storms reaching southern England in the winter of 2013/2014 caused severe floods and £451 million insured losses. In a large ensemble of climate model simulations, we find that, as well as increasing the amount of moisture the atmosphere can hold, anthropogenic warming caused a small but significant increase in the number of January days with westerly flow, both of which increased extreme precipitation. Hydrological modelling indicates this increased extreme 30-day-average Thames river flows, and slightly increased daily peak flows, consistent with the understanding of the catchment’s sensitivity to longer-duration precipitation and changes in the role of snowmelt. Consequently, flood risk mapping shows a small increase in properties in the Thames catchment potentially at risk of riverine flooding, with a substantial range of uncertainty, demonstrating the importance of explicit modelling of impacts and relatively subtle changes in weather-related risks when quantifying present-day effects of human influence on climate.