Publication Date:
2017-07-06
Description:
In December 2015, 195 countries agreed in Paris to hold the increase in global mean surface temperature (GMT) well below 2.0 °C above pre-industrial levels and to pursue efforts to limit the temperature increase to 1.5 °C. Since large financial flows will be needed to keep GMTs below these targets, it is important to know how GMT has progressed since pre-industrial times, taking short-term and long-term (decadal) natural variability into account. However, the Paris Agreement is not conclusive as for methods to calculate it. Should trend progression be deduced from GCM simulations or from instrumental records by (statistical) trend methods? Which trend model should be chosen and what is pre-industrial? Does trend progression depend on the specific GMT dataset chosen? To find answers to these questions we performed an uncertainty and sensitivity analysis where datasets and model choices have been varied. For all cases we evaluated trend progression since pre-industrial, along with uncertainty information. To do so, we analysed four trend approaches and applied these to the five leading GMT products. As a parallel path, we calculated GMT progression from an ensemble of 106 GCM simulations, corrected for natural variability. We find GMT progression to be largely independent of various trend model approaches. However, GMT progression is significantly influenced by the choice of GMT datasets. Both sources of uncertainty are dominated by natural variability. Mean progression derived from GCM-based GMTs appears to lie within the range of the trend-dataset combinations. A difference between both approaches lies in the width of uncertainty bands: bands for GCMs are much wider. Results appear to be robust as for specific choices for pre-industrial. Our Paris policy recommendation would be to choose a spline or IRW trend model and estimate it on the average of the five leading GMT datasets, where 1880 is taken as base year. Given this choice trend progression for 2016 accounts for 1.01 ± 0.13 °C (2-σ).
Print ISSN:
1814-9340
Electronic ISSN:
1814-9359
Topics:
Geosciences
Permalink