...
One of the two papers doesn't even bother, instead focusing on trends. It uses 50 years of weather data to identify single-day and five-day instances of extreme precipitation in the Northern Hemisphere. They then compared the historic record to ensembles of eight climate models, run under three different conditions: a stable climate, natural forcings like solar and volcanic activity, and natural forcings plus human-induced ones. The work suggests that, given the trends in solar and volcanic activity, we'd actually expect to see fewer extreme weather events. Instead, there's been an upward trend in one-day events that the authors ascribe to anthropogenic forcings. Five-day events show a much weaker linkage.
Still, that's just a general trend. The second paper looks at a specific event, a series of extreme floods that took place in England and Wales in the year 2000. Here, the authors used a single model of the Northern Hemisphere climate and fed it the relevant data on things such as ocean temperatures and ice cover; they also created a control scenario with conditions approximating those in 1900. The outputs of these scenarios were fed into a runoff model of the affected area in order to determine how often events of similar magnitude occurred.
...
All the scenarios undoubtedly included flooding. But, by comparing the results from runs based on the actual conditions to ones without anthropogenic forcing, the authors could make an estimate of how climate change has influenced the risks of an extreme event. In nine out of 10 cases, that risk went up by 20 percent or more; in two out of three cases, it was raised by more than 90 percent. Given those probabilities, the authors conclude that climate change is very likely to have contributed to the floods of 2000.
...
