When we first heard much about atmospheric rivers (AR) they were dumping incredible amounts of rain onto 2 parts of Britain. David A Lavers and his colleagues from the Universities of Reading and Iowa have now released information on how these water vapour carriers extend pole-ward and become more extreme as the thermodynamic responses of the planet to our global warming.
The rivers are narrow, at a height of 1 - 2.5km. They can be thousands of km long, and measure about 300km in width. The west of North America and the western European areas are most affected, during winter. Extra-tropical cyclones venture north at the same time and help to produce the floods (caused by the AR phenomenon). The amount of water actually exceeds normal river flows, being up to 4500 times greater in flow! When this amount of water falls on impermeable upland areas, the effect is disastrous. Eastern low river basins see less affected by the flows.
The number and frequency of ARs is likely to increase, causing more disasters for those who live in "the wrong place." From 2-14 events every year, there is likely to be an increase in number. While air temperatures continue increasing, there will be more capacity to carry water vapour. That means potentially higher rainfall totals and we all know what kind of flood that will produce, given recent experience.
For California, there are 7 projected climate models that predict such increases. European papers don't assess AR mechanisms so much. The CMIP5 (Fifth Climate Model Intercomparison Project) shows slightly stronger storm tracks near Britain. With detailed 6-hourly input, the North Atlantic patterns predicted become more obvious. More of NW Europe could become involved in this pattern of real disaster. Sources, flows and sinks were evaluated from 1980 till 2005, using reanalysis to identify faults in the 5 models employed. A 3-D representation of the "rivers" was produced which then represented actual conditions as nearly as possible. As no attempt was made to represent pre-industrial climate situations, the frequent mistakes experienced in those models were avoided!
When tested the AR events that actually took place were accurately represented in the AR structures within the models. This shows that predictions for 2074 -2099 should be taken seriously. The amount of damage in both Cumbria and western England and Wales was large enough to worry insurance companies who are concerned with more than 200,000 high-risk properties.
The models strikingly agreed that double the number of ARs would occur, while the winds responsible would remain directionally similar. Some difficulties in calculating the future intensity were down to simplified thermodynamic calculations. The intensity nevertheless is likely to increase because of the rising amount of water carried by the warming atmosphere
This is an open access paper worth looking at for the models' graphs alone; the authors publish in Environmental Research Letters.