Atmospheric Rivers

Figure 1. IR satellite image from METEOSAT at 8th January 2005. Clouds are in white and the red arrow shows the direction of the air flow in the WCB.

The advent of satellite imagery opened our eyes to many unseen aspects of the airflow in the atmosphere which had been invisible to most scientific instruments. It gave us the possibility to trace the transportation of moisture from tropical regions to the higher latitudes. Further, we could integrate the amount of the water vapour throughout the depth of the troposphere (lowest layer of the Earth’s atmosphere) everywhere, especially over the oceans. The latter approach proved to be an invaluable tool to discover a weather phenomenon called “atmospheric river”.

It all began with the measurements of carbon monoxide for the Air Pollution from Satellite (MAPS) project in 1990. A puzzling feature was the presence of high carbon monoxide values well removed from their sources. Newell et al. (1992) were first who used this satellite data to shed some light on the transport problem. They found corridors of vertically integrated water vapour fluxes over the ocean and used the term “tropospheric rivers” to emphasize the large amounts of water vapour being transported in the troposphere. This striking finding was further investigated by Zhu and Newell (1998) at MIT using observations from polar-orbiting satellites and research aircraft over the eastern Pacific Ocean in the winter of 1997-98. Their result showed that these filamentary features, which they called atmospheric rivers (ARs), constitute a significant fraction of the total moisture transport, and almost all the meridional (south to northward) transport atmid-latitudes.

Figure 2. Flooding in Norway 14-15th september 2005.

The interest in studying ARs has increased in recent years because of their strong link to several flood events in the extratropics (latitudes beyond the tropic but not polar regions). In these regions, atmospheric large-scale low pressure systems are associated with extreme precipitation during the cold seasons. They form where cold-dry polar and warm-moist tropical air masses meet. This result in a very unstable atmosphere. The unstable atmosphere then responds by rearranging the air in the boundaries between two air masses. Consequently, warm and cold fronts are formed in pairs as the low system moves poleward. In this way, low-pressure systems transport the moist tropical air toward higher latitudes. As the cyclone moves over the ocean, it may take up additional moisture which increases the potential for intense precipitation later when it hits land. In the 1960’s, observations in the the troposphere revealed that the ascending air stream over the advancing warm front is responsible for the transport of the moisture. Carlson (1980) termed this warm and wet flow the warm conveyor belt (WCB). The IR (infra-red) image in the figure (1) displays the WCB (red arrow) and its associated cloudiness coverage over the North Atlantic.

Thanks to satellite imagery,we now have a better understanding of ARs. We know that they are narrow regions of increased moisture transport in the troposphere, having lengths several times their widths and translating throughout the troposphere. They have a transient nature which means their lifetime is short (about 1-3 days). Satellite pictures show that they are always present somewhere over the globe. They are embedded within the extratropical cyclones and are a wider part of the WCBs. In fact, circulations in the low pressure systems are the large scale driver of ARs, and WCBs are like expressways for the passage of water vapour coming from the equator.The strongest ARs, which are at around 2000-3000 kilometres long and several hundred kilometres across, carry more water than the Amazon! Therefore, an atmospheric river mega-storm could be a hugely destructive event.

However, those systems that strike the coastal area (called “land-falling ARs”) are of major interest for weather forecasters. When they encounter coastal mountains, the air which is rich in water vapour, rises up the mountain, cools, condenses and falls as intense rain or snow. Therefore, western coasts of US , UK and Norway are vulnerable to this type of weather situation. It is difficult to predict this type of precipitation in the sense that numerical models have problems with their timing and duration. Nevertheless, satellite images help us in nowcasting and very short-range forecasting of precipitation events associated with the presence of ARs.

Vertically integrated water vapour over ocean at 13th September 2005. The red colour implies where the water vapour is concentrated.

Western Norway is prone to extreme weather mainly because it is located at the end of the North-Atlantic storm tracks (the regular and frequent passage of extratropical low pressure systems). An example is the extreme precipitation event on southwest coast of Norway in September 2005 (Figure 2). There is no climatological study concerning the relation between the ARs and flooding events in Norway, but a few case studies show that extreme precipitation may coincide with the presence of moisture plumes originating in the tropics. Coastal regions in Norway experience less number of ARs, compared to US and UK, because they are further from the equator. On the other hand, the high storm activity over North Atlantic can provide the conditions to increase the number of ARs. The figure (3) shows the AR system associated with the 2005 flood in western Norway. The red colour implies where the water vapour is concentrated. The narrow in-red region stretched south-west to north-eastward indicates the position of the AR.

A schematic configuration of weather systems observed during AR events hitting Norway.

It seems that the role of a large scale high pressure system over Europe is also crucial for the formation and persistence of ARs that reach Norway. The figure (4) illustrates a configuration in which the moist tropical air (green arrow) is forced toward the higher latitudes. The clockwise (red) and anticlockwise (blue) circulations induced by high and low surface pressures, respectively, push the moist air north-eastward at lower latitudes over North Atlantic. Such a pattern is observed in almost all ARs that hit the western Europe.

One of the major concerns related to the global warming is whether powerful weather systems which cause flooding will become more frequent in the 21st century. As it is always the case, the question is whether ARs will be more frequent or more intense in the future. The answer may lie in changes in the storm tracks, the atmospheric river storms, and the capability of atmosphere in holding the moisture.