Across the vast expanse of the Middle East, the climate varies with the terrain, but it is typically hot and dry throughout most of the region. Extreme precipitation events, although rare, are a crucial part of the region’s climate. These sporadic storms recharge the area’s freshwater stocks, supporting agriculture in the otherwise dry environment. But they can also cause deadly and economically devastating floods. Researchers have long sought a better understanding of the factors driving extreme precipitation events in the Middle East to improve our ability to predict major storms. This endeavor has been made both more complicated and more urgent by climate change.
A new study led by researchers from the Max Planck Institute for Chemistry (MPIC) in Mainz, Germany presents a new way to detect extreme precipitation events based on the combination of two key meteorological processes. (Note by the MPIC-editors)
Previous studies have linked heavy-rainfall events to two atmospheric features. Stratospheric potential vorticity (PV) intrusions, which occur when air from the stratosphere invades the upper troposphere, have been shown to precede extreme precipitation events in, for example, the Alps, southern Africa, and northwestern Africa. Atmospheric rivers, in which moisture travels along plume-like tracks, have been linked to rain and flooding in North America and Europe. Atmospheric rivers are often detected using vertically integrated water vapor fluxes (IVT), but few studies have looked at data on both IVT and PV intrusions.
Andries de Vries and coauthors combined both into a single algorithm to detect extreme precipitation in an area covering eastern Egypt, southern Israel, Jordan, and parts of Saudi Arabia—all regions with low annual rainfall that have, nonetheless, been hit repeatedly with devastating rainfall events. The authors used data from a global reanalysis project, Interim-ERA, to identify stratospheric PV intrusions and IVT structures from 1979 to 2015 as well as extreme precipitation events, defined as days with rainfall that exceeds the 97.5th percentile.
First, the team demonstrated the success of the algorithm for four major precipitation events that inflicted severe damage in the Middle East: a 1979 storm that killed 50 people and displaced over 66,000, a 1994 storm that killed 600 people, another in 2005 that killed 29 people, and another in 2009 that caused 161 deaths and $900 million in damages. The authors found that in all four cases, stratospheric PV intrusions and IVT structures were clearly apparent. Next, a climatological analysis showed that PV and IVT features coincided with nearly 90% of all extreme precipitation days over the 37-year study period. Both PV and IVT occurrences were highly associated with the precipitation severity and mirrored seasonal rain trends, frequently appearing in October and tapering off around May. The longer the features persisted, the farther south the PV intrusions extended, and the larger the IVT values were, the more extreme the rainfall was. IVT incursions were a particularly strong predictor of extreme precipitation days.
False alarms—or days in which both IVT and PV intrusions were present but the weather was dry—were not infrequent, but the authors found that in many of those cases, a key ingredient for extreme rainfall was missing: high tropospheric moisture content.
Finally, the authors evaluated the performance of the algorithm against precipitation observations. PV and IVT features contributed to about 40%–70% of the annual rainfall amounts and 50%–90% of the extreme precipitation days. This result provides strong evidence of the importance of PV and IVT in extreme precipitation events in the region.
"Our study combines for the first time two cutting edge research methods for detection of heavy rainfall in a region that received only little scientific attention. Apart from the Middle East, this approach can be applied to regions elsewhere and, moreover, introduces new perspectives for operational weather forecasting and future studies on climatic changes of extreme precipitation events", explains Andries de Vries first author of the paper.
Ultimately, the algorithm allows researchers to identify local-scale extreme rainfall events on the basis of large-scale meteorological features. The tool could help to improve both weather forecasts and warning systems and could help researchers evaluate how extreme precipitation events respond to a changing global climate. (Journal of Geophysical Research: Atmospheres, doi.org/10.1002/2017JD027587, 2018)