Ehud Strobach
Eddies in the Mediterranean Sea
The ocean regulates the long-term climate conditions due to its long-term memory. The mass and heat capacity of this vast body of water are such that the current ocean state continues to impact the climate system for years. Therefore, our ability to predict the atmosphere’s future evolution at time scales longer than weather forecasts (>10days) highly depends on understanding processes controlling ocean circulation. It is well established that surface winds drive large-scale ocean circulation and that the atmosphere responds to large-scale ocean circulation through the sea surface temperature. However, the sea surface is organized in small-scale features known as mesoscale eddies and submesoscale current. The shapes of these features range from coherent vortices to strong currents and filaments. Like the global ocean, the Mediterranean Sea is composed of small-scale features. Evidence from recent years suggests that small-scale ocean features affect the atmospheric circulation in different regions worldwide. I study how small-scale sea features affect the climate in the Mediterranean Sea.
Air-Sea interactions
The air-sea interface is the medium separating two important components of the climate system. This interface has not received due attention in model-based studies. Partially, this has been because of the sophisticated coupled model systems required, optimally, ones with a mesoscale resolving ocean component (<25 km). The misrepresentation of air-sea interactions in general circulation models has negative impacts on surface fluxes for the major uncoupled atmospheric and oceanic reanalyses and state estimates.
Ocean fronts
The temperature at the ocean surface can vary by more than 10 degrees C in twenty kilometers. These fronts can affect the large-scale atmospheric circulation by diverting the propagation direction of atmospheric storms and increase moisture supply to storms.
Climate modeling
Climate models are the main tool used to simulate future climate conditions. However, the predictions made by these models are still subject to large uncertainties, and there is still an ongoing debate about what constitutes the best information that can be derived from these models and how to gain it.
surface temperature uncertainty
Machine learning
​
Learning algorithms can be used for various purposes in climate research. These include weighting climate models based on their past performances and classifying large-scale circulation patterns.
