The life cycles of potential vorticity cutoffs: climatology, predictability and high impact weather

Abstract

Stratospheric potential vorticity (PV) cutoffs are ubiquitous synoptic- to meso-scale cyclonic vortices at upper-tropospheric levels that occur frequently from subtropical to polar latitudes. PV cutoffs are usually identified as quasi-circular regions of stratospheric air isolated from the main stratospheric reservoir on an isentropic surface. Their genesis is the result of breaking synoptic-scale Rossby waves and their life cycle ends either via decay by diabatic processes or via advection back to the stratospheric reservoir (in this thesis referred to as `reabsorption'). PV cutoffs have been in the focus of dynamical meteorologists for more than 70 years mainly because of two reasons: (i) They frequently result in high impact surface weather, in particular heavy precipitation, and (ii) their decay due to diabatic processes results in the transport of stratospheric air masses into the troposphere and thereby affects the concentration of trace gases (e.g. water vapour, ozone) in the atmosphere. While many case studies of PV cutoffs exist and their climatological frequencies are well known, various climatological aspects of their life cycle and their surface impacts are still poorly understood. Also, a comprehensive global analysis of PV cutoffs is yet missing. However, this not only required in order to better assess their fundamental role for various aspects of the atmospheric circulation and chemistry (e.g. for cyclogenesis, the transport of moisture and aerosols, mixing of stratospheric ozone into the troposphere), but also to quantify their relevance for (high impact) surface weather in current and future climates. Current understanding of the relevance of PV cutoffs for high impact weather implies that their accurate prediction by state-of-the-art operational forecasting systems is essential to avoid socio-economic harm. However, only few case studies have investigated forecast uncertainties related to PV cutoffs. In summary, these studies pointed at three key aspects: (i) the direct effect of uncertainties related to PV cutoffs for surface weather, (ii) the role of upstream processes for uncertainties in the formation of PV cutoffs downstream, such as the extratropical transition of tropical cyclones and warm conveyor belts (WCBs), and (iii) the role of PV cutoff reabsorption for downstream uncertainties in the wave guide. The relevance of these aspects have not yet been addressed systematically and only little is known about the dynamics governing the propagation of uncertainties in these situations.\\ This thesis consist of two parts. The first part addresses various climatological aspects of the life cycle of PV cutoffs as well as their contribution to (extreme) precipitation and the second part focuses on forecast uncertainties related to PV cutoffs. To study the life cycle of PV cutoffs, a novel method is introduced to track PV cutoffs as three-dimensional objects. As this method is based on isentropic trajectories, cross-tropopause mass fluxes can also be quantified. Its application to 39 years of the ECMWF reanalysis dataset leads to the first global climatology of PV cutoffs that is independent of the selection of a single vertical level. In addition to known geographical frequency maxima, large regions over subtropical ocean basins in the summer hemispheres and a circumpolar band around Antarctica are identified. The three-dimensional life cycles of more than 40'000 PV cutoffs are identified and analyzed in detail, including their link to surface cyclones. Remarkable regional differences are found which provide the basis for a new classification of PV cutoff life cycles into three types (Types I, II, and III). Type I PV cutoffs form mostly in summer over subtropical ocean basins as a result of anticyclonic Rossby wave breaking equatorward of the jet stream. They can be relatively stationary and long lived, can grow downward in their vertical extent, and are only rarely connected to surface cyclones. Their life cycle ends mostly with diabatic decay. Type II PV cutoffs form from anticyclonic followed by cyclonic Rossby wave breaking between the polar and the subtropical jets. Their formation often occurs simultaneously with surface cyclogenesis and strong diabatic activity results in rapid decay at lower isentropic levels. At the end of their life cycle, diabatic decay and reabsorption are equally likely. Finally, Type III PV cutoffs form poleward of the jet stream by cyclonic Rossby wave breaking in the storm track regions. Typically, they are associated to surface cyclones that formed a few days earlier and their evolution is often rather adiabatic, i.e. without frequent diabatic decay and vertical displacement. Their life cycle ends most frequently with reabsorption. This dataset of PV cutoffs is used to attribute (extreme) precipitation to PV cutoffs and study the evolution of precipitation along their life cycle. It is found that Type I PV cutoffs often result in only little precipitation while Type II and III PV cutoffs are frequently accompanied by high precipitation volumes. Further, we find that enhanced meridional moisture transport is crucial for large precipitation amounts related to PV cutoffs. The first global quantification of the contribution of PV cutoffs to average and extreme precipitation reveals that PV cutoffs are particularly relevant in semi-arid subtropical regions (so-called Mediterranean climate regions), which are in fact the regions that are expected to experience the strongest changes in the hydrological cycle due to anthropogenic climate change. The second part of the thesis contains two separate studies, which are both based on ECMWF operational ensemble forecasts and analyses. First, a detailed case study shows how forecast uncertainty can propagate from the North Atlantic along the wave guide into the Mediterranean and lead to an uncertain PV cutoff genesis position and, as a result, an uncertain development of an intense Mediterranean tropical-like cyclone (medicane). In particular, it is highlighted, that the uncertainties initially emerge from a North Atlantic jet streak and propagate and amplify into the Mediterranean during Rossby wave breaking. Further, it shows how uncertainties in the PV cutoff genesis position substantially affect the vertical thermal structure of the Mediterranean cyclone. These results contribute to improved understanding of the predictability of PV cutoffs and medicanes. The second study aims to investigate more systematically how PV cutoffs affect forecast uncertainties. It is shown that PV cutoff genesis over the Mediterranean is systematically accompanied by enhanced forecast uncertainty which is largest at genesis time. PV cutoff reabsorption, however, does not show a systematic signal in the region considered and with the method used. Further, a systematic link between strong North Atlantic WCBs and forecast uncertainty in both, the North Atlantic region and the Mediterranean is found. This link is particularly strong for the Mediterranean where the effect of strong North Atlantic WCBs is twofold: First, WCBs introduce (or amplify pre-existing) forecast uncertainties in the North Atlantic wave guide and second, they support the establishment of large-scale conditions that direct the North Atlantic jet stream towards the Mediterranean, which provides a corridor for the propagation of forecast uncertainties. Overall, this thesis provides new perspectives on the life cycle of PV cutoffs. It proposes a new classification of PV cutoffs which stresses the regional variability of their genesis and lysis dynamics, their vertical evolution, and surface impacts. Furthermore, it points out the relevance of PV cutoffs for (extreme) precipitation, especially in Mediterranean climate regions. On the other hand, this work offers new insight into the origin and dynamics of medium-range forecast uncertainty for the North Atlantic and Mediterranean regions in general, and, more specifically, for PV cutoffs. The thesis points out the relevance of strong North Atlantic WCBs for enhanced forecast uncertainty and demonstrates the tight link between forecast uncertainties over the North Atlantic and the Mediterranean under suitable conditions. The results imply that the uncertainties associated to PV cutoffs, e.g. in their position or intensity, often originate from upstream processes and are handed over to PV cutoffs during Rossby wave breaking. PV cutoffs subsequently transfer these uncertainties to (high impact) surface weather. Hence, with this work we underline that PV cutoffs are highly relevant flow features that contribute substantially to the complex dynamics of the extratropical circulation, to high impact weather and its predictability, as well as the hydrology of Mediterranean climate regions.