Yair Cohen
Hebrew University of Jerusalem, Israel

Gradient wind imbalance at the outflow of Hurricanes


The gradient wind balance is an essential balance in topical meteorology and in the dynamic of Hurricanes. It is only due to this balance, that the otherwise ineffectively-large deformation radius in the tropics can be locally reduced to allow balance flow in a scale of a Hurricane. To zeros-order the tangential wind in a Hurricane is considered to be closely approximated by the gradient balance. Therefore the basic understanding of Hurricanes was sought in axi-symmetric models, in which this balanced is imposed (e.g. Charney and Eliassen, 1964; Ooyama, 1969; Emanuel, 1986). However, in an axi-symmetric Hurricane contours of geopotential height and of temperature are parallel and thus at the top-center of a hurricane a High pressure center may develop. Around a High with a small enough length scale, the sum of the outwards-pointing centrifugal and pressure-gradient forces cannot be balanced by the inward-pointing Coriolis force. In this case the gradient wind balance has no physical solutions (gradient imbalance) and the flow becomes highly divergent. In this work, some fifty three Hurricanes during 2004-2015 in the E-Pacific and W-Atlantic basins are tracked in a 12km-resolution North American Meso-scale model (NAM12). This data set is used in order to examine the imbalance at the top of Hurricanes and the associated divergence mechanism. We find that the storms in the data are almost equally divided into two groups: 1. Small, organized (i.e. equivalent barotropic), storms with gradient imbalance at the outflow and 2. Large, less organized, storms with gradient balance that holds anywhere in the free atmosphere. An index that quantifies the organization of a storm is defined. Using this index it is found that in the smaller storms the intensity of the imbalance is well correlated with outflow intensity, minimum 850mb geopotential height and maximum wind speed. In the group of large storms similar correlations are very poor. These findings motivate a relaxation of the assumption of gradient balance in idealized theories.

Time and place
Tuesday 10 May 2016, 11.15
Room C609, Arrhenius Laboratory, 6th floor