Search

A gusty downslope windstorm blowing at the eastern Adriatic coast is named bora. Related hourly mean wind speeds surpassing 20 m/s, with gusts reaching up to 50 or even 70 m/s, in the coastal mountain lee areas are typical (hurricane speeds). Substantial prospects in bora observations and measurements, understanding, modeling and its more detailed prediction during the last ~25 years will be discussed. It was generally thought before that bora was a falling, mostly thermodynamically driven wind ; however, bora is governed by mountain-wave breaking. The overall prospect mentioned would not be feasible without airborne data, nonlinear theory and advancements in computational techniques, especially mesoscale numerical models. However, the role of the boundary layer and waves in the upwind side in the bora evolution and the consequent lee-side flow structures are not understood adequately ; similarly, dynamical switches between weak-to- moderate to strong-to-severe bora flows and vice versa are understood inadequately.

There are several modifiers of the bora wind at the eastern coast of the Adriatic, but common to the majority of situations with bora are the cross-coast pressure gradient and cold advection on the continent. In this paper a detailed analysis of numerical model simulations of two cases with bora is presented in terms of the two aforementioned background state parameters, and additionally on the warm advection into the Adriatic basin. The results indicate that the speed of the bora wind offshore does not depend on the origin of the air within the jet. It is found that warm air from aloft can enter the accelerating streamline at the mountain ridge and continue as bora jet offshore. The pressure gradient across the mountain barrier is found to be an excellent indicator of the bora wind speed with correlation of up to 0.98.

Quasi-periodic pulsating behavior of wind gusts is a phenomenon regarded as inherent to downslope windstorms. These pulsations have periods of the order of 1 - 10 min and have so far been observed in the Boulder downslope windstorm, downslope windstorms on Iceland and in the bora wind. The bora wind blows off the east Adriatic coast and has been shown to greatly influence the dynamics of the Adriatic Sea. As the most prominent feature of the bora wind is its gustiness, which is responsible for the appearance of even hurricane-strength wind speeds, it is expected that the effect of gusts on the sea is also significant. This study deals with the pulsating behavior of the bora wind gusts. High-resolution numerical simulation of a long bora episode is presented and compared with measurements, with the goal to resolve the quasi-periodic pulsations. The model reproduced well the approximately 7 min periodicity of pulsations and the upstream structure of the atmosphere, and is thus used for detailed dynamical considerations. The results of previous studies that reported on the possibility of disappearance of pulsations within a bora episode are confirmed. Specifically, it is shown that the positive tropospheric wind shear related to the upper-level cross-mountain jet diminishes the low-level wave breaking and so the pulsations cease. Additionally, it seems that the situations with absence of pulsations may be related to the formation of the mountain-wave-induced rotor. The origin of the pulsations seems to be related to the Kelvin-Helmholtz instability appearing in the region between the low-level bora jet and wave breaking aloft.

The focus is on analytical and numerical modeling of simple katabatic flows blowing over short and long cool slopes. Since it was verified that classical scaling via Monin-Obukhov length performs over such slopes rather poorly (a low-level jet usually determines the surface fluxes), a modified Prandtl model is used instead. This linear analytic 1D model is extended for 1.) almost any gradual eddy diffusivity K(z) via the WKB method, and 2.) it includes the Coriolis effect. Its asymptotic (WKB) solution is checked against its more complete numerical solution and versus a mesoscale model (MIUU model). The Prandtl model is further used to determine one of the coefficients in the MIUU model turbulence length-scale for stable “ z-less“ conditions, L ; L = min[2a(TKE)ˆ (1/2)/N, a(TKE)ˆ (1/2)/S] where N and S are buoyancy frequency and absolute vertical wind shear and 2a = 0.537 is the standard coefficient in this higher-order turbulence closure scheme. The result can be useful in parameterizing shallow persistent katabatic flows in NWP and climate models, in data interpretation and for improving characteristic length-scales in other numerical models.

Little is known about mesoscale motions in the stable boundary layer, except in special cases of monotonic gravity waves, solitons and density currents. Models are unable to capture the more typical mesoscale structures in the stable boundary layer. Before models can be seriously evaluated, more information on such structures are required. Analysis of twenty data sets reveal that mesoscale motions in the nocturnal boundary layer are spread continuously over a range of scales and do not follow any spectral or local similarity. The strength of the mesoscale motions (here on time scales less than one hour) can vary by an order of magnitude or more between nights at a given site. Transient mesoscale motions tend to be stronger in complex terrain compared to flat terrain. One exception is that transient mesoscale motions tend to be weaker with drainage flows. The large variation of the strength of the mesoscale motions between nights is largely unpredictable with present understanding, although large vertical shear of wind direction tends to correspond with greater temporal variation of the wind direction. The change of wind direction with time for weak winds is more often a sequence of abrupt events rather than gradual meandering of the wind direction back and forth. The frequency distribution of the one-minute wind direction for the one-hour records with weak winds is often bi-modal or multi-modal. Models are generally unable to capture such mesoscale motions apparently due to explicit and implicit horizontal diffusion. Simulations using COAMPS reveal the role of explicit horizontal diffusion on the degree of mesoscale motion in the model. Comparisons between the model and field results in complex terrain are presented for different strengths of the horizontal diffusion.

Satellite-tracked drifter data and meteorological wind measurements indicated summer bora event of 22 - 25 June 1995 as interesting from both meteorological and oceanographic point of view. The studied summer situation was controlled by a mesoscale atmospheric disturbance that imparts strong cyclonic vorticity to the wind field, with the orographic effect being of secondary importance. During the bora event currents of 70 cm s-1 were measured by drifter movements tracked along the western Adriatic coast. Drifter trajectories for the selected period were simulated by an oceanographic model forced by the outputs of two mesoscale meteorological models, ALADIN and MM5, and by river discharges. The atmospheric models provided realistic space- and time-varying atmospheric fields (wind, air temperature and humidity) with ca 10 km horizontal resolution by ALADIN and with 9 and 3 km by MM5. Several numerical experiments with different combinations of the three forcing agents – wind stress, river discharges and surface heat fluxes – were performed. Proper formulation of the drag coefficient was crucial for the correct simulations of the drifter trajectories, as its increased values do reproduce well the effects of the sea surface roughness and atmospheric stability to the surface ocean layer with respect to the observed drifter movements. Numerical experiments revealed that strong current along the western Adriatic coast resulted from a combined influence of the wind stress and river discharges. The importance of the baroclinic, buoyancy driven contribution to the current field was particularly pronounced in the periods with weak winds. At the same time, drifter movements in the open sea area in front of the eastern Istrian coast were the result of the wind stress acting solely.

Numerical simulation of a long bora episode is presented and compared with measurements. The goal is to resolve the quasi-periodic oscillations of the bora gusts, i.e. the pulsations. The model reproduced well the approximately 7 min periodicity of pulsations and the upstream structure of the atmosphere, and can thus be used for the detailed dynamical considerations. The results of previous studies are confirmed, as well as the hypotheses on the mechanisms responsible for the disappearance of pulsations. Specifically, it is shown that the upper-tropospheric jet induces strong positive shear throughout the troposphere and consequently the local nonlinearity of the incoming flow weakens. Henceforth, the low-level wave breaking is diminished and so the pulsations cease. From the three previously proposed generating mechanisms of pulsations, the results obtained point to the Kelvin-Helmholtz instability as the primary mechanism in this case. Additionally, it seems that the situations with absence of pulsations may be related to the formation of the mountain-wave-induced rotor.

Potential and real evapotranspiration for stations in the mountain area of Croatia during the period 1981-1990. have been analysed. The mean decade values oj potential evapotranspiration heve been calculated according Eagleman relation. These values heve been used to estimate the decade values of real evapotranspiration according the Palmer method. Estimated potential and real evapotranspiration have been compared to decade values of precipitation amount. It is shown thjat the greatest differences between potential and real evapotranspiration are in the summer.

The mean annual and daily measured wind speed as well as the distribution of wind speed and direction in the mountain area of Croatia have been analysed. The month-by-month relative deviations from the mean speed are shown, too. The WAsP model, a model for calculation of the wind energy potential, has also been applied by wind data for Gospi}; ; and Ogulin stations in the period 1981-1990. These stations are situated in the mountain area of Croatia. The results show that the utilization of energy at the considered stations could not be economically justified.