OBSERVATIONAL AND NUMERICAL STUDY OF NORTHERN WIND FLOW THROUGH THE MAIN GAP OF CRETE ISLAND I. Koletsis (1, 2), K. Lagouvardos (1), V. Kotroni (1), and A. Bartzokas (2) (1) National Observatory of Athens - IERSD, Greece (2) University of Ioannina – Laboratory of Meteorology - Department of Physics, Greece European Geosciences Union 11th Plinius Conference on Mediterranean Storms Barcelona, Spain, 7 – 11 September 2009 University of Ioannina Introduction Numerical Study RS KS Investigate the role of the complex topography of Crete Island on the flow modification as well as on the development of a gap flow between the two highest mountains of the island (Lefka Ori and Idi) during a strong northerly flow over Aegean Sea. For that purpose observational and very fine model resolution simulations were used in order to investigate: a. the dynamics as well as the role of the elevated major gap on the airflow modification. b. the modification of the wind and temperature field. Scope RS KS Figure 4. Model results from the 1-km model domain at 0600 UTC 24 August 2007 (6 h forecast) showing (a) the 10-meter wind speed in m s-1 (full barb corresponds to 5 m s-1 ) and (b) the vertical cross section (following the bold line AA′ in Fig. 4a) of the potential temperature (solid black line at 2 K intervals) and of the parallel to the gap axis wind component (dashed line at 2 m s-1). The arrows indicate the position of the gap entrance (RS) and the gap exit (KS). Data Observational data 5-min data from four ground stations located along the gap (Fig. 1b). High-resolution (12.5 km) satellite QuikSCAT data. ECMWF analysis fields at 0.5 deg resolution. Model simulated data Hourly MM5 model outputs at 1-km resolution, using 39 vertical levels. A case with strong Etesian winds (24-25 August 2007) is investigated. The model reproduces adequately the observed wind flow modification and also the wakes south of Crete during the time of the maximum winds over the gap exit area. Offshore of the major gap the wind speed exceeds 15 m s-1 and extends in a region up to 40-50 km with a gradually deceleration as the flow moves farther downstream (Fig. 4a). The flow rapidly descended near the gap exit, creating an area of strong winds (~16 m s-1) at low levels below ~940 hPa. At 850 hPa an area of weak flow (~ 0 m s-1) above the rapidly descending isentropes is evident (Fig. 4b). The vertical structure over the gap exit is reminiscent of hydraulic flow associated with downslope windstorms, suggesting that the relatively low topographic features through the gap have excited a mountain wave (Durran, 1990). The highest simulated temperatures are located just downwind of the mountains as result of the descending warmer air. Moreover, trajectories analyses reveal that air parcels rise up over the terrain before descending rapidly in the southern coasts of the island resulting in an increase of surface temperature and a decrease in relative humidity. Also, the sea level pressure field reveals a pressure difference between the windward (higher values) and the leeward coasts (lower values) of the island, which is a common characteristic in areas of airflow blocked by high mountains (Kotroni et al., 2001; Smith, 1982), while the pressure difference between the entrance and the exit of the gap is calculated on 3 hPa (not shown). Figure 1. (a) Location of the Crete Island, (b) topography and geographical location of the four ground stations : Rethymno (RS), Armeni (AS), Mourne (MS) and Kerame (KS). The location of the elevated gap is also marked by the topography profile of the island. Observational Study Figure 5. Wind speed at 10 meters as observed from the ground station (red solid line) and as simulated by MM5 ( blue solid line) from 0000 UTC 24 August 2007 to 0000 UTC 26 August 2007 (48 h time period simulation) at (a) RS, (b) AS, (c) MS, (d) KS. Figure 2. QuikSCAT satellite wind fields at 10 meters valid at 0420 UTC 24 August 2007 over southern Aegean Sea at a resolution of 12.5 x 12.5 km, closer to the time of the maximum observed wind speed at the gap exit (KS) (at about 0600 UTC 24 August 2007). A full barb corresponds to 5 m s-1 . The qualitative comparison between the observed and the simulated wind speeds show a good agreement (Fig. 5). An overestimation of the model wind speed is also evident. At the gap exit unlike the observations, the simulated gap flow reached its maximum value around the noon hours of 25 August. This discrepancy could be attributed to the model pressure difference which is 1-2 hPa greater than observed (not shown), resulting in increasing the wind speeds. Furthermore, the model cannot capture the inland calm winds during nighttime (Fig. 5b). References Conclusions Main characteristics of general flow an upstream deceleration of the wind speed over the maritime area north of Crete as well as an upstream leftward turn of the wind (Fig. 2) higher values of the mean wind speed were observed at the southern coasts as a result of the wind flow channeling within the gaps (Fig. 2) a weak wind flow at the leeward areas of the highest mountains ranges (wakes) (Fig. 2) Durran, D. R.: Mountain waves and downslope winds, Atmospheric Processes over Complex Terrain, W. Blumen, Ed., Amer. Meteor. Soc., 59-81, 1990. Gaffin, M. D.: Unexpected Warming Induced by Foehn Winds in the Lee of the Smocky Mountains, Wea. Forecasting, 17, 907-915, 2002. Kotroni, V., K. Lagouvardos, and D. Lalas: The effect of the island of Crete on the Etesian winds over the Aegean Sea. Q.J.R. Meteorol. Soc., 127, 1917-1937, 2001. Sharp, J., and C. F. Mass: Columbia Gorge Gap Winds: Their Climatological Influence and Synoptic Evolution, Wea. Forecasting, 19, 970-992, 2004. Smith, R. B.: Synoptic observations and theory of orographically disturbed wind and pressure, J. Atmos. Sci., 39, 60-70, 1982. The observational analyses indicate that the strongest winds are located in the exit area of the gap (KS). This fact implies that the primary cause of the strong winds is not the Venturi effect associated with mass conversation (Sharp and Mass, 2002). Also, during the event a significant pressure difference between the gap entrance and exit was observed . This pressure difference was strongly correlated with the wind speed at KS, while the wind speed and pressure difference changes were in phase. The simulated horizontal analyses reveal that the observed structures related with the general flow around the island were realistically simulated. The comparison between the observed and the simulated time series of wind speed inside the gap indicates that the model was in qualitative agreement with the observations; however, an overestimation of the wind speed was produced due to the overestimated pressure differences. The vertical analyses present structures similar to downslope windstorms, suggesting that the gap topography can excite a mountain wave. Therefore, the gap flow descended abruptly at the gap exit creating the observed strong winds. Trajectories analyses indicate that the abrupt increase/decrease in temperature/relative humidity at the southern coasts of the island as well as at the KS are related to föhn conditions (Gaffin, 2002). Figure 3. Time series of 5-min average (a) mean wind speed at 10 meters, (b) sea level pressure difference between RS and KS as well as the wind speed at KS at 23-26 August 2007. The strongest winds occurred during the daytime, with the maximum intensity observed at the gap exit (KS) while decreasing progressively towards the northern station (RS) (Fig. 3a). The major characteristic of the mean sea level pressure time series was the persistent lower pressure at KS, which was evident throughout the event (not shown). The pressure difference between the entrance (RS) and the exit (KS) of the gap were in phase with the mean wind speed at KS (Fig. 3b). The temperature time series analysis reveals that the daily temperature maxima of RS, MS and AS are 4-5 ⁰C lower than at KS (not shown). Contact Ioannis Koletsis, National Observatory of Athens I. Metaxa & Vas. Pavlou, Lofos Koufou, GR 152 36, P. Pendeli, Athens, Greece. Tel.: +30 210 8109140 - koletsis@meteo.noa.gr