The heat fluxes at the sea and ice upper boundaries are the sum of long-wave backward radiation, sensible and latent heat fluxes at the sea surface. The momentum, sensible heat and evaporation fluxes are calculated through the air-sea interaction sub-model based on the Monin-Obukhov similarity theory. The bulk transfer coefficients depend on universal functions relevant to the given stability conditions of the atmospheric boundary layer. Inputs for the air-sea interaction sub-model are the air and dew point temperature at 2 m above the sea surface, wind speed at 10 m and the sea surface temperature. The method of iterative flux calculations is based on the approach of Launiainen and Vihma (1990).
Whenever thermal conditions are favourable to form ice, air-sea fluxes are modified to account for the effects of sea-ice, based on the thermodynamic sea-ice submodel of Schrum and Backhaus (1999).
Although more than half of the incoming solar radiation that enters the ocean in the long wave spectral band is absorbed within the top half meter, the remaining short wave fraction, as it penetrates through the surface waters, modifies SST by absorption, which in turn affects the rate of evaporation, leading to an impact on the water balance of the sea. The subsurface profile for solar radiation is computed using the two-band approximation of Paulson and Simpson (1977):
I(z) = Qs[R • exp(-z/Zi) + (1 - R) • exp(-z/Z2)] (6.3.20)
where Qs is the downward flux of incoming solar radiation; R is an empirical constant; Z1, Z2 are respectively the attenuation lengths for long wave and short wave spectral bands of solar radiation. For a one-dimensional model, Martin (1985) has found the simulations sensitive to the optical properties of the given type of seawater. For enclosed and semi-enclosed seas, Timofeev (1983) adopts a value of R = 0.53 for the empirical constant. Attenuation lengths for long wave radiation are typically small (Z1 = 0.033 m is used, as proposed by Timofeev (1983)), so that total absorption occurs in the first model layer. The attenuation length for short wave band of solar radiation strongly depends on turbidity and differs between coastal and offshore regions. For the Caspian Sea its value is estimated to be about 10-15 m in the central parts of the MCB and SCB, and about 1-5 m in the NCB (Terziev et al., 1992). The short wave attenuation length is parameterized depending on local depth,
Z2 = 15 m, forH > 100 m and Z2 = (15m/100m) H, forH < 100 m, where H is the depth.
The grid resolution of the model is (1/12)° in latitude and (1/9)° in longitude, which gives a grid size of about 9.3 km. There are 22 vertical model levels defined at depths of 1, 3, 7, 11, 15, 19, 25, 35, 50, 75, 100, 125, 150, 200, 250, 300, 400, 500, 600, 700, 800, 900 m.
The maximum depth in the model is 950 m, and a minimum depth of 5 m occurs in the shelf region of the NCB. The bottom topography and coastline correspond to the conditions during 1940-1955, when the mean sea level was 28 m below the global ocean level. The model bottom topography in Fig. 6.17 realistically represents the flat NCB shelf, the steep topographic slopes of the SCB and of the western part of the MCB, as well as a number of islands.
The model is initialized from a state of rest corresponding to the climatologic state of the sea in November (Fig. 6.20) in order to overcome the lack of temperature and
salinity data in winter in regions under ice cover. The data of sea surface temperature and salinity for January is shown in Fig. 6.20.
The values of vertical and lateral mixing coefficients were selected to have the following values: (am0, amb, ah0, ahb) = (50., 1., 10., 0.02) ■ 10-4 m2s-1, a = 1, n = 1, Am = 150 m2s-1, Ah = 0.1m2s-1. The time step of integration was 30 min.
The model was run for four years with perpetual seasonal forcing, to ensure that the basin averaged kinetic energy, temperature and general circulation reach quasi-stationary periodical states.
The model forcing is computed from monthly mean atmospheric surface variables based on the ECMWF ERA15 reanalysis data (wind velocity at 10 m height, air and dew point temperatures at 2 m height, incoming solar radiation and thermal back radiation).
The seasonal dynamics was simulated by applying perpetual yearly forcing corresponding to a selected year. Since drastic sea level changes in the last two decades have resulted from imbalances in the external forcing, we select a year with the lowest net sea level change in the period of interest covered by the ECMWF data. Analyses of hydro-meteorological data from Makhachkala, Fort-Shevchenko, Kras-novodsk and Baku indicate 1982 to be a year with small change in mean sea level, the temperature difference between January and December is +6.75 °C. For testing the validity of ECMWF ERA15 data we compare them with climatologic data and statistics from hydrometeorological atlases of Samoilenko and Sachkova (1963) and the books of Kosarev and Yablonskaya (1994) and Terziev et al. (1992) (hereinafter briefly referred to as SS, KY and TKK).
Monthly mean river runoff data were obtained from routine hydrometeorological observations.
The characteristic air temperature patterns in winter and summer are shown in Fig. 6.21. In winter, the temperature has a meridional gradient, decreasing from about +8 °C in the SCB to -1 °C in the NCB, with a local minimum near the mountainous west. Air temperature in July has a zonal gradient resulting from contrasts between the desert and mountain regions, increasing from about 22 ° C in the northwest to 27 °C in the east.
Throughout the year, vapour pressure is higher in the SCB compared to the other sub-basins and also in the interior of the sea compared to the coastal regions. Maximum vapour pressure occurs in July, reaching values of 27 and 23 mb respectively at the centers of the SCB and the NCB, and decreasing to 11-15 mb along the eastern coast. In February, the vapour pressure decreases from 8 mb at the center of SCB to 1-2 mb in the NCB and in the coastal regions.
Monthly mean air temperature and vapour pressure distributions for 1982 are close to the climatology provided by TKK, except for winter in the NCB, where air temperature from ECMWF is 3-5 °C higher than the values given by the climatology.
The wind speed is typically about 4 m/s during the summer and increases up to 5-6 m/s in winter. Wind speed in winter increases from south to north, exceeding 6.5 m/s in the north (Fig. 6.22a). In summer the maximum wind speed occurs to the east of the Apsheron peninsula. The annual cycle of the monthly mean wind can be divided into three periods: (a) December-January with convergence of winds in the MCB and SCB resulting from the high land-sea temperature contrast in winter, producing local cells of atmospheric circulation with upward motion of the relatively warmer air in the middle of the basin (Fig. 6.22a). (b) February-July when
large-scale anti-cyclonic winds prevail over the Caspian Sea (Fig. 6.22b), with south south-southwest-ward winds and divergence in the SCB. The local atmospheric circulation in summer in the SCB appears to be the opposite of the winter situation, as a result of the reversed land-sea temperature differences, when the land temperature in the surrounding deserts and steppes exceed 30-40 °C, while the sea is relatively cool. (c) August-November, when average wind direction gradually changes from south-, southwest-ward to westward (not shown here).
Substantial agreement is observed between monthly mean winds computed from the ECMWF reanalysis data for 1982 and the climatologic winds provided by SS on the basis of measurements made at ships and 72 coastal meteorological stations. The consistency between the climatologic means of SS from the 1950's and those derived from ECMWF ERA15 data for the 1980's suggest relatively small climatic change in the character of winds during the 30 years.
Was this article helpful?