Jökull


Jökull - 01.12.1966, Blaðsíða 23

Jökull - 01.12.1966, Blaðsíða 23
convection is the prevailing agent of heat trans- port can be stable against free convection. The transition between the two regimes is probably not sharp, and there is a certain overlap. In a stratified fluid a measure of the ratio between the buoyant and the dynamic forces is given by the Richardson number r. _ g((d0/dz)-|3) _ _ «gy o(dú/dz)2 (dú/dz)2 ’ where z is the vertical coordinate, o the density, fi the density lapse rate at adiabatic conditions and ú(z) the mean horizontal velocity. The second expression for Ri is derived on the basis of the assumption that the density de- pends only on the temperature. In the case of sea water this would imply constant salinity. Since the bottom waters of the deep oceans are locally homogeneous, the following considera- tions will be based on this assumption The Richardson number may possibly provide the basis for a criterion for the transition from fully forced to dominantly free convection. It is well known that both regimes of forced and free convection of heat can be observed in the lowest layers of the atmosphere where the winds are generally turbulent. At relatively high winds, forced convection conditions may prevail in spite of superadiabatic temperature lapse rates. The enhanced stability is due to the eddy transport of momentum and heat. On the other hand, high lapse rates and strong buoyant forces are built up on clear hot days with moderate or low winds. The lower layers of the atmosphere become unstable with the onset of free convection as the result. This situation is especially conspicuous in desert areas where very high ground temperatures are reached during the day. Large bubbles of hot air break away from the ground and penetrate up to great heights. These bubbles form the well known thermal winds, or thermals as they are generally referred to. The space near the ground, vacated by the thermal, becomes oc- cupied by cooler air coming from above, result- ing in a rather sudden local temperature drop at the ground. Conditions of free convection in the lower atmosphere are theref'ore characteriz- ed by relatively large temperature fluctuations near the ground. Priestley (1955) has investigated the transi- tion between the regimes of fully forced and dominantly free convection in the atmosphere near the ground. Normally, there will be a regime of forced convection in a boundary layer above the ground, whereas free convection may prevail at higher levels. Using the Richardson number at an elevation of 1.5 m, he finds that at this level the transition occurs when 0.02 < — Ri < 0.05. The same critical number will probably not apply to other levels. (3) DIFFUSION TYPE MODELS In modern meteorology (see Sutton, 1953) eddy transport processes are frequently describ- ed and studied on the basis of diffusion type models. In the present case of a horizontal layer of fluid this leads to the following model. Let the coordinate system be placed at the bottom of the layer with the z-axis pointing upward. Moreover, let the fluid flow in the direction of the x-axis and the velocity u(z) be indepen- dent of x and y. If @(P) is the potential tem- perature at a point P, the diffusion model leads to the following equation 3 / 30 \ 3 ( 30 \ (3) cv" V‘v 0y / +'?z Y'* (7 ) where ax, ay and az are the coefficients of eddy diffusion which are functions of space and time. Molecular conduction and sources within the fluid are being neglected. Initial and boundary conditions have to be prescribed in accordance with the given situation. For the present purpose where only order-of- magnitude considerations are involved, equa- tion (3) can be simplified considerably. It should be possible to obtain adequate approx- imations by assuming that the velocitv u and the eddy coefficients are constants and equal to average values. Moreover, in the case of ex- tensive horizontal layers, the vertical diffusion predominates and the horizontal component can be neglected. These simplifications lead to the following equation 30 30 S20 + u -— = a, ——, (4) JOKULL 177
Blaðsíða 1
Blaðsíða 2
Blaðsíða 3
Blaðsíða 4
Blaðsíða 5
Blaðsíða 6
Blaðsíða 7
Blaðsíða 8
Blaðsíða 9
Blaðsíða 10
Blaðsíða 11
Blaðsíða 12
Blaðsíða 13
Blaðsíða 14
Blaðsíða 15
Blaðsíða 16
Blaðsíða 17
Blaðsíða 18
Blaðsíða 19
Blaðsíða 20
Blaðsíða 21
Blaðsíða 22
Blaðsíða 23
Blaðsíða 24
Blaðsíða 25
Blaðsíða 26
Blaðsíða 27
Blaðsíða 28
Blaðsíða 29
Blaðsíða 30
Blaðsíða 31
Blaðsíða 32
Blaðsíða 33
Blaðsíða 34
Blaðsíða 35
Blaðsíða 36
Blaðsíða 37
Blaðsíða 38
Blaðsíða 39
Blaðsíða 40
Blaðsíða 41
Blaðsíða 42
Blaðsíða 43
Blaðsíða 44
Blaðsíða 45
Blaðsíða 46
Blaðsíða 47
Blaðsíða 48
Blaðsíða 49
Blaðsíða 50
Blaðsíða 51
Blaðsíða 52
Blaðsíða 53
Blaðsíða 54
Blaðsíða 55
Blaðsíða 56
Blaðsíða 57
Blaðsíða 58
Blaðsíða 59
Blaðsíða 60
Blaðsíða 61
Blaðsíða 62
Blaðsíða 63
Blaðsíða 64
Blaðsíða 65
Blaðsíða 66
Blaðsíða 67
Blaðsíða 68
Blaðsíða 69
Blaðsíða 70
Blaðsíða 71
Blaðsíða 72
Blaðsíða 73
Blaðsíða 74
Blaðsíða 75
Blaðsíða 76
Blaðsíða 77
Blaðsíða 78
Blaðsíða 79
Blaðsíða 80
Blaðsíða 81
Blaðsíða 82
Blaðsíða 83
Blaðsíða 84

x

Jökull

Beinir tenglar

Ef þú vilt tengja á þennan titil, vinsamlegast notaðu þessa tengla:

Tengja á þennan titil: Jökull
https://timarit.is/publication/1155

Tengja á þetta tölublað:

Tengja á þessa síðu:

Tengja á þessa grein:

Vinsamlegast ekki tengja beint á myndir eða PDF skjöl á Tímarit.is þar sem slíkar slóðir geta breyst án fyrirvara. Notið slóðirnar hér fyrir ofan til að tengja á vefinn.