780 
the windward coast. Orographic showers or rain may 
thus be superimposed upon the idealized precipita- 
tion areas of the models. Leeward of the mountain 
range, on the other hand, the foehn effect may lead to 
a dissolution of the idealized precipitation and cloud 
areas of the models. Mountain chains in general ob- 
struct the rapid motion of all fronts; the hindrance is 
greater the higher the obstruction and the less the 
vertical extent of the front itself. But a swiftly moving 
cold front will flow over a low obstruction without 
especial deformation. 
Frontolysis, Cyclolysis, and Anticyclolysis. When the 
cold air lies to the right of the general flow, frontolysis 
occurs in the lower portion of the front. A front which 
is traveling normal to the general flow aloft is very 
subject to frontolysis. A front dissipates as it departs 
from a deep pressure trough. 
As a consequence of the convective mixing of the 
two air masses below and above a frontal surface, the 
thermodynamic effect generally will tend to smooth 
and delete any weak fronts rather than to create new 
ones. A mediwm- or high-speed cold front (mostly found 
in the interior of an already existing cyclone) will 
generally be a katafront in the upper layers (see Fig. 2). 
It will then be subject to frontolysis, tending to make 
the frontal surface diffuse and thus dynamically in- 
effective (even if the effect of friction upon the motion 
tends to increase its slope and thereby presumably 
the horizontal shear). A mediwm- or high-speed warm 
front is generally an anafront, but the friction will 
tend to diminish its slope markedly, at least in the 
lowest kilometer of the atmosphere and thus, accord- 
ing to synoptic experience, also reduce the horizontal 
shear. 
After occlusion, the deepening of the cyclone ceases 
and its fillmg commences, provided that a false warm 
sector does not occur. The existing pressure gradient 
weakens and the cyclone fills more or less rapidly 
when it possesses supergeostrophic, divergent flow 
(Hesselberg [40] and Guilbert [86]). In particular a 
cyclone with marked supergeostrophic wind on its for- 
ward side fills during the following 12-24 hr. With 
weak upper flow around its periphery, a cyclone will 
fill. If an upper low reaches to the surface without 
especial tilting of the axis, the surface low sometimes 
fills. 
A young anticyclone weakens when approached by a 
cold front, particularly with weak upper flow around 
its periphery. The degeneration of the stable anticy- 
clone, unfortunately, can seldom be anticipated in ad- 
vance. This fact is especially regrettable, for seldom 
is a reliable forecast more insistently demanded than 
during the prolonged periods of abnormal weather due 
to the blocking action of the mature anticyclone. The 
anticyclolysis and the movement of such an anticy- 
clone is generally desultory. Sometimes the weather 
regime of the blocking anticyclone appears to be at 
last giving way, as the anticyclone starts to move and 
dissolve, only to be renewed again as it restrengthens 
and meanders back to its original position. In fact, in 
closing this section on surface map prognosis, we men- 
WEATHER FORECASTING 
tion that another weather forecasting deficiency on 
which much future study is needed concerns our lack 
of knowledge about anticyclone development. 
THE PROGNOSIS OF THE UPPER-AIR 
MAPS 
The upper-air prognosis of the pressure field proceeds 
in the same way as the analysis. By the layer method, 
we first prognosticate the partial relative hypsography 
of a mandatory surface, then graphically add it to the 
prognostic absolute hypsography of the adjacent and 
lower mandatory surface. By the layer method the 
forecaster builds upward one prognosticated mandatory 
layer on top of another, beginning with the prognostic 
surface map as the base. This method not only re- 
quires that the prognostic absolute hypsographies of 
all mandatory surfaces for a given time be mutually 
consistent and hydrostatically interrelated, but also 
provides a means of developing the baric prognosis 
upward from the more reliable prognosis of the surface 
map to the less predictable future state aloft. The main 
attention is directed toward prognosticating’ the pres- 
sure field at sea level and the relative hypsographies 
of each mandatory surface. The upper-air prognosis of 
the pressure field is then based solely upon predicting 
locally the rate of change in the partial relative height. 
Upper-air prognosis is thus concerned with extrapolating 
the relative hypsography on the basis of an estimate of 
its future over-all rate of change. 
The Geometrical Extrapolation of the Relative 
Hypsography. Closed centers of both low and high 
values in the relative hypsography, so-called thermal 
lows and thermal highs, respectively, and formerly 
termed ‘“‘cold-air drops” (Kaltlufttropfen) and ‘‘warm- 
air drops,” are often observed to occur in the partial 
and total relative hypsographies of higher level con- 
stant-pressure maps. These thermal lows are important 
conservative features of the constant-pressure maps. The 
same holds for the thermal highs. In prognosticating 
the relative hypsographies of the constant-pressure 
maps, therefore, attention should be given to presery- 
ing the real contimuity m time of these models. Stage 
three, then, in the order of our prognostic operations is 
the geometrical extrapolation of these thermal lows and 
highs. 
A series of actual surface and upper-air maps show- 
ing the typical features of the thermal centers are 
shown in Fig. 6, which is taken from a paper by 
Schwerdtfeger [73]. Following the cold front, the 
thermal low C in the 500/1000-mb hypsography moves 
(southeast) for 244 days with remarkable regularity 
and little change in size and value—536 gpDm to 540 
epDm (geopotential decameters). As a second example 
we may refer to the 24-day movement of the 500/1000- 
mb low in Fig. 8b, the innermost closed 516 gpDm 
isohypse of which remained about the same size dur- 
ing that period. It has been generally observed that 
this low moves not with the velocity of the air in 
which it exists but roughly with the horizontal velocity 
of the cold-front surface underneath it. Although the 
island of cold air in a moving upper low-pressure center 
