A PROCEDURE OF SHORT-RANGE WEATHER FORECASTING 
in the confluence region, negative in the delta region, 
as shown in Fig. 3b. 
In a study of the relation between sea-level pressure 
change and the upper-air density change, Hess [39] 
has found that intensifying cyclones are associated 
with upper-air local density decreases, primarily in 
the stratosphere. Corroborating Hess’s observations, 
Vederman [80] particularizes that the partial relative 
height of the upper one-third of the atmosphere (by 
weight) increases markedly over rapidly deepening 
lows. Moreover, Austin [5] observed that there is a 
tendency for the stability to increase over deepening 
cyclones in the lower stratospheric layers from 400 
mb to 100 mb. These two relationships reported by 
Hess and Austin both corroborate the previously men- 
tioned fact that the exit (divergence) region ‘above a 
fairly low level of nondivergence determines the sign 
of the pressure tendency at the ground, namely falling 
pressure on the surface map. The entrance region sim- 
ilarly should be associated with rising pressure at the 
ground. 
Upon the relationship between the ageostrophic 
upper flow and the local changes in sea-level pressure, 
Scherhag [72, pp. 23-26] already in 1943 had intro- 
duced some interesting rules for predicting the deepen- 
ing and fillmg of sea-level pressure systems. These 
rules were formulated from his synoptic experience 
which had been guided somewhat by Ryd’s earlier 
(1923) theoretical contributions on the subject. Al- 
though some of his rules still seem plausible in the 
light of Sutcliffe’s more recent investigations [76], a 
good deal of skepticism regarding them now exists. 
Still, for historical and illustrative purposes, we have 
summarized below some of his more popular rules. 
1. In the region where, on the constant-pressure 
maps, the absolute isohypses diverge markedly, such 
as at F in Fig. 3a, there is falling pressure on the surface 
map (cf. at F in Fig. 3b). On the other hand, the con- 
vergence of the upper absolute isohypses (cf. R in 
Fig. 3a) is associated with rising pressure on the surface 
map (cf. R in Fig. 3b). 
2. If the divergence of the upper absolute isohypses 
is vertically above a low center (.e., if the low in Fig. 
3b were displaced rightward to the position of Ff’), then 
this center has a tendency to deepen. A high center 
beneath converging absolute isohypses likewise in- 
creases. 
3. Cyclogenesis on the surface map usually occurs 
in those regions where, on an upper constant-pressure 
map, the absolute isohypses diverge. Anticyclogenesis 
is similarly associated with converging absolute iso- 
hypses. On the surface map, frontal waves which move 
toward the exit (entrance) region will intensify 
(weaken). 
4. If (as in Fig. 3) the frontal strip, the jet stream 
above the frontal strip, and the entrance and exit 
regions of the jet stream all have the same orienta- 
tion, then the sea-level cyclones will not deepen. Many 
stable waves on such a front are likely, but no waves 
are likely when the upper flow is mainly perpendicular 
to the front. 
775 
5. If, as indicated by the increased local crowding 
of the relative isohypses into the frontal strip (e.g., 
the sharpening of the frontal zone by the approach of a 
new front), a front becomes more sharply defined, the 
jet stream is intensified, and so are the convergence 
and divergence in its exit and entrance regions. (As 
will be shown in the later subsection on wind (pp. 
791), an increased local crowding of the absolute iso- 
hypses tends to produce large positive or negative 
angles of geostrophic deviation in the exit and entrance 
regions, respectively.) Then the fall® in the exit region 
and the rise in the entrance region will both intensify. 
6. If, as in Fig. 3a, the warm-air advection in the 
exit region is less than that in the entrance region, 
the absolute isohypses in the upper jet stream then, 
according to Scherhag, diverge more in the exit region 
of the jet stream than they converge in the entrance 
region. As a result the wave cyclone LZ in Fig. 3 in- 
tensifies and deepens rapidly. Similarly, if the warm- 
air advection appears to be greater in the exit region 
than in the entrance region (strong warm front, weak 
cold front), the absolute isohypses in the upper stream 
then converge more in the entrance region than they 
diverge in the exit region. As a result the cyclone will 
weaken and fill. 
With warm-air advection in the entrance region, 
the spacing of the relative isohypses decreases and 
their cyclonic curvature increases. With cold-air ad- 
vection in the exit region, the spacing and anticyclonic 
curvature of the relative isohypses increases. Where 
there is a sharp reversal of the upper flow, the fall is 
usually weakened, the rises strengthened. 
7. Deepening occurs in a low where there is cyclonic 
shear in the cyclonic flow aloft (Fig. 4a). A surface 
frontal wave does not develop where there is anticy- 
clonic shear in the cyclonic flow aloft (Fig. 4b). Surface 
anticyclogenesis tends to occur where there is anticy- 
clonic shear in the anticyclonic flow aloft (Fig. 4c). 
Highs do not develop where there is cyclonic shear in 
the anticyclonic flow aloft (Fig. 4d). If such a type of 
upper flow moves over a high, or if it forms there, the 
high will dissipate. 
The foregoing rules may be modified by several 
compensating processes. 
1. As will be shown in the later subsection on winds 
(see p, 791), the conditions for a relatively small ageo- 
strophic flow (leading to small amounts of divergence) 
are (1) that the horizontal shear of the geostrophic 
wind be cyclonic or only slightly anticyclonic, and (2) 
that, if the horizontal shear is anticyclonic, the curva- 
ture of the path be cyclonic or only slightly anticy- 
clonic. A pronounced cyclonic curvature of the ab- 
solute isohypses at a stationary trough must therefore 
be associated with their marked divergence. Long and 
narrow V-shaped troughs in the upper absolute hypsog- 
raphies, such as in Fig. 5, should be quasi-stationary. 
6. The expression fall refers to the center of maximum 
decrease in sea-level pressure. If, as in this case, the length 
of the period is not indicated, it is assumed that the change 
under consideration is valid for any period from three to 
twenty-four hours. 
