Hawaiian Climatology— LEOPOLD and STIDD 
217 
over the islands with a low center to the N¥, 
N, or NE. Their "anticyclonic” situations are 
clearly the normal or trade wind weather. They 
showed that the frequency of the two types 
varied inversely: the cyclonic was greatest in 
the December -February period while the anti- 
cyclonic was greatest in summer. 
Because winter is usually the rainiest season 
and also the one when cyclonic storms are most 
frequent, Wallen and Yeh had expected to find 
a correlation between a yearly precipitation 
index for Hawaii and the frequency of cyclonic 
situations for the same year. This correlation, 
however, was not revealed by their data. They 
concluded that variations in trade wind rainfall 
are sufficiently important to mask any direct 
relation of total rain to the frequency of cyclonic 
storm types. 
The importance of local convective clouds 
has generally been overlooked in view of the 
obvious importance of trade wind orographic 
rain. Leopold (1948) analyzed the time of day 
of rainfall occurrences on Oahu and Lanai and 
showed that the leeward or drier portions of 
these islands receive, primarily, afternoon rain- 
fall. Stations on coasts exposed to the trade 
winds and those directly in the lee of the moun- 
tains receive nighttime rain predominantly, a 
characteristic of orographic rain in Hawaii 
(Loveridge, 1924; Jones, 1938). The afternoon 
maximum was interpreted as resulting from 
local convective clouds which build up over the 
drier areas with daytime heating of the ground 
surface. 
Another characteristic type of rain in Hawaii 
is that of the "naulu storm” (Leopold, 1948), 
a summertime cumulus congestus or possibly 
cumulonimbus, which may produce short-dura- 
tion, high-intensity rainfalls over leeward or dry 
portions of the islands. The naulu rains result 
from local convergence where the sea breeze 
meets the trade wind. 
PROBLEMS AND TECHNIQUES IN ANALYSIS 
In the analysis of map types, a difficulty 
which occurs in low latitudes is that the historic 
series of weather maps does not give sufficient 
detail on the various low latitude disturbances 
which have been defined in recent years. 
Riehl’s studies in Puerto Rico (1945) can 
be applied with profit to synoptic analysis in the 
Hawaiian area. An important tool used by Riehl 
and others for synoptic analysis in these latitudes 
is a time cross section of winds plotted to the 
maximum observed height. Pressure and tem- 
perature data are also plotted at standard levels. 
Slight changes in winds, pressures, and temper- 
atures will often show the passage at high levels 
of minor troughs which may go undiscovered 
or be poorly defined on the weather maps. 
Using a similar type of chart, Stidd (MS.) 
noticed that in the winter months no rain fell 
in the drier localities of Hawaii as long as the 
trade winds were present in the lowest 2,000 or 
3,000 feet. Each occurrence of rain was pre- 
ceded by an almost total disappearance of the 
trade wind. A positive correlation between 
rainfall and surface pressure tendencies was also 
demonstrated, the rainfall occurring in conjunc- 
tion with rising pressures. Situations in which 
the axis of the nearest high pressure cell was 
passing to the east of the station as it sloped 
toward the equator with height were more pro- 
ductive of rain than those situations in which 
the axis was passing to the west. This is in 
accord with general knowledge of the distribu- 
tion of convergence and divergence about the 
high cell. 
Another aid to synoptic analysis made use of 
the vast number of weather reports from air- 
planes flying Pacific routes (Stidd, 1948). This 
is essentially a graph of time versus distance 
along a given route, the plane reports being 
spotted in their appropriate positions. Points of 
minimum "delta value” (true altitude minus 
pressure altitude) were found to lie on nearly 
straight lines in the summertime. The positions 
of these lines were found to correspond to the 
positions of easterly waves and the slopes of the 
lines were a function of the speed of such 
waves. The full use of all pilots’ reports is a 
time-consuming operation, however, and a thor- 
ough test of this method has not yet been made. 
