AIRCRAFT METEOROLOGICAL INSTRUMENTS 
as compared with the total 17. The second curve in Fig. 
2 gives the collection efficiency of a large area, in this 
case the nose of a B-17 airplane. A collector in the 
SPHERE OF 
| CM. DIAMETER 
SPHERE OF 
134 CM. DIAMETER 
Ip 2 3456 810% 20 3040 6080/00n 200 400600 !MM. 
DROP DIAMETER 
Fie. 2.—Collection efficiencies for cloud and raindrops at 
the stagnation points of two spheres. The calculations were 
made by R. M. Cunningham for a 1-em sphere (showing the 
performance of a small capillary collector) and for a 134-cm 
sphere (indicating the collection efficiency for a collector 
centered at the nose of a B-17 airplane). Pressure was assumed 
to be 800 mb, temperature +-5C, and air speed 200 mph. 
center of the nose would be less than 60 per cent effi- 
cient for drops of less than 100 u in diameter. 
Having removed the water from the air, we must 
measure the rate of collection. The most obvious means 
is to pipe it to a flow meter, but unfilled tubes of any 
length cause lag. The most widely used M meter is 
called a “capillary collector” [50] because it employs 
a capillary porous surface as a collector so as to keep 
its plumbing full of water at all times. The lag of such 
a system is limited only by the method of measuring 
flow rate. Many such methods have been devised, 
none of them entirely satisfactory [12, 52]. 
Evaporation from the collecting surface is a serious 
limitation of the capillary collector and similar types of 
M meters. Evaporation losses are assumed to be neg- 
ligible in dense cloud, but may be appreciable in rain; 
and evaporation makes it nearly impossible to use de- 
icing heat for supercooled water particles. Because of 
evaporation in clear air, the plumbing system should 
permit reverse flow or correct for the loss of water in 
some way. 
Another method is to collect both water and air by 
means of a scoop and determine the amount of water 
by measuring the amount of heat required to evaporate 
it, or by measuring the dew points of the air before and 
after all the water has been evaporated. The latter 
procedure is inaccurate because it determines a small 
quantity as a difference between two large quantities 
[27]. Other types using heat evaporate the water as it 
mmpinges on the collecting surface. One can either 
supply heating power at a constant rate and measure 
the temperature, or maintain a constant temperature 
and measure the heating power. Another variation is a 
heated surface whose electrical conductivity is a func- 
tion of its wetness. For all of these, the major problems 
are constancy of calibration and dependence on factors 
other than J, for example, air temperature, water- 
drop temperature, and humidity. The heated types, 
however, can be designed for use under icing conditions. 
1227 
An icing-rate meter is a useful instrument for meas- 
uring M under icing conditions in supercooled clouds 
[32]. Such meters are not readily available, but several 
types have been built, the most successful being some 
variation of the rotating-dise type [52], in which ice 
accretion is continuously measured on the edge of a 
disc rotated at a uniform rate. This type, however, is 
usually built with a thin dise which has good collec- 
tion efficiency for cloud drops but does not measure 
icing rate in rain due to splash and “‘run-back”’ effects. 
Another method is to expose a rotating cylinder or 
series of rotating cylinders of different diameters to 
the air stream and weigh the ice accumulation after a 
known time at a known air speed [89, 52]. This process 
also gives a measure of the drop-size distribution but is 
awkward and discontinuous. 
For reconnaissance aircraft a simple recording in- 
strument must be chosen. The N.A.C.A. reports suc- 
cess with one of the heated types just mentioned [32]. 
A small heated cylinder is exposed at right angles to 
the air stream with a thermocouple measuring and 
recording its surface temperature at the stagnation 
point. A sharp drop m recorded temperature indicates 
entry into a cloud, and subsequent variations give a 
rough measure of M7. The instrument reacts less and 
somewhat differently to snow, so that snow can be 
differentiated from rain under some conditions. By 
using two such heated cylinders, one centered at the 
nose of the aircraft and one exposed so as to collect 
small drops efficiently, the records should give a rough 
idea of drop size as well as of J. 
Electromagnetic-radiation devices may prove to be 
useful instruments for measuring liquid-water content 
if the drop-size distribution is known [22]. A light- 
transmission device working in the visible spectrum 
(a type of visibility meter) seems the simplest type. 
Assuming a uniform drop diameter d, WM is propor- 
tional to d log. Ip/I, where Ip/I is the ratio of trans- 
mitted light intensity in clear air to that in the cloud. 
Flight models of such instruments have been successful 
[28, 35]. An instrument operating on back-scattered 
light may also be possible [33]. Microwave radar may 
be used as well as light, since the intensity of back- 
scattered radiation at those frequencies is proportional 
to Md? and since radar detection of clouds is possible 
over short ranges.2 The strong dependence on drop 
size is unfortunate for this application. 
Radar, of course, is a very complex instrument; but 
if it is to be carried on an aircraft for other purposes, 
it should be considered for J measurement. It would 
have the advantage of indicating the value of the 
quantity Md? over any selected neighboring region 
instead of only at the aircraft. 
In summary, it is clear that rather complex and spe- 
cialized instruments are necessary for accurate meas- 
urement of liquid-water content, but simple and useful 
instruments such as the N.A.C.A. “cloud indicator” 
[32] have been devised which can dependably record 
2. See ‘‘Radar Storm Observation” by M. G. H. Ligda, pp. 
1265-1282 in this Compendium, 
