THE RELATION OF SEA ICE FORECAST AND 
AERIAL PHOTOGRAPHS IN ARCTIC REGIONS 
by Geza Teleki 
Professor of Geology at the George Washington 
University, Researcher for the Arctic Institute 
of North America. 
ABSTRACT 
Sea ice forecast is a predicting of ice conditions 
for ships passing arctic waters. Its effectiveness de- 
pends on the quality of the quasi-synoptic data of vis- 
ual observation techniques which are mostly subjective. 
Aerial photographs systematically taken during the arc- 
tic season along the same flight routes and at equal 
time intervals can help to cut down the subjectivity of 
visual observation. This is true mainly for area ele- 
ments of ice distribution: concentration, number and 
size of floes, etc. These elements can easily be con- 
verted from photographs into objective data with the 
aid of electronic scanning. But the elements which 
contain relief, age, thickness or drifting of ice floes 
require the personal evaluation of à photo-interpreter. 
Color photographs are of & very limited value. 
Aerial photographs, in general, are of great value 
in the coordination of data made during ice recon- 
naissance, trips of the ice observers, as well as in 
the training of ice observers and in the studies con- 
cerning long range ice forecasts. 
Meereis-Prognose ist ein Voraussagen von Eiszustan- 
den fur die arktischen Gewasser durchkreuzenden Schiffe. 
Die Gute einer solchen Prognose steht oder fallt mit der 
Qualitat der quasi-synoptischen Daten von meist subjek- 
tiven visuellen Aufnanmemethoden. 
Die wahrend der arktischen Saison entlang gewisser 
Fluglinien und zu gleichen Zeitinterwallen systematisch 
genommenen Flugaufnahmen konnen die Subjektivitat der 
Eisbeobachter einschranken, und zwar im Falle wo eine 
Messung der Flachenelemente ermoglicht ist: Konzent- 
ration, Zahl und Flachengrosse der Eisschollen, etc, 
Solche Elemente kann man verhaltnismassig schnell 
durch elektronisches Abtasten in nutzbare Angaben uber- 
setzen. Was jedoch die Eiselemente mit Relief, Alter, 
Machtigkeit oder Bewegungen der Eisschollen angeht 
scheint es unmoglich ohne einen guten Eisfotografie- 
Interpretator auszukommen. Auch Farbenaufnahmen sind 
von sehr beschrankter Hilfe, 
Im allgemeinen macht das Studium von Flugaufnahmen 
die koordinierung der Berichte verschiedener Eisbeo- 
bachter moglich, ist hilflich in der Erziehung von 
Eisbeobachtern und kahn die auf langere Dauer strebende 
Prognosen von Eiszustanden unterstutzen. 
* * * 
Navigation in arctic water8 is concerned with the 
task of transiting sea ice covered areas. To forecast 
the conditions of the "ice-crust" which floats, freezes 
&nd melts, shifts and drifts constantly it is necessary 
to evaluate and to organize the synoptic oceanographic, 
meteorological, geophysical and glaciological data into 
suitable directives for the transiting of icebreakers, 
single ships and convoys. Ice forecast is mainly con- 
cerned with constructing and transmitting ice charts 
which include the possible changes in ice conditions 
in the near or far future. 
  
Note: The studies on the application of aerial photo- 
graphs to sea ice forecast were aided by a contract be- 
tween the Office of Naval Research, Department of the 
Navy, and the Arctic Institute of North America. The 
opinions expressed in this article are those of the 
author, and do not necessarily reflect those of the 
U.S. Navy or the Arctic Institute of North America. 
Charting is based on observation and recording of 
ice-patterns in their successive stages during a given 
time interval: a few days, a month, a year. Until re- 
cently the quickest charting in arctic regions has been 
aerial visual reconnaissance and reporting. On the 
quasi-synoptic data of visual techniques depends the 
ice forecasting. 
The most important features reported by aerial ice 
observers are the concentration or density of ice, the 
concentration by size of floes, age or kind of ice, 
thickness, relief and topographic features, and the 
main boundaries of ice patterns or of leads, polynyas, 
and crack patterns. 
Skill and experience of the observer enable him to 
recognize patterns and features from various altitudes 
and sighting angles. The accuracy of the data reported, 
Will therefore depend on the subjective response of the 
observer to the combined factors. The lower the angle 
of vision or the higher the altitude from which obser- 
vation is carried out, the higher the estimated concen- 
tration becomes. With increasing flight altitude, the 
&ccuracy of judging ice floe sizes decreases; lower 
flight altitudes, on the other hand, prevent an accurate 
estimate of flow sizes and patterns in oblique viewing 
towards the horizon. Estimates of sizes of floes which 
are greater than one mile are extremely difficult. In 
general, the average difference of estimates of two ob- 
Servers is 30 per cent, but in some cases as much as 
100 per cent. 
The few facts listed in the preceeding paragraph ex- 
plain why the reconnaissance data used in sea ice fore- 
casting procedures are more or less inaccurate and the 
error margin is far greater than in the acquired meteor- 
ological and oceanographic data. To remedy this situa- 
tion, i.e. to at least narrow the error margin of visual 
"ice charting" instrumental methods have to be introduced. 
One of these technical means is the use of aerial photo- 
graphs for sea ice forecasting. However, the photographic 
method is not without limitations and it cannot solve all 
existing problems. For example, low hanging clouds and 
dense haze will obstruct the taking of photographic ex- 
posures. Since the incidence of cloud cover is usually 
high during operating seasons in arctic regions the 
gtatistical material illustrating seasonal changes of 
ice patterns over the same area is extremely difficult 
to evaluate. Photographs of pack ice or fast ice bound- 
aries cannot be taken except in good weather.  Neverthe- 
less these boundaries as well as polynyas, shore leads 
and flow leads can be received with fair accuracy by 
radar in spite of the undercast. 
What is then the importance of aerial photographic 
methods? How can aerial photographic coverage of cer- 
tein areas help ice forecasting and navigation? And - 
this is one of the most important questions - can it 
narrow the margin of error of the human, subjective 
visual observation and speed up control and evaluation 
of ice reconnaissance data? 
To answer these questions we have first to realize 
that a photograph is a black-and-white or color picture 
image of an area and its ice patterns as seen by the 
camera in the fraction of a second. Therefore, it is 
static. Movements will not show up unless we return 
several times over the same spot (which requires the 
difficult task of determining exact location in polar 
regions) and use a series of photographs which record 
the changes in patterns and drift. To do this for 
  
  
  
* : “Archives 6 
* 
3 = 
Now 
  
ua— © ee EEE