The median limits (Figure 1 1 ) show a simi- 

 lar, but smaller, seasonal progression to the ex- 

 treme limits. Their southward extent agrees well 

 with the bathymetry of the Grand Banks and the 

 flow of the Labrador Current. 



Comparison with climatologies based on 

 sightings 



Data presented in the IIP annual reports 

 for 1945 through 1974 varies from the analysis 

 described above, as only iceberg sightings were 

 reported. Aircraft flight tracks and sightings were 

 available for 1967 to 1974. Ship reports were not 

 published for the period, but the flight tracks are 

 assumed to cover the areas where icebergs had 

 been reported by ships. From 1945 to 1966, the 

 annual report plots the position of all targets re- 

 gardless of sighting source. Iceberg limits were 

 prepared based on the most recent sightings in 

 the same manner in the previous analysis for two 

 dates, April 15 and May 30. A summary of the 

 available data is given in Table 2. 



Limits based on sighting data alone are 

 less extensive than those based on sightings plus 

 drift (Figures 12, 13). For April 15, the median 

 limit from sighting data is 120 nm north of the 

 median limit from current IIP practices (Figure 3). 

 The extreme limit based on sightings is near the 

 30th percentile limit from the recent data. The 

 southern extent of the median limits for May 30 

 shows a similar 120 nm difference, however the 

 extreme southeastern and eastern limits are com- 

 parable. While 1972 and 1974 were extreme ice 

 seasons (based on the number of icebergs pass- 

 ing south of 48N), the geographic extent of the 

 limits far less extensive than that found for 1975- 

 1995. 



The Mudry (1991) frequency of iceberg 

 sightings compares well with the sightings-only lim- 

 its determined here. The region where icebergs 

 were sighted with 40% to 60% frequency extends 

 south to 43N, 50W (the Tail of the Grand Bank), 

 and icebergs were sighted as far south as 39N, 

 with one outlier at 36-30N, 49W. To the east, the 



median frequency extends to 45W near Flemish 

 Cap. 



Conclusions 



This paper presents a climatology of the 

 extreme distributions of icebergs in the North At- 

 lantic based on the International Ice Patrol Limits 

 of All Known Ice for the period of 1 975-1 995. The 

 limits presented here are one limited representa- 

 tion of the iceberg distribution. They represent the 

 cumulative effects of reconnaissance, reporting, 

 HP's knowledge of the regional oceanic circula- 

 tion, wind data, and the activities of IIP operations 

 personnel. The limits are also only instantaneous 

 presentations for discrete times. For example, an 

 extreme iceberg may be reported on the second 

 of the month, and be deleted due to predicted 

 deterioration or reconnaissance on the 14th, and 

 not be represented in the basic data used here. 



Another realization of the iceberg distribu- 

 tion is the iceberg sighting positions over time, but 

 sighting data is also limited by reconnaissance dis- 

 tribution, frequency and effectiveness. While this 

 climatology seeks to forecast the fate of sighted 

 icebergs, errors in the drift calculations arising from 

 unknowns in ocean circulation are potentially large. 



Substantial differences exist between the 

 1945-1974 limits climatology based on sightings 

 alone, and the more recent data. The limits based 

 on sightings alone are more compact than the 

 modern limits. Changes in the distribution of 

 sightings, and changes in reconnaissance areas, 

 may be investigated through the sighting data 

 base. These may provide insight as to whether 

 these differences are due to changes in the true 

 iceberg distribution or simply to changes in the 

 areas searched. 



61 



