IIP ICEBERG COUNTS
International Ice Patrol’s Iceberg Counts
Donald L. Murphy
Each year, the International Ice Patrol (IIP) estimates the number of icebergs that pass south of 48° N, the latitude south of which icebergs are considered a menace to North Atlantic mariners. The dataset (Table 1) extends from 1900, 12 years before the sinking of RMS Titanic
, to the present.
For several reasons, these iceberg counts do not constitute a rigorous, scientific data set and should be interpreted with great care. For example, IIP’s reconnaissance operations focus on the icebergs closest to the transatlantic shipping routes, and rarely does IIP conduct a comprehensive survey of the area south of 48° N. In addition, the methods of observation have changed radically over the years as new technologies became available to detect and track icebergs. The earliest data were obtained from visual observations from early 1900s sailing vessels, while the recent information is obtained from visual and radar observations from modern ships, aircraft, and satellites.
Origins of the Data
The origin of Table 1 can be traced to the 1926 IIP Annual Report (IIP, 1926) in which monthly iceberg counts were presented for 1900 through 1926. The early part of the iceberg history was reconstructed with data gleaned from mariner’s reports to the United States Hydrographic Office, mostly those printed in the Hydrographic Bulletin, and IIP records. In all the years that followed, IIP has presented monthly iceberg counts from its own records and published them in its annual reports.
Changes in the Definition of the Ice Year
Table 1 reflects the current definition of the ice year, which extends from October through the following September. The yearly total is simply the summation of the monthly values. For each line, the year is based on the year of the January that appears in the line.
At the time of the 1926 reconstruction, the ice year was the same as the calendar year. IIP maintained this convention until 1967 when its definition was changed to September to August. There were two reasons for this change. First, it recognized the seasonal nature of IIP’s operations. Icebergs usually stopped arriving on the Grand Bank in late summer and didn’t resume until winter, possibly as early as November or December. Thus, the late summer is a natural break in the movement of icebergs onto the Grand Banks. Second, and on a more pragmatic level, IIP did not have wait until the end of the calendar year to prepare its annual report.
The second change followed the 1982 iceberg season, when the ice year was redefined as October through September. This small change was undertaken for two reasons. First, it reduced the likelihood that an ice season would carry over into the following ice year. Second, it fell into line with the U. S. Government's fiscal year. The first reason probably carried more weight; since the 1982 season ended very late (1 September 1982), and IIP was concerned about the carry-over issue.
Each time the definition of the ice year was changed, the yearly totals were recomputed, so the new values would be internally consistent. These computations caused some small inconsistencies with the original yearly totals presented in the various IIP annual reports. For example, in 1940 and 1941 the ice year was the calendar year. In both years it was reported in IIP’s annual reports (IIP, 1940 and IIP, 1941) that two icebergs passed south of 48° N during the year. One of these icebergs passed south of 48° N in November 1940 and was originally counted as a 1940 observation. It is now counted as a 1941 observation. Thus, in Table 1 there is one iceberg listed for 1940, and three in 1941.
Observations versus Estimates
The earliest counts were simply a total number of icebergs observed south of 48° N. The icebergs were seen by vessels traversing the northwest Atlantic and reported to the U. S. Hydrographic Service or, after 1913, to the Ice Patrol vessel. The details of the counting process are not known, but it is likely that efforts were made to avoid counting duplicate observations. This task is more challenging than it seems. An iceberg’s appearance can change dramatically from day to day and the complex ocean currents make it difficult to predict the movement of an iceberg accurately, even for short periods. As a result, IIP was careful to refer to the monthly iceberg counts as estimates (IIP, 1927).
Beginning in 1932, IIP conducted routine oceanographic surveys to determine the ocean currents near the Grand Banks. The resulting current maps were used by the patrol vessels to guide their iceberg searches and help determine whether a reported iceberg had been seen before. This may have improved the ability to recognize whether an iceberg report was for a new detection or a re-sighting of a previously reported iceberg, but the process was far from precise.
Anderson (1993) describes the details of the iceberg counting process from 1960 through 1991. For most of the period, 1960-1988, the estimates were determined by hand counting the paper records of the iceberg reports and the model output. The model output sometimes included icebergs that were seen north of 48° N but were estimated by the model to have drifted south of that latitude, without being seen again. It is likely that this became more common when the computerized version of the iceberg model made it possible to predict the movements of a large number of icebergs.
In 1989, the iceberg-counting process was automated, and has changed little since. The monthly estimates of the number of icebergs passing south of 48° N are determined using the output of IIP’s iceberg drift model. The counts include icebergs seen south of 48° N as well as icebergs observed north of 48° N but estimated by the model to have drifted south of 48° N.
There is striking year-to-year variability evident in the 112-year record of IIP’s iceberg counts (See Figure 1 below and Table 1
(a PDF which will open in a new window)). The mean number of icebergs estimated to have passed south of 48° N is 474. The greatest number of icebergs (2202) occurred in 1984, while twice in IIP’s history (1966 and 2006) no icebergs were estimated to have passed south of 48° N. Five times in IIP’s history there has been at least one iceberg estimated to have passed south of 48° N during each of the months of the ice year: 1915, 1919 through 1921, and 1985. April and May are, by far, the months with the most icebergs entering the shipping lanes.
Number of icebergs estimated by IIP to have passed south of 48° N since 1900.
Anderson, I., 1993, International Ice Patrol’s Iceberg Sighting Data Base 1960-1991, Appendix D in Report of the International Ice Patrol in the North Atlantic, Bulletin No. 79.
International Ice Patrol, 1926. International Ice Observation and Ice Patrol Service in the North Atlantic Ocean, Season of 1926, Bulletin No. 15, 127 p.
International Ice Patrol, 1927. International Ice Observation and Ice Patrol Service in the North Atlantic Ocean, Season of 1926, Bulletin No. 16, 119 p.
International Ice Patrol, 1940. International Ice Observation and Ice Patrol Service in the North Atlantic Ocean, Season of 1940, Bulletin No. 30, 89 p.
International Ice Patrol, 1941. International Ice Observation and Ice Patrol Service in the North Atlantic Ocean, Season of 1941, Bulletin No. 31, 62 p.