Iceberg sightings, shapes and management techniques

for offshore Newfoundland and Labrador:

Historical data and future applications

Denise Sudom and Garry Timco Adrienne Tivy

Oceans, Coastal and River Engineering Canadian Ice Service
National Research Council Environment Canada
Ottawa, Canada Ottawa, Canada
Denise.Sudom@nrc-cnrc.gc.ca Adrienne.Tivy@ec.gc.ca
Garry.Timco@nrc-cnrc.gc.ca

Abstract: For safe and efficient operations in the iceberg- databases were developed with a focus on the key needs for
infested waters offshore eastern Canada, accurate information on operating and decision-making on Canada’s east coast.
icebergs is needed. Databases on iceberg sightings, shapes and Quantitative information on iceberg populations and frequency
management techniques have been developed in order to bring is needed for probabilistic analysis of potential impacts with
all relevant iceberg information into one repository. Iceberg offshore platforms and subsea facilities. The geometry and size
sightings have been recorded offshore Newfoundland and of icebergs is a key component that dictates the local and
Labrador since the 1600s. Sighting methods, locations, yearly global loads that will be generated during an iceberg impact.
variability and uncertainties are discussed. In more recent times, Knowledge of iceberg management techniques is important in
detailed 2D and 3D measurements have been made of iceberg
order to maintain safe operations offshore, as well as to
geometries, which are useful for structural load calculations.
mitigate economic effects (i.e. the need to disconnect an FPSO
Techniques to deflect iceberg drift from critical offshore
locations have also evolved over the past 40 years. The various
to avoid an oncoming iceberg).
methods that have been used for iceberg management are Historical iceberg population data can also be applied in the
discussed, as well as the factors that affect their success rates. development of predictions of the number of icebergs in future
Relationships between historical iceberg populations and sea ice seasons. An understanding of the main factors that influence
can be used to forecast iceberg severity in future seasons; year-to-year changes in the iceberg population on the Grand
updated correlations have been made between sea ice coverage Banks, particularly relationships with some lead time, is of
and iceberg severity.
interest to the offshore oil and gas industry for planning
Keywords: iceberg sightings; iceberg management; iceberg operations. Previous studies have showed that iceberg severity
shapes; sea ice; Grand Banks; Labrador. (or the frequency of icebergs) on the Grand Banks is affected
by sea ice extent and concentration, and to a lesser extent by
I. INTRODUCTION atmospheric pressure fields and air temperature. In this paper,
updated correlations have been made between historical sea ice
The petroleum industry offshore Newfoundland and
coverage and iceberg severity.
Labrador faces unique challenges for successful operation.
During several months of the year the region is infested with II. ICEBERG SIGHTINGS
icebergs, which can produce extremely high loads on vessels,
offshore platforms and seabed installations. Accurate A. Available data
information is needed on iceberg population densities, iceberg The NRC-PERD Iceberg Sightings Database is the result of
shapes and sizes, and methods of iceberg management. In the a major effort to collect visual and radar-detected iceberg
early days of petroleum development on the Grand Banks, each sightings from ships, offshore structures, aircraft and satellite
of the petroleum companies collected and managed their own from the past 400 years. The Iceberg Sightings Database is a
iceberg data sets, and the data were scattered and isolated. The compilation of iceberg sighting entries from the years 1619 to
National Research Council of Canada (NRC), with the support 2012, and has become the industry standard for historical
of the Program of Energy Research and Development (PERD), iceberg sightings offshore Newfoundland and Labrador.
sought to remedy this situation. They began a program to work
with both industry and the offshore regulators to develop and Prior to the creation of this database, iceberg information
maintain industry standard databases on key iceberg was scattered in various sources and industry operators
information. maintained individual datasets. The database was first
compiled in 1998 from data sets that were kindly provided by
Three NRC-PERD databases were created: Iceberg the petroleum companies operating on the Grand Banks.
Sightings, Iceberg Shapes and Iceberg Management. These AMEC (Newfoundland) and BMT Fleet Technology

978-1-4799-4918-2/14/$31.00 ©2014 Crown

developed the initial version of the database under contract to 25,000
Occasional ship reports

the NRC. Since then, BMT Fleet has regularly maintained and More regular ship reports
Aircraft sightings
updated it under contract to NRC. The latest version of the Regular IIP sightings
Petroleum industry
database was published in 2013 [1]. An abbreviated description 20,000 sightings

Number of yearly iceberg sightings

was published by Verbit et al. [2]. Re-sighted icebergs
No known re-sights
The database includes the date and time of sightings,
15,000
geographic coordinates, iceberg shape and size categories (with
dimensions if available), along with information on the source
of the sighting. A total of 387,781 iceberg sightings are
included, of which 93,812 are confirmed re-sights (the same 10,000

iceberg tracked more than once) and 20,771 are possible re-
sights. As an example of one season of data, Fig. 1 shows the
location of all iceberg sightings in the database for the 2012 5,000

season.
Iceberg sightings come from visual or radar observations 0
made from ships, offshore structures, aircraft and satellite

1850

1860

1870

1880

1890

1900

1910

1920

1930

1940

1950

1960

1970

1980

1990

2000

2010
imagery. Fig. 2 shows the timeline of annual iceberg sightings Ice season
and times during which various data sets are available. Fig. 2. Reported iceberg sightings since the year 1850, and times during
Information from the International Ice Patrol (IIP) makes up a which various data sets are available.
large part of the NRC-PERD database. Aircraft sighting data is
available since 1941, and regular sightings from IIP began in
1960 [3]. Data from the offshore oil operators has been B. Analysis of database
compiled and provided by Provincial Aerospace Ltd. (PAL) Fig. 3 shows the annual number of iceberg sightings both
since 1973. Satellite data is also compiled since 2011. In above and below 48°N since the year 1960. The figure includes
addition, historic iceberg sighting data are included – mainly only the location of the first sighting of the iceberg, not of any
collected by Brian Hill of NRC [4]. Sporadic ship-ice collision subsequent sightings (or “re-sights”). The definition of the
and other reports date back to the 1600s, and more regular annual “ice season” is October 1 to September 30. For
sightings have been made beginning in the mid- to late 1800s. example, icebergs sighted on November 1, 2000 and February
1, 2001 would both be included in ice season 2001. The
Iceberg sightings offshore Newfoundland and Labrador are
Iceberg Sightings Database is not a statistical representation of
not necessarily made on a standard basis, even in modern
the true yearly number of icebergs offshore Newfoundland and
times. Sightings may be made from a ship that happens to pass
Labrador. The IIP data discussed in Section V of this paper –
an iceberg, or from an aircraft that covers different areas in
yearly number of icebergs crossing the 48th parallel – is
different years. More sightings are made during periods of
somewhat of a better representation of annual iceberg severity.
increased shipping or offshore petroleum activity.
However, IIP does not cover the exact same area every year,
and visibility from the aircraft and distances travelled can vary.

10,000

9,000
N of 48
S of 48
8,000

7,000
Number of icebergs

6,000

5,000

4,000

3,000

2,000

1,000

0
1960
1962
1964
1966
1968
1970
1972
1974
1976
1978
1980
1982
1984
1986
1988
1990
1992
1994
1996
1998
2000
2002
2004
2006
2008
2010
2012

Ice season

Fig. 3. Yearly frequency of iceberg sightings from all sources since 1960,
excluding known re-sights. The data is divided into icebergs north or
south of 48°N, based on the location of the first sighting of the iceberg.
Fig. 1. Plot of all icebergs in the NRC-PERD Iceberg Sightings Database for
the 2012 ice season.
 Table I shows the number of icebergs that are first sighted iceberg geometries, with sufficient data to allow for 3D
by each method, along with the total numbers of sightings. rendering. Fig. 6 gives examples of icebergs with less detailed
Sightings may be made from ships, aircraft, or the offshore information. All images are available in the database.
industry; the sighting may first be made using either visual and
radar methods, or both. Recently, satellite sightings have also The information in the Iceberg Shapes Database was
become part of the database. (Numbers in Table I do not add compiled with the intention that it would be used for iceberg-
up to the totals, since for some sightings the detection method structure load calculations and iceberg drift modeling. Iceberg
is not recorded.) In Table II, the numbers of icebergs of mass can be used for calculation of kinetic energy. The length
various sizes first sighted north and south of 48°N are given. and width of the iceberg are used in the calculation of impact
rates and pack ice loads. High-resolution data is necessary in
TABLE I. ICEBERG SIGHTING METHODS order to better design local members of offshore ships and
structures for impacts with various iceberg shapes. Detailed 3-
Number of records by sighting method, dimensional shape of the iceberg is needed for the calculation
Total
Source of excluding re-sights
sightings, of maximum load during the indentation (from the
iceberg Visual
sighting Visual Radar & Total
including instantaneous area of contact and ice failure pressure). The
re-sights moment of inertia is also useful; most iceberg-structure
radar
Ship 77,543 1,388 664 86,464 104,917 impacts are off centre and the translational kinetic energy is
absorbed in both crushing of the ice and rotational kinetic
Aircraft 66,760 13,442 17,071 100,823 141,987 energy.
Offshore
129 725 1,176 4,039 27,079
industry
Satellite - - - 1,844 3,054

TABLE II. SIZE CLASSIFICATIONS OF SIGHTED ICEBERGS

Number of records, excluding

Size classification (and known re-sights
waterline length)
North of 48°N South of 48°N
Growler (<5 m) 16,255 15,960
Bergy bit (5 - 15 m) 1,583 841
Small (15 - 60 m) 27,772 15,421
Medium (61 - 120 m) 30,737 12,812
Large (121 - 200 m) 15,492 11,115
Very large (> 200 m) 2,139 833
Ice island 64 283

III. ICEBERG SHAPES

A. Available data
Fig. 4. Locations of profiled icebergs in the NRC-PERD Iceberg Shapes
A database on Iceberg Shapes has been compiled from Database.
detailed 2D and 3D information on over 800 iceberg
geometries. The database was originally compiled by Canatec
[5] under contract to NRC; it has since been updated at NRC
[6]. Data were sourced from 1976 to 1985 from the Hibernia
and Terra Nova development studies and various projects on
iceberg characterization (e.g., by C-CORE, Fisheries and
Oceans Canada, and Fenco). Eighty percent of the data are
from Grand Banks region, with the remainder from the
Labrador Sea. Fig. 4 shows the geographic area over which
measurements have been made. The iceberg shapes data is also
included (in lesser detail) in the Sightings Database.
The database includes the measurement date and location
of the iceberg, along with general information on the size and
shape classification. The height, waterline length and width,
and draft are summarized, and 2D or 3D (x,y,z coordinates) are
given where available. Fig. 5 shows several examples of
Fig. 5. Examples of 3D geometries in the Iceberg Shapes Database.
 operations offshore Newfoundland. Fig. 7 shows a successful
towing operation that took place near the oil installations on the
Grand Banks in 2000. An iceberg passed between the Hibernia
and White Rose oil fields, and was then towed by a support
vessel as it was predicted to drift near the Terra Nova drilling
site. Eventually the iceberg was released at a safe distance
from Terra Nova.
The NRC-PERD Iceberg Management Database is an
extensive compilation of records of iceberg management
operations from 1973 onward, bringing together information
from numerous oil and gas exploration and development
companies. The Iceberg Management Database has been
compiled by Provincial Aerospace Ltd. (PAL) under contract
to the NRC starting in 2001, and has since been regularly
updated. Rudkin et al. [7] have given a summary of the
development of the database. The raw data mostly came from
various “Well History Reports” that the offshore regulator
Fig. 6. Examples of 2D and topographic sketches in the Iceberg Shapes required to be filed at the completion of a drilled well offshore
Database.
eastern Canada. Before the creation of the database, a
stakeholders’ workshop was held to determine the types of
information that would be needed. The earlier data came from
B. Analysis of database handwritten notes, which were often very detailed; the more
The Iceberg Shapes Database contains measured modern computerized records generally contain less
dimensions from 872 icebergs, including the following: information. Some of the older data, however, had lower
accuracy for iceberg drift and tow tracks which needed to be
• 3D geometry for 28 iceberg keels from sonar profiling taken into consideration.
• 3D geometry for 566 sails, from stereo photography
The current version includes data up to and including the
(digital or image files) 2012 season [8]. It contains over 1,700 records of iceberg
• 2D profiles for 155 iceberg sails and keels (image management operations: 949 from the Grand Banks, 721 from
files) offshore Labrador and 77 from West Greenland. The total
• 3D animations for full views of 76 icebergs (sails, number of individual icebergs managed is 1172; some icebergs
keels or both). require more than one management attempt. Fig. 8 illustrates
the type of iceberg parameters and trajectory information
The profiled icebergs are fairly representative of the included in the database.
different types and sizes that may be encountered offshore
Newfoundland and Labrador. The database includes The iceberg information from the Management Database is
geometries of various shapes of icebergs, including domed, also included (in lesser detail) in the Iceberg Sighting
drydocked, pinnacled, tabular, and wedged; iceberg sizes range Database.
from bergy bits to very large icebergs. The range of sizes of
icebergs in the database is given in Table III.
The Iceberg Shapes Database needs to be updated to
include data for recent years, and this effort is underway.

TABLE III. RANGE OF MEASURED VALUES FOR ICEBERG GEOMETRIES IN

SHAPES DATABASE

Parameter Minimum (m) Maximum (m)

Waterline length 3 450

Height above water 8.5 107
Draft 10 185

IV. ICEBERG MANAGEMENT

A. Available data
Iceberg management operations have been documented
Fig. 7. Example of a successful iceberg management operation on the Grand
since the 1970s and are now a routine part of oil development Banks, reproduced from [8].
 TABLE IV. SIZE CLASSIFICATIONS OF MANAGED ICEBERGS

Size classification (and Number of recorded management

waterline length) events

Growler (<5 m) 43
Bergy bit (5 - 15 m) 117
Small (15 - 60 m) 627
Medium (61 - 120 m) 525
Large (121 - 200 m) 337
Very large (> 200 m) 8
Size unknown / not
93
recorded

Various techniques have been used for iceberg

management. By far, the most common method is towing with
Fig. 8. Example of iceberg management information record created by PAL
for the NRC-PERD Iceberg Management Database, reproduced from
a single vessel, as shown in Table V. Towing with a net or
[8]. washing by ship propeller have also been utilized on many
occasions.
The definition of a “new” iceberg management operation as
B. Analysis of database opposed to another attempt within the current operation is
subjective. For the database, the original documentation was
Fig. 9 shows the annual number of recorded iceberg studied, and if it was unclear then a time-based method was
management events on the Grand Banks, Labrador Shelf and used [8]. If more than 12 hours elapsed between management
offshore West Greenland. The sizes of the managed icebergs attempts, then the next attempt was considered a new
range from growlers to one iceberg with a waterline length of operation. Most icebergs will be sufficiently deflected at the
350 m. As seen in Table IV, most managed icebergs are in the end of a single management operation, or after the second
small and medium categories. Management attempts have attempt. Looking at all management events in the database, an
been made on icebergs of all shape categories (wedge, dry- iceberg will on average require 1.5 attempts. The maximum
dock, pinnacle, tabular, etc.). recorded number of management attempts on one iceberg is 17.
There are numerous reasons why an iceberg management
attempt or operation may end; these can be divided into the 11
250
Grand Banks
categories shown in Table VI (adapted from [8]). The majority
Labrador Shelf
of iceberg management events are purposely ended when the
tow is released. In terms of an unexpected end to operations,
Number of iceberg management events

West Greenland
200
the most common occurrence is the slippage of a tow line,
followed by the rolling of an iceberg.
150
TABLE V. FREQUENCY OF USE OF VARIOUS ICEBERG MANAGEMENT
METHODS

100
Number of events
Iceberg management method
(and percentage of total records)

50 Single vessel tows 1418 (81%)

Net tows 136 (8%)
0 Prop-washing 96 (6%)
1973
1975
1977
1979
1981
1983
1985
1987
1989
1991
1993
1995
1997
1999
2001
2003
2005
2007
2009
2011

Water cannon management 49 (3%)

Year
Fig. 9. Annual number of iceberg management events in the NRC-PERD Two-vessel tows 38 (2%)
Iceberg Management Database, by geographic area. Ramming 5 (<1%)
Other techniques 8 (<1%)
 TABLE VI. REASONS FOR ENDING AN ICEBERG MANAGEMENT Table VII shows the success rates of iceberg management
OPERATION
events in the database, using the two different definitions. In
Reason for ending Number of Percentage of 11% of the records, the iceberg management operation was not
operation events total records technically successful, in that the course of the iceberg was not
Tow Released 1017 58% altered as planned. However, if the iceberg does not cause any
downtime in offshore petroleum operations, then the operation
Rope Slipped 404 23%
is still an “operational success”.
Berg Rolled 156 8.9%
PAL has attempted to develop correlations between iceberg
Ice Priorities 36 2.1% management success (as calculated by their algorithm) and the
Berg Broke 29 1.7% various factors involved in management operations [8]. Single
vessel rope tows generally result in more successful
Berg Grounded 30 1.7% management compared to the other commonly used methods.
Vessel Priorities 21 1.2% Iceberg shape appears to have somewhat of an effect on the
Change Method 18 1.0% tow outcome; wedge, dome and pinnacle icebergs have a lower
calculated success compared with tabular, blocky or dry-dock
Rope Failure/Parting 11 0.6% icebergs. Iceberg size does not seem to play an important role
Change Vessels 7 0.4% in the success of a management operation, perhaps because
vessels are chosen based upon the size of iceberg to managed.
Other 21 1.2%
TABLE VII. ICEBERG MANAGEMENT SUCCESS RATES
Timing is a critical factor in iceberg management, as
Percent of total management
support vessel time is very costly and vessels may have duties Success of operation
events
unrelated to ice management. The average time required to
tow an iceberg is 7 hours for Labrador and 12 hours on the C-CORE definitions:
Grand Banks. The connection of a vessel to an iceberg
Operational success 99%
generally takes between 1 and 2 hours. The time to the closest
point of approach (TCPA) is the calculated time required by Technical success 88%
the iceberg to reach its closest point to an offshore facility. On
PAL definitions:
average for the Grand Banks, iceberg management operations
began 13 hours before TCPA. Of the 949 management Complete success 7%
operations on the Grand Banks, 39 were started with a TCPA
Successful 25%
of less than 2 hours.
Acceptable 39%
The overall success of an iceberg management operation
depends on a number of factors related to the iceberg Poor 28%
characteristics, wind and waves, type of management
attempted, timeline of operations, and response of the iceberg.
These factors are included in the database. There are several V. FACTORS AFFECTING YEARLY ICEBERG FREQUENCY ON
methods of defining the success of an iceberg management THE GRAND BANKS
operation. C-CORE [9] has made a distinction between An understanding of the main factors that influence year to
technical and operational success: year changes in the iceberg population on the Grand Banks,
particularly relationships with some lead time, is of interest to
• Operational success: Downtime in operations was
the offshore oil and gas industry for planning operations.
avoided Previous studies have focused on predicting the number of
• Technical success: A demonstrated change in course icebergs drifting south of 48°N, a time-series maintained by the
was achieved and the towed iceberg achieved a course International Ice Patrol (IIP) that extends back to 1900.
made good with one or multiple attempts. Atmospheric pressure fields and air temperature were found to
influence iceberg severity at lead times of 1 to 6 months (e.g.
Since this definition offers no quantification of values for [10]; [11]. Marko et al. showed that spring ice extent in the
deflection or change in course to establish a “technical Labrador Sea, followed by winter ice extent in Davis Strait, are
success”, PAL has created an algorithm to quantify the relative the critical parameters for forecasting iceberg severity [12].
success of an iceberg management operation [8]. Numeric This is supported by Peterson [13] who reports squared
values are assigned to the following key fields: correlation coefficients (r2) as high as 0.72. That is to say,
winter sea ice extent variability (or year to year change) in
• Number of connections or attempts
Davis Strait can explain 72% of the variability in iceberg
• Vessel tow heading versus the iceberg’s course made severity.
good
• Was there a change in the iceberg’s drift course? Fig. 10 shows the total number of icebergs, between
• Was there a change in the iceberg’s drift speed? October and September, drifting south of 48°N. This time-
series is highly variable, with a maximum of 2202 in 1984 and
• What was the final outcome?
a minimum of 0 in both 1966 and 2006. Overall the mean is
491 with a standard deviation of 567. An important explained by the correlation pattern in February ice
consideration is the high level of uncertainty in this time-series. concentration.
The mission of the International Ice Patrol is to define the
extreme southern limit of icebergs rather than to determine the
total number of icebergs. The change in methodologies for
iceberg detection over time is another source of uncertainty, for
example the introduction of technology such as SLAR (side-
looking airborne radar).
The IIP data shown in Fig. 10 differ from that in the NRC-
PERD Iceberg Sightings Database. The iceberg sightings
shown in Fig. 3 are based on the location of the first sighting of
an iceberg only; many icebergs are first sighted north of 48°N
and subsequently drift further south. More importantly, the
NRC-PERD Database includes all IIP data plus data from
numerous other sources. Since these sources vary from year to
year depending mainly on the petroleum industry, the IIP data
was used as a somewhat less biased representation of the
annual iceberg severity.
2500
Number of icebergs south of 48N

2000

1500

Fig. 11. Map of correlation between iceberg severity and February sea ice
1000 concentration (SIC) offshore Newfoundland and Labrador for 1950-
2013. Monthly sea ice concentration on a 1-degree grid is from the UK
500 Hadley Centre [14].

0
Improvements in the quality of the time-series data on
iceberg counts could be made to account for changes in aircraft
1950
1953
1956
1959
1962
1965
1968
1971
1974
1977
1980
1983
1986
1989
1992
1995
1998
2001
2004
2007
2010
2013

survey areas over time. Peterson [13] started to address this

Ice season
issue by normalizing the data based on the number of flights in
Fig. 10. Number of icebergs drifting south of 48N for iceberg seasons each year. In the past, BMT Fleet, under contract to the NRC,
(October 1 – September 30) 1950-2013. developed a statistical analysis product to calculate the density
of icebergs as viewed by aircraft. Flight paths were plotted and
In the present study, updates have been made to past work iceberg counts were determined within the estimated visible
on identifying statistical relationships between atmospheric sighting areas [2].
pressure fields, air temperature and ice extent with the IIP
iceberg severity time-series (Fig. 10) using a new statistical VI. SUMMARY AND RECOMMENDATIONS
forecasting program developed by the Canadian Ice Service
(CIS) and the NRC. The forecasting program uses an Accurate information on icebergs is essential for safe and
automated regression scheme to identify empirical efficient operations in the Iceberg Alley offshore eastern
relationships between climate variables and a time-series of Canada. Quantitative information on iceberg sightings is
interest, with some lead time. The predictors tested to forecast needed for assessing the frequency or severity of icebergs from
iceberg severity include regional sea level pressure, surface air year to year. The geometry and size of icebergs are key
temperature and sea ice concentration from lead times ranging components in calculating the local and global loads that could
from 1-month (February) to 12 months (previous March). be generated during an iceberg impact. Also, information on
Consistent with previous work (e.g. [12]; [13]), winter sea ice iceberg management techniques is important to help maintain
extent is the dominant predictor of seasonal iceberg severity. safety offshore, as well as to mitigate the risk of disruption or
Fig. 11 shows the correlation map between iceberg severity downtime for petroleum operations.
based on the IIP data (Fig. 10) and February sea ice In the early days of petroleum development on the Grand
concentration. Regions of high correlations or regions where Banks, the offshore operators collected and managed their own
there is a strong relationship between ice concentration and iceberg data sets, and the data were scattered and isolated. In
iceberg severity (red regions in Fig. 11), are along the marginal the late 1990s the NRC began a program to work with both
ice zone. This highlights, as in previous studies, the predictive industry and the offshore regulators to develop and maintain
capability of ice extent or the location for the winter ice edge. industry standard databases on key iceberg information. Three
From the present analysis, approximately 55% of the databases have been created to provide a comprehensive source
variability in the total number of icebergs south of 48°N is of iceberg information: the NRC-PERD Databases on Iceberg
Sightings, Iceberg Shapes, and Iceberg Management. These
databases are available in Microsoft Access and include built- [2] Verbit, S., Comfort, G. and Timco, G., 2006. “Development of a
in query functionality and links to pop-up images or Database for Iceberg Sightings off Canada’s East Coast”, Proceedings
of IAHR’06, Vol. 2, pp. 89-96, Sapporo, Japan.
animations, and are available from the NRC by contacting the
[3] National Snow and Ice Data Center (NSIDC), 2014. International Ice
lead author of this paper. Patrol (IIP) Iceberg Sightings Database. Available:
An understanding of the main factors that influence http://nsidc.org/data/g00807. Last accessed 1 July 2014.
seasonal changes in the iceberg population on the Grand [4] Hill, B., 2012. Ice data. Available: http://www.icedata.ca/. Last accessed
1 July 2014.
Banks, particularly relationships with some lead time, is of
[5] Canatec 1999. “Compilation of Iceberg Shape and Geometry Data for
interest to the offshore oil and gas industry for planning the Grand Banks Region”, PERD/CHC Report 20-43, Calgary, AB,
operations. In the present study, consistent with previous work, Canada.
winter sea ice coverage was found to be the dominant predictor [6] Barker, A., Skabova, I. and Timco, G.W., 1999. “Iceberg Visualization
of seasonal iceberg severity. Approximately 55% of the Database”, NRC Report HYD-TR-042, Ottawa, ON, Canada.
variability in the total number of icebergs south of 48°N can be [7] Rudkin, P., Young, C., Barron, P. Jr. and Timco, G.W., 2005. “Analysis
explained by the pattern of February ice concentration. This and Results of 30 Years Of Iceberg Management”, Proceedings of the
correlation is useful because the majority of icebergs drift 18th International Conference on Port and Ocean Engineering under
offshore Newfoundland and Labrador in the months of March Arctic Conditions (POAC’05), Vol. 2, pp. 595-604, Potsdam, NY, USA.
to June, so the February ice concentration can be used to [8] Provincial Aerospace Ltd., 2013. “PERD Iceberg Management Database
Update Through 2012 Ice Season”, PAL Report ESD-P0036 to NRC.
predict iceberg severity later in the season.
[9] C-CORE, 2002. “Integrated Ice Management R&D Initiative-Year
With improvements to the NRC-PERD Iceberg Sightings 2001.” C-CORE Report R-01-24-605 to Chevron, ExxonMobil, Husky,
Database there is an opportunity to create a new seasonal Norsk Hydro, PERD and Petro-Canada.
iceberg severity time-series that includes more observations [10] Schnell, I.I., 1962. “On the iceberg severity off Newfoundland and its
prediction”, Journal of Glaciology, Vol. 4, pp. 161-172.
and takes into consideration the frequency and location of
[11] Walsh, J.E., W.I.Wittmann, L.H.Hester and W.S.Dehn, 1986. “Seasonal
iceberg surveys. It is expected that iceberg severity prediction prediction of iceberg severity in the Labrador Sea”, Journal of
skill will improve with efforts to refine the iceberg count time- Geophysical Research, Vol.19, No.C8, pp. 9683-9692
series to better represent the “true” iceberg population in each [12] Marko, J. R., D. B. Fissel, P. Wadhams, P.M. Kelly and R.D. Brown,
year. 1994. "Iceberg severity off eastern North America: its relationship to
sea-ice variability and climate change", Journal of Climate, Vol. 7, pp.
ACKNOWLEDGMENTS 1335-1351.
This work was supported by the Program of Energy [13] Peterson, I.K., 2004. “Long-range forecasting of the iceberg population
on the Grand Banks”, Proceedings of the 14th International Offshore and
Research and Development (PERD) and the NRC Arctic Polar Engineering Conference (ISOPE), pp. 831-836, May 23-28,
Program. Toulon, France.
[14] Rayner, N.A., D.E. Parker, E.B. Horton, C.K. Folland, L.V. Alexander,
REFERENCES D.P. Rowell, E.C. Kent, and A. Kaplan, 2003. “Global analyses of sea
[1] BMT Fleet Technology Ltd., 2013. “PERD Iceberg Sighting Database surface temperature, sea ice, and night marine air temperature since the
Update 2013”, BMT Report 7909FR to NRC. late nineteenth century.” J. Geophys. Res. (Atmospheres), 108(14), DOI:
10.1029/2002JD002670.