From Sea, to Air, to Space: Nearly 100 Years of IIP Iceberg Tracking Technology and Beyond
 
Byline: by LT Erin Christensen, Operations Officer, International Ice Patrol, U.S. Coast Guard and MSTCS John Luzader, Command Senior Chief, International Ice Patrol, U. S. Coast Guard.
 
Introduction
 
The maritime community has long recognized the need for iceberg information to assist mariners in navigating safely through the waters of the North Atlantic. As shipping technology advanced, carrying more cargo and people at faster speeds, knowing the location of the iceberg danger became more important in avoiding collisions. The formation of the International Ice Patrol (IIP) shortly after the 1912 sinking of RMS Titanic provided mariners the source of information they needed to navigate safely. In keeping with its core value of “Improvement”, IIP has aggressively pursued the latest technology to assist in gathering, analyzing, and distributing maritime safety information over its nearly 100-year history. What started as ships without radars or precise navigation searching for icebergs in the notorious Grand Banksfog is on the threshold of fusing information from satellite-based radars with data from automated ship tracking systems and using vastly improved numerical models of iceberg drift and deterioration to provide a comprehensive picture of the iceberg distribution.
 
This article describes how IIP has used technological advances to improve iceberg tracking over its history and presents a vision of future IIP operations. A companion article entitled: “International Ice Patrol: For the purpose of Safeguarding Life and Property at Sea” provides a brief history of IIP with a focus on its foundation and operations.
 
By Sea: Titanic to World War II
           
The first IIP ships, U.S. Revenue Cutter (USRC) Seneca and USRC Miami, conducted reconnaissance using lookouts. To enhance the lookouts’ ability, they experimented with several ways to find icebergs day or night: “At night we experimented with the searchlight, but found it to be of little use. We could see farther at night with the naked eye and marine glasses…” [i] They also made observations about the characteristics of ice and investigated events such as ice blink to determine if there were better ways to spot bergs: “The phenomenon of ice blink was quite prevalent around ice fields. On clear nights, especially when the moon was up, the sky along the horizon in the direction of the ice was markedly lighter than the rest of the horizon; this effect would be noted before the ice was sighted.”[ii] Unfortunately, they found that while ice blink was useful for detecting the presence of sea ice, it was not helpful for detecting icebergs.
 
The earliest efforts at using indirect methods of detecting icebergs met with mixed success. In 1913, the crews of the Ice Patrol vessels tried using the ship’s steam whistle to detect icebergs with little success. They concluded that the “existence of an echo means an obstruction, but its absence proves nothing”. They also tried to determine if an iceberg was near by measuring the sea and air temperature. They found that a sudden decrease in the temperature of the water means nothing as far as icebergs are concerned, concluding that “(the temperature of) sea water is streaky as a rule”.[iii] Also, they found that if a ship was close enough to feel the cooling effect that an iceberg has on the air, it was too close to the iceberg and the ship was in danger of collision. In a more promising 1914 effort, Professor Reginald Fessenden of the Submarine Signal Company used a “submarine electric oscillator” to detect icebergs using the propagation of sound through the water. He was able to detect an iceberg 0.5 to 2.5 miles away. He also successfully used this technology to determine the depth of the water, which later led to the invention of the fathometer.[iv]
 
Ultimately, standing at the highest point of the vessel above water and using the naked eye and binoculars was the primary method of iceberg observation. In addition, it should be noted that from the earliest days of Ice Patrol, ships transiting the area have reported ice and weather information and continue to do so today. Reports from these ships, sometimes referred to as “the thousand eyes of the Ice Patrol”,[v] assist IIP in creating accurate products for all transatlantic mariners.
 
Visual iceberg detection remained the only useful method of locating icebergs from ships until the advent of shipboard radars during World War II (WWII). In 1946, U. S. Coast Guard Cutter (USCGC) Mojave was equipped with both a 10-cm radar and 3-cm radar.[vi] The numbers refer to the wavelength of the microwaves the radar transmits. In general, longer wavelength radars have poorer resolution than short wavelength radars, but are less affected by sea clutter. The Mojave crew faced several challenges evaluating the radars. The weather on the Grand Banksfrequently made Mojave pitch and roll putting the radar equipment through heavy vibrations, which impacted radar calibration and the ability. It was also a challenge to conduct necessary maintenance on the radar equipment when the equipment had to be turned off in order to repair it. The radars were regularly operated for 24 hours a day to augment navigation during periods of reduced visibility resulting from poor weather.
 
At the end of the 1946 season, IIP determined that smaller sea chop seemed to have a greater radar return than larger sea swells. The study found that radar sea clutter could mask detecting medium size icebergs. In addition, by using radar alone there was no way to differentiate between an iceberg and a vessel. Ultimately, they concluded that the heavier the sea state (sea clutter) and winds the greater the chance that a radar would miss an iceberg. They concluded that “a vigilant watch must be kept by any vessel, naval or commercial, operating in areas which ice is likely to be encountered.”[vii]
 
In 1959, IIP conducted another, more extensive study of iceberg detection by shipboard radar [viii], but by then it was becoming clear that aircraft had replaced ships as the preferred platform for iceberg reconnaissance. As a result, the focus of the iceberg detection studies turned toward aircraft-based systems, which is discussed later.
 
The early Ice Patrol cutters carried Oceanographers onboard to gather information on the North Atlantic Ocean. They conducted hydrographic surveys near the Grand Bankscollecting a wide range of data including: water depth, salinity, water temperature, water samples, and water movement. [ix] In 1920, scientific observer Albert L. Thuras noted that icebergs were carried south along the coast of Labradorby the Labrador Current until it mingled with the north-flowing Gulf Stream. Icebergs would generally remain south of the Grand Banks, where the two currents met, until they melted. He also noted that the area of “mixed waters” was usually very foggy, held icebergs, and was extremely dangerous for vessels.[x] The Commanding Officers and Oceanographers drew conclusions from their data and learned much about the North Atlantic Ocean circulation and how it impacted iceberg drift and deterioration. These early observations laid the framework for newer data collection and forecasting technologies in iceberg tracking.
 
By Air: World War to Present
 
Immediately following WWII, IIP began to use aircraft for iceberg reconnaissance. The reason for the shift from sea to air patrols was that aircraft could cover a much larger area in a fraction of the time of its seagoing counterparts. In 1946, the first aircraft used was the PBY-5A “Catalina” flying boat and the PB4Y-1 “Liberator”, better known as the B-24. By the end of the 1946 season, the first PB-1G “Flying Fortress” was introduced into the Ice Patrol aircraft inventory. This airplane, famous for its WWII bombing missions, was to become the workhorse of IIP aerial reconnaissance for the next 12 years.
 
The patrol aircraft were equipped with Loran for navigation and radar that augmented ice observers’ visual sightings. In 1946, the aircraft flew an average of 7.7 hours per flight at 25-nm track spacing.[xi] The types of radar used were the AN/APS-3 and AN/APS-15A.[xii] Similar to shipboard radars, sea state and clutter such as sea ice presented target identification challenges because they had radar returns which made it difficult to identify icebergs. Because the aircraft moved so quickly over the surveillance area, there was also no accurate way to estimate the speed of a radar target to determine if it might be a vessel or an iceberg. Icebergs and the wooden baroque commonly used for fishing near the Grand Bankspresented similar radar returns. However, the 1946 annual report does note that “steel vessels give a somewhat sharper and brighter echo than do icebergs.”[xiii] Additionally, small dories around a larger fishing schooner had similar radar returns as an iceberg surrounded by growlers (smaller pieces of ice).[xiv] IIP concluded that aircraft-based radars were useful tools to determine a target’s existence, but the targets had to be visually confirmed as ice because the radar returns from an iceberg did not differ much from the radar returns from vessels. [xv]
 
Initially, IIP considered aircraft to be a supplement rather than a replacement for surface patrol vessels. Their major concern was the difficulty airplanes had distinguishing between icebergs and vessels when the visibility was poor. As a result, aerial reconnaissance missions were scheduled during late winter and early spring when visibility was more likely to be good.  Surface patrol vessels were deployed later in the spring when fog is typically more prevalent near the Grand Banks. As IIP gained more confidence in aerial reconnaissance throughout the 1950s, this attitude reversed. Aircraft became the primary reconnaissance platform and surface patrols were used only in severe iceberg seasons.
 
In 1963, Ice Patrol began using the HC-130B “Hercules” based out of Elizabeth City, North Carolina (a partnership that began shortly after WWII and continues to this day) for iceberg reconnaissance.[xvi] As the airframes improved, so did the radar systems; however, until 1983 the primary method of locating icebergs remained visual detection. This severely limited the number of days a patrol could be conducted because of the requirement for reasonably good visibility conditions in the planned search area. All that changed in 1983 when IIP began using the AN/APS-135 Side-Looking Airborne Radar (SLAR) for iceberg detection. The SLAR was a powerful radar that had the ability cover a wide area on each side of the aircraft. It recorded a radar image on film and in later years digitally. Careful analysis of the radar image allowed an experienced operator to distinguish between an iceberg and a vessel by using several cues, including overall target movement and characteristics of the radar return (target intensity, the presence of wakes, etc.). This system provided IIP with near all-weather reconnaissance capability because in many cases IIP could distinguish between ships and icebergs without visual confirmation. In some cases, however, making the distinction was difficult and the aircraft had to descend beneath the clouds to visually confirm the identity of a radar target.
 
The ability to distinguish between an iceberg and a ship using radar alone took another major step forward with the introduction of the AN/APS-137 Forward-Looking Airborne Radar (FLAR) in 1993. This system, used in conjunction with the SLAR, had an Inverse Synthetic Aperture Radar (ISAR) mode, which gave IIP the ability to create a Doppler image of an individual, unknown target. In this mode, the FLAR focused on the object as it tossed in the seaway. The resulting image of the radar return was used by an experienced operator to identify the object without visual confirmation. The fact that SLAR, later augmented with FLAR, was used for over two decades by IIP is a testament to the system’s effectiveness.
 
In 2009, IIP began using the HC-130J with the ELTA-2022 Multi-ModeRadar (MMR) for iceberg reconnaissance.  IIP uses the ELTA’s search mode, which scans 360 degrees, for detection and its ISAR mode for identification.
 
Finally, IIP is in the early stages of testing the Coast Guard’s newest airframe, the HC-144A “Ocean Sentry”, to determine its ability to conduct iceberg reconnaissance. The HC-144A is equipped with the Telephonics APS-143C (V) 3 Multi-Mode Radar that shares similar capabilities to the ELTA. This airframe is based out of Aviation Training Center Mobile, Alabama.   
 
By Space: Into the Future
 
Since the first weather satellite was put into orbit in 1960, IIP has been eager to use space-based systems for iceberg reconnaissance.[xvii] (Jarboe, Appendix D 2002 AR,)[ERC1] The first efforts at detecting icebergs using satellites was met with little success. The sensors on the early satellites were visual and infrared, neither of which can reliably detect objects through clouds.  In addition, the sensors were designed to monitor large scale weather systems so they were incapable of detecting anything but the largest icebergs.
 
By the mid-1990s, satellites with synthetic aperture radars (SAR) were being placed into orbit, providing the ability to detect targets regardless of cloud cover.  Newer SARs have improved resolution and varying beam modes that provide additional characteristics for each target.  Even with these advancements in technology, the classification of targets detected by satellite-based SAR is an active area of research.  IIP is evaluating the effectiveness of using other maritime domain awareness data to resolve target ambiguity.  Continued evaluation and research has gradually increased IIP’s confidence in incorporating the available information into the drift and deterioration model.  27% of the icebergs modeled by IIP were derived from SAR satellites in 2011.
 
Several challenges must be surmounted before satellites can contribute significantly to IIP’s reconnaissance.  The ability to distinguish between ships and icebergs must become reliable.  More satellites with high enough resolution to detect small icebergs must become available.  Currently, there is no single SAR-equipped satellite or constellation of satellites that can provide complete and frequent high-resolution coverage over IIP’s area responsibility that exceeds 450,000 square nm.
 
Forecasting
 
Iceberg tracking combines detection and forecasting. The previous sections focused on the improvements in the technology of iceberg detection, which is simply gathering the iceberg’s location, size, and shape. Forecasting, the process of estimating changes in the iceberg’s location and size, is important for two reasons. First, it helps IIP determine if a reported iceberg is new or an iceberg already in the database. Second, it allows IIP to estimate where an iceberg will be in the future and how much it will have melted.
 
Like iceberg detection, forecasting has made significant advances over IIP history. When the first IIP vessels patrolled the vicinity of the Grand Banks, they knew little about the currents, so most iceberg reports were viewed as new observations. There was little effort to estimate where an iceberg would drift. By the early 1930s, IIP Oceanographers were routinely conducting oceanographic surveys and using those results to calculate by hand the ocean currents. They used this information to help search for icebergs near the transatlantic shipping lanes. By the late 1940s, the IIP oceanographic vessel was providing monthly maps of the ocean currents and sea surface temperature. These oceanographic surveys continued until 1979 when IIP began using satellite-tracked oceanographic buoys to provide ocean current information.
 
In 1979, IIP began using a computer model to predict the movement of icebergs. The model was enhanced in 1983 to include predictions of iceberg melt, and thus predict the sizes of sighted icebergs over time.[xviii] IIP’s current forecasting system is called the iceBerg Analysis and Prediction System (BAPS). This model allows IIP to estimate the movement and deterioration of icebergs using environmental data such as winds, waves, currents, and sea surface temperature.
 
International Ice Patrol’s vision for the future
 
Twenty years from now, how IIP tracks icebergs will be drastically different than how it has been done in the past.  The use of radar satellite imagery, Vessel Movement Reporting Systems (VMRS), Automatic Identification System (AIS), and ship and sir Reports will automatically be ingested into a modernized modeling system. The system will differentiate between icebergs, vessels, and industry platforms so that it can accurately determine the iceberg limit. The modeling system will automatically apply real time environmental data to its forecast modeling to predict how the icebergs will move. The iceberg limit and warnings that IIP personnel currently create and disseminate will be computer generated and automatically distributed to the maritime community. What will not change is the IIP’s mission to monitor the iceberg danger near the Grand Banks of Newfoundland and provide the iceberg limit to the maritime community. The IIP is proud to say that no vessel that has heeded the Ice Patrol’s published iceberg limit has collided with an iceberg. Ice Patrol will continue to be ever vigilant and do the best job possible to provide mariners with information to help prevent another Titanic disaster.
 
About the authors:

LT Erin Christensen took over as the IIP Operations Officer in July 2011, after serving in Valdez, Alaskaat the Valdez Marine Safety Unit. Managing vessel traffic in the vicinity of ice and providing iceberg information to help mariners make informed navigation decisions has been a central component in LT Christensen’s Coast Guard Career.
 
MSTCS John Luzader has served in the Coast Guard for over 21 years. He has over 12 years of experience responding to oil spills, hazardous materials releases and natural disasters. This is his second assignment at IIP and has over seven years of experience tracking icebergs.
 
Key words: Technology, Air, Sea, Space, Tracking, Icebergs, Reconnaissance.
 
Example: port security, marine safety, environmental protection.