For people living in northern environments, ice is a common phenomenon that affects our local activities. Most of us however, don't consider its larger regional or global implications. Ice covers a substantial part of the Earth's surface and is a major factor in commercial shipping and fishing industries, Coast Guard and construction operations, and global climate change studies. Polar sea ice seasonally covers an even larger area, roughly equal in size to the North American continent, 25 million km².
Its extensive distribution means that sea ice plays a large role in the albedo of the earth. Albedo is a term referring to the measure of reflectivity of the Earth's surface. Ice and snow are highly reflective and changes in their distribution would affect how much solar energy is absorbed by the earth. Under warming conditions, the ice would melt, and less incoming energy would be reflected, thereby potentially increasing the warming trend. The opposite may also be true - an increase of ice due to cooler conditions would reflect even more of the incoming solar energy, potentially propagating even colder conditions. Of course these potential changes in sea ice distribution are of concern to scientists studying global climate change, as are sea ice interactions with the ocean and atmosphere.
During winter in the northern hemisphere, ice creates a substantial barrier to both lake and ocean going vessels trying to reach ports or navigating along coastlines. Ice floes, pack ice and icebergs create potential hazards to navigation, while landfast ice hinders access to the shore. Ice breakers are often used to create routes for ships to follow from the open water to their ports. In Canada, two important locations for this type of operation are the Gulf of St. Lawrence /Great Lakes and the Canadian Arctic. The Gulf is the main route for international cargo vessels headed for Montreal and Québec, and is affected by ice through the winter and spring. Canada's Arctic is home to mineral and hydrocarbon reserves that require shipping for construction equipment, supplies, and transport of resources to refineries and populated markets. In addition, the main method of re-supply for northern communities is by sea. In both areas, information regarding ice conditions, type, concentration and movement are required.
To address these demands, ice analysis charts, daily ice hazard bulletins, seasonal forecasts, and tactical support for observation are provided. In Canada, the Canadian Ice Service is responsible for acquiring and distributing this information and appropriate products. They also maintain an ice information archive which contains useful data for environmental impact assessments, risk assessment, short-term and seasonal route planning for ships, efficient resource transportation and infrastructure development.
Remote sensing data can be used to identify and map different ice types, locate leads (large navigable cracks in the ice), and monitor ice movement. With current technology, this information can be passed to the client in a very short timeframe from acquisition. Users of this type of information include the Coast Guard, port authorities, commercial shipping and fishing industries, ship builders, resource managers (oil and gas / mining), infrastructure construction companies and environmental consultants, marine insurance agents, scientists, and commercial tour operators.
Examples of sea ice information and applications include:
"...GPS = Good Protection Sidekick..."
Accidents like the sinking of the Titanic are virtually eliminated now, with iceberg reconnaissance (provided by the International Ice Patrol) and GPS navigation onboard ships. And even if a ship did collide with an iceberg, search and rescue operations using remote sensing and GPS navigation could save many lives in such an incident.
Ships navigating through high latitude seas (both northern and southern) are often faced with obstacles of pack ice and moving ice floes. Ice breakers are designed to facilitate travel in these areas, but they require knowledge about the most efficient and effective route through the ice. It is important to know the extent of the ice, what type of ice it is, and the concentration and distribution of each type. This information is also valuable for offshore exploration and construction activities, as well as coastal development planning.
Ice isn't simply ice!
Sea ice isn't a uniform, homogeneous unit. What appears to be a single cover of ice can vary in roughness, strength, salinity, and thickness. Pack ice and ice floes consist of assemblages of different ice types patchworked together, intersected by dynamic leads or cracks. Ice is usually defined by its age - either as new, first-year or multi-year ice. New ice is smooth and relatively thin (5-30 cm)and provides the least resistance to ice breakers . First year ice is older and thicker than new ice (30-200cm) and can pose a significant hazard to all vessels, including icebreakers. When deformed into rubble fields and ridges, first year ice types can become impassable. Ice that survives into a second and later years, generally becomes thicker (>2m) and declines in salinity, increasing the internal strength. This ice is a dangerous hazard to ships and off-shore structures. Ice charts are maps of different ice types and concentration of ice, which are distributed to those working in marine environments where ice affects their operations.
Why remote sensing?
Observing ice conditions is best from a ground perspective, but this doesn't allow for determining the extent or distribution of the ice. Remote sensing from airborne or spaceborne sensors provides this very valuable view. The areas of ice can be easily mapped from an image, and when georeferenced, provide a useful information source. Remote sensing technology is capable of providing enough information for an analyst to identify ice type (and thus infer ice thickness), and from this data, ice charts can be created and distributed to those who require the information.
Active radar is an excellent sensor to observe ice conditions because the microwave energy and imaging geometry combines to provide measures of both surface and internal characteristics. Backscatter is influenced by dielectric properties of the ice (in turn dependent on salinity and temperature), surface factors (roughness, snow cover) and internal geometry / microstructure. Surface texture is the main contributor to the radar backscatter and it is this characteristic which is used to infer ice age and thickness. New ice tends to have a low return and therefore dark appearance on the imagery due to the specular reflection of incident energy off the smooth surface. First year ice can have a wide variety of brightness depending on the degree of roughness caused by ridging and rubbing. Multi-year ice has a bright return due to diffuse scattering from its low salinity, and porous structure.
Coarse resolution optical sensors such as NOAA's AVHRR provide an excellent overview of pack ice extent if atmospheric conditions are optimal (resolution = 1km).
Passive microwave sensing also has a role in sea ice applications. Objects (including people!) emit small amounts of microwave radiation, which can be detected by sensors. Sea ice and water emit substantially different amounts of radiation, so it is relatively easy to delineate the interface between the two. The SSM/I onboard the shuttle collected data in this manner. The main drawback of passive microwave sensors is their poor spatial resolution (approx. 25km) which is too coarse for tactical ice navigation.
Ocean ice occurs in extreme latitudes - the high Arctic and Antarctica. But ice also covers prime sea and lake shipping routes in northern countries, particularly Canada, Russia, Japan and northern European and Scandinavian countries. High latitude areas experience low solar illumination conditions in the winter when the ice is at a maximum. This has traditionally hindered remote sensing effectiveness, until the operationalization of radar sensors. The all weather / day - night, capabilities of SAR systems, makes radar remote sensing the most useful for ice type and concentration mapping.
To provide sufficient information for navigation purposes, the data must be captured frequently and must be processed and ready for use within a very short time frame. High resolution data covering 1-50 km is useful for immediate ship navigation, whereas coarse resolution (1-50km), large area coverage (100 - 2000km²) images are more useful for regional strategic route planning. For navigation purposes, the value of this information has a limited time window. However, for playing a role in increasing our knowledge about climate dynamics and ice as an indicator of global climate change, the data has long term value.
RADARSAT has orbital parameters and a radar sensor designed to address the demands of the ice applications community. The Arctic area is covered once a day by RADARSAT and systems are in place to efficiently download the data direct from the ground processing station right to the vessel requiring the information, in a time frame of four hours. Airborne radar sensors are also useful for targeting specific areas and providing high resolution imagery unavailable from commercial spaceborne systems. Airborne radar is more expensive but has the benefit of directly targetting the area of interest, which may be important for time critical information, such as tactical navigation in dynamic ice. Winter is the preferred season for acquiring radar scenes for ice typing. Melting and wet conditions reduce the contrast between ice types which makes information extraction more difficult.
Future remote sensing devices are planned to provide comprehensive measurements of sea ice extent.
"...I like my eggs on ice..."
Creating an Ice Chart
The Canadian Ice Service of Environment Canada (CISEC) creates charts for ice type that are distributed to their clients on a near-real time basis. These charts are essentially ice maps with Egg Codes superimposed, which explain the development stage (thickness), size, and concentration of ice at both regional and site specific scales. The codes used to represent the ice information are displayed in an oval symbol, resembling an egg, hence the term Egg Code . Egg codes are used not only for sea ice, but also lake ice. Also they conform to the WMO (World Meteorological Organization) standards.
Once you understand the meaning of the various codes, the interpretation of the ice charts is relatively easy.
For more detailed information about the coding procedure and terminology, go to the Canadian Ice Service homepage.
Case study (example)
RADARSAT Expedites Expedition to the Magnetic North Pole!
In March of 1996, teams of Arctic adventurers set out on an expedition to reach the magnetic North Pole, located on the west coast of Ellef Ringnes Island, in Canada's high Arctic. Travelling across sea ice by ski, the teams required a route on smooth first year ice in order to haul their gear and conserve energy. Ice blocks, rubble and irregular relief made deformed and multi-year ice virtually impassable. One team relied on remote sensing - image maps created from RADARSAT data - to plan their route.
The ScanSAR image covered the entire extent of the route, from Resolute Bay on Cornwallis Island to the pole (78°6'N, 104°3'W). The resolution of 100m provided information about the ice cover and type, and mapped coastlines were added following geometric processing, to provide a geographic reference. The team was also equipped with GPS and communication technologies.
On the image map, passable ice appears uniformly dark, due to the specular reflection of incident radiation from the radar on the smooth surface. Rubbly, rough ice that often contained enough relief to make skiing impossible appears bright, due to the reflection of the radar energy back to the sensor.
The team using RADARSAT image maps was the only one to complete their journey to the magnetic North Pole. The other teams were hindered by rough ice and could not efficiently plan their route without the synoptic view provided by remote sensing. RADARSAT, with its sensitivity to ice type, far northern coverage, and reliable imaging was the most suitable sensor for this type of application. Its success bodes well for future exploration endeavors!
Reference: Lasserre, M., 1996. RADARSAT Image Maps Make Arctic Expedition a Success, Remote Sensing in Canada, Vol. 24, No. 1, June, 1996. Natural Resources Canada.
Ice moves quickly and sometimes unpredictably in response to ocean currents and wind. Ice floes can move like tectonic plates, sometimes breaking apart like a rift valley or colliding in a style similar to the Indian and Asian plates, creating a smaller version of the Himalayan Mountains - a series of ridges and blocky ice rubble. Vessels can be trapped or damaged by the pressure resulting from these moving ice floes. Even offshore structures can be damaged by the strength and momentum of moving ice. For these reasons it is important to understand the ice dynamics in areas of construction or in the vicinity of a shipping/fishing route.
Why remote sensing?
Remote sensing gives a tangible measure of direction and rate of ice movement through mapping and change detection techniques. Ice floes actually have individual morphological characteristics (shape, structures) that allow them to be distinguished from one another. The floes can be mapped and their movement monitored to facilitate in planning optimum shipping routes, to predict the effect of ice movement on standing structures (bridges, platforms).Users of this type of information include the shipping, fishing, and tourism industries, as well as engineers involved in offshore platform and bridge design and maintenance.
Monitoring of ice movement requires frequent and reliable imaging. The revisit interval must be frequent enough to follow identifiable features before tracking becomes difficult due to excessive movement or change in appearance. Active microwave sensing (radar) provides a reliable source of imaging under all weather and illumination conditions. RADARSAT provides this type of sensor and is a spaceborne platform, which is advantageous for routine imaging operations. The orbital path ensures that Arctic areas are covered daily which meets the requirement for frequent imaging.
The resolution and imaging frequency requirements for ice motion tracking vary with the size of floes and the ice dynamics in a region. In areas of large slow moving floes (e.g. Beaufort Sea), 1km resolution data over 10 day intervals is adequate. In dynamic marginal ice zones (e.g. Gulf of St. Lawrence), 100m resolution data over 12-24 hr intervals is required.
Case study (example)
The significance of the force and potential effect of ice movement was brought to light recently with the design and construction of the Confederation Bridge, a 13km link from Prince Edward Island, in Canada's Maritimes, across Northumberland Strait to New Brunswick on Canada's mainland. Crossing a strait that endures ice floes moving in response to winds, currents and tides through a narrow arm of the Gulf of St. Lawrence, the bridge will have to withstand tremendous forces from moving ice impacting its supports.
Ice floes in Northumberland Strait are dynamic due to oceanic and atmospheric forces, yet constricted in their movement. The result is compression collisions creating large rubbly ice masses that extend vertically above and below the water level up to 20 m1 (each direction). These ice masses have the potential of critically damaging any structure impeding its movement back and forth in the strait. The design and engineering of the bridge had to take into account both the thickness and actual constant movement of the ice. Ice information archived at the Canadian Ice Service contributed to the understanding of the ice dynamics in the strait, and its tensile properties, critical for setting engineering parameters.
During construction, a radar image of the bridge site was obtained to observe the impact of the bridge supports on the flow of ice around the site. Due to the design of the supports, which are cone-shaped at the waterline to help bend and break the ice, the ice cracked and flowed around the supports. This is one image where ice movement can be inferred from a single image and does not require multi-temporal scenes. In the image, the ice can be seen flowing from bottom to the top with the wakes of rubble created by the bridge supports clearly visible.
Remote sensing will be used to monitor the effect of the bridge on the ice movement and ensure that ice build up isn't occurring beyond expectations. As exemplified in the image, the bridge will have an impact on the ice dynamics, by breaking up large floes into smaller pieces which may accumulate on the shore in piles. This effect will be monitored, as will any subsequent effects on microclimate, which might affect the agriculture or fishing industries of PEI1.