The Polar TEP (Thematic Exploitation Platform) is a public site to which registered users are allowed to upload content and anyone may view or download content. Users are responsible for knowing their rights to distribute information and for ensuring the appropriateness of information that they upload to the site.
Polar View Earth Observation Limited (Polar View) is developing a Polar Thematic Exploitation Platform (Polar TEP) for the European Space Agency (ESA). Polar TEP will provide polar researchers with access to computing resources, earth observation (EO) and other data, and software tools in the cloud. This new approach will remove the need to transfer large EO data sets around the world, while increasing the analytical power available to researchers and operational service providers.
In the traditional workflow for the analysis of earth observation (EO) data, users download the data to their local site and then process it using their available software and computing resources. With the increasing volume of data available from missions such as Sentinel, and the resulting need for powerful computing resources for processing, the existing methods of working are inefficient and restrict the use of EO data.
ESA’s Thematic Exploitation Platform (TEP) concept aims to provide a working environment where users can access algorithms and data remotely, providing them with computing resources and tools that they might not otherwise have, and avoiding the need to download and store large volumes of data. This new way of working will encourage wider exploitation of EO data.
EO is especially import in the polar regions at a time when climate change is having a profound impact and excitement about new economic opportunities is driving increased attention and traffic, resulting in concerns about the state of the region’s delicate ecosystems. Developing tools to model, understand and monitor these changes is vitally important in order to better predict and mitigate the resulting global economic and environmental consequences.
Polar View Earth Observation Limited is developing a TEP for ESA that will be customized for polar research and operational needs. Polar View brings a team to this project that combines a unique mix of polar domain experts, scientists, operational service providers, ICT experts, and user representatives. Polar TEP will integrate data discovery and access for a rich set of polar themed EO and complimentary datasets, a scalable computing environment, a suite of analytical tools and the ability for users to supply their own models, plus tools to allow sharing of results and promote collaboration. These resources will be accessed through a web portal.
The platform architecture will be open, scalable and independent of any specific IT infrastructure. This approach will allow easy expansion of the platform’s capabilities, and encourage the development of a wider network of other thematic exploitation platforms.
The polar regions are vast and have a growing significance for the entire globe. They play an important role in regulating and driving the global climate, but are also the regions experiencing the fastest increases in temperature associated with global climate change. Monitoring and understanding these changes is vitally important for everyone, not only the indigenous populations, since these changes will have global effects.
In the coming decades and as global climate change progresses, the world’s polar regions will become significantly more important. As a source of natural resources and current low population density, the Arctic is of increasing interest to politicians and industry. Global interest is fueled by environmental concerns for the delicate ecosystems and by excitement over perceived abundance of oil and gas. Where the Antarctic Treaty currently protects the southern polar regions from some of this attention, nevertheless there are still real pressures to ensure the Antarctic environment remains protected from increasing development.
Observations from space provide unique information which is essential for the successful understanding and management of climate change. The polar regions are remote and hostile environments where efforts to collect required observations and data are limited by very real constraints such as the weather, lack of infrastructure and long periods of polar darkness during the winters. As a consequence satellite platforms provide the only source of consistent, repeatable, regional scale, calibrated, year-round data of the polar regions. The large number of EO satellites observing various aspects of the polar regions provide a comprehensive monitoring system for the maritime and land cryosphere.
As shown in the figure, the Polar TEP platform will include:
Antarctic ice shelves play a vital role in the global climate, producing significant water masses that drive ocean circulation and buttressing the rate at which inland ice sheets drain into the ocean. A recent study [RD06] suggests that the Filchner Ice Shelf region in Antarctica might soon experience a wholesale change in oceanographic regime with potentially dramatic consequences for the ice shelf and its tributary ice streams.
A newly initiated large-scale study will use glaciological and oceanographic measurements to improve and validate a set of regional models to predict the regions contribution to 21st Century sea level rise. The lack of the ability to predict these changes was highlighted in the latest IPCC report [RD07] noting the risk of “Abrupt and irreversible ice loss from a potential instability of marine-based sectors of the Antarctic Ice Sheet”. The observational inputs to this study are wide ranging, but will include satellite observations of ice stream surface velocities to be validated by in situ GPS units. Satellite derived data provide important details of spatial and temporal variation in ice sheet and ice stream velocity.
[Figure 4: Filchner Ice Shelf study area showing surface ice sheet and ice stream velocities.]
Derivation of ice sheet and ice stream surface velocities is a computationally intensive process that requires large volumes of satellite SAR imagery. To cover the required area and time periods, data from a number of different satellite sensors will be involved. This raises the issue of inter-comparison of velocity data derived from SAR sensors with different characteristics (e.g., how do velocities derived from C-band Radarsat2 data compare with data from X-band TerraSAR-X data?) and using different techniques (e.g., how do velocities derived from interferometry compare with velocities from feature tracking?). The P-TEP will provide capabilities to answer a number of the issues involved.
National ice centres throughout the world have mandates from their respective governments to monitor and chart sea ice and iceberg conditions in their particular areas of interest. This is a significant activity and involves the manual interpretation of large volumes of satellite data for ice chart production, which is labour intensive, slow and lacks consistency between ice centres as well as inside the same ice centre.
With the large volume of Sentinel-1 SAR data expected from early 2015 onwards, a more automated and rigorous analysis method that allows the combination of SAR, microwave radiometer data, infrared data, scatterometer data and visible data, as well as state of the art numerical weather prediction data is needed.
Current effort is directed at collation of data from multiple sources and (building on the previous development of satellite data assimilation in an ice analysis system) establishing an automated ice analysis system with the aim of being able to produce a regularly updated Arctic wide ice analysis every six hours.
The data assimilation system will build on previous ESA efforts such as SAR Ice constellation (STSE), Sea-ice CCI, SMOSICE etc, since radiative transfer models and backscatter models developed in these projects will be key elements in quantitatively relating satellite observations to surface ice conditions.
[Figure 3: Greenland ice chart produced by DMI.]
Development and operations of such a large scale and complex data assimilation system will depend on established access to all required datasets and a programming environment within which to develop, deploy and operate the necessary algorithms. The storage and processing requirements required for this task are significant and most readily provided in close proximity to the required data resources.
In addition production of a regular ice analysis will require regular validation using independent data, which will be more easily achieved within the established processing chain given access to the necessary observation data.
Polar View, originally developed with funding from ESA and now operating independently, aims to continue as the leading organisation for operational satellite-based monitoring of the polar environments and cryosphere. Through many years of service delivery, Polar View services have demonstrated a significant impact on operations safety, environmental protection and sustainable development to a large and diverse group of end users.
Polar View services provide accurate, real time cost-saving information in a number of areas including glaciers, ice edge monitoring, iceberg & sea ice monitoring, lake & river ice and snow cover. These services provide valuable information supporting sustainable economic development, mitigating the risk of accidents in harsh polar environment and informing efforts to monitor and protect the polar ecosystems.
More recently through European funded projects including ICEMAR and Polar Ice, significant improvements are being made to develop new sea ice information resources and to improve the delivery of information to ships operating in the polar regions.
A facility such as the proposed Polar TEP has significant advantages for improved service delivery and development of new services. Operational services rely on stable and reliable IT resources to ensure high levels of service performance.
Polar View near real-time information services use a large volumes of satellite imagery that needs to be accessed, processed and delivered as quickly as possible. The Polar TEP will provide fast access to the Sentinel-1 data stream and access to a range of other data resources. Importantly these data will be made available together with processing capacity. This environment will allow service providers to establish services for end users and improve their reliability and performance.
The Polar TEP environment also establishes an environment for development and operation of new information services, making it a key component for supporting commercial and open information services and their end users.
The ESA CCI Sea Ice project will combine and extend ongoing research to develop improved and validated time series of ice concentration and ice thickness for use in climate research. Since sea ice is a sensitive climate indicator with large seasonal and regional variability, the climate research community requires long-term and regular observations of these key ice parameters in both the Arctic and Antarctic. The ESA CCI Sea Ice project will deliver global data sets to support climate research and monitoring according to the GCOS requirements for generation of satellite-based data sets and products.
Users of sea ice thickness estimates also include commercial players planning ship and offshore operations in ice covered waters. As ice thickness influences loads on structures, it is much sought after during infrastructure design phases and when planning operations.
Advances in the processing of ERS and Envisat radar altimeter data has though shown that these instruments can be used to calculate low-resolution sea ice thickness data from measurements of ice freeboard. In addition to ERS and Envisat, in CCI Phase 2 estimates will also be derived from Cryosat-2 data.
Operational processing chains will benefit from easier access to data, combined with access to scalable processing resources.
The Polar TEP is of special interest to the sea ice thickness CCI since the European radar altimeter data is an input both for ice sheet elevation change and sea ice thickness retrieval. Some of the processing steps are identical or easily adapted from one application to another. In addition functionality for sea ice related altimeter data processing (such as sea ice dedicated re-trackers and ice type detection from waveform shape) may also be implemented. Also the iceberg detection from radar altimeter data is based on the same input data and waveform characterisation as sea ice thickness retrieval as the “iceberg echoes” are recognised and discarded from the SIT processing.
Computationally, the heaviest part of the altimeter to SIT chain is the so called “retracking”. A retracker fits a mathematical model to the waveform, and using this model calculates the range representing the target surface.
Lastly, the developers of altimeter sea ice thickness retrieval algorithms would benefit from the TEP. Currently, researchers tend to build and test their tools on a local computer with data downloaded on a local disk. Having a TEP that would allow the testing of their modules (for example retrackers or freeboard to thickness conversion) jointly with other researchers, which would allow the researchers to share and demonstrate their ideas and novel tools within their research community much faster than is currently possible.
Furthermore the Polar TEP should include visualisation and data export tools for sea ice thickness. These will include averaging (both spatial and temporal) of processor outputs.
