Home >> July 2011 Edition >> Focus On EO + Imagery – The DMC Grows Up: David Hodgson
Focus On EO + Imagery – The DMC Grows Up: David Hodgson
Managing Director of DMCii

Surrey Satellite Technology Ltd’s Disaster Monitoring Constellation is a fleet of small low-cost Earth observation satellites, but despite — or because of — that fact, it has undergone rapid evolution in the last decade, a process set to accelerate further in the next few years.

‘The perfect is the enemy of the good’ is a saying that should apply to space as much as anywhere else. Except that designers of standard space missions aim for as close to perfection as they can get, with exhaustively-tested space-qualified parts serving as the basis of multiple-redundant subsystems. The ensuing satellites promise to operate optimally for years on end — but such performance comes at an astronomical cost in terms of mass, complexity and time as well as cash. Not to mention the top-of-the-range launchers needed to fly such heavy payloads (the launch typically amounting to upwards of 40 percent of mission costs).

HodgsonFig1 Is any alternative even feasible for working in space, the most unforgiving environment imaginable? Yes, declares the burgeoning NewSpace community, made up of businesses focused on innovative activities such as lower-cost launch services and space tourism. NewSpacers criticize the traditional space sector’s obsession with maximizing mission performance while failing to control costs — a tendency hardly discouraged by decades worth of cost-plus government procurement contracts.

In the end the only way to grow space markets is slashing the cost of doing business in space. When squeezing out that tricky extra 20 percent performance ends up costing 80 percent of your mission budget, why not be pragmatic, and accept 80 percent of performance for a modest 20 percent pricetag instead?

Of course it’s easy to discount the arguments of unproven NewSpace firms that are, with a few honorable exceptions, years away from turning a profit, if at all. However, 26 years ago, long before the term ‘NewSpace’ was even coined, a small U.K. firm called Surrey Satellite Technology Ltd. (SSTL) was founded with that same pragmatic philosophy in mind. Today, SSTL is the world’s leading small satellite company.

Founder Sir Martin Sweeting realized that the increasing ability of commercial-off-the-shelf (COTS) devices could still offer useful functionality within physically small satellite structures. He led a small team at the University of Surrey to design and build the U.K.’s first amateur satellite: UoSAT-1 launched on October 6, 1981, carrying a reprogrammable onboard computer and a 256x256 pixel CCD to perform Earth observation. A second micro-satellite followed, but funding was tight. SSTL began as a means of helping to finance future small satellite research — a goal it turned out to accomplish in spades. Today, this multi-million pound company employs more than 300 people.

HodgsonFig2 A total of 36 SSTL-built missions have gone into orbit since then, manufactured for customers all over the world. Some things remain the same, however, including the small size of the satellites. This means missions can be developed much more quickly: a typical 100-kg SSTL satellite can be designed, built, tested and made ready for launch in less than two years, compared to a five to 10 year timespan for their standard multi-ton equivalents.

It also makes them much cheaper to build using very much smaller teams, and their performance also benefits greatly from the very latest COTS innovations. SSTL satellites make use of COTS terrestrial technologies to offer much better performance and lower cost than would be the case if only space-specific technologies were used. SSTL has honed its processes and created designs to ensure that the use of COTS does not impact the reliability of the satellites. Of course, the real test is in space, and SSTL’s mission success rate speaks for itself, demonstrating that it is possible to offers both low cost and low risk to customers. As small satellites can be built and flown so much more frequently, technology can be developed on an incremental basis, following the same innovative approach as the microelectronics and software industries. New hardware flight-tested on one mission can be added as standard on the next.

Earth monitoring also remains a core area of interest: Of the 36 SSTL spacecraft flown to date, 27 have been Earth observation satellites, culminating in the Disaster Monitoring Constellation (DMC). The DMC is a constellation of currently six small satellites which operate together to offer wide-area 32m ground sample distance (GSD) coverage of anywhere on Earth within 24 hours, as compared to up to 16 days for Landsat and similar times for comparable, single satellites.

The DMC’s highly-responsive nature means it is often tasked to acquire images of disaster zones for civil protection teams and UN agencies through the International Charter ‘Space and Major Disasters’. A steady stream of images are also acquired for the national owners of the individual DMC satellites and commercial customers. The constellation is operated by DMC International Imaging (DMCii), a subsidiary of SSTL that also commercially markets DMC imagery.

HodgsonFig3 The evolution of the DMC reveals how SSTL’s pragmatic approach works in practice. The project began with a recognition of the need for timely access to low-cost Earth observation data for disaster management, following a call for improved response to natural and man-made disasters at the Third United Nations Conference on the Exploration and Peaceful Uses of Outer Space (UNISPACE III) in 1999. Test missions such as UoSAT-12 and TsingHua-1 proved the concept of acquiring medium-resolution multispectral imagery using small satellites, compatible to Landsat in resolution and spectral bands but acquired much more cheaply.

Employing the company’s 100-kg class workhorse SSTL-100 satellite platform, multiple assets could be flown for the equivalent cost of a single large mission — but delivering more capacity than an individual satellite ever could, the combined constellation offering daily access to all parts of the globe. To further boost the constellation’s reach, the satellites were given an extremely wide 650 km viewing swath-based on double banks of imagers able to acquire complete regions in a single along-track sweep, instead of having to mosaic together many smaller images acquired at separate times. Such wide-area imaging was the DMC’s one essential function — unnecessary capabilities were to be avoided, to control costs.

The low price tag per unit means many different governments could afford to purchase an individual sovereign satellite, gaining technical expertise and hands-on training. In addition, all members of the DMC consortium benefit from shared access to all data acquired by each others’ missions. As there is also a revenue stream from commercial sale of DMC data, the constellation is partly self-sustaining, helping to underpin its continued existence. This is in marked contrast to many government-run Earth observation systems, where future data continuity cannot be assured (even the venerable Landsat program risks data gaps due to a repeated delays in completing its follow-up), making DMC data increasingly attractive to long-term users.

The first DMC satellite reached orbit in 2002, with three more launched the following year. In December 2004 the constellation demonstrated its usefulness in dramatic fashion by performing rapid damage mapping across all the far-flung nations affected by the Asian tsunami. The DMC has also developed incrementally along the way. When China’s Beijing-1 joined the constellation in 2005, this larger SSTL 150 satellite included an additional high-resolution (4m GSD) panchromatic imager to zoom in on areas of interest.

HodgsonFig4 The 2009-launched U.K.-DMC-2 and Deimos-1 satellites were based on improved versions of the standard 100-kg class platform, known as SSTL 100 v2.1. Improved technology gave them enhanced imaging resolution of 22m GSD, effectively doubling the number of pixels they acquire per hectare. Improved onboard memory, power generation and data compression techniques mean these satellites can store an order of magnitude more imagery aboard and downlink data 10 times faster to the ground.

These ‘second generation’ DMC satellites will be joined this July by the 100-kg-class 22m GSD NigeriaSat-X and the very-high resolution 2.5-m GSD NigeriaSat-2, the latter’s enhanced vision necessitating an enlarged 300-kg class SSTL 300 satellite platform. The issue here is not only aperture size but the precision attitude control required to point the imager at each required 20x20 km area target — performing rapid roll maneuvers of up to 45 degrees while also compensating for the rotation of Earth’s surface.

To make the satellite sufficiently agile for its ambitious variety of imaging modes (including stereo imaging and artificial image widening based on maneuvering), NigeriaSat-2 has no deployable sections, such as extended antennas or solar wings, to maximize its stability and does without liquid propulsion altogether, which avoids sloshing effects. Instead, an attitude and orbit control subsystem based on momentum wheels overseen by a combination of star trackers, GPS sensors and compact MEMS-based gyro devices derived from terrestrial automobile stability control systems is used.

In SSTL terms, NigeriaSat-2 is something of an oversized outlier. However, with the majority of DMC missions yet to come, most will retain the heritage SSTL 100 platform. SSTL continues to harness the ongoing advances of the terrestrial electronics industry, so that SSTL 100 v2.2 will offer a further step change in performance. It will have the same 22m GSD imaging resolution, but a combination of extra onboard memory, faster downlinks and sophisticated data compression techniques mean this next generation will be an ‘always-on’ satellite capable of generating 62 million km2 of image data daily. Every five days a single ‘EarthMapper’ spacecraft will image the entire land surface of the globe, or a constellation of five such spacecraft can provide global daily coverage.

Satellite tasking procedures will become redundant, simplifying satellite operations. EarthMapper will be able to capture dynamic environmental phenomena such as crop growth, deforestation, urban sprawl, floods or disasters more rapidly and comprehensively than ever before. An ‘always-on’ constellation would represent a significant resource for the entire remote sensing community.

Agricultural forecasting requires a minimum of three images per growing season, which EarthMapper will easily deliver. Urban expansion could be monitored as it happens, home by home, while illegal forest logging pinpointed for intervention by ground authorities. In disaster situations, before-and-after maps would chart the damage done within a single day. And other previously impractical applications come into reach: imagine wide-area monitoring of water resources upon national or even continental scales, with weekly checks on the evolving surface area of lakes, rivers, dams and irrigation systems.

HodgsonFig5 EarthMapper will have 32 Gigabytes of onboard storage provided by a pair of High Speed Data Recorders (Flash memory systems as used in smart phones and cameras are also in development, offering much higher levels of non-volatile storage). SSTL engineers are deciding the best trade-off between data compression techniques and the number of ground stations required — the less ‘lossy’ the data the better its quality, but the number of ground stations needed would swell as a consequence, from a minimum of two to a maximum of five, increasing the cost of operations.

Doubling data downlink rates from the current 80 Mbit/s to 160 Mbit/s would allow lossless compression to a single pair of ground stations, although this transmitter would be too power-hungry for the standard SSTL 100 platform. This could be accommodated by switching the satellite’s body-mounted single-junction solar cell panels to more efficient triple-junction cells, so-called because successive layers are tuned to different segments of the overall light spectrum, as well as adding extra solar panels. U.K. DMC-2 has already flown a deployable triple-junction solar panel. The best solution for EarthMapper might well be a combination of high- and low-powered transmitters on the satellite, offering users a choice between lossless and lossy imagery, with a tried and tested system backing up the newer high-power design to reduce mission risk. The first EarthMapper satellite could be flying as early as next year, with a full constellation in operation in the second half of this decade.

The EarthMapper SSTL 100 v2.2 will become the new standard for the DMC, but that standard will, in turn, give way by mid-decade to the SSTL 100 v3, planned to incorporate a new imager that is currently at the prototype stage, offering increased resolution and access to two new spectral bands.

This new design combines the red, green and near infrared (NIR) channels of current DMC missions with the addition of short-wave infrared (SWIR) — capable of identifying the water content of crops and soil — and blue. The visible and NIR bands have 10 – 15m GSD, with SWIR of 20 – 30m GSD, with a swath of 440 km for all bands. The SSTL 100 v3 will continue to provide information to the current users, but will better respond to the needs of some core users such as the agricultural sector, adding important capabilities to the DMC.

Another planned evolution of the basic platform, the SSTL 100 HR would use the same avionics as the EarthMapper to provide 3-m GSD panchromatic-sharpened RGB (red green blue) imagery with a 24 km swath. Any greater resolution than that and the laws of physics kick in to demand a larger aperture. For the next iteration of the DMC, the DMC3, SSTL intends to offer 1m panchromatic resolution imagery with 4-m RGB and NIR using a still larger version of its agile SSTL 300 platform, the SSTL 300 S1, differentiated by a large telescope extending from its body. DMC3 will also follow a new business model, with customers leasing access to satellite capacity.

The past and future evolution of the DMC demonstrates how incremental improvements of ‘fit for purpose’ engineering can give rise to a broadened suite of capabilities. Previous constraints on image resolution and data timeliness of data are both being tackled while preserving the constellation’s low cost and sustainable nature. Beyond that, the fundamental limitations on DMC imagery remain only clouds and darkness. This, too, is being addressed, with a new synthetic aperture radar (SAR) mission in the planning stages.

Power-hungry SAR has traditionally been the province of large, multi-ton satellites such as Canada’s 2.7-ton Radarsat, Germany’s 800 kilogram SAR-Lupe, or the European Space Agency’s 8.2-ton Envisat. Past Russian military SAR missions even required orbital nuclear reactors to make them practically feasible. SSTL, by contrast, has been developing a small satellite SAR to add an all-weather night and day capability to the DMC, offering affordable SAR data to expand its user base, following SSTL’s same approach to its optical missions.

Throughout last year, SSTL worked with its long-term SAR partner EADS Astrium U.K. on a mission and payload design, with airborne trials using a demonstrator being carried out last summer and a prototype payload unit being readied for tests to prepare for a proposed launch as early as 2013. The payload will offer multi-mode SAR imagery at a variety of resolutions and swaths, from 3-6 m at 20-30 km swath up to 30 m at 750 km. It will operate in S-band in multiple polarization’s to add ‘color’ to its images.

SSTL’s NovaSAR-S platform borrows the avionics heritage of the SSTL 300 with a novel wedge-like structure: The 400-kg satellite’s Earth-facing side incorporates a 3x1m radar antenna while its other, Sunward side is covered with triple-junction solar panels. Power constraints mean ‘always-on’ operation is not an option, but the NovaSAR-S will be able to cover upwards of one million km2 daily, making its radar imagery available at a cost comparable to its optical DMC counterparts, with a multi-satellite constellation envisaged.

Potential applications include more frequent monitoring of cloud-mired rain forests, monitoring ships in protected marine zones (vessels appear as bright points in radar images) surveying rice crops and also flooding — the world’s most single most costly type of natural disaster.

By 2017, all these DMC satellites should be operational in orbit, offering customers a complete portfolio of Earth observation imagery at the same traditional low cost. The take-home lessons from the DMC’s success apply to NewSpace and old space alike: building a sustainable business around space is much more an economic challenge than a technical one. Maintain a laser focus on a narrow but marketable field of endeavor while keeping costs low. Innovative business models — such as the DMC’s multi-owner jointly-operated satellite constellation — are at least as important as technical innovations. Incremental development offers a way to increase capabilities and grow markets, but costs must still be kept in line.

Finally, and most important of all: Offer something useful and inexpensive and customers will gladly pay for it. HodgsonHead

About the author
David Hodgson is the Managing Director of DMC International Imaging Limited. He has worked in satellite related industries for over 19 years and is the Past Chair of the British Association of Remote Sensing Companies (BARSC). He holds a degree in computing and a Masters degree in business administration. David also serves on the Executive Secretariat of the International Charter, ‘Space & Major Disasters’.