1.INTRODUCTIONIn the development of new hyperspectral technologies, the observation of our Earth and its atmosphere has become one of our focal points. Reliable information is needed to determine its state, analyze its utilization and finally draw conclusions from it concerning the environment’s and our well-being. Recorded data is best made available shortly after its detection or in real-time. This is for example crucial for early warnings and reactions in disaster applications like floods and wild fire. The demand on information, retrieved from hyperspectral Earth observation, is rapidly increasing in the last 20 years and drives the advancement of related instrument technologies. Miniaturization, as well as higher spatial, temporal and spectral resolution are in pursuit of current technology studies and developments. This is enabled by progress in various relevant technology fields, as electronics miniaturization, new CMOS sensor developments, new materials and processes, hyperspectral filters, free-form optical design, free-form manufacturing and testing capabilities, on-board data processors and deep learning algorithms for image enhancement. Simultaneously, the reduction of cost for instruments and information provision is desired to feed and explore new data usage ideas and application markets. In addition to the already launched or currently developed hyperspectral instruments in the frame of ESA missions, also smaller activities have an impact on the hyperspectral technology evolution. In the frame of the ESA General Support Technology Programme (GSTP), Basic Research Technology Programme (TRP) and Announcement of Opportunities (AO), several compact hyperspectral instruments already have been and are being developed. These R&D activity programs support small development activities like CHIMA, where the realization of a new free-form grating based, high resolution instrument breadboard for hyperspectral atmosphere sensing applications was designed, built and tested, and also include instrument developments reaching flight-readiness maturity. One example is the compact, 6-unit CubeSat compatible HyperScout instrument, providing a wide swath for natural disaster monitoring applications, which was launched in 2018 for hardware and processing functionality testing. 2.APPLICATIONS AND DESIGN DRIVERSHyperspectral instruments in remote sensing have a wide range of applications, which is ever growing. For about 20 years, when the first hyperspectral remote sensing instruments started collecting data, the number of publications steadily increased. In Transon’s survey on hyperspectral Earth observation applications1, a ranking for the application areas of most interest is given by the distribution of number of articles found with the Scopus platform. With 31%, Natural Vegetation takes the first rank, then geology with 18%, water resources with 15%, agriculture with 14%, land cover with 9%, disaster with 5% and last, followed by the urban areas with 4%1. The growing interest in information obtained by hyperspectral data also opened new markets to data processing companies. Among the service areas that are enabled or enhanced by hyperspectral imaging is the displacement category, under which landslide, subsidence and infrastructure stability is covered. Further the marine service area, comprising sea water monitoring, algae bloom and fishing area. Examples for land services would be land cover, wildfire, illegal crops and soil sealing. And also urban change, infrastructure building and urban heat island are categories of urban dynamics and can be obtained as a service. Potential application areas could be listed almost indefinitely. The demand is obvious and the need for significant high-quality data from cost-effective instruments conflicts with high cost, large instrument applications. Technologies, data processing and applications stand here in close relation and are driven by each other. Each of them functions as a receiving and also as suggesting force in the current dynamic of hyperspectral data acquisition. A clear trend goes towards cost-effective, miniaturized and high performant instruments. Miniaturization with steady performance does not happen without technology advancement, hence many upcoming technologies support exactly this path. A good number of technologies and instrument concepts have already been developed in the frame of ESA’s R&D activities. However, high performance instruments enable applications out of reach for compact systems and may uncover uses that were not anticipated beforehand. Spatial and temporal resolution, spectral resolution and signal-to-noise ratio are key parameters to perform trade-offs that determine cost and complexity of the instrument. In case of compact instruments, the trade-off is done by constraining the design in the mass, volume and power available on a picosatellite or a smallsat. Therefore, the development consisted in identifying technologies that make compact hyperspectrals achieving acceptable performance and innovative functionalities, such as on-board real-time processing. For what concerns large instruments, the development is geared to implement a technology step to match the challenging requirements of the Copernicus Space Component Expansion programme. 3.INSTRUMENT DEVELOPMENTS3.1Launched hyperspectral instrumentsESA’s history of bringing hyperspectral instruments into orbit starts in 2001, when CHRIS, an imaging spectrometer, on board of the PROBA-1 space platform, was launched to measure bidirectional reflectance distribution function data for a better understanding of spectral reflectance from Earth’s vegetation. Since then, more instruments with hyperspectral capabilities have been launched for the observation of Earth’s land and atmosphere. SCIAMACHY, two sounders for nadir and limb atmospheric observation have been launched on Envisat the year after. In 2004, OMI on board of Aura was also launched for monitoring of ozone and other trace gases. GOME-2 on a MetOp satellite (2006) and Tropomi on Sentinel-5P (2017) have also been launched to observe the atmospheric chemistry with the detection of several non-continuous hyperspectral bands. With the ocean color, vegetation and aerosols in focus, the instrument OLCI was launched in 2016 on Sentinel-3, providing a partly continuous selection of hyperspectral bands. Finally, in 2019 the PRISMA instrument was launched on board of a mission with the same name, mainly focused on in-orbit technology demonstration, and for further monitoring of natural resources and the atmosphere. Some of these instruments were already achieving spectral resolutions of down to 0.22 nm. Table 1:Launched hyperspectral instruments2,3,4,5,6,7,8,9. Instrument | CHRIS | SCIAMACHY | OMI | GOME-2 |
---|
Mission/Satellite | PROBA-1 | Envisat | Aura | MetOp | Organization | Sira Technology Ltd | Dornier Satellitensysteme GMBH, Fokker Space | KNMI, NIVR, FMI, TNO, VTT | Leonardo | Purpose | Land, vegetation, ocean, atmosphere | Limb / nadir for atmospheric chemistry | Monitor ozone and other trace gases | Atmospheric chemistry, ozone, climate | Launch date | 2001 | 2002 | 2004 | 2006 | Altitude | 615 km | 774 km | 705 km | 827 km | Swath | 14 km | 960km (nadir) | 2600 km | 960 km (main channels) | Design | Prism spectrometer | Grating spectrometer | Grating spectrometer | Grating spectrometer | Spectralbands/range | 410 - 1050 nm18/63 bands | 214 - 2386 nm8 bands (non-continuous) | 270- 500 nm3 bands | 240 - 790 nm 4 bands (non-continuous) | Spectral resolution | 1.3 - 1 2 nm | 0.22 - 1.48 nm | 0.5 nm | 0.24 – 0.53 nm | Spatial resolution | 18 m/36 m | 3 km vertical (limb) 30/60 km (nadirALT/ACT) | 13 km/24 km | 40 km (main channels) | Temporal resolution | 1 week orbital repeat cycle | 3 day global coverage | Daily global coverage | 1.5-3 days global coverage | Weight | 14 kg | 198 kg | 65 Kg | 73 kg | Power | 8 W | 122 W | 66 W | 42 W |
Instrument | OLCI | Tropomi | PRIMSA |
---|
Mission/Satellite | Sentinel-3 | Sentinel-5P | PRISMA | Organization | TAS | Airbus, TNO, SRON, KNMI | Selex Galileo | Purpose | Ocean and land color, vegetation aerosol, clouds | Atmospheric chemistry, air quality | Technology demonstration, and to monitor natural resources and atmosphere | Launch date | 2016 | 2017 | 2019 | Altitude | 810 km | 824 km | 614 km | Swath | 1270 km | 2600 km | 30 km | Design | Grating spectrometer | Grating spectrometer | Prism spectrometer | Spectral bands/range | 400 - 1020 nm 21 bands (partly continuous) | 270 - 2385 nm 3 bands (non-continuous) | 400-2500 nm | Spectral resolution | 2.5 - 40 nm | 0.25 - 0.55 nm | 12 nm | Spatial resolution | 300 m | 7 km | 30 m | Temporal resolution | 2 days global coverage | Daily global coverage | TBC | Weight | 150 kg | 220 kg | 90 Kg | Power | 122 W | 170 W | 110 W |
Some the fore-mentioned hyperspectral devices, due to their success, had been used in following-up missions, as for example the GOME-2 sent initially in 2006 on board of the MetOp-A, but also later used with a similar design in the MetOp-B (2012) and the MetOp-C (2018) missions. 3.2Hyperspectral in developmentSome of the hyperspectral instruments in development are the UVN/Sentinel-4 and UVNS/Sentinel-5, dedicated to atmospheric chemistry. The former is designed to operate from geostationary orbit, while latter on sunsynchronous low Earth orbit. They are scheduled for launch for the coming few years. Further, the Floris instrument for vegetation fluorescence mapping will be launched in 2024 in the frame of the Fluorescence Explorer (FLEX) mission. Just recently, the phase B2 has been initiated for a Copernicus candidate mission for a potential Next-Generation Sentinel satellite. The goal of this provisioned instrument CHIME is to be useful for agriculture and the exploration of natural resources. This includes for example crop health assessment, species identification, soil property and mineral mapping. In order to provide this kind of information a spatial resolution of 20 m to 30 m and a spectral resolution of 10 nm is requested for a continuous spectrum recording from 400 nm to 2500 nm. This, in combination with a very high signal-to-noise ratio performance and a rather short revisit time, states a considerable challenge for the instrument designer and will drive new technology developments. In addition to CHIME, other hyperspectral missions are proposed, such as the Copernicus candidate mission, named CO2M, with hyperspectral sensing capabilities for atmospheric CO2 detection, and TRUTHS, aiming at establishing an SI-traceable space-based climate and calibration observing system to improve confidence in climate-change forecasts (currently in Phase A). Table 2:Hyperspectral instruments in development phase10,11,12,13,14,15,16,17,. Instrument | UVN | UVNS | Floris | CHIME | CO2M | HIS |
---|
Mission/Satellite | Sentinel-4 on MTG | Sentinel-5 on MetOp SG | Earth explorer missions / FLEX | CSC Expansion study for NG Sentinels | CSC Expansion study for NG Sentinels | Earth explorer / TRUTHS | Organization | Airbus | Airbus | Leonardo | TAS, OHB, Leonardo | TBC | Airbus | Purpose | Atmospheric chemistry | Atmospheric chemistry | Vegetation fluorescence mapping | Agriculture, raw materials, environment | Atmospheric CO2 detection | SI-traceable space climate and calibration system | Status | Launch planned 2023 | Launch planned 2022 | Launch planned 2022 | Phase B2 | Phase B2 | Phase A | Altitude | 35786 km | 835 km | 800 km | 632 km | TBC | 600 km | Swath | 8 km | 2715 km | 150 km | 128 km | ~200 km | 100 km | Design | Grating spectrometer | Grating spectrometer | Grating spectrometer | TBC | TBC | TBC | Spectralbands/range | 305 -775 nm3 hyperspectral bands | 270 - 2385 nm7 hyperspectral bands | 500 - 780 nm9 hyperspectral bands | 400 - 2500 nm | 747 - 2095 nm3 hyperspectral bands | 320-2400 nm | Spectral resolution | 0.12 - 0.5 nm | 0.25 -1 nm | 0.1 – 2 nm | 10 nm | 0.1 - 0.55 nm | TBC | Spatial resolution | 8 km | 7.5 km | 300 m | 20-30 m | 2 km | TBC | Temporal resolution | 60 min | Repeat cycle 29 days | 30 days | 10-15 days | ~3 days | TBC | Weight | 150 kg | 250 kg | TBC | 1640 kg | TBC | TBC | Power | 180 W | 220 W | TBC | TBC | TBC | TBC |
ESA also contributes to the development of novel innovative technologies in the field of optical design, manufacturing, materials, hyperspectral filters, CMOS detectors, electronics miniaturization and on-board data processing capabilities. These in turn enable a more compact design of such instruments and enable their accommodation on smaller platforms. Since several years these developments go towards miniaturization, cost-effectiveness and short revisit, while maintaining a high performance. 3.3HyperScout 1&2One of the instruments, developed in the frame of an ESA In-Orbit Demonstration GSTP activity, is the VNIR (400 nm to 1000 nm) hyperspectral instrument HyperScout. It has a large field of view (31° across track), a spectral resolution of about 15 nm and it takes advantage of a hyperspectral linear variable filter technology in combination with a CMOS detector. The wide field three-mirror anastigmat (TMA) telescope designing and manufacturing capabilities, in combination with an overall athermal design and the use of COTS-based miniaturized electronics contribute to its compact size and its ease of alignment. In addition, intelligent on-board processing shortens the time from data acquisition to the access of relevant information widening the scope of land and vegetation operational applications. In 2018 the HyperScout Proto-Flight Model was launched on a 6-unit Gomx-4B CubeSat in order to test the hardware and processing functionalities and is now operational with a swath of 220 km and a ground sampling distance of 70 m at 500 km altitude. Since any CubeSat platform has limited downlink capabilities, one of the main functionalities of this mission was to prove the L0-L2A on-board processing, to cope with the 1 TB hyperspectral data generation per orbit. Only application-specific high level processed hyperspectral cubes are therefore downloaded to ground. The company’s future target is to achieve a higher revisit with a satellite constellation18,19. HyperScout-1 has been proven operative well beyond its commissioning phase. Last image was taken in Spring 2020. In the recent years, cosine implemented the concept by adding a Thernal Infrared channel, and the device was updated also with enhanced Artificial Intelligence capabilities. The resulting Hyperscout-2 was launched in 2020 and is being successfully operating since20,21. Figure 1:Left: HyperScout-2 flight model. Right: HyperScout-2 design [credit cosine]22. 3.4STREEGO & HYPERSTREEGOAnother instrument, developed during an ESA R&D activity, is STREEGO. The main focus was to develop an innovative optical design with specifications that cover specific remote sensing product market needs. The result was a smallsat compatible multispectral imager with upgrade capabilities for hyperspectral performance. The optics consist of an unobscured three-mirror anastigmat, which can achieve hyperspectral performance in combination with integrated linear variable filters and newly developed sensors, such as a 12 Mpx CMOS detector with 5.5 μm pixel size. With this composition, up to 150 spectral bands over a spectral range of 430 nm to 880 nm can be provided. With a 200 mm aperture diameter and a field of view of 2° (F-number 6), a spatial resolution of 5.5 m to 11 m can be achieved, depending on the number of selected spectral bands. Further, the design is athermal due to the utilization of CTE (coefficient of thermal expansion) matched materials for mirrors and structure, which simplifies the platform thermal control. The identified data products for STREEGO are related to agriculture, forestry, resources monitoring and urban development. Currently, the instrument is at Proto-Flight Model maturity level and a 2-satellite constellation could be thinkable in order to reduce the revisit time. Below (Figure 2), the STREEGO instrument is shown22,23. Figure 2:Left: packaged STREEGO instrument; Right: HyperSTREEGO artist’s impression. In white, STREEGO main body; in purple, HyperScout/Panorama [credits Media Lario, cosine]24. A further development is the HyperSTREEGO concept, which uses the combination of the two already mentioned instruments, namely the HyperScout/Panorama (with a pointing ahead and image processing capabilities) and the STREEGO (with multi/hyperspectral capabilities). 3.5ELOISELOIS is a hyperspectral instrument at Engineering Model maturity, targeted for land observation, developed and tested within a General Support Technology Programme. This instrument covers the spectral range from 400 nm to 2450 nm and provides a spatial resolution of 35 m by incorporating three latest technology advances. First, a complex blazed free-form grating manufactured via single point diamond turning. Second, free-form optical design is applied; and third, hyperspectral back-side illuminated CMOS sensors are used, which enhances the instrument sensitivity. Another parameter, improving the instrument’s sensitivity, is the low F-number of 2.1, which aids to maintain a high signal-to-noise ratio. The summary of these technologies enables a smaller, cost-effective instrument with less components and less power consumption. Currently the development of the Qualification Model is ongoing. The non-exhaustive applications are in agriculture, forestry, environment and water monitoring, natural resources and disaster management and coastal monitoring19,25. Figure 3:ELOIS spectrometer breadboard [credits AMOS]. 3.6CHIMAWith ELOIS as precedent activity, the ESA Basic Research Technology Programme CHIMA was initiated to modify the instrument design (Figure 4) for atmospheric chemistry observation purposes. Especially in atmospheric sensing, a high spectral resolution and a high signal-to-noise ratio is needed. Following, the feasibility of a modified Offner-spectrometer with a holographically replicated blazed convex free-form grating of 1000 lp/mm, a curved slit and a keystone corrector were studied. The output was an athermal breadboard, covering the 600 nm to 800 nm spectral range with a spectral resolution of 0.5 nm. During assembly and performance tests, an excellent optical quality, easy alignment and expected spatial and spectral performances were proven19. Figure 4:Left: CHIMA spectrometer breadboard; Right: CHIMA breadboard CAD drawing [credits AMOS, Horiba Jobin-Yvon] 3.7CSIMBAThe Compact Smart spectral Imager for Monitoring Bio-agricultural Areas (CSIMBA) is an instrument suited for a 12-unit CubeSat In-Orbit Demonstration mission. The manufacturing of the Proto-Flight Model is currently in progress. CSIMBA is based on the development achieved with previous development activities based on the design that uses thin film interference filters directly deposited on a 12 Mpixel CMOS 2D detector array (Figure 5, right). The covered spectrum is from 475 nm to 900 nm and with a typical spectral resolution of 5 nm. The instrument provides swath of 80 km and a spatial resolution of 20 m from 500 km satellite orbit. The technology of filters directly deposited on the CMOS detector enables geometric filter designs and pixel precise spectral response control. Each spectral band covers 12 pixels to increase the signal-to-noise ratio. The detector read-out electronics provide a very high frame rate for image acquisition. The system will be provided with on-board data processing for real time compression. Further possibilities of image processing are planned. The long-term goal is to build up a constellation of hyperspectral nanosats, capable of enhanced vegetation monitoring of agricultural fields and the biodiversity status with an hourly revisit rate19, 26. Figure 5:Left: CSIMBA TMA; Right: Hyperspectral chip (+LVF) [credits VITO, AMOS, imec] All mentioned technology studies and built systems in the frame of a General Support Technology Programme or Basic Research Technology Programme activity are listed below with a selected subset of characteristics. Table 3:Hyperspectral instrument and technology developments under General Support Technology Programme or Basic Research Technology Programme study20,21,22,23,24,25,26. Instrument | ELOIS | HyperScout-1 | CHIMA | STREEGO | HyperScout-2 | CSIMBA |
---|
Platform | Micro-or minisatellite | 6-unit Gomx-4B CubeSat | Micro-or minisatellite | Micro-or minisatellite | 6-unit PhiSat-1CubeSat | 12-unit CubeSat | Mission/Activity | GSTP | GSTP(In Orbit Demonstration) | TRP | GSTP | In-orbit demonstration | GSTP | Involvement | AMOS, Spacebel, Centre Spatial de Liege, | Cosine research B.V., VITO,S [&] T AS, | AMOS, HoribaJobin-Yvon | Media Lario, TSD Space | Cosine research B.V | VITO, IMEC, DELTATEC, AMOS, | | Deltatec, Caeleste | TU Delft | | | | Aerospacelab, Spacebel | Purpose | Land observation | Vegetation, natural disaster monitoring | Atmospheric sensing | Agriculture, forestry, hydrology, resources, urban | Measure soil moisture, iceextent, ice thickness and detect melting ponds | Vegetation, agriculture, biodiversity | Status | Engineering Model | Launched in 2018 | 2019: Tested breadboard | Proto-Flight Model | Launched in 2020 | ~2022 (Proto-Flight Model) | Altitude | 600 km | 500 km | 800 km | 600 km | 540 km | 500 km | Swath | 70 km | 220 km | 1 km | 11 km | 310 km | 80 km | Design | Modified Offner + free-form grating | Three-mirror anastigmat + linear variable filter | Offner-based (free-form) + free-formgrating | Three-mirror anastigmat + linear variable filter | Three-mirror anastigmat + beam splitter to divide channels | Three-mirror anastigmat + linear variable filter | Spectral range | 400 - 2450 nm | 400 - 1000 nm | 600 - 800 nm | 430 - 880 nm | 400 – 1000 nm & 8.0 - 14 μm | 475 - 900 nm | Spectral bands | Up to 120 | 45 | TBC | Up to 150 | 48 | 50(470-620 nm)104(600-900 nm) | Spectral resolution | 2.5 nm | 12 nm | 0.5 nm | ~3 nm | 12 nm | 5 nm (typical) | Spatial resolution | 35 m | 70 m | 1 km | 5.5-11 m | 75 m and 490 m | 20 m | Temporal resolution | TBC | 45 days (global), daily revisit | TBC | ~1 week | Order of hours | ~1 week | Constellatio n ideas | Small satellite constellations | With e.g.constellation of 16, 2/hour revisit or 2/day global coverage | Small satellite constellations | Constellation for Sentinel-2equivalent bands in high resolution | Small satellite constellationsfor Earth Observation | Constellation for Sentinel-2 data complementatio n with a 2-3 days revisit | Instrument weight | 35 kg | 1.3 kg | TBC | 20 kg | 1.7 Kg | 4.5 kg | Instrument dimensions | 500 x 500 x 350 mm3 | 1-unit CubeSat compatible | 350 x 250 mm2 envelope optical instrument | 615 x 600 x 348mm3 | 19 x 13 x 12cm3 | 210 x 223 x 212 mm3 | Power | 8.27 W | 11 W | TBC | 17 W | 12 W | ~5 W |
4.KEY TECHNOLOGIESDuring the course of ESA Basic Research Technology Programme and General Support Technology Programme activities, a good number of new key technologies for hyperspectral instruments were identified. Some of them are discussed in more detail in the following sections. 4.1Free-form mirrorsA free-form surface is defined as a surface without rotational symmetry. This provides more degree of freedom to the optical designed to compensate for aberrations, furthermore it gives more flexibility for the positioning of optical components and of the image plane. Additional benefits are the avoidance of designs with obscuration, an enhanced image quality at steady volume and a larger field of view. Usually, a high signal-to-noise ratio can be maintained at low distortion values and designs can be made more compact or even reduced in number of components. The reduction of instrument components has again a positive influence on the ease of alignment, volume and cost. Such geometries could be also manufactured by the implementation of new additive manufacturing techniques using craved internal volumes and materials such AlSi40 or Scalmalloy, which would help to reduce the final components mass. 4.2Free-form gratingsGratings on free-form surfaces have been developed to provide one more parameter in the optimization of the optical design to reduce keystone and smile of spectrometer. Although this solution may be the source of additional complexity to a delicate optical element such as the grating, it proved to be a key component to minimize the number of optical elements. 4.3Hyperspectral Linear Variable FilterInstead of using a grating or a prism as dispersive spectrometer element, a hyperspectral system can also be designed by placing a linear variable filter in front of a CMOS detector. These filters are thin film Fabry-Perot interference filters with varying thickness and a narrow wavelength band that gradually changes across the filter. Usually, these filters are deposited on a glass substrate and then mounted on the detector, with potential alignment errors. Recent technology developments allowed to deposit the filters directly on the detector with nanometer accuracy, therefore resolving alignment problems. Furthermore, the filter application on a back side illuminated design allows for a higher quantum efficiency, and therefore a better signal-to-noise ratio. These technologies were developed within the frame of ESA Basic Research Technology Programme and General Support Technology Programme activities. 4.4On-board data processing and super resolution deep learningHigh resolution hyperspectral instruments generate a very large amount of data, in the range of one TB per orbit, exceeding by far satellite’s downlink capabilities, especially in the case of smallsat platforms. Instead of finding compromises to achieve data compression to achieve a manageable data volume for a small satellite, the path followed with the HyperScout development has been to pursue an on-board data processing to extract Level 2 data on-board in real time. An algorithm to calculate the Normalized Difference Vegetation Index in real time has been developed and tested. This work demonstrated that this mode of operation is within reach of a miniaturized payload using commercial off-the-shelf components currently available. Further device updates lead to the creation of HyperScout 2, which was designed including an Artificial Intelligence (AI) Eyes of Things processing unit27. Cosine B.V, claims that this new development represented the first hyperspectral orbiting device that using AI22. In addition to the reduction of data volume, the possibility of extracting information in real time on board will offer significant advantages for any phenomena where a prompt alert can trigger quick reaction, such as flooding, landslides, volcanic eruptions and so on. 4.5CoatingsAnother key technology in continuous development are structural and optical coatings. Enhancement in new materials and coating techniques will provide more competitive devices, by increasing their current optical efficiency and wavelength band capabilities, while structural black-painting coating would help to reduce stray light related issues. Examples of it are the recent years improved efficiency in UV coatings, extending their working ranges down to 50-250 nm, while still achieving efficiencies up to 70-80% by using MgF2 or LiF materials. Or advances in black coatings as provided by Acktar or Vantablack, which could serve to reduce stray light effects while also using lighter baffling components. 5.OTHER RELEVANT HYPERSPECTRAL DEVICE OUTSIDE ESAAlthough in this updated review authors showed the hyperspectral devices supported by ESA, one should not forget other Space Agencies where important steps are also being made. Some of the most relevant launched and in development devices are:
• Hyperion. From NASA and launched in 20001. • TianGong-1. From China28. • DESIS. From DLR and mounted on the ISS in 201829. • EnMAP. From DLR and planed for 2021-202230. • SHALOM. From ASI-ISA planed for 20221. • Amongst many more that slowly start populating our skies.
6.CONCLUSIONThis article presents an overview of the hyperspectral instruments related to ESA missions. Hyperspectral technologies and instruments, developed under ESA R&D contracts, were summarized and described in detail. Hyperspectral instruments and supporting technologies are currently experiencing a very dynamic phase. Applied technologies, processes, designs and application possibilities are far from being exhausted. Clearly, the trend of technology developments allows compact and cost-effective instrument designs with a short revisit time and quick information access, while keeping the data processing flexible for the use in different applications. At the same time, significant efforts are put into the implementation of the technology steps to match the challenging requirements of the Copernicus Space Component Expansion programme. ACKNOWLEDGEMENTSAuthors want to thank all industry partners and other members of the ESA/ESTEC TEC-MMO (Mechatronics & Optics) section for their support and dedication. REFERENCESTranson, J., D’andrimont, R., Maugnard, A. and Defourny, P.,
“Survey of Hyperspectral Earth Observation Applications from Space in the Sentinel-2 Context,”
Remote Sensing, 10
(2), 157
(2018). https://doi.org/10.3390/rs10020157 Google Scholar
“Sentinel Online: Home › Technical Guides › Sentinel-3 OLCI › OLCI Instrument › Description,”
ESA, “2019). Google Scholar
“Home › Missions › ESA EO Missions › Envisat › Instruments › SCIAMACHY › Design,”
ESA, “ Google Scholar
Van Mola, B. and Ruddicka, K.,
“The Compact High Resolution Imaging Spectrometer (CHRIS): the future of hyperspectral satellite sensors,”
Imagery of Oostende coastal and inland waters, Bruges,
(2004). Google Scholar
World Meteorological Organization,
“Instrument: GOME-2,”
World Meteorological Organization,2018). Google Scholar
Bazalgette Courrèges-Lacoste, G., Ahlers, B., Guldimann, B., Short, A., Veihelmann, B. and Stark, H.,
“The Sentinel-4/UVN instrument on-board MTG-S,”
https://www.eumetsat.int/cs/idcplg?IdcService=GET_FILE&dDocName=pdf_conf_p59_s1_01_bazalget_p&allowInterrupt=1&noSaveAs=1&RevisionSelectionMethod=LatestReleased Google Scholar
Coppo, P., Taiti, A., Pettinato, L., Francois, M., Taccola M. and Drusch, M.,
“Fluorescence Imaging Spectrometer (FLORIS) for ESA FLEX MIsison,”
Remote Sensing, 9
(7), 649
(2017). https://doi.org/10.3390/rs9070649 Google Scholar
“eoPostral Directory: Home › Directory › Satellite Missions › C › Copernicus: Sentinel-5,”
ESA, “2019). Google Scholar
“eoPortal Directory: Home › Directory › Satellite Missions › G › GomX-4,”
ESA, “ Google Scholar
Zuccaro Marchi, A., Maresi L., and Taccola, M.,
“Technologies and Designs for Small Optical Missions,”
ICSO, Chania
(2018). Google Scholar
De Clercq, C., Moreau, V., Jamoye, J.-F., Zuccaro Marchi, A., and Gloesner, P.,
“ELOIS: an innovative spectrometer design using a free-form grating,”
SPIE Optical Systems Design, Jena,2015). Google Scholar
“PhiSat-1 Nanosatellite Mission. HyperScout-2,”
Online, ESA, “ Google Scholar
Li, X., Wu, T., Liu, K., Li, Y. and Zhang, L.,
“Evaluation of the Chinese Fine Spatial Resolution Hyperspectral Satellite TianGong-1 in Urban Land-Cover Classification,”
Remote Sens., 8 438
(2016). https://doi.org/10.3390/rs8050438 Google Scholar
Krutz, D., Müller, R., Knodt, U., Günther, B., Walter, I., Sebastian, I., Säuberlich, T., Reulke, R., Carmona, E., Eckardt, A., Venus, H., Fischer, C., Zender, B., Arloth, S., Lieder, M., Neidhardt, M., Grote, U., Schrandt, F., Gelmi, S. and Wojtkowiak, A.,
“The Instrument Design of the DLR Earth Sensing Imaging Spectrometer (DESIS),”
Sensors, 19 1622
(2019). https://doi.org/10.3390/s19071622 Google Scholar
Guanter, L., Kaufmann, H., Segl, K., Foerster, S., Rogass, C., Chabrillat, S., Kuester, T., Hollstein, A., Rossner, G., Chlebek, C., Straif, C., Fischer, S., Schrader, S., Storch, T., Heiden, U., Mueller, A., Bachmann, M., Mühle, H., Müller, R., Habermeyer, M., Ohndorf, A., Hill, J., Buddenbaum, H., Hostert, P., Van der Linden, S., Leitão, P.J., Rabe, A., Doerffer, R., Krasemann, H., Xi, H., Mauser, W., Hank, T., Locherer, M., Rast, M., Staenz, K. and Sang, B.,
“The EnMAP Spaceborne Imaging Spectroscopy Mission for Earth Observation,”
Remote Sens, 7 8830
–8857
(2015). https://doi.org/10.3390/rs70708830 Google Scholar
|