EnMAP Environmental Mapping and Analysis Program Kaufmann et al., 2005 GeoForschungsZentrum Potsdam charly@gfz-potsdam.de 4th ESA Chris Proba Workshop ESA-ESRIN Frascat, Italy
Introduction In 2003 DLR-Agency started a selection process for a future Earth observation mission 9 different missions have been proposed 2 proposals have been selected for a phase A study: EnMAP: Hyperspectral Mission TanDEM-X: SAR Interferometry Mission Phase A study for EnMAP accomplished end of 2005, URD and MRD provided Both proposed missions have been selected for implementation Financial budget for EnMAP => 65 Mio (Instrument + Bus + Launcher) Start of phase B for EnMAP in October 2006 Envisaged launch date for EnMAP end of 2010
Overall Mission Goals To provide high-spectral resolution observations of bio-geochemical and geophysical variables To observe and develop a wide range of ecosystem parameters encompassing agriculture, forestry, soil/geological environments and coastal zones/inland waters To enable the retrieval of presently undetectable, quantitative diagnostic parameters needed by the user community To provide high-quality calibrated data and data products to be used as inputs for improved modeling and understanding of biospheric /geospheric processes Environmental Mapping and Analysis Program
Multi- versus Hyper-spectral / Potentials Multi-spectral Hyper-spectral N chlorite calcite dolomite alunite gypsum kaolinite appr. 1 km Makhtesh Ramon color composite of bands 1, 20, 48 Few fixed bands minimum identification field knowledge and labanalysis required low confidence Contiguous bands maximum identification increased classification accuracy high confidence
Identification/Quantification => Diagnosis 100 Each material on the Earth s surface has a unique spectral characteristic Reflectance [%] 90 80 70 60 50 kaolinite D = 1 - R min / R c R c R min 40 2.0 2.05 2.1 2.15 2.2 2.25 2.3 2.35 Wavelength in microns Individual Absorptions of pigments, minerals, man made objects Shape Position Depth Identification Quantification
Simulation Approach (SWIR-Range) Simulation parameters: 1950 nm - 2450 nm 510 spectra of 51 minerals Six different bandwidth: 5,10,15,20,30,50 (FWHM) All possible band centrings (steps of 1 nm) Noise Model 50, 100, 200 (SNR) 780 spectral libraries 23409 spectra each Classifier: Spectral Feature Fitting (SFF) Reduction of feature depth to simulate exposed natural surfaces
Simulations for opt. Band Design & SNR full resolution field spectrum full resolution radiance resampled radiance MODTRAN Spectral Response Function retrieved EnMAP spectra SNR 50 noisy resampled radiance SNR 100 Spectral Response Function resampled field spectrum SNR 200 MODTRAN Noise Model
SWIR - Results for Detectability of Minerals Results of detectability calculations Final results indicating optimum detectability at 10-15 nm for SNR ~150:1 and nearly independent from centering position Model for increasing SNR at increasing FWHM
Spectral Requirements Spectral Sampling interval VNIR-range Spectral Bandwidth (FWHM) Number of Bands (total of 218) I 420 nm - 500 nm 10 nm 10 ± 1 nm 8 II 500 nm - 850 nm 5/10 nm 5/10 ± 1 nm * 70/35 III 850 nm - 1030 nm 10 nm 10 ± 1 nm 18 SWIR-range Ia 950 nm - 1390 nm 10 nm 10 ± 1 nm 44 Ib 1480 nm - 1760 nm 10 nm 10 ± 1 nm 28 II 1950 nm - 2450 nm 10 nm 10 ± 1 nm 50 * Water/Land Mode
VNIR Detector Trade-off Investigations and meetings with manufacturers like E2V (UK) / Atmel (F) / Dalsa (NL) / Fillfactory (B) / Fairchild + Rockwell (USA) Simulations have shown non-compliances in Frame Rate / Quantum Efficiency / Full Well Capacity / Noise / Pixel Size off-the-shelf detector for EnMAP VNIR wavelength is not available Customized arrays from E2V and Fairchild can provide sufficient SNR performance (1024 x 512, 24 µm pixels, on-chip binning by 2, 8 output channels 227 Hz frame rate) E2V : known ESA/DLR supplier (+), existing technology (+), lower margin (-) Fairchild : best performance (+), relative new technology under development (-) E2V detector chosen as baseline
SWIR Detector Trade-off First choice of state-of-the-art detector materials : MCT (Mercury Cadmium Telluride, HgCdTe) Consultations of major manufacturers for scientific imaging : AIM (D) / Sofradir (F) / Rockwell (USA) / XenICs (B) Simulations : Sofradir detector meets EnMAP requirements with low margin AIM (current GENSIS development) shows best performance Sofradir : known ESA/DLR supplier (+), existing technology (+), low margin (-) AIM : best performance (+), Integrated Detector & Cooler Assy (IDCA, +) German supplier, dedicated technology development under DLR-BO funding (GENSIS) AIM detector chosen as baseline 1024 x 256, 24 µm pixels, 8 output channels 227 Hz frame rate
Sensor Parameters Signal-to-noise ratio (SNR) at 30% reflectance; 30 sun zenith angle; visibility 21 km; target 500 m a.s.l. Spectral calibration accuracy Spectral stability Requirements VNIR: > 500:1 (at 495 nm) SWIR: > 150:1 (at 2200 nm) 0.5 nm 0.5 nm Radiometric calibration accuracy < 5 % Quantification / Radiometric stability 14 bit / < 2.5 % Spectral smile/keystone effect Ground sampling distance (GSD) Swath width Geometric co-registration Swath length (at least) < 20 % of detector element 30 m x 30 m (at nadir; sea l.) 30 km 0.2 x GSD 1000 km /orbit; 5000 km /day Data Rate / Mass Memory Weight 860 Mbit/s / 512 Gbit 170 kg (780 kg incl. bus)
Orbit Parameters Requirements Coverage Global coverage in near nadir mode ( 5 ); Polar caps may be excluded Repeat cycle 23 days Pointing capability Target revisit time Local crossing time Pointing knowledge Orbit altitude Orbital period ± 30 o cross-track pointing 4 days with pointing capability 11.00 hrs ± 15 min 100 m at sea level 643 km ca. 98 minutes Orbit inclination 97.96
EnMAP Instrument Architecture SAR-Lupe Bus design heritage
Science Program / Fields of Applications EnMAP management of agricultural and forest ecosystems hazard assessment ARES parameter extraction and modeling Co-operative Network GFZ DLR inland water urban development dryland degradation
ARES Aircraft Support Scan Principle Whisk-Broom FOV 65 IFOV 2.0 mrad Oversampling 1.43 GSD 2.5 m - 10 m No. of Pixel 813 No. of Bands 151 (121+30) Rad. Resolution 14 bit Co-Registr. (refl.) < 0.05 pixel Co-Registr. (em.) < 0.05 pixel Co-Registr. (refl. - em.) < 0.10 pixel Wavelengths No. of Spec. Sampl. Band-Width Rad. Requirements Bands Interval (FWHM) [W m -2 sr -1 µm -1 ] 470-890 nm 32 15-13 nm 15-13 nm 0.09 NER 890-1350 nm 29 18-16 nm 16-14 nm 0.04 NER 1360-1800 nm 30 16-14 nm 14-13 nm 0.03 NER 2020-2420 nm 30 15-13 nm 14-12 nm 0.02 NER 8.1-12.1 µm 30 147-131 nm 130-117 nm 0.1-0.3K NE T
Interested Parties Participating Countries Germany Canada India Spain France Australia USA Netherlands Israel China South Africa Poland Italia New Zealand Sweden Switzerland Hungary Finland Denmark Represented Disciplines Agriculture/Forestry Biodiversity Ecology Wetlands Climate Change Water Soils/Landdegradation Geology/Mineralogy Arid Zones Cartography Urban Fisheries Meth. Development Cal./Val.
EnMAP - Project Structure Mission Advisory Board Project Management (DLR Agency) Scientific PI (GFZ) Science Team Space Segment (Industry PI) Ground Segment (DLR) Commercial Service Element (Value Adder) Instrument Operations Support (IOS) (IMF-AP) Mission Operation Segment (MOS) (RB-MB) Processing and Cal/Val (PCV) (IMF-PB) Payload Ground Segment (PGS) (DFD-BI)
Final Remarks The EnMAP mission as presented is a purely national financed solution starting into phase B in 10/06 fully compliant with the DLR AO fully compliant with the requirements according to the EnMAP User Requirements Document (URD) elaborated in phase A feasible with the allocated financial budget operated by DLR-GSOC (German Space Operations Center) and DLR-DFD (Deutsches Fernerkundungs-Datenzentrum) open for international partnerships with respect to data utilization and data downlink and may include the operation of additional international ground stations
Thanks for you patience Contact: http://www.gfz-potsdam.de