ASI-SRV General purpose modules for the pre-processing of

ASI-SRV General purpose modules for the pre-processing of remote sensed optical data Massimo Musacchio, Sergio Teggi, Fabrizia Buongiorno, Angelo Amodio, Marco Gregnanin, Giulia De Marzi, Stefano Vignoli, Sergio Perelli, Vincenzo Santacesaria USERe. ST' 08 18/03/2018 1

The ASI-SRV provides support to the following volcanic activity phases addressed by the Italian Civil Protection Department (DPC): › Surveillance and early warning › Sin-eruption phase › Post-eruption phase USERe. ST' 08 18/03/2018 2

How does ASI-SRV works EO-non EO data Providers USERe. ST' 08 18/03/2018 Added value Products 3

ASI-SRV system will be developed in 3 phases › In the first version the core of the system has been realized, including modules/algorithm well known and consolidated (TES, VAOT, Water Vapour, Effusion Rate, SO 2 and LAOT) READY TO BE OPERATIVE › During the following phases RTD based module will be developed and implemented (VAMP, Surface change detection, CO 2 , Multiparametric analysis ) USERe. ST' 08 18/03/2018 4

To optimize the computer processing time the ASISRV architecture is based on two data processing chains in order to get advantage of partnership infrastructures and laboratories The Optical data processing chain is localized in Rome at INGV The SAR data processing chain is localized in Naples at CNR-IREA This presentation is aimed to the “Optical-based” modules. USERe. ST' 08 18/03/2018 5

Several spaceborne based EO optical data will be acquired and processed: USERe. ST' 08 NASA EO-1 Hyperion NASA Terra/Aqua MODIS NASA Terra ASTER NOAA AVHRR EROS SPOT Quickbird Each data is furnished with specific spectral and spatial characteristics 18/03/2018 6

Pre crisis Surface Thermal anomalies monitoring ASTER VAOT Volcanic Aerosol Optical Thickness estimation Hyperion Water vapour estimation Hyperion Crisis Effusion rate AVHRR, MODIS Ash clouds optical characteristics (VAMP) AVHRR, MODIS Low resolution Aerosol Optical Thickness (LAOT) AVHRR MODIS Degassing plumes SO 2 Characterization AVHRR MODIS Post Crisis Change detection on surface characteristics due: Lava flow Hyperion and HI-RES Ash cover Hyperion and HI-RES USERe. ST' 08 18/03/2018 7

Sensor Provider Revisit time Production time Provision time QUICKBIRD Digital Globe Eurimage/Telespazio On demand hours SPOT IAMGE On demand hours EROS Image. Sat International IPT On demand hours ASTER-TERRA NASA 16 days hours Day and Night time passage On-demand Note Lower Revisit Time Higher Spatial Resolution Less than 1 mt HYPERION USERe. ST' 08 hours NOAA Direct Broadcast 4 hour minutes hours Directed broadcasted by INGV MODIS 18/03/2018 16 days AVHRR About 1 km NASA It follows the ASTER-TERRANASA orbit On demand ESA 2 per day minutes hours Night/day 8

Coreg Calibration L 1 A Digital Count L 1 B L 1 C Sensor Radiance Sensor Reflectance Level 1 Atmospheric Correction Auxiliari L 1 D Sensor Radiance Sensor Reflectance DEM Shaded Slope Aspect Bo. A Radiance Bo. A Reflectance Apparent Radiance Apparent Reflectance DEM Shaded Slope Aspect Correction terms Lup Ld. . . Level 2 Vector generation Generic Processor L 2 A L 2 B L 2 C Map Classified Mask sensor Geometry DEM Geometry Classified Mask DEM Geometry L 2 D Vectorial Layer DEM Geometry Level 3 L 3 A USERe. ST' 08 sensor Geometry 9 18/03/2018 Time, spatial Processor L 3 B DEM Geometry

General Purposes modules USERe. ST' 08 18/03/2018 10

General purposes modules To have EO “standardized” data, ready to be processed, by means of Radiometric calibration, Resized with a defined geographic extension and coverage, Atmospheric USERe. ST' 08 18/03/2018 and topographic effect removed 11

USERe. ST' 08 18/03/2018 12

GIS Referencing Tool The need to have common activities performed on the SRV product led to the development of a set of common tools, in order to perform, on different image standards, the same operations. The common tools (named GTR) are devoted to: cut and mosaic of the input images coregistration of input DEM and georeferencing of final SRV products USERe. ST' 08 18/03/2018 13

Within the ASI-SRV project a well defined geographic window has been defined for each of the three area of interest. (Vesuvio Campi Flegrei, Etna). This requirement leads to cut the low-res images, since they span over a geographic area wider than the one of interest; while each hi-res image may cover an area which is more narrow than the desired window, so a mosaic of several images followed by a cut of the temporary image obtained is needed. The tasks are made more complex by the fact that input images are provided in sensor geometry, so the cut has to be done in an appropriate way, in order to avoid the loss of points falling into the desired geographic window. USERe. ST' 08 18/03/2018 14

ETNA tes site Satellite/sensor Km dimension Upper Left Lower Right Width Height Longitude Latitude MODIS regional 350. 00 450. 00 11° 40’ 00” 39° 50’ 00” 16° 50’ 00” 36° 00’ 00” MODIS local 40. 00 14° 45’ 32” 37° 56’ 40” 15° 13’ 04” 37° 34’ 43” AVHRR regional 350. 00 450. 00 11° 40’ 00” 39° 50’ 00” 16° 50’ 00” 36° 00’ 00” AVHRR local 40. 00 14° 45’ 32” 37° 56’ 40” 15° 13’ 04” 37° 34’ 43” ASTER 62. 00 49. 50 14° 42’ 41” 37° 58’ 14” 15° 16’ 32” 37° 24’ 36” Hyperion 8. 50 35. 00 14° 58’ 00” 37° 50’ 00” 14° 59’ 11” 37° 30’ 00” Very High Resolution 26. 00 N/A N/A USERe. ST' 08 18/03/2018 15

Vesuvio and Campi Flegrei test site Satellite/sensor Km dimension Upper Left Lower Right Width Height Longitude Latitude MODIS regional 55. 00 33. 00 13° 59’ 47” 40° 55’ 07” 14° 36’ 06” 40° 39’ 12” AVHRR regional 55. 00 33. 00 13° 59’ 47” 40° 55’ 07” 14° 36’ 06” 40° 39’ 12” ASTER 65. 00 42. 00 14° 00’ 00” 41° 00’ 54” 14° 46’ 28” 40° 37’ 14” Hyperion (Vesuvio) 7. 60 12. 00 14° 23’ 32” 40° 51’ 50” 14° 27’ 29” 40° 45’ 02” Hyperion (Campi Flegrei) 16. 00 18. 50 14° 01’ 26” 40° 54’ 52” 14° 13’ 05” 40° 44° 51” Very High Resolution 26. 00 N/A N/A USERe. ST' 08 18/03/2018 16

The DEM Co-registration tool implements the two major functions of orthorectification and georeferentiation of raw images. The following two different models will be used: › Satellite Sensor Rigorous Orbital Model › Rational Polynomial Coefficients (RPC) Dem coregistration module produces HDF file as output. This output will be used as input for all DPS. The output level of the generated product is 1 C (e. g. ASTER_1 C, AVHRR_1 C) The tools are automated and do not require human interaction. The tools execution is scheduled by the SRV system. USERe. ST' 08 18/03/2018 17

• No specific algorithms are required. The module consists in the mere application of calibration coefficients. • The ASTER and HYPERION data will be calibrated, while MODIS data will be not calibrated because level 1 B data products contain calibrated radiances for all 36 MODIS bands and reflectances for the reflective Solar bands (Bands 1 through 19 and 26). USERe. ST' 08 18/03/2018 18

Radiometric Calibration Module • ASTER Input: Level 1 B in Digital Number (DN) USERe. ST' 08 18/03/2018 19

ATMOSPHERIC AND TOPOGRAPHIC CORRECTION TOOLS: CIRILLO USERe. ST' 08 18/03/2018 20

Using CIRILLO the spectral values of atmospheric terms, (i. e. transmittances, reflectance contribution due to solar radiance scattered by the atmosphere and downward spherical albedo of the atmosphere) are computed The second reason to prefer CIRILLO is due to the capability to evaluate altitude and b factor for each pixel of the image. For these calculations CIRILLO requires as input three images (files), geographically registered with the image to be corrected and with the same spatial resolution, containing elevation, slope and aspect USERe. ST' 08 18/03/2018 21

G A-B-C-D-E-F A. The sun radiance that reaches directly the pixel viewed by the sensor (target) and that is directly reflected by the target to the sensor; B. The sun radiance that reaches directly the pixel viewed by the sensor (target) and that is reflected by the target to the sensor following a multiple scattering path; C. The sun radiance that reaches the target following a multiple scattering path and that is directly reflected by the target to the sensor; D. The sun radiance that reaches the target following a multiple scattering path and that is reflected by the target to the sensor following a multiple scattering path; E. The sun radiance that directly reaches the surface surrounding the target and that is reflected by the surface to the sensor following a multiple scattering path; F. The sun radiance that reaches the surface surrounding the target following a multiple scattering path and that is reflected by the surface to the sensor following a multiple scattering path; G. The sun radiance that is directly scattered by the atmosphere to the sensor without reaching the ground. All of these terms, with the exception of G), are also influenced by the orientation of the surface with respect to the sun illumination direction. USERe. ST' 08 18/03/2018 22

USERe. ST' 08 18/03/2018 23

USERe. ST' 08 18/03/2018 24

ASI-SRV OPTICAL SENSOR DATA PROCESSING MODULES Massimo Musacchio, Malvina Silvestri, Claudia Spinetti, Stefano Corradini, Valerio Lombardo, Luca Merucci, Maria Fabrizia Buongiorno, Sergio Pugnaghi, Gabriele Gangale, Lorenzo Guerrieri, Sergio Teggi, Vincenzo Santacesaria, Sergio Perelli USERe. ST' 08 18/03/2018 25

For each product a historical series of remote sensed data have been processed By means of specific algorithm and using Auxiliary and Ancillary… …the product is obtained Shapefile Before to post on the foreseen WEBGIS each raster needs to be converted in a ESRI like shapefile USERe. ST' 08 18/03/2018 26

USERe. ST' 08 18/03/2018 27

USERe. ST' 08 18/03/2018 28

USERe. ST' 08 18/03/2018 29

USERe. ST' 08 18/03/2018 30

USERe. ST' 08 18/03/2018 31

USERe. ST' 08 18/03/2018 32

USERe. ST' 08 18/03/2018 33

Objective of the GTA is the generation of a synoptic view of a huge volume of heterogeneous data Discontinuous measurements Webcam network UV-scanner network for SO 2 flux Seismic network GPS permanent network Gravimetric network Magnetic network Analysis of the erupted ash Analysis of the erupted products Geologic and structural surveys Thermal mapping from helicopter Lava mapping from field USERe. ST' 08 18/03/2018 34

At the end of the scientific processing chain, raster classified products are available, and they need to be georeferenced before their transformation in vector products, in order to be displayed as a geographic information layer on a map. This task is performed by the “Map Projection” module, in order to warp them in an UTM projection. This georeferenced product is then stored in a Geo-TIFF file, that contains all the information needed for a correct visualization on a GIS. The product obtained is then ready to be analyzed by the Operator using the GTA, in order to validate or discard it before the publishing to the end user. USERe. ST' 08 18/03/2018 35

The GTA is a customization of ESRI Arc. GIS Desktop to support the operator in performing the Validation Process Workflow. (and its extensions): Arc. GIS is powerful to manage a huge amounts of SRV products Arc. GIS is widely used within DPC and INGV Arc. GIS includes a wide variety of programmable components, so that plug-in functionalities to validate SRV products can be integrated Arc. GIS, using some extensions, is compliant with the OGS protocols to be used in the frame of the project (LDS) GTA will use ESRI Arc. GIS USERe. ST' 08 18/03/2018 36

SRV Process Validation Workflow USERe. ST' 08 18/03/2018 37

USERe. ST' 08 18/03/2018 38

USERe. ST' 08 18/03/2018 39

For further information on ASI-SRV project contact: Maria Fabrizia Buongiorno: buongiorno@ingv. it USERe. ST' 08 18/03/2018 40