Satellite imagery Low resolution satellite imagery is generally available for free while highresolution imagery must be purchased and licensing usually restrict its distribution Freely available low resolution multispectral amp radar satellite imagery ID: 1024417
Download Presentation The PPT/PDF document "GIS Training Introduction to Remote Sens..." is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
1. GIS TrainingIntroduction to Remote Sensing
2. Satellite imageryLow resolution satellite imagery is generally available for free, while high-resolution imagery must be purchased and licensing usually restrict its distributionFreely available low resolution multi-spectral & radar satellite imageryLandsat 7 ETM+, Landsat 8… : Earth Explorer (http://earthexplorer.usgs.gov)Sentinel-1,2,3 : Copernicus Open Access Hub (https://scihub.copernicus.eu)Ocean monitoringSea Surface Temperature, Chlorophyll-a (SeaWIFS), MODIS : NOAA/NASA (https://oceancolor.gsfc.nasa.gov)Commercial high-resolution imageryIKONOS, GeoEye, QuickBird, WorldView-1,2,3,4Pléiades,…Capricorn Seamount
3. Image acquisitionAs there is no reception station in the South Pacific, image acquisition must be scheduled ahead, stored on-board and transferred when above a receiving station.As a result there are not many images available for a given area, and they might be cloudyCapricorn SeamountSchedule acquisitionAcquire pictureTransfer picture
4. Image acquisitionImagery is acquired along an orbit, and the desired area may overlap several orbitsThe Landsat 8 and Landsat 7 satellites both maintain a near-polar, sun-synchronous orbit, following the World Reference System (WRS-2). They each make an orbit in about 99 minutes, complete over 14 orbits per day, and provide complete coverage of the Earth every 16 days
5. Obtaining free imageryFor freely available imagery (Landsat, Sentinel), you just browse through the catalogue of archived images and select one with not too many clouds above the area of interest (using quicklooks as guidance). Then you submit the request and receive a link to download the image. An image would typically be around 1 GB
6. Ordering commercial imageryFor high resolution imagery, you create a polygon for the area of interest and query available imagery for that polygon. Because the image is billed per sq km, you only select the area you need, yet there are restrictions on the shape of the polygon and a minimum area for the order (25 km² for WorldView imagery for example)Total area 25 km²Min width
7. Ordering commercial imageryThe price per km² depends on the resolution, satellite, number of bands (4 or 8), licensing and processing.A multi-licensed image cost around USD 50 per km² at public priceSPC is a Digital Globe reseller and as such get bargain prices on some imagery, but the license restrict the use of images to the government for which they have been purchased : (commercial imagery you obtain through SPC is generally not shareable with third parties)Contact SPC Geoscience, Energy and Maritime Division (GEM) of you need to get access to already purchased imagery or order new satellite imagery. A lot of imagery has been purchased during the last 10 years, you might be lucky
8. Image mosaicBecause of paths and cloud cover, a combination (mosaic) of images taken at different times is generally necessary to cover a big area
9. Satellite sensorsImagery is the output of the satellite sensors, at a given spatial resolution with bands corresponding to a range of visible, infrared (or radar wavelengths for radar satellites)Capricorn Seamount
10. Radar imageryWater is opaque to radar wavelengths. Radar imagery can be used to detect boats but is onerous and often impractical for coastal applicationsA small boat would only be a few pixels on a radar imageThe image is not received in real timeOnce a boat is detected and there is suspicion of illegal activity, you still need to send aerial and naval forces to control the boat and gather evidences.Capricorn SeamountBoats
11. Multispectral imageryCapricorn SeamountSatellite Resolution BandsLandsat-8 30m/15m Pan+8 MS +TIRSentinel-2 10m-60m 12 bandsIKONOS 4m/1m Pan+4 MS (defunct)QuickBird 3m/.7m Pan+4 MS (destroyed)GeoEye-1 1.8m/.5m Pan+4 MSWorldView-1 .6m PanWorldView-2 2m/.5m Pan+8 MSWorldView-3 1.2m/.3m Pan+8 MS +SWIR+CAVIS
12. Hyperspectral imageryCapricorn SeamountAirborne (AVIRIS, CASI) and satellite (EO-1, HySIS)
13. Sea surface temperatureCapricorn Seamounthttps://www.ospo.noaa.gov/Products/ocean/sst/50km_night/index.html
14. WindsCapricorn Seamounthttps://manati.star.nesdis.noaa.gov/datasets/ASCATData.php
15. Satellite bandsCapricorn SeamountBand 1: Coastal blueBand 5: RedBand 8: NIR 2Panchromatic – Res .5mWorldView-2 : MSS bands 1-8 have a resolution of 2m, while panchromatic is 50 cm
16. Pan-sharpeningCapricorn SeamountBands 4,3,2 - Resolution 2mPanchromatic – Resolution .5mPan-sharpened 4,3,2 Resolution .5m+Pan-sharpening increases the spatial resolution of the multispectral image by merging it with the higher resolution panchromatic image
17. Image rectificationCapricorn Seamount
18. Image rectification
19. Image rectification
20. Image enhancement: dehazeCapricorn Seamount
21. Image classification: isodataCapricorn Seamount
22. Image classification: k-meansCapricorn Seamount
23. QGIS: Processing ToolboxCapricorn SeamountEnable the Processing Toolbox panel to access advanced tools
24. Image classification: k-meansCapricorn SeamountOpen the Landsat 7 ETM image and launch k-means clustering from the toolbox
25. Image classificationCapricorn SeamountImage classification can be supervised or unsupervised
26. Image classification: k-meansCapricorn SeamountSelect the Landsat 7 ETM image as Grid and set the number of clusters to 12Run k-means clustering12
27. Image classification: k-meansCapricorn SeamountSet render type to palette and classify with random colors. Remove colors corresponding to deep ocean
28. Image classification: k-meansCapricorn SeamountCompare the classified image with the original image
29. Image classification: raster to vectorCapricorn SeamountPolygonize the clusters image
30. Image classification: raster to vectorCapricorn SeamountSet the symbology and remove classes for deep water
31. Image classification: surface areasCapricorn SeamountUse Group Stats to calculate the surface area by class
32. Spectral processing: PCACapricorn Seamount
33. Capricorn SeamountApplication to North Efate mangrove4 km PCA on mangrove area
34. Sampling points in the mangroveSampled points for the areaWeb database interface
35. Species composition for sampled points
36. PCA vs species composition for sampled points
37. Delineation of zones based on PCA and species composition and extraction of polygon area in Ha
38. Intersection of mapping points and polygonsAcAuAcSpAvMaBaRaExAgHeLiLuLiRhApRhSaXyGrPoly idArea Ha200005020001415.94900002000001415.944000030300001415.9410001010001415.9420100020001415.9400002050001415.94002000000001415.94004000000001415.940020010100001415.9400000010950137.15001000000400137.150000001003020120.4000010110020120.4000000007020137.15
39. Averaging species composition by polygonand calculation of a weighted average (using area Ha) for the whole area