Blue tab
Green tab
Brown Tab

Earth Observation

 

 

Earth Observation in the frame of EO-MINERS - Overview of remote sensing methods, sensors and applications

Remote sensing methods and sensors

Remote sensing sensors vary in their platform (ground, air or space), in their spatial resolution, ground swath width, spectral resolution, spectral coverage and more. This section will give a brief description of the major EO remote sensing methods and sensors available today. For a more detailed description please refer to D2.1 (Ben-Dor et al. 2012) of the EO-MINERS project. For a detailed and updated sensor list with links to vendors or archives please refer to the web site of the Faculty of Geo-Information Science and Earth Observation (ITC) of the University of Twente (http://www.itc.nl/research/products/sensordb/searchsen.aspx).

Optical High Resolution Sensors - Sensors which can be either spaceborne or airborne with high spatial resolution (GSD pixel size<5m). The use of those sensors is recommended when a visible small size change in the environment is to be assessed mainly Land Use and Land Cover changes (LULC). (It can be referred also to non-visible changes but, in practice, those need hyperspectral sensors which ground sampling distance is still rather high due to the need to collect large amounts of photons). These are also the sensor systems to be used when addressing geometrical properties instead of spectral properties

Usually all information described above are provided by the service provider of each sensor. An example from the WorldView-2 sensor in orbit is given bellow. WordView-2 is a commercial panchromatic and VNIR multispectral spaceborne scanner, utilized by the many users for LULC applications.

The characterises of Worldview-2 properties as provided by DigitalGlobe (http://worldview2.digitalglobe.com/)

http://worldview2.digitalglobe.com/

Launch Date

October 8, 2009

Orbit Altitude

770 kilometers

Orbit Type

Sun synchronous, 10:30 am (LT) descending Node

Orbit Period

100 minutes; 7.25 year mission life, including all consumables and degradables (e.g., propellant)

Spacecraft Size, Mass, & Power

4.3 meters (14 feet) tall x 2.5 meters (8 feet) across, 7.1 meters (23 feet) across the deployed solar arrays; 2800 kilograms (6200 pounds); 3.2 kW solar array, 100 Ahr battery

Sensor Bands

Panchromatic
8 Multispectral (4 standard colors: red, blue, green, near-IR), 4 new colors: red edge, coastal, yellow, near-IR2

Sensor Resolution GSD

Ground Sample Distance Panchromatic: 0.46 meters GSD at Nadir, 0.52 meters GSD at 20° Off-Nadir
Multispectral: 1.8 meters GSD at Nadir, 2.4 meters GSD at 20° Off-Nadir
(note that imagery must be resampled to 0.5 meters for non-US Government customers)

Time Delay Integration (TDI)

Panchromatic - 6 selectable levels from 8 to 64
Multispectral - 7 selectable levels from 3 to 24

Swath Width

16.4 kilometers at nadir

GPS Position Accuracy & Knowledge

< 500 meters at image start and stop
Knowledge: Supports geolocation accuracy below Retargeting

Agility Acceleration

1.5 deg/s/s
Rate: 3.5 deg/s
Time to slew 300 kilometers: 9 seconds

Max Viewing Angle

Accessible Ground Swath Nominally +/-40° off-nadir = 1355 km wide swath
Higher angles selectively available
Per Orbit Collection: 524 gigabits
Max Contiguous Area Collected in a Single Pass: 96 x 110 km mono, 48 x 110 km stereo

Revisit Frequency

1.1 days at 1 meter GSD or less 3.7 days at 20° off-nadir or less (0.52 meter GSD)

Geolocation Accuracy

(CE 90) Specification of 12.2m CE90, with predicted performance in the range of 4.6 to 10.7 meters (15 to 35 feet) CE90, excluding terrain and off-nadir effects

With registration to GCP's in image: <2.0 meters (6.6 ft)

As seen from the table, the WorldView-2 sensor is dedicated to provide high spatial resolution with limited (relatively good) spectral resolution. An example to the spatial spectral products is given in figures 2 and 4.

Figure 1: A RGB WorldView-II image
Figure 1: A RGB WorldView-II image

Optical Multispectral Sensors – As was seen in WorldView-2, the 8 bands were able to provide reasonable spectral information. Other multispectral sensors available to that end are mostly the LANDSAT series. The importance of the LANDSAT program is based on the huge archive of imagery provided by NASA. The LANDSAT images can thus be useful for long range temporal change detection with limited spectral information since 1972. Airborne multispectral sensors have relatively high spatial resolution and are useful for the monitoring of rough changes in spectral features or fine changes in land cover.

LANDSAT TM (VNIR-SWIR-TIR multispectral spaceborne scanner)

LANDSAT characteristics
  LANDSAT 4,5 (1-3) LANDSAT 4,5 LANDSAT 7
Instrument Multispectral Scanner
(MSS)
Thematic Mapper
(TM)
Enhanced Thematic
Mapper Plus (ETM+)
In operation since 1972 since 1982 since 1999
Pixel size 79 x 79 m² 30 x 30 m² 30 x 30 m²
Spectral channels 1 (4) 0,50 - 0,60 µm, Green
2 (5) 0,60 - 0,70 µm, Red
3 (6) 0,70 - 0,80 µm, Near infrared
4 (7) 0,80 - 1,10 µm, Near infrared

1 0,45 - 0,52 µm, Blue-Green
2 0,52 - 0,60 µm, Green
3 0,63 - 0,69 µm, Red
4 0,76 - 0,90 µm, Near infrared
5 1,55 - 1,75 µm, Mean infrared
7 2,08 - 2,35 µm , Mean infrared

1 0,45 - 0,52 µm, Blue - Green
2 0,53 - 0,61 µm, Green
3 0,63 - 0,69 µm, Red
4 0,78 - 0,90 µm, Near infrared
5 1,55 - 1,75 µm, Mean infrared
7 2,09 - 2,35 µm , Mean infrared
Thermal channel   6 10,4 - 12,5 µm (120 x 120 m²) 6 10,4 - 12,5 µm (60 x 60 m²)
Panchromatic channel     8 0,52 - 0,90 µm (15 x 15 m²)

 

Figure 2. satellite images
Figure 2: Spatial resolution variations: very high resolution panchromatic WV-2 image on the left (GSD=0.5m),
high resolution Vis-NIR multispectral WV-2 image on the center (GSD=2m) and a low resolution
Vis-NIR-SWIR LANDSAT 7 image on the right (GSD=30m).

 

Figure 3: LANDSAT 5 image
Figure 3: LANDSAT 5 false colour composite image (source: Satellite Imaging Corporation)

Optical Hyperspectral Sensors - Hyperspectral data is divided into pin-point spectral measurements and imaging spectrometers. Ground pin-point spectral instruments do not create an image, but obtain an extremely high spectral resolution over a single solid angle. The output of those instruments is often referred to in the literature as Vis-NIR-SWIR Spectroscopy or simply NIR Spectroscopy. Vis-NIR-SWIR spectroscopy can be used as a calibration tool. After data is calibrated, imaging spectrometers can be used to map the mineralogy and chemistry of soil, water and air (e.g. acid mine drainage). Recent hyperspectral sensors that can be operated from the ground are available. They merge spatial information provided by regular cameras with the spectral information extracted from a point spectrometer. Still the SNR of these sensors is lower than regular point spectroscopy but as technology progress, this option will be more and more important to remote sense the environment from the ground. As for now, high spatial resolution hyperspectral spaceborne sensors are not available. As a result, the only way for collecting high spatial resolution hyperspectral data is by an airborne campaign. 

Example: Hyperion (spaceborne hyperspectral scanner):

Hyperion main characteristics

• 220 10nm bands covering

400nm - 2500nm

• 6% absolute rad. accuracy

• Swath width of 7.5 km

• IFOV of 42.4 μradian

• GSD of 30 m

 

Figure 4: Spectral resolution variations diagram
Figure 4: Spectral resolution variations: A reflectance spectrum of green grass as collected by a
Vis-NIR-SWIR hyperspectral sensor, such as Hyperion or HyMap (purple), by a Vis-NIR-SWIR multispectral
sensor such as LANDSAT TM (red) and by a Vis-NIR multispectral sensor such as WV-2 (blue).

TOP