Remote Sensing

Revision as of 03:03, 8 November 2017 by Alexis (talk | contribs)
Jump to: navigation, search

Remote Sensing involves the collection and interpretation of information sensed from distant objects, using usually aircrafts and satellites [1]. It is used to sense the energy that is naturally emitted or reflected by the earth’s surface, from the atmosphere or from other devices [2]. Any object with a temperature above the absolute zero (-273°C) emits energy called electromagnetic radiation that depends on the object’s temperature. The higher the temperature of the object, the shorter it is the reflected electromagnetic radiation wavelength [3].

Read First

This section is a complement of the Geo Spatial Data Wiki Page to explain more details of the satellite products, available datasets and how to extract data for impact evaluation.

There are several open source platforms that are useful to process images that have been produced using remote sensing: Erdas Imagine, TerrSet, Orfeo Toolbox, Opticks, ILWIS, Whitebox and Google Earth Engine. In this Wiki Page, we focus on the use of the Google Earth Engine platform. Sentinel-2-1.jpg Photo credits: NASA

Guidelines

Most used satellite products

There is a number of satellite products that are accessible to the public at different spatial and temporal resolutions that can be used to monitor changes in land cover in any part of the world. The following list shows the collection of images that are frequently used for remote sensing given their multispectral and multitemporal characteristics, all of them publicly available:


USGS Landsat 7 Raw Scenes (Orthorectified)

This dataset contains orthorectified images from Jan 1, 1999 to Apr 30, 2017 16 days with the following spectral bands (all bands are available at a resolution of 30 meters per pixel, except for band 8 that comes at a resolution of 15 m., and B6_VCID_1 and B6_VCID_2 were generated with a sensor of 60 m. resolution):

  • B1: Blue (0.45 - 0.52 µm)
  • B2: Green (0.53 - 0.60 µm)
  • B3: Red (0.63 - 0.69 µm)
  • B4: Near Infrared (0.77 - 0.90 µm)
  • B5: Short-wave Infrared 1 (1.55 - 1.75 µm)
  • B6_VCID_1: Low-gain Thermal Infrared 1 (10.40 - 12.50 µm). This band has expanded dynamic range and lower radiometric resolution (sensitivity), with less saturation at high Digital Number (DN) values.
  • B6_VCID_2: High-gain Thermal Infrared 1 (10.40 - 12.50 µm). Higher radiometric resolution (sensitivity), although it has a more restricted dynamic range.
  • B7: Short-wave infrared 2 (2.09 - 2.35 µm)
  • B8: Panchromatic (0.52 - 0.90 µm)


USGS Landsat 8 Raw Scenes (Orthorectified)

This dataset contains orthorectified images from Apr 11, 2013 - Apr 30, 2017 every 16 days with the following spectral bands (all bands are available at a resolution of 30 meters per pixel, except for band 8 that comes at a resolution of 15 m., and bands B10 and B11 were generated from sensor data with a resolution of 100 m.):

  • B1: Coastal aerosol (0.43 - 0.45 µm)
  • B2: Blue (0.45 - 0.51 µm)
  • B3: Green (0.53 - 0.59 µm)
  • B4: Red (0.64 - 0.67 µm)
  • B5: Near Infrared (0.85 - 0.88 µm)
  • B6: Short-wave Infrared 1 (1.57 - 1.65 µm)
  • B7: Short-wave infrared 2 (2.11 - 2.29 µm)
  • B8: Panchromatic (0.50 - 0.68 µm)
  • B9: Cirrus (1.36 - 1.38 µm)
  • B10: Thermal Infrared 1 (10.60 - 11.19 µm)
  • B11: Thermal Infrared 2 (11.50 - 12.51 µm)
  • BQA: Data quality assessment band


Sentinel-2: MultiSpectral Instrument (MSI), Level-1C

This dataset contains images representing TOA reflectance from Jun 23, 2015 - Nov 2, 2017 every 5 to 10 days (at the Equator) with the following spectral bands:

  • B1: Aerosols (443nm), 60m resolution
  • B2: Blue (490nm), 10m resolution
  • B3: Green (560nm), 10m resolution
  • B4: Red (665nm), 10m resolution
  • B5: Red Edge 1 (705nm), 20m resolution
  • B6: Red Edge 2 (740nm), 20m resolution
  • B7: Red Edge 3 (783nm), 20m resolution
  • B8: NIR (842nm), 10m resolution
  • B8a: Red Edge 4 (865nm), 20m resolution
  • B9: Water vapor (940nm), 60m resolution
  • B10: Cirrus (1375nm), 60m resolution
  • B11: SWIR 1 (1610nm), 20m resolution
  • B12: SWIR 2 (2190nm), 20m resolution


Remote sensing datasets and composites

This subsection list some examples of remote sensing datasets that have been calibrated and their accuracy have been assessed over a different locations and time periods via ground-truthing and validations.

MOD11A2.005 Land Surface Temperature and Emissivity 8-Day Global 1km

The MODIS global Land Surface Temperature (LST) and Emissivity 8-day data are composed of the daily 1-kilometer LST product (MOD11A1) and stored on a 1-km Sinusoidal grid as the average values of clear-sky LSTs during an 8-day period. It has the following bands:

  • LST_Day_1km: Daytime Land Surface Temperatures (K), Scale 0.02
  • QC_Day: Daytime Surface Temperature quality control assessments, see QC bit flags
  • Day_view_time: Daytime LST Observation Times (Hours), Scale 0.1
  • Day_view_angl: Daytime View Zenith Angles (Degrees), Offset -65.0
  • LST_Night_1km: Nighttime Land Surface Temperatures (K), Scale 0.02
  • QC_Night: Nighttime Surface Temperature quality control assessments, see QC bit flags
  • Night_view_time: Nighttime LST Observation Times (Hours), Scale 0.1
  • Night_view_angl: Nighttime View Zenith Angles (Degrees), Offset -65.0
  • Emis_31: Bands 31 Emissivity, Scale 0.002, Offset 0.49
  • Emis_32: Bands 32 Emissivity, Scale 0.002, Offset 0.49
  • Clear_sky_days: Clear Sky Day Coverage, see Clear Sky Flags.
  • Clear_sky_nights: Clear Sky Nighttime Coverage, see Clear Sky Flags.


Hansen Global Forest Change v1.3 (2000-2015)

  • treecover2000: tree canopy cover for year 2000 defined as canopy closure for all vegetation taller than 5m in height and encoded as a percentage per output grid cell, in the range 0–100.
  • loss: forest loss during the period 2000–2015 defined as a stand-replacement disturbance, or a change from a forest to non-forest state, encoded as either 1 (loss) or 0 (no loss).
  • gain: forest gain during the period 2000–2012 defined as the inverse of loss, or a non-forest to forest change entirely within the study period.
  • lossyear: the year of gross forest cover loss is a disaggregation of total forest loss to annual time scales, encoded as either 0 (no loss) or else a value in the range 1–14, representing loss detected primarily in the year 2001–2015, respectively.
  • datamask: the data mask has three values representing areas of no data (0), mapped land surface (1), and permanent water bodies (2).
  • Bands “first_30”, “first_40“, “first_50“, and “first_70“ are reference multispectral imagery from the first available year, typically 2000.
  • Bands “last_30”, “last_40“, “last_50“, and “last_70“ are reference multispectral imagery from the last available year, typically 2014.


Landsat 7 8-Day NDVI Composite

The Normalized Difference Vegetation Index is generated from the Near-IR and Red bands of each scene as (NIR - Red) / (NIR + Red), and ranges in value from -1.0 to 1.0. These composites are made from Landsat 7 Level L1T orthorectified scenes, using the computed top-of-atmosphere (TOA) reflectance for a periodicity of 8 days from Jan 1, 1999 to May 1, 2017.


Landsat 7 8-Day NDWI Composite

The Normalized Difference Water Index (NDWI) is sensitive to changes in liquid water content of vegetation canopies. These composites are made from Landsat 7 Level L1T orthorectified scenes, using the computed top-of-atmosphere (TOA) reflectance for a periodicity of 8 days from Jan 1, 1999 to May 1, 2017.


Landsat 7 8-Day EVI Composite

The Enhanced Vegetation Index (EVI) is generated from the Near-IR, Red and Blue bands of each scene, and ranges in value from -1.0 to 1.0. These composites are made from Landsat 7 Level L1T orthorectified scenes, using the computed top-of-atmosphere (TOA) reflectance for a periodicity of 8 days from Jan 1, 1999 to May 1, 2017.

Google Earth Engine

Google Earth Engine (GEE) combines a multi-petabyte catalog of satellite imagery and geospatial datasets with planetary-scale analysis capabilities and makes it available for scientists, researchers, and developers to detect changes, map trends, and quantify differences on the Earth's surface. It stores satellite imagery, organizes it, and makes it available for the first time for global-scale data mining. The public data archive includes historical earth imagery going back more than forty years, and new imagery is collected every day.

The GEE platform provides APIs in JavaScript and Python where it is possible to run own codes loading available datasets or uploading own imagery (e.g. UAV collected images or purchased images from private vendors). It is important to note that the algorithms, results and private images used in the platform are a property of the account holder and remain private until the owner publishes them[4].

The platform looks like this: GEE.png

To learn how to use it, please refer to the GEE Developers website where you can explore different documents explaining step by step, how to build codes within the API.


Back to Parent

This article is part of the topic Data Sources

Additional Resources

  • list here other articles related to this topic, with a brief description and link