Radiometric correction is done to reduce or correct errors in the digital numbers of images. The process improves the interpretability and quality of remote sensed data. Radiometric calibration and correction are particularly important when comparing data sets over a multiple time periods. The energy that sensors onboard aircrafts or satellites record can differ from the actual energy emitted or reflected from a surface on the ground. This is due to the sun's azimuth and elevation and atmospheric conditions that can influence the observed energy. Therefore, in order to obtain the real ground irradiance or reflectance, radiometric errors must be corrected for.
Radiometric Correction and Calibration
The value recorded for a given pixel includes not only the reflected or emitted radiation from the surface, but also the radiation scattered and emitted by the atmosphere. In most cases were are interested in the actual surface values. To achieve these values radiometric calibration and correction must be applied.
A sensor records the intensity of the electromagnetic radiation for each pixel as a digital number (DN). These digital numbers can be converted to more meaningful real world units like radiance, reflectance or brightness temperature. Sensor specific information is needed to carry out this calibration. In the case of Landsat data, the metadata file contains this information. Most image processing software packages have radiometric calibration tools. In ENVI some Landsat data can be converted directly to reflectance, with out needing to first calculate radiance.
Converting DNs to Radiance and Reflectance
The raw digital numbers (DN) in the images can be converted to what is known as top-of-atmosphere (TOA) radiance or reflectance. Equations rescale the data based on sensor specific information and removes the effects of differences in illumination geometry (different solar angle, Earth-sun distance). Most software packages have tools to calibrate the data.
Atmospheric correction is the process of removing the effects of the atmosphere to produce surface reflectance values. Atmospheric correction can significantly improve the interpretability and use of an image. Ideally this process requires knowledge of the atmospheric conditions and aerosol properties at the time the image was acquired.
Atmospheric Correction Models
Atmospheric models can be used to account for the effects of scattering and absorption in the atmosphere. A number of parameters are required to accurately apply atmospheric correction, including properties such as the amount of water vapor, distribution of aerosols. Sometimes this data can be collected by field instruments that measure atmospheric gases and aerosols, but this is often expensive and time consuming. Other satellite data can also be used to help estimate the amount and distribution of atmospheric aerosols. Many software packages include special atmospheric correction modules that use atmospheric radiation transfer models to produce an estimate of the true surface reflectance.
Dark Object Subtraction Method
This method is used when there is no available data on atmospheric conditions and aerosol properties at the time the image was acquired. The basic assumption of this method is that within the image some pixels are in complete shadow and their radiances received at the satellite are due entirely to atmospheric scattering (path radiance). This path radiance value is then subtracting from each pixel value in the image. The accuracy of these techniques are generally lower than physically-based corrections, but they are useful when no atmospheric measurements are available.
Landsat Land Surface Reflectance Products
Recently the USGS has developed software to apply calibration and atmospheric correction routines to Landsat level 1 data products. This data is known as Surface Reflectance data and is available for Landsat 4-5 Thematic Mapper (TM), Landsat 7 Enhanced Thematic Mapper Plus (ETM+) and Landsat 8 Operational Land Imager (OLI) data. This data can be ordered through EarthExplorer.
Mosaics created using Landsat 8 Level 1 data (left) and Provisional Landsat 8 Surface Reflectance data (right). Images created December 2014.
Image Source: USGS