Summary
Contents
Subject index
Remote sensing acquires and interprets small or large-scale data about the Earth from a distance. Using a wide range of spatial, spectral, temporal, and radiometric scales remote sensing is a large and diverse field for which this Handbook will be the key research reference. This Handbook is organized in four key sections: • Interactions of Electromagnetic Radiation with the Terrestrial Environment: chapters on Visible, Near-IR and Shortwave IR; Middle IR (3-5 micrometers); Thermal IR; Microwave • Digital sensors and Image Characteristics: chapters on Sensor Technology; Coarse Spatial Resolution Optical Sensors; Medium Spatial Resolution Optical Sensors; Fine Spatial Resolution Optical Sensors; Video Imaging and Multispectral Digital Photography; Hyperspectral Sensors; Radar and Passive Microwave Sensors; Lidar • Remote Sensing Analysis: Design and Implementation: chapters on Image Pre-Processing; Ground Data Collection; Integration with GIS; Quantitative Models in Remote Sensing; Validation and accuracy assessment; • Remote Sensing Analysis: Applications: LITHOSPHERIC SCIENCES: chapters on Topography; Geology; Soils; PLANT SCIENCES: Vegetation; Agriculture; HYDROSPHERIC and CRYSOPHERIC SCIENCES: Hydrosphere: Fresh and Ocean Water; Cryosphere; GLOBAL CHANGE AND HUMAN ENVIRONMENTS: Earth Systems; Human Environments & Links to the Social Sciences; Real Time Monitoring Systems and Disaster Management; Land Cover Change Illustrated throughout, an essential resource for the analysis of remotely sensed data, The SAGE Handbook of Remote Sensing provides researchers with a definitive statement of the core concepts and methodologies in the discipline.
Quantitative Models and Inversion in Optical Remote Sensing
Quantitative Models and Inversion in Optical Remote Sensing
Keywords
- radiative transfer
- land
- modeling
- inversion
- biogeophysical parameters.
Introduction
In the past several decades, vast amounts of remotely sensed Earth observations have been acquired and accumulated. With a series of new satellite programs planned, the data volume will continue to increase. Automating the procedures for processing and analyzing these data is critical. Furthermore, various numerical models characterizing Earth's environments, each associated with a different decision support system, have to be calibrated and run with spatially and temporally explicit data sets produced only from remotely sensed data at the appropriate scales.
Quantitative estimation of land surface variables from satellite observations is challenging. It requires not only the understanding of the remotely sensed signals through the physical modeling approaches, but also effective inversion algorithms. This chapter focuses on models and inversion algorithms for estimating land surface variables with the emphasis on the visible and near-infrared (IR) spectrum. The next section provides the overview of physical modeling of remote sensing signals. It starts with atmospheric radiative transfer, followed by surface radiation modeling and the sensor models. The following section presents the state-of-the-art inversion algorithms and summarizes some of the key products generated using these algorithms, followed by a section on validation with ground measurements.
Modeling Techniques
To understand remote sensing signals, land surface variables must be linked to at-sensor radiance recorded at the top (or middle) of the atmosphere using various physically based models (Liang 2004). These models can be used for simulating remote sensing signals by changing surface and atmospheric conditions, and for developing practical inversion algorithms to estimate land surface variables from satellite observations. Physically based models in three areas are discussed below: atmosphere, land surface, and sensor.
Atmospheric Radiative Transfer Modeling
The signals recorded by a sensor include both atmosphere and surface information. Atmospheric gases, aerosols, and clouds scatter and absorb incoming solar radiation and the reflected and/or emitted radiation from the surface. The atmosphere greatly modulates the spectral dependence and spatial distribution of the surface radiation. In order to estimate land surface variables, it is very important to understand the radiative transfer within the atmosphere so that atmospheric effects can be effectively removed from remotely sensed data.
If the atmospheric particles can be assumed to be isotropic, the one-dimensional (1D) radiative transfer equation for radiance I(μ, φ) at any direction (Ω: {μ, φ}) in the solar spectrum can be written as:
where μ is the consine of the viewing zenith angle, φ is the viewing azimuth angle, x is called optical depth or optical thickness, depending on the geometric height z and the extinction coefficient (σe)τ = ∫z0 σe(z)dz, ω is the single scattering albedo, and P is the phase function. These optical parameters (σe, ω, and P) can be determined from the aerosol models that are characterized by the particle size distribution, refractive index and so on.
Eq. (1) cannot be solved unless the boundary conditions are specified. For a 1D atmosphere, two boundary conditions are needed. At the top of the atmosphere, there is only direct solar radiation. The lower boundary condition at the Earth surface is characterized by the surface bidirectional reflectance factor (BRF).
...
- Loading...
Get a 30 day FREE TRIAL
-
Watch videos from a variety of sources bringing classroom topics to life -
Read modern, diverse business cases
-
Explore hundreds of books and reference titles
Read next
More like this
Sage Recommends
We found other relevant content for you on other Sage platforms.
Have you created a personal profile? Login or create a profile so that you can save clips, playlists and searches