Terrestrial Ecology
Table of contents
Data Acquisition
ALOS PALSAR
Launched on January 24th, 2006 aboard ALOS, the PALSAR instrument has promising applications for natural resource and land applications including parameters applicable to terrestrial nutrient cycle estimates. Research has shown that SAR data by itself, or combined with other optical or active systems, can enhance land characterization with information not otherwise available from passive remote systems. Because Radar emits its own signal, imaging can occur any time of the day or night independent of sun angle. This contrasts passive imaging systems that require the Sun’s illumination. Due to its longer wavelength than visible light, the microwaves used in Radar also have the advantage of not being impeded by cloud cover or other atmospheric contamination. Some examples of PALSAR land applications include estimating and mapping vegetation above ground biomass, deforestation mapping, wetland (including high latitude) characterization, and cropland monitoring.
Radar Imaging Basics
The PALSAR instrument is a type of SAR that emits energy in the long wavelength L-Band (1270 MHz) frequency. SAR radar systems can generate high-resolution imagery with a synthetic aperture (or virtual long antenna) by combining signals received by the physically short (real) antenna as it moves along its flight track.
As an imaging radar system moves along a flight path, it emits and receives pulses in a single particular microwave wavelength and orientation (waves polarized in a single vertical (V) or horizontal (H) plane). The radar pulse interacts with the Earth’s surface and is scattered in all directions, with some energy reflected back toward the radar’s antenna. Known as backscatter, the returned signal is received by the antenna a fraction of a second later and in a specific polarization (H or V). The brightness, or amplitude, of the backscatter is measured and recorded, and the data are used to derive an image. Radar waves interact differently with soil, vegetation, water, ice, and man-made objects such as buildings and roads because the backscatter is affected by the surface properties of objects. For a smooth surface such as water or a road, most of the incident energy is reflected away from the radar system resulting in a very low return signal. In contrast, rough surfaces will scatter the emitted energy in all directions and return a significant portion back to the antenna. In general, vegetation is usually moderately rough with most radar wavelengths.
PALSAR Terrestrial Biophysical Applications
The data can be used for a number of purposes: (1) to validate the SAR measurements using flux tower site characterization data; (2) to examine the impacts of vegetation dynamics on climate; (3) to understand human impacts on vegetation at a local scale; (3) to detect deforestation and forest degradation; (4) to map and differentiate growth stages and change; (5) to retrieve woody biomass and structural attributes; and (6) to characterize, map and monitor ecoregions such as mangroves and wetlands.
Satellite radar can be important to Earth system monitoring because the properties of the signal return are better suited for certain vegetative biophysical estimates and are more accurate or not otherwise obtainable by passive remote sensing systems. A number of studies have shown a significant relationship between L-Band SAR backscatter coefficients and forest structure parameters including above-ground biomass and vegetative structural attributes. Other examples of terrestrial applications include wetland characterization, mapping, and monitoring forest change analysis.
The PALSAR subsets provided in this data set might be useful for visual interest and preliminary analysis of the field area. For in-depth analyses, such as biomass estimation, vegetation characterization, etc., users might have to download the lower level products from ASF.
Quantitatively Comparing Multi-Temporal Data
The data values in the image are Digital Numbers (DN) that can be used in the following equation to extract the Normalized Radar Cross Section (NRCS).
NRCS (dB) = 10*log10(<DNˆ2>) + CF
Where the Calibration Factor (CF) is a constant -83.
The cross section parameter is useful to quantitatively compare multi-temporal data.
Data formats and files
Each scene is available in GeoTIFF format with a XML file containing metadata. A PDF document file is also available for each site and contains image-specific metadata, image analysts notes about channel assignments and colors, and a thumbnail of the SAR image. For sites with multiple images, only one thumbnail is included as images are quite similar in appearance.
Example PDF File:
GeoTIFF: Baltimore_Ecosystem_Study_20090728.tif
Imagery Date: 28-Jul-2009, 03:32:51
Location: Maryland
The SAR image has greens, pinks, blacks and some gray colors. By assigning the HH to the red and blue channels
and the HV polarization to the green, we get these slightly more intuitive colors. Green is tree canopy. Black is water.
Polarization: FBD 34.3 HH+HV
Bands: HH (red and blue)
Bands: HV (green)