Mapping Global Aerodynamic Roughness Length of Land Surface at 1 km Scale (Aerodynamic Roughness and Vegetation Structure of Amazon Basin)
Regina Célia dos Santos Alvalá Centro de Previsão de Tempo e Estudos Climáticos/INPE (SA-PI)
Sassan Sepehri Saatchi JPL/CALTECH (US-PI)
atmospheric boundary layer problems over various landscapes, the fluxes of
sensible heat, latent heat and carbon dioxide are determined in terms of surface
characteristics. One of the
main parameters used in Global Circulation Models (GCM) or in
soil-vegetation-atmosphere transfer (SVAT) models is the aerodynamic roughness
length (Z0). This
parameter is important to determine the vertical gradients of mean wind speed
and the conditions for momentum transfer over a vegetated or bare rough surface.
Over vegetated surfaces, the aerodynamic roughness length has a simple
one-to-one relationship with the rms height
variation of the vegetation at the top of the canopy.
Once this roughness length is determined for a surface, it does not
change with wind speed, stability
or stress. Because the ground
measurement of this parameter is difficult, time consuming and expensive,
reliable data on its distribution and quantity over global land cover types,
even at the coarse resolution of GCM models, do not exist.
As a result, even the spatially distributed models use values from very
limited and sparse field measurements reported in the literature.
In this work, we propose to use the SRTM data in conjunction with limited
calibration data from the VCL system to generate a global aerodynamic roughness
parameter at 1 km spatial scale to be used in regional and global scale climate
and land-atmospheric models. The
phase and amplitude of the cross correlation of the two complex images obtained
by SRTM is known to be sensitive to surface height which is defined as the
distance between the scattering phase centers and a reference line.
Even though the estimation of absolute height of vegetation from the data
can be erroneous due to surface variability and the limited penetration of the
signal into the canopy, but the variation of the height can be obtained
accurately. By averaging the estimated values from 30 meter resolution of SRTM
to 1 km pixels, the estimation of the rms height of the surface can be improved
to a centimeter accuracy. An algorithm has been developed to estimate this
parameter from SRTM data at the Jet Propulsion Laboratory and has been tested
using the airborne interferometric data (AIRSAR). To improve this algorithm and
validate the estimation of the parameter from the SRTM imagery, the Amazon basin
has been chosen as the primary test site. This
will allow the use of field measurements, satellite data, and process models to
test the application of the algorithm. The
LBA data sets used for this are: (1) vegetation structure data from various
primary, secondary and woodland sites, (2) tower measurements of aerodynamic
roughness, (3) VCL (Vegetation Canopy Lidar) data of vegetation height from
limited orbital data processed immediately after the mission launch, (4)
airborne laser altimeter data and possible AIRSAR data acquired by the LBA project,
(5) ecosystem/atmospheric process models for sensitivity analysis and
integration of spatial distribution of roughness data over the basin.
To produce a global map of aerodynamic roughness at 1km scale.
To produce a validated map of the aerodynamic roughness over the Amazon
To augment the vegetation map of the Amazon basin with structural
information derived from SRTM data.
To incorporate the aerodynamic roughness distribution in regional
atmospheric modeling system (RAMS).
The work consists of (1) physically based modeling of SRTM interferometric data over the
Amazon basin in order to enhance the application of the algorithm to various
vegetation types; (2) SRTM data analysis
and the estimation of aerodynamic roughness over the entire Amazon; (3) laser data analysis and the estimation of vegetation
height and roughness will be performed as soon as the data from VCL or from
airborne laser altimetry becomes available.
(3) ground studies at various
sites within the LBA project to determine the relationship between the
aerodynamic roughness and the physical roughness of the vegetation and to
produce a data set for validation; (4) modeling
to integrate the aerodynamic roughness distribution derived from SRTM in
regional atmospheric processes and to assess the sensitivity of processes to
variations of roughness length over different vegetation types.
Last Updated: August 30, 2000