 |
| Titel |
SMOS L1c and L2 product validation with airborne and in-situ observations over south-eastern Australia |
| VerfasserIn |
Christoph Rüdiger, Jeffrey P. Walker, Yann H. Kerr, Olivier Merlin, Arnaud Mialon, Edward J. Kim |
| Konferenz |
EGU General Assembly 2011
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 13 (2011) |
| Datensatznummer |
250050114
|
|
|
|
|
|
| Zusammenfassung |
| In January/February and September 2010, the Australian Airborne Cal/val Experiments for SMOS (AACES), two extensive airborne field campaigns, were undertaken in south-eastern Australia for the validation of the L1c (brightness temperature) and L2 (surface soil moisture) products of the SMOS mission. The field sites chosen are found within the 82,000km2 Murrumbidgee River catchment, located between 33-37°S and 144-151°E. The environmental conditions across the catchment - ranging from semi-arid plains in the west, to irrigated cultivated areas in the central part, and temperate subalpine conditions in the east - along with the existing soil moisture monitoring infrastructure make it an exceptionally well suited catchment for such satellite validation activities.
During the summer campaign (January/February 2010), a total area of 50,000km2 was covered with an airborne monitoring platform (the Polarimetric L-band Multichannel Radiometer), requiring 10 flight days. In September 2010, the winter campaign covered a smaller are of 25,000 km2, including 20 SMOS footprints. Each individual flight day was scheduled to coincide with a SMOS ascending overpass (6am local time).
In support of the airborne data collection, additional ground measurements were taken across a number of focus farms, including soil moisture and temperature, leaf area index (LAI), hyperspectral measurements of the vegetation, destructive samples of vegetation for vegetation water content and dry biomass, and thermogravimetric samples for calibration and soil property analysis. These ground measurements were obtained through in-situ measurements across two focus farms in each 100km x 50km area, with an extent of 5km x 2km each. While the soil moisture and temperature measurements were taken along 6 parallel transects every 50m, the vegetation measurements were collected within smaller (representative) vegetation sampling areas.
The initial release of the SMOS L1c brightness temperature data set was compared directly to the airborne data for validation purposes. It was found that a persistent over- (brightness temperature) and underestimation (soil moisture) exists within those early SMOS data throughout the measurements (for the data sets released operationally during the campaigns). This bias was found across all vegetation and climate conditions and may therefore not be related to the surface conditions.
In addition, the airborne data set was also used to study the spatial coverage requirements for airborne validation campaigns. Here it was found that the minimum fraction of a satellite footprint to be covered so as to achieve an uncertainty in the average of airborne data (due to incomplete sampling) not greater than the 4K design accuracy of SMOS was typically 50% of the footprint size. Moreover, it can be shown that this value is linearly dependent on the spatial variability within the full SMOS-sized footprint. |
|
|
|
|
|
|