 |
| Titel |
Evaluation of Soil Moisture Downscaling Algorithms for the SMAP Mission |
| VerfasserIn |
Xiaoling Wu, Jeffrey Walker, Christoph Rüdiger, Rocco Panciera |
| Konferenz |
EGU General Assembly 2014
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 16 (2014) |
| Datensatznummer |
250087966
|
| Publikation (Nr.) |
 EGU/EGU2014-2024.pdf |
|
|
|
|
|
| Zusammenfassung |
| The Soil Moisture Active Passive (SMAP) satellite is scheduled for launch by NASA in
November 2014, with the aim to provide a medium-resolution soil moisture product at the
global scale and with 2-3 days revisit frequency. The rationale behind this mission is that
the synergy between 3 km resolution active (radar) and 36 km resolution passive
(radiometer) observations can be used in a downscaling approach to overcome
the individual limitations of each observation, ultimately providing soil moisture
data at a resolution suitable for hydro-meteorological applications, on the order
of ~9 km. Two soil moisture downscaling approaches were tested in this study:
i) the baseline downscaling algorithm proposed for SMAP, which is based on an
assumption of linear relationship between radiometer and radar observations, with the
downscaled radiometer data then converted to a soil moisture product using the
passive microwave retrieval method; ii) the optional downscaling algorithm for
SMAP, which is based on an assumption of a directly linear relationship between soil
moisture and the radar observations. Data used to evaluate these two approaches
were collected from the Soil Moisture Active Passive Experiments (SMAPEx) in
south-eastern Australia, which closely simulate the SMAP data stream using airborne
observations for a single SMAP radiometer pixel over a 3-week interval. Both approaches
were compared to a reference soil moisture map retrieved from 1 km resolution
radiometer data. Results indicated that radar observations at vv-polarization had
the best correlation with radiometer observations or soil moisture data than hh- or
hv-polarization, thus having best performance during downscaling procedure. These two
downscaling approaches showed similar performance in terms of accuracy, with a
Root-Mean-Square Error (RMSE) in downscaled soil moisture data around 0.02 cm3/cm3,
when downscaled to 9 km resolution. This increased to 0.043 cm3/cm3 when applied at 1 km
resolution. Results indicated both downscaling methods had the ability to fulfill the
error target of SMAP, with a RMSE less than 0.04 cm3/cm3 at 9 km resolution. |
|
|
|
|
|
|