|
Titel |
Copula based assimilation of modeled and observed soil moisture on catchment scales |
VerfasserIn |
Christof Lorenz, Stefanie Vogl, Harald Kunstmann |
Konferenz |
EGU General Assembly 2013
|
Medientyp |
Artikel
|
Sprache |
Englisch
|
Digitales Dokument |
PDF |
Erschienen |
In: GRA - Volume 15 (2013) |
Datensatznummer |
250081413
|
|
|
|
Zusammenfassung |
For several years, satellite missions like SMOS or AMSR-E provide spatially distributed
information on soil moisture. However, this data is mainly restricted to the surface
or the first few centimeters in the soil e.g. due to a limited penetration depth of
L-Band radar and might show large uncertainties over heavily vegetated areas.
This holds also true for airborne observations, while they further suffer from a very
limited temporal availability. Therefore, one still has to rely on models and in situ
observations, if the water content of deeper soil layers is of interest. Significant differences
certainly occur when modeled and observed total soil moisture is compared due to the
models’ discrete soil levels or uncertainties in the driving fields like precipitation
or evaporation. This study focuses on the assimilation of soil moisture from the
high-resolution hydrologic models NLDAS NOAH and NLDAS MOSAIC with in situ
observations from the International Soil Moisture Network (ISM) over a sub-region of the
Mississippi basin. The spatio-temporal dependence structure between time-series of
terrestrial observations and model-based fields of soil moisture is analyzed. First
investigations, including the application of suitable goodness of fit tests, show that standard
Copula models, such as the Frank or Clayton Copula, are well suited to describe this
dependency. This information is used to perform an event-based assimilation of
modeled and observed soil moisture which incorporates (1) a bias correction to
improve the absolute values and (2) an improvement of the spatial patterns based on
the derived dependence structures. The assimilated fields are then cross-validated
with observed soil moisture in order to quantify the gain of the applied method. |
|
|
|
|
|