Geoscience Reference
In-Depth Information
In the tropical oceans, large vertical salinity gradients can develop in the upper few
metres of the ocean after heavy rainfall as evidenced by Soloviev and Lukas (
1996
);
Schl¨ ssel et al. (
1997
); and Wijesekera et al. (
1999
). Signatures of these intense precipi-
tation regimes can be detected in the SMOS Sea-Surface Salinity (SSS) data in the form of
freshwater patches, as clearly shown by Reul et al. (
2013
).
3 Soil Moisture
Numerous soil moisture products are available from active experiments (ERS, ASCAT)
(Naeimi et al.
2009
; Wagner et al.
1999
; Loew et al.
2006
), or from passive sensors (AMSR-E,
SMOS) (Kerr et al.
2010
; Njoku et al.
2003
; Loew et al.
2013
) and will be available as well
from combined passive/active microwave remote sensors (e.g., the planned Soil Moisture
Active Passive (SMAP) mission). These operate at coarser ([25 km) resolutions and span
altogether more than three decades of data. Recently, a multi-decadal blended dataset was
developed that is expected to further enhance the understanding of the water balance in
hydrological models (Dorigo et al.
2012
; Liu et al.
2011
).
The application of coarse resolution soil moisture data in hydrological models is con-
troversial. There seems to be no obvious approach in river run-off studies that would
explain under which conditions an improvement could be achieved. Recently, however,
several studies demonstrated positive impact when Soil Water Index (SWI) was assimi-
lated (Brocca et al.
2010a
,
b
; Matgen et al.
2011
; Meier et al.
2011
; Wagner et al.
1999
).
SWI represents the profile of soil moisture in the root zone which is the hydrological most
important zone in terms of run-off generation (Parajka et al.
2006
).
Semi-operational products are also available at medium-resolution scale ([1 km) (Pathe
et al.
2009
). Nevertheless, assimilation of such data into models was restricted by poor
radiometric resolution and revisit period. A soil moisture product from Sentinel-1 has been
foreseen with coverage every 6 days globally, nearly daily over Europe and Canada
(depending on latitude) (Hornacek et al.
2012
). With its remarkably improved radiometric
accuracy, it has the potential to be of great benefit for data assimilation, anomaly and
threshold detection as well as direct input into models operating at medium-resolution
scales (Doubkov
´
et al.
2012
).
4 Evapotranspiration
Neither evapotranspiration (ET) nor any of its components can be directly sensed from
satellites, as heat fluxes do not absorb nor emit electromagnetic signals directly. None-
theless, the last three decades have seen substantial progress in the combined field of
evaporation and remote sensing. Current methodologies concentrate on the derivation of
ET by combining some of the satellite-observable physical variables that are linked to the
evaporation process. Some of the existing algorithms differ in their purpose of application,
which to a certain extent defines the type of remote sensing data used and the amount of
required ancillary data. The majority use some form of thermal and visible data, with only
a few applying microwave observations. Some of these methodologies are fully empirical,
and others are based on more physically based calculations of ET via formulations like the
ones of Penman (
1948
), Monteith (
1965
), and Priestley and Taylor (
1972
), or focus on
solving the surface energy balance targeting the accurate determination of the sensible heat
flux (H). Most of the early methods were designed for local-scale studies and agricultural
