Soil Moisture and Ocean Salinity (SMOS)
(MIRAS on RAMSES)
The baseline SMOS payload is an L-band (1.4 GHz), Y shaped, 2D interferometric
radiometer. It is proposed to launch SMOS on a sun synchronous orbit.
The main objective of the SMOS mission is to deliver crucial variables of the land
surfaces: soil moisture, and of ocean surfaces: sea surface salinity fields. The mission should
also deliver information on root zone soil moisture, vegetation and biomass, and lead to
significant research in the field of the cryosphere.
Jean-Christophe Calvet is co-ordonator of the land climatology and meteorology sub-group.
SCIENTIFIC OBJECTIVES OVER LAND
From all the lower boundary conditions which drive the atmosphere, land-surfaces are
of particular interest to mankind as their direct and local impact is of great importance to
human activities. The challenges posed by land-surfaces for all meteorological and
climatological applications lie in the fact that they are very variable over a broad range of
temporal and spatial scales. In contrast to oceans for instance, diurnal variations of
temperature and fluxes are an order of magnitude larger. Another major difference is that
moisture for evaporation is available in limited supplies but constitutes at the same time a
memory for the system. The surface hydrology is one of the keys to our understanding of the
interaction between the continental surfaces and the atmosphere as it determines the
portioning of energy between the different fluxes. The proposed SMOS project is a research
mission intended to apply a new instrumental technique to tackle a number of scientific
objectives, mainly in the field of meteorology and climatology: weather forecast, climate
change sensitivity studies, and data analysis. The science issues considered in this proposal
are related to the parameterization of land surface processes and to the development of
methods to retrieve surface variables from satellite data, in order to improve the
representation of surface fluxes, soil moisture content, soil hydraulic characteristics, and plant
stress in mesoscale and global models. The initialisation of soil moisture in atmospheric
models, including numerical weather forecast models, is of great concern and a subject of
active research. The current methods to estimate soil moisture are all very indirect and other
ways of inferring soil moisture, with global coverage, are needed.
For watershed hydrologic model applications, there is an urgent need to have access to
distributed soil water fluxes at regular temporal resolutions over large areas. The yearly
integrated land surface and base flow water budgets are generally well predicted by the new
generation of hydrologic models. However, the estimation of the ratio between base flow and
surface runoff, as well as the ratio between deep drainage and soil moisture content are still
very imprecise. The soil stratum and in particular the unsaturated zone between the soil
surface and the groundwater table (vadose zone) plays a crucial role: the estimation of soil
moisture in the vadose zone is an important issue for short and medium term meteorological
modelling, hydrological modelling, and the monitoring of plant CO2 assimilation and plant
growth. The vadose zone hydrology being inaccurately described, attempts to monitor water
quality and flooding risks often fail. New ways to parameterise effective soil characteristics
are needed.
In most cases, the vadose zone hydrology and the surface fluxes are controlled by
vegetation. Modelling the rate of soil water extraction by the plant roots and the stomatal
feedback is important for either atmospheric, hydrologic, and environmental studies. The
current models manage to describe first order responses but do not encompass the complete
behaviour of the plant, especially at the mesoscale, where several landscape units may
contribute to the surface fluxes. In most parts of the world, plant water supply is the dominant
factor that affects plant growth and crop yields. Monitoring soil moisture is an interesting
way to detect water stress period (excess or deficit) for yield forecasting or biomass
monitoring, especially in areas where the density of climatic stations is low. Time series of
soil moisture at the mesoscale would be a very interesting input to the representation of
vegetation in land surface schemes, also.
Concerning the characterization of the atmosphere, there is a need to estimate the
surface emissivity at the wavelengths of the atmospheric sounders in order to improve the
retrievals. The all-weather surface characterisation capability of L-band radiometry could
help estimate these values.
These objectives may be achieved by estimating near-surface soil moisture content
(wS). In most cases, the root-zone or the vadose-zone soil moisture (wVZ) is required, also,
together with effective hydraulic soil characteristics. In the next section, it is explained how
the latter variables can be inferred from wS, provided this variable is monitored with a good
(3-4 days) temporal resolution.
The use of the near surface soil moisture wS to help characterise the surface fluxes,
bulk soil moisture and plant stress, must be considered through assimilation and aggregation
or disaggregation techniques.
Water storage in the soil, either in the top surface layer (e.g. 5 cm) or in deeper layers,
affects not only direct evapotranspiration but the heat storage ability of the soil, its thermal
conductivity, and the partitioning of energy between latent and sensible heat flux. It is
therefore a key variable of landsurface-atmosphere interaction.
The value of the top surface layer wS conditions direct evaporation from bare soil
or soil partially covered by vegetation, and determines the possibility of surface runoff
after rainfalls. This parameter is currently not extensively measured.
The vadose zone (wVZ) is the hydrological connection between the surface water
component of the hydrological cycle and the groundwater component. Evaporation,
infiltration and recharge of the groundwater usually occurs through the unsaturated zone.
Because of root water extraction, the vadose zone is the interface between the vegetation and
the hydrological systems: the value of wVZ conditions plant transpiration and CO2 uptake
through stomatal aperture and possible damage to the photosynthesis apparatus. Furthermore,
wVZ is directly linked to the ability of the soil to produce drainage after a rainfall. The soil-
vegetation-atmosphere transfer (SVAT) schemes now employed in meteorology and
hydrology are designed to describe the basic evaporation processes at the surface together
with the water partitionning between the vegetation transpiration, the drainage, the surface
runoff and the soil moisture increase or decrease. The current trend in SVAT modelling is the
integration of biological processes such as photosynthesis and plant growth, and hydrological
transfers, in the same surface model. The 'classical' part of the SVAT performs the
atmosphere interface calculations, while new modules provided by the research in physiology
and hydrology simulate interactive vegetation and river flow. This is why improving SVAT
modelling would benefit to meteorology, climatology, hydrology, and agronomy. In
operational simulations, a realistic initial value of wVZ must be provided to the SVAT model.
One of the main difficulties in the use of such parameterisations is the initialisation of wVZ:
soil wetness is one of the least understood and poorly simulated components of the climate
system and is also one of the most sparsely measured.
Water movement in the unsaturated zone is affected by intrinsic parameters such
as hydraulic characteristics depending upon structural properties and, to a lesser
extent, upon texture. The textural properties are characterised by a smaller underlying
variability than the structural properties, which is mainly due to the fact that biological and
anthropogenic factors have a much larger impact on structural properties than on textural
ones. Most SVAT models rely on the use of pedotransfer functions which estimate soil
characteristics from readily available data, such as texture (i.e. particle size distribution)
which are the most common measured soil data across the world. However, pedotransfer
functions are doomed to fail when used for the estimation of structural parameters.
Since microwave techniques provide information about the moisture wS of a shallow
surface layer (about 5 cm at L-band), only, it was investigated to what extent vadose zone soil
moisture wVZ and soil hydraulic characteristics can be inferred from near surface soil
moisture.
Time series of surface soil moisture content (wS) assessed by L-band microwaves
allow for the determination of wVZ and for that of the surface fluxes
(evaporatranspiration). When dealing with bare (or sparecely covered) soils, evaporation
rate and runoff can be calculated from wS time series. When dealing with soil surfaces
covered with vegetation, information on the vadose zone soil moisture (wVZ) is generally
needed to describe water and energy fluxes in the soil-plant-atmosphere continuum.
Furthermore, the inclusion of CO2 assimilation algorithms in SVAT schemes made it
potentially possible to simulate vegetation growth and, hence, to explore biosphere feedback
mechanisms in response to changes in rainfall patterns, temperature, and soil water store.
Initially, the problem was tackled through assimilation studies using meteorological satellites
observations. However it was shown that only under dry conditions the diurnal change in
surface skin temperature could be used to retrieve the soil moisture store. Hence, the
applicability of satellite infra-red data is limited. Attempts were made to retrieve the root-
zone soil moisture from screen level variables (air temperature and humidity) by inverting
simple surface schemes: in some conditions, it is possible to adjust the value of the root-zone
soil moisture in order to minimize the forecast error on the low-level atmospheric parameters.
A difficulty of this method is that the link between screen-level parameters and the root-zone
soil moisture is rather indirect. The consequence is that the method does not work when the
energy available at the surface is too low (e.g. short diurnal cycles in winter) or when the
wind speed is too high. In a recent study
Calvet et al. (1998) tested the possibility of using an
operational SVAT scheme to retrieve root-zone soil moisture and evapotranspiration fluxes
from in situ measurements of the near-surface soil moisture content (the 5 cm top layer)
applying atmospheric and precipitation forcing. The SVAT scheme chosen for the inversion
procedure was the ISBA code developed by
Noilhan and Planton (1989). The experimental
data collected during the MUREX (South of France) field campaign (3 years) were used for
ground validation. The authors derived assimilation rules for either surface soil moisture or
surface temperature and showed that 4 or 5 wS values measured once every four days are
sufficient to retrieve wVZ as well as the evapotranspiration flux.