documentation - CP34-BEC

File: BEC-SMOS-0001-PD.pdf , version 1.3
Title: SMOS-BEC Ocean and Land Products Description.
Authors: SMOS-BEC Team.
Contact: [email protected]
Date: 19/09/2014
File: BEC-SMOS-0001-PD.pdf
Version: 1.3
Date: 19/09/2014
Page: 1 of 20
SMOS-BEC OCEAN AND LAND PRODUCTS DESCRIPTION
Abstract: This technical note describes the products distributed by the SMOS-BEC team through its
data visualization and distribution service CP34-BEC http://cp34-bec.cmima.csic.es
File: BEC-SMOS-0001-PD.pdf
Version: 1.3
Date: 19/09/2014
Page: 2 of 20
Title: SMOS-BEC Ocean and Land Products Description.
Contents
1 Introduction
4
2 Ocean Products
5
2.1
SMOS ocean salinity data filtering . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5
2.1.1
Geophysical filters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5
2.1.2
Retrieval filters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5
2.1.3
Geometrical filters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
6
Ocean salinity Level 3 products . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
6
2.2.1
Binned products . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
6
2.2.2
Optimal interpolation products . . . . . . . . . . . . . . . . . . . . . . . . . . .
6
Ocean salinity Level 4 products . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
8
2.3.1
Fused products using singularity analysis techniques . . . . . . . . . . . . . . .
8
2.4
Ocean salinity reprocessing campaign . . . . . . . . . . . . . . . . . . . . . . . . . . . .
8
2.5
Ocean auxiliary data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
9
2.5.1
Singularity exponents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
9
2.6
Ocean files structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
11
2.7
Ocean products list . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
12
2.2
2.3
3 Land Products
3.1
3.2
3.3
Soil moisture Level 3 products
13
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
13
3.1.1
Soil moisture data filtering . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
13
3.1.2
ISEA land product . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
14
3.1.3
Binned land products . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
14
Soil moisture Level 4 products
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
16
3.2.1
High resolution soil moisture: delayed . . . . . . . . . . . . . . . . . . . . . . .
17
3.2.2
High resolution soil moisture: near real-time
. . . . . . . . . . . . . . . . . . .
17
Land files structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
17
File: BEC-SMOS-0001-PD.pdf
Version: 1.3
Date: 19/09/2014
Page: 3 of 20
Title: SMOS-BEC Ocean and Land Products Description.
3.4
Land products list . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
18
File: BEC-SMOS-0001-PD.pdf
Version: 1.3
Date: 19/09/2014
Page: 4 of 20
Title: SMOS-BEC Ocean and Land Products Description.
1
INTRODUCTION
The ESA’s Soil Moisture and Ocean Salinity (SMOS) mission is an innovative Earth Observation
satellite launched on November 2009 to remotely sense soil moisture over the land surfaces and sea
surface salinity over the oceans ([Kerr et al., 2010], [Font et al., 2010]). The SMOS single payload
is the Microwave Imaging Radiometer using Aperture Synthesis (MIRAS), a L-band 2D synthetic
aperture radiometer with multiangular and full polarimetric capabilities. It is a completely new type
of instrument, a technological challenge that has required the development of dedicated calibration
and image reconstruction algorithms ([McMullan et al., 2008]). The SMOS Barcelona Expert Center
(BEC) is an ESA Expert Support Laboratory dedicated to developing and testing new algorithms to
improve the baseline SMOS Level 2 products. Also the BEC aims at generating higher added-value
products of interest for a broad range of users. The SMOS-BEC products for sea surface salinity
and soil moisture are generated and distributed through the Production Center of Level 3 and 4
(CP34) since the beginning of the mission in an operational way. In the near future, the inclusion of
complementary remotely sensed products is envisaged.
This document describes the products currently created and distributed by the BEC through the
CP34.
File: BEC-SMOS-0001-PD.pdf
Version: 1.3
Date: 19/09/2014
Page: 5 of 20
Title: SMOS-BEC Ocean and Land Products Description.
2
OCEAN PRODUCTS
2.1
SMOS ocean salinity data filtering
The SMOS data used to compute the Ocean products described in section 2.2 (and in addition to
their derivatives of Level 4) come from Level 2 Ocean Salinity User Data Product (UDP) and Ocean
Salinity Data Analysis Product (DAP). These UDP and DAP files are generated by ESA and include geophysical parameters, a theoretical estimate of their accuracy, flags, and descriptors for the
product quality for three different roughness models (see [DPG, 2012, section 4.2.2.1] for a detailed
description of this product). All products developed at BEC are based on the third roughness model
[Guimbard et al., 2012].
The quality flags and descriptors from UDP and DAP files allow discarding unreliable Sea Surface
Salinity values. In order to create Level 3 and Level 4 products, three categories of filters are applied
to ocean Level 2 data: geophysical filters, retrieval filters and geometrical filters. Each filtering process
is coded using a 7 characters string that appears in the name of the resulting products (string EEEEEEE
in section 2.6). The current filtering process, coded as 2013001, follows the rules described in sections
2.1.1, 2.1.2 and 2.1.3
2.1.1
Geophysical filters
These filters are related to the geophysical conditions present in the area (grid point) and the time of
measurement [DPG, 2012, tables 4-19 to 4-21]. Retrieved salinity in a given gridpoint is discarded if
any of the following conditions is accomplished:
• Suspect ice presence (more than 50% of measures having a positive test ice)
• Rain (rain rate larger than 0.01 mm/h)
• High number of outliers (more than 20% of measures)
• Too many measures flagged for sunglint or moonglint (10%)
• Salinity is retrieved using a too low number of valid measures (less than 30 brightness temperature valid measures)
• Wind speed is larger then a given threshold (set to 12 m/s)
• Grid point is suspected of being contaminated by RFI (more than 33% of RFI outlier)
2.1.2
Retrieval filters
The iterative retrieval scheme implemented in the L2 processor provides information about its own
reliability. This information is summarized in some retrieval flags stored in Level 2 UDP files. The
conditions used to discard a SSS value in the filtering process are:
• Iterative scheme returns an error
• Retrieved value is outside range (SSS must be positive and lower than 42 psu)
• High retrieval value of sigma (theoretical uncertainty computed for SSS larger than 5 psu)
File: BEC-SMOS-0001-PD.pdf
Version: 1.3
Date: 19/09/2014
Page: 6 of 20
Title: SMOS-BEC Ocean and Land Products Description.
• Normalised cost function at the last iteration is below the significance level (5%)
• Maximum number of iteration (20) reached before convergence using forward model
• Iterative loop ends because Marquardt increment reaches a given threshold (100)
• The total number of available measures is too low (16)
2.1.3
Geometrical filters
It is known that the external parts of the swath provide lower quality data [Zine et al., 2007]. Thus,
only measures taken up to 360 km from the satellite track are considered to generate Ocean Salinity
products.
2.2
Ocean salinity Level 3 products
According to the direction of the SMOS orbit passes, SMOS products can be classified in ascending,
descending and both products. These products are created in a variety of averaging periods: 3 days
and 9 days generated every 3 days, monthly, seasonal (quaterly) and annual. The spatial averaging is
computed by default in a regular lat-lon grid of 0.25o × 0.25o
2.2.1
Binned products
The binned maps are constructed by simple weighted averaging of the filtered L2 SSS values. The
weight average of Sea Surface Salinity in the cell k is given by the expression [Boutin et al., 2012]:
N
SSS
k
=
wi SSSi , where wi =
1
Ri2 σi2
N
i=1
j=1
,
(1)
1
Rj2 σj2
σi is the theoretical uncertainty computed for SSS at a grid point i, Ri is the equivalent footprint size
(diameter of the equivalent circle) centered on the grid point i and N is the number of grid points
contained in the cell k.
Each netCDF file contains:
• Sea Surface Salinity
• Number of L2 grid points averaged in each cell
• Variance of these SSS values
• SSS anomaly with WOA 2009
The WOA 2009 used to compute anomaly is linear interpolated to the center of the averaging period
of each product.
2.2.2
Optimal interpolation products
SMOS Level 2 SSS data are optimally interpolated (Objective Analysis) to produce maps of higher
consistency and fewer gaps as compared to the L3 binned products. To reduce the computational
File: BEC-SMOS-0001-PD.pdf
Version: 1.3
Date: 19/09/2014
Page: 7 of 20
Title: SMOS-BEC Ocean and Land Products Description.
(a) Binned product
(b) Optimal Interpolated product corresponding to the above binned product
Figure 1: Example of distributed Level 3 products
cost, 0.25o × 0.25o grid binned L2 data are used to feed the OI algorithm. The OI is performed using
monthly WOA 2009 data as background field.
L3 products, as well as L4 products, are validated with near-surface measurements provided by Argo
profilers, which allow us to define several quality metrics. We have found that the implementation of
objective analysis significantly improve data accuracy with respect to binned maps.
The resulting product contains:
• Sea Surface Salinity analysis
• SSS anomaly with WOA 2009
• Background field
File: BEC-SMOS-0001-PD.pdf
Version: 1.3
Date: 19/09/2014
Page: 8 of 20
Title: SMOS-BEC Ocean and Land Products Description.
The WOA 2009 used to compute anomaly is linear interpolated to the center of the averaging period
of each product.
2.3
2.3.1
Ocean salinity Level 4 products
Fused products using singularity analysis techniques
These products are obtained with a singularity analysis based fusion technique. A template variable
of good quality (Sea surface temperature, SST, in our case, see section 2.5.1) is used as template to
restore the multifractal structure of singularity fronts in a noisy variable (SSS in our case). Further
information on the multifractal structure of ocean scalars can be found in [Turiel et al., 2009].
Singularity analysis based fusion can be used not only to improve the signal level, but also to increase
the spatial and time resolution of fused maps, provided that the template (SST for us) has the target
space and time resolutions following the local relationship
SSS = a × SST + b
(2)
where a and b are known as the local slope and intercept coeficients respectively.
The resulting product contains:
• Sea Surface Salinity analysis
• SSS anomaly with WOA 2009
• Local slope coeficient (a from equation 2)
• Local intercept coeficient (b from equation 2)
• Local regression coeficiet
The WOA 2009 used to compute anomaly is linear interpolated to the center of the averaging period
of each product.
2.4
Ocean salinity reprocessing campaign
The ESA reprocessing campaign covers all the SMOS data available from January 2010 until December
2013 up to L2. The processors used are: L1 Operational Processor L1OP v5.04 and L2 Operational
Processor L2OSOP v5.50, which are the ones used in the SMOS DPGS operational chain, from
December 2011 to date. As a part operational chain, Level 1 data (brightness temperature at antenna
level) bias is corrected by applying an Ocean Target Transformation (OTT) [Tenerelli and Reul, 2010].
During the reprocessing campaign, the method to ingest the OTT into the processing chain was
different from the DPGS operation chain:
• In the DPGS operational chain the OTT is computed once per month, using 6 days of filtered
data from the equatorial Pacific. Then the OTT is used to process data few weeks after its
computation.
• In the reprocessing campaign the OTT is computed every 2 weeks in the same region, but the
computation period coincides with the period in which the OTT is applied. This is possible
since reprocessing is pbviously perfored in delayed mode.
File: BEC-SMOS-0001-PD.pdf
Version: 1.3
Date: 19/09/2014
Page: 9 of 20
Title: SMOS-BEC Ocean and Land Products Description.
Figure 2: Fused product from the binned one shown in figure 1(a)
This modification on the OTT validity time (shift backward) leads to more consistent L2 SMOS SSS
data than the DPGS data. In particular, temporal inconsistencies (biases) are much reduced, as shown
in figure 3.
So the reprocessed L2 data set is more stable and of higher quality than the DPGS operational
chain data. The L3/L4 reprocessing campaign (coded as 2013001) was performed at SMOS-BEC by
filtering Level 2 data as described in section 2.1. The maps are built for ascending, descending and
both (ascending+descending) orbits. The products described in sections 2.2 and 2.3 have been also
generated with L2 reprocessed data.
It is worth noting that SMOS commissioning phase ended on May 20, 2010. Therefore, it is recommended to avoid the use of SMOS data prior to June 2010. Note also that, due to technical problems,
SMOS have not acquired reliable measures from December 27, 2010 to January 10, 2011.
2.5
2.5.1
Ocean auxiliary data
Singularity exponents
For any given ocean scalar (SST, SSS, SSH, Chlorophyll Concentration and even Water Leaving
Radiances) singularity exponents can be calculated. Singularity exponents are dimensionless measures
of the degree of regularity or irregularity of a function at each of its domain points. They extend the
concept of Holder exponents, such that positive exponents imply that the function is continuous and
has a given number of derivatives, while negative exponents imply that the function is irregular and
therefore experiences transitions, jumps and eventually divergences to infinity.
For obtaining singularity exponents we follow the theory explained in [Turiel et al., 2008a] and
[Turiel et al., 2008b]. The modulus of the gradient of the scalar is evaluated at each point in the
domain. The resulting field is projected on a given wavelet at different resolution scales, such that the
dependence of the projection on the resolution scale can be assessed by means of a log-log regression,
the slope of which is the singularity exponent.
Singularity exponents derived from regular scalars such as SSS, SST or SSH are lower-bounded at -1,
File: BEC-SMOS-0001-PD.pdf
Version: 1.3
Date: 19/09/2014
Page: 10 of 20
Title: SMOS-BEC Ocean and Land Products Description.
(a) Mean (ascending case)
(c) Standard deviation (ascending case)
(b) Mean (descending case)
(d) Standard deviation (descending case)
Figure 3: Differences between SMOS and ARGO for the reprocessed L3 binned maps in the latitude
range 60o N-60o S. Each point stands for a 9-days map at 0.25 x 0.25 lat-lon resolution
since they are finite variation functions (see [Turiel and Parga, 2000]). They have no upper bound,
although values beyond 2 are rare.
It has been verified ([Turiel et al., 2005]; [Isern-Fontanet et al., 2007]; [Turiel et al., 2009]) that singularity exponents derived from SST maps track with remarkable precision the streamlines of the general
circulation of the ocean. In fact, there is some evidence ( [Isern-Fontanet et al., 2007]; [Nieves et al., 2007])
that different ocean scalars have the same singularity exponents – what should be expected if singularity exponents are the result of flow advection, regardless of the specific process of any particular
ocean scalar.
This correspondence of singularity exponents can be exploited to reduce the effects of noise and
artefacts on a given scalar map using the information conveyed by the singularity exponents derived
from a different, higher-quality map. We have implemented a numerical algorithm capable of using
the singularity exponents of one scalar field to improve the quality of a different scalar field.
The singularity exponent product distributed has been generated from the daily OSTIA SST product
(downloadable at MyOcean webpage http://www.myocean.eu). The algorithm described above together with the OSTIA SST product are used to derive our L4 product, which outperforms the SMOS
SSS L3 products. The resolution of the product has been degraded to one fourth of degree to match
that of SSS L3 maps used to generate the L4 product.
Singularity exponent maps are also useful for front identification, eddy tracking and assessing mesoscale
activity. For such reason we distribute them here along with the other products (for instance, to
produce the SMOS L4 SSS products descibed in 2.3.1 using OSTIA SST products as template).
File: BEC-SMOS-0001-PD.pdf
Version: 1.3
Date: 19/09/2014
Page: 11 of 20
Title: SMOS-BEC Ocean and Land Products Description.
Figure 4: Singularity exponents derived from OSTIA SST maps at 0.25o resolution
2.6
Ocean files structure
The resulting Level 3 and Level 4 products are distributed in netCDF format and the name of each
file follows the layout:
BEC_AAAAAA_B_CCCCCCCCCCCCCCC_DDDDDDDDDDDDDDD_EEEEEEE_FFF_GGG.nc
Where each field of the filename is as follows:
• AAAAAA: is the product’s name:
– BINNED: Binned product
– OI
: Optimal Interpolation product
– L4 SST: Fused product using singularity analysis techniques derived from SST
– EXPSST: Singularity Exponents
• B: Indicates the orbit composition of the product.
– A for products composed by ascending orbits
– D for products composed by descending orbits
– B for products composed by both types of orbits
• CCCCCCCCCCCCCCC: Starting UTC time (YYYYMMDDThhmmss) of the first L2 product used
to create the L3/L4 product. This is an inherited value in products not derived directly from
Level 2 orbits.
• DDDDDDDDDDDDDDD: Ending UTC time (YYYYMMDDThhmmss) of the last L2 product used to
create the L3/L4 product. This is an inherited value in products not derived directly from Level
2 orbits (Optimal Interpolation and L4 products).
File: BEC-SMOS-0001-PD.pdf
Version: 1.3
Date: 19/09/2014
Page: 12 of 20
Title: SMOS-BEC Ocean and Land Products Description.
Spatial resolution
Type
Generation Rate
Averaging Period
Product
3 days
binned
3 days
9 days
optimal interpolated
fused using singularity analysis
binned
monthly
0.25 degrees
1 natural month
Reprocessed / Near Real Time
optimal interpolated
fused using singularity analysis
binned
quaterly
seasonal
optimal interpolated
fused using singularity analysis
binned
annual
annual
optimal interpolated
fused using singularity analysis
Orbit passes
ascending
descending
both
ascending
descending
both
ascending
descending
both
ascending
descending
both
ascending
descending
both
ascending
descending
both
ascending
descending
both
ascending
descending
both
ascending
descending
both
ascending
descending
both
ascending
descending
both
ascending
descending
both
ascending
descending
both
Code
XXXBIN003D025A
XXXBIN003D025D
XXXBIN003D025B
XXXBIN009D025A
XXXBIN009D025D
XXXBIN009D025B
XXXOI 009D025A
XXXOI 009D025D
XXXOI 009D025B
XXXFUT009D025A
XXXFUT009D025D
XXXFUT009D025B
XXXBIN001M025A
XXXBIN001M025D
XXXBIN001M025B
XXXOI 001M025A
XXXOI 001M025D
XXXOI 001M025B
XXXFUT001M025A
XXXFUT001M025D
XXXFUT001M025B
XXXBIN003M025A
XXXBIN003M025D
XXXBIN003M025B
XXXOI 003M025A
XXXOI 003M025D
XXXOI 003M025B
XXXFUT003M025A
XXXFUT003M025D
XXXFUT003M025B
XXXBIN001Y025A
XXXBIN001Y025D
XXXBIN001Y025B
XXXOI 001Y025A
XXXOI 001Y025D
XXXOI 001Y025B
XXXFUT001Y025A
XXXFUT001Y025D
XXXFUT001Y025B
Table 1: Ocean products distributed by BEC. Three first letters of the code indicated as XXX are NRT
for near real time products and REP for reprocessed products. Code string is necessary to automatically
download a given product using getBEC tool
• EEEEEEE: Internal code that designates the filtering applied. This is an inherited value in products
not derived directly from Level 2 orbits.
• FFF: Grid size of the product in a lat-lon grid multiplied by 100
• GGG: Version number of the file starting at 001
2.7
Ocean products list
The list of ocean products distributed by CP34 is summarized in table 1
In order to automatically download a given type of product, a Linux-based tool named getBEC is
offered to users. Registered users can download this tool from http://cp34-bec.cmima.csic.es/
bec-tools/
File: BEC-SMOS-0001-PD.pdf
Version: 1.3
Date: 19/09/2014
Page: 13 of 20
Title: SMOS-BEC Ocean and Land Products Description.
3
LAND PRODUCTS
Two different kinds of land products are generated and distributed by the BEC:
• Soil Moisture Level 3 products: The SMOS data used to compute Level 3 products is the Level
2 Soil Moisture User Data Product (UDP). The UDP files are operationally generated by ESA
and they are received at the BEC in near real time. They include geophysucal parameters, a
theoretical estimate of their accuracy, flags and descriptors for the product.
• Soil Moisture Level 4 products: Level 4 products combine the SMOS Level 1 C brightness temperature measurements from ESA, with Terra/Aqua MODIS Land Surface Temperatur (LST)
and NDVI (Normalized Difference Vegetation Index) data.
3.1
Soil moisture Level 3 products
In order to generate Soil Moisture Level 3 products, Level 2 soil moisture UDP are first filtered. Then
they are combined into maps where the spatial resolution is the same than the daily Level 2 SMOS
product (ISEA product) or into spatial averaged regular lat-lon grid of 25 km (binned products).
Figure 5: SMOS soil moisture L3 binned map corresponding to the annual mean of 2012 for ascending
(left) and descending (right) orbits.
3.1.1
Soil moisture data filtering
The quality flags and descriptors from UDP files allow discarding unreliable Soil Moisture values. The
following filters have been applied to create Level 3 products:
• Grid points with soil moisture negative values are discarded
• Grid points with Data Quality Index (DQX) values greater than 0.07 are discarded.
• Grid points with the no product flag raised are discarded. It indicates that the retrieval has
failed either due to retrieved geophysical data is not of an acceptable quality or other factors.
• Grid points with the probability of RFI flag set to high are discarded.
• Grid points with the out-range flag raised are discarded. It indicates that the retrieved geophysical data are outside the extended range.
File: BEC-SMOS-0001-PD.pdf
Version: 1.3
Date: 19/09/2014
Page: 14 of 20
Title: SMOS-BEC Ocean and Land Products Description.
3.1.2
ISEA land product
Daily maps of soil moisture, optical thickness and dialectric constant (real and imaginary part) are
constructed from level 2 UDP products with neither spatial nor temporal averaging. Ascending and
descending orbits are processed separately. The resulting product contains:
• Latitude
• Longitude
• Grid point ID (ISEA grid point identifier)
• Soil Moisture value (m3 /m3 )
• Data Qualiyty Index value for the soil moisture estimate. It is a measure of the standard
deviation error in the estimate (m3 /m3 )
• Optical thickness at the nadir direction (N p)
• Data Qualiyty Index value for the optical thickness estimate (N p)
• Real part of retrieved dielectric constant
• Data Quality Index value for the real part of retrieved dielectric constant
• Imaginary part of retrieved dielectric constant
• Data Qualiyty Index value for the imaginary part of retrieved dielectric constant
3.1.3
Binned land products
Daily soil moisture maps in EASE-ML 25km grid are constructed by DQX-weighted averaging. The
averaging of soil moisture in the cell k is computed following the expression:
1
DQXi2
N
SM
k
=
wi SMi , where wi =
N
i=1
j=1
.
(3)
1
DQXj2
The averaging of the associatd DQX ( DQX k ) is computed as:
1
DQX
N
2
k
=
i=1
1
.
DQXi2
(4)
The averaged spatial variance of the soil moisture estimates (V ark ) is computed as:
N
V ark =
i=1
N
i=1
1
DQXi2
2
1
DQXi2
N
N
−
i=1
1
DQXi4
i=1
SMi2
− SM
DQXi2
N
2
k
i=1
1
DQXi2
(5)
Ascending and descending orbits are processed separately. These products are created in a variety
of generation rates and averaging periods: 1 and 3 days -generated daily-, 9 days -generated every 3
days-, monthly, seasonal (quaterly) and annual (see Table 2).
The fields given per grid cell are:
File: BEC-SMOS-0001-PD.pdf
Version: 1.3
Date: 19/09/2014
Page: 15 of 20
Title: SMOS-BEC Ocean and Land Products Description.
Figure 6: SMOS soil moisture L3 3-days binned maps. The plots show the soil moisture evolution
during the Colorado in September 2013. Heavy rain was received from 11 to 16 of September.
Figure 7: SMOS soil moisture L3 monthly binned maps. The plots show the mean values of September
for the four years of the mission in the same region where inundation happened.
File: BEC-SMOS-0001-PD.pdf
Version: 1.3
Date: 19/09/2014
Page: 16 of 20
Title: SMOS-BEC Ocean and Land Products Description.
• Soil Moisture ( SM
• DQX ( DQX
k
k
of equation 3)
of equation 4)
• Variance of SM averaged in each cell (V ark of equation 5)
• Number of L2 soil moisture estimates used in the computation (N of equation 3)
3.2
Soil moisture Level 4 products
Soil moisture is a key state variable that links the Earth’s water, energy and carbon cycles, and its
variations affect the evolution of weather and climate over continental regions. The ESA SMOS is the
first satellite mission ever designed to measuring this variable, and its accurate observations of soil
moisture are helping to improve our understanding of water and energy fluxes interactions between the
atmosphere, the soil surface and subsurface at a global scale. However, its spatial resolution (on the
order of 40 km) prevents SMOS data from being applied in small scale applications, such as on-farm
water management, flood prediction or meso-scale weather forecasting.
Figure 8: Disaggregated SMOS soil moisture map at 1 km spatial resolution over the Iberian Peninsula,
from July 7, 2012 (6 A.M.) using the proposed algorithm. Empty areas in the image correspond to
clouds masking MODIS observations or quality-filtered SMOS TB.
One key research line at SMOS-BEC is the development of data fusion algorithms to provide downscaled SMOS-based soil moisture information resolving the dynamics within 100 m to 1 km catchments.
Accurate knowledge of the soil moisture status at these scales is essential to understand how to manage
and utilise soil water -one of the Earth’s scarcest and most valuable natural resource- to its maximum
potential.
An innovative downscaling approach for SMOS has been developed, which combines MODIS Visible/Infrared data with SMOS brightness temperatures into high-resolution soil moisture maps. To
File: BEC-SMOS-0001-PD.pdf
Version: 1.3
Date: 19/09/2014
Page: 17 of 20
Title: SMOS-BEC Ocean and Land Products Description.
date, validation results from comparison with in situ data over a selected suite of representative sites
support the use of this technique; high resolution soil moisture maps are shown to nicely reproduce
soil moisture dynamics at 1 km without a significant degradation of the root-mean-squared error with
respect to the SMOS L2 product [Piles et al., 2011], [Sanchez-Ruiz et al., 2014], [Piles et al., 2014].
This algorithm has been implemented at SMOS-BEC facilities and high resolution soil moisture maps
over the Iberian Peninsula are being distributed: maps from the first three years of SMOS in orbit are
available (delayed mode) and two near real-time maps are daily generated corresponding to ascending
and descending overpasses with a delay of less than 12 hours. These maps are already being used as
supporting information for forest brigades within the Catalonia region.
3.2.1
High resolution soil moisture: delayed
A data set of soil moisture maps covering the Iberian Peninsula at 1km spatial resolution since January
2011 up to the most recent processing date is provided. It contains two maps per day, corresponding to
SMOS ascending (6 A.M.) and descending (6 P.M.) passes. Maps are obtained using the downscaling
algorithm in [Piles et al., 2014], which combines the brightness temperature measurements from ESA
SMOS, with Land Surface Temperature and NDVI (Normalized Difference Vegetation Index) data
from Aqua MODIS day passes. The latest released SMOS data is available at SMOS-BEC facilities;
MODIS version 5 MYD11A1 products are freely distributed by the U.S. Land Processed Distributed
Active Archive Center (http://www.lpdaac.usgs.gov).
3.2.2
High resolution soil moisture: near real-time
Soil moisture maps covering the Iberian Peninsula at 1km of spatial resolution are provided in near real
time (delay <12 h). Two maps per day are generated, corresponding to SMOS ascending (6 A.M.) and
descending (6 P.M.) passes. Maps are obtained using the downscaling algorithm in [Piles et al., 2011],
which combines the brightness temperature measurements from ESA SMOS, with Land Surface Temperature and NDVI (Normalized Difference Vegetation Index) data from Terra/Aqua MODIS day
passes. The use of MODIS Terra LST is prefered. Nevertheless, downscaled maps using LST yield
broadly consistent results in [Piles et al., 2014]. Hence, Aqua is used when Terra LST is not available
(i.e. masked by clouds). SMOS latest released data in near-time time is available at SMOS-BEC
facilities; MODIS data in near real-time is kindly provided by LATUV (http://www.latuv.uva.es),
Valladolid University.
3.3
Land files structure
SMOS BEC Land products are distributed in netCDF format with the following naming convention:
BEC_AAAAAA_B_CCCCCCCCCCCCCCC_DDDDDDDDDDDDDDD_EEEEEEE_FFF_GGG.nc,
where each field of the filename is as follows:
• AAAAAA: is the product’s name:
– BIN SM: L3 Soil Moisture products
– HDE SM: L4 high resolution delayed soil moisture products
– HNR SM: L4 high resolution near real time soil moisture products
File: BEC-SMOS-0001-PD.pdf
Version: 1.3
Date: 19/09/2014
Page: 18 of 20
Title: SMOS-BEC Ocean and Land Products Description.
• B: indicates the orbit composition of the product.
– A for ascending orbits
– D for descending orbits
• CCCCCCCCCCCCCCC: starting UTC time (YYYYMMDD hhmmss) of the half-orbit used to create
the product.
• DDDDDDDDDDDDDDD: ending UTC time (YYYYMMDD hhmmss) of the half-orbit used to create
the product.
• EEEEEEE: internal code
– NOMINAL: for L3 product indicates that the nominal filter (described in section 3.1.1) has
been applied to L2 product
– AQUA1 : for L4 product indicates that LST data at 1km spatial resolution from AQUA has
been used
– TERR1 : for L4 product indicates that LST data at 1km spatial resolution from TERRA
has been used
• FFF: grid indicator
– 025: Indicates that EASE-ML grid of 25 km is considered
– 4H9: ISEA grid resolution
– IBE: Indicates that the product is provided for the Iberian Peninsula
• GGG: version number of the file starting at 001
3.4
Land products list
The list of land products is summarized in table 2.
In order to automatically download a given type of product, a Linux-based tool named getBEC is
offered to users. Registered users can download this tool from http://cp34-bec.cmima.csic.es/
bec-tools/
File: BEC-SMOS-0001-PD.pdf
Version: 1.3
Date: 19/09/2014
Page: 19 of 20
Title: SMOS-BEC Ocean and Land Products Description.
Spatial resolution
Type
Product
Generation Rate
Averaging Period
1 days
1 days
3 days
0.25 degrees
ISEA
Reprocessed / Near Real Time
binned
3 days
9 days
monthly
1 natural month
annual
annual
Reprocessed / Near Real Time
single value
1 days
1 days
Near Real Time
High Resolution
1days
1days
Delayed
High Resolution
1days
1days
1km
Orbit passes
ascending
descending
ascending
descending
ascending
descending
ascending
descending
ascending
descending
ascending
descending
ascending
descending
ascending
descending
Code
XXXSMB001D025A
XXXSMB001D025D
XXXSMB003D025A
XXXSMB003D025D
XXXSMB009D025A
XXXSMB009D025D
XXXSMB001M025A
XXXSMB001M025D
XXXSMB001Y025A
XXXSMB001Y025D
XXXSMB001D4H9A
XXXSMB009D4H9D
XXXSMH001DIBEA
XXXSMH001DIBED
XXXSMH001DIBEA
XXXSMH001DIBED
Table 2: Land products distributed by BEC. Three first letters of the code indicated as XXX are NRT
for near real time products, DEL for delayed products and REP for reprocessed products. Code string
is necessary to automatically download a given product using getBEC tool
References
[DPG, 2012] (2012). SMOS Level 2 and Auxiliary Data Products Specifications SO-TN-IDR-GS-0006.
INDRA. version 6.1.
[Boutin et al., 2012] Boutin, J., Martin, N., Yin, Y., Font, J., Reul, N., and Spurgeon, P. (2012).
First assessment of SMOS data over open ocean: Part II-sea surface salinity. IEEE Trans. Geosci.
Remote Sens., vol. 50, no. 5. pp. 1662-1675.
[Font et al., 2010] Font, J., Camps, A., Borges, A., Martin-Neira, M., Boutin, J., Reul, N., Kerr, Y.,
Hahne, A., and Mechlenburg, S. (2010). Smos: the challenging sea surface salinity measurement
from space. Proceedings of the IEEE, 98:649.
[Guimbard et al., 2012] Guimbard, S., Gourrion, J., Portabella, P., Turiel, A., Gabarr´o, C., and Font,
J. (2012). SMOS Semi-Empirical Ocean Forward Model Adjustement. IEEE Trans. Geosci. Remote
Sens., vol. 50, no. 5. pp. 1676-1687.
[Isern-Fontanet et al., 2007] Isern-Fontanet, J., Turiel, A., Garc´ıa-Ladona, E., and Font, J. (2007).
Microcanonical multifractal formalism: Application to the estimation of ocean surface velocities.
Journal of Geophysical Research: Oceans, 112(C5):2156–2202.
[Kerr et al., 2010] Kerr, Y., Waldteufel, P., Wigneron, J.-P., Delwart, S., Cabot, F., Boutin, J., Escorihuela, M.-J., Font, J., Reul, N., Gruhier, C., Juglea, S., Drinkwater, M., Hahne, A., Martin-Neira,
M., and Mecklenburg, S. (2010). The smos mission: new tool for monitoring key elements of the
global water cycle. Proceedings of the IEEE, 98(5):666–687.
[McMullan et al., 2008] McMullan, K. D., Brown, M., Martin-Neira, M., Rits, W., Ekholm, S., Marti,
J., and Lemanczyk, J. (2008). Smos: The payload. Geoscience and Remote Sensing, IEEE Transactions on, 46(3):594–605.
[Nieves et al., 2007] Nieves, V., Llebot, C., Turiel, A., Sol´e, J., Garc´ıa-Ladona, E., Estrada, M., and
Blasco, D. (2007). Common turbulent signature in sea surface temperature and chlorophyll maps.
Geophysical Research Letters, 34(23):1944–8007.
File: BEC-SMOS-0001-PD.pdf
Version: 1.3
Date: 19/09/2014
Page: 20 of 20
Title: SMOS-BEC Ocean and Land Products Description.
[Piles et al., 2011] Piles, M., Camps, A., Vall-llossera, M., Corbella, I., Panciera, R., Rudiger, C., Kerr,
Y., and Walker, J. (2011). Downscaling smos-derived soil moisture using modis visible/infrared data.
Geoscience and Remote Sensing, IEEE Transactions on, 49(9):3156 –3166.
[Piles et al., 2014] Piles, M., Sanchez, N., Vall-llossera, M., Camps, A., Martinez Fernandez, J., Martinez, J., and Gonzalez-Gambau, V. (2014). A downscaling approach for smos land observations:
evaluation of high resolution soil moisture maps over the iberian peninsula. IEEE Journal of Selected
Topics in Applied Earth Observations and Remote Sensing. in press.
[Sanchez-Ruiz et al., 2014] Sanchez-Ruiz, S., Piles, M., Sanchez, N., Martinez-Fernandez, J., Vallllossera, M., and Camps, A. (2014). Combining smos with visible and near/shortwave/thermal
infrared satellite data for high resolution soil moisture estimates. Journal of hydrology. in press.
[Tenerelli and Reul, 2010] Tenerelli, J. and Reul, N. (2010). Analysis of L1PP Calibration Approach
Impacts in SMOS Tbs and 3-Days SSS Retrievals over the Pacific Using an Alternative Ocean
Target Transformation Applied to L1OP Data. Technical report, IFREMER/CLS.
[Turiel et al., 2005] Turiel, A., Isern-Fontanet, J., Garcia-Ladona, E., and Font, J. (2005). Multifractal
method for the instantaneous evaluation of the stream function in geophysical flows. Phys. Rev.
Lett., 95:104502.
[Turiel et al., 2009] Turiel, A., Nieves, V., Garc´ıa-Ladona, E., Font, J., Rio, M.-H., and Larnicol, G.
(2009). The multifractal structure of satellite sea surface temperature maps can be used to obtain
global maps of streamlines. Ocean Science, 5(4):447–460.
[Turiel and Parga, 2000] Turiel, A. and Parga, N. (2000). The multifractal structure of contrast
changes in natural images: From sharp edges to textures. Neural Computation, 12(4):763–793.
[Turiel et al., 2008a] Turiel, A., Sol´e, J., Nieves, V., Ballabrera-Poy, B., and Garc´ıa-Ladona, E.
(2008a). Tracking oceanic currents by singularity analysis of microwave sea surface temperature
images. Remote Sensing of Environment, 112(5):2246 – 2260. Earth Observations for Terrestrial
Biodiversity and Ecosystems Special Issue.
[Turiel et al., 2008b] Turiel, A., Yahia, H., and P´erez-Vicente, C. (2008b). Microcanonical multifractal
formalism geometrical approach to multifractal systems: Part i. singularity analysis. J. Phys. A:
Math. Theor., 41(1).
[Zine et al., 2007] Zine, S., Boutin, J., Waldteufel, P., Vergely, J., Pellarin, T., and Lazure, P. (2007).
Issues About Retrieving Sea Surface Salinity in Coastal Areas From SMOS Data. IEEE Trans.
Geosci. Remote Sens., vol. 45, no. 7. pp. 2061-2072.

Download Report