While satellite-derived Downwelling Surface Radiation (DSR) products offer coarse spatial resolution (100 -101 km2), regional and local applications (e.g. exploitation of solar energy, efficient building) require much finer spatial resolutions (⪅ 103 m2). Although several DSR downscaling studies were conducted during the last decades considering topographic effects, only a few of them achieved the required spatial resolution for local applications and none of them evaluated the contribution of each effect. The current study proposes the downscaling of the MSG/SEVIRI DSR product (LSA-207), which recently included the fraction of diffuse radiation, from the original approximately 3 km × 3 km spatial resolution to approximately 30 m × 30 m based on topographic corrections. In addition, the effect of each correction (i.e. elevation, shadowing, and sky obstruction) in the final result is evaluated. Results are validated against ground measurements taken in 142 stations from the Servei Meteorologic de Catalunya –Catalan meteorology center– in North-East Spain, under different conditions (elevation, sky state, solar altitude) at 30 minutes temporal resolution. Shadowing caused the most noticeable changes in downscaled DSR across all elevations and sky states: about one percentage point (pp) reduction in Mean Bias Error (MBE). Correlation and Root Mean Squared Error (RMSE) remained similar pre- and post-downscaling. Greater differences were observed between clear and cloudy skies (respectively, correlations of 0.98 and 0.89-0.90, 3-4% and 6-7% relative MBE, 13-14% and 40% relative RMSE), and for solar altitude smaller than 20 degrees (30-40 pps in relative MBE).
The Satellite Application Facility on Land Surface Analysis (LSA SAF) produces and provides access to remotely sensed variables for the characterization of terrestrial ecosystems, such as land surface fluxes and biophysical parameters, taking full advantage of the EUMETSAT satellites and sensors. In this work, a procedure for the joint estimation of LSA SAF vegetation parameters is proposed. The approach includes the use of multi-task learning with gaussian processes (MTGP). The MTGP learns a shared covariance function on input features and a covariance matrix over tasks. Unlike the single output approaches, the proposed multi-task captures the inter-task dependencies among outputs. Two comparison exercises were undertaken to assess the estimation power of the MTGP as compared to single output algorithms such as standard gaussian processes regression (GPR), neural networks (NN), and random forest (RF). First, we evaluate the performance of MTGP in the context of deriving CO2 fluxes such as the gross primary production (GPP), net ecosystem exchange (NEE), and total ecosystem respiration (TER) blending SEVIRI/MSG and eddy covariance (EC) data. In addition, the MTGP prediction power was also assessed for the joint estimation of LAI, FAPAR, and FVC in a hybrid approach using radiative transfer model simulations and AVHRR/MetOp observations. The results show that MTGP outperforms the single output approaches in terms of accuracy. The MTGP multi-task optimization links outputs in such a way that the relationships among the biophysical parameters are better described obtaining a more robust model and therefore improving the accuracy of the estimates. The findings pave the way for future multi-task implementations in order to derive consistent outputs and accurate estimates of vegetation properties from remote sensing.
The LSA-SAF produces and disseminates variables for the characterization of terrestrial ecosystems and their role in the energy balance of Earth, such as land surface fluxes and vegetation parameters, taking full advantage of remotely sensed data from EUMETSAT satellites and sensors. All LSA SAF products are distributed according with EUMETSAT data policy and have been classified as essential and are distributed free of charge. This work provides an overview of the SEVIRI/MSG and AVHRR/MetOp LSA-SAF vegetation products. The LSA-SAF vegetation products provide consistent long-term data records with well-characterized uncertainty, which are required by the scientific community to model terrestrial ecosystems and energy cycles at regional and global scales. Three vegetation products (FVC, LAI, FAPAR) are provided from SEVIRI/MSG and AVHRR/MetOp observations. The vegetation products are routinely validated and provide pixel-wise uncertainty estimates and quality flag information to identify unreliable observations. The entire archive with the latest version of the several retrieval algorithms has been reprocessed in recent years in order to generate a homogeneous Climate Data Records (CDRs) of these vegetation variables. LSA-SAF has also developed recently two new products, SEVIRI/MSG GPP and EPS/AVHRR CWC. The future generation of new LSA-SAF products derived from the future MTG and EPS-SG satellites, with higher spatial and spectral resolution, will guarantee the continuity of the service.
The 2020+ Common Agricultural Policy encourages the use of Copernicus remote sensing data for the monitoring of agricultural parcels. In this work, a procedure for automatic identification of land use from remote sensing data is proposed. The approach includes the use of spectral information of Sentinel-2 time series over the Valencia province (Spain) during the agronomic year 2027/2018, and deep learning recurrent networks. In particular, a bi-directional Long Short Term Memory (Bi-LSTM) network was trained to classify active land uses and abandoned lands. A comparison exercise was undertaken to assess the classification power of the Bi-LSTM as compared to the random forest (RF) algorithm. The Bi-LSTM network outperformed the RF, and provided and overall accuracy of 97.5% when discriminating eleven land uses including abandoned lands. The results suggest the proposed methodology could potentially be implemented in an automated procedure to supervise the CAP requirements to access subsidies. In addition, the classification process also supports the continuous update of the Land Parcel Identification System (LPIS), which allows paying agencies to uniquely identify land parcels in space, store records of land uses (and assess its evolution), and ultimately ease the declaration procedure to both farmers and paying agencies.
Biogeochemical ecosystem models describe the energy and mass exchange processes between natural systems and their environment. They normally require a large amount of inputs that present important spatial variations and require a parameterization. Other simpler ecosystem models focused on a single process only need a reduced amount of inputs usually derived from direct measurements and can be combined with the former models to calibrate their parameters. This study combines the biogeochemical model Biome-BGC and a production efficiency model (PEM) optimized for the study area to calibrate a key parameter for the simulation of the ecosystem water balance by Biome-BGC, the rooting depth. Daily gross primary production (GPP) time series for the 2005-2012 period are simulated by both models. First, the optimized PEM is validated against GPP derived from four eddy covariance (EC) towers located at different ecosystems representative of the study area. Next, GPP time series simulated by both models are combined to optimize rooting depth at the four sites: different values of rooting depth are tested and the one that results in the lowest root mean square error (RMSE) between the two GPP series is selected. Explained variance and relative RMSE between Biome- BGC and EC GPP series are respectively augmented between 3 and 14 percentage points (pp) and reduced between 1 and 33pp. Finally the methodology is extrapolated for the whole study area and an original rooting depth map for peninsular Spain, which is coherent with the spatial distribution of vegetation type and GPP in the study area, is obtained at 1-km spatial resolution.
National forest inventories provide measurements of forest variables (e.g. growing stock) that can be used for the estimation of above ground biomass (AGB). Mapping growing stock brings knowledge about spatial distribution and temporal dynamics of ABG, which is necessary for carbon cycle analysis. Several studies have been conducted on the integration of ground and optical remote sensing data to map forest biomass over Europe. Nevertheless, more direct information on forest biomass could be obtained by LiDAR techniques, which directly assess vertical forest structure by measuring the distance between the sensor and the scattering elements located inside the canopy volume. Thus, global 1-km maps of forest canopy height have been recently obtained from the Geoscience Laser Altimeter System (GLAS). The current study aims to produce a forest growing stock map in Spain. Five different forest type areas were identified in three provinces along a North – South gradient accounting for different ecosystems and climatic conditions. Growing stock ground data from the Third Spanish National Forest Inventory were assigned to each forest type and aggregated to 1-km spatial resolution. GLAS-derived canopy height was extracted for the locations of selected ground data. A relationship between inventory growing stock and satellite canopy height was found for each class. The obtained relationships were then extended all over Spain. The accuracy of the resulting growing stock map was assessed at province level against the Third Spanish National Forest Inventory growing stock estimations (R = 0.85, RMSE = 21 m3 ha-1).
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.