Françoise Gellens-Meulenberghs

The Satellite Application Facility for Land Surface Analysis

Information on land surface properties finds applications in a range of areas related to weather forecasting, environmental research, hazard management and climate monitoring. Remotely sensed observations yield the only means of supplying land surface information with adequate time sampling and a wide spatial coverage. The aim of the Satellite Application Facility for Land Surface Analysis (Land-SAF) is to take full advantage of remotely sensed data to support land, land–atmosphere and biosphere applications, with emphasis on the development and implementation of algorithms that allow operational use of data from European Organization for the Exploitation of Meteorological Satellites (EUMETSAT) sensors. This article provides an overview of the Land-SAF, with brief descriptions of algorithms and validation results. The set of parameters currently estimated and disseminated by the Land-SAF consists of three main groups: (i) the surface radiation budget, including albedo, land surface temperature, and downward short- and longwave fluxes; (ii) the surface water budget (snow cover and evapotranspiration); and (iii) vegetation and wild-fire parameters.

Estimating crop-specific evapotranspiration using remote-sensing imagery at various spatial resolutions for improving crop growth modelling

By governing water transfer between vegetation and atmosphere, evapotranspiration (ET) can have a strong influence on crop yields. An estimation of ET from remote sensing is proposed by the EUMETSAT ‘Satellite Application Facility’ (SAF) on Land Surface Analysis (LSA). This ET product is obtained operationally every 30 min using a simplified SVAT scheme that uses, as input, a combination of remotely sensed data and atmospheric model outputs. The standard operational mode uses other LSA-SAF products coming from SEVIRI imagery (the albedo, the downwelling surface shortwave flux, and the downwelling surface longwave flux), meteorological data, and the ECOCLIMAP database to identify and characterize the land cover. With the overall objective of adapting this ET product to crop growth monitoring necessities, this study focused first on improving the ET product by integrating crop-specific information from high and medium spatial resolution remote-sensing data. A Landsat (30 m)-based crop type classification is used to identify areas where the target crop, winter wheat, is located and where crop-specific Moderate Resolution Imaging Spectroradiometer (MODIS) (250 m) time series of green area index (GAI) can be extracted. The SVAT model was run for 1 year (2007) over a study area covering Belgium and part of France using this supplementary information. Results were compared to those obtained using the standard operational mode. ET results were also compared with ground truth data measured in an eddy covariance station. Furthermore, transpiration and potential transpiration maps were retrieved and compared with those produced using the Crop Growth Monitoring System (CGMS), which is run operationally by the European Commissions Joint Research Centre to produce in-season forecast of major European crops. The potential of using ET obtained from remote sensing to improve crop growth modelling in such a framework is studied and discussed. Finally, the use of the ET product is also explored by integrating it in a simpler modelling approach based on light-use efficiency. The Carnegie–Ames–Stanford Approach (CASA) agroecosystem model was therefore applied to obtain net primary production, dry matter productivity, and crop yield using only LSA-SAF products. The values of yield were compared with those obtained using CGMS, and the dry matter productivity values with those produced at the Flemish Institute for Technological Research (VITO). Results showed the potential of using this simplified remote-sensing method for crop monitoring.

Evapotranspiration monitoring with Meteosat Second Generation satellites: improvement opportunities from moderate spatial resolution satellites for vegetation

The Satellite Application Facility on Land Surface Analysis proposes a land evapotranspiration (ET) product, generated in near-real time. It is produced by an energy balance model forced by radiation components derived from data of the Spinning Enhanced Visible and Infrared Imager aboard Meteosat Second Generation geostationary satellites, at a spatial resolution of approximately 3 km at the equator and covering Europe, Africa, and South America. In this article, we assess the improvement opportunities from moderate spatial resolution satellites for ET monitoring at the Meteosat Second Generation satellite scale. Four variables, namely the land cover, the leaf area index (LAI), the surface albedo, and the open water fraction, derived from moderate-resolution satellites for vegetation monitoring are considered at two spatial resolutions, 1 km and 330 m, corresponding to the imagery provided by Satellite Pour l’Observation de la Terre (SPOT)-VEGETATION and future Project for On-Board Autonomy – Vegetation (PROBA-V) space-borne sensors. The variables are incorporated into the ET model, replacing or complementing input derived from the sensor aboard the geostationary satellite, and their relative effect on the model output is analysed. The investigated processes at small scales unresolved by the geostationary satellite are better taken into account in the final ET estimates, especially over heterogeneous and transition zones. Variables derived from sensors at 250–300 m are shown to have a noticeable effect on the ET estimates compared to the 1 km resolution, demonstrating the interest of PROBA-V 330 m-derived variables for the monitoring of ET at Meteosat Second Generation resolution.

Landscape heterogeneity around flux measurement stations investigated through Sentinel-2 and PROBA-V satellite imagery

This study proposed the exploitation of Sentinel-2 and Proba-V imagery to assess the heterogeneity in the surroundings of EC stations through the multitemporal analysis of NDVI. The observations by these platforms allow computation of NDVI at 10 m (Sentinel-2) and 100 m (Proba-V) spatial resolutions. Such levels of spatial detail allowed the comparison of pixel values within two relevant geographic extents: the footprint of EC stations and the extent of satellite imagery cells commonly used to force flux modelling. The satellite platforms considered for this study exhibit different but complementary strengths. Proba-V allowed fast processing over a large number of stations in order to screen or rank EC stations as function of the spatial heterogeneity. The fine spatial resolution of Sentinel-2 allowed more in-depth spatial analysis. Three methods were implemented to explore the spatial heterogeneity with Sentinel-2 NDVI: the analysis of simple NDVI histogram, spatial heterogeneity index (SHI) and fitting a semi-variogram. The analysis with Proba-V resulted in a ranking of flux tower stations on the basis of NDVI Interquartile Range. This study was conducted in the framework of a research programme aiming at improving estimates of net primary productivity across different biomes. The methods appeared to complement each other: histograms were simple to interpret and offered a clear view on the spread and shape of data (often bi-model in cropped sites); SHI plots enable the visualization of heterogeneity changes throughout a period of time and semi-variograms allow the analysis at different distances and orientations.

Daily evapotranspiration at sub-kilometre spatial resolution by combining observations from geostationary and polar-orbit satellites

ABSTRACT Evapotranspiration (ET) estimates based on remote sensing are constrained by the spatial and temporal resolution of spaceborne observations. Therefore, the choice to be made is whether modelling at high temporal resolution with geostationary satellites at the cost of the spatial detail, or spatially detailed instantaneous modelling at overpass-time with polar-orbit satellites. This study aimed at exploiting the strength of both modelling approaches and derive daily ET estimates with cell size 1.0 km. The study was conducted in four sites with different land cover arrangements and catalogued as wetlands. The methodological approach was based on the algorithm behind the operational ET product of the Satellite Application Facility on Land Surface Analysis (LSA-SAF) initiative. The examined period comprised the end of the Spot-Végétation (Spot-V) mission and the first year of Proba-V operational modus. Two aspects were investigated: 1. The adequacy of combining observations from geostationary and Spot-V and Proba-V satellite missions for obtaining the target product; and, 2. The intercompatibility of ET estimates based on Spot-V and Proba-V so that continuous series over the operational period of both missions could be built. The comparison of simulated ET with in situ measured fluxes (eddy covariance) showed that the introduction of more spatial detail through polar-satellite observations resulted in the reduction of bias and root-mean-square error in ET estimates, as compared with estimations where Earth observation data were taken from geostationary satellites only. Moreover, the signals derived from Spot-V and Proba-V exhibited similarity during the period of operational overlap; especially in non-cultivated surfaces. This suggests that continuity objective of these two missions can be exploited in the estimation of ET time series.

Continuous monitoring of evapotranspiration (ET) overview of LSA-SAF evapotranspiration products

Evapotranspiration (ET) is the flux of water between the surface (vegetation, soil and water bodies) and the atmosphere. Monitoring this water loss may be of crucial importance for applications in hydrology, agriculture, water use efficiency studies and drought monitoring. We introduce one of the few satellite-based operational evapotranspiration products, generated continuously and in near real-time over Europe, Africa and part of South America. The ET products (30 minutes and daily) are generated at the EUMETSAT’s Satellite Application Facility on Land Surface Analysis (LSA-SAF) operations centre (http://landsaf.ipma.pt). Following our commitments to our user’s community, we are continuously looking for new ways to improve the product. To accomplish this, the feedback from users and potential users of the products is of great interest. In this contribution we present the ET products characteristics and recent improvements gained thanks to the inclusion in our ET algorithm of new variables derived from Earth observation by MSG SEVIRI. We show examples of the ET products and we highlight their potential in droughts detection and monitoring. Some examples of possible applications are presented to invite users and researchers to explore the possibilities offered by LSA-SAF evapotranspiration products.

Evaluating an energy balance setting and random forest-based downscaling for the estimation of daily ET at sub-kilometer spatial resolution

A large number of processes are impacted by the magnitude and timing of water fluxes. A proper monitoring of these processes would benefit from periodic estimation of evapo-transpiration (ET) with enough spatial detail to discriminate the different components of the landscape. However, high temporal and high/moderate spatial resolution are rarely present in the same ET product. In this study, we aimed at merging the strengths of geostationary and polar-orbit satellites to estimate daily ET at the native spatial resolution of Proba-V, 0.3 km. Two modelling paths were examined: the conduction of an energy balance at the target spatial resolution, and; energy balance at a coarser resolution followed by a statistical downscaling procedure based on building a Random Forest. The study area was the upper catchment of the Biebrza river in North Eastern Poland. The comparison of the modelled values with ET derived from in situ measurements (eddy covariance) indicate that both approaches performed good (Nash-Sutcliffe efficiency coefficient of 0.7589 and 0.7857 for first and second approach, respectively). The modelled spatial and temporal patterns were similar but the statistical downscaling path tended to generate lower ET estimates (bias equal to 0.0071 and −0.2254 for first and second approach, respectively).

SENSITIVITY ANALYSIS OF AQUACROP EVAPOTRANSPIRATION TO WEATHER STATION DISTANCE

Water balance calculation is essential for reliable agricultural management, and the actual evapotranspiration (ET) is the most complicated balance term to estimate. In agriculture, the most common method used is based on Penman-Monteith reference evaporation is determined from weather conditions for an unstressed grass cover, further multiplied by crop specific and soil water availability coefficients to obtain the actual evapotranspiration. This approach is also used in the AquaCrop model. This model has proven to be accurate when all weather data are locally available. However, in many cases, weather data can’t be collected on the site due to the limited number of stations and the vast region covered by each of them. Instead, data are often collected at many kilometers from the study site. The question we want to study is: how does evapotranspiration accuracy evolves with respect to weather station distance? A winter wheat plot in Lonzee (Belgium) was studied during the 2014-2015 agricultural seasons. Actual evapotranspiration was simulated with AquaCrop thanks to the weather data collected at 3 different distances from the study site: on the site (data collected by a fluxnet station), 20 km, 50 km and 70km from the site. The non-on-site weather data were derived from spatially interpolated 10 km grid data. These results were then compared to the fluxnet station evapotranspiration measurements to assess the impact of the weather station distance. Substantial differences, which were found between the four cases, evoking the importance of assimilating satellite derived ET products (e.g. MSG) into AquaCrop.

Retrieving daily evapotranspiration from the combination of geostationary and polar-orbit satellite data

Near-real time estimates of evapotranspiration (ET) are generated every 30 minutes over the whole field of view of the Meteosat Second Generation (MSG) satellite. It is a product that has been designed and is operated in the framework of the Land Surface Analysis Satellite Application Facility (LSA-SAF). Its temporal resolution and continental coverage is unprecedented. Nevertheless, the spatial resolution of the LSA-SAF ET might be too coarse for ecosystem monitoring; certainly in fragmented landscapes. This study evaluated the feasibility of generating ET estimates at moderate spatial resolution (~1 km) by incorporating vegetation data and using the spatial grid from a polar-orbit satellite platform in the estimation process. We used SPOT-Végétation data and followed two methodological paths. The first method was based on including SPOT-Vgt data in the forcing of the LSA-SAF ET algorithm. The second method consisted in deriving an empirical relationship between daily ET and major ET covariates (radiation, vegetation, surface temperature and soil moisture). The estimates were contrasted against measurements at flux towers in different biomes across Europe. The resulting ET estimates exhibited general good correlation with the validation data and allowed the detection of periods of the growing season where improvement is needed. The performance of the first method was better in most of the cases.