OBSERVATIONS – Evapotranspiration (LIAISE – GEWEX-dET)

  Oral Presentations                    
No. Name Affiliation Co-authors Abstract L-WG1 L-WG2 L-WG3 OBS RS MOD
17 Altava-Ortiz, Vicent Servei Meteorològic de Catalunya –
Benaiges, Aleix – Servei Meteorològic de Catalunya, Spain;
Barnolas, Mercè – Servei Meteorològic de Catalunya, Spain;
Calculation of reference evapotranspiration using different Penman-Monteith equation approximations and radiation parameterizations

It helps monitor hydric balances or study the evolution of an ongoing drought event at a local scale. ETo values from meteorological data are also used in the calibration of products derived from satellite or lidar measurements or even in the verification of ETo values obtained with Numerical Weather Prediction models. However, ETo values may differ considerably depending on the selected method-equation.
Among the available equations in the bibliography, we want to highlight the Penman-Monteith (PM) method due to its exhaustive approach. However, this exhaustivity also demands much meteorological information which is not always available for long periods. This problem is partially solved by using parameterizations.
In this study, we have centered on the ETo differences using hourly or daily meteorological data provided by the Automatic Weather Station Network (XEMA) of the Meteorological Service of Catalonia (SMC). In particular, we have studied the contribution of the nocturnal period to the total daily values. Differences among calculations and the nocturnal contribution have been analyzed according to the seasonal cycle and evaluated in differentiated meteorological synoptic contexts. 
We have also compared different calculations of ETo at hourly time resolution using some radiation parameterization proposed in FAO 98. Concretely, we have used one parameterization for short-wave radiation and two for large-wave radiation using soil temperature values at different depths. This study is carried out using 6 stations of the XEMA with complete measurement equipment.
0.5 1 0 0.75 1 0 0
26 Altés, Víctor Department of Environment and Soil Sciences, Universitat de Lleida / isardSAT –
Bellvert, Joaquim – IRTA, Spain;
Pascual, Miquel – Universitat de Lleida, Spain;
Villar, Josep Maria – Universitat de Lleida, Spain;
Drainage Dynamics and Irrigation Management in the Algerri-Balaguer Irrigation District (Lleida, NE Spain)

Irrigation is one of the main users of water worldwide and its overusemay affect the natural regimes of water systems. To avoid this, drainage and irrigation management needs to be improved. This study aims to determine the amount of water lost to drainage in a semi-arid Mediterranean irrigated area. Water use, rainfall and drainage were monitored for 12 months (2019–2020) in a 425 ha sub-basin in the Algerri-Balaguer irrigation district (8000 ha, NE Spain). In addition, irrigation requirements were estimated using the single-crop FAO-56 method and a two-source energy balance model (TSEB) was used to estimate actual evapotranspiration in the sub-basin. Water lost to drainage in the sub-basin was estimated as 18% of the total water that entered the perimeter as irrigation and rainfall, which are almost five time higher than theoretical requirements of leaching. Out of the total drainage water, 57% was estimated to be irrigation water and 43% rainwater. The average amount of irrigation water used was 614 mm and irrigation efficiency in the sub-basin was estimated
at 80.2% and averaged actual evapotranspiration with the TSEB at 1144 mm. The available margin of improvement is between 19.3% of the present irrigation drainage ratio and the 3.8% estimated with the leaching requirement model.
0.75 1 0 0.5 1 0 0
25 Bellvert, Joaquim Institut de Recerca i Tecnologia Agroalimentàries (IRTA) –
M. Pamies-Sans – IRTA, Spain;
P. Quintana-Seguí – Observatori de l’Ebre, Spain;
J. Casadesús – IRTA, Spain;
Analysis and forecast of crop’s water demand in the irrigation districts of Lleida: estimation of remote sensing ET through Copernicus-based Inputs

The sustainability of irrigated agriculture is strongly linked to the improvement of water use efficiency. This can be primarily achieved by improving water management and use at farm and/or irrigation district (ID) level. To improve the efficiency of water management in IDs passes by closing the gap between water supply and demand. Knowledge of water supply is based on information on impounded volumes, current flows, historical series and information from snow reserve models. It is in relation to crop water demands where the uncertainty is greater, because despite there are historical series, there is no knowledge of accurate crop potential water consumption in real time, so it is impossible to characterize demands throughout the irrigation campaign. ID managers ask for the development of tools to support decision-making in planning and water management. Analysis of the temporal and spatial variability of water demands are necessary to encompass the diversity of irrigators and to evaluate the operational and management decisions of the irrigation districts. Also, modeling water demands in future climate change scenarios will be useful for ID managers to make decisions about changes in land use and/or levels of maximum water restrictions to adopt in conditions of water scarcity. The present study aims to take advantage of the analysis of time-series of actual evapotranspiration (ETa) obtained with the two-source energy balance (TSEB) applied as part of the SEN-ET program, to quantify, compare, and analyze differences in crop water demands in several ID located in the Ebro basin during six consecutive growing seasons (2017-2022). A particular analysis has been conducted comparing 2022 as a year with a severe drought with the average. IDs are characterized by having different water allocations and regulations in water management, irrigation systems and crops. A deep comparison analysis between IDs and crop’s water demand by crop type has been carried out. In addition, projections of water demands in the area has been assessed until 2100 based on the simulation of ETa under six different global climatic models (bcc-cms1, CNRM-CM5, inmcm4, MIROC-ESM, MPI-ESM-MR, MRI-CGCM3) in two contrasting Representative Concentration Pathways (RCP) scenarios. Results showed that rainfall in 2022 was 13% less and ETa 8% higher in comparison with the average. This corresponded with 14% more in gross irrigation requirements. Projections of potential crop water demand indicated an increase of 20 and 35% by 2100, respectively for an scenario RCP4.5 and 8.5.
1 1 0 1 1 1 0.5
23 Company, Jaume University of the Balearic Islands –
Cristòbal, Jordi – IRTA, Spain;
Fortesa, Josep – University of the Balearic Islands, Spain;
García-Comendador, Julián – University of the Balearic Islands, Spain;
Estrany, Joan – University of the Balearic Islands, Spain;
Surface energy fluxes retrieval using a two-source energy balance model to improve water resources in a Mediterranean insular region: Preliminary results

Mediterranean island regions are witnessing unprecedented changes in response and impacts to climate warming. However, the feedbacks between the hydrosphere and extreme episodes, whether of anthropogenic or climatic origin, require significant improvements in their quantification, especially in the face of extreme episodes such as droughts and forest fires. The present study aims to investigate the spatio-temporal patterns of energy fluxes, especially latent heat, controlled by changes in the conditions of the soil surface and plant cover, under the effects of global change at the local (drainage basins) and regional scales. As a previous step, a two-source energy balance (TSEB) method (Kustas and Norman, 1999) was calibrated and evaluated using flux tower data from 2021 to 2022 to retrieve energy fluxes (ET, sensible heat flux, net radiation, and soil heat flux) at the local scale in a gauged watershed equipped with three gauging stations in integrated basins to carry out continuous monitoring of the export of water and sediments at Sa Font de la Vila (with an area about 4.8 km2). This watershed is characterized by afforestation processes of marginal farmland similar than those in Mallorca. Moreover, the watershed been affected in the last 20 years by two large forest fires which make it particularly interesting to understand the surface energy fluxes have change due to forest fires. 
An experimental 4 m tall flux tower located in a Mediterranean shrub plot representative of the natural vegetation present in the watershed was used to provide the surface temperature and meteorological inputs at half-hour timesteps for TSEB. As vegetation properties inputs, remote sensing estimates of LAI and green fraction from Aqua and Terra MODIS products were used. Preliminary results yielded an acceptable agreement between TSEB model and flux tower data. Average RMSE for all fluxes was about 50 W •m-2 which agrees with other studies using similar methods. Further efforts will be focused on the daily energy flux integration by means of the implementation of the Dual-Temperature-Difference models and SEN-ET.
0.25 1 0 1 1 1 0.5
21 Cristóbal, Jordi Institut de Recerca i Tecnologia Agroalimentàries (IRTA) –
Bellvert, Joaquim – IRTA, Spain;
Martínez-Villagrasa, Daniel, Un de les Illes Balears, Spain;
Pàmies-Sans, Magí – IRTA, Spain;
Martí, Belén, Universitat de les Illes Balears, Spain;
Cuxart, Joan, Universitat de les Illes Balears, Spain
Estimating daily evapotranspiration using the SEN-ET modelling framework during the LIAISE field campaign (Lleida, 2021)

In the coming decades, world food production will have to coexist with water’s growing scarcity, extreme climatic events such as droughts or heatwaves, the limited availability of arable land and unpredictable costs, among many other considerations, while still meeting the needs of a world population that is expected to have increased around 70% by 2050. Good and effective agricultural water management is therefore essential. In this regard, evapotranspiration (ET) is a fundamental variable as is one of the key parameters in the water balance equation, making it an indispensable factor in the monitoring of crop water status and the optimization of irrigation scheduling, especially in Mediterranean semi-arid regions. Currently, satellite remote sensing offers a unique perspective for operational ET monitoring that complements the in situ-based climate and hydrologic data and tools that are traditionally used, including for example the eddy covariance (EC), the Bowen ratio (BR) and lysimeters. Moreover, as the LIAISE project states, the understanding of the impact of anthropization and its representation in models have been inhibited due to a lack of consistent and extensive observations. This is especially relevant for ET, which is a key variable of the hydrological cycle, in which regional estimates are needed at regional scales.
Currently, to improve the lack of spatio-temporal ET estimates there are several initiatives. One of them is the SEN-ET modelling framework (http://esa-sen4et.org) based on the Two-Source Energy Balance model (TSEB) that allows estimating high-resolution ET daily estimates at a spatial resolution of 20 m by sharpening thermal observations from Sentinel-3 satellites (1km, daily) and optical observations from Sentinel-2 satellites (20m, every 5 days) in an operative way.
In this study SEN-ET was applied and validated using the LIAISE flux towers and field observations for different crops during the whole LIAISE campaign. Preliminary results in an apple orchard located at the IRTA-Mollerussa Experimental Field Station and in a corn field also in Mollerussa from May to October 2021 yielded an RMSE of around 1 mm·day-1 for a total 132 and 67 days, respectively. Moreover, SEN-ET showed a robust time series consistency and sensitivity when compared to the apple orchard flux tower in a water stress induced experiment carried out for 15 days in mid-July 2021. Although the model is currently being evaluated over other covers, these results suggest that this modelling framework can estimate ET in an accurate and feasible way.
0.75 1 0 1 1 1 0.5
12 Cuxart, Joan Universitat de les Illes Balears –
Boone, A.A., CNRM – MeteoFrance, France;
Le Moigne, P., CNRM – MeteoFrance, France;
Martí, B, Un of the Balearic Islands & CNRM – MeteoFrance, France;
Price, J., UK MetOffice, UK;
Miro, J.R. – Meteorological Service of Catalonia, Spain;
Groh, J., University of Bonn & Forschungszentrum Jülich, Germany;
Martínez-Villagrasa, D., University of the Balearic Islands, Spain;
ET: spatial variability of EC values depending on the state of the surface

During the experimental campaign organised by the LIAISE Initiative in summer 2021 in the Eastern Ebro basin, several surface energy budget stations were deployed over surfaces with different characteristics. In this study we employ data from seven of them: three at the Mollerussa site (corn, grass and apple orchard), two at or near the rainfed site of Els Plans (natural vegetation and non-irrigated almond orchard), a growing irrigated alfalfa field at the site of La Cendrosa and the station over the Ivars lake. All these stations were within a 10 km-radius circle centred in the town of Bellpuig.
For dry weather conditions, the values of the latent heat flux (or evapotranspiration, ET) were very different depending on the surface, varying from more than 500 W/m2 over the irrigated corn field to less than 30 W/m2 over dry mostly bare soil at the rainfed site. These different values implied also different sensible heat fluxes, with the irrigated surfaces showing thermal stable stratification in the afternoon well before sunset due to the cooling effect of evaporation.
The contrast between the irrigated and the rainfed areas was very well-marked, inducing eventually mesoscale circulations and local transport of humidity, which were overruled by the frequent sea breeze arrival in the afternoon. At the rainfed site and the sub-hourly scale, ET proceeded very irregularly in time and may challenge the fundamental hypothesis of stationarity for the computation of the turbulent fluxes. Over all the well-watered surfaces, classical estimates of Potential ET, such as Penman-Monteith or Priestley-Taylor, overestimated the observed values.
1 1 0 0.25 1 0 0
18 Girona, Joan Institut de Recerca i Tecnologia Agroalimentàries (IRTA) –
del Campo, Jesus – IRTA, Spain;
Mata, Merce – IRTA, Spain;
Heat waves effects on an apple orchard evapotranspiration evaluated by a weighing lysimeter.

Heat waves have increased considerably in recent years and according to current forecasts, they will become more and more frequent. In a generic way, there is the perception that every time one of these heat waves appears, the crops suffer a water deficit higher than what could be expected due to the climatic conditions. The aim of this work has been to assess whether the plants' water demand follows the guidelines proportional to the Reference Evapotranspiration (ETo) provided by the weather stations or if there is something different in these circumstances. Comparing the behavior of an apple tree plantation in which a weighing lysimeter has been installed, it has been found that during heat waves there is a 25% higher water demand than in more usual conditions, which 9 % (of this 25%) would be explained by the increase in ETo, but that the rest is due to a differential behavior of the plant. The work analyzes possible causes that could induce this significant increase in the water demand of the apple tree during a heat wave.
0.75 1 0 0.75 1 0 0
22 Gómez-Giráldez, Pedro J. CSIC-EBD –
Cristóbal, J – IRTA, Spain;
Nieto, Héctor – CSIC, Spain;
Díaz-Delgado, Ricardo – CSIC-EBD, Spain;
Anderson, Martha – USDA-ARS, USA;
Kustas, William – USDA-ARS, USA;
Estimation of evapotranspiration through remote sensing in a Juniper tree ecosystem at the Doñana National Park

Doñana National Park is located in the southwest of the Iberian Peninsula, where water scarcity is recurrent together with high heterogeneity in species and ecosystems. Monitoring of evapotranspiration (ET) has important implications because it plays a key role in global and regional climate modeling, the knowledge of the hydrological cycle as well as to assess the water stress that affects agricultural and natural ecosystems. Remote sensing can provide the necessary spatial information for the calculation of the evapotranspiration at global scales and in an economic way. In this work, a first evaluation of the modeling of water and energy fluxes is performed through the Two Sources Energy Balance (TSEB) model using Terra/Aqua MODIS images during the period 2014-2020 in an experimental plot of juniper (Juniperus phoenicea) in the Doñana Biological Reserve, where there is an eddy covariance tower installed for validation. TSEB yielded acceptable results (an average RMSE of 50 W·m-2) when measuring  surface  energy fluxes, taking into account the high heterogeneity that this type of vegetation presents. Future improvements in the determination of the leaf area index and the inclusion of more images and Eddy-Covariance stations will help to implement and improve the model in this type of vegetation.
0 1 0 1 1 1 0.5
19 Jomé, Mathilde Laboratoire d’Aérologie, CNRS, Université de Toulouse –
Lohou, Fabienne – Lab d’Aérologie, CNRS, Un de Toulouse, France;
Lothon, Marie – Lab d’Aérologie, CNRS, Un de Toulouse, France;
Kelley, Jason – Asperatus Consulting, USA;
Pardyjak, Eric – University of Utah, USA
Using Artificial Neural Network to estimate surface convective fluxes.

Earth’s surface and its properties impact the atmosphere on different scales. Thus, understanding the interactions between the surface and the atmosphere is important to establish and control global and regional numerical models. In February 2019, the Working Group on Numerical Experimentation (WGNE) reported that the bias observed in surface convective fluxes were the second largest source of errors in global and regional numerical models.
Reducing these errors by gaining a better understanding of the impact of surface-atmosphere interactions over heterogenous land surfaces is one of the main objectives of the Models and Observations for Surface-Atmosphere Interactions (MOSAI) project. Because an experiment set-up designed to study the impact of surface heterogeneity on surface convective fluxes is quite expensive, we tested a new method, based on Artificial Neural Networks (ANNs), that has been proven to be efficient in previous studies in estimating surface convective fluxes and evapotranspiration based on lower-cost measurements. Standard lower-cost meteorological stations co-located with higher-cost Eddy-Covariance (EC) flux stations can be paired and used to train an ANN estimate surface fluxes based only on classical meteorological variables. Using this method, one may then estimate fluxes for various vegetation covers.
After having experimented with this method on a few ideally chosen days to define the influence of the training dataset, the topography of the network and the choice of the input meteorological variables, we ran those tests on a year-long dataset. The aim was to propose an experimental deployment plan for a network of meteorological stations covering the various surfaces. A reference EC-station was moved around from one low-cost station to another in order to train each station in estimating the most accurate fluxes. We used this year-long dataset to test several rotation frequencies of the reference station for application of the method to future campaigns. The first results proved that estimating surface convective fluxes with ANNs using only a few variables and a simple neural network topography is possible and should be valid for long-term monitoring of surface energy fluxes over heterogeneous surface using a three-week rotation period of the training EC-station.
0 1 0 0.75 1 0 0.25
20 Le Page, Michel Centre d’Etudes Spatiales de la Biosphère (CESBIO) –
Mwangi, Samuel – CESBIO, Un de Toulouse, CNES, France;
Boulet, Gilles – CESBIO, Un de Toulouse, CNES,  France;
Bellvert Joaquim – IRTA, Lleida, Spain;
Fanise, Pascal – CESBIO, Un de Toulouse, CNES,  France;
Lemaire, Baptiste – CESBIO, Un de Toulouse, CNES,  France;
Tous, David – SafSampling, Verdu, Spain;
Assessment of evapotranspiration over a vineyard during  the LIAISE campaign

Evapotranspiration (ET) is one of the more important water fluxes of cropped areas in semi-arid areas. An experiment has been set up on a vineyard in northeastern Spain from April to September 2021. The system consisted in measuring 1- the components of the energy budget (four-component net radiometer, eddy covariance system for the monitoring of turbulent fluxes, soil conduction plates), 2- soil moisture on the row and inter-row of the grapevine, 3- rainfall and irrigation for the water budget, and 4- the components of the surface temperature with two NDVImeters and three thermal infrared cameras. Various approaches for assessing ET with remote sensing observations have been tested. Those include using NDVI as a proxy for basal crop coefficient (Kcb), NDVI for the assessment of the geometrical structure, and energy budget schemes (SPARSE, and TSEB applied as part of the SenET program). Those tests showed that a good assessment of the grapevine geometry leads to a less empirical and transposable satellite remote sensing model of Kcb, at least when there is no inter-row cover. It also showed that while the use of a homogeneous radiative scheme may be sufficient when using near-nadir thermal measurements, better modeling of the canopy should allow better flux consistency, especially for higher view angles.
1 1 0 0.75 1 0.75 0.5
16 Martí, Belén CNRM, Météo-France/CNRS, GMME, Toulouse, France –
Martínez-Villagrasa, Daniel – Un of the Balearic Islands, Spain;
Price, Jeremy – UKMO, United Kingdom;
Cuxart, Joan – University of the Balearic Islands, Spain;
Estimation of latent and sensible heat flux through similarity theory in semi-arid conditions

Turbulent fluxes characterize the exchanges of matter and energy in the boundary layer between air and the underlying surface. The estimation of these fluxes with standard meteorological measurements can increase the number of turbulent flux estimations in a given area since it implies much less expensive instrumentation than the eddy-covariance (EC) systems. Monin-Obukhov’s similarity theory (MOST) can be applied for this purpose using two measurements of wind, temperature and humidity in the surface layer at different heights. This method employs empirically adjusted functions from measurements taken at instrumented towers. To assess the goodness of the estimations it is necessary to compare them to fluxes obtained from EC systems. It is worth noting that for the latent heat flux no widely accepted specific adjusting coefficients exist, using instead those of the functions for the sensible heat flux. The later assumption implies that the turbulent transport of heat and humidity is the same (the so-called Lewis number being equal to one), a hypothesis under question.
The Els Plans site of the Land surface Interactions with the Atmosphere over the Iberian Semi-arid Environment (LIAISE) campaign and the ECUIB site in Mallorca provide the contrast between two semi-arid locations with different soil water availability and natural vegetation cover. It is found that for the sensible heat flux new coefficients are necessary and adjusted when the latent heat flux predominates, mainly in conditions of high humidity. For the latent heat flux it is found that, with the current form of the function, it is necessary to provide different adjusted coefficients depending on the Bowen ratio (Bo, ratio between the sensible and latent heat fluxes). The relationship between Bo and the functions used is explored though the Lewis number, finding that it is close to one for low values of Bo and it rises as Bo increases.
1 1 0 0.5 1 0 0
13 Martinez-Villagrasa, Daniel Universitat de les Illes Balears –
Martí, Belén – Un of Balearic Islands, Spain / CNRM, France;
Groh, Jannis – Forschungszentrum Jülich, Germany;
Price, Jeremy – Met Office, UK;
Wrenger, Burkhard – TH Ostwestfalen-Lippe, Germany;
Cuxart, Joan – University of Balearic Islands, Spain;
The role of WPL correction in determining the latent heat flux through the eddy-covariance method at the rainfed site of Els Plans

The estimation of surface turbulent fluxes with the eddy-covariance (EC) technique is done through instrumentation that accounts for the fast response of primary meteorological variables measured at a certain distance from the surface. In practice, these measurements are typically subjected to additional corrections to mitigate the source of errors related to known micrometeorological effects. One of these effects is related with the density fluctuations produced by temperature and humidity variations, a process that has to be considered when determining the flux of any air constituent. In consequence, this correction (called WPL after Webb, Pearman and Leuning work in 1980) affects the estimation of CO2 and water vapour fluxes.

The characterization of the latent heat flux through the EC systems at different heights has been evaluated at the rainfed site of Els Plans during the LIAISE campaign and their results are compared against the observations from a smart-field-lysimeter. Although the WPL correction in daytime accounts for a couple of tens of W/m2, it doubles the initial value provided by the EC system and gets closer to the observations from the smart-field lysimeter, suggesting that this correction is fundamental when dealing with arid and dry conditions. The WPL correction is less important after a rain event, when the latent heat flux is larger due to enough water availability. The current work will present a detailed interpretation of the WPL correction and its effect on a semi-arid rainfed site such as Els Plans.
1 1 0 0.25 1 0 0
15 Moonen, Robbert Utrecht University –
Netherlands, The
Agmuas, Getachew – Utrecht University, the Netherlands;
Hartogensis, Oscar – Wageningen University, the Netherlands;
Vilà, Jordi – Wageningen University, the Netherlands;
Röckmann, Thomas – Utrecht University, the Netherlands;
Investigating diurnal ecosystem scale H2O and CO2 isotope fluxes in an irrigated semi-arid environment during the LIAISE 2021 field campaign

Validation of gas exchange fluxes in models has been challenging due to the lack of ecosystem scale exchange fluxes partitioned into soil, plant and atmospheric components. One promising method to partition turbulent fluxes uses the exchange process dependent fractionation of molecules like CO2 and H2O. When applying this method to short spatiotemporal scales, an isotope flux (δ-flux) needs to be resolved. Few have attempted to measure this δ-flux as the required instrumentation only became available in recent years. In our presentation we will discuss observations made during the LIAISE 2021 field campaign using an EC system, Picarro L-2130i H2O isotope analyser, and Aerodyne TILDAS-CS CO2 isotope analyser. This campaign took place in the summer of 2021 in the heavily irrigated Ebro River basin near Lleida, Spain embedded in a semi-arid region.
We will present procedures to estimate and scrutinize the central δ-flux variable. To this end we calculated co-spectra of the relevant signals and compared their frequency dependent contributions. One relevant finding is that mole fractions and isotope ratios measured with the same instrument can be offset in time by more than a minute, thereby impacting the resulting δ-fluxes. Additionally, we found asymmetric signal loss between net ecosystem fluxes and δ-fluxes. We will show that such effects impact flux partitioning severely and indicate how they can be tackled using physically sound corrections. Only when such corrections and verifications are made, ecosystem flux partitioning can be applied to validate conceptual land-atmosphere exchange models. Such models will calculate the diurnal variability of CO2 and H2O isotopologue concentrations, and link local to regional scales, all with the purpose of better constraining current and future exchange fluxes.
1 1 0 0.25 1 0 0
14 Oliver-Manera, Jordi Institut de Recerca i Tecnologia Agroalimentàries (IRTA) –
  Estimation of transpiration using sap flow sensors calibrated with whole canopy transpiration measurements

Approximately 95% of the plant water consumption is transpired which is coupled with photosynthesis since both are dependant of the stomatal resistance. However, measuring whole plant transpiration is a big challenge. One of the most useful tools used to estimate transpiration are sap flows sensors by measuring sap velocity and using plant architectonic and hydraulic characteristics to upscale to whole plant. However, estimating whole plant transpiration may contain many source of errors. First, sap flows only take measurements of a a part of the xylematic network of the plant. Second, some architectonic and hydraulic plant parameters are estimated inaccurately. Therefore, calibration against a real whole plant measurement is necessary to upscale for sap flow to whole plant. Differently to other systems of measuring transpiration, whole canopy gas exchange (WCGC) chambers can measure the whole plant transpiration of many individual plants for a period and can be changed to other plants the next period of measurements. We compared two methods of measuring transpiration on grapevines irrigated at 100% of the estimated evapotranspiration: 1) connected to a portable gas exchange system; 2) by the difference in water vapour in and out the WCGC Then we used the two methods to calibrate sap flows using the compensation heat-pulse method and to estimate the daily vine transpiration and the basal crop coefficients (kcb).
0.25 1 0 0.25 1 0 0
27 Paolini, Giovanni isardSAT –
Escorihuela Maria Jose – isardSAT, Spain;
Merlin Olivier – CESBIO, France;
Pamies-Sans Magi – IRTA, Spain;
Bellvert Joaquim – IRTA, Spain;
Classifying Irrigation System at Field-level using Remote Sensing Data and Machine Learning

This research presents a new methodology for classifying irrigation systems at the field level using time-series of high-resolution remotely sensed data. While previous research have focused on simply detecting irrigated areas, this study aims to classify different types of irrigation techniques. Precise knowledge of different irrigation systems is needed in order to correctly model the anthropogenic impact in various land surface models. Additionally, these maps are also useful for administrative purposes, to estimate and monitor changes in irrigation practices and consequently encourage more sustainable use of the freshwater resources. Three different ML models for time-series classification were used and compared for this task: time-series forest, Rocket and ResNET. The models were trained using data collected in a field campaign in Catalunya, Spain in 2020, which included over 300 fields with different crop types and four different irrigation system classes: sprinkler, flood, drip/subsurface and non-irrigated. The study demonstrates that irrigation systems can be accurately classified using actual evapotranspiration (ETa) and surface soil moisture (SM) at field scale (20m) and that a combination of these variables with the ResNET model yields the best results, with an accuracy of around 90%. Another finding was that all the ML models employed for the classification were able to distinguish different types of irrigation regardless of the different types of crops or trees present in each field, so the classification model can be used without specifying crop types. Finally, annual irrigation systems maps were created for different years, and a comparison with historical and administrative data was performed. This analysis revealed a general trend, across different irrigation districts, in moving from non-irrigated and traditional irrigation systems (i.e. flood irrigation) to modern systems (drip and sprinkler).
1 0 0 1 1 1 0
24 Quintanilla, Manuel Institut de Recerca i Tecnologia Agroalimentàries (IRTA) –
Joaquim Bellvert – IRTA, Spain;
Ana Pelechá – IRTA, Spain;
Jaume Casadesus – IRTA, Spain;
Omar García-Tejera – Universidad de La Laguna, Spain;
Xavier Miarnau – IRTA, Spain;
Assessment of transpiration in different almond production systems with two-source energy balance models using high-resolution aerial imagery

During last decade, new irrigated almond areas has increased in Spain, rising water demands.  In a context of water scarcity, the estimation of crop evapotranspiration (ET) is crucial to monitor irrigation. The ET can be retrieved from surface energy flux modeling approaches, such as the Two-Source Energy Balance (TSEB), using very high-resolution imagery. In this study, two TSEB models were compared to assess transpiration (T) in almonds planted under different production systems (PS) and irrigation strategies: TSEB Priestley-Taylor (TSEB-PT) and contextual (TSEB-2T). The PSs consisted on open vase with minimal pruning (5.5 x 3.5 m spacing), central axis (5 x 3 m spacing), and hedgerow (4.5 x 3 m spacing), while irrigation treatments were full irrigation, mild stressed (receiving 50% of full irrigation) and stressed (receiving 20% of full irrigation). An Unmanned Aerial Vehicles (UAV) equipped with thermal and multispectral cameras allowed us to obtain the main parameters required for both TSEB models. The T was validated with calibrated sap-flow sensors. Five UAV flights were conducted from March to July of 2021. During the flights Leaf area index (LAI), stem water potential (Ψstem) and fractional intercepted photosynthetically active radiation (fIPAR) were also measured. The PS showed significant differences in fractional canopy cover, tree height, LAI and measured transpiration (Tsf). The Tsf and the Ψstem also showed significant differences between irrigation treatments. Both the TSEB-2T and TSEB-PT models overestimated T with a bias of 0.99 and 1.22 mm/day, respectively. However, TSEB-PT had a higher error. The R2 decreased in the more intensive production system. We hypothesize that model overestimation can be associated with the radiative transfer model used to estimate radiation interception. Modelled fIPAR showed a RMSE and a bias of 0.14 and 0.12 in comparison to in-situ measured fIPAR. Our results suggest the use of the TSEB-2T model instead of TSEB-PT. Besides, further refinements on the radiation transfer modelling approaches are needed, especially for row crops with different production systems.
0.5 1 0 1 1 1 0.5
  Poster Presentations                    
No. Name Affiliation Co-authors Abstract L-WG1 L-WG2 L-WG3 OBS RS MOD
P9 Bozorgi, Mahsa Institut de Recerca i Tecnologia Agroalimentàries (IRTA) –
Cristóbal, Jordi – IRTA, Spain;
Pelachá, Ana – IRTA, Spain
Estimation of Evapotranspiration Using SEBAL Algorithm and Landsat-8 Data – A Case Study in a Mediterranean pistachio orchard

Evapotranspiration (ET) is amongst important variables in hydrological cycle and surface energy balance (SEB), especially for irrigated agricultural regions. According to climate models, global temperature is expected to rise which leads to changes in precipitation and ET that respectively add water and remove it from the water cycle. So, global warming changes the balance between water supply and water demand by shifting energy and water balance and eventually bring on water and food security.
Although numerous methods have been developed to quantify ET as a critical prerequisite for water management and increasing water consumption efficiency, estimating ET accurately is still challenging. Measuring actual ET directly only could be for a specific point and time and it can be expensive and time consuming. Hence, ET estimation models by satellite data are used to overcome the limitations. Remote sensing-based ET models can be divided into three categories. Surface Energy Balance Algorithm for Land (SEBAL) is a one-source energy balance model, and it is one of widely used models for estimating ET in large farms. As it has different efficiency by changing crops and environmental condition, 
The purpose of this study is to evaluate the daily ET accuracy obtained by SEBAL Algorithm using six Landsat 8 Operational Land Imager (OLI) images in comparison with in-situ flux tower measurements in a Mediterranean young pistachio tree orchard in Lleida (North-East of the Iberian Peninsula) from March to December 2022. Moreover, remote sensing data was used for estimating meteorological variables and the whole calculation is done in Google Earth Engine. It can be concluded that the performance of the algorithm in estimating the actual evapotranspiration has a robust time-series consistency compared to the flux tower measurements. As the SEBAL Algorithm is a very convenient method, it can be easily assimilated into irrigating systems.
0.5 1 0 0.5 1 1 0
P6 Canut, Guylaine CNRM – Météo-France, CNRS –
Aressy, P. – CNRM-Météo-France, France;
Barbié, J. – CNRM-Météo-France, France;
Bouhours, G. – CNRM-Météo-France, France;
Couzinier, J. – CNRM-Météo-France, France;
Etienne, J.C – CNRM-Météo-France, France;
Garrouste, O. – CNRM-Météo-France, France;
LeMoigne, P. – CNRM-Météo-France, France;
Moulin, E. – CNRM-Météo-France, France;
Rieublanc, S. – CNRM-Météo-France, France;
Surface energy balance and thermodynamic measurements over a mobil platform on Lake Ivars during the LIAISE field campaign.

During the LIAISE campaign, a surface energy balance station was deployed on Ivars lake in the irrigated area near the La Cendrosa super site for several months from May to October 2021. The main purpose of this station was to measure sensible and evaporative fluxes from this lake to provide a reference for the irrigated area. It is important to correctly simulate the energy balance of lakes at the local scale (Le
Moigne et al., 2013) so that this modeling can be applied in numerical weather prediction or climate models. Observations on the lake are quite rare, but they are essential to evaluate and improve the parameterization of the processes in the numerical models. The platform was positioned in the middle of the lake and was powered by a combination of a fuel cell and solar panels. This platform moved with the lake surface. Low frequency measurements (every minute) of temperature, humidity and wind were recorded at several levels in the lake and in the atmosphere to study the processes at the lake/atmosphere interface. Fast measurements (20hz) measurements of the 3 components of wind, temperature and water vapor from a sonic anemometer and a licor were made to calculate sensible and latent heat fluxes. The sonic anemometer was coupled to an inertial motion sensor (100hz) to correct the wind for the motion of the ship. This poster presents an overview of the data over the whole period, the quality of the latent and sensible turbulent fluxes after motion correction and eddy-covariance method with the aim of initiating a study using this dataset to explore and evaluate the parameterization of lake surface fluxes in the SURFEX model at CNRM. Le Moigne, P, Legain, D, Lagarde, F, Potes, M, Tzanos, D, Moulin, E, Barrie, J, Salgado, R, Messiaen, G,
Fiandrino, A, Donier, S, Traulle, O, Costa, MJ, 2013 : Evaluation of the lake model FLake over a coastal lagoon during the THAUMEX field campaign. TELLUS SERIES A-DYNAMIC METEOROLOGY AND OCEANOGRAPHY, Volume: 65, Article Number: 20951, http://dx.doi.org/10.3402/tellusa.v65i0.20951 
1 1 0 0.25 1 0 0
P8 Groh, Jannis Forschungszentrum Jülich, Leibniz Centre for Agricultural Landscape Research –
Pütz, Thomas – Forschungszentrum Jülich, Germany;
Graf, Alexander – Forschungszentrum Jülich, Germany;
Vanderborght, Jan – Forschungszentrum Jülich, Germany;
Price, Jeremy – UKMO, United Kingdom;
Martí, Belén – CNRM-MeteoFrance, Un of the Balearic Islands, Spain;
Martínez-Villagrasa, Daniel – Un of the Balearic Islands, Spain;
Cuxart, Joan – University of the Balearic Islands, Spain;
Interactions between the atmosphere and the land surface at night under semi-arid conditions

Most studies on water flow dynamics within the soil-plant-atmosphere system focus on the daytime, but not on the processes at night. Evapotranspiration (ET), which is the second largest water flow in the global terrestrial water cycle after precipitation, occurs mainly during the day. There is, however, evidence that a considerable amount of water loss through ET also occurs at night. Nocturnal water losses are counteracted by the formation of non-rainfall water inputs (NRW) from dew formation, fog, and water vapour adsorption. To better capture the interplay of different nocturnal water fluxes, observations and models need to account not only for evaporation and transpiration processes, but also on formation of NRW inputs. This is particularly important for arid and semi-arid areas, where every drop of water counts, as the ecosystems there can only draw on limited water resources. We used a high precision weighable lysimeter, a leaf wetness sensor and meteorological observation to quantify nocturnal water fluxes and identify which mechanisms govern nocturnal water fluxes under water limited conditions during the intensive measurement campaign of LIAISE in July 2021.
1 1 0 0.5 1 0 0
P5 Hartogensis, Oscar Wageningen University –
Netherlands, The
Mangan, Mary-Rose – Wageningen University, the Netherlands;
Cuxart Rodamilans, Joan – Université des Iles Baleares, Spain;
Martinez Villagrasa, Daniel – Université des Iles Baleares, Spain;
Martí, Belén – Université des Iles Baleares, Spain; / CNRM-MeteoFrance, Un de Toulouse, CNRS, France;
Bellvert, Joaquim – IRTA, Spain;
Cristobal, Jordi – IRTA, Spain;
Sobrino, Jose Antonio – Universidad de Valencia, Valencia, Spain;
Groh, Jannis – Forschungszentrum Jülich, Germany;
Siegmann, Bastian – Forschungszentrum Jülich, Germany;
Rascher, Uwe – Forschungszentrum Jülich, Germany
De Boer, Hugo – Utrecht University, the Netherlands;
Gonzalez Armas, Raquel – Wageningen University, the Netherlands;
Goulas, Yves – Laboratoire de Météorologie Dynamique, CNRS/IPSL, Ecole Polytechnique, Palaiseau, France;
Miró, Josep Ramon – Meteorological Service of Catalonia, Spain;
Mercader Carbó, Jordi – Meteorological Service of Catalonia, Spain;
Boone, Aaron – CNRM-Un de Toulouse, Météo-France/CNRS, France;
Evapotranspiration methods inter-comparison at LIAISE

This study aims to present an overview of field observation based methodologies to estimate evapotranspiration (ET). It focusses on instrumentation deployed at the agricultural research station IRTA in Mollerussa, Spain during the LIAISE field campaign in July of 2021. Given the number of methodologies deployed at IRTA during LIAISE, this study can be seen as a current state of the art in determining evapotranspiration.
The fields at and around IRTA covered by the instrumentation during LIAISE include apple orchards (full-irrigated and half-irrigated), corn, and irrigated grass (FAO-type meteorological station). The measurement methodologies cover different spatial scales ranging from leaf, to canopy to integrated measurements over multiple fields. Most methods measure evapotranspiration (ET) as the combined effect of soil evaporation (E) and plant transpiration (T) but individual terms were measured as well. Measurement techniques that will be considered include: eddy-covariance, gradient-method, optical-microwave scintillometer, tree lysimeters, micro-lysimeters, FAO Penman-Monteith method, Satellite-, aircraft- and drone remote sensing, and transpiration estimates based on plant measurements such as stomatal conductance and stem water potential.
This is work in progress and at the workshop we will present, at best, very preliminary results from one golden day (22 July 2021) from as many ET products available and organize them in a simple overview (daily time-series, daily total sum). The idea is to discuss the results further during the working group meeting at the workshop. As a follow-up this study will also serve as a reference for land surface modelling community within LIAISE.
1 1 0 0.25 1 0 0
P10 Jomé, Mathilde Laboratoire d’Aérologie, CNRS, Université de Toulouse –
Lohou, Fabienne – Lab d’Aérologie, Un de Toulouse, France;
Lothon, Marie – Lab d’Aérologie, Un de Toulouse, France;
Canut, Guylaine – CNRM-CNRS, Météo-France, France;
Couvreux, Fleur – CNRM-CNRS, Météo-France, France;
Derrien, Solène – Lab d’Aérologie, Un de Toulouse, France;
Maurel, William – CNRM-CNRS, Météo-France, France;
Etienne, Jean-Claude – CNRM-CNRS, Météo-France, France;
Vial, Antoine – Lab d’Aérologie, Un de Toulouse, France;
Garrouste, Olivier – CNRM-CNRS, Météo-France, France;
Dupont, Jean-Charles – LMD, CNRS, IPSL, Paris, France
How to evaluate the representativity of reference long-term surface flux measurements in an heterogeneous landscape ?

The meterological phenomena draw their energy from the Earth’s surface and dissipate most of their energy close to the surface. The land surface, through its topography, soil moisture, temperature or vegetation activity, impacts the atmosphere from daily to seasonal time scale. The Working Group on Numerical Experimentation survey on systematic errors established that the outstanding errors in the modelling of surface fluxes of momentum and sensible and latent heat is the second most important issue. Therefore to reduce these biases, an accurate assessment of the Land-Atmosphere (L-A) exchanges, and their correct representation, are essential for weather and climate forecasts.
The evaluation of L-A exchanges in numerical weather and climate prediction models in the context of heterogeneous surfaces is the main objective of the Models and Observations for Surface-Atmosphere Interactions (MOSAI) project. The main objective of our study corresponds to the first scientific objective of MOSAI : to investigate and determine the uncertainty and representativeness of L-A exchanges measured over heterogeneous landscapes by reference towers. To achieve these objectives, dedicated field experiments are needed to document the variability of the L-A exchanges within a grid mesh around the ACTRIS sites. A set of a one year-field campaign per site are currently performed at Meteopole (Toulouse, 2021), at SIRTA (Palaiseau, 2022) and at P2OA (Lannemezan, 2023).
The final objective is to build two new indicators to complete the measured surface flux at the ACTRIS sites. The first one concerns the horizontal representativeness of the local measured surface flux in the heterogeneous landscape at the scale of the ESM or NWP grid and needs two steps. The second indicator will quantify the surface fluxes uncertainties (random and systematic errors, and, surface-energy-balance non-closure). After carrying out an homogeneisation work of the dataset in order to facilitate the analysis and use of the so-called dataset, we documented the arrangements of the heterogeneities as a first step to understand the heterogeneity of our landscape. From there, we defined two indicators that aim to quantify the surface flux heterogeneities : one concerning the footprint and the other concerning the flux local spatial variability. Those heterogeneity indicators and the measurement uncertainty due to SEB non-closure will be cross-analysed to test the possible effect of significant horizontal advective flux, and/or an insufficient sampling of large-scale turbulent eddies due to landscape heterogeneity.
0 1 0 1 1 0 0
P7 Mangan, Mary Rose Wageningen University –
Netherlands, The
Hartogensis, Oscar – Wageningen University, the Netherlands;
Guylaine, Canut – CNRM – Météo-France, Un de Toulouse, France;
Le Moigne, Patrick – CNRM – Météo-France, Un de Toulouse, France;
Etienne, Jean-Claude, CNRM – Météo-France, Un de Toulouse,
Vilà, Jordi – Wageningen University, the Netherlands;
First Results of Optical-Microwave Scintillometer Measurements for Open Water Evaporation

Optical-Microwave Scintillometry (OMS) offers an alternative to eddy covariance for directly measuring latent heat flux.  Unlike eddy-covariance, OMS is not constrained by the same necessity for a sufficiently long averaging time for computing fluxes, so fluxes can be computed at high temporal frequency.  It also measures path-averaged flux instead of a point, so the OMS has a larger footprint than eddy covariance-based measurement techniques. While the OMS method has been verified over land, there have been few studies verifying the method over water with the exception of Lobos-Roco et al., 2022 who tested OMS over a saline lake in the Atacama Desert. In this analysis, we expand the work of Lobos-Roco et al., 2022 by comparing the results of sensible and latent heat fluxes as well as structure parameters between the OMS and the eddy covariance systems over Lake Ivars from the LIAISE field campaign. We will discuss the first results for the turbulent flux intercomparison from Lake Ivars, which can help to provide a critical validation of the OMS method over open water.
1 1 0 0.25 1 0 0