Human imprint on Land surface Interactions with the Atmosphere over the Iberian Semi-arid Environment – HILIAISE

HILIAISE advanced our understanding of the impact of semi-arid natural and anthropogenic processes on land atmosphere interactions and the regional scale hydrological cycle by conducting process studies using synergy between multiple state-of-the-art land surface models (LSMs) and several 3D high-resolution mesoscale models and LES confronted by detailed observations and remote sensing-derived products from the LIAISE field campaign. To this end, the project was divided into three workpackages (WPs).

In WP1, the field campaign observations using eddy-covariance techniques, scintilometers, surface temperature (LST) and soil temperature and moisture (SSM) probes were used to evaluate and improve LSM model representations of semi-arid (such as baresoil evaporation) and anthropized (irrigation) processes/exchanges with the atmosphere at the field scale for mutliple representative surfaces in the region, such as cash-crops consisting in flood-irrigated corn, alfalfa, drip irrigated apple trees and grapes, and rainfed almonds, and a natural grass field along with a lake surface.

In WP2, the atmospheric boundary layer observations from the field campaign were processed and case studies were undertaken by multiple state-of-the-art mesoscale atmospheric models (used for both research and operational short term weather forecasting) and compared to observations, and new LSM parameterizations of irrigation have been evaluated in terms of the impact on low level regional scale meteorology and circulation patterns using radio sounding profiles and aircraft atmospheric turbulence data from the field campaign. Spatially distributed high resolution SSM, LST and surface flux maps were produced a combination of satellite data, in-situ observations and data assimilation. This data was used to study the significant surface heterogenity over this highly anthropized region, and these products were also used as low boundary conditions in large eddy simulations in order to study how such heteorgenities impact the lower atmosphere.

In WP3, data from the soil moisture networks and remote sensing data (satellite, SLAP and GLORI products) were used to estimate the timing and amount of irrigation over a sub-region of the irrigated zone. Land data assimuilation of LAI into a LSM was also used to improve the mapping of irrigated regions compared to static maps. The resulting products and methodologies are being used both as input and to evaluate the performance of land surface hydrological model irrgiation modules/algorithms. Finally, LSM/hydrological models have been used to study high resolution hydrological processes over the Ebrob basin. They will contribute to improve the representation of the effects of anthropization on the energy and water cycles by operational NWP and GCM models.

The LIAISE campaign advanced understanding of evapotranspiration (ET), soil moisture, and irrigation impacts in semi-arid environments. At the field scale, an analysis of observations revealed large variations in surface energy partitioning which arose from differences in soil moisture, vegetation type, cover, and heterogeneity. Transpiration-driven ET produced daytime stable stratification. Land surface models (LSMs) tend to overestimate bare-soil evaporation, so that work was done to incorporating a dry surface layer (DSL) resistance into the SURFEX LSM which resulted in a reduction in latent heat flux errors by ~30% and better capturing post-rain dynamics. In addition, a study using lidar data determined that the turbulent diffusivites used to compute ET in LSMs are in need of improvement.

At the regional scale, surface soil moisture (SSM) was estimated with airborne GNSS-R at 100 m resolution and compared with other products. Sentinel-1/2-derived SSM consistently outperformed disaggregated Soil Moisture Active and Passive mission estimates, though both struggled with tree crops and small-scale irrigation. Sentinel-1 products captured large-scale wetting events but not field-scale irrigation, while GNSS-R mapping showed strong potential for agricultural monitoring. Irrigated-area mapping using Sentinel-1/2 confirmed the feasibility of SAR–optical approaches for near-real-time irrigation detection, albeit with caveats (e.g. dependence on vegetation cover, event size >10 mm, masking by rainfall).

Comparisons of spatially distributed ET products revealed that GLEAM performs well in rain-fed zones but poorly in irrigated areas, whereas SEN-ET showed higher accuracy for irrigated zones. While remote-sensing ET algorithms captured broad irrigated patterns, discrepancies among products highlight remaining uncertainties in actual ET. LSMs were then used to simulate the energy and water fluxes over the entire basin, but they struggled at high resolution, reflecting limitations in representing irrigation and surface water redistribution. Remotely sensed surface temperature was found to hold promise to help understanding model behaviors.

A mesoscale model intercomparison revealed major errors over irrigated regions. Observations showed irrigation dominated surface energy partitioning, suppressed buoyancy flux, lead to cooling of the boundary layer (several degrees), increased specific humidity by ~50%, and reduced convection. Improving the representation of irrigation in one mesoscale model significantly improved simulations, capturing irrigation-induced circulations and even revealing that irrigation modifies regional sea-breeze circulations. Altogether, the LIAISE results demonstrate that irrigation profoundly alters ET, surface fluxes, boundary-layer structure, and regional circulations, with critical implications for weather forecasting, climate modeling, and sustainable water resource management.

The LIAISE database of observations and scientific findings underpins several international modeling initiatives that will soon be open to external groups. The first is a land surface model (LSM) intercomparison project, involving offline simulations of surface energy budgets and land processes at eight contrasting sites during the Long Observation Period (LOP), spanning key months of the plant growth cycle. These sites, ranging from irrigated crops and grasslands to natural drylands, aim to improve the representation of irrigation in LSMs for hydrological and climate projections.

Building on the first LIAISE mesoscale intercomparison (Jimenez Cortez et al., 2025), a second initiative focuses on including irrigation processes, crucial for operational numerical weather prediction (NWP) at kilometer-scale resolution. A third initiative targets multi-model basin-scale water budget estimates for the Ebro basin, explicitly accounting for human influences. Preliminary regional LSM simulations at kilometer resolution (1989–2013, with WFDEI 0.5° forcing) used default configurations with little or no irrigation; future experiments will integrate irrigation and river routing to assess the benefits of high-resolution forcing data. Support from the GEWEX project under the WCRP will be sought to maximize international participation and visibility.

Beyond these intercomparison efforts, work will continue to refine physical processes in land surface and turbulence components of atmospheric models, particularly heterogeneity, through collaborations among French partners and researchers abroad, especially in Spain. A doctoral project beginning in 2025 (SDU2E) will use aircraft data from LIAISE to evaluate physical processes in high-resolution atmospheric models (large-eddy simulation). Additional work at CNRM will improve representation of semi-arid processes such as bare-soil evaporation. Both CNRM and IPSL will further advance land surface and hydrological modeling in coupled climate models for future water resource projections. Finally, remote sensing–based algorithms for irrigation decision-making will be improved, building on progress from HILIAISE.

The overarching objective of HILIAISE is to better understand and model the human imprint on semi-arid energy and water cycles.
To obtain this objective:
1) The project will have a long term field campaign with a 15-day Special Observing Period during summer 2020 over the Ebro basin in northeastern Spain when land surface heterogeneities are at their maximum. This campaign will focus on surface and boundary layer contrasts between irrigated and non-irrigated natural regions, and quantifying the water resource demand.
2) The project will use a multidisciplinary approach using a suite of hydrological, land-surface and meteorological models focusing on both using existing and improving parameterizations of anthropization and semi-arid surfaces.
The improved representation of anthropogenic and semi-arid processes in models will form the foundation for improving the understanding and prediction of water resource changes for the recent past, present and under future climate change.