We live in a world of rapid social, economic and ecosystem change, facing major environmental challenges such as global warming, biodiversity loss and pressures on natural resources. Addressing these topics requires world-class ecosystem research by a well-connected, extensive community of experts, supported by advanced sites and facilities, openly shared and easily accessible data and capacity building programs. This is the goal of the Integrated European Long-Term Ecosystem, critical zone and socio-ecological system Research Infrastructure (eLTER RI).

eLTER RI will adopt a fundamentally systemic approach to observe and analyse the environmental system, encompassing biological, geological, hydrological and socio-ecological perspectives. It will be the first research infrastructure capturing and analysing holistically the integrated impacts of climate change alongside other pressures on a wide variety of European ecosystems. Ca. 200 eLTER research sites will provide a wide scale and systematic coverage of major European terrestrial, freshwater and transitional water ecosystem. eLTER RI will allow in-situ, co-located gathering of Essential Variables ranging from bio-physico-chemical to biodiversity and socio-ecological data. Ecosystem change caused by long-term pressures and short-term pulses will be investigated in a nested design from the local to the continental scale. With the huge number of in-situ sites and platforms and the harmonized and standardized observation concept, the eLTER RI offers outstanding new perspectives for hydrological research in Europe.

One of the major aims of long term ecosystem monitoring and research is to provide quality controlled and reliable data to support scientific analyses and enable input for designing environmental policies and assessing their impacts. Both the concept and in-situ design as well as the basic architectures and tools of the eLTER RI to support data providers and data users will be presented.

How to cite: Zacharias, S., Vereecken, H., Bäck, J., and Mirtl, M.: eLTER RI – A new European Research Infrastructure addressing today’s environmental challenges – a new perspective for European hydrological research, A European vision for hydrological observations and experimentation, Naples, Italy, 12–15 Jun 2023, GC8-Hydro-25, https://doi.org/10.5194/egusphere-gc8-hydro-25, 2023.

Inland surface water, especially lakes and small water bodies, are essential resources and have impacts on biodiversity, greenhouse gases and health. This is particularly true in the semi-arid Sahelian region, where these resources remain largely unassessed, and little is known about their number, size and quality. Remote sensing monitoring methods remain a promising tool to address these issues at the large scale, especially in areas where field data are scarce. Thanks to technological advances, current remote sensing systems provide data for regular monitoring over time and offer a high spatial resolution, up to 10 metres.

Several water detection methods have been developed, many of them using spectral information to differentiate water surfaces from soil, through thresholding on water indices (MNDWI for example), or classifications by clustering. These methods are sensitive to optical reflectance variability and are not straight forwardly applicable to regions, such as the Sahel, where the lakes and their environment are very diverse. Particularly, the presence of aquatic vegetation is an important challenge and source of error for many of the existing algorithms and available databases.

Deep learning, a subset of machine learning methods for training deep neural networks, has emerged as the state-of-the-art approach for a large number of remote sensing tasks. In this study, we apply a deep learning model based on the U-Net architecture to detect water bodies in the Sahel using Sentinel-2 MSI data, and 86 manually defined lake polygons as training data. This framework was originally developed for tree mapping (Brandt et al., 2020, https://doi.org/10.1038/s41586-020-2824-5).

Our preliminary analysis indicate that our models achieve a good accuracy (98 %). The problems of aquatic vegetation do not appear anymore, and each lake is thus well delimited irrespective of water type and characteristics. Using the water delineations obtained, we then classify different optical water types and thereby highlight different type of waterbodies, that appear to be mostly turbid and eutrophic waters, allowing to better understand the eco-hydrological processes in this region.

This method demonstrates the effectiveness of deep learning in detecting water surfaces in the study region. Deriving water masks that account for all kind of waterbodies offer a great opportunity to further characterize different water types. This method is easily reproducible due to the availability of the satellite data/algorithm and can be further applied to detect dams and other human-made features in relation to lake environments.

How to cite: de FLEURY, M., Kergoat, L., Brandt, M., Fensholt, R., Kariryaa, A., Kovács, G. M., Horion, S., and Grippa, M.: Sentinel-2 MSI for mapping Sahelian water bodies using a U-Net network, A European vision for hydrological observations and experimentation, Naples, Italy, 12–15 Jun 2023, GC8-Hydro-22, https://doi.org/10.5194/egusphere-gc8-hydro-22, 2023.

The land components of Earth System Models (ESMs) are increasingly used to predict changes in surface climate and hydrological processes at the global scale. However, due to their coarse resolution, these models still struggle in representing the fine-scale spatial variability of key land variables important for hydrological applications, such as snow cover, land surface temperature, and soil moisture. To address this limitation, we test here a sub—grid model structure recently developed for the Geophysical Fluid Dynamics Laboratory (GFDL) ESM4.1 land model. This novel model structure employs a machine learning technique and high-resolution terrain data to partition each land surface model grid in a set of land ‘tiles’ with homogeneous physical properties, which are then used to learn land processes at scales finer than the nominal land model resolution. This technique is especially relevant over complex terrain, where we can use elevation information to refine model predictions of the local energy balance and hydrological processes. Over each topography-aware model tile, the land model can thus provide local estimates of relevant land variables which can be then combined to produce high resolution maps and learn their spatial variance. As a proof of concept, here we will compare this modelling approach with satellite observations of land surface temperature, evaluating the skill of the model in reproducing land heterogeneity over complex topography regions. This analysis can be extended to other variables of hydrological interest, in particular soil moisture. As land models of increasing spatial resolution are being developed, our analysis here underlines the importance of evaluating not only grid average model output, but also predicted spatial variability using observational datasets.

How to cite: Zorzetto, E., Malyshev, S., Chaney, N., and Shevliakova, E.: Can Earth System Models represent the spatial variability of land surface processes over complex terrain?, A European vision for hydrological observations and experimentation, Naples, Italy, 12–15 Jun 2023, GC8-Hydro-83, https://doi.org/10.5194/egusphere-gc8-hydro-83, 2023.

Soil-vegetation-atmosphere transfer (SVAT) schemes explicitly consider the role of vegetation in affecting water and energy balance by considering its physiological properties. However, most current SVAT schemes and hydrological models do not consider vegetation a dynamic component. The seasonal and monthly evolution of the physiological parameters is kept constant year after year. This fact is likely crucial in transient climate simulations for hydrological models used to study climate change impact. Therefore, the analysis of vegetation dynamics became crucial to study these scenarios.

Vegetation dynamics, especially over large scales, can be monitored using remote sensing. The Normalised Difference Vegetation Index (NDVI) is still the most well-known and frequently used spectral indices derived from remote sensing, identifying vegetated areas and their condition. NDVI is based on plants' differential reflectance for different parts of the solar radiation spectrum.

In this work, we present a classification of rangelands in Spain based on the NDVI time series using them, like the result of SVAT and defining metrics and the Hurst Exponent from detrended fluctuation analysis. These areas are located in different precipitation and temperature regimen but with a Mediterranean climate with different aridity grades: Huescar, Castuera and Lozoya. K-means and unsupervised random forest were used to cluster the pixels using time series metrics and Hurst exponents. The clustering results will be discussed by comparing them to climate and topographical data.

References

Sanz E, Sotoca JJM, Saa-Requejo A, Díaz-Ambrona CH, Ruiz-Ramos M, Rodríguez A, Tarquis AM. Clustering Arid Rangelands Based on NDVI Annual Patterns and Their Persistence. Remote Sensing. 2022; 14(19):4949. https://doi.org/10.3390/rs14194949

Acknowledgements

Financial support from the project "CLASIFICACIÓN DE PASTIZALES MEDIANTE MÉTODOS SUPERVISADOS - SANTO" code RP220220C024, by Universidad Politécnica de Madrid, is highly appreciated.

How to cite: Sanz, E., Martín Sotoca, J. J., Saa-Requejo, A., Díaz-Ambrona, C. H., Ruiz-Ramos, M., Rodríguez, A., Almeida, A., Moratiel, R., and Tarquis, A. M.: Clustering Rangelands Based on NDVI Annual Patterns with different aridity grades, A European vision for hydrological observations and experimentation, Naples, Italy, 12–15 Jun 2023, GC8-Hydro-126, https://doi.org/10.5194/egusphere-gc8-hydro-126, 2023.