Over the past two decades, the technique of microwave remote sensing has made tremendous progress to provide robust estimates of surface and deeper-layer soil moisture at different scales. From local to landscape scales, several field or aircraft experiments have been organised to improve our understanding of active and passive microwave soil moisture sensing, including the effects of soil roughness, vegetation, spatial heterogeneities, and topography. At continental scales, a series of several passive and active microwave space sensors, including SMMR (1978-1987), AMSR (2002-), ERS/SCAT (1992-2000) provided information on surface soil moisture. Current investigations of L-band passive microwave observations with SMOS (2009-) and SMAP (2015-), and active C-band microwave observations with the Metop/ASCAT series (2006-) and Sentinel-1, enable an accurate quantification of the soil moisture at regional and global scales. Building on the legacy of these mission operational programmes like Copernicus but also novel developments will further enhance our capabilities to monitor soil moisture, and they will ensure continuity of multi-scale soil moisture measurements from agricultural to climate scales. At the same time, research has put a new focus on establishing rigorous guidelines for the installation, calibration, operation, maintenance, and use of in situ soil moisture measurements that are informed by metrological practices, as well as on the development of advanced quality control procedures for an ever growing suit of global in situ soil moisture measurement networks to obtain so-called fiducial reference measurements (FRMs) for soil moisture.
Over the last decade, efficient machine learning-based observation operators have been advocated to assimilate remotely sensed soil moisture observations, such as neural networks and gradient boosting trees.
Root-zone soil moisture (RZSM) represents water available for plant uptake and evaporation, making it a key variable for agricultural applications and drought monitoring. While satellite observations are restricted to near-surface soil moisture, modeling approaches exist that can approximate soil moisture conditions in deeper layers. The validation of such deeper-layer products requires reliable ground reference measurements at corresponding depths. However, existing soil moisture monitoring networks, such as those available from the International Soil Moisture Network (ISMN), are highly inconsistent in sensor depth placement and vertical coverage.
In this study, we first investigate the spatial and vertical distribution of in situ sensors available within the ISMN. We then propose a best practice method for identifying Fiducial Reference Measurements (FRMs) for RZSM dynamics, which can be used to validate RZSM products for any assumed soil layer depth. Finally, we compare these RZSM FRMs against existing satellite-derived and modeled RZSM products from the ESA Climate Change Initiative Soil Moisture (CCI SM) dataset and the GLDAS-NOAH land surface model.
Results reveal substantial gaps and inconsistencies in spatial coverage and sampling depth among ISMN networks, with a bias towards agriculturally dominated regions in temperate and cold climate zones, where most stations measure only within the top 10 cm. Out of almost 3000 ISMN stations, a small subset of less than 300 FRM-labeled stations provides sufficiently consistent and deep vertical sampling within the root-zone , assumed as top 1 m soil layer. Those stations are selected to determine an optimal vertical sampling design for the reliable monitoring of specific soil layers. RZSM FRMs are derived for four different soil layers (0-0.1 m, 0.1-0.4 m, 0.4-1 m, and 0-1 m) and used to validate ESA CCI SM and GLDAS-NOAH RZSM estimates for the same layers.
How to cite: Ilina, D., Lems, J., Gruber, A., Dorigo, W., and Crapolicchio, R.: Identifying reliable in situ reference measurements for satellite root-zone soil moisture validation from the International Soil Moisture Network, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-17805, https://doi.org/10.5194/egusphere-egu26-17805, 2026.
The validation of remotely sensed Soil Moisture (SM) products against ground reference data is strongly affected by the spatial scale mismatch between the large satellite footprints and point-scale in situ measurements, exacerbated by differences in station density across in situ monitoring networks. A common but inconsistently applied practice is to average multiple in situ sensors falling within a single satellite grid cell prior to validation. However, there is currently no clear consensus on whether spatial averaging provides a more reliable reference for satellite validation.
In this study, we systematically assess the impact of spatially averaging in situ soil moisture measurements from different sensors of the International Soil Moisture Network (ISMN) on the validation of soil moisture products from the ESA Climate Change Initiative (CCI). The ESA CCI SM product is provided on a 0.25° grid (approximately 25 × 25 km). More than 20% of the CCI grid cells, with in situ stations on them, contain two or more in situ sensors, with an average of 9 sensors per such grid cell. Averaging these sensors within single grid cells may provide a more reliable proxy for grid cell average soil moisture dynamics, but only if their measurements are mutually consistent. Alternatively, satellite products may be compared against each sensor individually, but this causes grid cells that contain multiple sensors to be disproportionately represented in validation summary statistics, potentially biasing validation metrics.
We therefore examine the implications of different spatial averaging choices to answer the question of when and how in situ measurements from dense networks should be averaged for the validation of satellite products.
Our results suggest that, in most cases, averaging measurements from multiple, spatially distributed sensors yields more reliable reference time series that improve in situ-satellite comparison metrics. However, we also see a considerable number of cases where averaging sensors reduces the reliability of the time series, most commonly when averaging measurements from different sensor types. Our findings highlight the importance of methodological consistency and provide guidance for the validation of current and future satellite soil moisture products.
How to cite: Lems, J., Dorigo, W., and Gruber, A.: To Average or Not to Average? Using ground reference networks to validate satellite soil moisture products , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-3899, https://doi.org/10.5194/egusphere-egu26-3899, 2026.
The advent of Cosmic Ray Neutron Sensing (CRNS) stations has represented a significant advancement in the field of soil moisture (SM) retrieval. With footprints of up to 300 m and measurements referring to soil depths of up to 50 cm, these stations can effectively contribute to improving the evaluation of satellite-derived and modelled SM products.
Recently, a CRNS station implemented by Finapp S.r.l. was installed in the peri-urban area of Tito (Contrada Carlone – 40.573425 N, 15.676042 E), located in the Basilicata region of southern Italy. The monitoring site, characterized by complex geomorphological conditions, features an integrated setup including time-lapse Electrical Resistivity Tomography (ERT) system, an array of hydrological sensors (tensiometers, soil moisture sensors, piezometers), and meteorological sensors (temperature, humidity, wind speed, and solar radiation). This open-air monitoring laboratory, supported by advanced methodologies for data integration, enables a multidisciplinary and multiscale approach to investigate SM variability and, more broadly, provides insights into the hydrogeological risks affecting the site. The laboratory was established within the framework of the “ITINERIS” project (PNRR M4C2 Inv.3.1 IR), funded by the EU’s Next Generation program.
Focusing on the CRNS station, SM data acquired since July 2025 have already been compared with different satellite-based SM products, such as the weekly Copernicus Surface Soil Moisture (SSM) derived from Sentinel-1 SAR data and the daily Soil Water Index (SWI) obtained from ASCAT (Advanced Scatterometer) acquisitions. Additional datasets and products acquired at different temporal and spatial scales, as well as based on diverse technologies (active, passive, and merged), will be considered in future analysis. Preliminary results are promising and highlight the strong potential of the laboratory to produce accurate SM measurements. These measurements will be scaled up to the regional level within the framework of the “Space It Up” project, funded by the Italian Space Agency and the Ministry of University and Research (contract No. 2024-5-E.0 – CUP I53D24000060005), to better investigate the impacts of climate change across the entire region. SM variability as well as spatiotemporal anomalies will be analyzed.
How to cite: Lacava, T., Albano, R., Calamita, G., Martino, L., Onorati, B., Satriani, A., and Perrone, A.: Evaluating soil moisture variability in complex environments through multiscale measurements: a multidisciplinary framework for ground-to-satellite data integration, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-17579, https://doi.org/10.5194/egusphere-egu26-17579, 2026.
Root-Zone Soil Moisture (RZSM) is critical for understanding hydrological processes, predicting droughts, and improving agricultural yield forecasts. As RZSM cannot be directly measured by satellites, a simple Exponential Filter (EF) model is applied to the ESA Climate Change Initiative (CCI) Soil Moisture (SM) product to derive a global satellite-based RZSM product. The EF model uses a single parameter, T, to smooths and temporally delay the surface SM signal to estimate RZSM at three depth layers (0–10 cm, 10–40 cm, and 40–100 cm). Previously, the T parameter was defined with vertical variability only (one value per depth layer), and was otherwise assumed to be globally constant. The T parameter is calibrated for each RZSM layer based on in-situ observations of the International Soil Moisture Network (ISMN). Consequently, the resulting RZSM product is strongly influenced by the in-situ data used for calibration. Given that ISMN stations are unevenly distributed globally, the calibrated T parameter may not be fully representative at the global scale.
To investigate how the choice of ISMN stations influences the optimal T per depth layer, pre-filtered ISMN stations were first characterized according to land cover class (ESA CCI), climate class (Köppen-Geiger classification) and soil texture (USDA soil triangle). Based on these characteristics, station selection methods were defined to better match the global distributions of the characterization variables. The resulting T values were then compared to those without deliberate station selection.
This method did not result in major changes to the T parameter across the different station selection methods, demonstrating the stability of the EF model. However, the effectiveness of station selection is constrained by the spatial coverage of the ISMN, with some land cover, climate and soil texture classes represented by few or no stations. This limited coverage leads to deviations from the corresponding global distributions. Increasing numbers of ISMN stations and improved representativeness of the limiting classes in the future calls for further research.
How to cite: Kroyer, B., Lems, J., and Dorigo, W.: Investigating the global representativeness of the ESA CCI RZSM product, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-10162, https://doi.org/10.5194/egusphere-egu26-10162, 2026.
Passive microwave soil moisture (SM) retrieval relies on an accurate representation of soil complex dielectric permittivity, yet the uncertainty introduced by dielectric model selection remains insufficiently quantified. In this study, we evaluated five widely used mineral soil dielectric models and two organic soil dielectric models, and assessed how errors and uncertainties from soil dielectric models propagated into SM retrievals using the single channel algorithm (SCA) based on L-band Soil Moisture Active Passive (SMAP) data. We revealed a substantial inter-model disagreement of retrieved SM with a global mean spread of 0.044 m3/m3. The largest divergence occurred in the tropics and northern high latitudes, where mean RMSE value exceeded 0.10 m3/m3. In generally, organic soil models outperformed mineral soil models, yielding significantly higher R (0.66 vs 0.64) and lower ubRMSE (0.068 m3/m3 vs 0.069 m3/m3) values. Among all models, the Mironov 2019 model that accounts for soil organic carbon (SOC) effect exhibited the best performance with a mean R value of 0.66 and ubRMSE value of 0.07 m3/m3. We further demonstrated that soil dielectric model choices overall contributed 27.6% of SM retrieval error, especially under high SOC conditions. Finally, we derived a global map of optimal dielectric model using triple collocation analysis, and showed that the R and ubRMSE metrics could be further improved by 0.04 and 0.006 m3/m3. compared with the SMAP SM product. Our results highlight the importance of dielectric model specific uncertainty characterization and support regionally adaptive dielectric parameterizations for more accurate L-band SM products.
How to cite: Shangguan, Y., Tong, C., and Shi, Z.: How much does the soil dielectric model matter for passive microwave soil moisture retrieval?, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-15397, https://doi.org/10.5194/egusphere-egu26-15397, 2026.
The boreal forest is the second largest terrestrial biome and owing to its vast extent, is a critical component of the global climate system, functioning as a major carbon reservoir and regulating land–atmosphere water and energy exchanges. However, boreal ecosystems are highly sensitive to climate-driven changes in water availability, exacerbating drought stress, wildfire risk, and widespread, drought-induced tree mortality, demonstrating the need for improved characterization of soil and plant water dynamics. More specifically, soil moisture is a fundamental control on boreal forest productivity and disturbance dynamics, governing water availability for transpiration, photosynthesis, and internal plant water storage. While microwave remote sensing instruments provide valuable large-scale soil moisture observations, their interpretation in forested environments remains challenging due to the combined influence of soil and vegetation water on the microwave signal and a lack of species-specific validation data. In particular, the contribution of internal plant water storage to microwave observations is poorly constrained in boreal ecosystems.
In this study we examined coupled soil–plant water dynamics in a mixed boreal forest in central Saskatchewan as part of the SMAPVEX22-Boreal field campaign. Hourly measurements of real dielectric constant (RDC) were collected from near-surface organic soil (5 cm), mineral soil, and tree xylem across 27 forested sites during the 2022 growing season. Measurements focused on three dominant boreal species representing contrasting functional types: jack pine (Pinus banksiana), black spruce (Picea mariana), and trembling aspen (Populus tremuloides). To independently characterize internal plant water storage, destructive vegetation sampling was conducted to quantify gravimetric water content in primary branches, secondary branches (including foliage), and whole branches. Soil water potential was estimated using texture-based parameterizations to better represent plant-available water.
Time series analyses revealed a strong and consistent relationship between soil RDC and tree xylem RDC, indicating tightly coupled soil–plant water dynamics throughout the growing season. Soil moisture exhibited greater short-term variability than tree RDC, while xylem RDC showed a gradual seasonal drydown and became less responsive to individual precipitation events as summer progressed. Pronounced species-specific differences were observed: trembling aspen exhibited significantly higher and more variable xylem RDC than the conifer species, whereas black spruce sites were characterized by persistently wetter soils associated with thicker organic layers. Lag-correlation analysis showed virtually no delay between soil moisture and tree RDC at an hourly timescale for jack pine and black spruce, and a short (~1 hour) lag for aspen, with the strongest correlations (~ 0.80) occurring in the mineral soil layer, suggesting the influence of relatively shallow rooting depth in water access strategies.
These results reflect contrasting species-specific hydraulic strategies, with jack pine adapted to drier conditions and black spruce and aspen maintaining greater internal water storage. The strong, near-synchronous coupling between soil and plant water at hourly timescales suggests limited temporal separation between soil wetting and vegetation uptake in boreal forests, constraining their use as a signal-separation mechanism in microwave remote sensing. Thus, species-level hydraulic differences should be explicitly considered in soil moisture retrieval and validation frameworks.
How to cite: Wicks, K., Roy, A., Cosh, M., and Berg, A.: Investigating Soil and Plant Xylem Dielectric Relationships in Boreal Forest, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-15414, https://doi.org/10.5194/egusphere-egu26-15414, 2026.
Soil moisture plays a crucial role in both ecosystems and human activities. It serves as a primary water source for plants and soil microorganisms. People use soil moisture information in irrigation strategies, predicting soil-borne plant diseases, and more. Soil moisture memory (SMM) refers to the soil’s ability to reflect signals of perturbations caused by anomalies such as intense storms or prolonged dry periods, over time. Its significance stems from its direct link to soil moisture dynamics, as understanding SMM characteristics helps predict soil moisture behavior over time. Many studies have investigated SMM, employing various metrics for its measurement, for example, the e-folding autocorrelation timescale. The time scale of SMM ranges from a couple of days to several months, but its duration and seasonality vary by location, depending on soil types, local hydrological settings, climatic regimes, and vegetation ecosystems. This study introduces a novel SMM metric based on the differences between the conditional and marginal distributions of soil moisture. First, a soil moisture simulation model is calibrated using modified ERA5 Potential Evapotranspiration (PET) and NASA’s GPM IMERG precipitation data as inputs, with SMAP soil moisture data as target values on a daily scale. Next, 2,000 years of daily precipitation and minimum/maximum temperature are generated using the stochastic weather generator WeaGETS, driven by GPM IMERG and CPC temperature data. PET is then estimated from the simulated temperature using the temperature-based Hargreaves-Samani equation. Using the generated 2,000-year input data, daily soil moisture is simulated. The simulation bias is then corrected using the CDF-matching method. With the bias-corrected daily soil moisture, the joint, marginal, and conditional probability distributions of soil moisture are analyzed at multiple lead times (3, 7, 14, 21 days) across four seasons and two study sites in Iowa and Ukraine. Results show that conditional distributions converge toward marginal distributions within 7-14 days in Iowa and 14-21 days in Ukraine in most seasons, with winter and spring exhibiting the longest SMM time scale for Iowa and Ukraine, respectively. This study shows how the conditional distributions of soil moisture gradually converge to the marginal distributions as lead prediction time increases. The time to convergence, dependent on soils, climate and season is a measure of the memory of soil moisture in the system. The conditional distributions are key to applications like irrigation scheduling.
How to cite: Takahiro, F., Soylu, M., Arnone, E., and Bras, R.: Deriving the conditional distribution of soil moisture and its use in estimating memory in the water-soil system, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-14482, https://doi.org/10.5194/egusphere-egu26-14482, 2026.
Since the launch of ESA’s Sentinel-1 mission, openly accessible soil moisture products at ~1 km spatial sampling have become popular for regional scientific studies and numerous applications in agriculture, on land use/change, and water resource management, among others. However, while community-agreed approaches exist for estimating data quality in the temporal domain - e.g., through comparison with in situ measurement time series - comparable quantitative assessments of spatial performance are still largely lacking. This is due to the limited availability of reference measurements and the lack of methods that can effectively exploit them for spatial evaluation.
Recently, a new Point-Scale-Downsampling (PSD) framework was proposed, which enables the computation of both temporal and spatial performance metrics between satellite observations and in situ point measurements. The framework uses coarse-scale benchmark data to assess relative differences between products across spatial scales.
In this presentation, we show results from a recent intercomparison study of native (Sentinel-1) and downscaled (SMAP, SMOS, ASCAT, ESA CCI) 1 km soil moisture products over Europe. We compute temporal and spatial performance metrics using the PSD framework with respect to reference in situ measurements from the International Soil Moisture Network (ISMN). We place our findings in the context of traditional temporal quality assessments and correlogram-based spatial variability characteristics. We conclude that, while high-resolution products overall outperform coarse-resolution benchmark products in terms of spatial information content, the downscaled products at this stage tend to show better spatio-temporal agreement with the available in situ measurements than native SAR retrievals. Additional reference measurements and novel, qualitative approaches to assess the suitability of satellite soil moisture for specific applications could further improve the understanding and reliability of these data in the future.
This study received funding from the European Space Agency (ESA) "Hyper-resolution Earth observations and land-surface modeling for a better understanding of the water cycle" (4Dhydro) project, with tender reference: ESA AO/1-11298/22/I-EF.
How to cite: Preimesberger, W., Stradiotti, P., Piles, M., Fan, D., Raml, B., Peng, J., and Dorigo, W.: Spatiotemporal Evaluation of Downscaled and Native High-Resolution Satellite Soil Moisture Products, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-19306, https://doi.org/10.5194/egusphere-egu26-19306, 2026.
Quality Assurance for Soil Moisture (QA4SM, available at https://qa4sm.eu/) is a centralized, cloud-based online tool for satellite soil moisture data validation that was launched in 2018. Via an easy-to-use web interface, the platform is designed to simplify the systematic intercomparison of satellite measurements, land surface models and reanalysis fields and validate them against globally available fiducial reference measurements (FRMs) from the International Soil Moisture Network (ISMN; https://ismn.earth). The platform delivers reproducible validation results, grounded in consistent methodology and community-agreed best practices, following requirements set forth by the Global Climate Observing System (GCOS) and the Committee on Earth Observation Satellites (CEOS).
QA4SM implements an extensive array of regularly-updated satellite products from missions including SMOS, SMAP, ASCAT, and Sentinel-1, along with multi-sensor datasets from the Copernicus Climate Change Services (C3S) and ESA Climate Change Initiative (CCI). It incorporates in situ reference data from the ISMN and reanalysis products such as ERA5(-Land) and GLDAS-Noah. Users can upload their own custom datasets in multiple formats for comparison against state-of-the-art reference products. The platform offers extensive customization options for validation, including dataset filtering, spatial and temporal sub-setting, scaling methods, temporal matching, and anomaly calculation. Results can be archived and published with digital object identifiers (DOIs) for traceability and scientific reproducibility.
QA4SM is continuously evolving to keep up with scientific developments and user needs. Several new capabilities are planned for the upcoming Release 4 in March:
- Automated validation of NRT data streams will enable stable, continuous monitoring of data quality.
- Spatial validation functionality will complement the existing temporal validation, addressing use cases that involve datasets with limited temporal coverage as well as high-resolution products.
- A programmatic API will provide command-line access to the platform, enabling seamless integration with automated workflows.
- A new interactive data viewer will facilitate intuitive exploration of validation results.
In this presentation, we demonstrate the functionalities of QA4SM, highlight the features introduced in Release 4, and discuss planned future developments.
QA4SM is developed as part of the European Space Agency’s Fiducial Reference Measurement for Soil Moisture (FRM4SM) project.
How to cite: Unterasinger, F., Aberer, D., Unterholzner, T., Shmeleva, S., Preimesberger, W., Ilina, D., Lems, J., Boresch, A., Bakcsa, Z., Mialon, A., Bousbih, S., Dorigo, W., Gruber, A., Crapolicchio, R., and Scipal, K.: Quality Assurance for Soil Moisture (QA4SM) - A Centralized Soil Moisture Validation and Inter-Comparison Platform, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-14005, https://doi.org/10.5194/egusphere-egu26-14005, 2026.
Soil moisture (SM) is a key variable in land-atmosphere interactions, and numerous efforts aim to produce consistent large-scale SM datasets. Satellite-based retrievals provide valuable complements to physically based approaches, particularly for achieving global coverage. Neural networks (NNs) have demonstrated strong potential for improving SM retrieval accuracy in recent years. This study benchmarks daily SM retrievals from Advanced SCATterometer (ASCAT) observations using multiple NN-based architectures, with varying degrees of localization, a strategy designed to help the models adapt to local conditions. Two model families are evaluated: multilayer perceptions (MLPs) and convolutional neural networks (CNNs). We examine configurations that incorporate physical variable augmentation, geographic coordinate inputs, and explicitly localized designs (pixel-scale MLP and locally-connected CNN) to assess the sensitivity of the model accuracy to input nature and localization strength. In non-localized settings, CNNs consistently yield higher spatial and temporal correlations, reflecting their ability to learn spatial hierarchies and local patterns. In strongly localized designs, the pixel-scale MLP and locally-connected CNN achieve very high overall correlations with substantially reduced local bias, highlighting the value of localized learning for capturing fine‑scale SM variability. In addition to providing improved daily SM estimates, our CNN-based retrieval can also capture intraday variability. This capability is particularly evident during intense precipitation events, offering new perspectives into short-term hydrological dynamics. Looking ahead, future efforts should focus on integrating complementary satellite measurements from other sensors (SMOS, SMAP, AMSR, CIMR) to further improve retrieval accuracy, robustness, and temporal resolution.
How to cite: Dinh, L. A., Aires, F., and Pellet, V.: Advancing ASCAT soil moisture retrievals: benchmarking neural-network models and exploring intraday estimation potential, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-17249, https://doi.org/10.5194/egusphere-egu26-17249, 2026.
Soil moisture plays a critical role in regulating the water and energy flux between land and atmosphere, influencing hydrological processes, vegetation dynamics, and climate feedbacks. The traditional data-driven approaches such as correlation and regression have been widely used for understanding the relationships between soil moisture and its controlling drivers, but these methods are limited in capturing complex nonlinear interactions between the multiple drivers, and not much robust enough to quantify the dominant key drivers that control soil moisture patterns. Machine learning algorithms can overcome this limitation by learning the complex non-linear interaction directly from observational datasets. However, the black box nature of the model limits its explainability, which can be improved by the integration of explainable artificial intelligence (XAI). The present study aims to understand the climatic and vegetation drivers controlling the seasonal variability of soil moisture patterns at a remote sensing scale across India, using a random forest modelling framework integrated with XAI. Satellite-derived soil moisture and other hydroclimatic and vegetation drivers were analysed at a large scale (36 km) across seasons. The result shows that the model performs well in capturing the grid-wise temporal variability of soil moisture based on seasons. The XAI based interpretation identifies precipitation as the dominant controlling driver during the monsoon season, covering nearly 70 percent of the areal extent across semi-arid and sub-humid regions. The diurnal land surface temperature and evaporative fraction are identified as the dominant drivers across the arid regions during the non-monsoon seasons. Our findings with the aid of model with XAI integrated explainability techniques, helps in understanding the complex drivers affecting soil moisture patterns across seasons in India, which is essential for improving weather and climate forecasting models and better preparedness for extreme events, including drought.
How to cite: Colin A, A. and Singh, S.: Understanding Seasonal Variability of Soil Moisture Patterns with Explainable Machine Learning, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-18242, https://doi.org/10.5194/egusphere-egu26-18242, 2026.
Soil moisture is critical for understanding hydrological processes, with applications spanning weather forecasting, agricultural management, and flood prediction. However, the fundamental requirement is not merely large data volumes, but gap-free observations with spatiotemporal continuity. Gap-free data is essential to preserve long-term time series characteristics and comprehensively understand hydrological processes. More critically, in climate modeling and extreme event forecasting under climate change, seamless observations are fundamentally required—even sporadic gaps can fundamentally alter predictions.
The SMAP L3 AM (descending) product is widely recognized as a representative remote sensing product at 36 km resolution. Even accounting for the 2-3 day orbital revisit cycle, temporal coverage is severely limited: land grids average 62% temporal gaps, rising to 71% after quality control, with numerous regions experiencing over 80% data unavailability. This extreme temporal sparsity, driven by orbital constraints, RFI, and retrieval algorithm limitations, fundamentally limits applications requiring continuous observations.
Various methods have been developed to interpolate these temporal gaps. Approaches include data fusion using ground and satellite observations, or employing data assimilation through hydrological modeling to fill gaps. Recently, methods using deep learning models—which demonstrate high predictive performance—have been extensively researched for gap-filling. However, these methods suffer from critical limitations: (i) many existing interpolation methods distort data by ignoring the inherent characteristics of the original observations; (ii) when using external data sources, uncertainties such as sensor inconsistencies, temporal misalignments, and simultaneous missing data issues are introduced; and (iii) deep learning often fails to reflect underlying physical processes, producing unexplainable black-box results.
To address these limitations, this study proposes a hybrid framework that combines a Water Balance Model (WBM) with MDN-ConvLSTM (Mixture Density Network-Convolutional Long Short-Term Memory). The framework employs residual learning, where the WBM provides physically consistent baseline predictions and the MDN-ConvLSTM learns systematic differences between model estimates and SMAP observations. The MDN captures residual characteristics inherent to SMAP, enabling reconstruction using only SMAP observations. This design maintains physical interpretability while leveraging deep learning for complex residual patterns, marking the first MDN application to satellite soil moisture reconstruction.
The study compares two learning strategies and three spatial scales (16×16, 32×32, 64×64 pixels): Closed-Loop (using actual SMAP when available) and Open-Loop (recursively using own predictions), evaluating model stability, long-term gap response, and feasibility as proxy observations. Validation against original SMAP demonstrates successful reconstruction of temporally seamless SMAP-like data (ubRMSE = 0.029 m³/m³, R = 0.726, KGE = 0.679). Notably, Open-Loop achieved comparable performance to Closed-Loop, demonstrating robustness with limited data and potential as reliable proxy observations during satellite outages. This physics-guided residual learning approach establishes a novel paradigm combining physics-based water balance modeling with data-driven residual learning using only the target satellite product.
(This work was supported by the National Research Foundation of Korea(NRF) grant funded by the Korea government(MSIT) (RS-2025-23523230))
How to cite: Hong, C. and Kim, S.: Reconstruction of SMAP L3 Soil Moisture Data Using a Physical-Deep Learning Based Hybrid Approach, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-2704, https://doi.org/10.5194/egusphere-egu26-2704, 2026.
Estimation of surface soil moisture (SSM) is fundamental for hydrological monitoring and agricultural management, particularly in regions affected by recurrent droughts such as central–northeastern Argentina. Data assimilation techniques provide a robust approach to estimating SSM variability by integrating numerical soil models—i.e., land surface models—with observations from satellite-based remote sensors. Satellite missions based on L-band microwave sensors, such as NASA’s SMAP and ESA’s SMOS, provide global SSM retrievals, yet these observations often exhibit systematic biases arising from instrument noise, indirect measurements (i.e., observational operator), and temporal and spatial heterogeneities. Within data assimilation techniques, the ensemble Kalman filter (EnKF) has been widely employed for SSM estimation. This algorithm assumes unbiased observations; thus, bias correction becomes necessary to ensure optimal assimilation. In this study, we evaluate three off-line observational bias-correction techniques within a land data assimilation framework based on the Noah-MP v4.0.1 land surface model and an EnKF. The assessment focuses on the 2022 dry season over the endorheic Pampas region. We introduce a bias correction approach to mitigate sampling errors in cumulative distribution function (CDF) matching: (i) the climatological statistics are computed using homogeneous soil texture pixels within the bin, and (ii) a 45-day moving temporal sampling window is used to give a smoother evolution of the CDF. During dry periods, we empirically demonstrate that soil moisture probability density functions are statistically distinguishable across different soil textures within the bin. Furthermore the monthly-fixed statistics exhibited jumps during the dry season. This approach is compared with the standard CDF matching and the normal deviate scaling. These three off-line bias correction techniques are applied to correct SMAP and SMOS satellite retrievals prior to data assimilation. We show that this improved statistical sampling for CDF matching has a non-negligible impact on the SSM estimates resulting from EnKF, particularly during dry periods. The corrected sampling of CDF matching shows better alignment and stronger correlation with the time series of the independent in-situ soil moisture measurements. Overall, the study emphasizes the need for context-aware bias-correction techniques to enhance SSM data assimilation in regions with strong seasonal precipitation variations. Moreover, SSM estimations influence deeper model layers through vertical propagation of the information. These results motivate future work exploring how surface corrections might lead to enhanced subsurface estimates.
How to cite: Obregon, F. M., Pulido, M., Lucini, M. M., Muller, O., and Ruscica, R.: Assessment of bias correction techniques for satellite soil moisture data assimilation in the dry season of Southeastern South America, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-8188, https://doi.org/10.5194/egusphere-egu26-8188, 2026.
In this study, we assess the performance of a recently proposed reflectivity index (IR), derived from Fresnel coefficients and implemented through a change-detection approach for soil moisture retrieval (Zribi et al., 2020). Initially developed for Sentinel-1 VV polarization, the IR is theoretically linked to surface dielectric properties and therefore to surface soil moisture (SSM). We extend its evaluation by integrating dual-frequency SAR observations from Sentinel-1 (VV) and EOS-04 (HH) acquired over a three-year period across three monitoring sites in the Berambadi watershed, India. At each site, multi-depth HydraProbe sensors were deployed to provide high-quality in-situ SSM measurements for validation. All satellite acquisitions from both missions were standardized to comparable spatial resolution (30 m) and matched for similar incidence angles to ensure consistency. Results show that the IR exhibits stronger sensitivity to SSM in VV polarization than in HH, yielding improved retrieval performance across heterogeneous land surface conditions. Nonetheless, the effectiveness of the IR decreases with increasing soil moisture, with a more pronounced reduction in sensitivity observed for VV polarization. Overall, the findings demonstrate the suitability of the IR for operational soil moisture monitoring and highlight the important role of polarization and moisture regime in controlling its performance.
How to cite: Gupta, V., Zribi, M., Baghdadi, N., and Muddu, S.: Fresnel-Based Reflectivity Index for Surface Soil Moisture Retrieval from Multi-Polarization SAR Observations, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-703, https://doi.org/10.5194/egusphere-egu26-703, 2026.
When retrieving the surface soil moisture over agricultural fields using Synthetic Aperture Radar (SAR), vegetation absorbs and scatters the signal, which then hinders the analysis of the underlying soil [1,2]. One method to circumvent this is by decomposing the radar signal into three components: surface, dihedral, and volume scattering [3,4]. Recent advancements have extended these models into a tensor framework and incorporated spatial information to then invert the geophysical parameters of the models and retrieve soil moisture for a wider range of crop scenarios [5]. The soil moisture is retrieved through numerical optimization of the models' geophysical parameters.
In this work, we compare the information loss when retrieving soil moisture using the tensor-based decomposition between full-polarisation and dual-polarisation inversion. We assess the ambiguity of parameter retrieval for different combinations of dual-polarisation channels and conclude on which set-up of dual-polarisations with VV, VH, and/or HH provides the most constrained and thus most optimal soil moisture retrieval with the tensor decomposition technique.
This work is implemented using the full-pol airborne F-SAR data from DLR and soil moisture retrieval from the inversion is compared with ground measurements taken during the AgriROSE-L campaign around Munich, Germany, in 2025.
[1] I. Hajnsek, E. Pottier and S. R. Cloude, "Inversion of surface parameters from polarimetric SAR," in IEEE Transactions on Geoscience and Remote Sensing, vol. 41, no. 4, pp. 727-744, April 2003, doi: 10.1109/TGRS.2003.810702.
[2] Dipankar Mandal, Vineet Kumar, Debanshu Ratha, Subhadip Dey, Avik Bhattacharya, Juan M. Lopez-Sanchez, Heather McNairn, Yalamanchili S. Rao, Dual polarimetric radar vegetation index for crop growth monitoring using sentinel-1 SAR data, Remote Sensing of Environment, https://doi.org/10.1016/j.rse.2020.111954.
[3] Freeman, Anthony, and Stephen L. Durden. "A three-component scattering model for polarimetric SAR data." IEEE transactions on geoscience and remote sensing 36.3 (2002): 963-973.
[4] Yamaguchi, Yoshio, et al. "Four-component scattering model for polarimetric SAR image decomposition." IEEE Transactions on geoscience and remote sensing 43.8 (2005): 1699-1706
[5] Basargin, N., Alonso-González, A., & Hajnsek, I. “Model-based tensor decompositions for soil moisture estimation.” Proceedings of the European Conference on Synthetic Aperture Radar, EUSAR (2024)
How to cite: Dugué, M., Basargin, N., and Hajnsek, I.: Soil Moisture Retrieval in the Presence of Vegetation Using Dual-polarisation Data, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-1588, https://doi.org/10.5194/egusphere-egu26-1588, 2026.
Monitoring soil moisture is critical for understanding Earth’s climate, hydrological variability, and vegetation dynamics, particularly in remote regions where ground observations are limited. We develop and validate a multi-sensor approach integrating Sentinel-1 radar (2016-2021) and Sentinel-2 optical data (2021-2025) within Google Earth Engine (GEE) to characterize surface soil moisture dynamics across the Falkland Islands. The aim was to evaluate temporal patterns, sensor consistency, and agreement with in-situ measurements. This will provide a continuous nine-year record of soil moisture dynamics to facilitate regional climate change adaptation and mitigation. We used Sentinel-1 synthetic aperture radar (SAR) backscatter to compute a 10-day interval time series of soil moisture index (SMI) through radar backscatter calibration, temporal compositing, and vegetation correction. While Sentinel-1 soil moisture estimates showed limited correlation with in-situ measurements at 20 cm depth (likely due to sensing depth differences) they exhibited moderate to strong correlations with rainfall, supporting the satellite’s ability to capture rainfall-driven hydrological variation. Due to data discontinuities in Sentinel-1 acquisitions after 2021, we used Sentinel-2 imagery to extend the analysis through September 2025. Both datasets were analysed for seasonal and interannual variability and validated against in-situ volumetric soil moisture (VSM) from Temperature Moisture Sensor (TMS) dataloggers installed across representative grassland and peatland habitats. Results reveal coherent seasonal cycles across all major regions of the Falklands, with recurring summer minima and winter maxima corresponding to drought and recharge periods. Overall, both sensors consistently detected the same hydrological patterns, including wet winters and dry summers with interannual variability linked to regional rainfall dynamics rather than spatially distinct behaviours between subregions. The integration of optical and radar observations provides a robust means of monitoring soil-moisture variability in remote environments. This multi-sensor framework supports future data assimilation, drought assessment, and climate-impact studies. Establishing long-term monitoring that integrates multi-sensor approaches is essential to understand the Falklands’ evolving hydrological and ecological trends.
Keywords: Soil moisture; Remote sensing; Sentinel-1; Sentinel-2; Google Earth Engine.
How to cite: Ko, N. T., Baylis, A., Davies, G. M., Barlow, D., and Evans, C.: Multi-sensor monitoring of soil moisture dynamics in subpolar environments using Sentinel-1 and Sentinel-2 data: A case study from the Falkland Islands, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-240, https://doi.org/10.5194/egusphere-egu26-240, 2026.
Accurate sub-daily soil moisture (SM) measurements at high spatial resolution on a global scale are essential for climate monitoring, agricultural management, and hydrological applications. Passive microwave missions including the Soil Moisture Active Passive (SMAP) and Soil Moisture Ocean Salinity (SMOS) satellites provide global SM products; however, their temporal revisit time (2–3 days) and coarse spatial resolution (36–50 km) limit their ability to capture short-term SM dynamics. In recent years, Global Navigation Satellite System Reflectometry (GNSS-R) has emerged as a promising alternative, offering a large number of sub-daily SM observations from low-cost, lightweight satellite platforms. Nevertheless, GNSS-R observations are spatially irregular and contain coverage gaps [1].
Here, we propose a weighted data fusion approach to integrate soil moisture estimates from NASA’s GNSS-R mission (CYGNSS) with SMAP Level-3 (36 km) and SMOS Level-2 (25–40 km) products independently, generating continuous sub-daily SM maps at a regular spatial resolution of 20 km. The weights are functions of the spatiotemporal distance between the output and input grid points, as well as the expected reliability of the input data. The proposed fusion framework aims to fill spatial gaps in CYGNSS-derived SM, improve its retrieval accuracy through the incorporation of passive microwave observations, and enhance the spatiotemporal resolution of SMAP and SMOS products.
The performance of the fusion model is evaluated over the Contiguous United States (CONUS) during the first five months of 2021. Strong spatial agreement is observed between CYGNSS–SMAP fused maps and SMOS products, as well as between CYGNSS–SMOS fused maps and SMAP products, demonstrating the model’s effectiveness in filling CYGNSS data gaps. Compared to CYGNSS-only SM estimates, the fused products show substantial improvements in accuracy. For the CYGNSS–SMOS fusion, the correlation with SMAP increases from approximately 0.64 to 0.80, while the RMSE decreases from about 0.07 to 0.04 m³/m³. Similarly, the CYGNSS–SMAP fusion improves the correlation with SMOS from about 0.43 to 0.58 and reduces the RMSE from approximately 0.10 to 0.06 m³/m³.
To further evaluate the model’s ability to generate sub-daily soil moisture observations, additional validation was performed using soil moisture time series from 11 in-situ stations obtained from the International Soil Moisture Network (ISMN) [2]. The fused products successfully capture the temporal variability observed in the in-situ measurements, with slightly better performance for the CYGNSS–SMOS fusion compared to the CYGNSS–SMAP fusion. Median correlation coefficients of approximately 0.60 and 0.56, and median RMSE values of about 0.076 and 0.083 m³/m³, are obtained for the CYGNSS–SMOS and CYGNSS–SMAP fused products, respectively.
[1] Senyurek, V., Gurbuz, A., Kurum, M., Lei, F., Boyd, D., & Moorhead, R. (2021). Spatial and temporal interpolation of CYGNSS soil moisture estimations. Paper presented at the 2021 IEEE International Geoscience and Remote Sensing Symposium IGARSS.
[2] Dorigo, W., Himmelbauer, I., Aberer, D., Schremmer, L., Petrakovic, I., Zappa, L., et al. (2021). The International Soil Moisture Network: serving Earth system science for over a decade. Hydrology and Earth System Sciences Discussions, 2021, 1-83.
How to cite: Rahmani, M., Di Simone, A., Di Martino, G., Iodice, A., and Riccio, D.: A Data Fusion Framework for Sub-Daily Soil Moisture Mapping Using CYGNSS, SMAP, and SMOS, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-15802, https://doi.org/10.5194/egusphere-egu26-15802, 2026.
Accurate, field-scale soil moisture estimates are critical for hydrological modeling and agricultural monitoring, yet they remain difficult to obtain from any single satellite system. L-band radiometers such as SMAP provide reliable large-scale soil moisture retrievals, but their coarse spatial resolution limits local applicability. In contrast, C-band SAR observations from Sentinel-1 offer fine spatial detail, though their sensitivity to surface roughness and vegetation requires careful calibration.
We develop a multi-sensor downscaling framework that combines SMAP morning soil moisture with ascending and descending Sentinel-1 VV/VH backscatter. To better represent soil moisture dynamics, historical VV backscatter minima and maxima are used to derive a Soil Moisture Index (SMI), alongside NDVI to account for vegetation. These variables are complemented by the Antecedent Precipitation Index (API) and evapotranspiration to consider surface water fluxes, as well as topographic information from a digital elevation model.
Model robustness is evaluated using a strict temporal split: data from 2020–2024 are used for training, while 2025 is reserved as an independent test year. Three non-parametric algorithms—Random Forest, XGBoost, and K-Nearest Neighbors—are assessed against in-situ measurements from the International Soil Moisture Network.
Including meteorological information and historical backscatter features leads to consistent performance gains across models. On the independent test set, coefficients of determination exceed 0.5, with XGBoost achieving the lowest RMSE and outperforming both Random Forest and KNN. These results demonstrate the value of combining complementary satellite observations and targeted feature engineering for reliable, high-resolution soil moisture mapping.
How to cite: Ma, J., Got, J.-B., Pan, Y., Tabibi, S., Craeye, C., and Lambot, S.: Synergistic Use of Sentinel-1, SMAP, and Ancillary Data for High-Resolution Soil Moisture Mapping, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-20060, https://doi.org/10.5194/egusphere-egu26-20060, 2026.
Seasonal ground deformations are caused by climatic and geophysical factors, including annual temperature fluctuations, elastic lithosphere response, and variations in soil moisture (SM) and groundwater levels. In particular, expansive clayey soils undergo volumetric changes depending on their moisture content, swelling with high SM and shrinking during dry periods. These processes can damage buildings, infrastructures, and hamper slope stability.
This study analyzes the impact of SM and land surface temperature (LST) on seasonal ground deformations in the Po Valley, characterized by clay–rich soils and highly vulnerable to climate-change drought. The analysis covers the period 2020–2023 at 500 m resolution, combining downscaled SMAP SM data, European Ground Motion Service (EMGS) deformation data, and MODIS LST.
SM is downscaled from 9 km to 500 m using an Extreme Gradient Boost model, trained on aggregated Sentinel-1, ALOS, and SAOCOM backscatter data and static variables. The model obtains a R²=0.80 and RMSE=0.012 m³/m³ on the test set, while validation against in-situ measurements shows improved correlation (0.6 vs 0.5) and reduced relative error (6% vs 10%) compared to the original SMAP product.
EGMS displacement time series are averaged within the 500 m cells, and seasonal deformations are extracted using a Loess method and analyzed using correlation and lagged correlation approaches. 33.5% of the samples show seasonal amplitudes greater than 2.5 mm, consistent with swelling–shrinking effects. For about half of the dataset, Spearman correlations with LST are above 0.6, while for SM are weaker (0.3). Time-lag analysis revealed that SM effects peak near zero lag but can persist up to 30 days due to groundwater influence, whereas LST effects are mostly instantaneous or exhibit lags up to 15 days.
Multivariate regression analysis quantifies the independent contributions of SM and LST: 16% of samples has R²≥0.8, indicating that these drivers explain most of the seasonal variability, 38% has 0.5≤R²<0.8, and the 46% shows R²<0.5, suggesting that other factors may be dominant. SM-driven deformations are located in valley areas, while LST-dominated signals are prevalent in mountainous zones.
These results demonstrate the effectiveness of the downscaling approach in improving SM estimation and show that SM and LST jointly explain most seasonal deformations in over half of the analyzed samples. However, SM-driven deformation is detectable in a limited number of samples, and can be masked by thermal expansion. Future work should integrate groundwater and geological data and exclude scatterers with high temperature sensitivity to better isolate SM-induced deformations.
This work was supported by the Università di Modena e Reggio Emilia – Fondazione di Modena Project “Ensembling SATellite monitoring and BIM methods in the SAFety assEssment of road infrastructure (SATSAFE)”, FAR 2024 - Bando per il finanziamento di progetti di ricerca interdisciplinari.
How to cite: Brunelli, B., Grassi, F., and Mancini, F.: Investigating the Climatic Drivers of Seasonal Ground Motion through Multisensor SAR Soil Moisture, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-5764, https://doi.org/10.5194/egusphere-egu26-5764, 2026.
Groundwater is the primary source of accessible freshwater globally, playing a vital role in food security, water supply, and industrial activities. However, excessive abstractions exacerbated by climate change and population growth scenarios threaten long-term availability of groundwater resources. In this context, Earth Observation technologies offer innovative monitoring solutions. Relying on the preliminary findings by Dari et al. (2025) through the SM-Inversion algorithm, this PhD project aims to expand the framework of monitoring groundwater dynamics through remotely sensed soil moisture. The assessment of the extent at which data resolution can be an issue, the upscaling across scale (regional, country, global), and the exploration of synergies with gravimetry missions are among the long-term objectives of the project. In fact, SM-Inversion method enables the estimation of aquifer recharge rates by integrating satellite soil moisture data at various spatial resolutions, in order to evaluate the accuracy of estimates across scales. Preliminary results obtained over eleven aquifers located in the Murray-Darling Basin (Australia) are presented here. Five remote sensing soil moisture products are evaluated: ASCAT (Advanced SCATterometer), CCI (Climate Change Initiative) Combined, CCI Passive, SMAP (Soil Moisture Active Passive), and SMOS (Soil Moisture and Ocean Salinity). Potential evaporation rates from GLEAM (Global Land Evaporation Amsterdam Model) and precipitation from ERA5 (European ReAnalysis-v5) are also used in the proposed aquifer-scale analysis.
References:
Dari, J., Filippucci, P., Brocca, L., Quast, R., Vreugdenhil, M., Miralles, D., Morbidelli, R., Saltalippi, C., and Flammini, A.: A novel approach for estimating groundwater recharge leveraging high-resolution satellite soil moisture, Journal of Hydrology, 652, 132678, https://doi.org/10.1016/j.jhydrol.2025.132678, 2025.
How to cite: Blois, M. R., Dari, J., Flammini, A., Morbidelli, R., and Brocca, L.: Monitoring groundwater dynamics across scales through satellite soil moisture: preliminary results from a novel PhD project, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-5483, https://doi.org/10.5194/egusphere-egu26-5483, 2026.
The European Space Agency (ESA) developed the Climate Change Initiative (CCI) programme to integrate observations from successive active and passive satellite missions and to generate the first multi-decadal global soil moisture data set. The Land Parameter Retrieval Model (LPRM) has been widely used to derive soil moisture from passive microwave observations. Traditionally, LPRM employs a single incidence angle for soil moisture retrieval and has been successfully applied across multiple passive microwave sensors. However, unlike other passive sensors used within the CCI programme, the Soil Moisture and Ocean Salinity (SMOS) mission measures brightness temperatures over a wide range of incidence angles, from 0° to 65°. However, the existing algorithm relies on single incidence angle observations; therefore, there is a need to exploit the multi-angular information provided by SMOS. This study focuses on advancing and adapting the soil moisture retrieval algorithm to incorporate multi-incidence angle observations while maintaining compatibility with other sensors, thereby enabling the production of a consistent long-term climatological soil moisture data record. The LPRM cost function is modified to integrate brightness temperature information acquired at dual polarizations and multiple incidence angles. This enhancement is expected to improve the operational retrieval algorithm and contribute to the generation of a more reliable multi-decadal soil moisture data record.
How to cite: Prajapati, R., Duchemin, D., Mialon, A., Moldenhauer, C., Gabor D Kovacs, D., de Jeu, R., Arnoud Dorigo, W., and Rodriguez-Fernandez, N.: Advancing Multi-Angular SMOS Retrievals within Land Parameter Retrieval Model for the Climate Change Initiative Programme, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-18843, https://doi.org/10.5194/egusphere-egu26-18843, 2026.
Soil moisture is a critical variable for hydrological and agricultural processes, yet its accurate characterization remains challenging due to its high spatial and temporal variability. Passive microwave-based soil moisture observations are routinely available from satellite platforms, but their spatial resolution limits their applicability for many field-scale applications, creating a need for intermediate-scale solutions. Recent advances in unmanned aerial vehicle (UAV) platforms provide new opportunities to observe soil moisture at relatively high spatial resolution offering a promising approach to bridge this spatial gap.
This study evaluates a new UAV-mounted L-band (1.4 GHz) polarimeter developed by Skaha Remote Sensing for monitoring soil moisture at high spatial resolutions. Weekly UAV flights were carried out from mid-April to early November 2025 over a long-term crop rotation and tillage experiment at the University of Guelph’s Elora Research Station. Over these plots, the long-term management practices have resulted in measurable differences in soil structure and hydrological properties across experimental plots. Passive microwave L-band brightness temperature observations were collected across 56 plots representing several crop types (soybeans, corn, winter wheat, barley, and alfalfa) and tillage treatments. The brightness temperatures were compared to in situ soil moisture measurements obtained from each of the plots.
Over the growing season, strong, statistically significant correlation relationships were observed between the brightness temperatures and measured soil moisture (r -0.38 – r -0.9). These relationships suggest that the UAV-based L-band polarimeter can measure moisture variability at relatively high spatial resolutions (~7m; dependent on flying height) and during all time periods of the growing season analyzed. Comparisons by crop type and phenological stage suggests the performance of the sensor for soil moisture retrieval varied by crop type and phenological stage. Crops such as corn and soybeans with relatively larger row spacing generally showed stronger (negative) correlations between the measured brightness temperatures and soil moisture (r < -0.7) while denser crops with more ground coverage (alfalfa, barley) exhibited weaker correlations (r between -0.6 and -0.38). Overall, the results demonstrate the potential of UAV-mounted L-band radiometry to potentially bridge the scale gap between point-based in situ measurements and coarse-resolution satellite observations.
How to cite: Berg, A., Rusk, H., McLaren, A., and Wolleben, M.: Influence of agricultural crop type on UAV-based passive microwave soil moisture retrieval at L-band, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-15908, https://doi.org/10.5194/egusphere-egu26-15908, 2026.
More than 80% of the world’s farms are small, with a farm size of less than 2 hectares. Under such highly fragmented agricultural systems, smallholder farmers require accurate, field-scale predictions of soil moisture and irrigation requirements that can capture local variability. Vegetation heterogeneity plays a crucial role in shaping soil moisture variability across various spatial scales. At the field scale, differences in crop types, cropping patterns, and management practices can substantially alter soil-water dynamics.
Traditional Land Surface Models (LSMs) simulate land surface processes with spatial, temporal, and physical consistency. However, their typically coarse spatial resolution (>10 km) fails to resolve the sub-grid heterogeneity relevant for small and marginal farms. While hyper-resolution (<100 m resolution) LSMs have the potential to simulate land surface processes at the field scale, many of their processes, including vegetation dynamics, are heavily parameterized. Previous studies have highlighted the sensitivity of soil moisture to Leaf Area Index (LAI), but the reliance of LSMs on lookup-table LAI parameterization introduces substantial errors in simulating crop phenology, yield, and growing-season length, especially in regions like India with a prevailing fragmented agricultural system. These errors arise from the assumption of uniform parameter values across different climatic regions and crop types, disregarding the scale effects of vegetation and soil heterogeneity. Addressing these limitations requires region-specific parameterization or using satellite-based LAI values in simulating soil moisture to enhance the accuracy of soil moisture predictions and improve the representation of vegetation dynamics in LSMs.
The present study evaluates the benefits of using satellite LAI data in generating a field-scale soil moisture simulation. Towards this, we set up HydroBlocks, a hyper-resolution LSM over Upper Bhima Basin, in India, to simulate 3-hourly 30 m resolution surface (0-5 cm)and root zone (0-30 cm) soil moisture. The Upper Bhima Basin is a sub-basin of the Krishna River, lying predominantly on the leeward side of the Western Ghats. The study area has the majority of its land under croplands and receives relatively low rainfall, making it an ideal location for studying soil moisture variability under varying vegetation conditions. In this study, we made four HydroBlocks simulation experiments with LAI data from different sources: 1) the default lookup table values, 2) monthly climatological LAI data derived from MODIS for each land use land cover class, 3) assimilating MODIS LAI data using the direct insertion technique, and 4) using the dynamic vegetation module. We compared the spatial and temporal dynamics of soil moisture simulation from different experiments. Further, we statistically evaluated both surface and root-zone soil moisture simulations against in situ observations.
How to cite: Vishnu, U. K., Vergopolan, N., Lanka, K., and Jayaluxmi, I.: Quantifying the benefits of incorporating vegetation heterogeneity in farm-scale soil moisture simulations, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-1263, https://doi.org/10.5194/egusphere-egu26-1263, 2026.
Agricultural drought occurs when soil moisture available to crops is inadequate to meet their water requirements, resulting in reduced crop yields and agricultural production. Therefore, it is crucial to accurately monitor agricultural drought for effective irrigation planning and scheduling, water resources management and allocation, and food security protection. Soil moisture is a fundamental variable for estimating agricultural drought indicators; thus, reliable, accurate, continuous, and long-term datasets are essential for effective drought monitoring. L-band radiometry is the most effective passive microwave remote sensing technique for estimating global soil moisture. The ESA Soil Moisture and Ocean Salinity (SMOS) and NASA Soil Moisture Active Passive (SMAP) missions, launched in 2009 and 2015, respectively, were developed specifically to retrieve Surface Soil Moisture (SSM) at L-band (1.4 GHz) for the top 5 cm of soil, with a target accuracy of 0.04 m³ m⁻³. These missions also generate Root Zone Soil Moisture (RZSM) at 0–100 cm depth through assimilation of SSM into land surface models. Agricultural drought can be detected using the Soil Water Deficit Index (SWDI) and Soil Moisture Drought Index (SMDI) derived from SMAP and SMOS satellite-based soil moisture products. The Karkheh River Basin is one of the most important watersheds and agricultural regions in southwestern Iran, with a semi-arid to arid climate, which has experienced frequent droughts in recent years. This study evaluates the efficiency of SMOS and SMAP products for agricultural drought monitoring using the SMDI and SWDI indices over the Karkheh River Basin.
How to cite: Jamei, M., Asadi Oskouei, E., and Jamei, M.: Evaluation of SMAP and SMOS Microwave Satellite Soil Moisture Products for Agricultural Drought Monitoring: Karkheh River Basin Case Study, Iran, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-660, https://doi.org/10.5194/egusphere-egu26-660, 2026.
Understanding crop water status and soil moisture dynamics in heterogeneous agricultural landscapes remains a major challenge for microwave remote sensing, especially when multiple crop types coexist within a single pixel. With the NISAR mission providing fully polarimetric L- and S-band SAR observations, there is a unique opportunity to evaluate its retrieval capability in complex mixed-cropping systems.
In this study, we conduct an intensive field campaign across irrigated and rainfed plots in southern India to assess how NISAR L- and S-band backscatter responds to variations in vegetation water content (VWC) and surface soil moisture (SSM) under heterogeneous conditions. Each satellite-aligned pixel in the study region typically contains 4-5 crop types with distinct canopy structures and rooting characteristics. For selected NISAR acquisition dates, we measure VWC through destructive sampling of each crop species present within the pixel. Concurrently, surface soil moisture is measured using both handheld probes and permanently installed soil moisture sensors deployed across the heterogeneous fields to capture intra-pixel variability.
By combining in-situ VWC, multi-depth soil moisture observations, and crop-wise metadata with co-located NISAR L- and S-band backscatter, we evaluate (i) the sensitivity of each band to mixed vegetation conditions, (ii) the ability to distinguish irrigated vs. rainfed water-use patterns, and (iii) the impact of intra-pixel crop diversity on retrieval accuracy. This work provides one of the first ground-based evaluations of NISAR performance in complex Indian agroecosystems and contributes toward developing improved retrieval approaches for crop water assessment and soil moisture estimation in heterogeneous landscapes.
How to cite: Anjali, P. S., Gupta, V., Judge, J., Dutta, D., Thiyagarajan Logambal, E., Gawai, D., Tikhe, P. V., Chothani, V., Katroth, P. S., Anand, N., Goyal, M., Singh, S., Murugeshan, S., and Muddu, S.: Behaviour of NISAR L and S band Backscatter for Soil Moisture and Crop Water Monitoring in India’s Mixed Cropping Landscapes, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-765, https://doi.org/10.5194/egusphere-egu26-765, 2026.
Records of soil moisture dynamics are the land register of the convolution of footprints of the multiple water fluxes that intervene in the soil. Therefore, the variability of soil moisture changes occurring at a specific location expresses not only the stochastic nature of soil moisture but, more importantly, differences in the statistics of their distribution in space and time, which can be used to classify patterns of soil moisture regime across space and time.
The analysis of hydrological extremes often deals with the scarcity of data (i.e. insufficient extreme events contained in the series, which biases the range of values from reality). Fortunately, earth observation data is increasingly providing reliable distributed datasets whose spatial detail can compensate for the limitation in time length. This is particularly true for soil moisture data whose ground records rarely surpass decades but whose distributed data is achieving resolutions that, despite persistent applicability limits below certain resolutions, can ease the analysis of soil moisture variability.
Accordingly, this study analyses the temporal changes in soil moisture of various types of soil moisture products across all cells of the adopted 5 x 5 Km grid across Europe and Africa, in both the rewetting and drying signs of change, with the aim of finding distinct ranges and frequency-magnitude characteristics along the distribution of soil moisture changes. Special attention is devoted to the soil moisture changes distribution’s upper tail (extreme events like floods (positive change) and flash drought (negative change)) and lower tail (detection limit, product sensitivity).
Three types of soil moisture products are used (remote sensing passive: ESA CCI passive subset; remote sensing active: EUMETSAT ASCAT; and model-based: LISFLOOD model integrated in the European Copernicus Emergency Monitoring System (CEMS)) to evaluate their ability to show consistency across ranges of the distribution of soil moisture changes so that it can ensure the efficacy of monitoring systems integrating earth observation and modelling data.
The analyses to extremes applied to soil moisture change data show results consistent and complementary to those published for rainfall and runoff generation, identify the areas where soil moisture mediation of the water cycle is more relevant in relation to hydrological regime classification and map thresholds of impactful events.
But more importantly, results reveal notable disparity in the estimation of the relevance of an event (magnitude (expressed as intensity of change) and occurrence (expressed in frequency or return period), particularly in the most impactful cases of extreme events, entailing gaps between the dynamics detected by current soil moisture products and their true dynamics. Such disparities among datasets must be prevented from propagating to monitoring systems.
Therefore, the approach provides insights for the continuous upgrading of the products’ consistency (i.e. remote sensing and model-based datasets), while encourages adopting metrics of distribution consistency in the early warning system pipelines, particularly across impactful ranges of soil moisture value change, to improve the monitoring accuracy according to the regional characteristics, with subsequent benefits to the efficacy of responses to the impacts.
How to cite: Gaona, J., Brocca, L., Kumar, V., Filipucci, P., Liaqat, M. U., and Serbouti, I.: Characterisation of extreme events through satellite and model-based soil moisture products over Europe and Africa , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-11438, https://doi.org/10.5194/egusphere-egu26-11438, 2026.
Surface soil moisture exhibits strong day-to-day variability in response to antecedent rainfall. On scales beyond a few kilometres, soil moisture can generate daytime mesoscale circulations via sensible heat flux gradients which may influence the development of new convective rain events, creating a feedback loop. Numerical model simulations struggle to capture such feedbacks, due to shortcomings in the representations of convection, the water stress control on evapotranspiration, and uncertainties in the soil moisture itself. On the other hand, recent analysis of satellite observations has illustrated the importance of such feedbacks on storm initiation across Sub-Saharan Africa (Taylor et al, 2026). Here we examine how well observed mesoscale soil moisture structures across Africa are captured in medium resolution (0.1 degree) products generated within the European Space Agency Climate Change Initiative Soil Moisture project.
We use two approaches to evaluate soil moisture products at the mesoscale. Both are based on simple spatial correlations at the sub-1 degree scale with independent observations of related variables. Firstly, we quantify the spatial consistency between changes in soil moisture over 12 hours (consecutive overpasses) and accumulated precipitation. Interestingly, this analysis highlights the shortcomings of well-used precipitation products (e.g. IMERG) at this scale compared to a recent deep learning-based product (Rain Over Africa; Amell et al 2025). Secondly, we compare patterns of anomalous soil moisture with daytime Land Surface Temperature (LST) from Thermal Infrared imagery. We find that the active microwave-based ASCAT soil moisture product (with effective spatial resolution ~15km) outperforms passive microwave-based products (SMAP, SMOS, AMSR-2) and model-based ERA5-Land data in this exercise, with consistently stronger negative soil moisture-LST correlations . Finally, we use medium resolution soil moisture to demonstrate an impact on convective activity.
Amell, A., Hee, L., Pfreundschuh, S., & Eriksson, P. (2025). Probabilistic Near-Real-Time Retrievals of Rain Over Africa Using Deep Learning. Journal of Geophysical Research: Atmospheres. https://doi.org/https://doi.org/10.1029/2025JD044595
Taylor, C. M., Klein, C., Barton, E. J., Hahn, S., & Wagner, W. (2026). Wind shear enhances soil moisture influence on rapid thunderstorm growth. Nature. https://www.nature.com/articles/s41586-025-10045-7
How to cite: Taylor, C., Ahmad, J., Harris, B., Kovács, D., Moldenhauer, C., Dorigo, W., and Amell, A.: Evaluation and application of medium resolution soil moisture data over Sub-Saharan Africa, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-10506, https://doi.org/10.5194/egusphere-egu26-10506, 2026.
The 2022 drought in Western Europe exposed major shortcomings in Europe's strategies for managing drought risk. In France, the drought was characterised by an unprecedented number of vegetation fires, as well as damage to buildings caused by clay shrink-swell. This has led to a growing awareness of the risks associated with drought and highlighted the need to improve climate risk management capabilities. While land surface models (LSMs) offer continuous temporal and spatial coverage, they may struggle to accurately represent certain processes due to their complexity. As LSMs cannot represent all relevant processes, data assimilation (DA) can be used to update them with observational data. This improves LSMs' capacity to monitor soil and vegetation variables driven by climatic and anthropogenic factors. In this study, we use the interactions between soil, biosphere and atmosphere (ISBA) land surface model within the SURFEX modelling platform, which was developed by Météo-France, to monitor land surface variables and characterise droughts. We analyse leaf area index (LAI) and root-zone soil moisture (RZSM) using sequential data assimilation (DA) and machine learning (ML) techniques in near-real time at a high spatial resolution, combining AROME weather forecast model forecasts with Copernicus Land Monitoring Service (CLMS) LAI observations derived from Sentinel-3. We use a satellite data assimilation system (LDAS, or Land Data Assimilation System) to correct ISBA model simulations by integrating LAI observations. We will present examples of applications related to monitoring natural hazards (e.g. clay shrinkage, wildfires and flash floods) in the context of the Integrated Research Insight into Climate Risks (IRICLIM) project (https://www.pepr-risques.fr/fr/projets-cibles/iriclim-recherche-integree-sur-risques-lies-au-climat).
How to cite: Calvet, J.-C., Baehr, Y., Bonan, B., and Vanderbecken, P.: Monitoring of vegetation and drought in France using the ISBA land surface model and satellite data, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-3292, https://doi.org/10.5194/egusphere-egu26-3292, 2026.
