Paul C. Vermunt

Forest water dynamics can be assessed on large spatio-temporal scales using satellite-based remote sensing. Vegetation optical depth (VOD), which indicates the vegetation's attenuation of microwaves, contains mainly information on the dry biomass, structure, and water content of the vegetation. Water dynamics can be reflected in short-term variations in VOD. Currently, VOD is operationally retrieved using passive microwave sensors, such as AMSR-2, SMAP, and SMOS, or radar-based sensors, such as ASCAT, with a coarse spatial resolution (tens of kilometers), which hinders the understanding of complex landscapes and is a major obstacle for VOD validation using ground sensors. To overcome this shortcoming of spatial resolution, we utilize synthetic aperture radar (SAR) sensors, such as Copernicus Sentinel-1 (S1) C-band (5.504 GHz), which enables a much higher spatial resolution (tens of meters).This study aims to estimate spatially high-resolution SAR-based VOD across different forest ecosystems, using both VH and VV polarizations, and ultimately assess forest water dynamics. We employ soil and vegetation physical scattering models (De Roo et al., 2001; Ulaby & Long, 2014) and constrain the effective scattering albedo (ω), which indicates the ratio of scattering to absorption of vegetation. In our dual-channel approach, we utilize in situ soil moisture from forest ecological observatories and co-polarized S1 backscatter as direct model inputs, and characterize the vegetation structure (ω) using cross-polarized S1 backscatter to estimate SAR-based VOD. We test our approach across two deciduous broadleaf and three evergreen needleleaf forest ecosystems in Central Europe for up to three years (2023-2025). In addition, we compare our SAR-based VOD with VOD estimates from Global Navigation Satellite System-Transmissometry (GNSS-T), derived from a pair of in situ receivers: one located at the top of the canopy and one on the ground for each test site (Brede et al., 2025). We validate our approach using in situ plant gravimetric moisture content (mg; [kgwater/kgwet biomass]) measurements of the tree canopy and remote sensing-based leaf area index. We will also transfer our dual-channel approach to the agricultural site of the Land-Atmosphere Feedback Initiative (LAFI) and other ecosystems in a later step. In the end, spatially high-resolution satellite-based SAR-based VOD enables not only analyses of forest water dynamics but also small-scale up to stand-based assessments of plant hydraulics. ReferencesBrede, B., Schellenberg, K., Camps, A., Chaparro, D., Damm, A., Forkel, M., Frankenberg, C., Ghosh, A., Hartmann, H., Herold, M., Humphrey, V., Jagdhuber, T., Konings, A., Kurum, M., Niederberger, M., Schmullius, C., Stassin, T., Steele-Dunne, S., Borght, N., …, Jonard, F. (2025). VODnet: a virtual GNSS-T VOD network for monitoring of forest water budget and structure. https://doi.org/10.13140/RG.2.2.17146.35522.De Roo, R. D., Du, Y., Ulaby, F. T., Dobson, M. C. (2001). A semi-empirical backscattering model at L-band and C-band for a soybean canopy with soil moisture inversion. IEEE Transactions on Geoscience and Remote Sensing, 39(4), 864–872. https://doi.org/10.1109/36.917912.Ulaby, F. T., Long, D. G. (2014). Microwave radar and radiometric remote sensing. University of Michigan Press.

The ITC Faculty at the University of Twente operates a soil moisture monitoring network consisting of approximately 20 stations that continuously measure volumetric soil water content and soil temperature at up to five depths between 5 and 80 cm. The network was originally established in 2009; over time, several stations have been removed, while others have been added. Its initial purpose was to support the calibration and validation of satellite-based soil moisture products. Recent applications use soil moisture and groundwater monitoring to support adapted water management practices, including adjustable weirs and controlled drainage. For this purpose, supplementary soil moisture stations have been installed in smaller clusters within projects conducted in collaboration with local farmers and the regional water authority Vechtstromen. The quality-checked dataset from 2009-2020 has been published by van der Velde et al. (2023) and also added to the International Soil Moisture Network (ISMN). Furthermore, real-time and historical soil moisture data contribute to the Dutch drought portal. Recently, the soil moisture network has been integrated into the development of a larger multi-sensor infrastructure at the ITC, supported by the NWO-funded Sectorplan in Earth and Environmental Sciences.The collected data will be analyzed to investigate long-term trends, responses to meteorological extremes, and spatial variability in soil moisture across the Twente region. Furthermore, data from soil moisture, meteorological, groundwater, and additional sensors, together with remote sensing observations, will serve as calibration and validation data for an integrated hydrological model. This framework aims to investigate the effects of local agricultural water management practices on water fluxes and water balance components, such as evapotranspiration, groundwater recharge, and surface runoff, and to scale up field-level adaptation measures and their effects to the regional scale. Insights from these investigations are expected to support the identification of sustainable and resilient water management practices from field to regional scales, helping to better cope with increasing water-related challenges such as droughts and flooding.References: van der Velde, R., Benninga, H.-J. F., Retsios, B., Vermunt, P. C., and Salama, M. S.: Twelve years of profile soil moisture and temperature measurements in Twente, the Netherlands, Earth Syst. Sci. Data, 15, 1889–1910, https://doi.org/10.5194/essd-15-1889-2023, 2023.

Vegetation Optical Depth (VOD) has become an increasingly important biogeophysical variable, and has been used to estimate aboveground biomass (AGB) or vegetation water content (VWC). In recent years, GNSS transmissometry (GNSS-T) has been developed as a tool to estimate VOD continuously on plot-level (e.g. Humphrey & Frankenberg (2023)). Currently, observations of GNSS-VOD throughout the globe are being brought together within VODnet, aiming to, amongst others, serve the calibration/validation of satellite VOD products. (Brede et al. (2025)).At the same time, we lack fundamental understanding of the drivers of temporal GNSS-VOD variability and their relative roles under different conditions, as well as the comparability with radiometry-based VOD. Here, we present (1) a detailed analysis of a 2.5-year record of low-cost (u-blox) GNSS-VOD observations from a dense Douglas fir forest in the Netherlands, including quantifying the biophysical drivers of temporal VOD dynamics (i.e. VWC, AGB, interception/dew, and temperature) and their relative importance, and (2) a cross-comparison of GNSS-VOD and L-band, upward-looking radiometer-based VOD measurements from a Scots pine forest in Switzerland.References:Humphrey, V. and Frankenberg, C.: Continuous ground monitoring of vegetation optical depth and water content with GPS signals, Biogeosciences, 20, 1789–1811, 2023.Brede, B., Schellenberg, K., Camps, A., Chaparro Danon, D., Damm, A., Forkel, M., Frankenberg, C., Ghosh, A., Hartmann, H., Herold, M., Humphrey, V., Jagdhuber, T., Konings, A., Kurum, M., Niederberger, M., Schmullius, C., Stassin, T., Steele-Dunne, S., Van der Borght, N., Strube, M., Vermunt, P., Yao, Y., Monteith, A., Richards, E., Persson, H., Lecart, B., and Jonard, F.: VODnet: a virtual GNSS-T VOD network for monitoring of forest water budget and structure, Living Planet Symposium '25, Vienna, https://doi.org/10.13140/RG.2.2.17146.35522, 2025.

The dynamics of water, biomass, and structure of forest ecosystems are challenging to assess on larger spatio-temporal scales with ground-based measurement techniques, particularly beyond individual tree stands. Here, satellite-based remote sensing provides solutions to enhance these assessments.Vegetation optical depth (VOD) is a remote sensing variable that measures the attenuation of microwaves by vegetation. The VOD signal contains information on dry biomass, structure, and water content of vegetation. These signal components can be disentangled using microwave scattering or emission models, depending on active or passive acquisition modes. Short-term variations in VOD time series primarily reflect water dynamics, while seasonal changes are associated with biomass variations. VOD is operationally retrieved globally, with a temporal revisit of 1-to-3 days, from passive satellite sensors like AMSR-2, SMAP, and SMOS, acquiring at a relatively coarse spatial resolution (~40 km) with enhanced interpolations providing ~10 km gridding products. Thus, coarse resolution limits studying forest stands at local scales with passive microwave techniques.This study aims to estimate spatially high-resolution Synthetic Aperture Radar (SAR)-based VOD in forest ecosystems based on Sentinel-1 C-band (5.504 GHz) backscatter data (10 m) from March to September 2023 in Germany. The focus is on two primary study sites, characterized by a deciduous broadleaf ("Leinefelde") and an evergreen needleleaf forest ("Wetzstein"), contrasting the most common forest types in Central Europe.Regarding the methodology, we disentangle C-band VOD in its core components to derive the water content of the upper tree canopy, where the C-band is most sensitive due to microwave penetration capabilities. For this purpose, we employ a combination of physically-based soil and vegetation scattering models (radiative transfer theory). Moreover, we compare our resulting SAR-based VOD time series, among others, against VOD estimates derived from Global Navigation Satellite System-Transmissometry (GNSS-T) at L-band (1.1-1.5 GHz), using in situ receivers, one at the top of the canopy and one on the ground. We further plan to validate our approach with in situ plant gravimetric moisture content (mg; [kgwater/kgwet biomass]) measurements of the tree canopy for both forest sites. Finally, our approach paves the way for further application in agriculture. This will be explored in the new Land-Atmosphere Feedback Initiative (LAFI) in detail.Retrieved satellite-based VOD at such high spatial resolution allows for small-scale up to stand-based analyses of forest water dynamics, biomass changes, and leaf water potential variations. Consequently, these SAR-based VOD dynamics hold potential for monitoring forest health, detecting drought and water stress as well as assessing plant phenology, biomass, and carbon storage.

European forests are experiencing a significant increase in mortality as a result of plagues and wildfires, often enhanced by droughts. Understanding, managing, and adapting our forests to future changes requires large-scale spatiotemporal monitoring of forest water status. Microwave remote sensing products, such as Vegetation Optical Depth (VOD), are a valuable addition to optical metrices due to their direct relationship to vegetation water content (VWC) and consistent time series. These products are increasingly used for studying drought-stressed vegetation. However, unlike soil moisture, there is little continuous ground-based data to validate VWC estimates from microwave instruments. To accelerate the development of VWC products, we need continuous ground-based measurements for various locations. In a Dutch coniferous forest, we have tested state-of-the-art sensors and approaches to measure water content and potential continuously (i.e. using frequency- and time domain reflectometers, GNSS receivers, microtensiometers, dielectric leaf wetness sensors, sap flow sensors, dendrometers). The objectives are (I) to perform long-term continuous measurements, (II) to use our insights to scale up the measurements to other locations. After multiple months of measuring, we here present our first results.

In situ soil moisture measurement networks are essential for developing, improving and validating satellite soil moisture products. In the east of the Netherlands, an area susceptible to droughts, a monitoring network for soil moisture and – temperature has been operational since 2009. Spread across an area of 45 by 40 km, twenty profile monitoring stations observe moisture and temperature in the root zone. Four field campaigns were conducted in order to calibrate the sensors and to assess the spatial representativeness of the measurements.The network has proven to be of great value for validation of satellite products (e.g. for SMAP soil moisture). In addition, continuation of the measurements will increase its value for climate studies. Currently, the network is being redesigned to better suit operational water management in the region, while preserving the value of long time series for climatological research, and taking into account its potential value for future missions. Here, we present an overview of the network and the observations since its establishment, including the adjustments that are being made to the network and research opportunities.Van der Velde, R., Benninga, H.J.F., Retsios, B., Vermunt, P.C., Salama, M.S. (2022) - Twelve years profile soil moisture and temperature measurements in Twente, the Netherlands. Earth System Science Data Discussions, 1-44.Velde, dr. ir. R van der (University of Twente) (2022): Twelve years profile soil moisture and temperature measurements in Twente, the Netherlands. DANS. https://doi.org/10.17026/dans-znj-wyg5

Spread across Twente and its neighbouring regions in the east of the Netherlands, a network of 20 profile soil moisture and temperature (5, 10, 20, 40, and 80 cm depths) monitoring stations was established in 2009. Field campaigns have been conducted covering the growing seasons of 2009, 2015, 2016, and 2017, during which soil sampling rings and handheld probes were used to measure the top 5 cm volumetric soil moisture content (VSM) of 28 fields near 12 monitoring stations. In this paper, we describe the design of the monitoring network and the field campaigns, adopted instrumentation, experimental setup, field sampling strategies, and the development of sensor calibration functions. Maintenance and quality control procedures and issues specific to the Twente network are discussed. Moreover, we provide an overview of open third-party datasets (i.e. land cover/use, soil information, elevation, groundwater, and meteorological observations) that can support the use and analysis of the Twente soil moisture and temperature datasets beyond the scope of this contribution. An indication for the spatial representativeness of the permanent monitoring stations is provided through comparisons of the 5 cm station measurements with the top 5 cm field-averaged VSM derived from the field campaign measurements. The results reveal in general reasonable agreements and root mean squared errors that are dominated by underestimations of the field-averaged VSM, which is particularly apparent for the grass fields and is strong after heavy rain. Further, we discuss the prospects the datasets offer to investigate (i) the reliability of soil moisture references that serve the development and validation of soil moisture products, and (ii) the water and energy exchanges across the groundwater–vadose-zone–atmosphere continuum within a lowland environment in a changing climate. The datasets discussed are publicly available at https://doi.org/10.17026/dans-znj-wyg5 (Van der Velde et al., 2022).

For a good interpretation of radar backscatter sensitivity to vegetation water dynamics, we need to know which parts of the vegetation layer control that backscatter. However, backscatter sensitivity to different depths in the canopy is poorly understood. This is partly caused by a lack of observational data to describe the vertical moisture distribution. In this study, we aimed to understand the sensitivity of L-band backscatter to water at different heights in a corn canopy. We studied changes in the contribution of different vertical layers to total backscatter throughout the season and during the day. Using detailed field measurements, we first determined the vertical distribution of moisture in the plants, and its seasonal and sub-daily variation. Then, these measurements were used to define different sublayers in a multi-layer water cloud model (WCM). To calibrate and validate the WCM, we used hyper-temporal tower-based polarimetric L-band scatterometer data. WCM simulations showed a shift in dominant scattering from the lowest 50 cm to 50–100 cm during the season in all polarizations, mainly due to leaf and ear growth and corresponding scattering and attenuation. Dew and rainfall interception raised sensitivity to upper parts of the canopy and lowered sensitivity to lower parts. The methodology and results presented in this study demonstrate the importance of the vertical moisture distribution on scattering from vegetation. These insights are essential to avoid misinterpretation and spurious artefacts during retrieval of soil moisture and vegetation parameters.

In recent years, radar remote sensing has been increasingly used for studies on interactions between vegetation and hydrology. New opportunities arise for more advanced studies, as unprecedented spatiotemporal monitoring capability will be offered by the next generation of spaceborne radar instruments. To avail of these opportunities, we need a better understanding of the water dynamics in a canopy which is captured in a radar signal.Here, we present our latest findings from two experimental campaigns. In these campaigns, we used a ground-based prototype L-band radar instrument to obtain sub-daily observations, and extensive hydrometeorological measurements to monitor the water flow and storage in the soil-plant-atmosphere continuum of a corn field. Estimating 15-minute fluctuations of vegetation water content (VWC), surface canopy water (dew, interception), and surface soil moisture allowed us to quantify backscatter sensitivity to each of these moisture stores.It will be shown that the nocturnal cycle of dew, and the diurnal cycle of VWC have a considerably higher effect on L-band backscatter than previously assumed. Both particularly affected the vertically polarized signals. Furthermore, we will demonstrate that the non-uniform vertical distribution of moisture in the canopy is dynamic, both on seasonal and diurnal timescales. A modelling study quantified the impact this has on backscatter. Our findings demonstrate the opportunities for spaceborne sub-daily radar observations to monitor rapid vegetation water dynamics. Moreover, they offer insights for future validation field campaigns. ReferencesVermunt, P. C., Khabbazan, S., Steele-Dunne, S. C., Judge, J., Monsivais-Huertero, A., Guerriero, L., & Liu, P. W. (2020). Response of Subdaily L-Band Backscatter to Internal and Surface Canopy Water Dynamics. IEEE Transactions on Geoscience and Remote Sensing, 59, 7322-7337.Vermunt, P. C., Steele-Dunne, S. C., Khabbazan, S., Judge, J., & van de Giesen, N. C. (2021). Reconstructing Continuous Vegetation Water Content To Understand Sub-daily Backscatter Variations. Hydrology and Earth System Sciences Discussions, 1-26.