Soil Moisture Estimation Research Articles

Surface soil moisture estimation in bare agricultural soil using modified Dubois model for Sentinel-1 C-band SAR data

Surface soil moisture has vital role in water energy balance, climate change and agriculture mainly for crop water requirements and irrigation scheduling. Microwave remote sensing with its unique characteristics of high penetration and sensitivity towards dielectric constant, has enabled the researchers to explore various techniques for soil moisture estimation. With the launch of Sentinel-1 (A&B) Synthetic Aperture Radar (SAR) satellites, the hindrance in accessing high spatial and temporal resolution data is eliminated. The current study focuses on surface soil moisture estimation for bare agricultural fields in the semi-arid region. Field soil moisture up to 5 cm depth using HydraGo Probe sensor and surface roughness synchronizing with satellite pass dates were collected from total 102 locations spanning four dates. Volumetric and sensor-based soil moisture are well correlated with R2 = 0.85. The Modified Dubois Model (MDM) was applied to obtain the relative permittivity of the soil for the backscattering coefficient (σ◦) for VV polarization, which is used as one of the inputs in universal Topp’s model for soil moisture calculation. Model derived soil moisture is well correlated with ground-based soil moisture for the entire range of the soil moisture (0.02-0.18 m3m-3) with R2 = 0.85 and RMSE=0.005. The entire soil moisture was categorized in three soil moisture ranges to evaluate the sensitivity. The highest correlation was observed for 0.06-0.1 m3m-3 with R2 = 0.73 and RMSE=0.003 followed by 0.015-0.6 m3m-3 with R2 = 0.81 and RMSE=0.001 and 0.11-0.18 m3m-3 with R2 = 0.48 and RMSE=0.019 which is significantly low. Performance accuracy of MDM is encouraging for bare soil moisture estimation for even the lower range of surface soil moisture.

Calibration of Dobson model for improving soil moisture retrievals from AMSR satellite imagery

Satellite-based retrieval of environmental variables has seen rapid advancements in recent years, although it remains challenged by various sources of uncertainty. In this study, we endeavor to enhance the accuracy of soil moisture estimation from AMSR imagery through parameter calibration. Our focus is on calibrating the Dobson dielectric mixing model, a component of the radiative transfer model that relies heavily on empirical relationships.
 The calibration process is based on black-box optimization techniques for the retrieval problem. We have adapted the CORS optimization algorithm to address the specific characteristics of our task. We also considered different target functions for calibration.
 To evaluate the efficacy of our framework, we conducted tests across a dataset comprising 118 ground stations in the United States. The outcomes reveal that optimizing parameter settings can provide limited improvement to the accuracy and, when configured accordingly, address specific issues such as bias correction. Calibration emerges as a potent tool for refining surface soil moisture retrievals, although its effectiveness tends to diminish in larger calibration areas.

Comparing quasi-3D soil moisture derived from electromagnetic induction with 1D moisture sensors and correlation to barley yield in variable duplex soil

Farms in Western Australia (WA) are highly variable in soil texture and water retention capacity; therefore, information of soil moisture spatial variability in the field is important for effective crop management. In practice, farmers often rely on point sensors to determine soil moisture in their fields for crop planning. The limitation of point measurements to account for spatial variability highlights the need to develop methods that can assess soil moisture across variable broadacre fields. This information can enable more effective site-specific crop management practices. In this study, we used a mobile non-intrusive electromagnetic induction (EMI) sensor to map soil apparent electrical conductivity (ECa) to predict soil moisture levels at three different depths (0–0.5, 0.5–0.8 and 0.8–1.6 m) across the field and compared it with barley yield production. To convert the depth specific electrical conductivity (EC) from EMI surveys into soil moisture estimates, calibrations between electrical resistivity tomography (ERT) to volumetric moisture were used. These calibrations were developed for the different soil textural classes of the field, with R2 of 0.97–0.99. There was no significant difference between the estimated soil moisture from EMI surveys and the point moisture measurements obtained from soil moisture sensors and soil sample extracts, with Pearson R values of 0.64 for 0–0.5 m depth and 0.73 for 0.5–0.8 m depth. Additionally, correcting the temperature drift relative to the measured soil temperature on EMI surveys by 10% in dry season and 31% in wet season, improved the moisture estimation results. This study successfully demonstrated the effectiveness of spatial soil moisture estimation using EMI sensor in a field characterized by horizontal and vertical soil texture variability. However, no significant correlation was found between the mapped soil moisture or soil texture with barley yield. This may be due to relatively high initial moisture levels resulting from two years of fallow rotation.

Bias-blind and bias-aware assimilation of leaf area index into the Noah-MP land surface model over Europe

Abstract. Data assimilation (DA) of remotely sensed leaf area index (LAI) can help to improve land surface model estimates of energy, water, and carbon variables. So far, most studies have used bias-blind LAI DA approaches, i.e. without correcting for biases between model forecasts and observations. This might hamper the performance of the DA algorithms in the case of large biases in observations or simulations or both. We perform bias-blind and bias-aware DA of Copernicus Global Land Service LAI into the Noah-MP land surface model forced by the ERA5 reanalysis over Europe in the 2002–2019 period, and we evaluate how the choice of bias correction affects estimates of gross primary productivity (GPP), evapotranspiration (ET), runoff, and soil moisture. In areas with a large LAI bias, the bias-blind LAI DA leads to a reduced bias between observed and modelled LAI, an improved agreement of GPP, ET, and runoff estimates with independent products, but a worse agreement of soil moisture estimates with the European Space Agency Climate Change Initiative (ESA CCI) soil moisture product. While comparisons to in situ soil moisture in areas with weak bias indicate an improvement of the representation of soil moisture climatology, bias-blind LAI DA can lead to unrealistic shifts in soil moisture climatology in areas with strong bias. For example, when the assimilated LAI data in irrigated areas are much higher than those simulated without any irrigation activated, LAI will be increased and soil moisture will be depleted. Furthermore, the bias-blind LAI DA produces a pronounced sawtooth pattern due to model drift between DA updates, because each update pushes the Noah-MP leaf model to an unstable state. This model drift also propagates to short-term estimates of GPP and ET and to internal DA diagnostics that indicate a suboptimal DA system performance. The bias-aware approaches based on a priori rescaling of LAI observations to the model climatology avoid the negative effects of the bias-blind assimilation. They retain the improvements in GPP anomalies from the bias-blind DA but forego improvements in the root mean square deviations (RMSDs) of GPP, ET, and runoff. As an alternative to rescaling, we discuss the implications of our results for model calibration or joint parameter and state update DA, which has the potential to combine bias reduction with optimal DA system performance.

A Typical Review of Current and Prospective Microwave and Optical Remote Sensing Datasets for Soil Moisture Retrieval

Soil Moisture content is a vital indicator of both the weather and the water cycle. It has been a long-standing difficulty for the field of remote sensing to make sense of soil moisture's spatial and temporal distribution. For over five decades, researchers across the world have exclusively investigated the optical and microwave datasets for estimating soil moisture by developing various models, and algorithms. Nevertheless, challenges are faced in the consistent retrieval of SM at local, and global scales with higher accuracy in space and time resolution. The review was conducted in-depth, looking at the methods using optical and microwave data to determine soil moisture, and outlining the benefits and drawbacks considering the current needs. With this research, a new age of widespread use of space technology for remote sensing of soil moisture has been ushered in. The study also acknowledges the scientific challenges of utilizing remote sensing datasets for soil moisture measurement.

Mapping spatiotemporal soil moisture in highly heterogeneous agricultural landscapes using mobile dual‐spectra cosmic‐ray neutron sensing

AbstractAccurate large‐scale soil moisture (SM) maps are crucial for catchment‐scale hydrological models used for water resource management and warning systems for droughts, floods, and wildfires. SM can be mapped by mobile cosmic‐ray neutron (CRN) systems of moderated detectors at homogeneous landscapes of similar soil and vegetation. In this study, we present a new approach for mobile CRN detection to perform to its full potential, where CRN measurements can also be converted to SM at heterogeneous landscapes. The approach is based solely on thermal and epithermal neutron datasets obtained from mobile dual‐spectra CRN detection, combined with theoretical developments using a particle transport model. For each measurement point, the land cover type is identified using the thermal‐to‐epithermal (T/E) ratio, and the relevant neutron‐count‐to‐soil‐moisture conversion function is estimated from CRN stations located at the main land cover types in the catchment. With this approach, the requirement of collecting 100+ soil samples for each point along the survey route is omitted. We use this T/E‐dependent approach to obtain SM maps from 12 CRN surveys and compare it with a simple approach where only the conversion function from the agricultural site is used. SM by the simple approach is comparable to the estimates of the agricultural stations of a capacitance sensor network, while the estimates of the T/E‐dependent approach also compare well with the heathland and forest stations. With accurate SM estimates for all landcover types, the average error is reduced from 0.089 to 0.038 when comparing CRN SM with space‐borne Soil Moisture Active Passive Mission estimates.

Assessing soil moisture levels using visible UAV imagery and machine learning models

The estimation of soil moisture (SM) as an important variable in the hydrological cycle of nature is necessary for the optimal management of water and soil resources. One of the indirect methods to estimate SM is using visible imagery with unmanned aerial vehicles (UAVs). This study aims to evaluate the potential of visible UAV imagery for estimating SM in a bare soil field in Iran. In this study, M5 tree (M5P), random forest (RF), sequential minimal optimization regression (SMOreg), and multilayer perceptron (MLP) methods have been used for SM modeling from RGB (Red, Green and Blue) bands and brightness and intensity indices of aerial imagery. Three evaluation methods were used to assess the accuracy of the models, including the coefficient of determination (R2), root mean square error (RMSE), and mean absolute error (MAE). Four different indices, including difference index (DI), ratio index (RI), normalized difference index (NDI), and perpendicular index (PI), were used to estimate SM. The green and red bands pair were found to be the optimal bands for SM estimation. The findings showed that the PI index provided the most accurate SM estimates (R2 = 0.51). The RF model predicted SM most accurately among the machine learning models tested (R2 = 0.67). However, all models underestimated SM content in high-moisture areas and overestimated it in low-moisture areas, with the MLP model showing the most significant overestimation. All the indices were saturated beyond 25% SM. In general, this study highlighted the potential of aerial RGB imagery and associated indices for assessing SM levels within bare soil fields. However, it should be noted that the use of individual bands and indices alone is not sufficient to make an accurate estimate of SM.

Estimation of soil moisture using environmental covariates and machine learning algorithms in Cathedral Peak Catchment, South Africa

AbstractSoil moisture (SM) is a fundamental constituent of the terrestrial environment and the hydrological cycle. Owing to its significant influence on catchment hydrological responses, it can be utilized as an indicator of floods and droughts to aid early warning systems. This study aimed to develop a field‐scale method to estimate SM using parametric and machine learning‐based methods to compare whether advanced artificial intelligence methods can give similar results as traditional methods. Considering this, monthly observed SM data (from the top 10 cm), environmental covariates, and remotely sensed data from March 2019 to July 2021 for the Cathedral Peak Research Catchments VI and IX in South Africa were obtained. From the 241 observations obtained across 12 sites, 160 (∼66%) were used for model training, while the remaining 81 (∼34%) were used for model testing. Employing 10‐fold cross‐validation, the individual machine learning models (viz., support vector machine [SVM], random forest (RF), k‐nearest neighbor, classification and regression trees [Rpart], and generalized linear model) displayed a satisfactory performance (R2 = 0.52–0.79; root mean square error = 3.79–5.80). In the validation phase, the RF model displayed a superior performance, followed by the SVM. Subsequent SM estimation using the hybrid model produced satisfactory results in training (R2 = 0.90) and testing (R2 = 0.45). The results obtained from this study can aid in predicting SM variations in catchments with limited monitoring. Furthermore, this model can be applied in drought monitoring, forecasting, and informing agricultural management practices.

Evaluation of Improvement Schemes for FY-3B Passive Microwave Soil-Moisture Estimates Retrieved Using the Land Parameter Retrieval Model

Satellite observations have provided global and regional soil-moisture estimates in the last four decades. However, the accuracy of these observations largely depends on reducing uncertainties in the retrieval algorithms. In this study, we address two challenges affecting the quality of soil-moisture estimates from a widely used soil-moisture-retrieval model, the land parameter retrieval model (LPRM). We studied two improvement schemes that were aimed at reducing uncertainties in open water signals (the LPRMv6_OWF) and vegetation signals (the LPRMv6_Veg), as well as a scheme to reduce their combined impacts (the LPRMv6_OWFVeg) on LPRM-retrieved soil moisture using the FengYun-3B (FY-3B) satellite observations. To assess the impacts of the improvement schemes, we utilized in situ soil moisture from the Jiangsu and Jiangxi provinces in China. We found that the retrievals (Rs) of the LPRMv6_Veg and the LPRMv6_OWFVeg were mainly in the range of 0.2 to 0.5 in Jiangsu and Jiangxi, with increases of 0.1 compared to those of the LPRMv6. The standard deviation (SD) of the LPRMv6_OWFVeg increased in Jiangsu, while the R of the LPRMv6_OWF increased in Jiangsu by 0.05–0.1 compared to that of the LPRMv6, but the SD tended to become worse. In Jiangxi, there was an increase of 0.1 in R. The results show that each of these algorithms improved the accuracy of soil-moisture inversion to some extent, compared to the original algorithm, with the LPRMv6_OWFVeg showing the greatest improvement, followed by the LPRMv6_Veg. The accuracy of both the LPRMv6_OWF and the LPRMv6_OWFVeg decreased to some extent when the open-water fraction (OWF) was greater than 0.2. Full areal extent analyses based on triple collocation showed significant improvements in correlations and minimized errors across different vegetation scenarios over the entire region of China in both the LPRMv6_OWF and the LPRMv6_Veg. However, reduced qualities were found in arid regions in northern China because of the nonlinear relationships between land-surface temperature, vegetation, and soil moisture in the LPRM. These results highlight important lessons for developing comprehensive improvement schemes for soil-moisture retrievals from passive microwave satellite observations.

Evaluation of Spire GNSS-R reflectivity from multiple GNSS constellations for soil moisture estimation

ABSTRACT Spaceborne Global Navigation Satellite System-Reflectometry (GNSS-R) has emerged as a pivotal tool with different land applications, prominently encompassing soil moisture estimation. In contrast to conventional radiometer satellites commonly used for this purpose, GNSS-R offers higher spatiotemporal coverage while maintaining cost-effectiveness. The potential of using Global Positioning System (GPS) reflections measured by the Cyclone Global Navigation Satellite System (CYGNSS) mission to retrieve soil moisture has been previously demonstrated. In 2019, Spire Global Inc. launched their first GNSS-R satellites, which now comprise a constellation of four CubeSats. These satellites track reflections from multi-constellation, encompassing GPS, Galileo, BeiDou Navigation Satellite System (BDS), and Quasi-Zenith Satellite System (QZSS). In this study, an analysis and validation of Spire GNSS-R L1B surface reflectivity for soil moisture retrieval within east Australia during an eight-month period in 2021 is presented. A comparison of the estimated Spire surface reflectivity to that of CYGNSS is performed, unveiling analogous behavioural patterns and biases across both missions. Soil moisture is estimated using observations from Spire GPS-only, Spire multi-constellation, and CYGNSS. The Soil Moisture Active and Passive (SMAP) retrievals are used as the reference, presuming a linear relationship between changes in soil moisture and changes in reflectivity. Our results indicate that the Spire GNSS-R mission can detect variations in soil moisture with a performance comparable to that of CYGNSS. A median unbiased root-mean-square difference (ubRMSD) of 0.062 m3.m−3 is found for both Spire GPS and multi-constellation when using 9-km products and SMAP as the reference.

Soil moisture simulation of rice using optimized Support Vector Machine for sustainable agricultural applications

The growth and development of rice crops primarily depend on appropriate soil water balance for which soil moisture is the key determinant. Soil moisture is a crucial parameter in the hydrological cycle, which has a vital role in optimal water management for sustainable agricultural growth as it has a significant impact on hydrological, ecological, and climatic processes. Thus, accurate estimation of soil moisture is important otherwise it will drastically reduce crop yields, intensifying the global food crisis. A novel soil moisture prediction model (SVM-COLGWO) that incorporates the Grey Wolf Optimizer (GWO) into Chebyshev chaotic maps and opposition-based learning to optimize the Support Vector Machine (SVM) model is proposed. The suggested model simultaneously increases the simulated model’s accuracy while speeding up global convergence. To evaluate the proposed model, the prediction performance is compared with other hybrid and standalone models where the feasibility of the proposed model is validated through superior simulation results (MAE = 0.167, MSE = 0.179, RMSE = 0.423, MAPE = 0.162, and R2= 0.949) including Shannon’s Entropy. Thus, based on accurate soil moisture simulation through the proposed model, irrigation can be effectively scheduled for sustainable rice growth.

Contactless estimation of soil moisture using leaky Rayleigh waves and a fully convolutional network

AbstractSoil moisture is a key factor that influences various aspects of ecosystem functioning. Measuring soil moisture without installing any objects in the soil is desirable because it allows for accurate characterizations of soil moisture while minimizing impacts on soil structure and ecology. In this study, we explored the potential of leaky Rayleigh waves as a proxy to contactlessly estimate soil moisture. We developed an ultrasonic system containing a transducer, receivers, and acoustic barrier. The specimens of sand, silt, and clay were utilized. Experiments were conducted over 4 months. We used a widely used soil‐embedded moisture sensor to compare and develop relationships between leaky Rayleigh waves and soil moisture. Our results showed that as soil moisture increased, the velocity and amplitude of leaky Rayleigh waves decreased because water molecules attracted to the soils led to their attenuation. However, their magnitudes were not considerable except for very dry soils. To overcome these limited relations to estimate soil moisture from leaky Rayleigh waves, we constructed authentic images based on the observed leaky Rayleigh waves and used them as inputs for a fully convolutional network. We found that the combination of the ultrasonic system and deep learning approach developed in this study were suitable for estimating soil moisture without soil disturbances (RMSE = 0.01 m3 m−3). This study suggests that leaky Rayleigh waves have the potential to serve as a reliable proxy for determining soil moisture without the need for physical contact.

Retrieval of Surface Soil Moisture over Wheat Fields during Growing Season Using C-Band Polarimetric SAR Data

Accurate estimation and regular monitoring of soil moisture is very important for many agricultural, hydrological, or climatological applications. Our objective was to evaluate potential contributions of polarimetry to soil moisture estimation during crop growing cycles using RADARSAT-2 C-band images. The research focused on wheat field data collected during Soil Moisture Active Passive Validation Experiment (SMAPVEX12) conducted in 2012 in Manitoba (Canada). A sensitivity analysis was performed to select the most relevant non-polarimetric and polarimetric variables extracted from RADARSAT-2, and statistical models were developed to estimate soil moisture. In fine, three models were developed and validated: a non-polarimetric model based on cross-polarized backscattering coefficient σHV0; a polarimetric mixed model using six polarimetric and non-polarimetric retained variables after the sensitivity analysis; and a simplified polarimetric mixed model considering only the phase difference (ϕHH−VV) and the co-polarized backscattering coefficient σHH0. The validation reveals significant positive contributions of polarimetry. It shows that the non-polarimetric model has a much larger error (RMSE = 0.098 m3/m3) and explains only 19% of observed soil moisture variation compared to the polarimetric mixed model, which has an error of 0.087 m3/m3, with an explained variance of 44%. The simplified model has the lowest error (0.074 m3/m3) and explains 53.5% of soil moisture variation.

A new variant of the optical trapezoid model (OPTRAM) for remote sensing of soil moisture and water bodies

Over the past few years, the Optical Trapezoid Model (OPTRAM) has been widely used as a means for high-resolution mapping of surface soil moisture using optical satellite data. In this paper, we propose a new variant of OPTRAM that can map not only soil moisture, but also water bodies such as lakes and rivers. The proposed variant was tested using laboratory experimental data as well as Landsat-8 reflectance observations. Results showed the new OPTRAM variant has greater skill than the original variant in separating land and water pixels. In addition, the new variant showed less sensitivity to the model parameters, and hence, is less user dependent. To quantitatively examine the user-dependency of the model, we analyzed OPTRAM soil moisture based on Landsat-8 satellite images in California, where we varied the model parameters in a plausible range. The correlations of the resulting maps in terms of R2 between two largely different sets of parameters were found in the range of 0.47-0.52 for the original variant and 0.67-0.76 for the new variant. Because some OPTRAM parameters can be quite uncertain, particularly in wet regions, the reduced sensitivity promises more consistent soil moisture estimates across the range of parameter choices.

Sensitivity of surface soil moisture retrieval to satellite-derived vegetation descriptors over wheat fields in the Kairouan plain

ABSTRACT Soil moisture estimation is a key component in hydrological processes and irrigation amounts' estimation. The synergetic use of optical and radar data has been proven to retrieve the surface soil moisture at a field scale using the Water Cloud Model (WCM). In this work, we evaluate the impact of staellite-derived vegetation descriptors to estimate the surface soil moisture. Therefore, we used the Sentinel-1 data to test the polarization ratio ( σ V H 0 / σ V V 0 ) and the normalized polarization ratio (IN) and the frequently used optical Normalized Difference vegetation Index (NDVI) as vegetation descriptors. Synchronous with Sentinel-1 acquisitions, in situ soil moisture were collected over wheat fields in the Kairouan plain in the center of Tunisia. To avoid the bare soil roughness effect and the radar signal saturation in dense vegetation context, we considered the data where the NDVI values vary between 0.25 and 0.7. The soil moisture inversion using the WCM and NDVI as a vegetation descriptor was characterized by an RMSE value of 5.6 vol.%. A relatively close performance was obtained using IN and ( σ V H 0 / σ V V 0 ) with RMSE under 7. 5 vol.%. The results revealed the consistency of the radar-derived data in describing the vegetation for the retrieval of soil moisture.

IMERG Precipitation Improves the SMAP Level-4 Soil Moisture Product

Abstract The NASA Soil Moisture Active Passive (SMAP) mission Level-4 Soil Moisture (L4_SM) product provides global, 9-km resolution, 3-hourly surface and root-zone soil moisture from April 2015 to the present with a mean latency of 2.5 days from the time of observation. The L4_SM algorithm assimilates SMAP L-band (1.4 GHz) brightness temperature (Tb) observations into the NASA Catchment land surface model as the model is driven with observation-based precipitation. This paper describes and evaluates the use of satellite- and gauge-based precipitation from the NASA Integrated Multi-satellitE Retrievals for the Global Precipitation Measurement (IMERG) products in the L4_SM algorithm beginning with L4_SM Version 6. Specifically, IMERG is used in two ways: (i) The L4_SM precipitation reference climatology is primarily based on IMERG-Final (Version 06B) data, replacing the Global Precipitation Climatology Project Version 2.2 data used in previous L4_SM versions, and (ii) the precipitation forcing outside of North America and the high latitudes is corrected to match the daily totals from IMERG, replacing the gauge-only, daily product or uncorrected weather analysis precipitation used there in earlier L4_SM versions. The use of IMERG precipitation improves the anomaly time series correlation coefficient of L4_SM surface soil moisture (versus independent satellite estimates) by 0.03 in the global average and by up to ∼0.3 in parts of South America, Africa, Australia, and East Asia, where the quality of the gauge-only precipitation product used in earlier L4_SM versions is poor. The improvements also reduce the time series standard deviation of the Tb observation-minus-forecast residuals from 5.5 K in L4_SM Version 5 to 5.1 K in Version 6. Significance Statement Soil moisture links the land surface water, energy, and carbon cycles. NASA Soil Moisture Active Passive (SMAP) satellite observations and observation-based precipitation data are merged into a numerical model of land surface water and energy balance processes to generate the global, 9-km resolution, 3-hourly Level-4 Soil Moisture (L4_SM) data product. The product is available with ∼2.5-day latency to support Earth science research and applications, such as flood prediction and drought monitoring. We show that a recent L4_SM algorithm update using satellite- and gauge-based precipitation inputs from the NASA Integrated Multi-satellitE Retrievals for the Global Precipitation Measurement (IMERG) products improves the temporal variations in the estimated soil moisture, particularly in otherwise poorly instrumented regions in South America, Africa, Australia, and East Asia.

A Deep Learning Data Fusion Model Using Sentinel-1/2, SoilGrids, SMAP, and GLDAS for Soil Moisture Retrieval

Abstract We develop a deep learning–based convolutional-regression model that estimates the volumetric soil moisture content in the top ∼5 cm of soil. Input predictors include Sentinel-1 (active radar) and Sentinel-2 (multispectral imagery), as well as geophysical variables from SoilGrids and modeled soil moisture fields from SMAP and GLDAS. The model was trained and evaluated on data from ∼1000 in situ sensors globally over the period 2015–21 and obtained an average per-sensor correlation of 0.707 and ubRMSE of 0.055 m3 m−3, and it can be used to produce a soil moisture map at a nominal 320-m resolution. These results are benchmarked against 14 other soil moisture evaluation research works at different locations, and an ablation study was used to identify important predictors. Significance Statement Soil moisture is a key variable in various agriculture and water management systems. Accurate and high-resolution estimates of soil moisture have multiple downstream benefits such as reduced water wastage by better understanding and managing the consumption of water, utilizing smarter irrigation methods and effective canal water management. We develop a deep learning–based model that estimates the volumetric soil moisture content in the top ∼5 cm of soil at a nominal 320-m resolution. Our results demonstrate that machine learning is a useful tool for fusing different modalities with ease, while producing high-resolution models that are not location specific. Future work could explore the possibility of using temporal input sources to further improve model performance.

Improvement and application on the estimation model of windbreak and sand fixation function based on remote sensing soil moisture factor

Soil moisture factor is one of the important parameters in the study of wind and sand fixation functions of ecosystems. Traditional methods often use potential evaporation, rainfall, and irrigation observed by meteorological stations to estimate soil moisture, which has significant limitations in terms of spatial continuity and data availability. Based on the development of remote sensing technology in soil moisture detection, we selected four remote sen-sing indicators for soil moisture (MODIS evapotranspiration ratio method, SMAP soil moisture ratio method, visible shortwave infrared drought index method, and remote sensing humidity index method) to improve the estimation of soil moisture factor in the modified wind erosion equation model (RWEQ), and used the improved algorithm to analyze the spatiotemporal variations and driving factors of wind prevention and sand fixation services in the northwest region of Liaoning Province from 2001 to 2021. The results showed that the MODIS evapotranspiration ratio method had the highest correlation with traditional meteorological methods in calculating soil moisture. The formula obtained by fitting the two could be used to improve the calculation of soil moisture factor in the RWEQ model. From 2001 to 2021, the wind prevention and sand fixation capacity in the northwest region of Liaoning Province showed strong spatial distribution characteristics in the northern and eastern regions, while weak in the central and western regions. According to Mann-Kendall trend testing, 72.7% of the regions in northwest Liaoning Province were showing an upward trend in their ability to prevent wind and fix sand. The application of geographic detector models for driving factor analysis showed that the change in wind and sand fixation capacity was a process of multiple factors interacting with each other, greatly influenced by soil type, annual wind speed, and economic development level. Moreover, the interaction between various driving factors had a higher impact on wind and sand fixation than that of single factors. The results could improve the RWEQ model estimation and provide technical support for the long-term analysis of ecological function formation mechanisms and driving forces in the northwest region of Liaoning.

Optimizing Soil Moisture Retrieval: Utilizing Compact Polarimetric Features with Advanced Machine Learning Techniques

Soil moisture plays a crucial role in various environmental processes and is essential for agricultural management, hydrological modeling, and climate studies. Synthetic Aperture Radar (SAR) remote sensing presents significant potential for estimating soil moisture due to its ability to operate in all weather conditions and provide day-and-night imaging capabilities. Among the SAR configurations, the Compact Polarimetric (CP) mode has gained increasing interest as it relaxes system constraints, improves coverage, and enhances target information compared to conventional dual polarimetric SAR systems. This paper introduces a novel approach for soil moisture retrieval utilizing machine learning algorithms and CP SAR features. The CP SAR features are derived from a series of RADARSAT Constellation Mission (RCM) CP SAR imagery acquired over Canadian experimental sites equipped with Real-Time In Situ Soil Monitoring for Agriculture (RISMA) stations. This study employs a diverse dataset of compact polarimetric SAR features and corresponding ground truth soil moisture measurements for training and validation purposes. The results of our study achieved a Root Mean Square Error (RMSE) of 6.88% with a coefficient of determination R2 equal to 0.60, which corresponds to a correlation R between true and predicted soil moisture values of 0.75, using optimized Ensemble Learning Regression (ELR) with a decision-tree-based model. These results improved, yielding an RMSE of 5.67% and an R2 equal to 0.73 (R = 0.85), using an optimized Gaussian Process Regression (GPR) model.

A double two-sources energy-water balance model for improving evapotranspiration estimates and irrigation management in fruit trees fields

Improving the use of water in irrigated agriculture is meaningful in a period of increasing water scarcity conditions. A more accurate estimate of evapotranspiration (ET) and its components thus becomes fundamental to better quantify the irrigation water volumes. Many existing models, based on different remote sensing data, provide daily estimates of latent heat flux (LE) from correlation between net radiation and instantaneous estimates of LE, computed as residual component of the energy balance equation or from a correlation of land surface temperature (LST) with vegetation indices., However, they mainly lack of solutions which are continuous in time (e.g. hourly), independent from satellite LST availability and a simultaneous estimation of soil moisture. Addressing this gap, a double two-sources energy-water balance model (FEST‐2×2‐EWB) is developed based on the decoupling of both water and energy fluxes between the bare soil or grass interrow and the trees rows. This novel parameterization approach enables the differentiation of water uptake from the root zone, which varies between tall trees and grass, considering the dynamics of soil moisture (SM) in the superficial and deep layers. Additionally, it provides partitioned values for transpiration and evaporation. The new model has been evaluated in two irrigated trees fields in the North of Italy, a walnut trees field from 2019 to 2021 and a pear trees field, for the year 2022. Results of the study showcased a root-mean-square-error (RMSE) of about 55 W m− 2 and a bias of about 40 W m− 2 for hourly latent and sensible heat fluxes when compared to the eddy covariance stations located in the fields, while a RMSE of 2 °C (bias of 1.5 °C) for LST and of 0.04 for SM. The FEST‐2×2‐EWB model significantly enhances the accuracy of ET simulation in fruits trees areas in respect to the original one source and one-layer version of the same model. Finally, the application of an irrigation optimization strategy with this new model, allowed to demonstrate its potentiality in water saving (about 90 mm in a year) in respect to farmers applied irrigation, and with a difference of about 60 mm between using the double two-sources model and single source one.