Validation and ensemble-based layer-wise correction of soil moisture observations from automatic stations

Soil moisture is one of the fundamental variables in land–atmosphere interactions, hydrological processes and vegetation dynamics. Accurate soil moisture information is of great significance for environmental and climate studies. In recent years, automatic monitoring stations have been increasingly deployed owing to their advantages of high-frequency and standardized measurements. However, their measurements suffer from non-negligible biases, limiting the reliability of automatic soil moisture datasets. To address these challenges, this study conducts a comprehensive validation and machine-learning-based correction of three automatic soil moisture systems: the automatic soil moisture station, the CRS-2000C regional soil moisture measurement system, and the soil temperature–moisture monitoring system at the Xilinhot National Climate Observatory, China. Using manual measurements as reference, the observations from the three automatic stations were validated. All the automatic soil moisture measurements revealed substantial biases, especially in deeper soil layers. To reduce the biases, an ensemble correction framework that employed generalized additive model to integrate Cubist, Random Forest, XGBoost, and CatBoost models was developed for layer-wise soil moisture correction. Five-fold cross-validation was applied to evaluate correction performance. After correction, the accuracy of all stations improved significantly, with R² increasing by 0.075–0.289, mean absolute error decreasing by 0.012–0.057 m³/m³, and root mean square error decreasing by 0.013–0.075 m³/m³, with particularly pronounced improvements in deeper layers. This study highlights the necessity of correcting automatic soil moisture observations and provides an effective framework for the correction.

1 Introduction

Soil moisture strongly influences evapotranspiration, surface temperature, precipitation feedback, and plant physiological status, and therefore regulates land–atmosphere interactions, hydrological processes, vegetation dynamics, and ecosystem functioning (Entekhabi, 1995; Vogel et al., 2018; Li Q. et al., 2022; Huang et al., 2025). As a fundamental component of the terrestrial water and energy balance, soil moisture plays a critical role in hydrology, ecology, agriculture, the carbon cycle, and climate-related processeschange (Engman, 1991; Vereecken et al., 2008; Van Der Molen et al., 2011; Liu et al., 2015; Singh et al., 2021; Furtak and Wolińska, 2023; Lv et al., 2025)^. Accurate and continuous soil moisture information is of fundamental importance, particularly in water-limited ecosystems (O’Donnell and Manier, 2022; Bessenbacher et al., 2023; Duarte and Hernandez, 2024).

Ground-based observation is the most reliable source of soil moisture information, including ground measurements and automatic measurements. Ground sampling offer high accuracy measurements, but its labor-intensive limits its suitability for high temporal frequency monitoring. As a comparison, automatic soil moisture stations can provide high-frequency, automated and standardized monitoring, and have been widely deployed in recent years. Despite these advantages, automatic soil moisture measurements are subject to certain biases due to sensor, installation and environmental conditions, resulting in systematic differences between automatic and manually measured soil moisture. Bogena et al. (2017) indicated that EC-5 automatic soil moisture measurements exhibited large inter-sensor variability and non-negligible systematic errors. Datta et al. (2018) found in field experiments in Oklahoma that the measurement accuracy of five commercial soil moisture sensors varied across soils with different salinity and clay content, and that the accuracy also differed depending on the method used to determine soil moisture thresholds. Zemni et al. (2019) found that 5 TE automatic soil mositure measurements without site-specific calibration showed substantial bias in the Jemna oasis, Tunisia.

Common soil moisture sensors typically estimate soil moisture indirectly by measuring the soil’s dielectric constant or electrical resistivity. Due to the influence of soil physical and chemical properties, sensor characteristics, and surrounding environmental conditions, the sensor readings often exhibit linear or nonlinear deviations from the actual soil moisture. By establishing a linear or nonlinear relationship between the sensor measurements and the true soil water content, soil moisture data can be effectively calibrated. To get reliable automatic soil moisture datasets, several studies have focused on the correction of these observations, using ground measurements as reference. Bogena et al. (2017) developed regression-based calibration equations to correct SMT100 soil moisture measurements against manual ground data. Patrignani et al. (2022) corrected the CS655 and CS650 soil moisture measurements using linear regression models. Chen et al. (2019) used multivariate adaptive regression splines (MARS) and Gaussian process regression (GPR) to correct frequency-domain reflectometry (FDR) soil moisture sensors. Ruszczak and Boguszewska-Mańkowska (2022) using an enemble learning algorithm to improve the moisture observation accuracy of 10HS sensors. Li B. et al. (2022) employed both a linear calibration model (LCM) and a universal calibration model (CCM) based on soil properties to calibrate the 5TM capacitive soil moisture sensor. Setiawan et al. (2023) used a third-order polynomial regression model to calibrate a capacitive soil moisture sensor. Adla et al. (2024) evaluated several regression models and machine learning models to correct SM100 sensor soil moisture measurements. Abdelmoneim et al. (2025) developed a polynomial calibration function to calibrate a low-cost capacitive soil moisture sensor (SEN0193, DFRobot). These correction studies mostly rely on single-factor approaches, using only a single regression or machine learning model and focused on correcting soil moisture at only one depth, while largely neglecting depth-dependent biases. Systematic correction studies of automatic soil moisture station data remain very limited, and there is a lack of integrated calibration approaches that effectively combine manual observations, environmental factors, and multi-model algorithms.

This study aimed to validate soil moisture observations from automatic meteorological stations using manually measured data, and to develop an ensemble-based layer-wise model to correct the automatic observations across different soil depths. Such corrections are expected to improve the reliability of soil moisture monitoring, supporting the development of high-precision, long-term soil moisture observation networks essential for climate, agricultural, and ecological applications.

2 Study site and data

2.1 Study site

The Xilinhot National Climate Observatory is located in the central Xilingol Grassland of Inner Mongolia, China (43.95°N, 116.12°E; elevation: 1,124 m), and serves as one of the national benchmark climate stations (Figure 1). The Xilingol region experiences a typical temperate continental climate with limited precipitation, with an annual rainfall of approximately 150–400 mm and annual evaporation ranging from 1,500 to 2,700 mm (Ye et al., 2021), indicating a pronounced water deficit characteristic of semi-arid regions. With long-term and continuous observational records, the station is representative of the climatic features of the typical temperate grassland ecosystem. The station is equipped with comprehensive observation systems, providing essential data support for regional climate change research and climate services. Grassland ecosystems are highly sensitive to soil moisture variability, as soil moisture directly affects vegetation growth, evapotranspiration, and overall ecosystem functioning (Su et al., 2020). Therefore, studying the calibration of soil moisture measurements from automatic stations at the Xilinhot National Climate Observatory is of significant importance.

Figure 1

Figure 1. Location of the xilinhot national climate observatory.

2.2 Ground measured soil moisture data

Manual ground measured soil moisture data from the Xilinhot National Climate Observatory, covering the period from 18 May 2019 to 21 August 2025, were used to validate the soil moisture observations from automatic meteorological stations. Soil samples were collected at five depth intervals (0–10 cm, 10–20 cm, 20–30 cm, 30–40 cm, and 40–50 cm). The gravimetric method was employed to determine the soil gravimetric water content at each depth. Manual ground soil moisture data were collected annually from March 28 to October 28, with daily sampling performed once on the 8th, 18th, and 28th of each month.

2.3 Automatic soil moisture data

The core device of the automatic soil moisture monitoring station is the DZN2 automatic soil moisture sensor (hereafter referred to as DZN2). Based on the FDR principle, the sensor measures changes in the frequency of electromagnetic waves emitted by the probe as they pass through media with different dielectric constants (Skierucha and Wilczek, 2010; Rasheed et al., 2022), and calculates the soil volumetric water content (VWC) using a mathematical model. The instrument features high resolution, up to 0.001 m³/m³, and measurement accuracy of ±0.02 m³/m³ VWC, covering a full range from 0 to soil saturation. It supports depth-profile measurements with 10 cm intervals and can monitor up to 16 soil layers. This system can be used to establish soil moisture monitoring networks, enabling real-time monitoring and dynamic analysis of soil moisture across different sites. In this study, the VWC data collected by the DZN2 span from 16 April 2020, to 23 August 2025, covering five soil layers at depths of 0–10 cm, 10–20 cm, 20–30 cm, 30–40 cm, and 40–50 cm, with measurements recorded daily.

The CRS-2000C regional soil moisture measurement system (hereafter referred to as CRS-2000C) is a professional instrument designed for mesoscale soil moisture monitoring, with its core measurement device being the COSMOS-001. The system indirectly estimates soil water content by detecting the concentration of fast neutrons produced near the ground during the moderation of cosmic rays. It can measure soil moisture to a maximum depth of approximately 70 cm, covering the full range from 0 to saturation. The system achieves a relative soil moisture measurement accuracy of ±0.03 m³/m³ and enables continuous and stable monitoring of soil moisture over large spatial scales. The VWC data obtained from this system span from 22 July 2022 to 23 August 2025. At the Xilinhot station, measurements were taken for the 0–50 cm soil layer, with a sampling interval of 30 min.

The soil temperature and moisture monitoring system (hereafter referred to as 5TM) primarily employs the 5TM sensor (Decagon Devices, United States) to measure soil VWC at depths of 0–50 cm. The 5TM sensor utilizes capacitance/frequency-domain technology to determine the soil dielectric constant, enabling accurate estimation of soil moisture. Operating at a frequency of 70 MHz, the sensor minimizes the influence of soil texture and salinity, ensuring reliable measurements across a wide range of soil types. It provides high measurement accuracy, with VWC precision of ±0.0008 m3/m3 within the 0–0.5 m3/m3 VWC range, and supports a measurement range of 0–1 m3/m3 VWC. The sensor also responds rapidly to dynamic changes in soil moisture, making it suitable for long-term, continuous monitoring. Additionally, the 5TM sensor features a compact and robust design, facilitating long-term field deployment and efficient data acquisition. The VWC data used in this study were collected by the 5TM from 18 May 2019, to 27 October 2024, at three soil layers: 0–10 cm, 10–20 cm, and 20–40 cm, with measurements recorded daily. The photographs of soil moisture monitoring instruments are shown in Figure 2.

Figure 2

Figure 2. Photographs of soil moisture monitoring instruments: ((a) DZN2; (b) CRS-2000C; (c) 5TM).

2.4 Other data

Daily meteorological data from the Xilinhot station, including near-surface air temperature, wind speed, relative humidity, evaporation, and precipitation, were obtained for 1 May 2019 to 21 August 2025. The data were automatically recorded by a DZZ4 automatic weather station.

This study utilized the Google Earth Engine (GEE) platform to acquire the Normalized Difference Vegetation Index (NDVI) data from the Moderate Resolution Imaging Spectroradiometer (MODIS) sensors. Specifically, we used the 16-day composite MOD13Q1 product, which provides NDVI at a spatial resolution of 250 m and a temporal resolution of 16 days (Huete et al., 2002). NDVI values corresponding to the locations of the meteorological stations were extracted for the period from 1 May 2019, to 21 August 2025.

2.5 Data preprocessing

To ensure the reliability and usability of the soil moisture data, this study implemented strict quality control measures (González-Rouco et al., 2001; Hubbard et al., 2005; Dorigo et al., 2013). First, a reasonable threshold range was set, limiting the VWC to 0 m3/m3–0.6 m3/m3 to remove clearly erroneous or invalid values. Second, based on meteorological principles, soil moisture variations were compared with precipitation patterns to maintain temporal consistency; that is, soil moisture should not exhibit significant fluctuations during periods without precipitation, and obvious outliers were manually removed. Data showing no variation for 10 consecutive days were also considered invalid. Finally, to further identify and eliminate outliers, the 3σ rule was applied for outlier detection and removal, thereby maximizing the accuracy and representativeness of the data.

Moreover, to ensure the reliability of the NDVI data, quality control was performed on the MOD13Q1 NDVI product in this study. Specifically, MOD13Q1 provides pixel-level quality information (SummaryQA) to indicate data reliability. During processing, pixels affected by clouds, cloud shadows, aerosol contamination, or low-quality observations were removed, thereby retaining only high-quality pixels (SummaryQA = 0) for direct use. In addition, inspection confirmed that there were no missing NDVI values at the meteorological station locations, and thus no interpolation was required.

3 Methods

3.1 Automatic soil moisture data validation

The accuracy of the automatic station soil moisture data was evaluated based on the manually observed soil moisture measurements. Ideally, manual ground and automatic soil moisture measurements should be matched at the same location, time, and depth. However, in practice, the manual ground sampling sites and the automatic station were not exactly at the same location, but both were situated within the same grassland area, where soil moisture conditions can be considered comparable. Since the automatic station records fewer soil layers, the manual ground observed multi-layer data were averaged to correspond to the station’s measurement depths. It should be noted that the manual ground measurements were originally recorded as gravimetric soil water content. To match the automatic station data, they were converted to volumetric water content using the corresponding soil bulk density for each layer. The formula is shown in Equation 1:

${\theta }_v={\theta }_g\times {\rho }_b(1)$

where θ_v is the volumetric water content (m3/m3), θ_g is the gravimetric water content (kg/kg), ρ_b is the soil bulk density (g/cm³).

For comparative analysis, the manual ground measured volumetric soil moisture was matched with the automatic station measurements at the same dates and soil layer depths, and line charts were plotted accordingly. The temporal variation patterns of both datasets were then compared.

Based on the paired dataset, the accuracy of volumetric soil water content measurements from the automatic stations was evaluated using the determination coefficient (R²), mean absolute error (MAE), root mean square error (RMSE), and mean bias (MB). The corresponding formulas are shown in Equations 2–5:

$R^2=1-\frac{{{\sum }_{i=1}^n\left(y_i-{\widehat{y}}_i\right)}^2}{{\sum }_{i=1}^n{\left(y_i-\overline{y}\right)}^2}(2)$

$MAE=\frac{1}{n}{\sum }_{i=1}^n\left|y_i-x_i\right|(3)$

$RMSE=\sqrt{\frac{1}{\mathrm{n}}{\displaystyle \sum _{\mathrm{i}=1}^{\mathrm{n}}}{\left(y_i-x_i\right)}^2}(4)$

$MB=\frac{1}{n}{\sum }_{i=1}^n\left(y_i-x_i\right)(5)$

where x_i is the manual measured volumetric soil water content, y_i is the automatic station measurements, x̄ and ȳ are their average values, and n is the number of paired samples.

3.2 Automatic soil moisture data correction

The VWC recorded by automatic soil moisture stations was corrected against manual ground measurements. VWC is often influenced by surrounding environmental conditions such as vegetation cover, soil texture, and microclimatic variability. Moreover, its relationship with near-surface meteorological factors (e.g., mean air temperature, precipitation, wind speed, and evapotranspiration) is inherently complex and nonlinear. Machine learning models can automatically capture these nonlinear relationships, thereby producing predictions that more closely reflect actual ground measurements. By incorporating auxiliary variables such as vegetation indices and meteorological data, the models are also able to account for systematic biases in automatic station measurements arising from local environmental heterogeneity. Different algorithms exhibit distinct strengths in capturing feature interactions and error patterns; therefore, to improve the accuracy of the automatic station measurements, multiple machine learning methods were employed for model fitting, including Cubist, Random Forest, XGBoost, CatBoost, and further developed a linear ensemble model based on these approaches.

The Random Forest algorithm, proposed by Breiman, is an ensemble learning method that employs decision trees as base learners (Breiman, 2001; Salman et al., 2024). It builds upon the Bagging framework by introducing random feature selection during the construction of each decision tree. The final classification or regression outcome is determined through majority voting or averaging across all trees. Random Forest is characterized by its simplicity, robustness, and reduced tendency to overfit.

The Cubist algorithm is a rule-based regression model developed as an extension of Quinlan’s M5 model tree. At each leaf node of the Cubist decision tree, a multivariate linear regressionmodel is fitted to the subset of data covered by the corresponding rule set (Chen et al., 2020). By combining decision trees with local linear regression, Cubist retains the interpretability of rule-based partitions while enhancing local fitting capability, making it particularly suitable for small-to medium-sized datasets.

XGBoost, proposed by Chen, is a gradient boosted decision tree (GBDT) algorithm (Chen and Guestrin, 2016). It builds an ensemble of decision trees sequentially, with each new tree fitted to the residuals of the previous trees, thereby progressively improving predictive accuracy. The algorithm is computationally efficient and can effectively handle large-scale datasets.

CatBoost, developed by Yandex, is another gradient boosting decision tree algorithm (Dorogush et al., 2018). It employs symmetric trees as the base structure and uses feature-splitting preprocessing to handle categorical data. In combination with an ordered boosting strategy, it effectively mitigates gradient bias and prediction shift issues common in gradient boosting (Cai et al., 2024). CatBoost requires fewer hyperparameters, is user-friendly, and is well suited for complex datasets while reducing the risk of overfitting (Fu et al., 2024)^.

To further improve predictive accuracy and fully leverage the strengths of individual models, this study developed an ensemble model based on a generalized additive model (GAM) (Hastie and Tibshirani, 1995). The GAM flexibly characterizes the nonlinear relationships between input features and the response variable through smooth functions. Specifically, the predictions of the four base models were used as new input features to train a second-level GAM, which learns the optimal combination of the base model outputs. Compared with individual models, this ensemble approach achieves a secondary optimization of the prediction results, effectively reducing both bias and variance and significantly improving the robustness and overall accuracy of the predictions (Ganaie et al., 2022; Mienye and Sun, 2022)^.

The appropriate selection of variables plays a critical role in ensuring the accuracy and stability of the model. In this study, meteorological, vegetation, and soil-related variables were selected as inputs for the automatic soil moisture correction model according to physical considerations and previous studies, including near-surface air temperature, surface temperature, wind speed, relative humidity, evapotranspiration, precipitation, atmospheric pressure, NDVI, and the soil moisture of adjacent upper layers as model input variables (Hide, 1954; Seneviratne et al., 2010; Trenberth, 2011; Vicente-Serrano et al., 2010; Wang et al., 2007; Zhang et al., 2018; Froidevaux et al., 2014; Wu et al., 2025; Elmotawakkil et al., 2025). These variables characterize atmospheric forcing, vegetation conditions, and soil state that jointly influence soil moisture variability, and were used to construct the model training dataset to develop four base models.

Precipitation events not only directly replenish soil moisture but also exert a sustained influence on VWC over the following days, resulting in a clear lagged response in soil moisture dynamics. Similarly, evaporation continuously depletes soil water, producing a cumulative effect on VWC in subsequent days. Therefore, considering only the precipitation or evaporation on a single day may fail to accurately capture the true soil moisture dynamics when analyzing the water balance and its driving factors. To address this, we introduced a time-weighting approach that accounts for the effects of precipitation and evaporation over the preceding 10 days on the current-day VWC. A linearly decreasing weighting scheme was applied (Equations 6, 7), giving progressively greater weight to more recent days so that their influence on soil moisture is emphasized. This approach allows a more comprehensive characterization of soil moisture responses to water inputs and losses, effectively capturing the delayed effects of water replenishment and depletion and improving the model’s ability to simulate the spatiotemporal dynamics of soil moisture.

$X_t^{*}={\displaystyle \sum _{i=1}^{10}}{w}_i·X_{t-i}(6)$

where $X_t^{*}$ is the weighted cumulative precipitation (or evaporation) over the preceding 10 days for day t; $X_{t-i}$ is the daily precipitation (or evaporation) on the ith day prior to day t; $w_i$ is the weight coefficient assigned to the ith day, calculated as follows:

$w_i=\frac{11-i}{{\sum }_{j=1}^{10}\left(11-j\right)}(7)$

During the modeling process, this study employed RandomizedSearchCV (Bergstra and Bengio, 2012) to optimize the hyperparameters of four base learners (Random Forest, Cubist, XGBoost, and CatBoost) with the aim of achieving optimal model performance. Parameter search spaces were constructed according to the structural characteristics of each model. For the Cubist model, the key parameters included neighbors, n_rules, and n_committees. The Random Forest model primarily tuned n_estimators, max_depth, min_samples_split, min_samples_leaf, and max_features. The XGBoost model focused on n_estimators, max_depth, learning_rate, subsample, colsample_bytree, and min_child_weight. The CatBoost model optimized core parameters such as iterations, depth, and learning_rate. All parameter search spaces were explored using RandomizedSearchCV with random sampling and five-fold cross-validation to identify the optimal parameter combinations. The detailed search ranges are presented in Table 1.

Table 1

Table 1. Parameter search space settings for the base learners.

This study used manually measured soil moisture as the dependent variable and constructed the training dataset with independent variables including the automatic station measurements for the same layer, near-surface air temperature, wind speed, relative humidity, the weighted cumulative precipitation over the past 10 days, the weighted cumulative evaporation over the past 10 days, the calibrated soil moisture from the adjacent upper layer (for deeper layers), and the NDVI index. Based on this training dataset, base models were constructed using machine learning algorithms such as RF, Cubist, XGBoost, and CatBoost, which were then integrated using a GAM model. Five-fold cross-validation was employed to evaluate model performance. Specifically, the entire dataset was randomly divided into five subsets. In each iteration, one subset was used as the validation set (without replacement), while the remaining four subsets served as the training set, and this process was repeated five times. Model performance was assessed using four metrics: the R2, MAE, RMSE, and MB.

4 Results

4.1 Validation results

Figure 3 present the comparison between manual measurements and the DZN2, the CRS-2000C, and the 5TM sensor. Overall, the sensor data follow similar trends as the manual ground measurements, although they exhibit fluctuations around the ground values, showing a certain degree of deviation.

Figure 3

Figure 3. Time series of soil VWC measured by the DZN2, CRS-2000C, 5TM, and manual ground observations. ((a) 0–10 cm; (b) 10–20 cm; (c) 20–30 cm; (d) 30–40 cm; (e) 40–50 cm; (f) 0–50 cm; (g) 0–10 cm; (h) 10–20 cm; (i) 20–40 cm).

Figure 4 show scatter plots between manual ground observations and the DZN2 (a-e), CRS-2000C (f), and 5TM sensor (g-i), respectively. For the DZN2, the R2 values in the 0–10 cm, 10–20 cm, 20–30 cm, 30–40 cm, and 40–50 cm soil layers were 0.418, 0.313, 0.218, 0.148, and 0.179, respectively, with a max–min difference of 0.270. The corresponding MAE values were 0.053, 0.057, 0.056, 0.058, and 0.062 m3/m3, with a max–min difference of 0.009 m3/m3. The RMSE values were 0.068, 0.069, 0.070, 0.075, and 0.079, with a max–min difference of 0.011 m3/m3. The MB values were −0.026, −0.023, 0.013, 0.022, and 0.037 m3/m3, with a max–min difference of 0.063 m3/m3. These results indicate that the agreement between DZN2 and manual ground observations was highest at the 0–10 cm depth, but declined notably with increasing soil depth. Moreover, automatic stations tended to underestimate soil water content in the shallow layers while overestimating it in the deeper layers.

Figure 4

Figure 4. Scatter plots of soil VWC measured by the DZN2, CRS-2000C, and 5TM sensors versus manual ground observations. ((a) 0–10 cm; (b) 10–20 cm; (c) 20–30 cm; (d) 30–40 cm; (e) 40–50 cm; (f) 0–50 cm; (g) 0–10 cm; (h) 10–20 cm; (i) 20–40 cm).

For the 0–50 cm soil layer, CRS-2000C measurements had a R2 of 0.774, a MAE of 0.027 m3/m3, a RMSE of 0.033 m3/m3, and a MB of 0.008 m3/m3 compared with manual ground observations. These results indicate that the CRS-2000C slightly overestimated VWC, particularly under higher soil moisture conditions, but achieved the highest measurement accuracy among the three systems.

For the 0–10 cm, 10–20 cm, and 20–40 cm soil layers, the 5TM sensor showed R2 of 0.673, 0.535, and 0.200, respectively, when compared with manual ground measurements. The MAE were 0.039, 0.032, and 0.076 m3/m3, the RMSE were 0.048, 0.041, and 0.099 m3/m3, and the MB were −0.027, −0.013, and 0.072 m3/m3, respectively. The 5TM sensor exhibited higher accuracy in shallow soil layers and lower accuracy in deeper layers. The measurement bias varied with depth: the system slightly underestimated VWC in the g and h layers, whereas it overestimated VWC in the 20–40 cm soil layer.

4.2 Correction results

Figure 5 present scatter plots comparing the corrected measurements from the DZN2 (a-e), CRS-200C (f), and 5TM (g-i) systems with the manual ground observations. As shown in Figure 5, after correction, the VWC measured by the DZN2 across the 0–10 cm, 10–20 cm, 20–30 cm, 30–40 cm, and 40–50 cm soil layers exhibits R2 values of 0.683, 0.507, 0.382, 0.437, and 0.294, respectively. The corresponding MAE values are 0.031, 0.027, 0.024, 0.020, and 0.018 m3/m3, while the RMSE values are 0.037, 0.036, 0.031, 0.026, and 0.024 m3/m3. The MB of all soil layers is close to zero. Overall, the correlations for all five layers improved substantially, with the largest increase in R2 reaching 0.289 (30–40 cm soil layer), indicating that the correction markedly enhanced the consistency between DZN2 and manual ground observations. Compared with the uncorrected data, the error metrics also decreased significantly. The maximum reduction in MAE reached 0.044 m3/m3 (40–50 cm soil layer), while the maximum decrease in RMSE reached 0.055 m3/m3 (40–50 cm soil layer). These results demonstrate that the correction effectively improved the measurement accuracy of the DZN2 across all soil depths, with particularly notable improvements in the deeper layers.

Figure 5

Figure 5. Scatter plots of the corrected soil VWC measured by the DZN2, CRS-2000C, and 5TM sensors versus manual ground observations. (a) 0–10 cm; (b) 10–20 cm; (c) 20–30 cm; (d) 30–40 cm; (e) 40–50 cm; (f) 0–50 cm; (g) 0–10 cm; (h) 10–20 cm; (i) 20–40 cm.

As shown in Figure 5, the corrected R² of the CRS-2000C increased to 0.849, representing an improvement of 0.075 compared with the uncorrected results. Meanwhile, the MAE and RMSE were reduced to 0.027 m³/m³ and 0.033 m³/m³, decreasing by 0.0143 m³/m³ and 0.0167 m³/m³, respectively, with the MB approaching zero. The These results indicate that the correction method significantly improved the measurement accuracy of the CRS-2000C and enhanced its agreement with manual ground observations.

As shown in Figures 5g–i, after correction, the 5TM achieved R² values of 0.800, 0.683, and 0.447 for the for the 0–10 cm, 10–20 cm, and 20–40 cm soil layers, respectively, representing improvements of 0.127, 0.148, and 0.247 compared with the uncorrected results. The corresponding MAE were 0.022, 0.020, and 0.019 m³/m³, decreased by 0.0167, 0.0121, and 0.0572 m³/m³, respectively. The RMSE values were reduced to 0.030, 0.028, and 0.024 m³/m³, corresponding to reductions of 0.018, 0.013, and 0.075 m³/m³, respectively. Overall, the correction substantially improved the consistency between 5TM measurements and manual ^ground observations, with notably greater enhancements in the deeper soil layer.

5 Discussion

In recent years, soil moisture monitoring has attracted increasing attention, and automated soil moisture sensors have become widely deployed. However, due to differences in device types, sensor accuracy, and installation conditions, the quality of automatic observations varies considerably. Existing studies have shown that measurements from automatic stations can deviate substantially from manual ground observations, and the results of this study further confirm this issue, particularly at deeper soil layers where the discrepancies become even more pronounced. Given the close relevance of soil moisture data to hydrological, ecological, and climate-related research, it is essential to calibrate measurements from automatic stations to improve their accuracy and reliability.

Existing correction methods for automatic soil moisture observations still have limitations. Wang et al. (2023) applied a Random Forest approach to calibrate 5 TE capacitance-based soil moisture sensors, achieving a post-calibration RMSE of 0.05 m3/m3, compared to 0.07 m3/m3 when using the modified Topp equation, corresponding to an error reduction of approximately 28.6%. Wang et al. (2023) employed a station-based linear calibration for 5TM sensors, reducing the average RMSE across 17 sites from 0.049 m3/m3 before calibration to 0.027 m3/m3 after calibration, representing a decrease of about 44.9%. Adla et al. (Adla et al., 2024) applied multiple least-squares and machine learning models to calibrate Spectrum Inc. SM100 sensors, achieving a post-calibration RMSE of 0.031 m3/m3, compared with 0.046 m3/m3 prior to calibration, corresponding to an error reduction of roughly 31.5%. These studies collectively demonstrate that regression and machine learning–based calibration methods can substantially improve the accuracy of automatic soil moisture sensors, although the degree of improvement varies and certain limitations remain. Most studies rely on a single regression model or a single machine-learning algorithm (Bogena et al., 2017; Patrignani et al., 2022; Li B. et al., 2022), which is insufficient to capture the complexity of soil moisture biases and thus constrains both correction accuracy and model robustness (Rowlandson et al., 2013). Some studies did not systematically account for surrounding environmental factors (e.g., temperature, pressure, vegetation cover, surface energy balance) when developing calibration models (Vaz et al., 2013; Nagahage et al., 2019; Qi et al., 2024). As a result, they were unable to correct the complex bias mechanisms jointly driven by atmospheric conditions, vegetation dynamics, and soil hydrothermal processes. In addition, prior research has predominantly focused on the correction of a single soil layer, overlooking the limited representativeness of surface environmental factors for deeper layers (Holzman et al., 2017; Mane et al., 2024). Surface soil moisture responds rapidly to precipitation and evapotranspiration changes, whereas deeper soil moisture exhibits pronounced lag effects and is more susceptible to sensor response delays and soil heterogeneity, making deep-layer bias correction particularly challenging. In response to the above issues, this study develops an ensemble correction framework that integrates multiple machine-learning algorithms, including Cubist, Random Forest, XGBoost, and CatBoost, to enhance overall correction performance and model stability. Furthermore, a hierarchical correction strategy is proposed. For deep-layer soil moisture correction, the framework fully leverages the inter-layer correlations by incorporating the corrected soil moisture from the upper layer as an auxiliary predictor for the lower layer, progressively constructing a layered correction system. This approach enables optimized correction of automatic soil moisture observations across different depths and substantially improves the consistency and accuracy of the correction results.

To further investigate the roles of input variables in soil moisture estimation models, this study employed the Permutation Importance method (Altmann et al., 2010), using MAE as the metric to evaluate each variable’s contribution to prediction accuracy. Figure 6 presents the variable importance for the four models applied to soil moisture data from three automatic stations. The results show that the moisture content of the immediately overlying soil layer is significantly more important than other variables, indicating that the models heavily rely on this feature. NDVI ranks second in importance, particularly for the CRS-2000C dataset, which contains only a single 0–50 cm soil layer, where NDVI becomes the most critical predictor. NDVI reflects vegetation cover and evapotranspiration, which directly affect soil moisture. Atmospheric pressure also indirectly influences soil moisture by regulating soil and vegetation evapotranspiration, while temperature and precipitation can affect soil moisture dynamics to some extent and improve model performance, although their relative importance is lower. Overall, the models effectively capture the inter-layer relationships and the influence of surface environmental factors on soil moisture.

Figure 6

Figure 6. Importance of input variables in soil moisture estimation across RF, Cubist, XGBoost, and CatBoost models. ((a) DZN2; (b) CRS-2000C; (c) 5TM).

In this study, three types of automatic soil moisture sensors (DZN2, CRS-2000C, and 5TM) were calibrated. To achieve optimal model performance, RandomizedSearchCV was employed to efficiently explore the hyperparameter space and identify the best configuration for each algorithm. During model training, a five-fold cross-validation scheme was applied to assess model accuracy. By randomly dividing the dataset into five subsets and iteratively using each subset as the validation set, this approach ensures that each fold serves as an independent and non-overlapping validation set, thereby enhancing the reliability of model evaluation. After calibration, the DZN2 exhibited a R2 of 0.294–0.683, MAE of 0.018–0.031 m3/m3, and RMSE of 0.024–0.037 m3/m3, corresponding to an increase in R2 of approximately 62%–195%, a reduction in MAE of 41%–70%, and a decrease in RMSE of 53%–65% relative to pre-calibration values. For CRS-2000C, R2 increased to 0.849, MAE decreased to 0.013 m3/m3, and RMSE decreased to 0.016 m3/m3, representing a 9.7% improvement in R2 and roughly a 51% reduction in both MAE and RMSE. The 5TM sensor showed post-calibration R2 values of 0.447–0.800, MAE of 0.019–0.022 m3/m3, and RMSE of 0.024–0.030 m3/m3, corresponding to increases in R2 of 18.9%–123.5%, reductions in MAE of 38%–75%, and decreases in RMSE of 32%–76%. Overall, the calibration method significantly enhanced the measurement accuracy of all three automatic soil moisture sensors and improved their agreement with ground-based observations, with the most pronounced improvements observed in deeper soil layers.

This study also has certain limitations. First, the research area is restricted to the Xilinhot National Climate Observatory, which represents a typical semi-arid region, and does not encompass diverse climate zones or soil types. Second, the manual ground measurements used for calibration were collected at relatively low sampling frequency, resulting in a limited sample size and allowing only partial calibration of the automatic soil moisture observations. Future work will expand the study to broader regions and longer time periods, covering multiple climate zones, vegetation types, and soil conditions, in order to develop calibration models that more comprehensively capture environmental variability and long-term sensor performance. In addition, more soil-related variables will be incorporated into the calibration framework to further enhance its accuracy and robustness, thereby providing stronger technical support for the high-quality application of automatic soil moisture observations.

6 Conclusion

Automatic soil moisture observation stations provide continuous, high-precision point measurements of soil moisture, serving as a critical reference for land–atmosphere interactions, hydrological processes, and vegetation dynamics. Accurate soil moisture information is of great importance for environmental and climate studies. In this study, manual soil moisture measurements were used as a reference, and multiple meteorological and environmental factors were employed as input variables. A GAM was used to integrate Cubist, Random Forest, XGBoost, and CatBoost to perform layer-wise modeling and correction for the DZN2, CRS-2000C, and 5TM sensors. After correction, the MAE of DZN2, CRS-2000C, and 5TM decreased by approximately 50%–67%, 51%, and 38%–75%, respectively, while the RMSE decreased by approximately 53%–65%, 51%, and 32%–76%, respectively, compared with the uncorrected data. Overall, the measurement accuracy of all stations improved significantly, with the most pronounced improvements observed in deeper soil layers. This study provides high-precision and reliable ground-based soil moisture data, which can support drought monitoring, soil moisture product validation, and hydrological and climate research.

Data availability statement

The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding author.

Author contributions

HL: Investigation, Data curation, Writing – original draft, Methodology, Conceptualization, Funding acquisition. YL: Formal Analysis, Visualization, Validation, Writing – original draft, Data curation. MJ: Writing – review and editing, Conceptualization, Resources, Methodology, Supervision. CX: Writing – review and editing, Data curation. YX: Resources, Funding acquisition, Writing – review and editing. CW: Resources, Writing – review and editing. MY: Writing – review and editing, Investigation. FC: Investigation, Writing – review and editing. WZ: Data curation, Writing – review and editing.

Funding

The author(s) declared that financial support was received for this work and/or its publication. This research was funded by the Scientific Experiment Foundation of Inner Mongolia Meteorological Bureau (nmqxkxsy202411), the Common Application Support Platform for National Civil Space Infrastructure “13th Five-Year Plan” Land Observation Satellites (2017–000052-73-01–001735), the Scientific and Technological Innovation Foundation of Inner Mongolia Meteorological Bureau (nmqxkjcx202421), and the Graduate and Innovation Projects of Jiangsu Province (KYCX25_1622).

Conflict of interest

The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Generative ai statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

References

Abdelmoneim, A. A., Al Kalaany, C. M., Khadra, R., Derardja, B., and Dragonetti, G. (2025). Calibration of low-cost capacitive soil moisture sensors for irrigation management applications. Sensors 25, 343. doi:10.3390/s25020343

PubMed Abstract | CrossRef Full Text | Google Scholar

Adla, S., Bruckmaier, F., Arias-Rodriguez, L. F., Tripathi, S., Pande, S., and Disse, M. (2024). Impact of calibrating a low-cost capacitance-based soil moisture sensor on AquaCrop model performance. J. Environ. Manag. 353, 120248. doi:10.1016/j.jenvman.2024.120248

PubMed Abstract | CrossRef Full Text | Google Scholar

Altmann, A., Toloşi, L., Sander, O., and Lengauer, T. (2010). Permutation importance: a corrected feature importance measure. Bioinformatics 26, 1340–1347. doi:10.1093/bioinformatics/btq134

PubMed Abstract | CrossRef Full Text | Google Scholar

Bergstra, J., and Bengio, Y. (2012). Random search for hyper-parameter optimization. J. Mach. Learn. Res. 13, 281–305. doi:10.5555/2503308.2188395

CrossRef Full Text | Google Scholar

Bessenbacher, V., Gudmundsson, L., and Seneviratne, S. I. (2023). Optimizing soil moisture station networks for future climates. Geophys. Res. Lett. 50, e2022GL101667. doi:10.1029/2022GL101667

CrossRef Full Text | Google Scholar

Bogena, H., Huisman, J., Schilling, B., Weuthen, A., and Vereecken, H. (2017). Effective calibration of low-cost soil water content sensors. Sensors 17, 208. doi:10.3390/s17010208

PubMed Abstract | CrossRef Full Text | Google Scholar

Breiman, L. (2001). Random forests. Mach. Learn. 45, 5–32. doi:10.1023/A:1010933404324

CrossRef Full Text | Google Scholar

Cai, Y., Yuan, Y., and Zhou, A. (2024). Predictive slope stability early warning model based on CatBoost. Sci. Rep. 14, 25727. doi:10.1038/s41598-024-77058-6

PubMed Abstract | CrossRef Full Text | Google Scholar

Chen, T., and Guestrin, C. (2016). “XGBoost: a scalable tree boosting system,” in Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, San Francisco California USA: ACM (IEEE), 785–794.

Google Scholar

Chen, L., Zhangzhong, L., Zheng, W., Yu, J., Wang, Z., Wang, L., et al. (2019). Data-driven calibration of soil moisture sensor considering impacts of temperature: a case study on FDR sensors. Sensors 19, 4381. doi:10.3390/s19204381

PubMed Abstract | CrossRef Full Text | Google Scholar

Chen, Q., Yang, X., Ouyang, Z., Zhao, N., Jiang, Q., Ye, T., et al. (2020). Estimation of anthropogenic heat emissions in China using cubist with points-of-interest and multisource remote sensing data. Environ. Pollut. 266, 115183. doi:10.1016/j.envpol.2020.115183

PubMed Abstract | CrossRef Full Text | Google Scholar

Datta, S., Taghvaeian, S., Ochsner, T., Moriasi, D., Gowda, P., and Steiner, J. (2018). Performance assessment of five different soil moisture sensors under irrigated field conditions in Oklahoma. Sensors 18, 3786. doi:10.3390/s18113786

PubMed Abstract | CrossRef Full Text | Google Scholar

Dorigo, W. A., Xaver, A., Vreugdenhil, M., Gruber, A., Hegyiová, A., Sanchis-Dufau, A. D., et al. (2013). Global automated quality control of in situ soil moisture data from the international soil moisture network. Vadose Zone J. 12, 1–21. doi:10.2136/vzj2012.0097

CrossRef Full Text | Google Scholar

Dorogush, A. V., Ershov, V., and Gulin, A. (2018). CatBoost: gradient boosting with categorical features support. doi:10.48550/ARXIV.1810.11363

CrossRef Full Text | Google Scholar

Duarte, E., and Hernandez, A. (2024). A review on soil moisture dynamics monitoring in semi-arid ecosystems: methods, techniques, and tools applied at different scales. Appl. Sci. 14, 7677. doi:10.3390/app14177677

CrossRef Full Text | Google Scholar

Elmotawakkil, A., Sadiki, A., Moumane, A., Jaldi, S., Batchi, M., Karkouri, J. A., et al. (2025). Machine learning techniques for soil moisture prediction in arid and semi-arid regions: a case study of Morocco. Intell. Geoengin. 2, 251–261. doi:10.1016/j.ige.2025.11.001

CrossRef Full Text | Google Scholar

Engman, E. T. (1991). Applications of microwave remote sensing of soil moisture for water resources and agriculture. Remote Sens. Environ. 35, 213–226. doi:10.1016/0034-4257(91)90013-V

CrossRef Full Text | Google Scholar

Entekhabi, D. (1995). Recent advances in land-atmosphere interaction research. Rev. Geophys. 33, 995–1003. doi:10.1029/95RG01163

CrossRef Full Text | Google Scholar

Froidevaux, P., Schlemmer, L., Schmidli, J., Langhans, W., and Schär, C. (2014). Influence of the background wind on the local soil moisture–precipitation feedback. J. Atmos. Sci. 71, 782–799. doi:10.1175/JAS-D-13-0180.1

CrossRef Full Text | Google Scholar

Fu, P., Zhang, J., Yuan, Z., Feng, J., Zhang, Y., Meng, F., et al. (2024). Estimating the heavy metal contents in entisols from a mining area based on improved spectral indices and catboost. Sensors 24, 1492. doi:10.3390/s24051492

PubMed Abstract | CrossRef Full Text | Google Scholar

Furtak, K., and Wolińska, A. (2023). The impact of extreme weather events as a consequence of climate change on the soil moisture and on the quality of the soil environment and agriculture – a review. CATENA 231, 107378. doi:10.1016/j.catena.2023.107378

CrossRef Full Text | Google Scholar

Ganaie, M. A., Hu, M., Malik, A. K., Tanveer, M., and Suganthan, P. N. (2022). Ensemble deep learning: a review. Eng. Appl. Artif. Intell. 115, 105151. doi:10.1016/j.engappai.2022.105151

CrossRef Full Text | Google Scholar

González-Rouco, J. F., Jiménez, J. L., Quesada, V., and Valero, F. (2001). Quality control and homogeneity of precipitation data in the southwest of Europe. J. Clim. 14, 964–978. doi:10.1175/1520-0442(2001)014%3C0964:QCAHOP%3E2.0.CO;2

CrossRef Full Text | Google Scholar

Hastie, T., and Tibshirani, R. (1995). Generalized additive models for medical research. Stat. Methods Med. Res. 4, 187–196. doi:10.1177/096228029500400302

PubMed Abstract | CrossRef Full Text | Google Scholar

Hide, J. C. (1954). Observations on factors influencing the evaporation of soil moisture. Soil Sci. Soc Amer J 18, 234–239. doi:10.2136/sssaj1954.03615995001800030002x

CrossRef Full Text | Google Scholar

Holzman, M., Rivas, R., Carmona, F., and Niclòs, R. (2017). A method for soil moisture probes calibration and validation of satellite estimates. MethodsX 4, 243–249. doi:10.1016/j.mex.2017.07.004

PubMed Abstract | CrossRef Full Text | Google Scholar

Huang, X., Liang, S., Ziegler, A. D., and Zeng, Z. (2025). Decoupling vegetation and soil-moisture interaction in evapotranspiration interannual variability. iScience 28, 113008. doi:10.1016/j.isci.2025.113008

PubMed Abstract | CrossRef Full Text | Google Scholar

Hubbard, K. G., Goddard, S., Sorensen, W. D., Wells, N., and Osugi, T. T. (2005). Performance of quality assurance procedures for an applied climate information system. J. Atmos. Ocean. Technol. 22, 105–112. doi:10.1175/JTECH-1657.1

CrossRef Full Text | Google Scholar

Huete, A., Didan, K., Miura, T., Rodriguez, E. P., Gao, X., and Ferreira, L. G. (2002). Overview of the radiometric and biophysical performance of the MODIS vegetation indices. Remote Sens. Environ. 83, 195–213. doi:10.1016/S0034-4257(02)00096-2

CrossRef Full Text | Google Scholar

Li, B., Wang, C., Gu, X., Zhou, X., Ma, M., Li, L., et al. (2022). Accuracy calibration and evaluation of capacitance-based soil moisture sensors for a variety of soil properties. Agric. Water Manag. 273, 107913. doi:10.1016/j.agwat.2022.107913

CrossRef Full Text | Google Scholar

Li, Q., Ye, A., Zhang, Y., and Zhou, J. (2022). The peer-to-peer type propagation from meteorological drought to soil moisture drought occurs in areas with strong land-atmosphere interaction. Water Resour. Res. 58, e2022WR032846. doi:10.1029/2022WR032846

CrossRef Full Text | Google Scholar

Liu, Y., Pan, Z., Zhuang, Q., Miralles, D. G., Teuling, A. J., Zhang, T., et al. (2015). Agriculture intensifies soil moisture decline in northern China. Sci. Rep. 5, 11261. doi:10.1038/srep11261

PubMed Abstract | CrossRef Full Text | Google Scholar

Lv, Y., Kuang, J., Li, R., and Shi, Z. (2025). Moisture fluctuations: an important but overlooked factor in the carbon cycle. J. Environ. Manag. 393, 127075. doi:10.1016/j.jenvman.2025.127075

PubMed Abstract | CrossRef Full Text | Google Scholar

Mane, S., Das, N., Singh, G., Cosh, M., and Dong, Y. (2024). Advancements in dielectric soil moisture sensor calibration: a comprehensive review of methods and techniques. Comput. Electron. Agric. 218, 108686. doi:10.1016/j.compag.2024.108686

CrossRef Full Text | Google Scholar

Mienye, I. D., and Sun, Y. (2022). A survey of ensemble learning: concepts, algorithms, applications, and prospects. IEEE Access 10, 99129–99149. doi:10.1109/ACCESS.2022.3207287

CrossRef Full Text | Google Scholar

Nagahage, E. A. A. D., Nagahage, I. S. P., and Fujino, T. (2019). Calibration and validation of a low-cost capacitive moisture sensor to integrate the automated soil moisture monitoring system. Agriculture 9, 141. doi:10.3390/agriculture9070141

CrossRef Full Text | Google Scholar

O’Donnell, M. S., and Manier, D. J. (2022). Spatial estimates of soil moisture for understanding ecological potential and risk: a case study for arid and semi-arid ecosystems. Land 11, 1856. doi:10.3390/land11101856

CrossRef Full Text | Google Scholar

Patrignani, A., Ochsner, T. E., Feng, L., Dyer, D., and Rossini, P. R. (2022). Calibration and validation of soil water reflectometers. Vadose Zone J. 21, e20190. doi:10.1002/vzj2.20190

CrossRef Full Text | Google Scholar

Qi, Q., Yang, H., Zhou, Q., Han, X., Jia, Z., Jiang, Y., et al. (2024). Performance of soil moisture sensors at different salinity levels: comparative analysis and calibration. Sensors 24, 6323. doi:10.3390/s24196323

PubMed Abstract | CrossRef Full Text | Google Scholar

Rasheed, M. W., Tang, J., Sarwar, A., Shah, S., Saddique, N., Khan, M. U., et al. (2022). Soil moisture measuring techniques and factors affecting the moisture dynamics: a comprehensive review. Sustainability 14, 11538. doi:10.3390/su141811538

CrossRef Full Text | Google Scholar

Rowlandson, T. L., Berg, A. A., Bullock, P. R., Ojo, E. R., McNairn, H., Wiseman, G., et al. (2013). Evaluation of several calibration procedures for a portable soil moisture sensor. J. Hydrology 498, 335–344. doi:10.1016/j.jhydrol.2013.05.021

CrossRef Full Text | Google Scholar

Ruszczak, B., and Boguszewska-Mańkowska, D. (2022). Soil moisture a posteriori measurements enhancement using ensemble learning. Sensors 22, 4591. doi:10.3390/s22124591

PubMed Abstract | CrossRef Full Text | Google Scholar

Salman, H. A., Kalakech, A., and Steiti, A. (2024). Random forest algorithm overview. Babylon. J. Mach. Learn. 2024, 69–79. doi:10.58496/BJML/2024/007

CrossRef Full Text | Google Scholar

Seneviratne, S. I., Corti, T., Davin, E. L., Hirschi, M., Jaeger, E. B., Lehner, I., et al. (2010). Investigating soil moisture–climate interactions in a changing climate: A review. Earth-Sci. Rev. 99, 125–161. doi:10.1016/j.earscirev.2010.02.004

CrossRef Full Text | Google Scholar

Setiawan, I., Musa, M. D.Th., and Putri, S. A. (2023). Re-Calibration of model-based capacitive sensor for IoT soil moisture measurements. JAIC 7, 150–155. doi:10.30871/jaic.v7i2.6809

CrossRef Full Text | Google Scholar

Singh, S., Mayes, M. A., Shekoofa, A., Kivlin, S. N., Bansal, S., and Jagadamma, S. (2021). Soil organic carbon cycling in response to simulated soil moisture variation under field conditions. Sci. Rep. 11, 10841. doi:10.1038/s41598-021-90359-4

PubMed Abstract | CrossRef Full Text | Google Scholar

Skierucha, W., and Wilczek, A. (2010). A FDR sensor for measuring complex soil dielectric permittivity in the 10–500 MHz frequency range. Sensors 10, 3314–3329. doi:10.3390/s100403314

PubMed Abstract | CrossRef Full Text | Google Scholar

Su, F., Wang, F., Li, Z., Wei, Y., Li, S., Bai, T., et al. (2020). Predominant role of soil moisture in regulating the response of ecosystem carbon fluxes to global change factors in a semi-arid grassland on the loess Plateau. Sci. Total Environ. 738, 139746. doi:10.1016/j.scitotenv.2020.139746

PubMed Abstract | CrossRef Full Text | Google Scholar

Trenberth, K. (2011). Changes in precipitation with climate change. Clim. Res. 47, 123–138. doi:10.3354/cr00953

CrossRef Full Text | Google Scholar

Van Der Molen, M. K., Dolman, A. J., Ciais, P., Eglin, T., Gobron, N., Law, B. E., et al. (2011). Drought and ecosystem carbon cycling. Agric. For. Meteorology 151, 765–773. doi:10.1016/j.agrformet.2011.01.018

CrossRef Full Text | Google Scholar

Vaz, C. M. P., Jones, S., Meding, M., and Tuller, M. (2013). Evaluation of standard calibration functions for eight electromagnetic soil moisture sensors. Vadose Zone J. 12, 1–16. doi:10.2136/vzj2012.0160

CrossRef Full Text | Google Scholar

Vereecken, H., Huisman, J. A., Bogena, H., Vanderborght, J., Vrugt, J. A., and Hopmans, J. W. (2008). On the value of soil moisture measurements in vadose zone hydrology: a review. Water Resour. Res. 44, 2008WR006829. doi:10.1029/2008WR006829

CrossRef Full Text | Google Scholar

Vicente-Serrano, S. M., Beguería, S., and López-Moreno, J. I. (2010). A Multiscalar Drought Index Sensitive to Global Warming: The Standardized Precipitation Evapotranspiration Index. J. Clim. 23, 1696–1718. doi:10.1175/2009JCLI2909.1

CrossRef Full Text | Google Scholar

Vogel, M. M., Zscheischler, J., and Seneviratne, S. I. (2018). Varying soil moisture–atmosphere feedbacks explain divergent temperature extremes and precipitation projections in central Europe. Earth Syst. Dynam. 9, 1107–1125. doi:10.5194/esd-9-1107-2018

CrossRef Full Text | Google Scholar

Wang, C., Gu, X., Zhou, X., Yang, J., Yu, T., Tao, Z., et al. (2023). Chinese soil moisture observation network and time series data set for high resolution satellite applications. Sci. Data 10, 424. doi:10.1038/s41597-023-02234-8

PubMed Abstract | CrossRef Full Text | Google Scholar

Wang, X., Xie, H., Guan, H., and Zhou, X. (2007). Different responses of MODIS-derived NDVI to root-zone soil moisture in semi-arid and humid regions. J. Hydrol. 340, 12–24. doi:10.1016/j.jhydrol.2007.03.022

CrossRef Full Text | Google Scholar

Wu, H., Fu, C., Zhang, L., Mekonnen, Z. A., Zhu, Q., Yu, K., et al. (2025). Significant sensitivity of global vegetation productivity to terrestrial surface wind speed changes. Nat. Commun. 16, 9315. doi:10.1038/s41467-025-65000-x

PubMed Abstract | CrossRef Full Text | Google Scholar

Ye, J., Hu, Y., Zhen, L., Wang, H., and Zhang, Y. (2021). Analysis on land-use change and its driving mechanism in xilingol, China, during 2000–2020 using the google Earth engine. Remote Sens. 13, 5134. doi:10.3390/rs13245134

CrossRef Full Text | Google Scholar

Zhang, H., Chang, J., Zhang, L., Wang, Y., Li, Y., and Wang, X. (2018). NDVI dynamic changes and their relationship with meteorological factors and soil moisture. Environ. Earth Sci. 77, 582. doi:10.1007/s12665-018-7759-x

CrossRef Full Text | Google Scholar

Zemni, N., Bouksila, F., Persson, M., Slama, F., Berndtsson, R., and Bouhlila, R. (2019). Laboratory calibration and field validation of soil water content and salinity measurements using the 5TE sensor. Sensors 19, 5272. doi:10.3390/s19235272

PubMed Abstract | CrossRef Full Text | Google Scholar