Monitoring Soil Moisture Content Research Articles

Monitoring Soybean Soil Moisture Content Based on UAV Multispectral and Thermal-Infrared Remote-Sensing Information Fusion.

By integrating the thermal characteristics from thermal-infrared remote sensing with the physiological and structural information of vegetation revealed by multispectral remote sensing, a more comprehensive assessment of the crop soil-moisture-status response can be achieved. In this study, multispectral and thermal-infrared remote-sensing data, along with soil-moisture-content (SMC) samples (0~20 cm, 20~40 cm, and 40~60 cm soil layers), were collected during the flowering stage of soybean. Data sources included vegetation indices, texture features, texture indices, and thermal-infrared vegetation indices. Spectral parameters with a significant correlation level (p < 0.01) were selected and input into the model as single- and fuse-input variables. Three machine learning methods, eXtreme Gradient Boosting (XGBoost), Random Forest (RF), and Genetic Algorithm-optimized Backpropagation Neural Network (GA-BP), were utilized to construct prediction models for soybean SMC based on the fusion of UAV multispectral and thermal-infrared remote-sensing information. The results indicated that among the single-input variables, the vegetation indices (VIs) derived from multispectral sensors had the optimal accuracy for monitoring SMC in different soil layers under soybean cultivation. The prediction accuracy was the lowest when using single-texture information, while the combination of texture feature values into new texture indices significantly improved the performance of estimating SMC. The fusion of vegetation indices (VIs), texture indices (TIs), and thermal-infrared vegetation indices (TVIs) provided a better prediction of soybean SMC. The optimal prediction model for SMC in different soil layers under soybean cultivation was constructed based on the input combination of VIs + TIs + TVIs, and XGBoost was identified as the preferred method for soybean SMC monitoring and modeling, with its R2 = 0.780, RMSE = 0.437%, and MRE = 1.667% in predicting 0~20 cm SMC. In summary, the fusion of UAV multispectral and thermal-infrared remote-sensing information has good application value in predicting SMC in different soil layers under soybean cultivation. This study can provide technical support for precise management of soybean soil moisture status using the UAV platform.

Sustainable forest development in the face of global warming: Optimizing Robinia pseudoacacia L. stands to alleviate soil moisture depletion

To foster sustainable forest development amid global warming, it is imperative to enhance the quality of existing forest stands while avoiding excessive depletion of soil moisture. This study established 24 Robinia pseudoacacia L. standard plots (40 × 40 m2) in the Caijiachuan watershed, located in the gully region of the Loess Plateau, Shanxi Province, China, to monitor soil moisture content from 2007 to 2021. Additionally, the study examined the influence of 19 ecological factors, including climate, topography, and stand characteristics, on soil moisture. The findings indicated a sustained decline in soil moisture storage across the 0–200 cm soil profile, with significant depletion at 0–140 cm. The soil moisture content at depths of 80–200 cm was critically low, resulting in a “dry soil layer” during years with low precipitation. Climatic factors (average annual temperature and annual precipitation) and stand factors (uniform angle index, stand density, litter thickness, and Simpson diversity index of the shrubs) emerged as dominant factors impacting soil moisture variations. An increase in average annual temperature due to global warming, compounded by suboptimal stand density, primarily drove soil moisture depletion. Using a multivariate optimization equation to model the relationship between dominant factors and soil moisture storage, this study recommends a dynamic stand management strategy. This strategy involves selective thinning or replanting of Robinia pseudoacacia L. trees, targeting a gradual reduction in optimal stand density from 1500 to 1000 plants·ha−1 over the next two decades. Additionally, transitioning stand spatial arrangements from uniform or clustered to random distributions, maintaining ideal litter thickness (2.43–2.47 cm), and enhancing shrub diversity are advisable. Such measures can alleviate soil moisture deficits in Robinia pseudoacacia L. stands attributable to global warming, thereby promoting the Loess Plateau's sustainable forestry advancement.

Monitoring soil moisture content in the root zone of winter wheat with multi-angle multispectral imagery

ABSTRACT Exploring the impact of different monitoring angles from unmanned aerial vehicle (UAV) on the monitoring accuracy of soil moisture content (SMC) is crucial for precision irrigation. To this end, experiment was conducted to monitor the SMC of winter wheat at different growth stages under different irrigation treatments in Yangling, Shaanxi, China. At a solar zenith angle of 45°, multispectral remote sensing data from a UAV were collected at thirteen different monitoring zenith angles. Simultaneously, the SMC in the wheat root zone was measured. From the UAV multispectral images, spectral reflectance was extracted for the construction of vegetation indices. Then the correlation between the vegetation indices and the measured SMC was analyzed. With the vegetation indices as input variables, SMC monitoring models were constructed using Extreme Learning Machine (ELM), Random Forest (RF), and Back Propagation Neural Network (BPNN). The study also examined the effect of specific angles (hotspot and dark spot angles) on the estimation accuracy of the SMC at the nadir angle. The results indicated that different monitoring angles significantly impact the SMC estimation accuracy. Band reflectance and vegetation indices exhibited significant peak values and angular effects at the monitoring zenith angle of 45°. The models achieved the optimal inversion accuracy at the hotspot angle (at a monitoring zenith angle of 45° in the solar principal plane), and the accuracy was ranked as follows: BPNN (R2 = 0.71; RMSE = 1.69) > ELM (R2 = 0.52; RMSE = 1.94) > RF (R2 = 0.48; RMSE = 2.10). By eliminating shadows at the nadir angle through the threshold of dark spot, inversion accuracy similar to the hotspot direction is achieved. This study provides a basis for appropriate selection of UAV flight angles for the monitoring of SMC in the root zone of winter wheat.

Enhancing precision of root-zone soil moisture content prediction in a kiwifruit orchard using UAV multi-spectral image features and ensemble learning

Accurate and real-time monitoring of soil moisture content (SMC) is of utmost importance for effective field irrigation and maximizing crop water productivity. However, a comprehensive investigation into the inversion study for determining suitable combinations of unmanned aerial vehicle (UAV) image features and enhancing the precision of SMC model prediction has yet to be fully validated within a kiwifruit orchard setting. This study addresses this gap by employing a pre-processing method and an optimal band combination algorithm to assess the impact of various combinations of kiwifruit canopy reflectance and fraction vegetation coverage (FVC) features on the sensitivity of root-zone SMC. Furthermore, an optimal ensemble learning (EL) framework was developed to monitor SMC at various root-zone depths (0–10 cm [SMC10], 0–20 cm [SMC20], 0–30 cm [SMC30], 0–40 cm [SMC40], 0–50 cm [SMC50], 0–60 cm [SMC60]). The key findings of this research highlight the successful derivation of 10 wavebands and FVC features, exhibiting a strong correlation with SMC at different root depths. The gradient boosting (GBDT) model demonstrated the exceptional accuracy in estimating SMC10, with an impressive R2 value of 0.963 ± 0.030 and low RMSE values of 0.238 ± 0.111. Similarly, the eXtreme Gradient Boosting (XGBoost) model outperformed in estimating SMC20 to SMC60, with R2 and RMSE values of 0.963 ± 0.024 and 0.117 ± 0.053, respectively. Additionally, the utilization of the optimal EL model allows for digital mapping of SMC at different depths across fruit growth stages, showcasing superior adaptability for SMC30 to SMC60 (with R2 and RMSE of 0.782 ± 0.090 and 0.037 ± 0.011) compared to SMC10 and SMC20 (with R2 and RMSE of 0.765 ± 0.097 and 0.056 ± 0.024). These results underscore the potential of the EL estimation framework in characterizing the spatial distribution of root-zone SMC at the individual kiwifruit plant level.

Complement time-series UAV spectral data based on Ambrals kernel-driven model to monitor soil moisture content

ABSTRACT Continuous time-series spectral data are important for inversion of crop or soil information. UAV remote sensing is usually selected under clear and windless weather conditions, but it is not possible to have such weather every day, which results in the UAV not collecting continuous daily spectral information. To explore this issue, we focused on summer maize with four irrigation levels as the research subject. A UAV platform with a multispectral sensor was used to acquire measured spectra of the maize canopy. The solar zenith angle was calculated and substituted into the Ambrals kernel-driven model to obtain simulated spectral data for the maize canopy, and the time-series UAV spectral data were complemented. Then, four vegetation indices (VIs) were established using simulated and measured spectral data, respectively, and the accuracy of the simulated VIs was evaluated. Finally, the simulated and measured VIs were used to monitor and evaluate variations in soil surface moisture content, respectively, and provide irrigation warning. The results demonstrated that Ambrals kernel-driven model can be used to simulate the reflectance of maize canopy collected by UAV. The simulated reflectance can complement time-series UAV spectral data and be used to establish VIs, among which Structure Intensive Pigment Index (SIPI) was established with the highest accuracy (R = 0.729). The VIs established by simulated reflectance can be used to monitor soil surface moisture content, and the monitoring effect is similar to the measured data (R2 = 0.642, RMSE = 0.42). It can evaluate the soil moisture a few days after irrigation and ensure the continuity and timeliness of soil moisture data, so as to improve the crop irrigation system and carry out irrigation warning. These results have certain reference for the supplementation of time-series spectral data and farmland irrigation using UAV multispectral remote sensing.

More Precise Monitoring of Soil Moisture Content in Agricultural Fields by Upscaling Conversion of Multispectral Image Data From Unmanned Aerial Vehicles

More Precise Monitoring of Soil Moisture Content in Agricultural Fields by Upscaling Conversion of Multispectral Image Data From Unmanned Aerial Vehicles

Inversion method for soil moisture content based on a distributed fiber optic acoustic sensing system.

The traditional measurement method can't achieve real-time monitoring of soil moisture content (SMC) within a two-dimensional area. To solve the above problems, we propose a rapid SMC monitoring method for two-dimensional area based on distributed acoustic sensing (DAS). DAS demodulates the backward Rayleigh scattering signal containing seismic wave sound velocity information from the active seismic source. The folding ruler approximation is employed to calculate the sound velocity of the soil, which is then inverted to determine the soil moisture content. The experiment measured the soil within a two-dimensional area formed by the seismic source and the acoustic sensing optical cable. The sensing optical cable and the active seismic source are organized into a two-dimensional area and the measurement range is 3 × 10 m with 33 points. The SMC ranges from 15% to 40%. The experiment shows that the absolute error between the measured values obtained by DAS and the water cut meter is 7%. This experiment verifies the feasibility of using the Rayleigh scattering properties to invert SMC and provides a new method for real-time monitoring of SMC in a large area.

Research on robust inversion model of soil moisture content based on GF-1 satellite remote sensing

This study aims to improve the accuracy and applicability of traditional linear regression and machine learning algorithms for monitoring soil moisture content by satellite remote sensing. Shahaoqu Experimental Station within Jiefangzha Irrigation District of Hetao Irrigation Area is selected as the study area. The dataset includes GF-1 satellite remote sensing images and measured soil moisture content. The study employs the best subset selection to determine the combination of sensitive variables for soil moisture content at different soil depths. Additionally, robust regression theory is introduced to traditional linear regression and machine learning algorithms to construct soil moisture content inversion models at different soil depths. The results show that all three improved robust algorithms effectively reduce the influence of outliers, improve the accuracy and stability of these inversion models. The improved robust nonlinear algorithm outperforms the improved robust linear algorithm. Compared with the unprocessed algorithm, at the soil depth of 0–20 cm, Robust Extreme Learning Machine (RELM) algorithm is the most optimal, with the validation set's R2adj increasing from 0.555 to 0.696 (approximately 25.33%) and the RMSE decreasing from 0.025 to 0.018 (approximately 26.54%). Robust Least Squares Support Vector Machine (RLSSVM) algorithm is the second best, with the validation set's R2adj increasing from 0.502 to 0.641 (approximately 27.75%) and the RMSE decreasing from 0.026 to 0.020 (approximately 24.59%). Robust Linear Regression (LMS-RLS) algorithm is relatively the worst, with the validation set's R2adj increasing from 0.491 to 0.659 (approximately 34.05%) and the RMSE decreasing from 0.026 to 0.019 (approximately 27.21%). The sensitivity of the five soil depths to soil moisture content follows the order of 0–20 cm, 20–40 cm, 0–60 cm, 0–40 cm, and 40–60 cm. The improved robust regression algorithms can improve the accuracy and applicability of soil moisture content estimation and provide a reference for monitoring soil moisture content using GF-1 satellite.

Infiltration and Failure Behavior of an Unsaturated Soil Slope under Artificial Rainfall Model Experiments

Slope disasters often include soil erosion and shallow landslides. These types of slope failure can occur in unsaturated soil. In this study, artificial rainfall is applied to a compacted soil slope in a laboratory to investigate the effects of rainfall intensity, initial moisture content and relative compaction on infiltration and failure behavior of an unsaturated soil slope subjected to rainfall infiltration. Soil moisture probes installed in the soil slope were used to monitor soil moisture content during the experiment. Test results show that the soil saturation at the toe of the slope was observed to be higher than other areas of the slope following the onset of rainfall. Gradually, the saturation advanced towards the crest, resulting in the formation of a wetting band along the slope surface. With increasing rainfall duration, the wetting band progressed further downwards along the slope. The failure of the slope began at the crest and junction of the slope, and the primary cause was soil erosion resulting from high-intensity rainfall. The most significant variable affecting infiltration behavior is relative compaction, followed by the initial moisture content of the compacted slope. During the experiment, the low hydraulic conductivity of the saturated soil caused most of the rainfall on the slope to become surface runoff, which minimized the impact of rainfall intensity on infiltration behavior. Specifically, the hydraulic conductivity of soil compacted at optimal moisture content and 90 R.C. is only 7.041 × 10-5 cm/sec or 2.53 mm/h, which is much lower than rainfall intensities of 80 mm/h and 160 mm/h. In addition, soil saturation contours facilitated visualization and quantification of the infiltration behavior of slopes, enabling a more detailed analysis of experimental results. These results help understand the behavior of unsaturated soil slopes under artificial rainfall conditions and aid in designing effective slope stabilization measures to prevent slope failure and minimize the risk of landslides.

Hyperspectral response and monitoring study of soil moisture content based on the optimized spectral index

AbstractSoil moisture controls the exchange of energy between the land surface and the atmosphere and significantly affects plant growth and productivity. Hyperspectral (350–2500 nm) monitoring of soil moisture content (SMC) could provide a theoretical basis for the real‐time estimation of spatial and temporal variations in soil moisture. During this study, we have developed SMC prediction models in different soil moisture ranges by using the combination of spectral preprocessing methods and optimized spectral indices. Results indicated that first‐order derivative (FD) preprocessing method can highlight the effective information of the spectra and improve the correlation between spectral reflectance and SMC. The spectral reflectance exhibited a strong correlation with SMC near 1400, 1700, 1900, and 2200 nm. The FD‐NDSI‐W0‐model had the best SMC prediction and applicability (R2 = 0.977, root mean square error = 3.413%, relative percent deviation = 6.171) to predict SMCs. Overall, pretreatment methods, combined with the two‐band random combination to optimize the spectral index to build a prediction model of SMC, could be applied for accurate monitoring of SMC.

Multilayer Model in Soil Moisture Content Retrieval Using GNSS Interferometric Reflectometry.

The global navigation satellite system-interferometric reflectometry (GNSS-IR) was developed more than a decade ago to monitor soil moisture content (SMC); a system that is essentially finished has emerged. The standard GNSS-IR model typically considers soil to be a single layer of medium and measures the average SMC between 1 and 10 cm below the soil surface. The majority of the SMC is not distributed uniformly along the longitudinal axis. This study is based on a simulation platform and suggests a SMC-stratified measurement model that can be used to recover the SMC at different depths in the sink and reverse osmosis to address the issue that conventional techniques cannot accurately measure soil moisture at different depths. The soil moisture of each layer was assessed by utilizing the GNSS signals reflected by various soil layers, and this study employed total transmission when the vertical linearly polarized component of the electromagnetic wave was conveyed by the GNSS signal reflected by the soil. This work employed the Hilbert transform to obtain the interference signal envelope, which increases the visibility of the interference signal's "notch" and reduces the burr impact of the interference signal brought on by ambient noise. The accuracy of the SMC measurement at the bottom declines due to the soil's attenuation of the GNSS signal power, but the correlation between the predetermined value and SMC retrieved by the GNSS-IR multilayer SMC measurement model similarly approached 0.92.

Evaluating soil moisture content under maize coverage using UAV multimodal data by machine learning algorithms

Timely and accurate estimation of soil moisture content (SMC) is essential for precise irrigation management at the farm scale. Unmanned aerial vehicle (UAV) remote sensing with a high spatiotemporal resolution has become a promising method for SMC monitoring. Many existing SMC models have only been tested at a specific crop growth stage using a single type of sensor, and the effects of growth stage and irrigation variation on SMC estimation accuracy remain unclear. To address these limitations, this study used UAV-based multimodal data to quantify SMC in a maize field under various levels of irrigation over two years using three machine learning algorithms (MLA): partial least squares regression (PLSR), K nearest neighbor (KNN), and random forest regression (RFR). The results demonstrated that multimodal data fusion improves the SMC estimation accuracy regardless of the MLA, especially the joint use of thermal and multispectral data. Among three SMC regression models, the RFR model produced the most accurate SMC estimation for the two growing seasons regardless of sensor combinations. The RFR model using all three types of data generated the most accurate and robust SMC estimation at the vegetative stage with R2 of 0.68 and 0.78, and rRMSE of 20.82% and 19.36% for 10- and 20-cm soil depths, respectively; it also produced the best SMC estimation accuracy under well-watered and mild to modest deficit irrigation treatments for both soil depths. The study shows that the high spatial–temporal maps of SMC using UAV-based multimodal data has promising potential for supporting decision-making in irrigation scheduling at the farmland scale.

Development of Soil Moisture Content Monitoring System for Precision Measurement of Soil Moisture in Sub-Optimal Land for Palm Oil Plantation

Indonesia has 15 million hectares of palm oil plantation, making it one of the world’s greatest palm oil and CPO producers. However, despite the large area of palm oil plantations, the production is not yet optimal. Soil moisture and environmental factors are the leading cause. Thus, a technology suited to monitor soil moisture, soil temperature, and microclimates needs to be developed. This research aims to develop a Soil Moisture Content (SMC) Monitoring System that can be used to monitor soil moisture, supported by soil temperature and microclimates variable and also operates automatically and in real-time. The device is also integrated with a cloud server. The implemented SMC was installed on the palm oil plantation of PT Kerry Sawit Indonesia, Wilmar International Plantation. On top of that, the SMC device can measure soil moisture up to three variations of depth. The acquired data was saved and sent to a cloud server. Then, a user can access the data to execute early water deficit anticipation. The resulting device is capable of measuring soil moisture accurately, which can be proven with these parameters, R square value of 0.9782, RMSE of 1.2%, and MAPE of 2.6%.

IoT based smart irrigation system for communal use

Kenya is experiencing a rapid human population growth and consequently rising demand for food. This is happening at the backdrop of adverse climatic changes currently being experienced. The problem is further aggravated by the fact that more than 80% of the country’s landmass is classified as Arid and Semi-Arid Lands (ASALs). In order to meet the food demand, there is need to divorce from rain-fed agriculture and put most of the arable land under irrigation. Since Kenya is a water scarce country, efficient irrigation systems are required. An efficient irrigation system that promotes use of rain water harvesting technologies and maximizes crop water productivity through use of appropriate sensors and wireless communication technology has been developed. The innovation involves use of farmers’ earth pans for harvesting and storing rain water and irrigating fields through drip lines. In addition, smart soil moisture sensors are fitted to each field under irrigation for real-time monitoring of soil moisture content and automatically notifying an irrigation administrator through mobile phone when crops require watering. The administrator is equipped with a tricycle fitted with solar powered water pumping system for pumping water from the earth pan to the crops. When the operator receives a signal request from water sensors in field he/she moves with the tri-cycle and services it. The smart sensors (IoT based) installed in the field will then notify him/her when sufficient amount of water has been received so that irrigation can be stopped. The operator waits for the next request from another farmer’s field to service. The technology has been tested in different portions of the university land under irrigation. The results indicate that crops receive timely watering and show no water stress. The innovation has the potential of making high tech irrigation system accessible to small scale farmers at an affordable cost and enhance food production and in turn food security in the country.

Verification on the mode of moisture transfer in the vadose zone of thick loess

AbstractPrecipitation is the only source of groundwater recharge in the Chinese loess region. However, rainfall has been the main driver of geological disasters and engineering challenges in China, such as landslides, slope collapse and sinking of foundations. The roles of preferential flow and piston flow in moisture transfer in the thick loess vadose zone remain controversial. This study used a 100‐m thick loess shaft to monitor soil moisture content and soil temperature over 5 years, whereas climate indices were monitored on the Earth surface. The soil water characteristic curves and hydraulic conductivity function of the loess and the palaeosol collected in the profile were measured in a laboratory. The soil profile moisture transfer process was simulated based on the climatic boundary conditions and measured hydraulic parameters. The monitored and simulated results suggest that loess moisture transfer is dominated by piston flow. The thick loess vadose zone above the groundwater level could be divided into two sub‐zones, namely the active and steady zones. The active zone has a thickness of ~7 m and shows a direct response to climate change, with soil moisture content regulated by rainfall infiltration, evaporation and transpiration and soil moisture moving as transient flow. Although preferential flow can move in the active zone along joints, fissures and holes when water accumulates in the lower ground during heavy rainfall, piston flow remains the main mode of water movement. The steady zone has a thickness of &lt;7 m and is above the groundwater level. Soil moisture content and soil temperature in the steady zone show temporal stability and soil moisture moves as a steady flow. Piston flow is the only mode of water movement in the steady zone. Since percolation velocity in the steady zone is estimated to be 73.3 mm/year, 1330 years of rainfall infiltration through the 97.5 m thick vadose zone would be required to recharge groundwater.

Inversion model of surface bare soil temperature and water content based on UAV thermal infrared remote sensing

Soil moisture content is important to agricultural monitoring, ecological restoration and landslide warning. Traditional methods cannot monitor soil moisture content in large area, high precision and continuous data. Unmanned aerial vehicle (UAV) remote sensing monitoring has enabled wide-scale water content monitoring, improving the spatial resolution accuracy to centimeter level and the temporal resolution accuracy to minute level, which makes UAV remote sensing favorable for soil moisture content monitoring. In this paper, to achieve rapid and accurate inversion data of bare surface soil moisture content via UAV thermal infrared remote sensing, UAV is used to obtain a visible light map of the experiment area and a thermal infrared radiation map at 9:00 am, 1:00 pm, and 6:00 pm in a given day. Gray value of thermal infrared radiation is based on Planck’s blackbody radiation law and quadratic polynomial fitting to invert temperature data; the ground soil temperature and humidity sensor provides sample areas of the measured temperature and soil moisture content. The linear relationship between gray value of thermal infrared radiation, brightness temperature, measured temperature, and soil moisture content was obtained by analyzing the data. The experimental results show that there is a good linear relationship between the brightness temperature and water content, and the brightness t temperature R2 at 6 PM is better than the measured one. Finally, a model for calculating the surface bare soil moisture content based on brightness temperature is established. RMSE and RPD were used to verify the accuracy of the inversion model. The model accuracy verification results show that the inversion model at 6 pm has a good accuracy, which verifies the feasibility and accuracy of unmanned aerial vehicle thermal infrared remote sensing monitoring of surface bare soil moisture content.

Estimating soil moisture content under grassland with hyperspectral data using radiative transfer modelling and machine learning.

The monitoring of soil moisture content (SMC) at very high spatial resolution (<10m) using unmanned aerial systems (UAS) is of high interest for precision agriculture and the validation of large scale SMC products. Data-driven approaches are the most common method to retrieve SMC with UAS-borne data at water limited sites over non-disturbed agricultural crops. A major disadvantage of data-driven algorithms is the limited transferability in space and time and the need of a high number of ground reference samples. Physically-based approaches are less dependent on the amount of samples and are transferable in space and time. This study explores the potential of (1) a hybrid method targeting the soil brightness factor of the PROSAIL model using a variational heteroscedastic Gaussian Processes regression (VHGPR) algorithm, and (2) a data-driven method employing VHGPR for the retrieval of SMC over three grassland sites based on UAS-borne VIS-NIR (399-1001 nm) hyperspectral data. The sites were managed by mowing (Fendt), grazing (Grosses Bruch) and irrigation (Marquardt). With these distinct local pre-conditions we aimed to identify factors that favor and limit the retrieval of SMC. The hybrid approach presented encouraging results in Marquardt (RMSE = 1.5 Vol_%, R2 = 0.2). At the permanent grassland sites (Fendt, Grosses Bruch) the thatch layer jeopardized the application of the hybrid model. We identified the complex canopy structure of grassland as the main factor impacting the hybrid SMC retrieval. The data-driven approach showed high accuracy for Fendt (R2 = 0.84, RMSE = 8.66) and Marquardt (R2 = 0.4, RMSE = 10.52). All data-driven models build on the LAI-SMC relationship. However, this relationship was hampered by mowing (Fendt), leading to a lack of transferability in time. The alteration of plant traits by grazing prevents finding a relationship with SMC in Grosses Bruch. In Marquardt, we identified the timelag between changes in SMC and plant response as the main reason of decrease in model accuracy. Yet, the model performance is accurate in undisturbed and water-limited areas (Marquardt). The analysis points to challenges that need to be tackled in future research and opens the discussion for the development of robust models to retrieve high resolution SMC from UAS-borne remote sensing observations.

In situ mesoscale soil moisture content monitoring based on global navigation satellite system interferometric reflectometry and ensemble modeling

The existing point-scale or large-scale soil moisture content (SMC) monitoring techniques cannot entirely satisfy the requirements of in situ SMC monitoring for agricultural purposes. We proposed a framework for in situ mesoscale SMC monitoring based on global navigation satellite system interferometric reflectometry (GNSS-IR) and ensemble modeling technique. The framework consists of a data collecting and preparing system and an SMC retrieval system. The amplitude, phase, and detection depth, which are the GNSS-IR signal-to-noise ratio parameters, were positively and negatively correlated with the volumetric SMC data, respectively. In the SMC retrieval stage, five kinds of models including linear regression, multilinear regression, k-neighbor regressor, support vector regressor, and random forest (RF) was tested as the first-order model. Two kinds of swarm optimization algorithms including sparrow search algorithm (SSA) and particle swarm optimization (PSO) were examined for models’ hyper-parameter optimization. The results show that the RF performed best and had a coefficient of determination (R2) of 0.798, a root mean square error (RMSE) of 0.043 cm3 / cm3, and a mean absolute error (MAE) of 0.034 cm3 / cm3 for the validation set. Both the SSA and PSO are effective for models’ optimization. After input variable selection, the second-order ensemble RF model outperformed the first-order RF model and had an R2 of 0.819, an RMSE of 0.040 cm3 / cm3, and an MAE of 0.031 cm3 / cm3 for the validation set. The proposed framework is potentially valuable for popularization because of its cost-effectiveness and high accuracy.

Developing a generalized vis-NIR prediction model of soil moisture content using external parameter orthogonalization to reduce the effect of soil type

Real-time soil moisture content (SMC) monitoring is an important parameter in precision agriculture that can be utilized to enhance soil and water management. Visible and near-infrared (vis-NIR) has been proposed as a promising method for SMC monitoring. However, vis-NIR reflectance response to soil moisture is strongly influenced by soil properties such as texture and organic matter content. Thus it is difficult to develop a general prediction model of vis-NIR that can estimate SMC of different soil types. To solve this problem, this study utilizes the External Parameter Orthogonalization (EPO) method to reduce the effect of soil type. Seven soils (CZ, GA, JN, PDS, PY, WF, and XZ) were collected from the North China Plain. A fiber-optic spectrometer (300–1000 nm) was used to obtain spectral response of seven soils adjusted to several SMC levels. The partial least square regression (PLSR) was used to model the pretreated vis-NIR spectra of the seven soils. The modeling results confirmed the feasibility of using vis-NIR spectroscopy to predict SMC on individual soils. However, the prediction models vary significantly between soil types. To obtain a generalized prediction model of SMC that can be applied to different soils, three soils were selected from the seven soils for calibration, while the rest were used for validation. The results showed that direct PLSR modeling failed to achieve a unified prediction model. To solve this problem, EPO was applied to minimize the influence of soil types. The results demonstrated that EPO-PLSR can realize a generalized SMC prediction model. It is suggested that the soils for building EPO-PLSR models should cover a wide span of soil organic matter (SOM) content. Soil with the least SOM content appeared to be best for developing an EPO transfer matrix.

Soil moisture content estimation in winter wheat planting area for multi-source sensing data using CNNR

Rapid and accurate estimation of soil moisture content (SMC) is an important part of precision agriculture, and it is also one of the key problems to be solved in field real-time monitoring and precision irrigation. Most of the existing studies are limited to SMC monitoring of bare soil, which can be obtained by optical remote sensing (e.g., multispectral, hyperspectral) or Synthetic Aperture Radar (SAR). However, for the soil covered by vegetation, such as farmland, there are some theoretical defects with only one of the measuring methods. Meanwhile, in order to break through the limitations of low spatial and temporal resolutions of satellite remote sensing, it is of great significance to study SMC retrieval based on multi-source remote sensing data for the near earth UAV remote sensing systems. Based on this, this paper, taking the winter wheat planting area in Guanzhong plain of China as the research area, combines the advantages of ultra-wideband (UWB) radar, and multispectral remote sensing data, to reduce the influences of vegetation coverage on the estimation accuracy. A one-dimensional regression convolution neural network model is constructed to realize the quantitative prediction and estimation of SMC in farmland. The carried out experiments show that the proposed CNNR model has a better performance than traditional SVR and GRNN models and the R2, RMSE and RPD are 0.7453, 0.0140 cm3/cm3 and 1.9246, respectively. After introducing NDVI, MSAVI and DVI vegetation indices generated from multispectral images, the accuracy of the three models increased significantly. Among the three models, the constructed CNNR model has the best performance, and its R2, RMSE and RPD reach 0.9168, 0.0089 cm3/cm3, and 3.0201. Furthermore, after adding different levels of Gaussian noise to the original radar echoes, the CNNR model constructed in this paper still has the highest prediction accuracy and the strongest noise robustness.