Developing pedotransfer functions using Sentinel-2 satellite spectral indices and Machine learning for estimating the surface soil moisture

Sedaghat, Azadeh; Shahrestani, Mahmoud Shabanpour; Noroozi, Ali Akbar; Nosratabad, Alireza Fallah; Bayat, Hossein

doi:10.1016/j.jhydrol.2021.127423

Cited by 21 publications

(10 citation statements)

References 91 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Comparing the results of this study to other studies, it can be noted that the RF had superior performance compared to the other models forming the ensemble, a result comparable to Adab et al (2020), who utilized terrain-related data and RS data to predict SM in a semi-arid region and found that the RF model exhibited a better performance from the selected models (viz., SVM). Similarly, Sedaghat et al (2022) also found that the RF method was considerably higher than the MLR method in estimating surface SM using pedotransfer functions and Sentinel 2 imagery. Chaudhary et al (2022) reported a similar finding, showing that the RF method was the best during the training stage; however, RF was second out of 12 machine learning methods during testing.…”

Section: Discussionmentioning

confidence: 79%

See 1 more Smart Citation

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

Kibirige,

Gokool,

Mkhize

2023

Vadose Zone Journal

View full text Add to dashboard Cite

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

show abstract

Section: Discussionmentioning

confidence: 79%

“…Similarly, Sedaghat et al. (2022) also found that the RF method was considerably higher than the MLR method in estimating surface SM using pedotransfer functions and Sentinel 2 imagery. Chaudhary et al.…”

Section: Discussionmentioning

confidence: 88%

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

Kibirige,

Gokool,

Mkhize

2023

Vadose Zone Journal

View full text Add to dashboard Cite

show abstract

“…Our study showed that the RF model outperformed the ELM and SVM models, which may be related to the advantages of the RF model itself. A previous study 44 pointed out that, compared with other machine learning models, the RF model in principle had the following main advantages: (1) it basically does not overfit, and its error decreases and converges as the number of trees increases; (2) when dealing with small sample data sets, it can handle unbalanced data and automatically process missing values to achieve more efficient performance; and (3) it is an integrated learning model that can resample and train on this basis.…”

Section: Discussionmentioning

confidence: 93%

Synergistic estimation of soil salinity based on Sentinel-1 image texture and Sentinel-2 salinity spectral indices

Yin

Chen

et al. 2023

J. Appl. Rem. Sens.

View full text Add to dashboard Cite

Soil salinization, one of the important factors leading to global land degradation, seriously affects sustainable agricultural development. Accurate and frequent monitoring of soil salt content (SSC) contributes to the management and restoration of salinized soil. Our study aimed to evaluate the potential to synthetically estimate the salinity in bare soil with different remote sensing sensors (medium-resolution Sentinel-1 and Sentinel-2). The salinity data of 134 surface soil samples were collected in the Shahaoqu irrigation area (SIA) in the Hetao Irrigation District of Inner Mongolia. Simultaneously, the remote sensing images of Sentinel-1 and Sentinel-2 of SIA were obtained. A total of 46 predictors, including 12 remote sensing data (10 multispectral bands, 1 VV, and 1 VH), 8 polarization combination indices, 16 salinity spectral indices, and 10 texture features, were obtained and calculated from the remote sensing image data. Three machine learning algorithms, random forest (RF), support vector machine (SVM), and extreme learning machine (ELM), were used for the construction of soil salinity prediction models based on different combinations of the 46 predictors. The results showed that the remote sensing image data from multiple sensors could be used to estimate soil salinity more accurately than that of a single sensor. Among the three regression models (RF, SVM, and ELM) in our study, RF was the best prediction model (R 2 v ¼ 0.71, RMSE v ¼ 0.140%, and MAE ¼ 0.094%). The RF model combining Sentinel-1 SAR and Sentinel-2 multispectral imagery was the most effective in revealing the spatial distribution of soil salinity. The importance analysis of the predictor variables showed that texture features were the main explanatory variables for SSC prediction, with contrast being the most important predictor. The results of our study verified the potential to predict SSC by combining multisource remote sensing data and machine learning models in arid and semiarid regions, which could provide some reference for soil salinization monitoring and control in the bare soil.

show abstract

“…Rabiei et al [9] employed Sentinel-1 radar and Sentinel-2 multispectral imagery with machine learning algorithms for surface soil moisture estimation, showcasing high accuracy. Sedaghat et al [10] employed spectral indices, RF, and MLR for surface soil moisture estimation, highlighting the superiority of the RF method. Sharma et al [11] examined trends in NDVI and its correlation with LST, SM, and precipitation, emphasizing NDVI's sensitivity to LST.…”

Section: Introductionmentioning

confidence: 99%

Soil Moisture Modelling Using Remote Sensing and Artificial Neural Networks: A Study of Devbhumi Dwarka Region, Gujarat, India

Gojiya,

Karangiya,

Chavda

et al. 2024

IJECC

View full text Add to dashboard Cite

For the arid and semi-arid region of Devbhumi Dwarka in Gujarat, where soil moisture issue is prominent, a GIS-based approach is needed to develop models for estimation of soil moisture. In this study, Landsat and Sentinel data were used to develop multiple soil moisture indices. Using these spectral indices, artificial neural network (ANN) models were developed using actual recorded soil moisture data. Total of 174 samples were collected in the study area of 3,77,731 ha. Values of soil moisture ranged from 3.40 % to 12.50 % with an average value of 8.42 %. The maps of soil moisture indices i.e. LST, NDVI, NDWI, NSDSI3, MI and VSWI were generated on 1:550000 scale using ArcMap software. Moisture index NSDSI3 was highest correlating index. Best ANN model for soil moisture content (SMC) estimation was developed using Sentinel data (8-14-1) with RMSE, R2 and NRMSE values of 0.85 %, 0.73 and 0.10 for training and 1.11 %, 0.54 and 0.13 for testing respectively.

show abstract

Developing pedotransfer functions using Sentinel-2 satellite spectral indices and Machine learning for estimating the surface soil moisture

Cited by 21 publications

References 91 publications

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

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

Synergistic estimation of soil salinity based on Sentinel-1 image texture and Sentinel-2 salinity spectral indices

Soil Moisture Modelling Using Remote Sensing and Artificial Neural Networks: A Study of Devbhumi Dwarka Region, Gujarat, India

Contact Info

Product

Resources

About