Estimation of soil salt content by combining UAV-borne multispectral sensor and machine learning algorithms

Abstract

Soil salinization is a global problem closely related to the sustainable development of social economy. Compared with frequently-used satellite-borne sensors, unmanned aerial vehicles (UAVs) equipped with multispectral sensors provide an opportunity to monitor soil salinization with on-demand high spatial and temporal resolution. This study aims to quantitatively estimate soil salt content (SSC) using UAV-borne multispectral imagery, and explore the deep mining of multispectral data. For this purpose, a total of 60 soil samples (0–20 cm) were collected from Shahaoqu Irrigation Area in Inner Mongolia, China. Meanwhile, from the UAV sensor we obtained the multispectral data, based on which 22 spectral covariates (6 spectral bands and 16 spectral indices) were constructed. The sensitive spectral covariates were selected by means of gray relational analysis (GRA), successive projections algorithm (SPA) and variable importance in projection (VIP), and from these selected covariates estimation models were built using back propagation neural network (BPNN) regression, support vector regression (SVR) and random forest (RF) regression, respectively. The performance of the models was assessed by coefficient of determination (R2), root mean squared error (RMSE) and ratio of performance to deviation (RPD). The results showed that the estimation accuracy of the models had been improved markedly using three variable selection methods, and VIP outperformed GRA and GRA outperformed SPA. However, the model accuracy with the three machine learning algorithms turned out to be significantly different: RF > SVR > BPNN. All the 12 SSC estimation models could be used to quantitatively estimate SSC (RPD > 1.4) while the VIP-RF model achieved the highest accuracy (Rc2 = 0.835, RP2 = 0.812, RPD = 2.299). The result of this study proved that UAV-borne multispectral sensor is a feasible instrument for SSC estimation, and provided a reference for further similar research.

Estimating soil salinity under sunflower cover in the Hetao Irrigation District based on unmanned aerial vehicle remote sensing

“…Different VIs and SIs are calculated from two or more bands, so the selection of appropriate VIs and SIs is crucial. It has been shown that Spectral indices such as NDVI and CRSI are highly sensitive to SSC (Wei et al, 2020;Qi et al, 2020). We selected six widely used VIs and four SIs for this study, and their calculation equations are given in Table 2, where, R, B, G, NIR and RE are the T A B L E 1 UAV imagery and ground measurements with respect to the day of year (DOY) and the number of ground sampling points (GSP) in each study area: Ground parameters include SSC, LAI, CHL and H…”

“…Hu et al (2019) used the random Forest (RF) regression method to quantitatively estimate SSC and drew soil distribution maps of three different vegetation coverage based on UAV hyperspectral imagery. Wei et al (2020) established an estimation model of SSC by using UAV multispectral imagery and three machine learning algorithms and concluded that the quantitative estimation of SSC by UAV multispectral remote sensing is feasible. Chen et al (2020) established estimation models of SSC in sunflower fields at the budding and flowering stages and at different soil depths by using UAV multispectral data.…”

“…Many studies have used machine learning nonlinear methods to establish quantitative characterization models between spectral indices and SSC, back propagation neural network (BPNN), support vector machine regression (SVR) and RF are commonly used modelling methods. Wei et al (2020) established SSC estimation models using these three methods and showed that RF model had the highest prediction accuracy (R 2 = 0.84). Zhu et al (2022) and Yu et al (2022) showed that the ANN model had lower error and better overall performance in predicting soil heat flux.…”

Potential and limits of vegetation indices compared to evaporite mineral indices for soil salinity discrimination and mapping

“…Nevertheless, they can also be integrated into semi-empirical or physical models for quantitative prediction of the salinity content classes in the soil (Al-Ali et al, 2021). To select the most informative soil salinity index, comparative studies have been completed by applying regression analyses between EC-Lab and SI derived from spectral measurements, satellite, airborne and drone images (Allbed et al, 2014;Bannari et al, 2018;Peng et al, 2019;Hu et al, 2019;Wei et al, 2020;Milewski et al, 2020;Gopalakrishnan and Kumar, 2020). Often, obtained results vary depending on the spectral wavebands integrated in the equation of each index.…”

Soil Salinity Inversion in Coastal Corn Planting Areas by the Satellite-UAV-Ground Integration Approach

“…Mohammad et al [14] implemented the monitoring of soil salinity in a large area of Qom County, Qom Province, Iran, based on Sentinel-2A data. Wei et al [15] took advantage of UAV equipped with Micro-MCA multispectral sensors to obtain images and realized the estimation of soil salinity in a small area of Hetao Irrigation District. Wang et al [16] utilized a portable spectrometer ASD to obtain soil hyperspectral to construct a soil salinity inversion model in Baidunzi Basin, China, and achieved high model accuracy.…”