Salinity Properties Retrieval from Sentinel-2 Satellite Data and Machine Learning Algorithms

Soil pH and electrical conductivity (EC) are two major chemical properties affecting nutrient availability and rice grain yield. Therefore, this study was conducted to investigate the variation of pH and EC in the topsoil layer (0-15 cm) of rice fields as affected by major water source used for rice cultivation, rice-based cropping system adopted and rice growing soil orders in different agro-climatic zones (ACZs) in Sri Lanka. A total of 998 soil samples were collected from lowland rice fields. Both pH and EC were measured in 1:5 soil: water extracts. The pH of soil samples was in the range of 3.0-7.7 with a mean value of 5.0. Moreover, 75% of the soil samples had pH values below the optimum range for rice cultivation (5.5-7.0). Values of pH observed in Dry Zone soils were higher (5.2) than those in Wet Zone (4.4). Soil EC values ranged between 1.0 - 3,100 μScm−1 with a mean value of 148.5 μScm−1. Soil EC was similar among climatic zones (P>0.05). Upcountry Intermediate zone recorded the highest soil EC than that in other ACZs (P<0.05). Moreover, 73%, 22%, 3%, 2% and 0.1% of soil samples recorded EC values in the ranges of less than 150 (non-saline), 150-400 (slightly saline), 400-800 (moderately saline), 800-2000 (highly saline) and more than 2000 (very highly saline) μScm−1, respectively. Considering micro (e.g. paddy track) and macro (e.g. ACZ) scale spatial heterogeneity in soil pH and EC, appropriate site-specific strategies need to be adopted to improve soil pH and EC to suit sustainable rice crop production.

Accurate monitoring of soil salinization plays a key role in the ecological security and sustainable agricultural development of arid regions. As a branch of artificial intelligence, machine learning acquires new knowledge through self-learning and continuously improves its own performance. The purpose of this study is to combine Sentinel-2 Multispectral Imager (MSI) data and MSI-derived covariates with measured soil salinity data and to apply three machine learning algorithms in modeling to estimate and map the soil salinity in the study sample area. According to the convenient transportation conditions, the study area and sampling quadrat were set up, and the 5-point method was used to collect the soil mixed samples, and 160 soil mixed samples were collected. Kennard–Stone (K–S) algorithm was used for sample classification, 70% for modeling and 30% for verification. The machine learning algorithm uses Support Vector Machines (SVM), Artificial Neural Network (ANN), and Random Forest (RF). The results showed that (1) the average reflectance of each band of the MSI data ranged from 0.21–0.28. According to the spectral characteristics corresponding to different soil electrical conductivity (EC) levels (1.07–79.6 dS m−1), the spectral reflectance of salinized soil in the MSI data ranged from 0.09–0.35. (2) The correlation coefficient between the MSI data and MSI-derived covariates and soil EC was moderate, and the correlation between certain MSI data sets and soil EC was not significant. (3) The SVM soil EC estimation model established with the MSI data set attained a higher performance and accuracy (R2 = 0.88, root mean square error (RMSE) = 4.89 dS m−1, and ratio of the performance to the interquartile range (RPIQ) = 1.96, standard error of the laboratory measurements to the standard error of the predictions (SEL/SEP) = 1.11) than those attained with the soil EC estimation models established with the RF and ANN models. (4) We applied the SVM soil EC estimation model to map the soil salinity in the study area, which showed that the farmland with higher altitudes discharged a large amount of salt to the surroundings due to long-term irrigation, and the secondary salinization of the farmland also caused a large amount of salt accumulation. This research provides a scientific basis for the simulation of soil salinization scenarios in arid areas in the future.

A farm's soil moisture and soil electrical conductivity (EC)readings are extremely valuable for a farmer. They can help her reduce water use and improve productivity. However, the high cost of commercial soil moisture sensors and the inaccuracy of sub-1000 dollar EC sensors have limited their adoption. In this paper, we present the design and implementation of a system, called Strobe, that senses soil moisture and soil EC using RF propagation in existing Wi-Fi bands. Strobe overcomes the key challenge of limited bandwidth availability in the 2.4 GHz unlicensed spectrum using a novel multi-antenna technique. It maps the propagation time and amplitude of Wi-Fi signals received by different antennas to the soil permittivity and EC, which in turn depend on soil moisture and salinity. Our experiments with USRP, WARP, and commodity Wi-Fi cards show that Strobe can accurately estimate soil moisture and EC using Wi-Fi, thereby showing the potential of a future where a farmer can sense soil in their farm without investing 1000s of dollars in sensing tools.

Assessing the performance of machine learning algorithms for soil salinity mapping in Google Earth Engine platform using Sentinel-2A and Landsat-8 OLI data

The soil-overturning properties is an important matter in the working results by rotary tillage, but the mesuring method is not still established in Japan. The purpose of this study is to obtain establishment of the mesuring method and the accurate ratio of soil displacement by rotary tillage. This paper deals with two methods calcurating the ratio of soil displacement.In this experiment, we adopted the method to use the soil mixed kalium chloride (KCI), as electrical conductivity and concentration of solution of KCI were in proportion to each other. After the electrical conductivity of the soil which mixed two level's soils differing in electrical conductivity was mesured, we discussed on a hypothetical equation to calculate the electrical conductivity of the mixed soils. The deviation between mesured electrical conductivity and calculated one showed the range of -3.4-2.3%. However it is assumed that this degree of errors are practically admitted.From application of the hypothetical equation, the two methods (simultaneous equations of three unknowns and conditional inequality) to obtain the ratio of soil displacement were derived. Mesuring the electrical conductivity of soils which were picked before and after rotary tillage, we can easily estimate by these methods the each ratio that a soil layer before tillage moved to the three layers after tillage.The method by simultaneous equations of three unknowns must perform the experiment of three plots in which the mixture soil of KCI was put at each layer before rotary tillage. Moreover, this method needs to make the uniform conditions of the three experimental plots.In case of the method by conditional inequality, the ratio of soil displacement is calculated with a certain range. Therefore if this range increases, the estimating inequality become useless. However, increasing the electrical conductivity of mixture soil of KCI, it is possible to decrease the range within 1%.

This research explores the use of soil electrical conductivity (EC) in rice cultivation, focusing on soil EC correlation with vegetation indices (NDVI, NDRE) and soil EC impact on yield. A total of 228 soil EC, NDVI, and NDRE data points were collected from an 11,000 m² field in Hau Giang and Can Tho City. Soil EC was measured at the beginning of the season, and at 39 and 49 days after sowing. NDVI was recorded on days 39, 49, 60, and 72, with initial values set to zero due to no crop presence. Results show that soil EC decreased from the start to day 49, while NDVI and NDRE increased. Spearman correlation analysis revealed strong relationships between early soil EC and NDVI at days 39 and 49 (r = 0.82, 0.71), and between soil EC at day 49 (EC 49) and NDRE at later stages (r = 0.594, 0.565). Additionally, 150 soil EC and yield samples from five Mekong Delta regions were used in an LSTM model (70% training, 30% testing), achieving high accuracy (RMSE = 0.0919, MAE = 0.0634, R2 = 0.8492). The findings suggest that soil EC is a valuable indicator for monitoring crop growth and managing rice production, particularly in areas affected by salinity due to climate change.

Electrical conductivity of some soils in Trinidad as affected by density, water and peat content

To clarify the nutrient characteristics of 'salt island' and 'fertile island' effects in saline-alkali soil, the native Tamarix chinensis of the Yellow River Delta (YRD) was selected to measure its soil pH,electrical conductivity (EC), organic carbon (SOC),total nitrogen (N),total phosphorus (P) and their stoichiometry characteristics at different soil depths. The results showed that soil pH and EC increased with the increasing soil depth. Soil EC and P in the 0-20 cm layer decreased and increased from canopied area to interspace,respectively. SOC,N,N/P and C/P in the 20-40 cm soil layer decreased, and C/N increased from the shrub center to interspace. SOC and N contents between island and interspace both decreased but P content decreased firstly and then increased with the increasing soil depth. Soil pH correlated positively with EC. In addition, pH and EC correlated negatively with C,N,P contents and their ecological stoichiometry.

Loss function is an integral part of any machine learning algorithm. Loss function is used to measure the current performance of machine learning model during training. Choosing a relevant loss function is therefore important. A better loss function can drive the model towards a good sub-optimal solution. Machine learning has been used to model relations between variables of complex systems, e.g. soil. Using a good machine learning algorithm, we can predict various soil properties like pH, soil electrical conductivity (SEC), etc. Prediction of these variables is important since this can help in deciding which food crops should be planted. SEC refers to the amount of soluble salt present in the soil. If the soil contains too much salt, then the vegetation of the area suffers. Also, the prediction of soil electrical conductivity is important because it is difficult to visit each site to estimate it. Using machine learning models to predict SEC, without any in-situ analysis, we can get an intuitive idea about the value of SEC of the soil. An attempt to predict SEC using neural network model is done in this study. To train the machine learning model, several loss functions were optimized that are generally used for prediction in machine learning—mean squared error loss (MSE or L2 loss), mean absolute error loss (MAE or L1 loss), Huber loss. All the loss functions—mean squared error loss, mean absolute error and Huber loss along with different optimizers like Adam and stochastic gradient descent (SGD)—were experimented in the model. Many techniques like dropout, batch normalization and adaptive learning rate were also attempted in order to improve the model. The error metric for evaluation of our predictions with the actual values was MSE. After a lot of iteration and optimizations, the best model estimated MSE of 0.029.

This study aimed to elucidate the response of soil microbial communities to saline soil amelioration via biological organic fertilizer. A year-long experiment was conducted on coastal saline soil, employing water and fertilizer strategies. Three treatments were compared: dry field (control, CK), paddy field (W), and combined dry and irrigated fields with biological organic fertilizer (BW). Soil DNA was extracted and sequenced using high-throughput methods, revealing significant reductions in soil electrical conductivity (EC) and pH with W and BW treatments. Moreover, the BW treatment notably increased soil organic carbon content by 17.2%, as well as soil urease and alkaline phosphatase activity. Fungal community richness increased, with the BW treatment showing a 36% rise in the ACE index and a 24% increase in the Shannon index, while the Simpson index decreased by 59%. Dominant fungal phyla were Ascomycota, Mortierellomycota, and Basidiomycota, with Basidiomycota prevailing at the genus level. Redundancy analysis (RDA) indicated that soil pH, EC, and organic carbon were key determinants of fungal community distribution, with the BW treatment correlating negatively with pH and salt and positively with soil organic carbon (SOC). Fungal functional groups varied among treatments, with saprophytic fungi predominating, but the BW treatment showed a higher relative abundance of animal pathogenic fungi. In summary, the integration of biological organic fertilizer with flooding ameliorates soil properties and influences the changes in soil fungal community structure and function in the short term. These results could enhance the scientific basis for the efficient utilization and development of saline soil resources in coastal areas.

The study’s specific objective is to identify prospective cashew nut (Anacardium occidentale) producing areas in the Bankura district based on soil availability and land suitability in red and lateritic soil. The index of crop suitability (ICS) analysis was carried out utilizing the Fuzzy-analytical hierarchy process (F-AHP) approach and 16 geo-environmental variables such as soil texture, soil organic carbon (OC), Soil electrical conductivity (EC), soil pH, slope, rainfall, etc. The criteria were also developed following FAO's crop suitability index for the distinction between S (suitable) and N (unsuitable) order. According to the cashew suitability class, 9.24% of cashew growth regions can be categorised as highly suitable, 34.57% as moderately suitable, and 19.59% as marginally suitable. The F-AHP weight shows the important factors of cashew cultivation that are soil OC and EC. The results demonstrate that cashew production is more beneficial in the red and lateritic soil.

Empirical estimation of saturated soil-paste electrical conductivity in the EU using pedotransfer functions and Quantile Regression Forests: A mapping approach based on LUCAS topsoil data

The electrical conductivity (EC) Resistively (calculated) and salinity of some selected soil samples collected from three landforms, namely Beach Ridge Sand (BRS), coastal plain sand (CPS), sandstone/shale Hill ridges (SHR) from Akwa Ibom State of the Federal Republic of Nigeria were measured the obtained data were subjected to spearman rank order statistics and result shows that there were positive correlation of 0.20 between soil EC and salinity. The electrical conductivity of the area under study falls within the range of 0.031 ds/m for SHR 3A t0 0.200ds/m for SHR 2A while the salinities are between 0.6mg/l for BRS 3B and 0.9mg/l for SHA 2A soil samples. Soil sample SHR 3A has the highest resistivity measurement of 32.Ωm, while SHR 2A soil sample has the lowest of 6. Ωm. from the EC data the signal strength of EC measurements decreases with soil depth and most of the EC values falls within the range of 20 - 80 (µs/cm) which is the optimum level for most plants, usually indicates well - fertilized soils and salt sensitive plants may be insured.

Crop monitoring over large areas with high accuracy is of great significance in precision agriculture. The study is an attempt to assess the potential of open-source high-resolution satellite datasets and open-source digital platforms in combination with AI/ML algorithms for near-real-time crop monitoring and yield estimation at the farm level. In this research, we used Sentinel-1 and Sentinel-2 datasets, by processing them on the Google Earth Engine platform and developed several crop-based indicators to assess crop phenology as well as the distinction between a well-managed field (demo plots) vs a normal farmers' practice-managed crop (control plots) using Sentinel-1 satellite data. Further, crop yields were estimated before the harvesting of the crop by using Sentinel-1 and Sentinel-2 data with machine learning algorithms. The findings demonstrate that the effect of an improved package of practices on rice was significantly different from the farmer's practice. Among the statistical yield models developed for yield estimation, the gradient tree boosting model performed better than other models. This study proposes a novel method of near-real-time remote crop monitoring right from sowing to harvest time to estimate crop yields with an accuracy of 77 percent. There is potential in using open-source satellite data for monitoring farm fields in the future.

A global meta-analysis of the effects of irrigation with water containing salts on saline-alkali soils