Evaluating machine learning performance in predicting sodium adsorption ratio for sustainable soil-water management in the eastern Mediterranean.

Soil salinization is a critical global issue for sustainable agriculture, impacting crop yields and posing a threat to achieving the Sustainable Development Goal (SDG) of ensuring food security. It is necessary to monitor it in detail and uncover its underlying factors at a regional scale. In this context, the present study aimed to evaluate soil health in the eastern Mediterranean region by using the Sodium Adsorption Ratio (SAR) as an indicator of soil salinity in three distinct soil horizons. The main objective of the research was to evaluate the performance of four machine learning (ML) models, including Random Forest (RF), Nu Support Vector Regression (NuSVR), Artificial Neural Network-Multi Layer Perceptron (ANN-MLP), and Gradient Boosting Regression (GBR), for accurate prediction of SAR following the Recursive Feature Elimination (RFE) as a feature selection method. Moreover, SHapely Additive exPlanations (SHAP) was applied as sensitivity analysis to identify the most influential covariates. Main findings of the research revealed that the average clay content in the surface horizon (H10-25cm) was 50.5% ± 10.4, which significantly increased to 57.5% ± 8.7 (p < 0.05). No significant mean differences were detected between the studied horizons for SAR and Na+. ML output revealed that NuSVR outperformed other algorithms in accurately predicting outcomes during both the training and testing stages. Moreover, Scenario 2 (SC2) with seven selected features from the RFE method facilitated highly accurate SAR predictions. Overall, the performance of ML models is ranked as NuSVR > GBR > ANN-MLP > RF. Lastly, SHAP sensitivity analysis identified CEC, Ca+2, Mg+2, and Na+ as the most influential variables for SAR prediction in both the training and testing stages. Hence, the research yielded valuable insights for efficient agricultural soil management at a regional level using state-of-the-art technology.

Does Short-Term Combined Irrigation Using Brackish-Reclaimed Water Cause the Risk of Soil Secondary Salinization?

Brackish water has to be used to irrigate crops for harvest due to the scarcity of freshwater resources. However, brackish water irrigation may cause secondary soil salinization. Whether the combined utilization of different non-conventional water resources could relieve the risk of secondary soil salinization has not been reported. In order to explore the safe and rational utilization of brackish water in areas where freshwater resources are scarce, a pot experiment was conducted to study the risk of secondary soil mixed irrigation and rotational irrigation using brackish water and reclaimed water or freshwater. The results indicated that: (1) Short-term irrigation using reclaimed water did not cause secondary soil salinization, although increasing soil pH value, ESP, and SAR. The indices did not exceed the threshold of soil salinization. (2) Compared with mixed irrigation using brackish-freshwater, the contents of soil exchangeable Ca2+, K+, and Mg2+ increased, and the content of soil exchangeable Na+ decreased under rotational irrigation using brackish-reclaimed water. In addition, the contents of soil exchangeable Na+ and Mg2+ under mixed irrigation or rotational irrigation were significantly lower, and the exchangeable K+ content of the soil was higher compared with brackish water irrigation. The exchangeable Ca2+ content under rotational irrigation was higher than that of brackish water irrigation, while the reverse was seen under mixed irrigation. (3) For different combined utilization modes of brackish water and reclaimed water, the ESP and SAR were the lowest under rotational irrigation, followed by mixed irrigation and brackish water irrigation. The ESP under brackish water treatment exceeded 15%, indicating a certain risk of salinization, while ESPs under other treatments were below 15%. Under mixed irrigation or rational irrigation using reclaimed-brackish water, the higher the proportion or rotational times of reclaimed water, the lower the risk of secondary soil salinization. Therefore, short-term combined irrigation using brackish water and reclaimed water will not cause the risk of secondary soil salinization, but further experiments need to verify or cooperate with other agronomic measures in long-term utilization.

Estimation of sodium adsorption ratio in a river with kernel-based and decision-tree models.

The control of surface water quality plays an important role in the management of water resources. In this context, the estimation and assessment of sodium adsorption ratio (SAR) are required which is one of the significant water quality parameters in the agricultural production sector. Chemical analysis might not, however, be feasible for a longer period of time in all the country-scale rivers. Therefore in this study, a support vector regression (SVR) model with different kernel functions; K nearest neighbour algorithm; and four decision-tree models, namely, Hoeffding tree, random forest, random tree, and REPTree, were used to estimate the SAR value with minimal parameters in the Aladag River in Turkey. In alternative scenarios, a correlation matrix and sensitivity analysis were used to ascertain the model inputs from among the 15 distinct parameters. All 15 parameters were utilized as model inputs in the first scenario, and only the sodium (Na) parameter was utilized as the model input in the final scenario. The accuracy of the aforesaid models was then assessed making use of correlation coefficient, Nash-Sutcliffe model efficiency coefficient, root mean square error, mean absolute error, and Willmott index of agreement. The results indicate that the SVR model with the poly kernel function provides the best estimates of SAR among the considered models. According to the findings, there is no considerable difference between the results acquired in the first and last scenarios, and one can determine the SAR value while making use of machine learning approaches taking into account only Na parameter.

The impact of olive mill wastewater spreading on soils using integrated approach of proximal soil survey, spatial, and multivariate analyses.

Olive oil production generates a large amount of olive mill wastewater (OMW), the most difficult to treat of agro-industrial effluents. Spreading of OMW across the soilscape has become the most frequently used practice in several Mediterranean basin countries but is hotly debated because of its potential to impair soil environs. The research hypothesis states that soil deterioration is correlated strongly with the spatiotemporal rate of OMW application; thus, the spatial pattern of the soil attributes should be established in conjunction with the rate of OMW spreading. The spatial pattern was ascertained using proximal soil sensing that measures the apparent electrical conductance (ECa). Eight representative locations were identified using grouping analysis of the ECa data. The soils were analyzed for selected physical and chemical attributes known to be affected by OMW spreading as well as ancillary parameters needed for the calibration of the ECa. Discriminant analysis successfully categorized 76% of the ECa groups, selecting CEC, sand, pH, and ESP as the most powerful discriminatory variables in the grouping analysis. The correlation coefficients between the measured ECa, and the calculated ECa, and soil moisture were very high (r > 0.77, P < 0.05), suggesting that the proximal soil survey results are well calibrated. The spreading of untreated OMW over more than 7 years did not impair the soils under study. No significant changes in soil chemo-physical properties such as pH (< 7.66), electrical conductivity in saturated paste (< 3.58 dS m-1), sodium adsorption ratio (< 2.3), potassium adsorption ratio (< 0.33), exchangeable sodium percentage (< 3.85%), and unsaturated hydraulic conductivity (< 0.3 cm h-1) were found in comparison with untreated soils. The results support the premise that moderate quantities of OMW (50-70 m3 ha-1 year-1) equally spread over the soilscape will have little impact on soil health.

Levels and Distribution of Pollutants in the Waters of an Aquatic Ecosystem in Northern Mexico.

The availability of good quality water resources is essential to ensure healthy crops and livestock. The objective of this study was to evaluate the level of pollution in Bustillos Lagoon in northern Mexico. Physical-chemical parameters like sodium, chloride, sulfate, electrical conductivity, nitrates, and the pesticide dichlorodiphenyltrichloroethane (DDT) were analyzed to determine the water quality available in the lagoon. Although DDT has been banned in several countries, it is still used for agricultural purposes in Mexico and its presence in this area had not been analyzed previously. Bustillos Lagoon was divided into three zones for the evaluation: (1) industrial; (2) communal lands; and (3) agricultural. The highest concentrations of sodium (2360 mg/L) and SAR (41 meq/L) reported in the industrial zone are values exceeding the United Nations Food and Agricultural Organization (FAO) irrigation water quality guidelines. DDT and its metabolites were detected in all of the 21 sites analyzed, in the agricultural zone ∑DDTs = 2804 ng/mL, this level is much higher than those reported for other water bodies in Mexico and around the world where DDT has been used heavily. The water in the communal zone is the least contaminated, but can only be recommended for irrigation of plants with high stress tolerance and not for crops.

Passive remediation of coalbed natural gas co-produced water using zeolite.

Coalbed natural gas (CBNG) co-produced waters can contain sodium (Na(+)) concentrations that may be environmentally detrimental if discharged to receiving bodies of water or applied to land surfaces. A field demonstration and companion laboratory studies were conducted to evaluate the use of a Bear River zeolite (BR-zeolite) for mitigating impacts associated with Na(+) in CBNG waters. Bench-scale kinetic and adsorption isotherm studies were performed to determine both the rate and extent of sodium Na(+) adsorption and assess the effects of bicarbonate (HCO3(-)) and chloride (Cl(-)) anions. Results of these studies showed that the adsorption of Na(+) on BR-zeolite followed the Langmuir adsorption model with maximum adsorption equal to 21 and 18 g Na(+)/kg zeolite with 0.0012 and 0.0006 L/mg Langmuir coefficients (KL) for sodium bicarbonate and sodium chloride, respectively. The kinetics study indicated that the sorption of Na(+) was inversely related to the size of the zeolite particles with significantly greater adsorption for smaller particles. The field demonstration evaluated the effectiveness of BR-zeolite for mitigating infiltration losses from Na(+) in CBNG waters. The field site utilized 12 open boreholes, each installed to a depth of approximately 1.8 m. Each borehole was lined with a 3.0 m long, 15 cm diameter schedule 40 PVC pipe and fitted with an automatic data logging pressure transducer for measuring water levels over time. The BR-zeolite was found to mitigate much of the deleterious effect that high sodium adsorption ratio (SAR = 27 (mol/m(3))(1/2)) CBNG co-produced water had on soil permeabilities.