A study was conducted to investigate the influence of Asian monsoon on chlorophyll-a (Chl-a) content in Sabah waters and to identify the related oceanographic conditions that caused phytoplankton blooms at the eastern and western coasts of Sabah, Malaysia. A series of remote sensing measurements including surface Chl-a, sea surface temperature, sea surface height anomaly, wind speed, wind stress curl, and Ekman pumping were analyzed to study the oceanographic conditions that lead to large-scale nutrients enrichment in the surface layer. The results showed that the Chl-a content increased at the northwest coast from December to April due to strong northeasterly wind and coastal upwelling in Kota Kinabalu water. The southwest coast (Labuan water) maintained high concentrations throughout the year due to the effect of Padas River discharge during the rainy season and the changing direction of Baram River plume during the northeast monsoon (NEM). However, with the continuous supply of nutrients from the upwelling area, the high Chl-a batches were maintained at the offshore water off Labuan for a longer time during NEM. On the other side, the northeast coast illustrated a high Chl-a in Sandakan water during NEM, whereas the northern tip off Kudat did not show a pronounced change throughout the year. The southeast coast (Tawau water) was highly influenced by the direction of the surface water transport between the Sulu and Sulawesi Seas and the prevailing surface currents. The study demonstrates the presence of seasonal phytoplankton blooms in Sabah waters which will aid in forecasting the possible biological response and could further assist in marine resource managements.
Droughts are recurring events in Australia and cause a severe effect on agricultural and water resources. However, the studies about agricultural drought risk mapping are very limited in Australia. Therefore, a comprehensive agricultural drought risk assessment approach that incorporates all the risk components with their influencing criteria is essential to generate detailed drought risk information for operational drought management. A comprehensive agricultural drought risk assessment approach was prepared in this work incorporating all components of risk (hazard, vulnerability, exposure, and mitigation capacity) with their relevant criteria using geospatial techniques. The prepared approach is then applied to identify the spatial pattern of agricultural drought risk for Northern New South Wales region of Australia. A total of 16 relevant criteria under each risk component were considered, and fuzzy logic aided geospatial techniques were used to prepare vulnerability, exposure, hazard, and mitigation capacity indices. These indices were then incorporated to quantify agricultural drought risk comprehensively in the study area. The outputs depicted that about 19.2% and 41.7% areas are under very-high and moderate to high risk to agricultural droughts, respectively. The efficiency of the results is successfully evaluated using a drought inventory map. The generated spatial drought risk information produced by this study can assist relevant authorities in formulating proactive agricultural drought mitigation strategies.
Termite nests have long been suggested to be good indicators of groundwater but only a few studies are available to demonstrate the relationship between the two. This study therefore aims at investigating the most favourable spots for locating groundwater structures on a small parcel of land with conspicuous termite activity. To achieve this, geophysical soundings using the renowned vertical electrical sounding (VES) technique was carried out on the gridded study area. A total of nine VESs with one at the foot of a termitarium were conducted. The VES results were interpreted and assessed via two different techniques: (1) physical evaluation as performed by drillers in the field and (2) integration of primary and secondary geoelectrical parameters in a geographic information system (GIS). The result of the physical evaluation indicated a clear case of subjectivity in the interpretation but was consistent with the choice of VES points 1 and 6 (termitarium location) as being the most prospective points to be considered for drilling. Similarly, the integration of the geoelectrical parameters led to the mapping of the most prospective groundwater portion of the study area with the termitarium chiefly in the center of the most suitable region. This shows that termitaria are valuable landscape features that can be employed as biomarkers in the search of groundwater.
This study develops an oil spill environmental vulnerability model for predicting and mapping the oil slick trajectory pattern in Kota Tinggi, Malaysia. The impact of seasonal variations on the vulnerability of the coastal resources to oil spill was modelled by estimating the quantity of coastal resources affected across three climatic seasons (northeast monsoon, southwest monsoon and pre-monsoon). Twelve 100 m3 (10,000 splots) medium oil spill scenarios were simulated using General National Oceanic and Atmospheric Administration Operational Oil Modeling Environment (GNOME) model. The output was integrated with coastal resources, comprising biological, socio-economic and physical shoreline features. Results revealed that the speed of an oil slick (40.8 m per minute) is higher during the pre-monsoon period in a southwestern direction and lower during the northeast monsoon (36.9 m per minute). Evaporation, floating and spreading are the major weathering processes identified in this study, with approximately 70% of the slick reaching the shoreline or remaining in the water column during the first 24 h (h) of the spill. Oil spill impacts were most severe during the southwest monsoon, and physical shoreline resources are the most vulnerable to oil spill in the study area. The study concluded that variation in climatic seasons significantly influence the vulnerability of coastal resources to marine oil spill.
Facial recognition has a significant application for security, especially in surveillance technologies. In surveillance systems, recognizing faces captured far away from the camera under various lighting conditions, such as in the daytime and nighttime, is a challenging task. A system capable of recognizing face images in both daytime and nighttime and at various distances is called Cross-Spectral Cross Distance (CSCD) face recognition. In this paper, we proposed a phase-based CSCD face recognition approach. We employed Homomorphic filtering as photometric normalization and Band Limited Phase Only Correlation (BLPOC) for image matching. Different from the state-of-the-art methods, we directly utilized the phase component from an image, without the need for a feature extraction process. The experiment was conducted using the Long-Distance Heterogeneous Face Database (LDHF-DB). The proposed method was evaluated in three scenarios: (i) cross-spectral face verification at 1m, (ii) cross-spectral face verification at 60m, and (iii) cross-spectral face verification where the probe images (near-infrared (NIR) face images) were captured at 1m and the gallery data (face images) was captured at 60 m. The proposed CSCD method resulted in the best recognition performance among the CSCD baseline approaches, with an Equal Error Rate (EER) of 5.34% and a Genuine Acceptance Rate (GAR) of 93%.
Groundwater hazard assessments involve many activities dealing with the impacts of pollution on groundwater, such as human health studies and environment modelling. Nitrate contamination is considered a hazard to human health, environment and ecosystem. In groundwater management, the hazard should be assessed before any action can be taken, particularly for groundwater pollution and water quality. Thus, pollution due to the presence of nitrate poses considerable hazard to drinking water, and excessive nutrient loads deteriorate the ecosystem. The parametric IPNOA model is one of the well-known methods used for evaluating nitrate content. However, it cannot predict the effect of soil and land use/land cover (LULC) types on calculations relying on parametric well samples. Therefore, in this study, the parametric model was trained and integrated with the multivariate data-driven model with different levels of information to assess groundwater nitrate contamination in Saladin, Iraq. The IPNOA model was developed with 185 different well samples and contributing parameters. Then, the IPNOA model was integrated with the logistic regression (LR) model to predict the nitrate contamination levels. Geographic information system techniques were also used to assess the spatial prediction of nitrate contamination. High-resolution SPOT-5 satellite images with 5 m spatial resolution were processed by object-based image analysis and support vector machine algorithm to extract LULC. Mapping of potential areas of nitrate contamination was examined using receiver operating characteristic assessment. Results indicated that the optimised LR-IPNOA model was more accurate in determining and analysing the nitrate hazard concentration than the standalone IPNOA model. This method can be easily replicated in other areas that have similar climatic condition. Therefore, stakeholders in planning and environmental decision makers could benefit immensely from the proposed method of this research, which can be potentially used for a sustainable management of urban, industrialised and agricultural sectors.
The diagnosis of leukemia involves the detection of the abnormal characteristics of blood cells by a trained pathologist. Currently, this is done manually by observing the morphological characteristics of white blood cells in the microscopic images. Though there are some equipment- based and chemical-based tests available, the use and adaptation of the automated computer vision-based system is still an issue. There are certain software frameworks available in the literature; however, they are still not being adopted commercially. So there is a need for an automated and software- based framework for the detection of leukemia. In software-based detection, segmentation is the first critical stage that outputs the region of interest for further accurate diagnosis. Therefore, this paper explores an efficient and hybrid segmentation that proposes a more efficient and effective system for leukemia diagnosis. A very popular publicly available database, the acute lymphoblastic leukemia image database (ALL-IDB), is used in this research. First, the images are pre-processed and segmentation is done using Multilevel thresholding with Otsu and Kapur methods. To further optimize the segmentation performance, the Learning enthusiasm-based teaching-learning-based optimization (LebTLBO) algorithm is employed. Different metrics are used for measuring the system performance. A comparative analysis of the proposed methodology is done with existing benchmarks methods. The proposed approach has proven to be better than earlier techniques with measuring parameters of PSNR and Similarity index. The result shows a significant improvement in the performance measures with optimizing threshold algorithms and the LebTLBO technique.
Ever increasing demand for water resources for different purposes makes it essential to have better understanding and knowledge about water resources. As known, groundwater resources are one of the main water resources especially in countries with arid climatic condition. Thus, this study seeks to provide groundwater potential maps (GPMs) employing new algorithms. Accordingly, this study aims to validate the performance of C5.0, random forest (RF), and multivariate adaptive regression splines (MARS) algorithms for generating GPMs in the eastern part of Mashhad Plain, Iran. For this purpose, a dataset was produced consisting of spring locations as indicator and groundwater-conditioning factors (GCFs) as input. In this research, 13 GCFs were selected including altitude, slope aspect, slope angle, plan curvature, profile curvature, topographic wetness index (TWI), slope length, distance from rivers and faults, rivers and faults density, land use, and lithology. The mentioned dataset was divided into two classes of training and validation with 70 and 30% of the springs, respectively. Then, C5.0, RF, and MARS algorithms were employed using R statistical software, and the final values were transformed into GPMs. Finally, two evaluation criteria including Kappa and area under receiver operating characteristics curve (AUC-ROC) were calculated. According to the findings of this research, MARS had the best performance with AUC-ROC of 84.2%, followed by RF and C5.0 algorithms with AUC-ROC values of 79.7 and 77.3%, respectively. The results indicated that AUC-ROC values for the employed models are more than 70% which shows their acceptable performance. As a conclusion, the produced methodology could be used in other geographical areas. GPMs could be used by water resource managers and related organizations to accelerate and facilitate water resource exploitation.
Gully erosion possess a serious hazard to critical resources such as soil, water, and vegetation cover within watersheds. Therefore, spatial maps of gully erosion hazards can be instrumental in mitigating its negative consequences. Among the various methods used to explore and map gully erosion, advanced learning techniques, especially deep learning (DL) models, are highly capable of spatial mapping and can provide accurate predictions for generating spatial maps of gully erosion at different scales (e.g., local, regional, continental, and global). In this paper, we applied two DL models, namely a simple recurrent neural network (RNN) and a gated recurrent unit (GRU), to map land susceptibility to gully erosion in the Shamil-Minab plain, Hormozgan province, southern Iran. To address the inherent black box nature of DL models, we applied three novel interpretability methods consisting of SHaply Additive explanation (SHAP), ceteris paribus and partial dependence (CP-PD) profiles and permutation feature importance (PFI). Using the Boruta algorithm, we identified seven important features that control gully erosion: soil bulk density, clay content, elevation, land use type, vegetation cover, sand content, and silt content. These features, along with an inventory map of gully erosion (based on a 70 % training dataset and 30 % test dataset), were used to generate spatial maps of gully erosion using DL models. According to the Kolmogorov-Smirnov (KS) statistic performance assessment measure, the simple RNN model (with KS = 91.6) outperformed the GRU model (with KS = 66.6). Based on the results from the simple RNN model, 7.4 %, 14.5 %, 18.9 %, 31.2 % and 28 % of total area of the plain were classified as very-low, low, moderate, high and very-high hazard classes, respectively. According to SHAP plots, CP-PD profiles, and PFI measures, soil silt content, vegetation cover (NDVI) and land use type had the highest impact on the model's output. Overall, the DL modelling techniques and interpretation methods used in this study proved to be helpful in generating spatial maps of soil erosion hazard, especially gully erosion. Their interpretability can support watershed sustainable management.
Studies relating to trends of vegetation, snowfall and temperature in the north-western Himalayan region of India are generally focused on specific areas. Therefore, a proper understanding of regional changes in climate parameters over large time periods is generally absent, which increases the complexity of making appropriate conclusions related to climate change-induced effects in the Himalayan region. This study provides a broad overview of changes in patterns of vegetation, snow covers and temperature in Uttarakhand state of India through bulk processing of remotely sensed Moderate Resolution Imaging Spectroradiometer (MODIS) data, meteorological records and simulated global climate data. Additionally, regression using machine learning algorithms such as Support Vectors and Long Short-term Memory (LSTM) network is carried out to check the possibility of predicting these environmental variables. Results from 17 years of data show an increasing trend of snow-covered areas during pre-monsoon and decreasing vegetation covers during monsoon since 2001. Solar radiation and cloud cover largely control the lapse rate variations. Mean MODIS-derived land surface temperature (LST) observations are in close agreement with global climate data. Future studies focused on climate trends and environmental parameters in Uttarakhand could fairly rely upon the remotely sensed measurements and simulated climate data for the region.
Water is a vital resource supporting a broad spectrum of ecosystems and human activities. The quality of river water has declined in recent years due to the discharge of hazardous materials and toxins. Deep learning and machine learning have gained significant attention for analysing time-series data. However, these methods often suffer from high complexity and significant forecasting errors, primarily due to non-linear datasets and hyperparameter settings. To address these challenges, we have developed an innovative HDTO-DeepAR approach for predicting water quality indicators. This proposed approach is compared with standalone algorithms, including DeepAR, BiLSTM, GRU and XGBoost, using performance metrics such as MAE, MSE, MAPE, and NSE. The NSE of the hybrid approach ranges between 0.8 to 0.96. Given the value's proximity to 1, the model appears to be efficient. The PICP values (ranging from 95% to 98%) indicate that the model is highly reliable in forecasting water quality indicators. Experimental results reveal a close resemblance between the model's predictions and actual values, providing valuable insights for predicting future trends. The comparative study shows that the suggested model surpasses all existing, well-known models.
Iris biometric detection provides contactless authentication, preventing the spread of COVID-19-like contagious diseases. However, these systems are prone to spoofing attacks attempted with the help of contact lenses, replayed video, and print attacks, making them vulnerable and unsafe. This paper proposes the iris liveness detection (ILD) method to mitigate spoofing attacks, taking global-level features of Thepade's sorted block truncation coding (TSBTC) and local-level features of the gray-level co-occurrence matrix (GLCM) of the iris image. Thepade's SBTC extracts global color texture content as features, and GLCM extracts local fine-texture details. The fusion of global and local content presentation may help distinguish between live and non-live iris samples. The fusion of Thepade's SBTC with GLCM features is considered in experimental validations of the proposed method. The features are used to train nine assorted machine learning classifiers, including naïve Bayes (NB), decision tree (J48), support vector machine (SVM), random forest (RF), multilayer perceptron (MLP), and ensembles (SVM + RF + NB, SVM + RF + RT, RF + SVM + MLP, J48 + RF + MLP) for ILD. Accuracy, precision, recall, and F-measure are used to evaluate the performance of the projected ILD variants. The experimentation was carried out on four standard benchmark datasets, and our proposed model showed improved results with the feature fusion approach. The proposed fusion approach gave 99.68% accuracy using the RF + J48 + MLP ensemble of classifiers, immediately followed by the RF algorithm, which gave 95.57%. The better capability of iris liveness detection will improve human-computer interaction and security in the cyber-physical space by improving person validation.
The prediction of suspended sediment load (SSL) within riverine systems is critical to understanding the watershed's hydrology. Therefore, the novelty of our research is developing an interpretable (explainable) model based on deep learning (DL) and Shapley Additive ExPlanations (SHAP) interpretation technique for prediction of SSL in the riverine systems. This paper investigates the abilities of four DL models, including dense deep neural networks (DDNN), long short-term memory (LSTM), gated recurrent unit (GRU), and simple recurrent neural network (RNN) models for the prediction of daily SSL using river discharge and rainfall data at a daily time scale in the Taleghan River watershed, northwestern Tehran, Iran. The performance of models was evaluated by using several quantitative and graphical criteria. The effect of parameter settings on the performance of deep models on SSL prediction was also investigated. The optimal optimization algorithms, maximum iteration (MI), and batch size (BC) were obtained for modeling daily SSL, and structure of the model impact on prediction remarkably. The comparison of prediction accuracy of the models illustrated that DDNN (with R2 = 0.96, RMSE = 333.46) outperformed LSTM (R2 = 0.75, RMSE = 786.20), GRU (R2 = 0.73, RMSE = 825.67), and simple RNN (R2 = 0.78, RMSE = 741.45). Furthermore, the Taylor diagram confirmed that DDNN has the highest performance among other models. Interpretation techniques can address the black-box nature of models, and here, SHAP was applied to develop an interpretable DL model to interpret of DL model's output. The results of SHAP showed that river discharge has the strongest impact on the model's output in estimating SSL. Overall, we conclude that DL models have great potential in watersheds to predict SSL. Therefore, different interpretation techniques as tools to interpret DL model's output (DL model is as black-box model) are recommended in future research.
Wetlands are regarded as one of the most important ecosystems on Earth due to various ecosystem services provided by them such as habitats for biodiversity, water purification, sequestration, and flood attenuation. The Al Hawizeh wetland in the Iran-Iraq border was selected as a study area to evaluate the changes. Maximum likelihood classification was used on the remote sensing data acquired during the period of 1985 to 2013. In this paper, five types of land use/land cover (LULC) were identified and mapped and accuracy assessment was performed. The overall accuracy and kappa coefficient for years 1985, 1998, 2002, and 2013 were 93% and 0.9, 92% and 0.89, 91% and 0.9, and 92% and 0.9, respectively. The classified images were examined with post-classification comparison (PCC) algorithm, and the LULC alterations were assessed. The results of the PCC analysis revealed that there is a drastic change in the area and size of the studied region during the period of investigation. The wetland lost ~73% of its surface area from 1985 to 2002. Meanwhile, post-2002, the wetland underwent a restoration, as a result of which, the area increased slightly and experienced an ~29% growth. Moreover, a large change was noticed at the same period in the wetland that altered ~62% into bare soil in 2002. The areal coverage of wetland of 3386 km(2) in 1985 was reduced to 925 km(2) by 2002 and restored to 1906 km(2) by the year 2013. Human activities particularly engineering projects were identified as the main reason behind the wetland degradation and LULC alterations. And, lastly, in this study, some mitigation measures and recommendations regarding the reclamation of the wetland are discussed. Based on these mitigate measures, the discharge to the wetland must be kept according to the water requirement of the wetland. Moreover, some anthropogenic activities have to be stopped in and around the wetland to protect the ecology of the wetland.
This study has highlighted the trend of recently-reported dengue cases after the implementation of the Movement Control Orders (MCOs) caused due to COVID-19 pandemic in Malaysia. The researchers used the dengue surveillance data published by the Malaysian Ministry of Health during the 3 phases of MCO (which ranged between 17th March 2020 and 28th April 2020) was used for determining the cumulative number of dengue patients. Thereafter, the dengue cases were mapped using the Geographical Information System (GIS). The results indicated that during the 42 days of MCO in Peninsular Malaysia, 11,242 total cases of dengue were reported. The daily trend of the dengue cases showed a decrease from 7268 cases that occurred before the MCOs to 4662 dengue cases that occurred during the initial 14 days of the COVID-19 pandemic (i.e., MCO I), to 3075 cases occurring during the MCO II and 3505 dengue cases noted during MCO III. The central peninsular region showed a maximal decrease in new dengue cases (52.62%), followed by the northern peninsular region (1.89%); eastern coastal region (1.25%) and the southern peninsular region (1.14%) during the initial MCO implementation. However, an increase in the new dengue cases was noted during the MCO III period, wherein all states showed an increase in the new dengue cases as compared during MCO II. The decrease in the pattern was not solely based on the MCO, hence, further investigation is necessary after considering different influencing factors. These results have important implication for future large-scale risk assessment, planning and hazard mitigation on dengue management.
Preparation of natural hazards maps are vital and essential for urban development. The main scope of this study is to synthesize natural hazard maps in a single multi-hazard map and thus to identify suitable areas for the urban development. The study area is the drainage basin of Xerias stream (Northeastern Peloponnesus, Greece) that has frequently suffered damages from landslides, floods and earthquakes. Landslide, flood and seismic hazard assessment maps were separately generated and further combined by applying the Analytical Hierarchy Process (AHP) and utilizing a Geographical Information System (GIS) to produce a multi-hazard map. This map represents the potential suitability map for urban development in the study area and was evaluated by means of uncertainty analysis. The outcome revealed that the most suitable areas are distributed in the southern part of the study area, where the landslide, flood and seismic hazards are at low and very low level. The uncertainty analysis shows small differences on the spatial distribution of the suitability zones. The produced suitability map for urban development proves a satisfactory agreement between the suitability zones and the landslide and flood phenomena that have affected the study area. Finally, 40% of the existing urban pattern boundaries and 60% of the current road network are located within the limits of low and very low suitability zones.
In some parts of tropical Africa, termite mound locations are traditionally used to site groundwater structures mainly in the form of hand-dug wells with high success rates. However, the scientific rationale behind the use of mounds as prospective sites for locating groundwater structures has not been thoroughly investigated. In this paper, locations and structural features of termite mounds were mapped with the aim of determining the aquifer potential beneath termite mounds and comparing the same with adjacent areas, 10 m away. Soil and species sampling, field surveys and laboratory analyses to obtain data on physical, hydraulic and geo-electrical parameters from termite mounds and adjacent control areas followed. The physical and hydraulic measurements demonstrated relatively higher infiltration rates and lower soil water content on mound soils compared with the surrounding areas. To assess the aquifer potential, vertical electrical soundings were conducted on 28 termite mounds sites and adjacent control areas. Three (3) important parameters were assessed to compute potential weights for each Vertical Electrical Sounding (VES) point: Depth to bedrock, aquifer layer resistivity and fresh/fractured bedrock resistivity. These weights were then compared between those of termite mound sites and those from control areas. The result revealed that about 43% of mound sites have greater aquifer potential compared to the surrounding areas, whereas 28.5% of mounds have equal and lower potentials compared with the surrounding areas. The study concludes that termite mounds locations are suitable spots for groundwater prospecting owing to the deeper regolith layer beneath them which suggests that termites either have the ability to locate places with a deeper weathering horizon or are themselves agents of biological weathering. Further studies to check how representative our study area is of other areas with similar termite activities are recommended.
The presence of a well-trained, mobile CNN model with a high accuracy rate is imperative to build a mobile-based early breast cancer detector. In this study, we propose a mobile neural network model breast cancer mobile network (BreaCNet) and its implementation framework. BreaCNet consists of an effective segmentation algorithm for breast thermograms and a classifier based on the mobile CNN model. The segmentation algorithm employing edge detection and second-order polynomial curve fitting techniques can effectively capture the thermograms' region of interest (ROI), thereby facilitating efficient feature extraction. The classifier was developed based on ShuffleNet by adding one block consisting of a convolutional layer with 1028 filters. The modified Shufflenet demonstrated a good fit learning with 6.1 million parameters and 22 MB size. Simulation results showed that modified ShuffleNet alone resulted in a 72% accuracy rate, but the performance excelled to a 100% accuracy rate when integrated with the proposed segmentation algorithm. In terms of diagnostic accuracy of the normal and abnormal test, BreaCNet significantly improves the sensitivity rate from 43% to 100% and specificity of 100%. We confirmed that feeding only the ROI of the input dataset to the network can improve the classifier's performance. On the implementation aspect of BreaCNet, the on-device inference is recommended to ensure users' data privacy and handle an unreliable network connection.
Lung cancer is the leading cause of cancer death and morbidity worldwide. Many studies have shown machine learning models to be effective in detecting lung nodules from chest X-ray images. However, these techniques have yet to be embraced by the medical community due to several practical, ethical, and regulatory constraints stemming from the "black-box" nature of deep learning models. Additionally, most lung nodules visible on chest X-rays are benign; therefore, the narrow task of computer vision-based lung nodule detection cannot be equated to automated lung cancer detection. Addressing both concerns, this study introduces a novel hybrid deep learning and decision tree-based computer vision model, which presents lung cancer malignancy predictions as interpretable decision trees. The deep learning component of this process is trained using a large publicly available dataset on pathological biomarkers associated with lung cancer. These models are then used to inference biomarker scores for chest X-ray images from two independent data sets, for which malignancy metadata is available. Next, multi-variate predictive models were mined by fitting shallow decision trees to the malignancy stratified datasets and interrogating a range of metrics to determine the best model. The best decision tree model achieved sensitivity and specificity of 86.7% and 80.0%, respectively, with a positive predictive value of 92.9%. Decision trees mined using this method may be considered as a starting point for refinement into clinically useful multi-variate lung cancer malignancy models for implementation as a workflow augmentation tool to improve the efficiency of human radiologists.
This study evaluates state-of-the-art machine learning models in predicting the most sustainable arsenic mitigation preference. A Gaussian distribution-based Naïve Bayes (NB) classifier scored the highest Area Under the Curve (AUC) of the Receiver Operating Characteristic curve (0.82), followed by Nu Support Vector Classification (0.80), and K-Neighbors (0.79). Ensemble classifiers scored higher than 70% AUC, with Random Forest being the top performer (0.77), and Decision Tree model ranked fourth with an AUC of 0.77. The multilayer perceptron model also achieved high performance (AUC=0.75). Most linear classifiers underperformed, with the Ridge classifier at the top (AUC=0.73) and perceptron at the bottom (AUC=0.57). A Bernoulli distribution-based Naïve Bayes classifier was the poorest model (AUC=0.50). The Gaussian NB was also the most robust ML model with the slightest variation of Kappa score on training (0.58) and test data (0.64). The results suggest that nonlinear or ensemble classifiers could more accurately understand the complex relationships of socio-environmental data and help develop accurate and robust prediction models of sustainable arsenic mitigation. Furthermore, Gaussian NB is the best option when data is scarce.