This study harnessed the formidable predictive capabilities of three state-of-the-art machine learning models—extreme gradient boosting (XGB), random forest (RF), and CatBoost (CB)—applying them to meticulously curated datasets of topographical, geological, and environmental parameters; the goal was to investigate the intricacies of flood susceptibility within the arid riverbeds of Wilayat As-Suwayq, which is situated in the Sultanate of Oman. The results underscored the exceptional discrimination prowess of XGB and CB, boasting impressive area under curve (AUC) scores of 0.98 and 0.91, respectively, during the testing phase. RF, a stalwart contender, performed commendably with an AUC of 0.90. Notably, the investigation revealed that certain key variables, including curvature, elevation, slope, stream power index (SPI), topographic wetness index (TWI), topographic roughness index (TRI), and normalised difference vegetation index (NDVI), were critical in achieving an accurate delineation of flood-prone locales. In contrast, ancillary factors, such as annual precipitation, drainage density, proximity to transportation networks, soil composition, and geological attributes, though non-negligible, exerted a relatively lesser influence on flood susceptibility. This empirical validation was further corroborated by the robust consensus of the XGB, RF and CB models. By amalgamating advanced deep learning techniques with the precision of geographical information systems (GIS) and rich troves of remote-sensing data, the study can be seen as a pioneering endeavour in the realm of flood analysis and cartographic representation within semiarid fluvial landscapes. The findings advance our comprehension of flood vulnerability dynamics and provide indispensable insights for the development of proactive mitigation strategies in regions that are susceptible to hydrological perils.

Coronavirus disease (COVID-19), caused by acute respiratory syndrome coronavirus 2 (SARS-CoV-2), is a worldwide challenge effecting millions of people in more than 210 countries, including the Sultanate of Oman (Oman). Spatiotemporal analysis was adopted to explore the spatial patterns of the spread of COVID-19 during the period from 29th April to 30th June 2020. Our assessment was made using five geospatial techniques within a Geographical Information System (GIS) context, including a weighted mean centre (WMC), standard deviational ellipses, Moran’s I autocorrelation coefficient, Getis-Ord General-G high/low clustering, and Getis-Ord $$G_{i}^{*}$$G i ∗ statistic. The Moran’s I-/G- statistics proved that COVID-19 cases in datasets (numbers of cases) were clustered throughout the study period. The Moran’s I and Z scores were above the 2.25 threshold (a confidence level above 95%), ranging from 2274 cases on 29th April to 40,070 cases on 30th June 2020. The results of $$G_{i}^{*}$$G i ∗ showed varying rates of infections, with a large spatial variability between the different wilayats (district). The epidemic situation in some wilayats, such as Mutrah, As-Seeb, and Bowsher in the Muscat Governorate, was more severe, with Z score higher than 5, and the current transmission still presents an increasing trend. This study indicated that the directional pattern of COVID-19 cases has moved from northeast to northwest and southwest, with the total impacted region increasing over time. Also, the results indicate that the rate of COVID-19 infections is higher in the most populated areas. The findings of this paper provide a solid basis for future study by investigating the most resolute hotspots in more detail and may help decision-makers identify targeted zones for alleviation plans.

Recurring wildfires pose a critical global issue as they undermine social and economic stability and jeopardize human lives. To effectively manage disasters and bolster community resilience, the development of wildfire susceptibility maps (WFSMs) has emerged as a crucial undertaking in recent years. In this research endeavor, two deep learning algorithms were leveraged to generate WFSMs using two distinct remote sensing datasets. Specifically, the Moderate-Resolution Imaging Spectroradiometer (MODIS) and Landsat-8 images were utilized to monitor wildfires that transpired during the year 2021. To develop an effective WFSM, two datasets were created by incorporating 599 wildfire locations with Landsat-8 images and 232 sites with MODIS images, as well as twelve factors influencing wildfires. Deep learning algorithms, namely the long short-term memory (LSTM) and recurrent neural network (RNN), were utilized to model wildfire susceptibility using the two datasets. Subsequently, four WFSMs were generated using the LSTM (MODIS), LSTM (Landsat-8), RNN (MODIS), and RNN (Landsat-8) algorithms. The evaluation of the WFSMs was performed using the area under the receiver operating characteristic (ROC) curve (AUC) index. The results revealed that the RNN (MODIS) (AUC = 0.971), RNN (Landsat-8) (AUC = 0.966), LSTM (MODIS) (AUC = 0.964), and LSTM (Landsat-8) (AUC = 0.941) algorithms demonstrated the highest modeling accuracy, respectively. Moreover, the Gini index was employed to assess the impact of the twelve factors on wildfires in the study area. The results of the random forest (RF) algorithm indicated that temperature, wind speed, slope, and topographic wetness index (TWI) parameters had a significant effect on wildfires in the study region. These findings are instrumental in facilitating efficient wildfire management and enhancing community resilience against the detrimental effects of wildfires.

Aflaj (plural of falaj) are tunnels or trenches built to deliver groundwater from its source to the point of consumption. Support vector machine (SVM) and extreme gradient boosting (XGB) machine learning models were used to predict groundwater aflaj potential in the Nizwa watershed in the Sultanate of Oman (Oman). Nizwa city is a focal point of aflaj that underlies the historical relationship between ecology, economic dynamics, agricultural systems, and human settlements. Three hyperparameter algorithms, grid search (GS), random search (RS), and Bayesian optimisation, were used to optimise the parameters of the XGB model. Sentinel-2 and light detection and ranging (LiDAR) data via geographical information systems (GIS) were employed to derive variables of land use/land cover, and hydrological, topographical, and geological factors. The groundwater aflaj potential maps were categorised into five classes: deficient, low, moderate, high, and very high. Based on the evaluation of accuracy in the training stage, the following models showed a high level of accuracy based on the area under the curve: Bayesian-XGB (0.99), GS-XGB (0.97), RS-XGB (0.96), SVM (0.96), and XGB (0.93). The validation results showed that the Bayesian hyperparameter algorithm significantly increased XGB model efficiency in modelling groundwater aflaj potential. The highest percentages of groundwater potential in the very high class were the XGB (10%), SVM (8%), GS-XGB (6%), RS-XGB (6%), and Bayesian-XGB (6%) models. Most of these areas were located in the central and northeast parts of the case study area. The study concluded that evaluating existing groundwater datasets, facilities, current, and future spatial datasets is critical in order to design systems capable of mapping groundwater aflaj based on geospatial and ML techniques. In turn, groundwater protection service projects and integrated water source management (IWSM) programs will be able to protect the aflaj irrigation system from threats by implementing timely preventative measures.

The aflaj systems represent unique irrigation technologies that have been implemented in the Sultanate of Oman. This innovative system, referred to as “falaj” in the singular form, is composed of a sophisticated network of underground tunnels and open-air channels designed to access shallow subterranean water tables, thereby providing water for residential and agricultural use. The aflaj systems have played a significant role in supporting sustainable water resource management in arid and semiarid regions, making a notable contribution to the socioeconomic development of the country. The alteration of land use and land cover (LULC) in arid and semiarid regions can have significant consequences for hydrological systems, affecting the ability of local ecosystems to manage fresh surface and groundwater resources. These changes are often caused by both natural and anthropogenic factors. To investigate the impact of LULC changes on aflaj systems in the northern part of Oman, we utilized satellite imagery, aflaj data, and spatial analytical and image processing techniques within the framework of geographic information systems (GIS) and remote sensing. In the first part of the study, we quantified the changes in LULC and their impact on aflaj systems in seven cities in Oman due to urban expansion. In the second part, we evaluated the effect of LULC on groundwater for four major aflaj between 1985 and 2021. The study area was divided into four primary LULC classifications: vegetation, bodies of water, metropolitan areas, and bare soil. The classification maps demonstrated a high overall accuracy of 90% to 95%, indicating satisfactory performance. Our results revealed a significant reduction in vegetation areas between 1985 and 2021, primarily shifting from bare soil (BS) to urban areas (UAs) and from vegetation cover (VC) to BS, due to the reduction of groundwater resources. Over the four study periods (1985–1990, 1990–2000, 2000–2013, and 2013–2021), the percentages of the total area of Falaj Al-Muyasser, Falaj Daris, Falaj Al-Maliki, and Falaj Al-Khatmeen that transformed from agricultural lands to UAs were 40%, 39%, 32%, and 8%, respectively. Our study highlights the need for appropriate land management and planning to ensure the most effective solutions are utilized to meet social and economic sustainability requirements. In conclusion, our study presents a comprehensive analysis of LULC changes and their impact on aflaj systems over a 36-year period, providing new insights into the potential effects of LULC changes on groundwater resources and offering a basis for informed decision making on land management in arid and semiarid areas.

Climate change poses a major threat to vegetation and land cover worldwide. The loss of vegetation as a result of climate change can alter the functions and structure of the environment and its ecological systems. In the first part of this study, Sentinel-2 data, normalised different vegetation index (NDVI), and multiple regression methods were used to examine the impacts of the climatic factors of humidity, rainfall, and air temperature on vegetation dynamics from 2015 to 2021 in Dhofar, Southern Oman. In the second part of this study, random forest regression was employed to model the relationships between the NDVI and temperature, humidity, rainfall, soil map, geology map, topographic wetness index, curvature, elevation, slope, aspect, distance to buildings, and distance to roads. The multiple regression values revealed significant associations between the spatial distributions of the NDVI and the abovementioned climatic factors. The findings also indicated an increase of 1 °C in air temperature fluctuations between 2018 and 2021 over all five sites, with a strong tendency over Qairoon Hairiti Mountain. The rainfall records clearly indicated an increasing tendency from 2018 to 2020 due to the impact of frequent cyclones. Therefore, the results revealed a significant increase of 0.01 in the vegetation cover trend in 2018, 2019, and 2020 along the Sadah Mountain range and the eastern part of the Jabal Qara Mountains under the areas directly impacted by the cyclone, whereas there was a decrease along the western mountain range consisting of Jabal Qara and Jabal Qamar Mountains due to the impact of warm, dry air. The results revealed that NDVI values were sensitive to heavy rainfall over Jabal Samhan Mountain. The 12 variables that influenced NDVI levels had different levels of importance. Soil types, elevation, slope, rainfall, curvature, humidity, and temperature had the highest importance, while topographic wetness index, distance to urban area, aspect, distance to roads, and geology map had the lowest. The findings provide a significant foundation for Oman’s planning and management of regional vegetation, water conservation, and animal husbandry.