The Indochina block, the most significant part of South East Asia, is bounded by the Ailaoshan-Song Ma suture zone in the north and Jinghong-Nan-Uttradit, Sra Kaeo suture in the west. These sutures result from the various tectonic events related to the rifting and closing of the Paleo-Tethys and successive collision of South China (Sino-Vietnam)-Indochina-Sukhothai arc-Sibumasu continental blocks. However, the timing of the commencement of the rifting eastern Paleo-Tethys, subduction, subsequent collision orogeny, duration of these stages, and their geodynamic mechanism remain controversial. Our review of the stratigraphic, sedimentary, magmatic, metamorphic, structural, and geochronological data available for the suture zones bounding and interior of the Indochina block indicate successive tectonic events from rifting through subduction to orogeny and provide new insights into the tectonic history from Late Devonian to Triassic that shape the Indochina block and its adjacent regions. The remnant of the Paleo-Tethys branch and deep marine cherts represented by the Ailaoshan-Song Ma and Nan-Uttaradit suture melanges record the rifting and oceanic crust formation stage at ~380-313 Ma and 316-310 Ma, respectively. The southwestward subduction of the South China block under the Indochina block resulted in the development of the magmatic arc at ~ 310-247 Ma in the north Indochina block; meanwhile, the subduction of the Sibumasu block below the Indochina gave rise to the Sukhothai magmatic arc and Jinghong-Nan-Luang Prabang-Loei back-arc basins in the western Indochina at ~278-235 Ma. The South China slab rollback may significantly affect arc magmatism distribution in the north Indochina block. The final closure of the eastern Paleo-Tethys resulted in Indosinian collision orogeny of the Indochina with South China blocks at ~250-230 Ma in the Ailaoshan-Song Ma suture, followed by the Sibumasu block with Indochina-South China continent at ~240-230 Ma in the Jinghong-Nan-Uttradit-Sra Kaeo sutures. The late to post-collisional orogeny occurred in the north and west Indochina at ~235-200 Ma and ~230-200 Ma, respectively.

Advanced machine learning and deep Learning modeling applications for landslide susceptibility mapping are becoming increasingly popular. This study applied a deep learning model (DL) with a multilayer neural network to landslide research in the Phuoc Son district, Quang Nam province. Two methods for selecting conditioning factors, Correlation Attribute and OneR, were used to choose 12 condition parameters for landslides (Slope, Relief, Elevation, Distance to road, Rainfall, Land use, Weathering crust, Geology, Aspect, Soil, Distance to fault, and Curvature). Comparing the predicted results with two standard models, Naïve Bayes (NB) and Support Vector Machine (SVM), showed that the DL model has higher and better prediction performance. Accordingly, the prediction performance of the DL model on the training dataset was ACC = 92.12%, AUC = 0.970, and on the validation dataset was ACC = 87.52, AUC = 0.944. The LSM developed based on the DL model indicates that areas with high landslide susceptibility are primarily concentrated in the southern part of the study area. These findings could be highly beneficial for urban planning management, risk management, and efforts to prevent and mitigate the damage caused by landslides in Phuoc Son.

The Lung Po basalt, dating to 0.93, 0.81, and 0.47 Ma, occupies approximately 1 km² and is situated 12 km west of the Ailao Shan Red River Shear Zone (ASRRSZ) and about 65 km south of the 12-0 Ma Maguan intraplate volcanic area in southwest Yunnan (SW China), within the ASRRSZ. This olivine-bearing phyric alkaline basalt is characterized by high TiO₂ (around 2.3 wt.%), MgO (8-10 wt.%), and K₂O (approximately 2.8-3 wt.%) with Na₂O/K₂O ratios ranging from 1 to 1.2. These features partially overlap with the Maguan mantle xenolith-bearing alkaline basalt but are distinct from the Pleistocene alkaline basalt of Vietnam's Western Highlands. The Lung Po basalts exhibit a typical oceanic island basalt (OIB) trace element distribution pattern and a 'crossing' rare earth element (REE) pattern, indicating magma generation possibly by melting of garnet peridotite. They have high ⁸⁷Sr/⁸⁶Sr isotopic ratios (around 0.706) and low ¹⁴³Nd/¹⁴⁴Nd ratios (approximately 0.5126), along with moderate ²⁰⁶Pb/²⁰⁴Pb and ¹⁷⁶Hf/¹⁷⁷Hf isotopic ratios (respectively <18.3-18.4 and 0.28295-0.2830). These isotopic characteristics, coupled with OIB trace element features, challenge the involvement of crustal material. The Lung Po Pleistocene basalt and the 12-0 Ma Maguan alkaline basalt differ significantly from the 42-24 Ma post-collision high-K magmas in the ASRRSZ, which are associated with crustal tectonic processes. Instead, the Lung Po (and Maguan) basalt likely originated from a newly emplaced, metasomatically altered fertile asthenosphere following localized lithosphere extension and delamination after the India-Eurasian collision events.

The tragic COVID-19 pandemic, while presenting numerous devastating consequences, has inadvertently provided a unique opportunity for studying air pollution. In this study, we specifically evaluate the spatiotemporal changes in aerosols before and during the COVID-19 lockdown from March to May 2020 over Northwest India, with a particular focus on two subregions in the vicinity of Delhi: Delhi-West and Delhi-East. The assessment was conducted using aerosol optical depth (AOD) data from the Moderate Resolution Imaging Spectroradiometer (MODIS) mission, aerosol profiles from the Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observation (CALIPSO) mission, and ground-based measurements of PM2.5 and PM10. Results demonstrated evident reductions in surface particulate matter and AOD during the lockdown. Approximately 40% of the contribution to the total AOD was from aerosols below 1 km. Different rates of change in AOD were observed for the two subregions across different lockdown phases, attributed to differences in emission sources: Delhi-East is more influenced by residential emissions, while Delhi-West is more affected by natural sources. The surface concentration and variabilities of PM2.5 and PM10 confirmed the differences between the two subregions, emphasizing the role of anthropogenic activities in PM2.5 emissions over the study area.

This paper presents the first observations of the occurrence rates of Spread F and GPS total electron content (TEC) index (ROTI) over Vietnam at the equatorial trough and the northern tropical crest of ionization anomaly in the Asian sector. The data have been examined for the monthly and nighttime variations in the occurrence of these two data at Bac Lieu (9.28°N, 105.73°E, dip: 1.73°N) and Phu Thuy (21.03°N, 105.95°E, dip: 14.49°N) during 2023. For Bac Lieu, the monthly variation in the occurrence of the range Spread F (RSF) has the maxima in the February, May, and September months, while the mixed Spread-F (MSF) and ROTI occurrences exhibit a semiannual asymmetry with peaks in March/April and October. For the nighttime variation, occurrence peaks at 1915-1930 LT for RSF, at about 1945-2100 LT for MSF, and between 2030-2330 LT for ROTI. Regarding the frequency Spread F (FSF) occurrence, the maximum values in the monthly variation are in April, and the nighttime variation peaks at about 2115-2315 LT. For Phu Thuy, the monthly variation of RSF, MSF, and ROTI occurrences also exhibit a semiannual asymmetry with peaks in March/April and October. These peak magnitudes are largest for ROTI, moderate for MSF, and smallest for RSF. The nighttime variation of RSF, MSF, and ROTI occurrence peaks show intense season changes from winter to autumn at pre-midnight, spring at post-midnight, and summer at post-midnight. The FSF occurrences are more significant in summer than in other seasons, mainly after midnight. The time order appearance of the Spread F types at Bac Lieu and Phu Thuy is first of RSF, then MSF, and finally FSF. This could reflect that the formation mechanisms of Spread F types are different and require further research. Our observations also showed that the post-midnight occurrence of Spread F is much larger than ROTI at Bac Lieu and Phu Thuy. The monthly variations in occurrence rates of Spread F and ROTI at Bac Lieu and Phu Thuy are similar, but these occurrence rates at Bac Lieu are usually larger than at Phu Thuy.

This study utilized the Weather Research and Forecast (WRF) model to forecast hail induced by the hailstorms on 17 March 2020 in western North Vietnam, using two microphysical schemes: the Thompson and Morrison schemes. Assessment of the WRF skill in predicting hail coverage and intensity was done for two predicted indices, namely UH (Updraft Helicity) and CTG (Column Total Integrated Graupel). Two predicted variables are DTh (hail diameter given by WRF using the Thompson Hail Algorithm) and DHc (hail diameter given by the HAILCAST submodel in WRF). The predicted hail coverage and intensity were compared with the products given by the Pha Din radar's Hail Size Discrimination Algorithm (HSDA) for three categories: small, large, and giant hail size. Using the Morrison scheme, the WRF model indicates that the hail-coverage forecast skills of UH, CTG, and DHc are highest, with an insignificant difference at the horizontal scale more significant than 60 km. However, the DHc variable given by the Morrison scheme provides the most successful forecast for both hail size and coverage compared with the HSDA products and field reports. This is because HAILCAST considers kinematic and microphysical processes to predict maximum hail size at the surface. The predicted hailstorms could occur in environments with moderate convective available potential energy but require robust moisture flux convergence over high mountains.

Hyperspectral imagery obtained from Unmanned Aerial Vehicles (UAVs) is increasingly employed to investigate nutrient concentrations in vegetation. The deployment of a hyperspectral camera on a UAV, flight planning, image acquisition, preprocessing of hyperspectral data, and the subsequent estimation of nutrient concentrations in vegetation are facing challenges. These challenges manifest as geometric, spectral distortions, and the abundance of numerous spectral bands. This study seeks to guide on mitigating the impact of issues encountered during an experiment to estimate nutrient concentrations in rice leaves using UAV hyperspectral images. An industrial hexagonal drone equipped with a push-broom hyperspectral camera featuring 122 bands within the Visible to Near-Infrared (VIS-NIR) wavelength range (400-960 nm) is employed to collect data over a 1-hectare testing rice field. Models for estimating Leaf Phosphorus Concentration (LPC) and Leaf Potassium Concentration (LKC) are developed based on the correlation between hyperspectral images, characterized by a 3 cm spatial resolution, and 162 LPC and 162 LKC reference data points. Utilizing various vegetation indices for LPC and LKC estimation, the outcomes reveal that a combination of band wavelengths at 838 nm and 734 nm is effective for LPC estimation, yielding a Root Mean Square Error (RMSE) of 27.1%. Conversely, LKC estimation exhibits an RMSE of 38.8% with an insignificant correlation between LKC and the current wavelength ranges. Above all, this study is a primary example of the utilization of UAV hyperspectral data in precision agriculture in Vietnam.

The prediction of permeability in porous media is a critical aspect in various scientific and engineering applications. This paper presents a machine learning (ML) model based on the XGBoost algorithm for predicting the permeability of porous media using microstructure characteristics. The seahorse optimization algorithm was employed to fine-tune the hyperparameters of the XGBoost algorithm, resulting in a model with predictive solid capabilities. Regression analysis and residual errors indicated that the model achieved good prediction results on the training and testing datasets, with RMSE values of 0.0494 and 0.0826, respectively. A SHAP value sensitivity analysis revealed that the essential inputs were the size of the inclusions, with the quantiles representing the maximum size of the inclusions being the most significant variables affecting permeability. The findings of this study have important implications for the design and optimization of porous media, and the XGBoost algorithm-based ML model provides a fast and accurate tool for predicting the permeability of porous media based on microstructure characteristics.

In this study, the main active faults in the territory of Laos were identified by analyzing the spatial relationship between the distributions of neo-tectonic faults and earthquake epicenters. The map of neo-tectonic faults was built by integrating the results of neo-tectonic faults research using geological-geomorphological data together with the lineament map obtained from remote sensing analysis. Nontectonic lineaments were eliminated by correlating the spatial distribution of the lineament field with a topographic map, DEM, and geological-geomorphological data. The earthquake data, including 4416 events in Laos and its surroundings, were collected from different sources: the International Seismological Center (ISC), the earthquakes recorded by the local seismic network in Laos, the seismic data in Vietnam, and the earthquake catalog provided by the Thailand Meteorological Department (TMD). Among these, 820 events were located using the hypocenter method, and the local network recorded the data. The magnitude conversion was applied to get a unified scale Mw. The catalog of 1617 main shocks obtained after eliminating foreshocks and aftershocks using the declustering technique was used for a spatial correlation with the neotectonic fault distribution to identify active faults. A total of 14 main active fault zones in the Laos territory were defined. Most are also seismogenic faults with Mw ≥ 5.0 occurring along their trace. Considering the characteristics of seismic activity and the active and neotectonic faults, the territory of Laos can be divided into six seismotectonic zones according to the decreasing level of seismic activity: the Western, the Northeastern Samnua, the Phongsali, the South Truong Son, the North Truong Son, and the Khorat zones. Each zone is characterized by relative homogeneity in the seismic activity and the characteristics of active and neotectonic faults.

This study evaluates the efficacy of five machine learning algorithms Support Vector Regression (SVR), Decision Tree (DT), Random Forest (RF), Light Gradient Boosting Machine Regressor (LGBM), and Linear Regression (LR) in predicting water levels in the Vietnamese Mekong Delta's tidal river system, a complex nonlinear hydrological phenomenon. Using daily maximum, minimum, and mean water level data from the Cao Lanh gauging station on the Tien River (2000-2020), models were developed to forecast water levels one, three, five, and seven days in advance. Performance was assessed using Nash-Sutcliffe Efficiency, coefficient of determination, Root Mean Square Error, and Mean Absolute Error. Results indicate that all models performed well, with SVR consistently outperforming others, followed by RF, DT, and LGBM. The study demonstrates the viability of machine learning in water level prediction using solely historical water level data, potentially enhancing flood warning systems, water resource management, and agricultural planning. These findings contribute to the growing knowledge of machine learning applications in hydrology and can inform sustainable water resource management strategies in delta regions.