The evolution of saltating tracks driven by strong wind remains unknown due to the low accuracy or recall rates of saltating particle tracking algorithms (SPTs). Manual identification of saltating tracks becomes a primary bottleneck because of low efficiency, restricting the development of new SPTs with high accuracy. Herein, we proposed an optimized tree model for automatically identifying saltating tracks in the high-speed video under strong wind through establishing the dataset with multiple statistical quantities of instant saltating velocity (MSQV) and the workflow embracing the Tree-structured Parzen Estimator (TPE). The optimized Categorical Boosting model by the D3 dataset (CatBoost-D3) could be considered the best classifier among the tree models, owning the higher accuracy (0.9352), precision (0.9348), recall (0.9352), F1-score (0.9350) and area under an receiver operating characteristics curve (AUC, 0.9730), and lower time cost. The best performances were associated with the ensemble effect of critical and secondary features, distinct from the previous finding which revealed only the effect of critical features on enhancing AUC value. Additionally, one observed that the present model was comparable to other optimized tree model by the dataset with double-class and outperformed the other tree model by the dataset with multi-class. The present work offers a new avenue for identifying hop trajectories and tracking sand particle flow via machine learning in the future, and a new channel for reunderstanding the relationship between midair collision and saltation under strong wind through automatic identification of saltating tracks.
The grain size characteristics of aeolian sediments are the combined result of sand sources, regional airflow regimes, dune morphology, etc., and are essential for understanding the formation and evolution of barchan dunes. Based on field investigations and laboratory experimental data, in this paper, we explored differences in the grain size characteristics of the surface sediments of barchan dunes at the southeastern edge of the Taklimakan Desert (TKLM-SE), the western (QB-W) and southern parts of the Qaidam Basin Desert (QB-S), the sand belt connecting the Badain Jaran Desert and the Tengger Desert (BT-B), and their responses to sand sources, dune morphology, and wind regimes. The main results were as follows: (i) The mean grian size distribution patterns of the windward slope toe to the leeward slope toe through the dune crest/ridge varied with different transects of the barchan dunes and different deserts, showing four types including gradually fining (GF), gradually coarsening (GC), coarsening followed by fining (CF), and fining followed by coarsening (FC). The patterns were GF and CF in the TKLM-SE; GF, GC, and FC in the BT-B; GF, GC, and CF in the QB-W; and GF in the QB-S. (ii) The interdune sediments provided the source material for the formation and development of barchan dunes and their grain size varied in different deserts. The interdune sediments were composed of gravel and very fine sand in the TKLM-SE, while they were composed of medium and fine sand in the QB-W, QB-S, and BT-B. (iii) The windward side of the barchans varied with different wind directions, and dune height affected dune surface airflow velocity and direction, changing the pattern of grain size distribution on the dune surface. The wind regime over a ten-day or half-month scale could explain the variance in the grain size distribution patterns better than that on an annual scale. (iv) Grain size characteristics of dune surface sands changed with dune shape due to changes in the surface airflow velocity and direction and the sediment-carrying capacity of the airflow. With an increasing ratio of dune height to dune width, the grain size of the dune crest sands became coarser. These results help advance our understanding of the grain size characteristics of barchan dunes and regional variabilities in their patterns.
Dust particles are considered as a very important way of soil contamination by heavy metals. Therefore, this study was conducted to evaluate the concentration of heavy metals and their health risk in windborne sediments. For this purpose, sediment traps were installed in five dominant wind directions including north, northeast, northwest, west, and southwest, and center of Tabas city (Iran) to collect the suspended sediments in the air. Sediment sampling was conducted monthly from January to December 2021. The concentrations of heavy metals were measured using atomic adsorption method followed by extraction by aqua regia, and the carcinogenic and non-carcinogenic health risks of heavy metals for children and adults were evaluated during different months of the year. According to the results, the maximum and minimum amounts of windborne sediments found in northwest (85.66 gm−2) and west (29.3 gm−2), respectively. Monthly variations in windborne sediments discharge also revealed that the maximum amounts of windborne sediments occurred in September and November 2021 from northeast (125 and 117 mgkg−1, respectively). The maximum concentrations of cadmium (0.82 mg/kg) were found in the west of Tabas, while those of lead (192.72 mg/kg), and nickel (227.34 mg/kg) were obtained in the city center. In addition, the highest carcinogenic and non-carcinogenic risks belonged to nickel and the lowest risks were obtained for lead. Also, the carcinogenic risk of cadmium was higher than lead but lower than nickel. In general, the carcinogenic and non-carcinogenic risks of the studied heavy metals were low (less than 1).
Dust events are caused by strong winds that lift dust particles into the air. Due to surrounding deserts and agriculture, West Texas experiences many dust events. This study examines dust events that occurred between 2000 and 2020 across four locations: Amarillo, Lubbock, Midland, and El Paso. A total of 1,834 dust events were identified across the four locations with an average of 22 dust events annually. 227 dust events were observed in Amarillo, 609 in Lubbock, 545 in Midland, and 453 dust events were observed in El Paso. A slight increasing trend of dust events over time was observed for Amarillo, Lubbock, and Midland while El Paso showed a decreasing trend. Most dust events occurred during the spring to early summer months and they lasted an hour or less. Many dust events occurred during times of drought and periods of La Niña. Separation of the dust events based on the meteorological disturbance that caused them (convective vs. synoptic) showed that synoptic disturbances contribute to >60 % of the dust events, while convective disturbances were responsible for most of the remaining. Synoptic disturbances were predominately in spring while convective disturbances were common in the early summer months. A comparison of meteorological parameters measured during each disturbance shows that synoptic dust events were associated with lower temperatures, dew point, and relative humidity, but with higher wind speeds and gusts.
The evolution of large star dunes, because of their remote location, size and surface complexity, is barely recorded and understood. This lack of understanding applies to surface and subsurface features alike. In order to detect the transformation, the detailed subsurface stratigraphy and the relative chronology of large star dunes, we used ground penetrating radar (GPR) on all major arms of a complex star dune of Erg Chebbi, south-eastern Morocco. We used a 350 MHz digital antenna from Geophysical Survey Systems, Inc (GSSI), reaching a depth of 12.5 m to identify main radar facies associated with former downwind dune flanks describing the depositional history. Our results enable the determination of former dune crest positions, their potential past movement and in consequence the construction of the paleo-dune topography. In accordance with simulated historical wind data, we found a potential sediment deficit on the south-eastern side of the dune. This also correlates with surface data describing an oblique form of the star dune and the spatial distribution frequency of its major arms. Our detailed recordings show, for the first time, the complex internal composition of all arms of one large star dune and surface sensitive form-flow interactions. Our results allow the discrimination of deposition phases and therefore, we have constructed a relative chronology as a basis for future sampling and the reconstruction of star dune evolution in general.