Beyond Conventional Monitoring: A Semantic Segmentation Approach to Quantifying Traffic-Induced Dust on Unsealed Roads

Road dust is a mixture of fine and coarse particles released into the air due to an external force, such as tire-ground friction or wind, which is harmful to human health when inhaled. Continuous dust emission from the road surfaces is detrimental to the road itself and the road users. Due to this, multiple dust monitoring and control techniques are currently adopted in the world. The current dust monitoring methods require expensive equipment and expertise. This study introduces a novel pragmatic and robust approach to quantifying traffic-induced road dust using a deep learning method called semantic segmentation. Based on the authors' previous works, the best-performing semantic segmentation machine learning models were selected and used to identify dust in an image pixel-wise. The total number of dust pixels was then correlated with real-world dust measurements obtained from a research-grade dust monitor. Our method shows that semantic segmentation can be adopted to quantify traffic-induced dust reasonably. Over 90% of the predictions from both correlations fall in true positive quadrant, indicating that when dust concentrations are below the threshold, the segmentation can accurately predict them. The results were validated and extended for real-time application. Our code implementation is publicly available.

Long-term variations of global dust emissions and climate control

Dust discharged from the surface into the air has an important impact on global climate change, the ecological environment, and human health. However, the spatiotemporal variations of global dust emissions and the climate control of dust emissions from different dust sources in recent decades are still unclear. This study explores the spatiotemporal variations of global dust emissions from 1980 to 2020 based on the MERRA-2 dust emissions dataset and provides a detailed investigation of the interannual variations of dust emissions from major dust sources in the world and their contribution to the global dust cycle. On this basis, the association between global dust emissions and average wind speed (AWS), surface air temperature (SAT), precipitation (Ppt), relative humidity (RH), soil evaporation (SE), soil moisture (SM), and solar radiation (SR) were explored. In particular, the comparative importance of these climatic factors and their combined structures on dust emissions from different dust sources. The results show that North Africa contributed the most to global dust emissions, contributing 58% of the total global emissions, while South Africa and North America contributed the least to global dust emissions, at less than 1%, respectively. Classification and Regression Tree (CART) analysis shows that SR was the major factor affecting the dust emissions of Australia, East Asia, South America, and Central Asia. AWS was the major factor influencing dust emissions in North Africa and South Asia. SAT, RH, and SM were the major factors affecting dust emissions in West Asia, North America, and South Africa, respectively. There were great differences in the climatic factors combinations on dust emissions intensity in different dust sources. These findings assist us in better understanding the control of climatic factors on dust emissions from global dust sources and have important scientific significance for accurately predicting dust events and reducing disaster risks.

Discrete Element Simulation on Sand-Bed Collision Considering Surface Moisture Content

The process of aeolian sand transport is an important mechanism leading to the formation and evolution of local landforms in coastal areas and desert lakes. For a long time, the role of surface moisture in incipient motion of sand grains by wind stress has been extensively studied but, in fact, sand-bed collision is the main mechanism in steady aeolian sand flow. At present, the lack of understanding of surface moisture content on sand-bed collision limits the application of aeolian sand transport models in wet coastal areas. In this paper, we adopt numerical simulations to discuss and analyze the effect of cohesive forces formed by surface moisture content on the sand-bed collision process based on discrete element method. High density contact forces appear with the surface moisture increasing, and form a closed structure around the edge of crater to resist the avulsion in horizontal direction. Under high moisture condition, even though the ejected sand grains saltate away from the surface, the tension forces will prevent from leaving. The ejected number trend with incident velocity shows some nonlinear characteristics due to the unequally distributed force chains and liquid bridges in the unsaturated sand bed surface.

Physical Crust Formation on Sandy Soils and Their Potential to Reduce Dust Emissions from Croplands

The sandy croplands in the Free State have been identified as one of the main dust sources in South Africa. The aim of this study was to investigate the occurrence and strength of physical soil crusts on cropland soils in the Free State, to identify the rainfall required to form a stable crust, and to test their impact on dust emissions. Crust strength was measured using a fall cone penetrometer and a torvane, while laboratory rainfall simulations were used to form experimental crusts. Dust emissions were measured with a Portable In-Situ Wind Erosion Laboratory (PI-SWERL). The laboratory rainfall simulations showed that stable crusts could be formed by 15 mm of rainfall. The PI-SWERL experiments illustrated that the PM10 emission flux of such crusts is between 0.14% and 0.26% of that of a non-crusted Luvisol and Arenosol, respectively. The presence of abraders on the crust can increase the emissions up to 4% and 8% of the non-crusted dust flux. Overall, our study shows that crusts in the field are potentially strong enough to protect the soil surfaces against wind erosion during a phase of the cropping cycle when the soil surface is not protected by plants.

Stepwise Assessment of Different Saltation Theories in Comparison with Field Observation Data

Wind-blown dust models use input data, including soil conditions and meteorology, to interpret the multi-step wind erosion process and predict the quantity of dust emission. Therefore, the accuracy of the wind-blown dust models is dependent on the accuracy of each input condition and the robustness of the model schemes for each elemental step of wind erosion. A thorough evaluation of a wind-blown model thus requires validation of the input conditions and the elemental model schemes. However, most model evaluations and intercomparisons have focused on the final output of the models, i.e., the vertical dust emission. Recently, a delicate set of measurement data for saltation flux and friction velocity was reported from the Japan-Australia Dust Experiment (JADE) Project, which enabled the step-by-step evaluation of wind-blown dust models up to the saltation step. When all the input parameters were provided from the observations, both the two widely used saltation schemes showed very good agreement with measurements, with the correlation coefficient and the agreement of index both being larger than 0.9, which demonstrated the strong robustness of the physical schemes for saltation. However, using the meteorology model to estimate the input conditions such as weather and soil conditions, considerably degraded the models’ performance. The critical reason for the model failure was determined to be the inaccuracy in the estimation of the threshold friction velocity (representing soil condition), followed by inaccurate estimation of surface wind speed. It was not possible to determine which of the two saltation schemes was superior, based on the present study results. Such differentiation will require further evaluation studies using more measurements of saltation flux and vertical dust emissions.

Estimation of saltation emission in the Kubuqi Desert, North China

The Kubuqi Desert suffered more severe wind erosion hazard. Every year, a mass of aeolian sand was blown in the Ten Tributaries that are tributaries of the Yellow River. To estimate the quantity of aeolian sediment blown into the Ten Tributaries from the Kubuqi Desert, it is necessary to simulate the saltation processes of the Kubuqi Desert. A saltation submodel of the IWEMS (Integrated Wind-Erosion Modeling System) and its accompanying RS (Remote Sensing) and GIS (Geographic Information System) methods were used to model saltation emissions in the Kubuqi Desert. To calibrate the saltation submodel, frontal area of vegetation, soil moisture, wind velocity and saltation sediment were observed synchronously on several points in 2011 and 2012. In this study, a model namely BEACH (Bridge Event And Continuous Hydrological) was introduced to simulate the daily soil moisture. Using the surface parameters (frontal area of vegetation and soil moisture) along with the observed wind velocities and saltation sediments for the observed points, the saltation model was calibrated and validated. To reduce the simulate error, a subdaily wind velocity program, WINDGEN was introduced in this model to simulate the hourly wind velocity of the Kubuqi Desert. By incorporating simulated hourly wind velocity, and model variables, the saltation emission of the Kubuqi Desert was modeled. The model results show that the total sediment flow rate was 1-30.99 tons/m over the last 10years (2001-2010). The saltation emission mainly occurs in the north central part of the Kubuqi Desert in winter and spring. Integrating the wind directions, the quantity of the aeolian sediment that deposits in the Ten Tributaries was estimated. Compared with the observed data by the local government and hydrometric stations, our estimation is reasonable.

The role of deep convection and nocturnal low-level jets for dust emission in summertime West Africa: Estimates from convection-permitting simulations

[1] Convective cold pools and the breakdown of nocturnal low-level jets (NLLJs) are key meteorological drivers of dust emission over summertime West Africa, the world's largest dust source. This study is the first to quantify their relative contributions and physical interrelations using objective detection algorithms and an off-line dust emission model applied to convection-permitting simulations from the Met Office Unified Model. The study period covers 25 July to 02 September 2006. All estimates may therefore vary on an interannual basis. The main conclusions are as follows: (a) approximately 40% of the dust emissions are from NLLJs, 40% from cold pools, and 20% from unidentified processes (dry convection, land-sea and mountain circulations); (b) more than half of the cold-pool emissions are linked to a newly identified mechanism where aged cold pools form a jet above the nocturnal stable layer; (c) 50% of the dust emissions occur from 1500 to 0200 LT with a minimum around sunrise and after midday, and 60% of the morning-to-noon emissions occur under clear skies, but only 10% of the afternoon-to-nighttime emissions, suggesting large biases in satellite retrievals; (d) considering precipitation and soil moisture effects, cold-pool emissions are reduced by 15%; and (e) models with parameterized convection show substantially less cold-pool emissions but have larger NLLJ contributions. The results are much more sensitive to whether convection is parameterized or explicit than to the choice of the land-surface characterization, which generally is a large source of uncertainty. This study demonstrates the need of realistically representing moist convection and stable nighttime conditions for dust modeling. Citation: Heinold, B., P. Knippertz, J. H. Marsham, S. Fiedler, N. S. Dixon, K. Schepanski, B. Laurent, and I. Tegen (2013), The role of deep convection and nocturnal low-level jets for dust emission in summertime West Africa: Estimates from convection-permitting simulations, J. Geophys. Res. Atmos., 118, 4385-4400, doi:10.1002/jgrd.50402.