An investigation into climatic and terrestrial drivers of dust storms in the Sistan region of Iran in the early twenty-first century

Intrinsic and extrinsic techniques for quantification uncertainty of an interpretable GRU deep learning model used to predict atmospheric total suspended particulates (TSP) in Zabol, Iran during the dusty period of 120-days wind

Total suspended particulates (TSP), as a key pollutant, is a serious threat for air quality, climate, ecosystems and human health. Therefore, measurements, prediction and forecasting of TSP concentrations are necessary to mitigate their negative effects. This study applies the gated recurrent unit (GRU) deep learning model to predict TSP concentrations in Zabol, Iran, during the dust period of the 120-day wind (3 June - 4 October 2014). Three uncertainty quantification (UQ) techniques consisting of the blackbox metamodel, heteroscedastic regression and infinitesimal jackknife were applied to quantify the uncertainty associated with GRU model. Permutation feature importance measure (PFIM), based on the game theory, was employed for the interpretability of the predictive model's outputs. A total of 80 TSP samples were collected and were randomly divided as training (70%) and validation (30%) datasets, while eight variables were used in the TSP prediction model. Our findings showed that GRU performed very well for TSP prediction (with r and Nash Sutcliffe coefficient (NSC) values above 0.99 for both datasets, and RMSE of 57 μg m-3 and 73 μg m-3 for training and validation datasets, respectively). Among the three UQ techniques, the infinitesimal jackknife was the most accurate one, while all the observed and predicted TSP values fell within the continence limitation estimated by the model. PFIM plots showed that wind speed and air humidity were the most and least important variables, respectively, impacting the predictive model's outputs. This is the first attempt of using an interpretable DL model for TSP prediction modelling, recommending that future research should involve aspects of uncertainty and interpretability of the predictive models. Overall, UQ and interpretability techniques have a key role in reducing the impact of uncertainties during optimization and decision making, resulting in better understanding of sophisticated mechanisms related to the predictive model.

Evaluating temporal sand drift potential trends in the Sistan region, Southeast Iran

The Sistan region in Southeastern Iran is one of the world's most sensitive areas when it comes to sandstorms and wind erosion. One of the most influential factors in interpreting sandstorms is sand drift potential (DP), which is directly related to wind speed. Accurately, monitoring this phenomenon is still being determined, considering various temporal scales. Therefore, the main aim of this research is to analyze the trend of DP on monthly and annual scales. Our results showed that monthly variations of DP reached the highest and lowest values in July (609 VU) and January (47 VU), respectively. Blowing sand predominantly moved southeast, and the directional index fluctuated from 0.88 to 0.94. The annual DP was measured equal to 2700 VU, signifying a relatively high value when compared to other arid regions worldwide. The trend analysis results obtained from the Mann-Kendall test revealed both positive trends during the period 1987-2001 and negative ones from 2002 to 2016). However, the positive trend was found statistically insignificant. Furthermore, Sen's slope test results demonstrated that a negative trend could be observed with a steeper slope during July, September, and August, while a positive trend could be observed with a steeper pitch during November, December, and June. We recommend that land managers and stakeholders involved in controlling blowing sand using biological and physical methods should consider these trends in the Sistan region. Implementing nature-based solutions or control strategies should focus on these temporal sequences.

Spatiotemporal variations of internal dust events in urban environments of Iran, Southwest Asia

This article investigates the Dust Storm Index (DSI) and its trend using the Mann-Kendall test, across urban areas of Iran on the monthly, seasonally, and annually scales from 2000 to 2018. The results showed that cities located in the humid region, especially Khoram Abad and Avaj, had the lowest DSI values, and the cities located in arid regions, particularly Zabol, Sarakhs, and Zahedan, had the highest DSI values during the study period. On a monthly basis, the positive trends were observed in most cities of Iran in March, October, and August, while the negative trends were mainly observed in Feb, May, and June. Birjand, Torbat Heydariyeh, Saveh, Shiraz, and Kerman showed an increasing trend of DSI in most months of the study period. On a seasonal scale, the autumn and summer DSI changes showed significant positive trends in 18% of the urban environments in Iran. A similar trend was observed for 17% and 15% of study urban areas in the spring and winter, respectively. On an annual scale, the significant upward trends in DSI were observed in 13% while its negative changes were found in 10% of study cities. These results can be useful for decision-makers and managers to take appropriate measures to reduce and control dust events in urban areas that have suffered from the increasing trend of dust events in the past years.

Impact of climate change induced future rainfall variation on dynamics of arid-humid zone transition in the western province of India

The transition of the Earth's climate from one zone to another is one of the major causes behind biodiversity loss, rural-urban migration, and increasing food crises. The rising rate of arid-humid zone transition due to climate change has been substantially visible in the last few decades. However, the precise quantification of the climate change-induced rainfall variation on the climate zone transition still remained a challenge. To solve the issue, the Representative Grid Location-Multivariate Adaptive Regression Spline (RGL-MARS) downscaling algorithm was coupled with the Koppen climate classification scheme to project future changes in various climate zones for the study area. It was observed that the performance of the model was better for the humid clusters compared to the arid clusters. It was noticed that, by the end of the 21^st century, the arid region would increase marginally and the humid region would rise by 24.28-36.09% for the western province of India. In contrast, the area of the semi-arid and semi-humid regions would decline for the study area. It was observed that there would be an extensive conversion of semi-humid to humid zone in the peripheral region of the Arabian sea due to the strengthening of land-sea thermal contrast caused by climate change. Similarly, semi-arid to arid zone conversion would also increase due to the inflow of dry air from the Arabian region. The current research would be helpful for the researchers and policymakers to take appropriate measures to reduce the rate of climate zone transition, thereby developing the socioeconomic status of the rural and urban populations.

A field study of the surface disturbance effects of animals and motor vehicles on aeolian sediment emission from a silty playa surface

Dry playa surfaces can be extremely vulnerable to disturbance which breaks their surface crusts resulting in increased aeolian sand and dust emissions. Trampling by livestock and motor vehicles is an important source of this disturbance. The Hamoun Lakes in the Sistan region of Iran are a major source area of dust storms which are causing damage to infrastructure and communities. This study performed portable wind tunnel tests of controlled surface disturbance by animals (cow, sheep) and motor vehicles (automobile, motorcycle) on a silty playa surface of Hamoun Saberi lake. The aim of this study was to assess the effect of different levels of trampling and disturbed surfaces by both vehicles and livestock on dust emission at wind speeds of 6 and 12 m s^-1. A significant increase of sediment emission was found with increased number of passes of vehicles and livestock and the degree of surface disturbance, more so at the high wind speed (12 m s^-1). No significant differences were observed between a 10-20% disturbance level and an undisturbed surface, but statistically significant differences became apparent when disturbances reached 50-60% to 90-100% compared to undisturbed surfaces. Greater emission rates were reached by disturbances provided by automobile and cow compared to motorcycle and sheep, indicating greater trampling effects of automobile and cow likely related to greater weight and larger footprint. The automobile was the most surface-destructive of the four anthropogenic disturbances, providing emission in a lower number of passes compared to the motorcycle, cow and sheep. Better management of vehicle and livestock allocation on playas subject to disturbance, such as the Hamoun Lakes, will be a useful strategy to reduce disturbance and the frequency and intensity of dust storms.

A new application of multi-criteria decision making in identifying critical dust sources and comparing three common receptor-based models

Dust storms are a common phenomenon in arid and semi-arid regions in West Asia, which has led to high levels of PM₁₀ in local and remote area. The Yazd city in Iran with a high PM₁₀ level located downstream of dust sources in the Middle East and Central Asia. In this study, based on meteorological and PM₁₀ monitoring data, backward trajectory modeling of air parcels related to dust events at Yazd station was performed using the HYSPLIT model in 2012-2019. The trajectory cluster analysis was used to identify the main dust transport pathways and wind systems. Three methods of Cross-referencing Backward Trajectory (CBT), Potential Source Contribution Function (PSCF) and Concentration Weighted Trajectory (CWT) were used to identify the most critical dust sources. Multi-Criteria Decision Making (MCDM) methods were also used to integrate the results. Nine dust sources affecting central Iran were determined, and six criteria from different aspects were considered. To prioritize the dust sources affecting central Iran from four new MCDM methods, including WASPAS, EDAS, ARAS and TOPSIS were used. The results showed that the Levar wind system (51%), the Shamal wind system (32%) and the Prefrontal wind system (18%) were the most important wind systems to cause dust events in central Iran. The MCDM approach to identify dust sources also showed that Dasht-e-Kavir in central Iran was the most critical dust source. The results also showed that in hot seasons (spring and summer), local and Central Asia dust sources and cold seasons (autumn and winter), Middle East dust sources have the greatest impact on dust events in central Iran. Also, a comparison of common receptor-based methods for identifying dust sources showed that CBT, CWT and PSCF were the most appropriate methods for identifying dust sources, respectively.

Long-Term Variability of Dust Events in Southwestern Iran and Its Relationship with the Drought

Dust storms represent a major environmental challenge in the Middle East. The southwest part of Iran is highly affected by dust events transported from neighboring desert regions, mostly from the Iraqi plains and Saudi Arabia, as well as from local dust storms. This study analyzes the spatio-temporal distribution of dust days at five meteorological stations located in southwestern Iran covering a period of 22 years (from 1997 to 2018). Dust codes (06, 07, 30 to 35) from meteorological observations are analyzed at each station, indicating that 84% of the dust events are not of local origin. The average number of dust days maximizes in June and July (188 and 193, respectively), while the dust activity weakens after August. The dust events exhibit large inter-annual variability, with statistically significant increasing trends in all of five stations. Spatial distributions of the aerosol optical depth (AOD), dust loading, and surface dust concentrations from a moderate resolution imaging spectroradiometer (MODIS) and Modern-Era Retrospective analysis for Research and Applications (MERRA-2) retrievals reveal high dust accumulation over southwest Iran and surrounding regions. Furthermore, the spatial distribution of the (MODIS)-AOD trend (%) over southwest Iran indicates a large spatial heterogeneity during 2000–2018 with trends ranging mostly between −9% and 9% (not statistically significant). 2009 was the most active dust year, followed by 2011 and 2008, due to prolonged drought conditions in the fertile crescent and the enhanced dust emissions in the Iraqi plains during this period. In these years, the AOD was much higher than the 19-year average (2000 to 2018), while July 2009 was the dustiest month with about 25–30 dust days in each station. The years with highest dust activity were associated with less precipitation, negative anomalies of the vegetation health index (VHI) and normalized difference vegetation index (NDVI) over the Iraqi plains and southwest Iran, and favorable meteorological dynamics triggering stronger winds.

Analysis of some factors related to dust storms occurrence in the Sistan region

Dust storms over the Sistan region in East Iran are associated with predominant northwest winds (called 120-day winds) which promote desertification, including drying of the Hamoun wetlands. These storms are more frequent in spring and summer seasons in the Sistan region. The study aims to examine the relationship between vegetation cover and wind speed with dust storms intensity in order to understand the behavior of dust sources using satellite remote sensing data (AOD) between 2000 and 2019. Based on the time series, the study period can be divided into three parts based on the following characteristics: high dust intensity (2004), moderate relative intensity of value in all parameters studied (2005 to 2014), and dust reduction (2015-2019). Time series analysis shows a negative relationship between AOD and wind speed owing presumably to vegetative cover changes during years that wind speed has increased. Based on multiple regression analysis by monthly time scales that conforms time series result, monthly NDVI is significantly related to AOD. Analysis of the 3 hourly wind data suggests a positive relationship between wind and dust, and effective thresholds for dust erosion based on wind speeds are proposed for the Sistan region.

Using multivariate adaptive regression splines and extremely randomized trees algorithms to predict dust events frequency around an international wetland and prioritize its drivers

This study aimed to evaluate the performance of multivariate adaptive regression splines (MARS) and extremely randomized trees (ERT) models for predicting the internal and external dust events frequencies (DEF) across the northeastern and southwestern regions of the Gavkhouni International Wetland. These models were also evaluated to model the internal DEF (IDEF) across the northwestern, southeastern, northern, and western regions around the wetland. Furthermore, the main factors controlling DEF and IDEF were identified based on the importance value (IV) of predictors in the best model. The results showed that the ERT model increased the prediction accuracies by an average of 40%, compared to the MARS model. According to the IV obtained from the ERT model, aerosol optical depth (IV = 0.28), wetland discharge (IV = 0.25), near-surface wind speed (IV = 0.08), and erosive winds frequency (IV = 0.07) were identified as the most important factors controlling DEF across the northeastern and southwestern regions of the wetland. However, the erosive wind speed was detected as the major factor affecting the IDEF in the northern (IV = 0.16), western (IV = 0.18), and southeastern (IV = 0.65) regions of study wetland. It was also found that vapor pressure with IV of 0.32 had the greatest effect on IDEF variations across the northwestern region of the wetland. Overall, the results demonstrate the effectiveness of the ERT model in modeling the factors affecting DEF and IDEF, and the results may be used to mitigate dust events hazards around the Gavkhouni Wetland, in central Iran.

Atmospheric Dynamics and Numerical Simulations of Six Frontal Dust Storms in the Middle East Region

This study analyzes six frontal dust storms in the Middle East during the cold period (October–March), aiming to examine the atmospheric circulation patterns and force dynamics that triggered the fronts and the associated (pre- or post-frontal) dust storms. Cold troughs mostly located over Turkey, Syria and north Iraq played a major role in the front propagation at the surface, while cyclonic conditions and strong winds facilitated the dust storms. The presence of an upper-atmosphere (300 hPa) sub-tropical jet stream traversing from Egypt to Iran constitutes also a dynamic force accompanying the frontal dust storms. Moderate-Resolution Imaging Spectroradiometer (MODIS) and Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observation (CALIPSO) observations are used to monitor the spatial and vertical extent of the dust storms, while model (Weather Research and Forecasting model coupled with Chemistry (WRF-Chem), Copernicus Atmospheric Monitoring Service (CAMS), Regional Climate Model-4 (RegCM4)) simulations are also analyzed. The WRF-Chem outputs were in better agreement with the MODIS observations compared to those of CAMS and RegCM4. The fronts were identified by WRF-Chem simulations via gradients in the potential temperature and sudden changes of wind direction in vertical cross-sections. Overall, the uncertainties in the simulations and the remarkable differences between the model outputs indicate that modelling of dust storms in the Middle East is really challenging due to the complex terrain, incorrect representation of the dust sources and soil/surface characteristics, and uncertainties in simulating the wind speed/direction and meteorological dynamics. Given the potential threat by dust storms, more attention should be directed to the dust model development in this region.