Fig 2 - uploaded by José M Baldasano
Content may be subject to copyright.
In the top and central panels, it is shown the source term based on the arid and semi-arid categories of the 10-min SSiB (a) and Olson World Ecosystem and the 1-km USGS land use datasets (b). In the panel c, regional distributions of the source function (G) from Ginoux et al. (2001) are shown in the grey scale. The red circles indicate dust emission source areas discussed in the text: (1) Bodeíe´Bodeíe´, (2) Mali, (3) Mauritania, (4) Western Sahara-Morocco, (5) Algeria-Adrar, (6) North of Algeria-Tunisia, (7) Lybia desert, (8) An-Nafud desert and (9) Rub' Al Khali desert.
Source publication
The BSC-DREAM8b model and its predecessor are analysed in terms of aerosol optical depth (AOD) for 2004 over Northern Africa, the Mediterranean and the Middle East. We discuss its performances and we test and analyse its behaviour with new components. The results are evaluated using hourly data from 44 AERONET stations and satellites seasonal avera...
Contexts in source publication
Context 1
... the original DREAM version (hereafter referred as M4, see Table 1), the source mask S is calculated from The codes denote the following references. B41: Bagnold (1941 remapping the arid and semi-arid categories of the Olson World Ecosystems database (EPA, 1992) at 10-min resolu- tion to the regional model domain (Fig. 2a). For soil textures, a combination of the Zobler near-surface global soil texture (Staub and Rosenzweig, 1987) at 18 resolution and the UNEP/GRID gridded FAO/UNESCO 4-km soil units is used. For each soil texture class the fractions of clay, small silt, large silt and sand are estimated from the soil texture triangle (Hillel, 1982). In ...
Context 2
... Pe´rez et al. (2006a) showing the positive impact of including dustÁradiation interactions in the short-term weather forecast. The main features of BSC-DREAM8b (hereafter referred as M8, see Table 1), described in Pe´rez Pe´rez et al. (2006a), are a new source function S based on the arid and semi-arid categories of the 1-km USGS land use dataset (Fig. 2b), a more detailed particle size distribution which includes eight size bins within the 0.1Á10 mm radius range according to Tegen and Lacis (1996) as listed in Table 2, a source size distribution derived from D' Almeida (1987), and dust radiative feedbacks (Pe´rezPe´rez et al., 2006a). Because important overestimations of the simulated ...
Context 3
... number r min Á r max (in mm) r eff (in mm) which is defined as the probability to have accumulated alluvium sediments in the grid cell i of altitude z i , and where z max and z min are the maximum and the minimum eleva- tions in the surrounding 108)108 topography, respectively (Fig. 2c). We analyse the impact of the preferential source mask of Ginoux et al. (2001) in conjunction with the two emission schemes in D8 and N8. In these model simula- tions, the source term S is multiplied by G in D8 and N8 to which we refer as DG8 and NG8, ...
Context 4
... maximum AOD in winter and spring over the An-Nafud desert in the northern part of the Arabian Peninsula (Fig. 4). In spring, the comparison with AERONET (Fig. 8) shows that M4 presents better seasonal correlations than M8 (0.59 for M4 and 0.53 for M8). This is partly linked to the different source mask used in each model version (see Table 1 and Fig. 2a and b). In both model versions, maximum summer activity shifts towards southern latitudes over the Rub' Al Khali desert (in the southwestern part of the Arabian Peninsula, see Fig. 4). M4 and M8 suffer strong overestimations (Fig. 11) Taylor's diagrams for M4, M8, D8, DG8, N8 and NG8 model version against the filtered AOD direct-sun AERONET ...
Context 5
... Preferential source approach. The inclusion of the topographic preferential source mask G from Ginoux et al. (2001) (see Fig. 2c) in both emission schemes (DG8 and NG8) enhances the dust emission in the Bodeíe´Bodeíe´during winter and autumn in better agreement with satellite observations (Fig. 6). Consequently, an increase of Atlantic dust transport during winter and autumn is observed as well as a better correspondence with AERONET observations in the Sahelian ...
Citations
... Detailed physics-based numerical modeling is typically used for dust forecasting 6,7 . These models are based on assumptions regarding the emission and transport of dust storms; these predictions are thus highly model-dependent and often suffer from low skill 8 . ...
Events of high dust loading are extreme meteorological phenomena with important climate and health implications. Therefore, early forecasting is critical for mitigating their adverse effects. Dust modeling is a long-standing challenge due to the multiscale nature of the governing meteorological dynamics and the complex coupling between atmospheric particles and the underlying atmospheric flow patterns. While physics-based numerical modeling is commonly being used, we propose a meteorological-based deep multi-task learning approach for forecasting dust events. Our approach consists of forecasting the local PM 10 (primary task) measured in situ, and simultaneously to predict the satellite-based regional PM 10 (auxiliary task); thus, leveraging valuable information from a correlated task. We use 18 years of regional meteorological data to train a neural forecast model for dust events in Israel. Twenty-four hours before the dust event, the model can detect 76% of the events with even higher predictability of winter and spring events. Further analysis shows that local dynamics drive most misclassified events, meaning that the coherent driving meteorology in the region holds a predictive skill. Further, we use machine-learning interpretability methods to reveal the meteorological patterns the model has learned, thus highlighting the important features that govern dust events in the Middle East, being primarily lower-tropospheric winds, and Aerosol Optical Depth.
... With the intention to facilitate the use of this method by any potential user not necessarily trained in the field, in the Diapason-Methodology-step 1 we opted to only make use of the numerical outcome of desert dust forecasts to select desert dust days, thus making its outcome operator-independent and fast. In particular, the daily-and site-resolved identification of desert dust-affected days (binary yes/no dust-flag) is made using numerical data of the BSC-DREAM8b model (Basart et al., 2012b). With respect to the EC-Method, this alternative procedure loses the advantage of a supervised evaluation and the countercheck of desert dust presence by visual comparison to additional resources as satellite images and/or different dust models. ...
... The BSC-DREAM8b model (Nickovic et al., 2001;Pérez et al., 2006;Basart et al., 2012b) predicts the atmospheric dust cycle including emission, transport and deposition along with dust-radiation interactions. BSC-DREAM8b has been used and evaluated for long time periods over Europe, Northern Africa and the Middle East (Basart et al., 2012b) and it is being used for dust forecasting and as a dust research tool. ...
... The BSC-DREAM8b model (Nickovic et al., 2001;Pérez et al., 2006;Basart et al., 2012b) predicts the atmospheric dust cycle including emission, transport and deposition along with dust-radiation interactions. BSC-DREAM8b has been used and evaluated for long time periods over Europe, Northern Africa and the Middle East (Basart et al., 2012b) and it is being used for dust forecasting and as a dust research tool. It is also one of the models whose forecasts should be evaluated within the EC-Methodology step 1 described above. ...
Desert dust storms pose real threats to air quality and health of millions of people in source regions, with associated impacts extending to downwind areas. Europe (EU) is frequently affected by atmospheric transport of desert dust from the Northern Africa and Middle East drylands. This investigation aims at quantifying the role of desert dust transport events on air quality (AQ) over Italy, which is among the EU countries most impacted by this phenomenon. We focus on the particulate matter (PM) metrics regulated by the EU AQ Directive. In particular, we use multiannual (2006–2012) PM10 records collected in hundreds monitoring sites within the national AQ network to quantify daily and annual contributions of dust during transport episodes. The methodology followed was built on specific European Commission guidelines released to evaluate the natural contributions to the measured PM-levels, and was partially modified, tested and adapted to the Italian case in a previous study. Overall, we show that impact of dust on the yearly average PM10 has a clear latitudinal gradient (from less than 1 to greater than 10 µg/m³ going from north to south Italy), this feature being mainly driven by an increased number of dust episodes per year with decreasing latitude. Conversely, the daily-average dust-PM10 (≅12 µg/m³) is more homogenous over the country and shown to be mainly influenced by the site type, with enhanced values in more urbanized locations. This study also combines the PM10 measurements-approach with geostatistical modelling. In particular, exploiting the dust-PM10 dataset obtained at site- and daily-resolution over Italy, a geostatistical, random-forest model was set up to derive a daily, spatially-continuous field of desert-dust PM10 at high (1-km) resolution. This finely resolved information represent the basis for a follow up investigation of both acute and chronic health effects of desert dust over Italy, stemming from daily and annual exposures, respectively.
... The model uses the thermal state of the atmosphere, near-surface winds, soil properties, and vegetation covers etc., to simulate dust. The model has proven accuracy in predicting dust storm events6,7,[87][88][89] and is well-validated with datasets from various satellite observations and observational networks6,90,91 . ...
This paper investigates the characteristics and impact of a major Saharan dust storm during June 14th–19th 2020 on atmospheric radiative and thermodynamics properties over the Atlantic Ocean. The event witnessed the highest ever aerosol optical depth for June since 2002. The satellites and high-resolution model reanalysis products well captured the origin and spread of the dust storm. The Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observation (CALIPSO) measured total attenuated backscatter and aerosol subtype profiles, lower angstrom exponent values (~ 0.12) from Modern-Era Retrospective Analysis for Research and Application—version 2 (MERRA-2) and higher aerosol index value from Ozone monitoring instrument (> 4) tracked the presence of elevated dust. It was found that the dust AOD was as much as 250–300% higher than their climatology resulting in an atmospheric radiative forcing ~ 200% larger. As a result, elevated warming (8–16%) was observed, followed by a drop in relative humidity (2–4%) in the atmospheric column, as evidenced by both in-situ and satellite measurements. Quantifications such as these for extreme dust events provide significant insights that may help in understanding their climate effects, including improvements to dust simulations using chemistry-climate models.
... The AOD outputs of the models related to their forecasting (initial time 24 h before the valid time) were extracted from the archives of their data. The models that are examined in this study are BSC-DREAM8b-V2 [71,72], DREAM8-MACC [25], DREAMABOL [73], EMA-RegCM4 [74], MACC-ECMWF [75], NASA-GEOS [76], NCEP-NGAC [77], NMMB-BSC [78] and NOA-WRF-Chem [79]. The main characteristics and numerical equations of these models (dust schemes, initial, boundary conditions, etc.) are briefly described in Appendix A. To evaluate the performance of these models in forecasting the examined dust storms, the AOD outputs over a selected spatial domain in the Middle East are qualitatively and quantitatively evaluated against the Terra-MODIS satellite data with 1° × 1° spatial resolution (level 3), retrieved by combined Dark Target and Deep Blue algorithms. ...
This study investigates four types of synoptic dust events in the Middle East region, including cyclonic, pre-frontal, post-frontal and Shamal dust storms. For each of these types, three intense and pervasive dust events are analyzed from a synoptic meteorological and numerical simulation perspective. The performance of 9 operational dust models in forecasting these dust events in the Middle East is qualitatively and quantitatively evaluated against Terra-MODIS observations and AERONET measurements during the dust events. The comparison of model AOD outputs with Terra-MODIS retrievals reveals that despite the significant discrepancies, all models have a relatively acceptable performance in forecasting the AOD patterns in the Middle East. The models enable to represent the high AODs along the dust plumes, although they underestimate them, especially for cyclonic dust storms. In general, the outputs of the NASA-GEOS and DREAM8-MACC models present greater similarity with the satellite and AERONET observations in most of the cases, also exhibiting the highest correlation coefficient, although it is difficult to introduce a single model as the best for all cases. Model AOD predictions over the AERONET stations showed that DREAM8-MACC exhibited the highest R2 of 0.78, followed by NASA_GEOS model (R2 = 0.74), which both initially use MODIS data assimilation. Although the outputs of all models correspond to valid time more than 24 h after the initial time, the effect of data assimilation on increasing the accuracy is important. The different dust emission schemes, soil and vegetation mapping, initial and boundary meteorological conditions and spatial resolution between the models, are the main factors influencing the differences in forecasting the dust AODs in the Middle East.
... • Results for dust optical depth, dust surface concentrations, dust dry and wet depositions forecasted by the models at the Barcelona Supercomputing Centre (BSC)-BSC-DREAM8b [13], NMMB/BSC-Dust [14], the horizontal resolution is 0.3°× 0.3°; • Results for aerosol optical depth (AOD) and dust surface concentrations, forecasted by the ensemble model at the World Meteorological Organization Sand and Dust Storm Warning Advisory and Assessment System (WMO SDS-WAS) Regional Center for Northern Africa, Middle East and Europe on a grid with resolution 0.5°× 0.5° [15]; • Results for the Dust AOD and total AOD at 550 nm, and PM10 concentrations forecasted by the global CAMS-ECMWF model [16] over Europe on a grid resolution 0.125°× 0.125°, available through the Copernicus Atmosphere Monitoring Service (CAMS) and the European Centre for Medium Weather Forecast (ECMWF) [17]; • Results for dust and PM10 over Europe obtained by the CAMS regional air quality ensemble model, with horizontal resolution 0.1°× 0.1°, [18]; • Maps based on multi-model results at global scale with resolution 0.1°× 0.1°at the Marine Meteorology Division of the Naval Research Laboratory USA (NRL), [19,20]; • HYSPLIT air mass backward-trajectories [21,22] calculated at three arrival heights (500, 1500 and 3000 m a.g.l.) for 96 h, using NCEP GDAS meteorological input with resolution 0.5°× 0.5°, and reanalysis data; ...
The objective of this work is to investigate the influence of Saharan dust events on the chemical composition of rain samples collected at three sites in Bulgaria during 2017–2018. Saharan dust intrusions were identified through a combination of satellite retrieved aerosol data and results from dust forecasting models and from backward trajectory model. The chemical composition of the samples (acidity pH, conductivity EC, main ions and elements) is analysed in view of the direction of the approaching air masses—“direct” influence (south-west), and “indirect” influence from other directions and regions, already impacted by Saharan dust. The samples were characterised by pH from 4.1 to 7.4, elevated values for EC (max 202 µS cm⁻¹) and for Si, Ca, Fe, Mg concentrations. For cases with direct influence Si and Ca values were up to 1.5 and 25 mg l⁻¹. In most of the indirect cases increased concentrations of sulphate, nitrate and ammonium were observed (up to 39.5, 23.1 and 8.3 mg l⁻¹).
... SDS-WAS models used include ground observations (particulate matter measurements progressively becoming available in NRT, 345 indirect information from regular weather reports and remote-sensing retrievals from sun photometers or vertical profilers) and satellite products (single-band images, qualitative multi-band products designed to improve dust identification or quantitative retrievals). Currently, the WMO SDS-WAS Regional Centre for Northern Africa, Middle East and Europe provides a multi-model platform with analysis and +54 hours forecasts for 12 dispersion models (Nickovic et al., 2001;Woodward et al., 2001;Zakey et al., 2006;Benedetti, et al., 2009;Morcrette et al., 350 200;Colarco et al., 2010;Pérez et al., 2011;Haustein et al., 2012;Basart et al., 2012Basart et al., , 2020Lu et al., 2016). SDS-WAS contributes to the International Cooperative for Aerosol Prediction (ICAP), an unfunded international forum for aerosol forecast centres, remote sensing data providers, and lead systems developers, which coordinates the first global multi-model Ensemble for aerosol forecasts, as described in Sessions et al. (2015). ...
The purpose of the EUNADICS prototype Early Warning System (EWS) is to proceed the combined use of harmonise data products from satellite, ground-based and in situ instruments to produce alerts of airborne hazard (volcanic, dust, smoke and radionuclide clouds), satisfying the requirement of ATM stakeholders (www.eunadics.eu). The alert products developed by EUNADICS EWS (i.e. NRT observations, email notifications and NetCDF Alert data Products, called NCAP) have shown shows the significant interest in using selective detection of natural airborne hazards from polar orbiting satellite. The combination of several sensors inside a single global system demonstrates the advantage of using a triggered approach to obtain selective detection from observations, which cannot initially discriminate the different aerosol types. Satellite products from hyperspectral UV and IR sensors (e.g. TROPOMI, IASI) and broadband geostationary imager (SEVIRI), and retrievals from ground-based networks (e.g. EARLINET, E-PROFILE and the regional network from volcanic observatories), are combined by our system to create tailored alert products (e.g. selective ash detection, SO2 column and plume height, dust cloud and smoke from wildfires). A total of 23 different alert products are implemented, using 1 geostationary and 12 polar orbiting satellite platforms, 3 external existing service, 2 EU and 2 regional ground-based networks. This allows the identification and the traceability of extreme events. EUNADICS EWS has also shown the interest to proceed a future relay of radiological data (gamma dose rate and radionuclides concentrations in ground-level air) in case of nuclear accident, highlighting the capability of operating early warnings with the use of homogenised dataset. For the four types of airborne hazard, EUNADICS EWS has demonstrated its capability to provide NRT alert data products to trigger data assimilation and dispersion modelling providing forecasts, and inverse modelling for source term estimate. All our alert data products (NCAP files) are not publicly disseminated. Access to our alert products is currently restricted to key users (i.e. Volcanic Ash Advisory Centres, National Meteorological Services, World Meteorological Organization, governments, volcanic observatories and research collaborators), as these are considered pre-decisional products. On the other hand, thanks to the SACS/EUNADICS web interface (https://sacs.aeronomie.be), the main part of the satellite observations used by EUNADICS EWS, are shown in NRT, with public email notification of volcanic emission and delivery of tailored images and NCAP files. All the ATM stakeholders (e.g. VAACs, NMSs, WMOs, Airlines and Pilots) can access and benefit of these alert products through this free channel.
... Together with growing interest in dust storms, the understanding of the physical processes associated with dust storms has increased rapidly over the last decades (World Meteorological Organization, 2018). Large efforts have been made to develop dust modeling systems (Marticorena and Bergametti, 1995;Shao et al., 1996;Marticorena et al., 1997;Alfaro et al., 1997;10 Wang et al., 2000;Liu et al., 2003;Basart et al., 2012), which mathematically simulate the life cycle of dust including emission, transport and deposition. Large scale global dust transport models, e.g., CAMS-ECMWF (Morcrette et al., 2009), or regional ones, e.g., NASA-GEOS-5 (Colarco et al., 2010) and BSC-DREAM8b (Mona et al., 2014), are essential parts of larger Earth system models. ...
When calibrating simulations of dust clouds, both the intensity and the position are important. Intensity errors arise mainly from uncertain emission and sedimentation strengths, while position errors are attributed either to imperfect emission timing, or to uncertainties in the transport. Though many studies have been conducted on the calibration or correction of dust simulations, most of these focus on intensity solely, and leave the position errors mainly unchanged. In this paper, a grid distorted data assimilation, which consists of an imaging morphing method and an ensemble-based variational assimilation, is designed for re-aligning a simulated dust plume to correct the position error. This new developed grid distorted data assimilation has been applied to a dust storm event in May 2017 over East Asia. Results have been compared for three configurations: a traditional assimilation that focuses solely on intensity correction, a grid distorted data assimilation that focuses on position correction only, and the hybrid assimilation that combines these two. For the evaluated case, the position misfit in the simulations is shown to be dominant in the results. The traditional emission inversion improves only slightly the dust simulation, while the grid distorted data assimilation effectively improves the dust simulation and forecast. The hybrid assimilation that corrects both position and intensity of the dust load provides the best initial condition for forecast of dust concentrations.
... DREAM8 model is embedded in the NCEP Nonhydrostatic Mesoscale Model (NMM) on the E-grid as meteorological driver (Janjic et al., 2001) using the ECMWF global forecast data for initial and boundary conditions. DREAM8 assimilates ECMWF dust analysis in the initial meteorological field, with static dust source function (Ginoux et al., 2001;Basart et al., 2012) and provides daily dust forecasts at the SEEVCCC website (Cuevas et al., 2015). The model follows the Janjic et al. (1994) dust scheme with 8 size bins (0.1-10 µm) and the dust uplifting is from Shao et al. (1993) using the USGS global land cover data (1-km resolution). ...
Ephemeral and dried lakes in southwest and central Asia, such as the Aral Sea and Hamouns in east Iran, are major sources of dust storms with an increasing tendency during the last decades due to anthropogenic influence and climate change. This study examines two characteristic dust-storm events originated from these areas on 12–15 July 2016 and 27–28 May 2018. Thermal low-pressure systems over topographic-low areas and the Caspian Sea High, along with pressure gradients and intense surface winds over the dust sources facilitated the dust outbreaks and transport. Three models (DREAM-NMME-MACC, CAMS and WRF-Chem) are synergized to simulate the spatial distribution of AOD and surface dust concentrations during the dust events. The results show that DREAM-NMME-MACC and WRF-Chem exhibit the highest discrepancies (R = 0.15–0.58; RMSE = 83%–125%) in representing the spatial and temporal distribution of dust compared to Terra-MODIS AODs, while CAMS reveals the best performance (R = 0.29–0.77; RSME = 67%–124%). All models significantly underestimate the high MODIS AODs, especially near the source areas, due to different dust schemes, soil conditions, meteorology and dynamic processes for dust emissions that they comprise. Furthermore, notable differences between the models are revealed in simulations of the PM10 concentrations in Zabol, east Iran, as the models fail to reproduce the temporal evolution and high intensity of the dust event. In general, all models represent better the dust storm originated from Sistan (13–14 July 2016) rather than the Aralkum dust storm (27–28 May 2018), indicating an incapability in representing the soil characteristics in a progressively drying terrain.
... The Earth Sciences Department from the Barcelona Supercomputing Center (BSC)-Centro Nacional de Supercomputacion (CNS) maintains a dust forecast operational system with the updated version of the former Dust Regional Atmospheric Model (DREAM) called BSC-DREAM8b v2.0 (Basart, Pérez, Nickovic, Cuevas, & Baldasano, 2012;Nickovic, Papadopoulos, Kakaliagou, & Kallos, 2001;Pérez, Nickovic, Baldasano et al., 2006;Pérez, Nickovic, Pejanovic, Baldasano, & Özsoy, 2006). The model predicts the windswept desert dust and was developed as a pluggable component of the Eta/NCEP (National Centers for Environmental Prediction) model. ...
This work presents a detailed study of the dynamical processes triggering the occurrence of the two heavy dust storms which occurred between 18 and 22 March 2012 over the Middle East. The dynamics of this event are related to the coupling of subtropical jet and polar jet over the Saudi Arabia region, resulting in massive dust storm generation and dust transport through Rub' al Khali and the Persian Gulf to the UAE region. AOD and PM 10 values showed a fourfold increase during the event reaching a maximum of 1.8 and 1653 μg/m 3 respectively. The spatial extent of the dust storm is evident from high values of MODIS AOD (~1.5) and OMI aerosol index (4.5) covering the entire Middle East. The total attenuated backscatter at 550 nm from CALLIPSO showed the vertical extent of dust up to 8 km. In addition, surface temperature showed a decrease of almost 15°C during the event signifying the intensity of the dust storm. Aerosol radiative forcing estimates during the dust storm showed a cooling at the surface and warming in the atmosphere, with a maximum forcing value reaching up to~− 210 Wm −2 (185 Wm −2). Hence, it is evident from the present study that the dust layer caused an additional warming of 150 Wm −2 in the atmosphere over this region. The present event showcases the importance of dust storm induced aerosol optical and physical processes, and associated atmospheric dynamics over UAE as well as other affected regions.
... The underestimation of dust concentrations over MR-A during the SD1 and SD2 events was also found in other simulation systems, for example as published by the SDS-WAS service (https://ess.bsc.es/bsc-dust-daily-forecast). As example, results for SD1 and SD2 from the forecast system BSC-DREAM8b (Basart and Carlos, 2012;Mona et al., 2014) are shown in the last 20 row of Fig. 4 and Fig. 5, respectively. These suggests that these emission models are also prone to underestimate the emission rate over Horqin desert. ...
Emission inversion using data assimilation fundamentally relies on having the correct assumptions on the emission background error covariance. A perfect covariance accounts for the uncertainty based on prior knowledge, and is able to explain differences between model simulations and observations. In practice, emission uncertainties are constructed empirically, hence a partially unrepresentative covariance is unavoidable. Concerning its complex parameterization, dust emissions are a typical example where the uncertainty could be induced from many underlying inputs, e.g., information on soil composition and moisture, landcover and erosive wind velocity, and these can hardly be taken into account together. This paper describes how an adjoint model can be used to detect errors in the emission uncertainty assumptions. This adjoint based sensitivity method could serve as a supplement of a data assimilation inverse modeling system to trace back the error sources, in case that large observation-minus-simulation residues remain after assimilation based on empirical background covariance. The method follows on application of a data assimilation emission inversion for an extreme severe dust storm over East Asia (Jin et al., 2019b). The assimilation system successfully resolved observation-minus-simulation errors using satellite AOD observations in most of the dust-affected regions. However, a large underestimation of dust in northeast China remained despite the fact the assimilated measurements indicated severe dust plumes there. An adjoint implementation of our dust simulation model is then used to detect the most likely source region for these unresolved dust loads. The backward modeling points to the Horqin desert as source region, which was indicated as a non-source region by the existing emission scheme. The reference emission and uncertainty are then reconstructed over the Horqin desert by assuming higher surface erodibility. After the emission reconstruction, the emission inversion is performed again and the posterior dust simulations and reality are now in much closer harmony. Based on our results, it is advised that emission sources in dust transport models include Horqin desert as a more active source region.
