Number Of Dusty Days Research Articles

Temporal and Spatial Variability of Dust in the Urmia Basin, 1990–2019

The living conditions in the Urmia Basin (northwestern Iran) face significant challenges due to dust events. This study investigates the spatial and temporal characteristics of dust phenomena in the Urmia Basin using MERRA-2 data and observational data from Tabriz, Urmia, Sarab, and Mahabad over a 30-year period (1990–2019). The findings reveal that despite several fluctuations, the annual number of dusty days increased from the 1990s to the 2010s in the Urmia Basin. The maximum number of dusty days was found to predominantly occur in May (spring) and October (autumn), driven by two distinct mechanisms. In early autumn, developing synoptic systems associated with increased wind speeds can cause dust emission from dry land sources. Consequently, an increase in dust wet deposition, precipitation, dust surface concentration, and the number of dusty days occurs in October. In contrast, a sharp decrease in precipitation from April to May leads to drying soil and dust emission in May. Among the studied cities, Tabriz experienced the highest number of dusty days (728) due to the combined effects of cross-border and local dust sources. The highest dust column density and dust dry deposition in the south and east of Urmia Lake indicate the impact of declining water levels, which resulted in a dry lakebed as the primary local dust source. The MERRA-2 spatial distribution reveals that dust surface concentration, and the number of dusty days decrease similarly from the southwest to the northeast of the Urmia Basin as the distance from cross-border dust sources increases. A positive correlation is observed between the number of dusty days and MEERA-2 data, including dust surface concentration, dust dry deposition, column mass dust, and total aerosol extinction, with coefficients of 0.74, 0.71, 0.69, and 0.68, respectively.

An experimental study for efficiency modeling of grid-connected photovoltaic system

ABSTRACT As a key factor, it is vital to mathematically model the electrical efficiency of photovoltaic (PV) systems. This paper investigates the efficient modeling of a 2.5 kW grid-connected PV system, located in Mahan, Kerman, Iran, with respect to different factors like solar radiation, humidity, dust deposition, temperature, and shading. For this aim, different nonlinear models are considered and a curve-fitting approach is used to optimally determine the unknown parameters. Over the case study, simulation results show that when the dust deposition is not included in the modeling (solar radiation, ambient temperature, cell temperature, humidity, and shading as inputs), the third-degree polynomial model leads to more accurate result than the other models. By use of the third-degree polynomial model, the value of root mean square error (RMSE) for train and test data sets is obtained 0.0123 and 0.0109, respectively. In this case, the average RMSE value is 0.0116. On the other hand, when dust impact is included in the modeling (solar radiation, ambient temperature, humidity, and number of dusty days as inputs), the accuracy of the exponential model is more than the other ones. In this case, the value of RMSE for both train and test data sets is obtained 0.0018. Sensitivity analysis shows that the system efficiency depends mainly on the radiation, cell temperature, and dust and less on the humidity.

An Investigation Into the Effect of Dust on Wheat Yield

Different regions of Iran are influenced by dust storms since they are located in the arid and semi-arid zone and due to their plain structure, the penetration of different synoptic systems, and adjacency with desert and important dust centers. This research investigates the effect of dust sedimentation on the yield of irrigated and rainfed wheat in four neighbouring provinces and on wheat production in Iran (Khuzestan, Ilam, Kermanshah, and Lorestan). Data are for the crop years from 2011–2012 to 2018–2019. For higher accuracy, the research studies irrigated and rainfed wheat systems separately with the panel model of the Just–Pope function. In the Just–Pope stochastic function, the effects of inputs on two indices of mean yield and yield variance are studied simultaneously. Based on the results, the risk function of irrigated wheat is significantly influenced by the variables of machinery, labour, and dust at the 1% level and by the variables of water and precipitation at the 10% level. The most influential variable in this function is dust (3.03%). In the risk function of rainfed wheat, all variables have a significant impact on production fluctuations at the 1% level. The most influential factor is the temperature (0.86%). Furthermore, the results reveal that the variable of the number of dusty days is significant in the production of irrigated and rainfed wheat at the 15% and 1% levels, respectively. The effect of this variable on rainfed wheat is negative and 0.101%. In this regard, it is necessary to develop dust-specific projects in the agricultural sector and estimate the cost imposed by this phenomenon.

Analyses of a Lake Dust Source in the Middle East through Models Performance

Drying lakes have become a new source of dust, causing severe problems in surrounding areas. From 2000 to 2017, a statistical study was conducted on Lake Urmia in Iran in the Middle East. The results indicated a significant increase in the annual number of dusty days in stations around the lake and the mean annual aerosol optical depth (AOD) at 550 nm. The sharp decrease in annual snowfall rate over the Lake Urmia area since 2007 has been linked to the lake’s decreasing water level and drying. During a dust storm event from 27 October to 31 October 2017, a local dust storm originated from Lake Urmia before another large-scale dust storm originated from the An-Nafud desert. According to MODIS true-color images, dust particles were lifted from Lake Urmia and transported eastward to the Caspian Sea and the HYSPLIT model. The comparison of the four models under the Sand and Dust Storm Warning Advisory and Assessment System (SDS-WAS) revealed that the models overestimated surface dust concentrations compared to ground-based PM10 measurements. Nevertheless, the NOAA/WRF-Chem and DREAMABOL models simulated higher dust concentrations during the dust period. More emphasis should be placed on the development of dust models for SDS-WAS models in Lake Urmia.

Linking ground-satellite observations with HYSPLIT Back trajectory modeling to identify dust sources affecting Western Iran: A case study in Lorestan province

Background: This study combined ground and satellite observations with the results of HYSPLIT model to identify the origin, transport, and deposition of sand and dust storms (SDSs) affecting Western Iran. Methods: Field-measured dust exposure data were obtained during 2000-2014 to analyze variability of dust concentration and dust intensity at annual, monthly, and daily scales. Remote sensing measurements in this research include the analysis of a total of eight (Level 1B Calibrated Radiances 1 km (MOD021KM)) MODIS tiles selected based on interpreting the results of ground observations to capture the major SDS events occurred between 2000 and 2014. Results: The results, indicating a sharp rise in the number of dusty days from 2008 onwards, were used as a basis to identify the spatial coverage and intensity of SDSs over the central part of the Middle East and the study province using a number of eight MODIS images. According to the back trajectory analysis of Hybrid Single-Particle Lagrangian Integrated Trajectory Model (HYSPLIT), the MODIS-derived SDSs were found to have both national and international sources. The most prevalent and powerful source was likely to initiate from deserts of Syria and Iraq by air masses coming from the Mediterranean Sea and terminate in Western Iran where the Zagros Mountains Chain blocks easterly winds. Conclusion: Despite the evident link between timing of regional SDSs and in situ observations, the contribution of local dust sources in the west of the province should be also investigated to provide insights into the development and spread of SDS events affecting Western Iran, especially Lorestan Province.

Predicting the number of dusty days around the desert wetlands in southeastern Iran using feature selection and machine learning techniques

In the past decades, some desert wetlands have become critical regions for dust production in the arid and semi-arid regions of the world. Accurate prediction of the number of dusty days (NDDs) in these areas is of great importance. The most popular method for predicting climatic and environmental variables is machine learning (ML). Although it has received more attention for spatial prediction, it has received less attention for the temporal prediction of these variables. This work is the first effort to predict NDDs in the major source of dust production in southeastern Iran using ML models and different feature selection (FS) techniques. For this purpose, monthly data of 21 predictor variables related to the study period (1988–2017) was used to predict the target variable (NDDs). The main aim was to evaluate the support vector machine (SVM), conditional inference random forest (CRF), and stochastic gradient boosting (SGB) models based on three FS algorithms, including Boruta, multivariate adaptive regression splines (MARS), and recursive feature elimination (RFE) techniques in predicting NDDs around the Hamoun wetlands. After analyzing the collinearity effect and removing the independent variables with a Tolerance < 0.11, the best attributes were selected to train the SVM, SGB, and CRF models. All datasets were randomly classified into training (70%) and verification (30%) sets. The performance of models was evaluated based on the determination coefficient (R-square), root mean square error (RMSE), mean absolute error (MAE), and Nash Sutcliffe efficiency (NSE) coefficient related to holdout data. The results indicated that SGB-MARS, SGB-RFE, and SGB-Boruta outperformed other models with different FS techniques, in terms of R2 (0.9), RMSE (2.5), MAE (1.9), and NSE (0.9). Furthermore, surface winds speed, maximum air temperature, relative humidity, wetland dried bed, and erosive winds frequency were detected as the most important factors for predicting NDDs in the study area. This study encourages us to use the SGB model with various FS techniques to predict NDDs around the desert wetlands. These results can help decision-makers reduce the risks of dust emission and increase the safety of residents around the desert wetlands.

Spatio-temporal and synoptic changes in dust at the three islands in the Persian Gulf region

The occurrence of severe and inclusive dust storms has caused a lot of damage every year in the Persian Gulf region and it has destructive environmental effects in this area. The aim of this study is to investigate the monthly changes in the frequency of dust storms over Khark, Kish and Qeshm islands located in the Persian Gulf region during the 10-year period (2007–2017) using satellite and model observations. The results show that the highest dust frequency was observed over Qeshm Island followed by Kish and Khark Islands. The highest number of dusty days was reported in all three stations in 2008 and most of the dust events occurred in May and July as compared to other months. The average wind speed in these two months (May and July) shows that the wind directions are mostly from northern and northwestern areas. The dust AOD evaluation of eight dust prediction models shows that the models simulated AOD patterns vary differently from each other for a severe dust storm on July 14, 2014 in the study area. Among the eight models, DREAM8-MACC model has a better performance and has good agreement with satellite observations. The analysis showed a high correlation (R = 0.74) between AERONET and DREAM8-MACC model AOD at the observation sites.

Spatiotemporal regionalization of atmospheric dust based on multivariate analysis of MACC model over Iran

Extraordinary dry conditions on the desert and arid areas in Iran, along with strong influences from surrounding dust sources in the Middle East and Arabia, have exposed the country to substantial amounts of dust aerosols with serious impacts on environment, health and socio-economic aspects. Considering this threat, daily aerosol optical depth (AOD) data over Iran obtained from the Monitoring Atmospheric Composition and Climate (MACC) reanalysis during 2003–2012 were used to analyze the spatial and temporal patterns of total AOD (TAOD) and dust AOD (DAOD). This study identifies the areas mostly affected by TAOD and DAOD in terms of daily average values and the areas associated with distinct temporal patterns, utilizing S-mode and T-mode principal component analysis (PCA). The analysis reveals six distinct sub-regions in southeast, west-northwest, northeast, east, central and southwest of Iran, which are among the major centers influenced by TAOD and DAOD extremes. The main findings of the T-mode results, consistent with TAOD and DAOD S-mode patterns, are (i) the maximum dust loading over the eastern, southeastern and southwestern subregions in March and May, (ii) dust loading with peaks in November over the eastern and northeastern sub-regions; (iii) maximum number of dusty days in July and August over southwest, (iv) maximum dust in July over southeastern and Central Iran, (v) a dust pattern with identical extent over the west and northwest sub-region in March and October and, (vi) a prevailing dust loading over the dry areas (Lut desert, Sistan plain and southwest edge of the territory) in October. The spatiotemporal evolution of dust is a function of the growth and expansion of dust extremes originated from various sources in Iran and the Middle East. The PCA methodology, through simplification of aerosol distribution to simpler structures, was also incorporated to provide a geographical interpretation of the highly sensitive territory of Iran.

Investigation of aerosols pollution across the eastern basin of Urmia lake using satellite remote sensing data and HYSPLIT model.

Desiccation of the world's second hypersaline lake, Lake Urmia, has drawn attention toward the feasibility of particle emissions from the lakebed to downwind regions. Therefore, this research was conducted to study spatiotemporal variations of aerosols across the eastern basin of the lake during 2001 to 2016. The Aerosol Optical Depth (AOD) products of Moderate-resolution Imaging Spectroradiometer (MODIS), extracted from Terra platform for 999 rural and urban areas and compared over adjacent (Shabestar, Osku, Azarshahr, Ajabshir, Bonab, Malekan), middle (Tabriz and Maragheh) and far (Bostanabad, Heris, and Sarab) counties. Number of dusty days and direction of high wind speeds (≥ 11m/s) were acquired from the East Azerbaijan Meteorological Organization and analyzed. Moreover, performing the backward trajectory model, the origin and distribution of aerosols were determined at altitudes of 500, 1000 and 2000m. The spatiotemporal variations of AOD provided statistically significant correlations (R2 ≥ 0.5 and p < 0.05) against the number of dusty days. AOD value was higher between 2009 and 2016 and estimated to be 0.36, 0.33 and 0.31 over adjacent, middle and far areas, respectively. Analysis of wind direction and trajectory plots implied that the particulate matter (PM) of study area was mainly transported from Iraq and Syria, especially in April, May and June months. PM has followed an increasing trend, while the adjacent areas have experienced higher pollution compared to far counties. The southwestern winds can play an important role in transportation of aerosols from either lakebed or western countries to the study area.

Spatial and temporal trends of dust storms across desert regions of Iran

Dust storms are among natural and anthropogenic hazards for socioeconomic resources, especially in desert regions. In recent years, dust storms have become a serious problem, especially in desert regions of Iran. This study investigates temporal and spatial variation of dust storm frequency in desert regions of Iran. The number of dusty days (NDD) are collected from 22 stations across the region. The statistical analysis of NDD time series is carried out to show both spatial and seasonal pattern of dust storm occurrence in the region. The regional map of statistical characteristics indicates a north to south increasing dust storm frequency. The spatial map also reveals higher year-to-year variation in south eastern Iran. The seasonality of NDD shows the highest frequency for summer followed by the spring and autumn seasons. The popular Mann–Kendall and the bootstrap MK test to consider serial correlation are then applied for Trend assessment. Results showed both negative (across the north and northwestern regions) and positive trend (across south and south eastern regions) in the annual and seasonal NDD time series. This north-to-south gradient in the spatial and temporal frequency NDD may arise from harsh dry and gusty winds as well as intense land use change in the south eastern territories of Iran. However, more careful and detailed studies are required to connect environmental conditions to change in NDD frequency.

Multivariate statistical analysis of heavy metals contamination in atmospheric dust of Kermanshah province, western Iran, during the spring and summer 2013

Atmospheric dust is known to contain heavy metals in different levels that can cause environmental pollution. In west of Iran, despite of the large number of dusty days per year, heavy metals content of the atmospheric dust has not been previously investigated. A total of 98 samples of atmospheric dust were collected during the spring (T1) and summer (T2) of 2013 from urban and suburban locations in Kermanshah province, and analyzed for Zn, Cu, Ni, Cr, Mn, and Fe total concentrations. Correlation, principal component and cluster analyses suggested probable natural and anthropogenic sources of the metals in the dust. The contamination levels of heavy metals were assessed on the basis of enrichment factor (EF), geo-accumulation index (Igeo), and ecological risk (RI). Comparing with the background values of world soils, elevated metal concentrations were found in dust samples, except for Mn and Fe. Manganese and Fe were mainly of natural origin with traces of anthropogenic influences; while Zn and Cu were mainly from traffic sources and probably were partly from industrial sources; Ni and Cr mainly resulted from industrial activities and probably in part from traffic sources. The analysis of EF revealed moderate enrichments for Mn and Cr, and significant enrichments for Zn, Cu, and Ni in T1 and T2, and similarly in urban and suburban areas. Based on Igeo index, the levels of Mn and Fe were classified as uncontaminated, while Zn, Cu, Ni, and Cr levels were evaluated to have moderate contamination in T1 and T2, and similarly in urban and suburban areas. The values of RI showed a low level of heavy metals pollution.

Investigating and comparing short period impact of dust on physiological characteristics of three species of Pinus eldarica, Cupressus sempervirens, and Ligustrum ovalifolium

The purpose of this study is to know about the short period response of three green area plants to deposit of dust on the plants and impact on their physiological characteristics. In recent decade, the number of dusty days increased in some parts of Iran because of drought and erosive wind blowing on erodible lands. In this study, three species of Pinus eldarica, Cupressus sempervirens, and Ligustrum ovalifolium were selected. The amount of cumulative falling dust of 1, 2, and 4 g/cm2 was deposited on samples using a dust simulator in three stages with time interval of 1 week. At the end of each stage, physiological characteristics of the three treatment seedlings and three control seedlings were measured from each species. To further investigate the effect of dust, the position of dust on the leaves of the three plant species was prepared by Scanner Electronic Microscopic (SEM). Measurement results show that dust increase has no significant effect on the physiological parameters of proline, chlorophyll and carotenoids, and soluble sugar of P. eldarica and C. sempervirens species. But two mechanisms have been used by L. ovalifolium plants to compensate for the lack of light absorption. This has increased chlorophyll content and its growth. Increasing dust in L. ovalifolium has significant effect on chlorophyll and morphological structures, such as stem length and number of leaves. The results also showed that the amount of soluble sugars did not change by increasing dust. Since the impact of growth factors in this study showed a significant increase with increase of dust, one can conclude that the soluble sugars from photosynthesis have been used for the structure of the three studied plants. According to the study, short period dusting have no effect on P. eldarica and C. sempervirens species but have significant effect on L. ovalifolium.

A Study on Dust Storms Using Wind Rose, Storm Rose and Sand Rose (Case Study: Tehran Province)

Sand and dust storms are natural events that occur widely around the world, mostly in dry and bare lands. Over thepast decade a large part of Iran has been affected by this phenomenon, and Tehran has not been excluded from thisevent. Therefore, having knowledge of spatiotemporal frequency variations can aid us in future management andstorm trajectory assessment. In this study the synoptic anemometric data of Tehran-Mehrabad stations over the past47 years were used. Annual and monthly wind and storm roses were drawn to determine the predominant winddirection. The primary results indicated that the prevailing wind direction is western, but in summer, it is southsoutheastern.According to the threshold wind velocity (6.5 ms-1) about 12% of recorded data was categorized intothe storm condition.. Results also indicated that western, northwestern, and southwestern winds have the three highestrates of sand drift potential (Dpt), respectively. The temporal variability of dust storms showed that between 1951 and2005, the largest number of days characterized by dust and thunderstorms was observed in the spring and summer.The average number of dusty days in spring and summer was four d/m, and the fastest winds ranged from 40 to 49m/s-1. An assessment of dusty days over the past five decades revealed that the average number of dusty days hassignificantly increased from 10 days to 80 days. It seems that cyclonic circulations in the Monsoon Trough togetherwith regional winds in central Iran are responsible for the transmission of dust particles over central Iran. Results anddata from this study can be helpful in determining and predicting the critical dust storm period and in formingmanagement strategies in order to minimize aftermath impacts.

Regional differences in the relationship between climatic factors, vegetation, land surface conditions, and dust weather in China’s Beijing-Tianjin Sand Source Region

This paper analyzes the regional differences in dust weather and its relationship with climatic factors and vegetation cover, using data at 27 observation stations in the Beijing-Tianjin Sand Source Region from 1960 to 2007. There was a decreasing trend in dust weather in the southeast, which was correlated with wind and temperature. In contrast, in the northwest, the number of dusty days did not decrease and was significantly correlated with precipitation and vegetation cover. These results suggest that, in addition to the climate and underlying conditions, physical geographic conditions also influenced the frequency of dust weather.

Transport routes and source regions of Asian dust observed in Korea during the past 40 years (1965–2004)

This paper investigates regional sources and transport routes of Asian dust observed in Korea over the past 40 years, that is from 1965 to 2004. These long-term changes were studied in relation to climatic changes and causes. About 87% of dust events in Korea occur in the spring and have been increasing steadily since the 1980s. These trends can be attributed to regional changes in dust sources to windward and the resultant transport pathways of dust storms. This study establishes that dust events have greatly increased during the 3 years from 2000 to 2002. It may be the result of not only an eastward shift in the regional source of dust but also changes in upper-level synoptic transport patterns. Intensified atmospheric circulation and an associated increase in wind speed as well as changes in snow cover, surface wetness and potential evaporation rates throughout Inner Mongolia and Manchuria are likely to be the main causes of the steadily increasing number of dusty days in Korea.

A simple formula for estimating global solar radiation in central arid deserts of Iran

Over the last two decades, using simple radiation models has been an interesting task to estimate daily solar radiation in arid and semi-arid deserts such as those in Iran, where the number of solar observation sites is poor. In Iran, most of the models used so far, have been validated for a few specific locations based on short-term solar observations. In this work, three different radiation models (Sabbagh, Paltridge, Daneshyar) have been revised to predict the climatology of monthly average daily solar radiation on horizontal surfaces in various cities in central arid deserts of Iran. The modifications are made by the inclusion of altitude, monthly total number of dusty days and seasonal variation of Sun–Earth distance. A new height-dependent formula is proposed based on MBE, MABE, MPE and RMSE statistical analysis. It is shown that the revised Sabbagh method can be a good estimator for the prediction of global solar radiation in arid and semi-arid deserts with an average error of less than 2%, that performs a more accurate prediction than those in the previous studies. The required data for the suggested method are usually available in most meteorological sites. For the locations, where some of the input data are not reported, an alternative approach is presented.

Study of desert dust events over the southwestern Iberian Peninsula in year 2000: two case studies

Abstract. Strong desert dust events occurring in 2000 over the southwestern Atlantic coast of the Iberian Peninsula are detected and evaluated by means of the TOMS Aerosol Index (A.I.) at three different sites, Funchal (Madeira Island, Portugal), Lisboa (Portugal) and El Arenosillo (Huelva, Spain). At the El Arenosillo station, measurements from an AERONET Cimel sunphotometer allow a more precise retrieval of the spectral AOD and the derived alpha Ångström coefficient. After using different threshold values of these parameters, we conclude that it is difficult to establish reliable and robust criteria for an automatic estimation of the number of dust episodes and the total number of dusty days per year. As a result, additional information, such as airmass trajectories, were used to improve the estimation, from which reasonable results were obtained (although some manual editting was still needed). A detailed characterization of two selected desert dust episodes, a strong event in winter and another of less intensity in summer, was carried out using AOD derived from Brewer spectrometer measurements. Size distribution parameters and radiative properties, such as arefractive index and the aerosol single scattering albedo derived from Cimel data, were analyzed in detail for one of these two case studies. Although specific to this dust episode, the retrieved range of values of these parameters clearly reflect the characteristics of desert aerosols. Back-trajectory analysis, synoptic weather maps and satellite images were also considered together, as supporting data to assess the aerosol desert characterization in this region of study.

Analysis of dust storms observed in Mongolia during 1937–1999

Climatology of dust storms in Mongolia is compiled based on observational data of 49 meteorological stations from 1960 to 1999 and compared them with data between 1937 and 1989. Three different maps of the distribution of dust storms, drifting dust and the number of dusty days are presented. The results of the analysis show that the number of days with dust storms is <5 days over the Altai, the Khangai and the Khentei mountainous regions and more than 20–37 days in the Gobi Desert and semi-desert area. The greatest occurrence of drifting dust arises around the Mongol Els area of west Mongolia. The number of dusty days, which is derived from the sum of the number of days with dust storms and drifting dust, is <10 days in the mountainous area and 61–127 days in the Gobi Desert and the Great Lakes hollow of west Mongolia. It is found that 61% of dust storms occur in the spring in Mongolia and a dust storm lasts on average from 3.1 to 6.0 h. About 65.5–91.0% of dust storms occur in daytime and 9.0–34.5% at night. Dust storms occur more frequently in the city region and are accompanied by surface wind speeds usually from 6 to 20 ms −1. Dust storms usually occur when soil and air are dry, and 70% of dust storms occur in dry soil conditions. When dust storms occur, relative humidity averages 20–40% in Mongolia. An important outcome of this study is the trend of dusty days between 1960 and 1999. It shows that the number of dusty days has tripled from the 1960s to 1990s and has decreased since 1990.

Response of the climatic temperature to dust forcing, inferred from total ozone mapping spectrometer (TOMS) aerosol index and the NASA assimilation model

Recently, Alpert et al. (Alpert, P., Shay-El, Y., Kaufman, Y.J., Tanre, D., DaSilva, A., Schubert, S., Joseph, J.H., 1998. Quantification of dust-forced heating of the lower troposphere, Nature 395 (6700), 367–370, (24 September).) suggested an indirect measure of the tropospheric temperature response to dust aerosols by using model updates — roughly speaking model errors — of the NASA Goddard Earth Observing System version No. 1 (GEOS-1) data assimilation system. They have shown that these updates, which provide information about missing physical processes not included in the predictive model, have monthly mean patterns, which bear a striking similarity to patterns of dust over the Atlantic. This similarity in the number of dusty days was used to estimate the atmospheric response to dust. Here, the study is extended for all the major subtropical deserts over Africa and Asia using the total ozone mapping spectrometer (TOMS) aerosol index ( AI) for dust recently derived by Herman et al. (Herman, J.R., Bhartia, P.K., Torres, O., Hsu, C., Seftor, C., Celarier, E., 1997. Global distribution of UV-absorbing aerosols from Nimbus 7/TOMS data, J. Geophys. Res. 102, 16911–16922.). It is shown that the TOMS dust is highly correlated with the model errors with a maximum at the altitude of about 580 hPa and for the month of June with average correlation coefficient of 0.69 reaching up to 0.8 for specific months. In contrast to the previous study where only dust over ocean was employed, here, much higher dust concentrations are detected and the linear heating for weak dust becomes quickly saturated for AI above 1.5, then drops for very high values of AI that exceed about 3. This result is consistent with the theoretical predictions.

Quantification of dust-forced heating of the lower troposphere

Aerosols may affect climate through the absorption and scattering of solar radiation and, in the case of large dust particles, by interacting with thermal radiation1,2,3. But whether atmospheric temperature responds significantly to such forcing has not been determined; feedback mechanisms could increase or decrease the effects of the aerosol forcing. Here we present an indirect measure of the tropospheric temperature response by explaining the ‘errors’ in the NASA/Goddard model/data-assimilation system. These errors, which provide information about physical processes missing from the predictive model, have monthly mean patterns that bear a striking similarity to observed patterns of dust over the eastern tropical North Atlantic Ocean. This similarity, together with the high correlations between latitudinal location of inferred maximum atmospheric heating rates and that of the number of dusty days, suggests that dust aerosols are an important source of inaccuracies in numerical weather-prediction models in this region. For the average dust event, dust is estimated to heat the lower atmosphere (1.5–3.5 km altitude) by ∼0.2 K per day. At about 30 dusty days per year, the presence of the dust leads to a regional heating rate of ∼6 K per year.