In zones where aerosol properties have been poorly characterized, satellite-based (MODIS) and ground-based (AERONET) aerosol optical depth (AOD) values typically differ. In this work, we use machine-learning based methods (artificial neural networks and support vector machines) to obtain corrected AOD values taken from MODIS in regions that are positioned far from AERONET stations. The method has been validated using several approaches. The area suitable for improvement covers 62

We present a method to correct aerosol optical depth (AOD) values taken from Collection 6 MODIS observations, which resulted in values closer to those recorded by the ground-based network AERONET. The method is based on machine learning techniques (Artificial Neural Networks and Support Vector Regression), and uses MODIS AOD values and meteorological parameters as inputs. The method showed improved results, compared with the direct MODIS AOD, when applied to nine stations in South America. The percentage of improvement, measured in terms of R2, ranged from 2% (Alta Floresta) to 79% (Buenos Aires). This improvement was also quantified considering the percentage of data within the MODIS expected error, being 91% for this method and 57% for direct correlation. The method corrected not only the systematic bias in temporal data series but also the outliers. To highlight this ability, the results for each AERONET station were individually analyzed. Considering the results as a whole, this method showed to be a valuable tool to enhance MODIS AOD retrievals, especially for locations with systematic deviations.

Many Chemistry Climate Models (CCMs) include a simplified treatment of brominated very short-lived (VSLBr) species by assuming long-lived methyl bromide (CH3Br) as a surrogate for VSLBr. However, given that VSLBr (i.e., bromoform CHBr3 and dibromomethane CH2Br2) decompose more rapidly than CH3Br, their impact on upper tropospheric chemistry and lowermost stratospheric ozone cannot be neglected. Thus, a mistreatment of VSLBr in CCMs may yield an unrealistic representation of their associated impacts. Here, we present a comprehensive intercomparison between various VSLBr chemical approaches with increasing degrees of complexity (i.e., surrogate, explicit, and full), and quantify the global impacts of these natural bromocarbons on tropospheric and stratospheric ozone, as well as on other oxidizing agents. Differences between chemical schemes maximize in the lowermost stratosphere and mid-latitude free troposphere, resulting in a latitudinally dependent reduction of ~1−7 DU in total ozone column and a ~5−15 % decrease of the OH/HO2 ratio, for full compared to surrogate. These bromine-driven changes in HOx abundances are expected to slow-down the oxidative processing of greenhouse gases (i.e., to increase the CH4 lifetime) in a region where these long-lived species have a final chance to undergo tropospheric degradation before injection to the stratosphere. Given the negligible additional computational cost and chemical complexity, we encourage all CCMs oriented to projecting the coupled evolution of stratospheric ozone within a changing climate to include a complete tropospheric representation of VSLBr sources and chemistry in the troposphere and stratosphere.

Four monitoring campaigns between the years 2009 and 2018 were conducted in Córdoba City, Argentina, to detect toxic metals in PM2.5 samples. The concentrations of As, Cd, Pb, Cu, Cr, Mn, Hg, Ni, and Zn, together with several other elements, were measured. The average metal concentrations followed the order: Zn > Cr > Cu > Mn > Pb > V > Ni > As ~ Sb > Cd > Tl > Pd > Hg > Pt. From the analysis of the temporal variation in the elemental concentration of PM2.5, results show seasonal variations that reach, in general, a maximum in the coldest seasons and a minimum in the warmer seasons. These differences could be explained by the different weather conditions during each season, the influence of the El Niño/La Niña regimen, and the presence of fires on certain sampling dates. The source apportionment analysis performed for the period 2017–2018 showed the contribution to PM2.5 of combustion of heavy fuel oil and diesel-powered vehicles, pet coke, metallurgical and nonferrous industries, paint plant factory, traffic, and natural sources like the soil and road dust. This last analysis completed the assignment of sources for the 10-year period of study. Thus, the results of this work contribute to the implementation of emission reduction strategies in order to decrease the impact of PM2.5 on the environment and the human health.

Accurate estimates of total global solar irradiance reaching the Earth’s surface are relevant since routine measurements are not always available. This work aimed to determine which of the models used to estimate daily total global solar irradiance (TGSI) is the best model when irradiance measurements are scarce in a given site. A model based on an artificial neural network (ANN) and empirical models based on temperature and sunshine measurements were analyzed and evaluated in Córdoba, Argentina. The performance of the models was benchmarked using different statistical estimators such as the mean bias error (MBE), the mean absolute bias error (MABE), the correlation coefficient ( r ), the Nash-Sutcliffe equation (NSE), and the statistics t test ( t value). The results showed that when enough measurements were available, both the ANN and the empirical models accurately predicted TGSI (with MBE and MABE ≤ |0.11| and ≤ |1.98| kWh m −2 day −1 , respectively; NSE ≥ 0.83; r ≥ 0.95; and | t values| < t critical value). However, when few TGSI measurements were available (2, 3, 5, 7, or 10 days per month) only the ANN-based method was accurate (| t value| < t critical value), yielding precise results although only 2 measurements per month were available for 1 year. This model has an important advantage over the empirical models and is very relevant to Argentina due to the scarcity of TGSI measurements.

A total of 175 aerosol samples were collected during the winter-spring months of 2014-2015 in Cordoba, Argentina. The samples were collected in five size fractions that ranged from < 0.25 to 10 mu m using a SIOUTAS impactor in a 6-h period and analyzed by synchrotron radiation X-ray fluorescence to obtain elemental aerosol mass concentrations for As, Ca, Co, Cr, Cu, Fe, K, Mn, Ni, Pb, Si, S, Ti, V, and Zn. An analysis of the distribution of the elements in the different collected fractions was performed to infer the most likely sources of particulate matter. From the analysis, it was found that the particulate matter fraction smaller than 0.25 mu m was mostly affected by emissions from traffic and combustion, while the particulate matter fraction ranging from 2.5 to 10 mu m was mostly influenced by re-suspension of road dust. In addition, the results indicated the presence of Si, typically found in coarse particulate matter and in large amounts in particles smaller than 0.25 mu m. To infer the sources of aerosols causing the deterioration of air quality, principal component analysis was performed. The smallest particles, due to their toxicity, are likely to produce more adverse health effects than larger size fractions. These results show the need to monitor their atmospheric emissions in order to preserve air quality. Synchrotron radiation X-ray fluorescence analysis of the elemental composition of particles smaller than 0.25 mu m in Cordoba city constitutes the first contribution to the knowledge base of the concentrations of toxic metals in this size fraction. These results are relevant because there is a lack of data in South American cities.

Global ultraviolet-B irradiance (UV-B, 280-315 nm) measurements made at the campus of the University of Cordoba, Argentina were analyzed to quantify the effects of ozone and aerosols on surface UV-B erythemal irradiance (UVER). The measurements have been carried out with a YES Pyranometer during the period 2000-2013. The effect of ozone and aerosols has been quantified by means of the Radiation Amplification Factor (RAF) and by an aerosol factor (AF, analogous to RAF), respectively. The overall mean RAF under cloudless conditions was (1.2 +/- 0.3) %, ranging from 0.67 to 2.10% depending on solar zenith angle (SZA) and on Aerosol Optical Depth (AOD). The RAF increased with the SZA with a clear trend. Similarly, the aerosol effect under almost-constant ozone and SZA showed that, on average, a 1% increase in AOD forced a decrease of (0.15 +/- 0.04) % in the UVER, with a range of 0.06 to 0.27 and no defined trend as a function of the SZA. To analyze the effect of absorbing aerosols, an effective single scattering albedo (SSA) was determined by comparing the experimental UVER with calculations carried out with the TUV radiative transfer model. (C) 2017 Elsevier Ltd. All rights reserved.

In this work, PM2.5 samples were collected in the winter-spring months of 2014-2016 at an urban site in Cordoba. Cordoba is the second largest city in Argentina and is an important industrial and touristic center. The collected samples were individually analyzed for chemical composition using different techniques. The soluble inorganic ions and carbonaceous particles were determined from bulk aerosol samples for the first time in the city. The mass concentrations of PM2.5, organic carbon, elemental carbon, inorganic ions and metals were determined according to the mass balance. The dominant mass components were organic matter and elemental carbon (54.8%), mineral dust (6.1%), secondary inorganic aerosols (3.0%), and salt (1.2%). A principal component analysis was applied to the samples and resulted in five major factors that explained 79% of the variance in PM2.5. These factors represented combustion, industrial sources, soil dust, secondary inorganic aerosol, and salt, and each explained between 11% and 20% of the variance. A comparison with the results from a previous campaign (2010-2011) revealed appreciable changes in the PM2.5 chemical composition. These changes were attributed to the two extreme meteorological conditions that prevailed in the region. The years 2014-2016 were largely dominated by the warm phase of the El Nifio Southern Oscillation, which leads to humid and cold weather in the Cordoba region, while the samples from 2010 to 2011 were collected during the dry and hot years resulting from the La Nifia regime. (C) 2017 Elsevier Ltd. All rights reserved.

AEROSOL Robotic Network (AERONET), Moderate Resolution Imaging Spectroradiometer (MODIS) and global UV-B (280–315nm) irradiance measurements and calculations were combined to investigate the effects of aerosol loading on the ultraviolet B radiation (UV-B) reaching the surface under cloudless conditions in Córdoba, Argentina. The aerosol radiative forcing (ARF) and the aerosol forcing efficiency (ARFE) were calculated for an extended period of time (2000–2013) at a ground-based monitoring site affected by different types and loading of aerosols. The ARFE was evaluated by using the aerosol optical depth (AOD) at 340nm retrieved by AERONET at the Cordoba CETT site. The individual and combined effects of the single scattering albedo (SSA) and the solar zenith angle (SZA) on the ARFE were also analyzed. In addition, and for comparison purposes, the MODIS AOD at 550nm was used as input in a machine learning method to better characterize the aerosol load at 340nm and evaluate the ARFE retrieved from AOD satellite measurements. The ARFE at the surface calculated using AOD data from AERONET ranged from (–0.11±0.01) to (–1.76±0.20) Wm–2 with an average of –0.61 Wm–2; however, when using AOD data from MODIS (TERRA/AQUA satellites), it ranged from (–0.22±0.03) to (–0.65±0.07) Wm–2 with an average value of –0.43 Wm–2. At the same SZA and SSA, the maximum difference between ground and satellite-based was 0.22 Wm–2.

In this work, we present a method to predict missing aerosol optical depth (AOD) values at an AERONET station. The aim of the method is to fill gaps and/or to extrapolate temporal series in the station datasets, i.e. to obtain AOD values under cloudy sky conditions and in other situations where there is a temporary or permanent lack of data. To accomplish that, we used historical AOD values at two stations, air mass trajectories passing through both of them (calculated by using the HYSPLIT model) and ANN calculations to process all the information. The variables included in the neural network training were the station numbers, parameters representing the annual average trend of meteorological conditions, the number of hours and the distance traveled by the air mass between the stations, and the arrival height of the air mass. The method was firstly applied to predict AOD at 440 nm in 9 stations located in the East Coast of the US, during the years 1999-2012. The coefficient of determination r(2) between measured and calculated AOD values was 0.855, which show the good performance of the method. Besides, this result represents a remarkable improvement compared to three simple approaches. To further validate the method, we applied it to another region (Iberian Peninsula) with different characteristics (lower density of AERONET stations, different meteorology, and lower wind field spatial resolution). Although the results are still good (r(2) = 0.67), the performance of the method was affected by these characteristics. Considering the obtained results, this method can be used as a powerful tool to predict AOD values in several conditions. The methodology can also be easily adapted to predict AOD values at other wavelengths or other aerosol optical properties. (C) 2015 Elsevier Ltd. All rights reserved.

This work presents the analysis of the long-term observations of aerosol optical properties in the central region of Argentina. Monitoring of aerosol parameters was carried out at the Cordoba-CETT AERONET site (31° 31′ S, 64° 27′ W, 730m.a.s.l.) from November 1999 until December 2010. Long-term measurements of aerosol optical depth, Ångström exponent, fine mode fraction and single scattering albedo were analyzed and compiled to describe the climatology of the optical properties of the aerosols of the region. The knowledge of the optical properties of aerosols and their spatial distribution is required to evaluate aerosol effects on the climate system. This information provides an opportunity for understanding how aerosols might influence the regional radiation budget. Results show that aerosol optical depth at 340nm is characterized by low values from February to April (monthly average of 0.15±0.05), very low values from May to June (monthly average of 0.08±0.03) and a sustained increase from July to September (monthly average of 0.20±0.09) reaching a value of 0.26. From this dataset, no long-term trends are observable. Results of the inter-annual variations of the Ångström exponent between 440 and 870nm reflect an important difference in the year 2004 compared to the other 11 years of the study. A possible explanation of this fact is elaborated with the help of back trajectory analysis. Finally, three episodes are described and analyzed, as they produced important increases of the daily aerosol optical depth value. We explained these episodes with a combination of air mass trajectory analysis, meteorology and the MODIS fire counts product.

Bulk aerosol samples collected during 2010 and 2011 at one receptor site in Córdoba City, Argentina, have been quantitatively analyzed to determine aerosol elemental composition by using SR-XRF. A receptor model analysis has been applied to ambient PM2.5 measurements. Four sources have been identified being their contributions: traffic: (13 ± 2) μg m−3, SO42−/combustion processes, including biomass burning: (15 ± 1) μg m−3, mineral dust: (7 ± 2) μg m−3 and industry: (8 ± 1) μg m−3. Source identification was carried out by inspection of key species in source profiles, seasonality of source contributions, comparison with literature data and the knowledge of the city; for the biomass burning contribution the MODIS burned area daily product was used to confirm wildfire events along the year. In addition, from May to August 2011, aerosols were collected in two additional size fractions (PM0.25–0.5, PM0.5–1) to investigate the toxic metal contributions to the finer fractions. An important result of this work is that toxic metals make an important contribution to the finest (PM0.25–0.5) size fraction. The results of the present analysis can help to demonstrate to local and national authorities the urgent need to carry out emission inventories, to implement air quality monitoring systems and to set regulations for PM2.5.

The main purpose of this work is to determine the relative contribution of different types of aerosols at an urban site by using two independent approaches: individual particle analysis, and radiative transfer calculations and irradiance measurements. To accomplish that purpose, we used our UV-B irradiance (280–315nm) data set, the AERONET (AErosol RObotic NETwork) database, the SEM (Scanning Electron Microscopy, LEO 1450VP) analyses of the collected particles and the Santa Barbara DISORT Atmospheric Radiative Transfer (SBDART) model. On one hand, the collected particles were analyzed by SEM-EDX (Energy Dispersive X-Ray, Genesis 2000) in order to determine their chemical composition. Then, by using a developed algorithm they were classified as rural or urban, resulting in a (24±3)% of rural and (76±8)% of urban. On the other hand, aerosols were incorporated into the SBDART model through two of its default profiles (urban or rural) and by using the Aerosol Optical Depth (AOD) provided by AERONET. The aerosol effect on experimental surface UV-B irradiance was reproduced by a linear combination of the irradiances calculated by using these profiles. From this analysis we found that, in average, a mix of aerosols of (30±3)% rural and (70±7)% urban explains the observed reduction in the experimental irradiance. Considering the agreement between the results obtained by using these two independent methodologies, the use of the irradiance as a surrogate variable to retrieve aerosol types is discussed. The methodology presented here is applicable to any site provided irradiance measurements and AOD are available.

Broadband measurements of global and diffuse UV-B irradiance (280–315nm) together with modeled and measured diffuse to global ratios (DGR) have been used to characterize the influence of different types of clouds on irradiance at the surface. Measurements were carried out during 2000–2001 in Córdoba City, Argentina. The Tropospheric Ultraviolet Visible (TUV) model was used to analyze the behavior of the modeled DGRs for different cloud optical depths and at different altitudes and solar zenith angles (SZA). Different cloud altitudes were also tested, although only the results for a cloud placed at 1.5–2.5km of altitude are shown. A total of 16 day with stratocumulus, 12 with cumulus, and 16 with cirrus have been studied and compared among them and also against 21 clear sky days. Different behaviors were clearly detected and also differentiated through the analysis of the averages and the standard deviations of the DGRs: 1.02±0.06 for stratocumulus, 0.74±0.18 for cumulus, 0.63±0.12 for cirrus, and 0.60±0.13 for the clear sky days, respectively. Stratocumulus clouds showed a low variability in the DGR values, which were concentrated close to one at all SZAs. DGR values for cumulus clouds presented a large variability at all SZAs, mostly associated with the different optical depths. Finally, the closeness between the DGR values for cirrus clouds and the DGR values for clear days showed that these clouds generally do not strongly affect the UV-B irradiance at the surface at any SZA. In the opposite side, stratocumulus clouds were identified as those with the largest effects, at all SZAs, on the UV-B irradiance at the surface.

24-h samplings of PM10 and PM2.5 have been carried out during the period July 2009 April 2010 at an urban and at a semi-urban site of Cordoba City (Argentina). The samples in the PM2.5 fraction weighted in the average 71 +/- 21 mu g m(-3) and 67 +/- 18 mu g m(-3) respectively, whereas the samples of the same sites in the PM10 fraction weighted 107 +/- 31 mu g m(-3) and 101 +/- 14 mu g m(-3). The chemical composition of aerosol particles was determined by synchrotron radiation X-ray fluorescence (SR-XRF). Elemental composition was different in the two fractions: in the finer one the presence of elements with crustal origin is reduced, while the anthropogenic elements, with a relevant environmental and health impact, appear to be increased. An important but unmeasured component is likely constituted by organic and elemental carbon compounds. Multivariate analysis (Positive Matrix Factorization) of the SR-XRF data resolved a number of components (factors) which, on the basis of their chemical compositions, were assigned physical meanings. (C) 2011 Elsevier Ltd. All rights reserved.

Ultraviolet (UV) actinic fluxes measured with two Scanning Actinic Flux Spectroradiometers (SAFS) aboard the NASA DC-8 aircraft are compared with the Tropospheric Ultraviolet-Visible (TUV) model. The observations from 17 days in July-August 2004 (INTEX-NA field campaign) span a wide range of latitudes (28 degrees N-53 degrees N), longitudes (45 degrees W-140 degrees W), altitudes (0.1-11.9 km), ozone columns (285-353 DU), and solar zenith angles (2 degrees-85 degrees). Both cloudy and cloud-free conditions were encountered. For cloud-free conditions, the ratio of observed to clear-sky-model actinic flux (integrated from 298 to 422 nm) was 1.01 +/- 0.04, i.e. in good agreement with observations. The agreement improved to 1.00 +/- 0.03 for the down-welling component under clear sky conditions. In the presence of clouds and depending on their position relative to the aircraft, the up-welling component was frequently enhanced (by as much as a factor of 8 relative to cloud-free values) while the down-welling component showed both reductions and enhancements of up to a few tens of percent. Including all conditions, the ratio of the observed actinic flux to the cloud-free model value was 1.1 +/- 0.3 for the total, or separately 1.0 +/- 0.2 for the down-welling and 1.5 +/- 0.8 for the up-welling components. The correlations between up-welling and down-welling deviations are well reproduced with sensitivity studies using the TUV model, and are understood qualitatively with a simple conceptual model. This analysis of actinic flux observations illustrates opportunities for future evaluations of photolysis rates in three-dimensional chemistry-transport models.

A description of the non-LTE intramolecular ozone relaxation cascade considering the relative importance of different transitions on the O(3)(v(1)v(2)v(3)) vibrational levels has been studied by quantifying the kinetic law of every process and transition that affect each level population. The analysis considers the inclusion for the first-time of bending-to-stretching (k(2D)) transitions in the energy relaxation cascade of ozone after it is formed by three-body recombination. In this way, the vibrational temperatures and the relative contribution of every transition are presented as a function of altitude. The results show that the inclusion of the bending-to-stretching transitions in the O(3) non-LTE relaxation scheme decreases the v(2)-mode overpopulation and therefore the stretching levels' population increases, as required for a correct radiance simulation of the 4.8 mu m ozone emission in the upper atmosphere. (C) 2010 Elsevier Ltd. All rights reserved.

The effect of clouds on total and UV-B irradiance in Cordoba, Argentina, was studied employing the TUV 4.1 model and measurements obtained with YES UVB-1 and YES TSP-700 radiometers, and a spectral radiometer Ocean Optics USB-4000. The experimental measurements were selected from a 10 years dataset (1999-2008). Clouds were classified by direct observation as cirrus, cumulus, and stratocumulus. The broadband Cloud Modification Factors (CMFs) have been calculated in the range of the total and the UV-B radiation for these types of clouds. The relations between them were analyzed for a significant number of days. The broadband CMF values range from around 0.1 up to 1.25, depending on the wavelength interval and on the cloud. type. The CMFUVB versus CMFT plots for different clouds have shown good adjustments and significant differences, which allows the distinction between them.Stratocumulus clouds show large attenuations and a linear relation with larger slopes as the solar zenith angle (SZA) increases. For this type of clouds an average slope of (1.0 +/- 0.2) was found. The relation between the CMF for cumulus clouds is linear with an average slope of (0.61 +/- 0.01). No dependence with the SZA was observed. Cirrus clouds plots show an exponential behavior with fit parameters equal to (0.48 +/- 0.08) and (0.68 +/- 0.15). However, when small SZA intervals are analyzed a linear relation is found. When the relations between the CMF were similar (cumulus and cirrus), the spectral variation in the UV range (320-420 nm) of a modified CMF (CMFm) was used to distinguish them. Hence, the spectral differences among the three types of clouds have been also analyzed for several days and SZA. Here, it was found that the effect of cirrus is essentially wavelength independent while cumulus and stratocumulus clouds show exponential decay relations but with different ordinates.In the analyzed relations the microphysical properties of the clouds seem to determine its behavior while the optical thickness leads to the different degrees of attenuation.The results obtained in this work are in agreement with those found for other authors. (C) 2009 Elsevier Ltd. All rights reserved.

The relative importance of different processes and transitions in the appearance of non-LTE populations for the fundamental levels of ozone was studied by quantifying the kinetic law of every process and transition that affect each level population. The vibrational temperatures and the relative contribution of every transition are presented as a function of altitude. The results show that the appearance of non-LTE populations for the fundamental levels is not produced by a direct imbalance between absorption and emission but as a consequence of imbalanced collisional transitions provoked by the overpopulation of O3(0v20) levels. This fact confirms that the relaxation cascade drives through the bending mode ν2.

The main purpose of this study was to perform an evaluation of the particles’ optical parameters’ influence on surface solar UV-B (280–315nm) irradiance in Córdoba, Argentina. To achieve this objective UV-B irradiance dataset, AERONET (AErosol RObotic NETwork) database, and TUV (Tropospheric Ultraviolet and Visible) model were used to analyze the effects of aerosols on surface irradiance on cloudless days during specific days of winter and spring of the period 1999–2006. Together with a direct observer, total irradiance (300–3000nm) measurements were used as an ancillary tool to verify the cloudless condition. Every year, during this period, important reductions in surface irradiance are observed due to the aerosol load. Aerosols were incorporated in the model through the aerosol optical depth at 340nm, the asymmetry parameter at 440nm, and the single scattering albedo at 440nm, all of them provided by AERONET Córdoba-CETT site. These factors vary from near to zero up to 1.4, from 0.56 up to 0.83 and from 0.43 up to 0.99, respectively. The behaviors of these factors along the year are analyzed considering the meteorology of Córdoba. When AERONET data are included in the TUV model they allow an accurate simulation of the UV-B irradiance, making the agreement with the experimental measurements substantially better. Only a small differences (±2%) remains, which can be attributed to diverse factors. As the AERONET site is 20km away from the irradiance measurement site, these results show the regional character of the aerosols in Córdoba, although small contributions of urban aerosols are not discarded. An episode of high aerosol and pollutants laden due to fires in the surrounding hills is briefly analyzed. A set of additional studies are needed to describe comprehensively the characteristics and behavior of the Córdoba aerosols. Some of them are being already carried out.