Aerosol Science and Engineering

, Volume 2, Issue 2, pp 74–91 | Cite as

Fine-Mode Aerosol Loading Over a Sub-Sahel Location and Its Relation with the West African Monsoon

  • Okechukwu K. Nwofor
  • Victor N. Dike
  • Zhaohui Lin
  • Rachel T. Pinker
  • Nnaemeka D. Onyeuwaoma
Original Paper


The aerosol fine-mode fraction (FMF) at the sub-Sahel AErosol RObotic NETwork (AERONET) site at Ilorin (8°32′N; 4°34′E) is found to be the highest (FMFmean = 0.487) compared to six upper Sahel AERONET sites. The fine-mode aerosol population at the site dominates the coarse mode in core West African Monsoon months of June–July–August (FMFmean = 0.581; Angstrom exponent derivative = 0.44). Correlations (r) of aerosol optical depth (AOD) time series with corresponding seasonal zonal wind (ZW) and meridional wind (MW) speeds of the European Centre for Medium-Range Weather Forecasts at the seven AERONET sites reveal a uniquely strong positive value (r = 0.6) of wet-season AOD and MW at Ilorin. The wet-season FMF distribution at the site is bimodal with a broad mode (peak center = 0.685; half-width = 0.521) attributed to a wide range of industrial/urban aerosols and a narrow mode (peak center = 0.338; half-width = 0.136) attributed to fine dust aerosols, while the dry-season distribution is mono-modal, attributed to a fairly broad dust/biomass burning aerosol mixture (peak center = 0.484; half-width = 0.394). These are corroborated with 7-day back trajectories calculated for core wet- and dry-season months over 2 years indicating mainly high altitude maritime and continental air masses in the wet season and lower altitude Sahara and Sudanian air masses in the dry season. Comparison of inter-annual rainfall and FMF trends indicates coherence of intensifying rainfall in traditional dry-season months (December, January, and February) with decreasing FMF distribution means and increasing FMF distribution widths which are consistent with reducing dust and biomass burning aerosols and growing industrial and urban aerosol sources.


Fine-mode aerosol WAM Ilorin AERONET Sahel 

1 Introduction

Knowledge of the relative atmospheric loading of aerosol particles of various size groups is crucial for improved understanding of roles of atmospheric aerosols in earth’s radiation budget, atmospheric visibility impairment, and environmental health. The fine-mode aerosols—which are aerosols typically of radius < ~ 1 µm (Husar 2005)—are particularly important in radiative forcing calculations because of their large absorption properties and cloud microphysical properties (Charlson et al. 1992; Boucher et al. 2013; Szopa et al. 2013; Smith et al. 2016). They greatly degrade visibility due to their strong scattering and absorption of visible light (Lee et al. 2005) and impact respiratory health severely, since they easily find their way into the air ways (Andrade-Filho et al. 2013). They are also implicated in various forms of lung cancer (see for, e.g., Tie et al. 2009). Information on fine-mode aerosol loading is very valuable in assessing human contributions to aerosol emissions. This is because fine aerosol mode is populated mainly by a range of carbonaceous aerosols of anthropogenic origin including biomass burning, fossil fuel combustion, bio-fuels, and gas flaring (Zhang et al. 2016), and the uncertainty in total anthropogenic radiative forcing is usually attributed to these aerosols (Boucher et al. 2013; Myhre et al. 2008, 2013). In many Sahel areas, fine-mode aerosols can be very intense because of the location of the areas within and downwind biomass burning sites as noted by Remer et al. (2009)—and from several unaccounted sources, huge inconsistencies between model calculations and aerosol-loading observations may be expected on account of the fine mode as is the case for Ilorin (see for, e.g., Lesins and Lohmann 2003 and Cesnulyte et al. 2014). Global aerosol modeling efforts would, therefore, benefit from more facts on the fine-mode aerosol sources particularly in relation with climate dynamics (e.g., Zuluaga, et al. 2012; Chung et al. 2016; Manish et al. 2017) and more importantly for the Ilorin area.

This study examines fine-mode aerosol fraction over the AErosol RObotic NETwork (AERONET) site of Ilorin relative to those of upper Sahel sites and its link with the West African Monsoon (WAM). The relation between WAM and Sahel aerosol loading is still an emerging body of knowledge and Ilorin is a very important site in that regard. Aerosol loading at the site was previously considered to be mainly of “dust” especially when considered in association with the annual harmattarn wind which is very intense in the area (see Pinker et al. 1994; Pandithurai et al. 2001) and subsequently as “mixed” owing to the phenomenal contributions from biomass burning aerosols—which combine almost equally with dust during the dry season (see for, e.g., Pinker et al. 2010; Eck et al. 2010; Giles et al. 2012). Although Ilorin is still a small town of ~ 910,000 people (Demographia 2017), and not highly industrialized yet, it is upwind (at least in the monsoon season) of mainly coastal urban and industrialized cities in West Africa such as Lagos, Cotonou, Lome, and Accra. Possibilities of aerosols from industrial emissions from these cities as well as from off-shore gas flaring being driven by the monsoon wind to the site were envisaged in Eck et al. (2010) and also considered in Onyeuwaoma et al. (2015). Consequently, Fawole et al. (2016) detected off-shore gas flaring signatures at Ilorin during the WAM months using AERONET data and backward trajectory analysis. This paper aims to contribute to the growing body of knowledge regarding WAM relations with aerosol loading generally and with fine-mode aerosol loading in particular especially over the sub-Sahel area. It considers Ilorin as a unique fine-mode aerosol-loading site, where in addition to dry-season biomass burning, the monsoon wind deposits aerosols from several urban locations in the south of Nigeria and other parts of West Africa in the wet season. It analyzes AERONET and wind field data for correlations between aerosol loading and wind direction. Multi-modal frequency distributions of seasonal fine-mode aerosol population were used with information from Angstrom exponent derivative to classify fine-mode aerosol types contributing to the aerosol loading during the seasons. Monsoon-season source attribution was supported with 7-day NOAA HYSPLIT back trajectories calculated for monsoon and dry-season months. Association between recent increases in rainfall as revealed using several precipitation data sets and fine-mode aerosol-loading characteristics at the site was examined.

2 Study Site, Data, and Methods

2.1 Climate and Aerosol Dynamics of the Study Site

Ilorin is a moderately commercial town located in the West African sub-Sahel (latitude 8°32′N; longitude 4°34′E; 307 m above sea level). The climate of the town derives from that of West Africa (longitudes 20°W and 20°E and latitudes 0° and 20°N) which is essentially tropical but with significant spatial variations which result in multiple climate zones that stretch from the Atlantic coast and progressively increasing in dryness towards the Sahara in the north. These different zones are distinguished mainly by differences in mean rainfall. The major divisions are the tropical rainforest, the sub-Sahel and the semi-arid and arid regions in the upper Sahel. Annual average rainfall is between 100 and 200 mm in the upper Sahel and between 500 and 600 mm in the lower Sahel (Nicholson 2013). The West African monsoon (WAM) moderates the rain-producing systems in the entire West Africa. WAM, often defined as seasonal changes in wind direction, is typically characterized by fluctuations of tropical maritime air mass (south-westerly wind) and the tropical continental air mass (north-easterly wind) separated by the Inter-Tropical Convergence Zone (ITCZ) over ocean areas and the Inter-Tropical Discontinuity (ITD) at the surface over continental West Africa. The rain-producing systems include the zonal African easterly jet (AEJ) (~ 600–700 mb; ~ 10 m s−1), the high level (~ 200 mb) meridional tropical easterly jet (TEJ) associated with the Asian monsoon outflow, African easterly waves (AEWs), and mesoscale convective systems (MCSs) (see Sylla et al. 2013). These phenomena interact in a complex way with the low-level monsoon flow, which transports moisture inland from the Atlantic Ocean, providing West Africa with most of its moisture for rainfall (Nicholson 2013). The Sahel ordinarily defines the ITCZ’s most northerly position and it is thought that a southern shift in the 70s and the 80s resulted in the much orchestrated and prolonged Sahel drought (see Giannini et al. 2003). The Sahel drought left in its wake great aridity up to the sub-Sahel area, where remnants of vegetation usually of grassland and shrubs were easily burnt as biomass fuel. This situation seems to be changing with recovery of vegetation from more rains.

Within Ilorin, the difference between average rainfall in the driest month and average rainfall of the rainiest month is about 200 mm. Dry-season wind speeds range from about 20 km/h to about 280 km/h with modal value at about 130 km/h and ~ 70% of the variance in December–January–February (DJF), while wet-season wind speeds range from about 60–260 km/h with mode at about 160 km/h and ~ 75% of the variance in June–July–August (JJA), and these tend to moderate the scenarios of emission, dispersion, and removal of aerosols at the site (see Nwofor et al. 2007; Pinker et al. 2010). These wind speeds fundamentally connected to the two seasonal convective systems, i.e., the monsoon and harmattarn, produce maximum aerosol loading at the frontal area of the inter-tropical discontinuity zone (ITZ) around the sub-Sahel area, where Ilorin is located (see Roberts et al. 2009; Liousse et al. 2010). A major fine-mode aerosol-loading process over the entire Sahel is associated with the north-easterly low-level jet (LLJ) of near-surface wind speeds exceeding 10 ms−1 (see Ben-Ami et al. 2010) transporting both dust from the Bodele axis (estimated to be responsible for much of global dust emissions) and biomass burning aerosols from the Sudania zone (see Flaounas et al. 2016). The monsoon south-westerly on the other hand is considered to be primarily responsible for the transport of sea salt aerosols to West Africa—but in more recent times, and considerable attention has been given to its role in the transport of fine-mode aerosol sources from fossil fuel burning to the lower Sahel (e.g., Eck et al. 2010; Onyeuwaoma et al. 2015; Fawole et al. 2016).

2.2 Data and Methods

The key aerosol parameters used in this study are the aerosol optical depth (AOD) which defines the monochromatic extinction of radiation within a column of atmospheric aerosol population and, therefore, provides a good indication of aerosol loading and the aerosol fine-mode fraction (FMF) defined as the proportion of fine-mode aerosols present in the entire aerosol volume. The optical depth of the coarse-mode aerosol (C-OD) was also employed as a complementary parameter of the fine-mode fraction, while the precipitable water vapor (PW) cm was used as indicator for moisture/season. The aerosol data were taken from the AErosol RObotic NETwork (AERONET) database. AERONET is operated by the National Aeronautics and Space Agency (NASA) ( as a ground-based remote-sensing program to assess aerosol optical properties and validate satellite data (Holben et al. 1998, 2001). Presently under the network are over 900 globally distributed CIMEL sun photometers providing temporal and spatial impute data for models. Although the level 2 AOD data, which are quality-assured (i.e., pre- and post-field calibrated and automatically cloud-screened and manually inspected—see Smirnov et al. 2000), were present as daily data for all the sites, the level 2 FMF data were not. To ensure uniform data sets for all locations considered, the level 1.5 AOD and FMF data from the provisions of the Spectral Deconvolution Algorithm (SDA) which are cloud-screened (O’Neill et al. 2003) were used. The number of aerosol data points employed in statistical determination of aerosol parameters differs from one site to the other (see Table 1). It is common knowledge that data ranges differ from one AERONET site to the other, and within sites in West Africa, several missing data points are common owing to instrument, internet, and power failure in addition to cloudiness as noted by Akoshile et al. (2016). The effects of such missing data points were removed by applying simple statistical techniques.
Table 1

Statistics of fine-mode fraction (FMF) and aerosol optical depth (AOD) for the AERONET sites

In the methodology adopted in the study, AOD (500 nm) and FMF (500 nm) of Ilorin are first compared to those of six upper Sahel AERONET sites. Figure 1 shows the location of Ilorin in relation with the upper Sahel sites at Djougou (1.59E; 9.76N), Ouagadougou (1.4W; 12.2N), DMN Maine Soroa (12.02E; 13.20N), IER Cinzana (5.93W; 13.27N), Banizoumbou (2.66E; 13.54N), and Dakar (16.95W; 14.39N). The sites are embedded in the regional rainfall climatology of the core monsoon months [June–July–August (JJA)]—as a way of introducing the apparent WAM connection to aerosol loading at Ilorin compared to the other sites. Banizoumbou, Ouagadougou, and Djougou (enclosed inside a blue triangle) are sites clearly located on a cluster that benefits maximally from advected dust and biomass burning aerosols from Bodele and Sudania axes. Considering that the FMF might not provide enough assessment of relative significance of fine- and coarse-mode aerosols for a given AOD value, the Angstrom exponent curvature [i.e., the derivative of the polynomial Angstrom exponent (αI)] was employed to provide additional assessment of dominant particle types in the various seasons. αI measures the wavelength dependence of the AOD using the analytical utility derived from spectral sensitivity of coarse and fine-mode aerosols, whereby its derivative (αI) is a simple method that provides reliable information on particle types (O’Neill et al. 2001, 2003; Schuster et al. 2006). It is particularly reliable for a bimodal size distribution associated with aerosol mixtures (Schuster et al. 2006) as one finds in the sub-Sahel area—with positive values of Angstrom derivative (αI > 0) often associated with significant fine-mode aerosols and large positive values (αI → 1) signifying dominance of fine-mode aerosols, while very low values (αI → zero) or negative values (αI < 0) indicate significant relative magnitude of coarse-mode aerosols and dominance over the fine mode, respectively (Kaskaoutis and Kambezidis 2006; Kaskaoutis et al. 2007; Soni et al. 2011). Aerosol types implicated in the seasonal fine-mode loading were hence classified by interpreting FMF frequency distributions of wet season and dry seasons in agreement with the Angstrom exponent derivative and using a classification scheme for fine-mode aerosol-type dominance as in Giles et al. (2012).
Fig. 1

Map of West Africa displaying the location of Ilorin AERONET site with six upper Sahel sites embedded in JJA rainfall climatology of West Africa. The cluster of sites with maximum fine-mode fraction, i.e., Djougou, Ouagadougou, and Banizoumbou (indicated in blue triangle) is located downwind the Bodele/Sudania dust and biomass burning aerosol transport trajectories (color figure online)

Since WAM is associated with dynamics of wind systems over the interior of West Africa, the hypothesis of WAM-driven aerosol loading at Ilorin is tested by correlating AOD (500 nm) with wind direction, i.e., wet-and dry-season zonal wind (ZW) and meridional wind (MW) speed anomalies for the various sites. The monthly means of wind speed anomalies (in ms−1) were obtained from ERA-Interim 6-hourly wind components of the global atmospheric reanalysis, produced by the European Centre for Medium-Range Weather Forecasts (ECMWF) of 0.5° × 0.5° grid resolution, which presences a finer atmospheric dynamics over the region. The configuration and performance of the data are described in Dee et al. (2011).

Attribution of the aerosol sources was supported using 7-day back trajectory analyses calculated for a total of 12 days—to cover the core wet-season months of June, July, and August and the core dry-season months of December, January, and February for 2 years (i.e., 2005 and 2006). The trajectory is based on the NOAA Air Resources Laboratory (ARL) HYSPLIT transport and dispersion model (see Stein et al. 2015; Rolph et al. 2017,

WAM influence on fine-mode aerosol loading was further examined by comparing rainfall data with AOD, coarse-mode AOD (C-OD), and FMF time series. Precipitation trends were assessed using a range of data which include: Nigerian Meteorological Agency (NIMET) real-time rainfall gauge data for the site (1971–2012) and the Global Precipitation Climatology Centre (GPCC) monthly data set from 1901 to present. The GPCC data are processed on a regular 0.5° × 0.5° grid resolution based on world-wide station records (Schneider et al. 2011). It is found to reproduce the precipitation scenario over West Africa (Salack et al. 2016). In the seasonal interpretation of data adopted, it is to be noted that West Africa unlike the temperate regions of the world has two major seasons: wet and dry seasons. The wet-season months are usually April, May, June, July, August, and September (AMJJAS) and dry-season months are usually October, November, December, January, February, and March (ONDJFM). In tying West African wet and dry periods to global seasonal dynamics, reference is often made to core wet months as June, July, August (JJA), and core dry months as December, January, February (DJF) corresponding to northern hemisphere summer and winter, respectively. September, October, November (SON) and March, April, May (MAM) constitutes “transition months” corresponding to northern hemisphere autumn and spring or the so-called “post-monsoon” and “pre-monsoon” periods, respectively. Due to possible differences in aerosol loading during these periods, we have utilized the 3-month season format in comparing aerosol data (e.g., FMF, PW, and AOD) for Ilorin in relation with the different sites. In the case where the interest was the precipitation increase in DJF at Ilorin which were not replicated in the other seasons at the site, the 3-season format was equally used. Due to common migration of wet and dry months into the “transition months” and in order not to mask important features which may be revealed by statistical analysis of longer data length, we employed the 6-month season format for correlations of AOD and wind directions as well as frequency analysis of FMF and other aerosol parameters.

3 Results and Discussion

3.1 Total and Fine-Mode Aerosol-Loading Statistics of Ilorin Compared to Upper Sahel AERONET Sites

Table 1 presents FMF (500 nm) and AOD (500 nm) data of Ilorin and those of six upper Sahel AERONET sites. The FMF means for Ilorin is enclosed in a green boundary together with those of Djougou, Ouagadougou, and Banizoumbou—considered alongside Ilorin as strong dust and/or biomass burning sites. From the data, the largest mean AOD occur at Djougou (mean = 0.611) and Ilorin (mean = 0.518), then Ouagadougou (mean = 0.466), Banizoumbou (mean = 0.43), IER-Cinzana (mean = 0.43), DMN Maine Soroa (mean 0.37), and Dakar (mean = 0.362). Very large AOD values (AODmax > 4) are present for DMN Soroa (5.38), Djougou (4.35), Ouagadougou (5.10), Banizombou (5.03), and Dakar (4.115). These sites are closer to the Sahara compared to Ilorin. Eck et al. (2010) have noted the highest FMF observations in the Sahel to be dominated by biomass burning aerosols. These are more common in locations downwind the Bodele dust/Sudania biomass burning axis. Hence, as previously mentioned, Djougou, Ouagadougou and Banizoumbou (enclosed in blue triangles) in addition to dust are usually associated with huge biomass burning episodes and have FMFmax = 1 in Fig. 1. Ilorin and IER-Cinzana tend to benefit minimally from rather spurious events responsible for maximum AOD and FMF values. IER-Cinzana, however, records lower FMFmean (0.365) compared to Ilorin which has the highest mean FMF of all the sites (FMFmean = 0.478). The fact of Ilorin having mean FMFmean higher than those of the biomass burning (BB) sites of Djougou, Ouagadougou, and Banizoumbou (FMFmean of 0.398, 0.375, and 0.373, respectively) is strong indication of fine-mode aerosol imputes from additional sources other than BB aerosols. Although from Table 1, the FMF average for Ilorin on account of its associated standard deviation (0.48 ± 0.18) is in the same range as those of Djougou (0.40 ± 0.19), Ouagadougou (0.38 ± 0.16), and Banizoumbou (0.37 ± 0.18), the table suggests that Ilorin may not have the same fine-mode aerosol-loading dynamics as these three sites. Ilorin from the table has the largest average FMF (0.47), but lower FMF range (0.08–0.99) compared to Djougou FMF range (0.06–1), Banizoumbou FMF range (0–1), and Ouagadougou FMF range (0–1). Again, FMF of 0 and 1 refer, respectively, to singly coarse- and fine-mode particles (associated with events such as dry-season dust outbreaks and biomass burning episodes, respectively) and these from the table are rare occurrences at Ilorin.

3.2 Seasonal FMF Trends of Ilorin Compared to Upper Sahel Sites

Since FMF contributed by BB aerosols is essentially pronounced during the dry season (i.e., DJF)—as dust, it is then apparent that the monsoon-seasonal wind may have outstanding role in the Ilorin FMF characteristics based on Fig. 2a. The figure shows JJA, MAM, DJF, and SON seasonal FMF and corresponding PW (cm) serving as indicator for season. The graph indicates JJA to be the key wet months at all the sites (see PW/JJA values), while the dry period occurs mainly in DJF (see PW/DJF). The wet-season FMF (FMF/JJA) is shown to be proportionally larger for Ilorin compared to other sites. Since FMF averages of Fig. 2a were evaluated for different data ranges, we plot in Fig. 2b the synchronized FMF time series of Ilorin and Banizoumbou (a dust and BB site with overlapping data range with Ilorin). It is evident from the figure that whereas FMF highs generally occur at Banizoumbou in DJF, FMF highs generally occur at Ilorin in JJA. This, therefore, supports the fact that the FMF levels at Ilorin are contributed substantially by monsoon dynamics compared to these other sites. In Fig. 2c, PW (cm), AOD (500 nm), FMF (500 nm), and Angstrom derivative (500 nm) are plotted for Ilorin. The figure shows FMF (JJA) to be highest (FMF > 0.5) compared to other seasons. The Angstrom derivative during JJA (αI ~ 0.5) is indicative of dominance by fine-mode particles. Although FMF is also significant in DJF (αI ~ 0.5), the small value (αI → 0) signifies dominance by coarse particles during the DJF season.
Fig. 2

a JJA, MAM, DJF, and SON seasonal FMF and corresponding seasonal PW(cm) serving as moisture/monsoon-season indicator. b Coherent FMF times series for Ilorin and Banizoumbou. c JJA, MAM, DJF, and SON seasonal averages of PW (cm), AOD, FMF, and Angstrom derivative (αI) for Ilorin

3.3 Aggregate Picture of Seasonal Fine-Mode Aerosol Loading at the Site

Following Fig. 2c, the following seasonal characteristics of fine-mode aerosol loading at Ilorin can be adduced:

SON: PW = 3.673; AOD = 0.384; FMF = 0.529; αI = 0.169: This is the season between the wet and dry periods, AOD is mainly from wet period fine mode as biomass transport from the Sudania region is not completely initiated at this period, and being a post-wet-season period wet vegetation does not support local biomass burning, locally lifted dust is, however, significant.

DJF: PW = 2.234; AOD = 0.954; FMFmean = 0.51; αI = 0.224: This is a very dry period when there is almost equal contributions from coarse-mode and fine-mode aerosols from advected desert dust and biomass burning aerosols from Sahara and Sudania regions especially during the harmattarn.

MAM: PW = 3.557; AOD = 0.752; FMF = 0.305; αI = 0.125: this is the season between the dry and wet periods and the only season, where FMF < 0.5 is encountered due to less intense bush burning after conclusion of planting season (bush burning is usually done in January/February) as well as seasonal weakness of southwesterly which transports fine-mode aerosol via polluted air from urban cities in the south. There is, therefore, outstanding coarse-mode dominance over the fine mode.

JJA: PW = 4.164; AOD = 0.352; FMF = 0.581; αI = 0.433: This is the core monsoon season; significant fine mode tends to dominate the coarse mode. This period is not a biomass burning period and considering that Ilorin is not highly industrialized, substantial FMF in this period is, therefore, transported to this site during JJA from non-biomass burning sources.

This seasonal picture especially for the monsoon season is corroborated and attributed in the sub-sections that follow.

3.4 Correlation Between Aerosol Optical Depth and Wind Direction

Wind has been known to influence atmospheric loading of aerosols in a manner that depends on a number of factors including wind direction, wind speed, season, and location (Huang et al. 2010; Manish et al. 2017). In Fig. 3, we evaluate this link, using values of correlations between AOD (500 nm) with zonal wind speeds (ZW) and meridional wind speeds (MW) anomalies for wet and dry seasons at all the sites. The time series data sets for Ilorin are shown in Fig. 4a, b. The wind fields are based on the European Centre for Medium-Range Weather Forecasts (ECMWF) reanalysis data of same temporal resolution as the AOD data. The following formalism has been adopted for ranking the correlations: r < 0 → anti-correlation; 0 < r < 0.09 → very low correlation; 0.1 < r < 0.29 → low correlation; 0.3 < r < 0.49 → moderate correlation; 0.5 < r < 0.69 → high correlation; 0.7 < r < 0.89 → very high correlation; 0.9 < r < 0.99 → perfect correlation. From the figure, all the AOD/wind field correlations are either low or very low except at IER-Cinzana, where AOD correlation with dry-season MW speed is moderate (r = 0.448) and at Ilorin, where wet-season correlation of AOD and MW is found to be high (r = 0.611). It is interesting that IER-Cinzana and Ilorin are two sites with the least AOD and FMF maxima which suggest less dust and biomass fire episodes at the two sites compared to the others.
Fig. 3

Correlation between AOD and Wind Direction-normalized zonal wind speed anomalies (ZW) and meridional wind speed anomalies (MW) for the sites arranged in order of increasing latitude. The most significant wet-season correlations are AOD/MW for Ilorin (see time series in Fig. 4)

Fig. 4

Time series of correlations of AOD with normalized zonal wind (ZW) and meridional wind (MW) speeds anomalies a ZW dry season, b ZW wet season, c MW dry season, and d MW wet season

3.5 Attribution of Seasonal FMF Features by Frequency Mode Classification

Figure 5a, b shows FMF frequency distributions for the wet and dry seasons, respectively, based on 1996–2016 data. The wet-season distribution is well fitted with a cumulative bimodal Gaussian shape. The two modes are: a high FMF mode (peak center = 0.685) of broad width (half-width = 0.521) and a low FMF mode (peak center = 0.338) of narrow width (half-width = 0.136). In the dry season on the other hand, FMF distribution is mono-modal with peak center = 0.484 and half-width = 0.394. The characteristics of these fitted distributions may be combined with appropriate FMF classification schemes to provide quantitative measures for attributing aerosol types implicated in the seasonal fine-mode loading.
Fig. 5

Frequency distributions of FMF with fitted Gaussian models: a wet-season months (i.e., AMJJAS) and b dry-season months (i.e., ONDJFM)

A wide range of aerosol optical and microphysical parameters may be deployed in association with the FMF to characterize and attribute seasonal fine-mode aerosol types. Notable amongst these are the Absorption Angstrom exponent (AAE) and the single scattering albedo (SSA) (see Bergstrom et al. 2007; Eck et al. 2010; Giles et al. 2012). The Angstrom exponent derivative \((\alpha^{\text{I}} )\), as introduced in Sect. 2.2, may be used particularly for the mixed aerosol types encountered at Ilorin. Moreover, the parameter has the advantage of easy computation from AOD values. For this work, \(\alpha^{\text{I}}\) was readily available for the site in good spread, density and quality. Analytically, we note that the spectral AOD \((\tau (\lambda ))\) is given by the Angstrom empirical relation:\(\tau (\lambda ) = \beta \lambda^{ - \alpha }\), where \(\alpha (\lambda i) = - \frac{{{\text{dIn}}\tau }}{{{\text{dIn}}\lambda }}\) is the Angstrom exponent and \(\beta\) is a constant. The Angstrom derivative is \(\alpha^{\text{I}} (\lambda i) = - \frac{{{\text{d}}\alpha }}{{{\text{dIn}}\lambda }}\). For multi-modal aerosol populations, there is departure from the linear form expressed by the \(\frac{{{\text{dIn}}\tau }}{{{\text{dIn}}\lambda }}\) relation, so that a polynomial of the form; \({\text{In}}\tau = \alpha_{2} ({\text{In}}\lambda ) ^{2} + \alpha_{1} ({\text{In}}\lambda ) + \alpha_{0}\) is more appropriate. The Angstrom derivative then becomes: \(\alpha^{\text{I}} (\lambda i) = \frac{{{\text{d}}\alpha }}{{{\text{dIn}}\lambda }} = - 2\alpha_{2}\), which provides indication of the particle regime from the sign of \(\alpha_{2}\) (see, e.g., O’Neill et al. 2001, 2003; Schuster et al. 2006; Kaskaoutis and Kambezidis, 2006; Kaskaoutis et al. 2007; Soni et al. 2011). Figure 6 shows a plot of daily values of \(\alpha^{\text{I}}\) versus FMF (1996–2016) with fitted polynomial.
Fig. 6

Correlation of daily FMF (500 nm) and Angstrom derivative (500 nm) from 1996 to 2016

Table 2 presents FMF ranges (adapted from Giles et al. 2012) and corresponding \(\alpha^{\text{I}}\) ranges (i.e., curvature trends) derived from the polynomial fits to Fig. 6:
Table 2

FMF classes defined based on Giles et al. (2012) and corresponding Angstrom derivative \((\alpha^{\text{I}} )\) ranges derived from polynomial fit of Fig. 5

FMF range (adapted from Giles et al. 2012)

Corresponding \(\alpha^{\text{I}}\) from fitted polynomial

0.66 < FMF ≤ 1.0

0.0 < FMF ≤ 0.33

0.33 < FMF ≤ 0.66

0.24 < \(\alpha^{\text{I}} \le\) 2.0

0.4 < \(\alpha^{\text{I}}\) ≤ − 0.31

− 0.31 < \(\alpha^{\text{I}}\) ≤ 0.24

Seasonal FMF range from fitted Gaussians of Fig. 6


Wet season (bimodal)

0.16 < FMF ≤ 1.20; 0.20 < FMF ≤ 0.47

Dry season (mono-modal)

0.09 < FMF ≤ 0.88

Lower panel: seasonal FMF ranges from Fig. 6

  1. 1.

    0.66 < FMF ≤ 1.0 → 0.24 < \(\alpha^{\text{I}} \le\) 2.0 corresponds to positive curvature (i.e., \(\alpha^{\text{I}}\) > 0). FMF is composed mainly of fine-mode aerosols (e.g., industrial pollution—shown in red boundary), and large spectral sensitivity of this FMF range indicates broad spectrum of fine-mode aerosols at the site.

  2. 2.

    0.0 < FMF ≤ 0.3 → 0.4 < \(\alpha^{\text{I}}\) ≤ − 0.31 corresponds to negative curvature (i.e., \(\alpha^{\text{I}}\) < 0). FMF is composed mainly of the coarse type (e.g., dust—shown in blue boundary), minimal spectral sensitivity indicate narrow and homogenous FMF mode.

  3. 3.

    0.33 < FMF ≤ 0.66 → − 0.31 < \(\alpha^{\text{I}}\) ≤ 0.24 corresponds to combination of positive and negative curvature, being at the turning point of the polynomial. FMF is composed of mixed (coarse–fine) particles (e.g., dust and biomass burning particles—shown in green boundary), spectral sensitivity is appreciable indicating broad spectrum of aerosols.


Using seasonal aerosol meteorology information for the site (see Sect. 3.3) and the FMF/\(\alpha^{\text{I}}\) classification above, it is deduced that the wet-season low and narrow FMF mode (0.20 < FMF ≤ 0.47) of Fig. 5 is representative of dust while the high and broad FMF mode (0.16 < FMF ≤ 1.20) is attributable to fine industrial aerosols. The large width of this mode also accommodates FMF in the coarse-dominated end of the spectrum characteristic of urban particulates. This is corroborated by Fawole et al. (2016) who computed SSA (675 nm) range of 0.88 < SSA < 0.96 for particles of FMF (500 nm) range of 0.23 < FMF < 0.59 and associated them to urban aerosols (also see Ezeh et al. 2014). The broad mono-modal shape of dry-season FMF distributions (0.09 < FMF ≤ 0.88), on the other hand, suggests a mixture of fine dust and biomass burning aerosols which usually occur during the same period.

3.6 Back Trajectory Analysis

For the purpose of tracking the sources of aerosols in the seasons at the site, we show in Fig. 7a, b, 7-day back trajectory sets for peak wet- and dry-season months: June, July, August and December, January, and March, respectively, chosen from 2005 to 2006 which were high AOD periods in the Ilorin data series. The days for the trajectory calculations were set at 15th day of each month which being roughly half of the month gives a good coverage for the entire seasons.
Fig. 7

a Back trajectories for Ilorin during core wet-season months (June, July and August) for 2005 and 2006. b Back trajectories for Ilorin during core dry-season months (December, January and February) for 2005 and 2006

3.6.1 June, July, and August

The SW trade wind prevails in Nigeria from April to October each year and it is the rain bearing wind. The result in Fig. 7a shows that the prevailing winds at Ilorin during these periods are found to originate mostly off-shore. Trajectories emanating from the coast at 100, 500, and 1000 m heights are traceable to the same source between latitudes 0°N and 10°S. These trajectories are constantly diffracted as they approach the equatorial line, forcing the trajectory to traverse some major coastal cities of West Africa such as Lome in Togo, Cotonou in Benin, Lagos, and Ibadan in Nigeria and Accra in Ghana. The implication of this is that urban and industrial emissions in these cities many of which are capital cities with hugely rising populations (see Table 3) would easily find their way to Ilorin. The trajectories also indicate that deposits at 1500 and 2000 m above ground level (AGL) have their sources from inlands, especially, the Sahara Desert and Bodele depression except in July and August 2005 (plates B and C), where the sources are traceable to the coast of Cameroon and Ghana, respectively. This implies that despite the dominance of the SW trade wind during the monsoon, a lot of dust particles are still deposited at Ilorin as a result of the intrusion of the NE trade wind at this of the year. This supports the wet-season FMF distributions of Fig. 6a.
Table 3

Major urban cities in the path of the calculated monsoon-season back trajectories en route to Ilorin and their urban population

(source: Demographia World Urban Areas, 13th Annual Edition:2017;

City location


Urban city population

Lagos 6°45′N; 3°38′E



Ibadan 7°23′N; 3°55′E



Accra 5°33′N; 0°12′W



Lome 6°75′N; 1°13′E



Cotonou 6°22′N; 2°26′E



3.6.2 December, January, and February

As shown in Fig. 7b, the North East (NE) from the Sahara and Bodele is the prevailing wind regime for much of the dry period. The trajectory lines show air masses originating mostly from heights below 1000 m AGL (with the exception of B and F, where it emanated at heights above 1000 m). A reversal of the wind regime is seen in plate C (December, 2005). Subsequently, by January and February 2006 as shown in plates D and E, the South West (SW) trade winds originating from the Atlantic became the prevailing wind regime with air masses originating at heights below 400 m. This implies that in addition to dust and biomass aerosols, there are possibilities of aerosols of different origins such as urban activities, ship effluents, sea salt, industrial effluents, etc. The wind regimes in C and D are rare occurrences in Nigeria at this time of the year. As we show in the succeeding section, the enhanced activity of the monsoon wind in the classical dry period as found above has the effect of resulting in multiple aerosol types especially of the fine mode.

3.7 Influence of Changing Rainfall Trends

A primary feature of WAM is the transport of both moisture and air into the continent. Association between Ilorin FMF with WAM as established in the above sections implies that changing monsoon systems as indicated in changing rainfall patterns would strongly impact both fine- and coarse-mode aerosol loading at the site. It is notable that the Sahel and sub-Sahel regions had witnessed rapid encroachment of the desert following the drought in the 1970s and the 80s leading to near total removal of vegetation and exacerbation of dust and biomass burning aerosol loading. In Fig. 8, we show the Nigerian Meteorological Agency (NIMET) time series of days with average rainfall (DAR) greater than 1, 5, 10, and 20 mm, first from 1971 to 1991 (Fig. 8a) and then from 1992 to 2012 (Fig. 8b). The slopes indicate that rainfall of all categories is growing in the second half of the data series (1992–2012) faster than in the first half (1971–1991) especially for the light rainfall category (DAR > 1 mm; slope − 0.24 → − 0.1, DAR > 5 mm; slope − 0.23 → − 0.14, DAR > 10 mm; slope − 0.36 → 0.08, DAR > 20 mm; slope − 0.02 → 0.12). Data from the Global Precipitation Climatology Centre (GPCC) monthly data set from 1994 to 2015 for the four seasons: DJF, MAM, JJA, and SON suggest that recovery of sub-Sahel rainfall is mainly on account of more rains being experienced during the classical dry period (i.e., DJF). We show trends of standardized rainfall anomalies based on the GPCC data in Fig. 9a–d. The figure shows that whereas there is negative rainfall trend during the wet season (JJA) (slope = − 0.00034), there is a positive linear trend with slope 0.012 in the dry period (DJF) which is the major dust and biomass aerosol-loading season.
Fig. 8

Inter-annual time series of days with Average Rainfall (DAR) greater than 1, 5, 10, and 20 mm at Ilorin Nigeria a 1971–1991, b 1992–2012

Source: Nigerian Meteorological Agency (NIMET) real-time rainfall gauge measurements

Fig. 9

Linear regressions of standardized rainfall anomalies of Ilorin for a: DJF, b MAM, c JJA, and d SON-showing increasingly more rains in the traditional dry period (DJF). The DJF positive rainfall anomaly around 2011–2016 corresponds to the sharp reduction in AOD, FMF, and C-OD during the same period (see Fig. 10)

We compare Fig. 9a with box-and-whisker plot trends of AOD, FMF, and C-OD evaluated in the mode of moving joint distributions from 1998 to 2016 (Fig. 10). The figures show that the DJF rainfall increase between 2010 and 2015 in Fig. 9 corresponds to the sharp decline in AOD, FMF and C-OD means and upper whisker values between 2012 and 2016 in Fig. 10. There is, however, enhancement in the FMF distribution width in the 2012–2016 periods compared to the preceding period (2010–2014). This indicates additional fine-mode sources. Evaluations of the moving joint box-and-whisker distributions in wet and dry seasons are shown in Fig. 11. The seasonal temporal trend shows that the FMF width during 2013–2016 period is broader compared to the earlier period of less rainfall (2010–2013) especially for the wet season which indicates additional fine-mode sources transported to the site from increased monsoon activity. The increasing rainfall in dry-season months shown in Fig. 9a is corroborated with data from the TRMM 3B43 product of the Tropical Rainfall Measuring Mission ( (see Fig. 12), showing higher rainfall rate at the site over the dry-season period of 2014/2015 compared to the dry-season period of 1998/1999. Although more rainfall is known to remove aerosols through wet deposition, the declining impact of intense previous years precipitation amounts on aerosol loading is more of a land cover reaction as reported in Wang et al. (2012) as it is a strong indicator that vegetation which improves in the succeeding year after increased rainfall might inhibit dust production (also see Cowie et al. 2014). Since drying of forests and their use as biomass for energy following the Sahel drought had added to the intensity of Sahel desertification (Cowie et al. 2014), given the increasing rainfall in the dry period, one may expect recovery of vegetation around the area. This is apparent from Fig. 13 which shows an Enhanced Infra-red Vegetative Index (EVI), from the MODIS-based land product: MOD13C2v5 (Huete et al. 1999) with spatial resolution of ~ 5600 m. The EVI maps show features of increased vegetation at the spot marked for Ilorin in 2013 map compared to the same spot in an earlier map (2001). This increase in land cover in response to rainfall increase in the dry period reduces the exposed surface for dust lifting (see Hui et al. 2008) and reduces burned biomass areas. This conclusion is without prejudice to notable adherence to government greening campaigns, which have seen reduction in local burning of vegetation in most parts of Nigeria’s sub-Sahel.
Fig. 10

Temporal trends of AOD, FMF, and C-OD, for Ilorin. The box-and-whisker plots were evaluated in the format of joint distributions histograms. Increase in FMF width between 2010–2014 and 2012–2016 is indicated

Fig. 11

Seasonal temporal trends of fine-mode fraction (FMF) and coarse-mode optical depth (C-OD) for Ilorin (1998–2016). Increases in FMF width between 2010–2013 and 2013–2016 for both seasons are indicated

Fig. 12

Comparison of average dry-season rainfall over a 15-year period evaluated from the TRMM 3B43 product of the NASA Tropical Rainfall Measuring Mission. a Average rainfall rate for November 1998 to March 1999; b average rainfall rate for November 2014–March 2015. (

Fig. 13

MODIS map of enhanced vegetative index (EVI) of Nigeria for 2001 and 2013, showing greening in the lower south west axis up to the sub-Sahel (Ilorin) and desertification in the lower Sahel region along the north-east axis over the 12-year period

4 Conclusions and Outlook

The large aerosol fine-mode fraction at Ilorin AERONET site compared to upper Sahel locations are found to be sufficiently contributed to by West African monsoon (WAM) activity especially when compared to the other notable dust and biomass burning sites. The positive correlation (r = 0.61 at 99% confidence) between wet-season AOD (500 nm) with corresponding meridional wind speed anomalies at Ilorin is unique to the site and underscores the outstanding relation of the WAM wind flow and aerosol depositions at Ilorin. Frequency distributions of seasonal FMF show a bimodal aerosol population in the wet season composed of a broad fine-size-dominated group and a narrow coarse-size-dominated group, while the dry-season FMF population has a single group of intermediate size aerosol mixture. Using Angstrom exponent derivative-based classification in agreement with FMF classification from the literature, the wet-season population is attributed to dust and aerosols both fine and coarse types from industries and urban activities, while the dry-season fine-mode population is found to be from a dust/biomass burning aerosol mixture. The above wet-season aerosol-type attribution is supported by a 7-day back trajectory calculation which confirms transport of particles by maritime wind traversing important coastal towns of West Africa in the wet season. Recovery in rainfall in recent years (as shown from Nigerian Meteorological Agency data) especially the growing wetness of the classical dry period (as indicated from the Global Precipitation Climatology Centre (GPCC) data) tends to explain the observed drop in mean AOD, C-OD and FMF in the last 5 years—consistent with reductions in local dust raising and biomass burning. Large FMF width observed during the same period suggests added fine-mode sources which might have been enhanced by increasing monsoon activity. With these results and with a good number of observational and model studies suggesting more rainfall across the entire Sahel (e.g., Fontaine et al. 2011; Dike et al. 2014; Dong and Sutton 2015; Raman et al. 2016; Park et al. 2016), a combination of reduction in dust and biomass burning aerosols and upsurge in industrial and urban aerosols from increased monsoon activity will be crucial to future model projections of aerosol loading in the area. Another outstanding subject of importance arising from this work is the possible implication which the observed long range transport of wet-season aerosols at high altitudes may have on aerosol layers over the region (see, e.g., Vernier et al. 2015; Tiwari et al. 2016). Analysis of aerosol vertical structure and characteristics especially from balloon and drone-borne in situ measurements in the sub-Sahel area and other West African locations are research propositions being explored in this regard.



The authors are grateful to the Principal Investigators of the Sahel AERONET sites used in the study for the data and for maintaining the stations. The authors gratefully acknowledge the NOAA Air Resources Laboratory (ARL) for the provision of the HYSPLIT transport and dispersion model and/or READY website ( used in this publication and other agencies whose data sets were employed. This study was undertaken during Okechukwu K. Nwofor’s stay at the International Center for Climate and Environment Sciences (ICCES), Institute of Atmospheric Physics (IAP), Chinese Academy of Science (CAS) in 2016 under the support of President’s International Fellowship Initiative (PIFI) from the Chinese Academy of Science. The support from the Chinese Academy of Sciences’ “The Belt and Road Initiatives” Program on International Cooperation (No. 134111KYSB20160010)is also appreciated. The authors thank the reviewers for their comments and suggestions which led to considerable improvement in the presentation of the ideas and results of this paper.

Compliance with Ethical Standards

Conflict of interest

There are no conflicts of interest declared.


  1. Akoshile CO, Ajibola TB, Falaiye OA et al (2016) A BSRN Station in a developing country: need, challenge, effort and progress. In: 14th BSRN Workshop, April 26–29, 2016, Canberra, AustraliaGoogle Scholar
  2. Andrade-Filho VS, Artaxo P, Hacon S et al (2013) Aerosols from biomass burning and respiratory diseases in children, Manaus, Northern Brazil. Rev Saúde Pública 47(2):1–8.
  3. Ben-Ami Y, Koren I, Rudich Y et al (2010) Transport of North African dust from the Bodele depression to the Amazon Basin: a case study. Atmos Chem Phys 10:7533–7544CrossRefGoogle Scholar
  4. Bergstrom RW, Pilewskie P, Russell PB, Redemann J, Bond TC, Quinn PK, Sierau B (2007) Spectral absorption properties of atmospheric aerosols. Atmos Chem Phys 7:5937–5943CrossRefGoogle Scholar
  5. Boucher O, Randall D, Artaxo P et al (2013) Clouds and aerosols. In: Stocker TF, Qin D, Plattner G-K, Tignor M, Allen SK, Doschung J, Nauels A, Xia Y, Bex V, Midgley PM (eds) Climate change 2013; the physical science basis. Contribution of working group I to the fifth assessment report of the intergovernmental panel on climate change. Cambridge University Press, Cambridge, pp 571–657. Google Scholar
  6. Cesnulyte V, Lindfors AV, Pitkänen MRA et al (2014) Comparing ECMWF AOD with AERONET observations at visible and UV wavelengths. Atmos Chem Phys 14:593–608CrossRefGoogle Scholar
  7. Charlson RJ, Schwartz SE, Hales JM et al (1992) Climate forcing by anthropogenic aerosols. Science 255(5043):423–430. CrossRefGoogle Scholar
  8. Chung CE, Chu J-E, Lee Y et al (2016) Global fine-mode aerosol radiative effect, as constrained by comprehensive observations. Atmos Chem Phys 16:8071–8080CrossRefGoogle Scholar
  9. Cowie SM, Knippert P, Marsham JH (2014) A climatology of dust emission events from northern Africa using long-term surface observations. Atmos Chem Phys 14:8579–8597CrossRefGoogle Scholar
  10. Dee DP, Uppala SM, Simmons AJ et al (2011) The ERA-Interim reanalysis: configuration and performance of the data assimilation system. Q J R Meteorol Soc 137:553–597CrossRefGoogle Scholar
  11. Demographia (2017) Demographia World Urban Areas, 13th Annual Edition: 04Google Scholar
  12. Dike VN, Shimizu MH, Diallo M, Lin Z, Nwofor OK, Chineke TC et al (2014) Modeling present and future African climate using CMIP5 scenarios in HadGEM2-ES. Int J Climatol. Google Scholar
  13. Dong B, Sutton R (2015) Dominant role of greenhouse-gas forcing in the recovery of Sahel rainfall. Nat Clim Change 5:757–760CrossRefGoogle Scholar
  14. Eck TF, Holben BN, Sinyuk A et al (2010) Climatological aspects of the optical properties of fine/coarse mode aerosol mixtures. J Geophys Res 115:D1920. CrossRefGoogle Scholar
  15. Ezeh GC, Obioh IB, Asubiojo OI et al (2014) Elemental compositions of PM10-2.5 and PM2.5 aerosols of a Nigerian urban city using ion beam analytical techniques. Nucl Instrum Methods Phys Res B 334:28–33CrossRefGoogle Scholar
  16. Fawole O, Cai X, Levine J et al (2016) Detection of gas flaring signature in the AERONET optical properties of aerosols at a tropical station in West Africa. J Geophys Res Atmos 121(24):14513–14524CrossRefGoogle Scholar
  17. Flaounas E, Kotroni V, Lagouvardos K et al (2016) Assessing atmospheric dust modeling performance of WRF-Chem: over the semi-arid and arid regions around the Mediterranean. Atmos Chem Phys. Google Scholar
  18. Fontaine B, Gaetani M, Ullman A (2011) Time evolution of observed July–September sea surface temperature-Sahel climate teleconnection with removed quasi-global effect (1900–2008). J Geophys Res 116:D041105. Google Scholar
  19. Giannini A, Saravanan R, Chang P (2003) Oceanic forcing of Sahel rainfall on interannual to interdecadal time scales. Science 302(5647):1027–1030. CrossRefGoogle Scholar
  20. Giles DM, Holben BN, Eck TF et al (2012) An analysis of AERONET aerosol absorption properties and classifications representative of aerosol source regions. J Geophys Res 117:D17203.,2012 CrossRefGoogle Scholar
  21. Holben BN, Eck TI, Slutsker I et al (1998) AERONET—a federated instrument network and data archive for aerosol characteristics. Remote Sens Environ 66:1–16CrossRefGoogle Scholar
  22. Holben BN, Tanre D, Smirnov A et al (2001) An emerging ground-based aerosol climatology: aerosol optical depth from AERONET 106. J Geophys Res 106(D11):12067–12097CrossRefGoogle Scholar
  23. Huang H, Thomas GE, Grainger RG (2010) Relationship between wind speed and aerosol optical depth over remote ocean. Atmos Chem Phys 10:5943–5950. CrossRefGoogle Scholar
  24. Huete A, Justice C, van Leeuwen W (1999) MODIS vegetation index (MOD 13) algorithm, theoretical basis, document version 3. University of Virginia Department of Environmental Sciences, Charlottesville, p 22903Google Scholar
  25. Hui WJ, Cook BI, Jose SR, Fuentes D, D’Odorico P (2008) Dust-rainfall feedbacks in the West African Sahel. Water Resour Res 44:W05202.,2008 CrossRefGoogle Scholar
  26. Husar RB (2005) The emergence of the bimodal distribution concept. In: Sem GJ, Boulaud D, Brimblecombe P, Ensor ES, Gentry JW, Marijnissen JCM, Preining O (eds) History & reviews of aerosol science. American Association for Aerosol Research, RestonGoogle Scholar
  27. Kaskaoutis DG, Kambezidis HD (2006) Investigation into the wavelength dependence of the aerosol optical depth in the Athens area. Q J R Meteorol Soc 132:2217–2234. CrossRefGoogle Scholar
  28. Kaskaoutis DG, Kambezidis HD, Hatzianastassiou N et al (2007) Aerosol climatology: dependence of the Angstrom exponent on wavelength over four AERONET sites. Atmos Chem Phys Discuss 7:7347–7397CrossRefGoogle Scholar
  29. Lee C-G, Yuan C-S, Chang J-C et al (2005) Effects of aerosol species on atmospheric visibility in Kaohsiung City, Taiwan. J Air Waste Manage Assoc 55:1031–1041CrossRefGoogle Scholar
  30. Lesins G, Lohmann U (2003) GCM aerosol radiative effects using geographically varying aerosol sizes deduced from AERONET measurements. J Atmos Sci 60:2747–2763CrossRefGoogle Scholar
  31. Liousse C, Guillaume B, Gregoire JM et al (2010) Updated African biomass burning emission inventories in the framework of the AMMA-IDAF program with an evaluation of combustion aerosols. Atmos Chem Phys 10:9631–9646CrossRefGoogle Scholar
  32. Manish K, Raju MP, Banerjee T (2017) Impact of drought and normal monsoon scenarios on aerosol induced radiative forcing and atmospheric heating in Varanasi over middle Indo-Gangetic Plain. J Aerosol Sci 113:95–107CrossRefGoogle Scholar
  33. Myhre G, Hoyle CR, Berglen TF et al (2008) Modeling of the solar radiative impact of biomass burning aerosols during the Dust and Biomass-burning Experiment (DABEX). J Geophys Res 113:D00c16. CrossRefGoogle Scholar
  34. Myhre G, Shindell D, Bréon F-M et al (2013) Anthropogenic and natural radiative forcing. In: Stocker TF, Qin D, Plattner G-K, Tignor M, Allen SK, Boschung J, Nauels A, Xia Y, Bex V, Midgley PM (eds) Climate change 2013: the physical science basis. Contribution of working group I to the fifth assessment report of the intergovernmental panel on climate change. Cambridge University Press, CambridgeGoogle Scholar
  35. Nicholson SE (2013) The West African Sahel: a review of recent studies on the rainfall regime and its interannual variability. ISRN Meteorol. (Article ID 453521) Google Scholar
  36. Nwofor OK, Chineke TC, Pinker RT (2007) Seasonal characteristics of spectral aerosol optical properties at a sub-Saharan site. Atmos Res 85:38–51CrossRefGoogle Scholar
  37. O’Neill NT, Dubovik O, Eck T (2001) Modified Angstrom Exponent for the characterization of sub-micrometer aerosols. Appl Opt 40:2368–2375CrossRefGoogle Scholar
  38. O’Neill NT, Eck TF, Smirnov A et al (2003) Spectral discrimination of coarse and fine mode optical depth. J Geophys Res 108(D17):4559–4573. CrossRefGoogle Scholar
  39. Onyeuwaoma ND, Nwofor OK, Chineke TC et al (2015) Implications of MODIS impression of aerosol loading over urban and rural settlements in Nigeria: possible links to energy consumption patterns in the country. Atmos Pollut Res 6:484–494CrossRefGoogle Scholar
  40. Pandithurai G, Pinker RT, Dubovik O, Holben BN, Aro TO (2001) Remote sensing of aerosol optical characteristics in sub-Sahel West Africa. J Geophys Res 106(D22):28347–28356CrossRefGoogle Scholar
  41. Park J, Bader J, Matei D (2016) Anthropogenic Mediterranean warming; essential driver for present and future Sahel rainfall. Nat Clim Change. Google Scholar
  42. Pinker RT, Idemudia G, Aro TO (1994) Characteristic aerosol optical depths during the Harmattan season on Sahara Africa. Geophys Res Lett 21(8):685–688CrossRefGoogle Scholar
  43. Pinker RT, Liu H, Osborne SR (2010) Radiative affects of aerosols in sub-Sahel Africa: dust and biomass burning. J Geophys Res 115:D15205. CrossRefGoogle Scholar
  44. Raman A, Arellano AF Jr, Sorooshian A (2016) Decreasing aerosol loading in the North American monsoon region. Atmosphere. Google Scholar
  45. Remer LA, Chin M, DeCola P et al (2009) Executive Summary. In: Chin M, Kahn RA, Schwartz SE (eds) Atmospheric aerosol properties and climate impacts, a report by the US Climate Change Science Program and the Subcommittee on Global Change Research. National Aeronautics and Space Administration, Washington, DCGoogle Scholar
  46. Roberts G, Wooster MJ, Lagoudakis E (2009) Annual and diurnal African biomass burning temporal dynamics. Biogeosciences 6:849–866CrossRefGoogle Scholar
  47. Rolph G, Stein A, Stunder B (2017) Real-time environmental applications and display sYstem: READY. environmental modeling and software. 95: 210–228.
  48. Salack S, Klein C, Giannini A et al (2016) Global warming induced hybrid rainy seasons in the Sahel. Environ Res Lett 11:104008. CrossRefGoogle Scholar
  49. Schneider U, Becker A, Finger P, Meyer-Christoffer A, Rudolf B, Ziese M (2011) GPCC full data reanalysis version 6.0 at 0.5°: monthly land-surface precipitation from rain-gauges built on GTS-based and historic data.
  50. Schuster GL, Dubovik O, Holben B (2006) Angstrom exponent and bimodal aerosol size distribution. J Geophys Res 111:D07207CrossRefGoogle Scholar
  51. Smirnov A, Holben BN, Eck TF et al (2000) Cloud screening and quality control algorithm for the AERONET database. Remote Sens Environ 73:334–337CrossRefGoogle Scholar
  52. Smith DM, Booth BB, Dunstone NJ et al (2016) Role of volcanic and anthropogenic aerosols in the recent global surface warming slowdown. Nat Clim Change 6:936–940. CrossRefGoogle Scholar
  53. Soni K, Singh S, Bano T et al (2011) Wavelength dependence of the aerosol Angstrom exponent and its implications over Delhi, India. Aerosol Sci Technol 45(12):1488–1498CrossRefGoogle Scholar
  54. Stein AF, Draxler RR, Rolph GD, Stunder BJB, Cohen MD, Ngan F (2015) NOAA’s HYSPLIT atmospheric transport and dispersion modeling system. Bull Am Meteorol Soc 96:2059–2077. CrossRefGoogle Scholar
  55. Sylla MB, Diallo I, Pal JS (2013) West African Monsoon in state-of-the-science regional climate models. In: Climate variability—regional and thematic patterns.
  56. Szopa S, Balkanski Y, Schulz M et al (2013) Aerosol and ozone changes as forcing for climate evolution between 1850 and 2100. Clim Dyn 40(9):2223–2250CrossRefGoogle Scholar
  57. Tie X, Wu D, Brasseur G (2009) Lung cancer mortality and exposure to atmospheric aerosol particles in Guangzhou. China Atmos Environ 43(14):2375–2377CrossRefGoogle Scholar
  58. Tiwari S, Mishra AK, Singh AK (2016) Aerosol climatology over the bay of Bengal and Arabian sea inferred from space-borne radiometers and lidar observations. Aerosol Air Qual Res 16:2855–2868CrossRefGoogle Scholar
  59. Vernier J-P, Fairlie TD, Natarajan M, Wienhld FG, Bian J, Martisnsson BG, Crumeyrolle S, Thomason LW, Bedka KM (2015) Increase in upper tropospheric and lower stratospheric aerosol levels and its potential connection with Asian monsoon. J Geophys Res Atmos 120:1608–1619. CrossRefGoogle Scholar
  60. Wang J, Zhao Q, Cui S et al (2012) Assessment of aerosol modes used in the MODIS ocean aerosol retrieval. J Atmos Sci 69:3595–3605CrossRefGoogle Scholar
  61. Zhang Y, Li Z, Qie L et al (2016) Retrieval of aerosol fine-mode fraction from intensity and polarization measurements by PARASOL over East Asia. Remote Sens 8(5):417. CrossRefGoogle Scholar
  62. Zuluaga MD, Webster PJ, Hoyos CD et al (2012) Variability of aerosols in the tropical Atlantic Ocean relative to African Easterly Waves and their relationship with atmospheric and oceanic environments. J Geophys Res Atmos 117:D16207. CrossRefGoogle Scholar

Copyright information

© Institute of Earth Environment, Chinese Academy Sciences 2018

Authors and Affiliations

  • Okechukwu K. Nwofor
    • 1
  • Victor N. Dike
    • 1
    • 2
    • 3
  • Zhaohui Lin
    • 2
  • Rachel T. Pinker
    • 4
  • Nnaemeka D. Onyeuwaoma
    • 5
  1. 1.Atmospheric Physics Group, Department of Physics and Industrial PhysicsImo State UniversityOwerriNigeria
  2. 2.International Center for Climate and Environment Sciences, Institute of Atmospheric PhysicsChinese Academy of SciencesBeijingChina
  3. 3.Energy, Climate and Environment Science GroupImo State Polytechnic UmuagwoOwerriNigeria
  4. 4.Department of Atmospheric and Oceanic ScienceUniversity of Maryland, College ParkMarylandUSA
  5. 5.Center for Basic Space Science (CBSS), National Space Research and Development Agency (NASRDA), Federal Ministry of Science and TechnologyUniversity of Nigeria, NsukkaNsukkaNigeria

Personalised recommendations