Introduction

The Indo-Gangetic Plain (IGP) encompasses 21% of the Indian subcontinent and hosts over 1/8th of the world’s population. The rapidly expanding economy in the IGP results in a wide range of emissions, including those from fossil fuel and biomass burning, industries, transport, mining, urbanization, and agricultural activities. Winter in this region is generally characterized by high aerosol optical depth (AOD), formation of persistent fogs, regional haze, and dew formation (Nair et al. 2007). Aerosol particles in the IGP region include those transporting in from overland and over ocean and are aged with continental as well as pristine ocean aerosol particles (Niranjan et al. 2006; Eck et al. 2001). The high aerosol loads make a significant contribution to the dew water chemical composition. Importance of the dew water chemistry has been reported in recent years (Lekouch et al. 2010; Yadav and Kumar 2014; Wentworth et al. 2016). Dew formation depends on the structure as well as chemistry of the substrate surface where condensation occurs (Beysens 1995). Contamination in condensed dew water is occurred by dry deposition, wet deposition (large fog droplet), and dissolution of gases from the atmosphere. Dew water collection is challenging, with many techniques for dew water collection having been reported, e.g., on wooden block (Duvdevani 1947), glass plate (Janssen and Römer 1991; Kidron 1998), nylon mesh (Zangvil 1996), aluminum sheet (Baier 1966), polyethylene sheet (Wagner et al. 1992), and Teflon sheet (Takeuchi et al. 2002).

Chemical composition of the dew water is strongly affected by the atmospheric condition of the sampling locations, e.g., temperature, humidity, dry deposition (Beysens et al. 2006; Takeuchi et al. 2003). Dew water is usually weakly acidic, having a pH close to 7, but can be more acidic (5.41–6.00) in urban locations. Concentrations of calcium and ammonium were found to be higher in previous studies (Table 1). Aerosol particles in the atmosphere act as condensation nuclei in the formation of dew and fog, making wet deposition a major pathway transporting pollutants from the atmosphere to the biosphere (Polkowska et al. 2008; Rubio et al. 2012; Lynam et al. 2015). Therefore, the chemical composition of dew water provides insight into local pollution and has been undertaken in most regions, including USA, Europe, Japan, China, etc. (Beysens et al. 2009; Polkowska et al. 2008; Xu et al. 2015). Unfortunately, there is very limited information on chemical composition of dew water in IGP region (Lakhani et al. 2012; Singh et al. 2006; Yadav and Kumar, 2014), even though winter time air pollution seriously impacts dew water chemical composition in IGP region. Severe air pollution situation in Bangladesh has already been reported (Salam et al. 2003a, 2008). For coastal Bhola, an IGP outflow location, chemical characterization of the aerosol particles was done, and relatively high concentrations of some chemical species were observed (Salam et al. 2003b). We decided to undertake a study of the dew water composition at this IGP outflow location in coastal Bhola, Bangladesh, and assess the impact of the aerosol particles on the dew water.

Table 1 Comparison of dew water composition collected at different locations of the world
Full size table

We collected dew water samples from December 2014 to January 2015 at this outflow location from the IGP into the Bay of Bengal. The objectives was to characterize the physicochemical properties of the dew water, to identify pollutant sources through apportionment studies, and to evaluate the influence of long range transported air pollutants on the dew water composition at this remote island location.

Experimental

Sampling location

Dew samples were collected at Bhola (22°10′01″ N, 90°45′00″ E), Bangladesh, a rural delta island in the Bay of Bengal (Fig. 1). This location rests in the outflow path for pollutants from the IGP region to the Bay of Bengal (Fig. 1), making it an important monitoring location. Bhola has an area of 1441 km2 and a population primarily employed in agriculture and fishing. There are no major industries in Bhola Island, and no traffic was observed within 1 km of the sampling location. Biomass burning for cooking is the primary source of local pollution.

Fig. 1
figure 1

Map of Bangladesh and Coastal Bhola (marked by red dots). Red circle indicates the sampling location (Source: Google map)

Full size image

Meteorology of coastal Bhola, Bangladesh

The climate in Bangladesh is characterized by high temperatures, excessive humidity, and distinctly marked seasonal variations in precipitation. Bangladesh can be divided into four seasons: pre-monsoon (March–May), monsoon (June–September), post-monsoon (October–November), and winter (December–February). The average meteorological conditions in coastal Bhola during different seasons have been reported by Salam et al. (2003b). The average temperature varies between 4 and 25 °C in the winter and 24 to 36 °C in the summer, with an average of 25.9 °C, and daily high temperatures generally observed in the afternoon. The highest relative humidity occurs in July (99%) and the lowest in December (36%), with light precipitation occurring in March and April, and average annual precipitation of 230 mm. Our dew water samples were collected during winter season (December, 2014–January, 2015) in Bhola, Bangladesh. The temperature ranged from 9 to 17 °C, no rain events were recorded, and air mass back trajectories (NOAA/ARL 2005) indicated transport was mostly from north to southeast during sampling period.

Sampling procedure and analysis

Dew water samples were collected nightly in a stainless steel bowl (Fig. 2) from 20 December 2014 to 30 January 2015 at the IGP outflow location in Bhola Island, Bangladesh. Our first attempts used a collector made of a poly(ethylene) sheet, but the collection efficiency was not satisfactory. Therefore, we modified our system to collect dew water using a stainless steel container and during our sampling period were able to collect ten dew water samples having adequate volume (>30.0 mL) to analyze all for parameters. On the remaining days, weather conditions did not support enough dew formation in our collector. The stainless steel container was cleaned daily with de-ionized water prior to nightly deployment, and field blanks were collected to minimize errors in analysis. The collection system was installed at the top of a 9.0-m tower, built in 2013 in collaboration with NASA AERONET and ONR Global, USA, for aerosol and precipitation studies. As there is generally no dew in daytime, our sampling period was 9 to 10 h nightly. Dew condensed on the inclined surface of the bowl and collected at the bottom, from which the samples were removed and stored daily around 6:00 a.m. to preclude evaporation due to sunlight. The collected dew water samples were filtered using Whatman 41 filter paper. Then, the samples were split, physical properties and ions were measured from one part (e.g., pH, conductivity, anions, cations, etc.), while the other part was acidified with ultra-pure nitric acid (1% V/V) and stored separately for trace metal analysis. Samples were stored in the refrigerator with sealed container to avoid any chemical transformation. The field blanks were collected using a clean stainless steel collector having the same size as that used for the real sample. The collector was filled with 50 mL de-ionized water, covered with air tight lid, and then kept beside the real sampler during the whole sampling time.

Fig. 2
figure 2

Stainless steel bowl used for dew water collection (area ≈0.800 cm−2) in Coastal Bhola, Bangladesh

Full size image

Field blanks were analyzed using the same procedure used for the dew, and results were excluded from the sample to avoid contamination error in the analysis. The pH and electrical conductivity were measured by using a pH meter (pH 211, Hanna Instruments) and conductivity meter (CM-5S, DKK-TOA Corporation). Concentration of water soluble cations (from filtered dew water) and trace metals (filtered dew water preserved with nitric acid) were determined by flame atomic absorption spectroscopy (model: AAnalyst 800, Perkin Elmer). Concentrations of water-soluble anions were determined by ion chromatography (model: 881 compact IC pro, Metrohm). Samples were filtered with a 0.45-μm Millipore membrane filter to exclude any remaining ultrafine particles before analysis of metals and ions. Concentration of HCO3 − was calculated from pH with Eq. (1) used by Kulshrestha et al. 2003:

$$ \left[{{\mathrm{HCO}}_3}^{-}\right]={10}^{-11.2+\mathrm{pH}} $$
(1)

Results and discussion

Physical properties

The dew water samples were analyzed for pH and electrical conductivity (EC). While slightly turbid before filtration, indicating presence of solid particulate material, the filtered dew water samples were transparent and odorless, indicating absence of volatile odorous species. The average pH of the dew water was 6.81 ± 0.3, nearly neutral. The acidity of dew water depends on dissolution of different gases (CO2, SOx, and NOx, etc.) and acidic aerosol (Beysens et al. 2006; Takeuchi et al. 2003), so the pH of the dew water is determined by the balance between acidic and neutralizing components, such as ammonium, bicarbonate, basic aerosol, etc. (Takeuchi et al. 2003). As a benchmark, rainwater pH is on the order of 5.6, owing to the formation of carbonic acid from CO2. The average EC of the collected dew water was 155 ± 62 μScm−1; corresponding to the total mineralization 119 ± 48 mg L−1 (≈0.77*EC). In Table 1, we compare our measured pH and EC to global dew water values. Our pH was comparable to that in Delhi and Agra (India) and rural Indiana (USA) but lower than in the other sites, except Bordeaux (France).

Chemical composition

The concentration of major ions in our dew water samples was measured and is presented in Table 1. The ionic concentrations of the measured cations and anions in our dew water samples reflect Ca2+ > Cl− > SO4 2− > Na+ > NO3 − > Mg2+ > K+ > HCO3 −. Concentration of measured total cations (1.3 × 103 μeq L−1) was 1.5 times higher than of total anions (0.84 × 103 μeq L−1). However, as we were unable to measure some ionic species, ion balance was not achieved in our data. From our measured ionic concentrations and the literature, in Table 1, we see that in Bhola, concentrations of NO3 − and K+ were higher than those in other locations, except for Negev, Israel. It is important to note here that most of the studies were conducted at urban sites. So, higher concentration of the anthropogenic species in dew water of Bhola is quite remarkable. In Bhola, K+ was 2–3 times higher than in Agra (India), Changchun (China), and Zadar (Croatia), while NO3 − was 1.5–2.5 times higher than in India (Delhi, Agra), Changchun (China), Yokohama (Japan), and Santiago (Chile) and much higher (5 to 27 times) than in all other locations. In Bhola, SO4 2− was 3–16 times higher than in Bordeaux (France), USA (Indiana, Arkansas), and Zadar (Croatia) but lower than in all other sites. Concentration of Na+ and Cl− were also higher in Bhola Island samples than in most other sites, except Zadar, Croatia (closer to the sea) and Negev, Israel (desert/semi desert region).

Trace metals, including Zn, Fe, Mn, Cu, Cr, Pb, and Ni were measured in the Bhola dew water samples. The concentration order (in μgL−1) of trace metals followed the sequence Zn (345.5 ± 154.5) > Fe(267.5 ± 320.1) > Mn(186.2 ± 96.5) > Cu(75.0 ± 21.4), with concentrations of Cr, Pb, and Ni below minimum detection limit. The instrumental detection limits for Cr, Pb, and Ni were 100.0, 200.0, and 60.0 μg L−1, respectively. In the aerosol samples previously collected at the coastal Bhola site (Salam et al. 2003b), Fe and Zn concentrations were also very high. Correlation among different chemical species in dew water is given in Table 2. Significant correlation was observed between soil tracers Ca2+ and Mg2+ (0.62) indicating their common crustal origin. Correlation between K+ and soil tracer Ca2+ (0.65) suggests that a significant fraction of the K+ is of crustal origin. Strong correlation was also found between Na+ and K+ (0.87), Na+ and Ca2+ (0.70), and Na+ and Mg2+(0.86), suggesting some fraction had marine origin. Cl− also showed significant correlation with Na+ (0.73) indicating probable sea salt.

Table 2 Correlation among different chemical species present in dew water
Full size table

Significant correlation between anionic species SO4 2− and NO3 − and their individual correlation (at p = 0.03) with biomass marker K+ indicates their anthropogenic sources. Correlation is also obtained between different anionic (SO4 2−, Cl−, NO3 −) and cationic species (K+, Ca2+, Mg2+) indicating their common sources or association of acidic and basic species after neutralization. Among trace metals, only Fe and Zn show strong correlation with each other. Mn shows strong correlation with SO4 2− and NO3 −. Among other trace metals, only Cu shows significant correlation with anionic species (Cl−). No significant correlation was obtained between trace metals and other metals (Na, K, Ca, Mg, etc.). Correlation analysis of trace metals suggests anthropogenic sources as they showed little correlation with soil and/or sea-derived species.

Fractional acidity and neutralization factor

Fractional acidity of the dew water was calculated using the following formula (Yadav and Kumar 2014):

$$ \mathrm{FA}=\left[{\mathrm{H}}^{+}\right]/\left[{{\mathrm{SO}}_4}^{2-}+{{\mathrm{NO}}_3}^{-}\right] $$
(2)

FA is 0 for maximum and 1 for no neutralization situation.

The average calculated value of FA in this study was 0.003 (highest 0.01, lowest 0.002), which indicates effective neutralization of acidic components (NO3 −, SO4 2−, etc.) in the dew water by basic components (NH4 +, Ca2+, Mg2+, and K+). The contribution of basic components to the neutralization process can be determined by calculating the neutralization factor (NF), via the equation (Yadav and Kumar 2014):

$$ {\mathrm{NF}}_{\mathrm{x}}=\left[ X\right]/\left[{{\mathrm{NO}}_3}^{-}+{{\mathrm{SO}}_4}^{2-}\right] $$
(3)

where X is the species whose neutralization factor is to be calculated.

The order by scale (larger to smaller) of neutralization factors found in dew water of Bhola Island is NFCa(1.42) > NFMg(0.42) > NF K (0.28). Dew forms near the ground, and thus collects more neutralizing particles (Ca2+, Mg2+) from soil than does water that forms at higher elevations (rain, cloud, etc.) (Yadav and Kumar 2014). Unfortunately, while NH4 + is another important neutralization species in dew water, its concentration was not measured in the current study.

Enrichment factor of major ions and trace metals

Enrichment factors were calculated to assist in identifying the origin of elements/ions in dew water (Kulshrestha et al. 1996). The calculation compares elemental ratios between ions/elements collected in the sample to reference material ratios. In coastal areas like Bhola, Na is commonly taken as a reference element for seawater since it is assumed to be primarily marine contribution in coastal regions (Kulshrestha et al. 2003). Al and Ca are typical lithophilic elements normally used as reference elements for continental crust (Cao et al. 2009). EFsea water and EFcrust of major cations and anions were calculated by the following equations, using Na and Ca as reference materials:

$$ {\mathrm{EF}}_{\mathrm{sea}\kern0.5em \mathrm{water}}={\left( X/\mathrm{Na}\right)}_{\mathrm{dew}}/{\left( X/\mathrm{Na}\right)}_{\mathrm{sea}} $$
(4)
$$ {\mathrm{EF}}_{\mathrm{crust}}={\left( X/\mathrm{Ca}\right)}_{\mathrm{dew}}/{\left( X/\mathrm{Ca}\right)}_{\mathrm{soil}} $$
(5)

where X is the concentration of the species of interest. (X/Na)sea is the ratio from seawater composition (Kidron and Starinsky 2012), and (X/Ca)crust is the ratio from crustal composition (Taylor 1964).

An EF value much less than or much higher than 1 is considered to be diluted or enriched respectively relative to the reference source (Safai et al. 2004). EFsea water and EFcrust values are listed in Fig. 3.

Fig. 3
figure 3

Enrichment factor of major ions and trace metals in dew water at Coastal Bhola, Bangladesh

Full size image

Cl− has EFsea water value of 1.02 but has an EFcrust 135.5, which clearly indicates its marine origin. And, although the Cl−/Na+ value of dew water (1.19) is slightly higher than of sea water (1.16), this merely suggests very small contributions from other sources of Cl−, such as K+. The average EFcrust and EFseawater value of K+ and Mg2+ suggested primarily crustal sources, with partial contribution from marine sources. The high EFcrust and EFseawater values of SO4 2− and NO3 − indicated that both were enriched in dew water compared to soil and seawater, suggesting that their primary sources was anthropogenic.

The EF value of trace metals with respect to the crustal material (e.g., Al, Fe) is often used to evaluate the degree of anthropogenic influence (Huang et al. 2013). Enrichment factors (EF) of trace metals were calculated in this study using Fe as reference material by the following equation (Tripathee et al. 2014):

$$ \mathrm{EF}={\left( X/\mathrm{Fe}\right)}_{\mathrm{dew}}/{\left( X/\mathrm{Fe}\right)}_{\mathrm{crust}} $$
(6)

where X is the concentration of the species of interest. The ratios of (X/Fe)crust are taken from Salam et al. (2003a).

Elements with an EF >10 are considered to be of mainly anthropogenic origin (Huang et al. 2013). Based on EF, the trace elements can be classified as non-enriched (EF = 1 to 10), moderately enriched (EF = 10 to 100), and highly enriched (EF > 100) (Al-Momani 2003). Enrichment factors of trace metals in dew water collected from Bhola (see Fig. 3) suggest that their source is mainly anthropogenic (EF > 10). Moreover, EF of Cu and Zn were greater than 100, indicating highly enriched conditions, suggesting severe contamination due to human activities (Duce et al. 1975).

Source contributions

Marine contribution

In order to determine marine contribution to the chemical species present in dew water, the sea salt fraction (SSF) and non-sea salt fraction (NSSF) have been calculated. Bhola is an island surrounded by the Bay of Bengal, so significant marine contribution is expected in dew water. The equations used for calculating SSF and NSSF are taken from (Kulshrestha et al. 1996)

$$ \%\mathrm{SSF}=100{\left( X/\mathrm{Na}\right)}_{\mathrm{sea}}/{\left( X/\mathrm{Na}\right)}_{\mathrm{dew}} $$
(7)
$$ \%\mathrm{NSSF}=100-\left(\%\mathrm{SSF}\right) $$
(8)

where X is the concentration of the species of interest, and (X/Na)sea is the ratio from seawater composition (Kidron and Starinsky 2012).

Na is used as reference, as this is a coastal site, and nearly all the sodium is of marine origin. From the value of %SSF, significant marine contributions are found for Cl− (97.47%), SO4 2− (11.57%), K+ (40.53%), and Mg2+ (5.23%) (see Fig. 4). The higher %NSSF value of SO4 2− (88.43%), K+(59.47%), Mg2+(94.77%), Ca2+(99.23%), and NO3 − (99.99%) suggest that these mostly came from non-marine sources.

Fig. 4
figure 4

Source contribution for different species present in the dew water at coastal Bhola, Bangladesh

Full size image

Crustal and anthropogenic contribution

In order to estimate the crustal contribution and anthropogenic contribution of different species present in dew water, crustal fraction (%CF) and anthropogenic fraction (%AF) were calculated using the following equations (Lu et al. 2011):

$$ \%\mathrm{CF}=100{\left( X/\mathrm{Ca}\right)}_{\mathrm{soil}}/{\left( X/\mathrm{Ca}\right)}_{\mathrm{dew}} $$
(9)
$$ \%\mathrm{AF}=\left(\left(100-\mathrm{SSF}\left(\%\right)-\mathrm{CF}\left(\%\right)\right)\right. $$
(10)

where X is the concentration of the species of interest. Here, Ca is used as reference element for continental crust, since it is a natural source for lithophilic elements and its composition in soil is stable (Cao et al. 2009).

Potassium occurs in the coarse particles of soil, but fine particles of potassium could result from biomass burning, industrial sources, or sea spray. Here, the primary source of NSSF K+ was considered crustal, as we were unable to distinguish between crustal sources and biomass/wood combustion in the present study (Cao et al. 2009; Yadav and Kumar 2014). The source contribution for different species is shown graphically in Fig. 4. The calculated value of %CF suggested small contribution of soil originated SO4 2−, Cl−, and NO3 − in dew water. The %NSSF of K+ and Mg2+ is considered as having pure crustal origin (Lu et al. 2011; Safai et al. 2004). Natural dust makes an important contribution to the ions present in dew water as the crustal sources of Ca2+, Mg2+, and K+ are dust aerosols, originating from soil, which are suspended in the lower layer of the atmosphere, and transported on our collection surface by wind/deposition (Xu et al. 2015). Ca2+ and Mg2+ could be derived from the dissolution of both primary and secondary minerals (CaCO3, CaCO3·MgCO3, and CaSO4·2H2O) commonly present as aerosol particles (Yadav and Rajamani 2006). From the values of (%AF) in Fig. 4, we see that while a small fraction of Cl− (1.8%) potentially comes from anthropogenic sources, SO4 2− and NO3 − have the highest anthropogenic contribution with %AF values of 83.5 and 99.3%, respectively.

Air mass trajectories analysis

Backward air mass trajectories were calculated for selected sampling days (for samples having the highest and lowest concentrations of SO4 2−, NO3 −, and trace metals) with the NOAA HYSPLIT model (hybrid single-particle lagrangian integrated trajectory model) (http://www.arl.noaa.gov/). The 48-h air mass trajectories were calculated at 500, 1500, 2500 meters AGL (above ground level). Figure 5a represents trajectories for the sample having the lowest concentrations of SO4 2−, NO3 −, and trace metals. That sample had 6.2 times lower concentration of (SO4 2− + NO3 −) than average and 5.8 times lower concentration of trace metals (sum of concentration of all trace metals) than average. From Fig. 5a, the calculated air masses at all levels were coming from the Bay of Bengal. On this sampling day, dew water was significantly influenced by clean air from over ocean, hence the observed very low concentrations of measured chemical species. The second case, having a lower than average concentration of (SO4 2−+NO3 −), is shown in Fig. 5b. Here, the air mass was coming from the IGP region to the sampling location carrying polluted air masses. The third case, Fig. 5c, represents sampling days when the concentration of (SO4 2− + NO3 −) was 1.25 times higher than average. Back trajectories for this case show that air masses came from the north (500 MAGL) and west directions (1500 and 2500 MAGL) on that sampling day. Figure 5d represents the fourth case: sampling days when concentration of (SO4 2− + NO3 −) was 1.63 times higher than average. Figure 5e represents our fifth case: the sampling day when the concentration of trace metals was highest in the dew water. The concentration of total trace metals in Fig. 5e was 2.02 times higher than the average total trace metal concentration. The air masses on that day came directly from north.

Fig. 5
figure 5

a Air mass trajectories when SO4 2−, NO3 −, and trace metals were lowest. b Air mass trajectories when SO4 2−+NO3 − was lower than the average value. c Air mass trajectories when SO4 2−+ NO3 − was higher than the average value. d Air mass trajectories when SO4 2− + NO3 − was highest. e Air mass trajectories when trace metals were highest

Full size image

Bhola is a remote rural Island of the Bay of Bengal and does not host much industrial activity.

The most probable major sources of the anthropogenic species in dew water in Bhola are air masses coming from different parts of IGP, as seen through the trajectory analysis. Although, biomass burning for cooking and agriculture activities could be responsible for some anthropogenic contributions; if it is the main source, then dew water samples should have very low concentrations of SO4 2−, NO3 −,and trace metals. Along the back trajectory of the air masses (Fig. 5), there are some big cities in Bangladesh and India. Moreover, winter time post-harvest agriculture burning could also have significantly impacted aerosol as well as dew water composition over the IGP (Rajput et al. 2011). High anthropogenic contributions of SO4 2− and NO3 − in dew water (the “Crustal and anthropogenic contribution” section) is an indicator for emissions of SOx and NOx into the atmosphere from different sources upwind in IGP, e.g., coal combustion, biomass burning, industrial emission, and vehicular exhaust (Yadav and Kumar 2014; Rubio et al. 2002). Dissolution of precursor gases or their oxidized products is likely responsible for the very higher concentrations of NO3 − as well as SO4 2− in dew water in coastal Bhola. The sources of trace metals in dew water are likely similar contamination from aerosol particles from upwind sources. It appears that dew water in coastal Bhola during winter time is contaminated when exposed to long-range transported air masses from polluted IGP regions.

Conclusion

Dew water collected in coastal Bhola, Bangladesh, during winter time was almost neutral with an average pH of 6.81. Concentrations of NO3 −, K+, Na+, and Cl− in dew water were relatively higher in Bhola. Further analyses for enrichment factor, percent source contribution, and correlation analyses revealed that the elemental composition of dew water at coastal Bhola had substantial contributions from anthropogenic sources. Air mass trajectories analyses showed that regional urban pollution has a significant influence on the dew water chemical composition in this remote, rural, IGP outflow location.