Start Submission Become a Reviewer

Reading: Atmospheric pathways of phosphorous to the Bay of Bengal: contribution from anthropogenic so...


A- A+
Alt. Display

Original Research Papers

Atmospheric pathways of phosphorous to the Bay of Bengal: contribution from anthropogenic sources and mineral dust


Bikkina Srinivas,

Physical Research Laboratory, IN
X close

M. M. Sarin

Physical Research Laboratory, IN
X close


The continental outflow from the Indo-Gangetic Plain and Southeast Asia, during the late NE monsoon (January–March), dominates the transport of chemical constituents to the marine atmospheric boundary layer (MABL) of the Bay of Bengal (BoB). During the rest of the year, prevailing wind regimes and meteorological conditions do not favour the atmospheric transport of continental products. Here we report on the spatio-temporal variability of inorganic phosphorous (PInorg= ) in the MABL and its dry-deposition flux to the surface BoB. On the basis of the abundance of PInorg in PM2.5 (0.1–0.8 nmol m−3) and PM10 (0.3–2.8 nmol m−3), we document its dominant occurrence in the coarse mode (Da≥2.5 µm). The analytical data also provide evidence for the chemical processing of mineral dust by acidic species and mobilisation of PInorg during the long-range atmospheric transport. However, significantly high PInorg/non-sea-salt Ca2+ ratios over the BoB suggest dominant contribution from anthropogenic sources (fertilisers and biomass burning emissions). PInorg concentration over the Arabian Sea is about 4 to 5 times lower and is primarily associated with the mineral dust from desert regions. The dry-deposition flux of PInorg to the BoB varies by one order of magnitude (0.5–5.0 µmol P m−2 d−1; Av: 0.02 Tg P yr−1). These results have implications to the air-sea deposition of phosphorous over oceanic regions downwind of the pollution sources and impact on the biogeochemistry of surface waters.

How to Cite: Srinivas, B. and Sarin, M.M., 2012. Atmospheric pathways of phosphorous to the Bay of Bengal: contribution from anthropogenic sources and mineral dust. Tellus B: Chemical and Physical Meteorology, 64(1), p.17174. DOI:
  Published on 01 Jan 2012
 Accepted on 20 Jun 2012            Submitted on 6 Jan 2012

1. Introduction

Phosphorous is one of the limiting nutrients for the primary productivity in both terrestrial and marine ecosystems (Chester, 2002; Fang, 2004). It has long been suggested that the marine primary productivity over the geological time scale was limited by the availability of phosphorous (Tyrrell, 1999). Recent observations also suggest that productivity in some of the oceanic regions (Mediterranean Sea and Sargasso Sea) and in marine basins, fed by large quantity of freshwater, is limited by the availability of phosphorous (Paytan and McLaughlin, 2007 and references therein). The role of phosphorus limitation in nitrogen fixation has also been emphasised for open ocean waters of the Atlantic and Pacific (Sanudo-Wilhelmy et al., 2001; Mills et al., 2004; Moutin et al., 2008). In this context, atmospheric mineral dust is considered as a major source of phosphorous to the ocean surface (Herut et al., 1999a; Ridame and Guieu, 2002; Krom et al., 2004; Anderson et al., 2010; Nenes et al., 2011).

However, the present-day increase in the supply of phosphorous from anthropogenic sources would lead to significant changes in the marine biogeochemical cycle of nutrients (Schlesinger, 1997; Mahowald et al., 2008). It has been argued that the global atmospheric cycle of phosphorous is significantly perturbed as a result of the extensive use of fertilisers and mining of phosphate rocks (Jahnke, 2000). The emissions from biomass burning has also been suggested to be a significant source of phosphorous (Mahowald et al., 2005, 2008; Anderson et al., 2010 and references therein). Recent studies have documented that the magnitude of atmospheric input of phosphorous may attain comparable levels with respect to its riverine supply (Compton et al., 2000; Paytan and McLaughlin, 2007). The uptake of anthropogenic species (non-sea-salt (nss)-, ) by Ca-rich mineral aerosols during the long-range transport has been documented in the literature (Wolf, 1984; Rastogi and Sarin, 2006; Saliba and Chamseddine, 2012). The chemical uptake of acidic species enhances the solubility of Ca-rich minerals such as carbonates and apatite (Ca3(PO4)2), and hence mineral dust is considered as a significant source of PInorg from ambient aerosols.

Therefore, atmospheric transport and air-sea deposition of phosphorous demand further investigation in the regional context. This article presents the first data set on the concentration of water-soluble inorganic phosphorus (PInorg) in ambient aerosols, its sources, size distribution and dry-deposition towards the Bay of Bengal (BoB).

2. Experimental methods

2.1. Study region and cruise tracks

The Bay of Bengal (BoB) and the Arabian Sea (ARS) are the two limbs of the northern Indian Ocean. The prevailing meteorological conditions during the late NE monsoon (January–March) favour the advective transport of pollutants from the Indo-Gangetic Plain and Southeast Asia to the BoB. In this context, earlier studies have highlighted the impact of continental outflow on the marine atmospheric boundary layer (MABL) of the northern Indian Ocean (Lelieveld et al., 2001; Ramanathan et al., 2001; Kumar et al., 2010).

Cruise tracks undertaken in the BoB, on board ORV Sagar Kanya, during March–April 2006 (Cruise: SK-223A) and January 2009 (SK-254) are shown in Fig. 1. During March–April 2006, prevailing winds over the BoB were initially north-easterly and changed to south-westerly in the later part of the cruise (a transition period), whereas the winds were predominantly north-easterly during the cruise in January 2009 (Fig. 1). A detailed description of meteorological conditions during all cruises has been reported in earlier publications (Alappattu et al., 2008; Kumar et al., 2010; Subrahamanyam et al., 2011). For comparison, chemical data on aerosol samples collected from the ARS (Fig. 1) have been used to characterise the contribution of inorganic phosphorous associated with the mineral dust from the desert regions. Although wet deposition is considered as a major source of nutrients to the ocean, wet precipitation events over ARS and BoB seldom occur during the late NE monsoon (January–March). We have, thus, addressed the dry-deposition fluxes of phosphorous to the surface waters of these two oceanic regions.

Fig. 1.   

Cruise tracks undertaken on board ORV Sagar Kanya: (a) Bay of Bengal (BoB; SK-223A: March–April 2006) and Arabian Sea (ARS: SK-223B: April–May 2006); (b) BoB (SK-254: January 2009); 7-d air mass back trajectories (AMBTs) are computed at an arrival height of 1000 m.

2.2. Sample collection and analysis

As a part of the Indian National Programme on ‘Integrated Campaign of Aerosols and Trace Gases Radiation Budget (ICARB)’, aerosol samples from MABL of the BoB and the ARS were collected using high-volume samplers (flow rate: 1.1 m3 min−1). During 19th March–12th April 2006, PM10 samples (n=23) were collected from the BoB. In a later cruise undertaken from 27th December 2008 to 28th January 2009, two size fractions of aerosols, PM2.5 (n=31) and PM10 (n=33), were collected over the bay region. In addition, PM10 samples (n=19) were collected from the ARS during 18th April–11th May 2006. All samples were collected using PALLFLEX®™ tissuquartz filters and were analysed for water-soluble ionic species (Na+, NH4+, K+, Mg2+, Ca2+, Cl and ), crustal elements (Al, Ca, Fe and Mg) and carbonaceous species (EC and OC). The relevant details and analytical procedures are described in earlier publications (Kumar et al., 2008, 2010). The nss components of , K+ and Ca2+ were assessed by correcting their measured concentrations for the contribution from sea salts (Keene et al., 1986), where nss-=−0.253×Na+; nss-K+=K+− 0.037×Na+; nss-Ca2+=Ca2+− 0.038×Na+, and 0.253, 0.037 and 0.038 are the ratios of , K+ and Ca2+ to Na+ in seawater, respectively.

For the analysis of PInorg () (this study), one quarter of the sample filter was extracted with Milli-Q water (specific resistivity ≤18.2 MΩ cm). The water extract was filtered through 0.45 µm (25 mm in diameter) Millipore filter and concentration of PInorg was measured using a 2-m-long liquid core waveguide capillary cell (LWCC, WPI Model) connected to USB-4000 spectrophotometer (Ocean Optics). Reagents and preparation of standard solutions for the determination of PInorg are similar to those described by (Li and Hansell, 2008). The absorption of the phosphate–molybdenum blue complex was measured at a wavelength of 710 nm (Zhang and Chi, 2002). The overall reproducibility in the measurement of PInorg is within 10%, assessed based on the repeat analysis of a commercial standard (Merck) and repeat checks of sample extracts. The detection limits, reported as three times the standard deviation of procedural blanks and normalised to average volume of air filtered (~1400 m3), for Na, Ca, Ca2+, K+, , and PInorg are 0.35, 0.38, 0.08, 0.17, 0.18, 0.31 and 0.013 nmol m−3, respectively.

We have estimated PDust by multiplying P/Al ratio of 0.008 [as the revised composition of Upper Continental Crust (UCC) (McLennan, 2001); P=700 ppm; Al=8.04%] with the Al concentration in aerosol samples (this study). In this calculation, it is implicit that Al is solely derived from the Continental Crust. A ratio of 0.012 for P/Al has been suggested for Saharan dust by Guieu et al. (2002). Similarly, Herut et al. (1999a) had suggested a ratio of about 0.011 for the Saharan aerosols. However, there are no data available for dust composition from the Thar Desert. We have used the P/Al ratio of UCC by Taylor and McLennan (1985), subsequently revised by McLennan (2001).

2.3. Statistical analyses

A summary of the chemical data (mean with ±1σ) obtained from cruises in the BoB (March–April 2006 and January 2009) and the ARS (April–May 2006) is presented in Tables 1 and 2. A comparison between the abundances of PInorg and chemical constituents is presented in terms of their statistical significance. Linear regression analysis of the measured chemical constituents is also performed to infer about the provenance of aerosol phosphorus. Furthermore, ANOVA analysis (including post-hoc tests) has been carried out to evaluate the seasonal variability in the molar ratios of nss-K+/nss-Ca2+, nss-/nss-Ca2+ and PInorg/nss-Ca2+ for the data (Table 2) obtained from all cruises (March–April 2006, April–May 2006 and January 2009).

3. Results and discussions

3.1. Air mass back trajectory (AMBT) analyses

The isentropic AMBTs were computed (for 7 d) using the HYSPLIT Model (Version-4; Draxler, 2002) by reanalysis of NCEP data set (GDAS, global data set from 2006 to present, Air mass back trajectories computed at an arrival height of 1000 m (and above) suggest that the origin of air masses during January 2009 cruise is mainly from the Indo-Gangetic Plain and Southeast Asia (Fig. 1). In contrast, air masses originate from the desert regions of Sahara, Arab, Oman, Somali and Thar, as observed during cruises undertaken from March to May 2006 (Fig. 1). The advective transport of mineral dust from the Middle East, Arabian Peninsula, Oman, Iran, Thar and Somalia to the MABL of ARS has been well recognised (Middleton, 1986; Tindale and Pease, 1999; Prospero et al., 2002). A detailed description on the transport pathways of aerosols to the ARS and BoB, during January to April, has been reported in earlier publications (Siefert et al., 1999; Dey et al., 2004; Kumar et al., 2008; Badarinath et al., 2010; Srinivas et al., 2011). However, a brief discussion is presented on the transport pathways of the mineral dust to these oceanic regions (BoB and ARS).

It has been reported that the high-altitude transport of mineral dust from the Middle East, Arabia, Iran and Thar Deserts to the ARS is a conspicuous feature during the summer months (March–May). However, at lower altitude, the Findlater jet counters the atmospheric transport from the desert areas to the ARS (Tindale and Pease, 1999). The periodicity and frequency of dust storms significantly increase during Spring and Summer months (Middleton, 1989). This observation is further corroborated by the total ozone mapping spectrometer absorbing aerosol index (TOMS AAI) data (based on the analysis of a 13-year data, from 1980 to 1992), suggesting that dust storm activity starts during March–May from the Arabian and Iranian Deserts and its magnitude peaks during June–July (Prospero et al., 2002). Likewise, the dust storm activity from the Thar region begins in the early Spring intermonsoon (March–April) and increases in strength during May–June as inferred from the intermonsoon TOMS Frequency of Occurrence and AAI (Prospero et al., 2002). Herman et al. (1997) have highlighted the application of satellite mapping of large dust events using TOMS of the NIMBUS-7 Satellite. A recent study has also emphasised the high-altitude transport (between 3 and 5 km) of dust storm originating from the Persian Gulf and ARS region and travel towards the Indian subcontinent (Badarinath et al., 2010). On the basis of LIDAR studies, an evidence for high-altitude dust transport (~3–3.5 km), during March–April 2006, has been reported by Raj et al. (2008). Likewise, high aerosol loading over the ARS observed during INDOEX field experiment was explained by invoking high-altitude transport of mineral dust from desert (viz., Arabia, Thar, Iran, Somalia) regions (Lelieveld et al., 2001; Ramanathan et al., 2001).

3.2. PInorg concentration over BoB

The mass concentration of water-soluble inorganic phosphorous (PInorg=) in PM10 averages about 1.1±0.3 nmol m−3 during March–April 2006 (Table 1). During sampling in January 2009, mean PInorg concentration centres on 0.3±0.2 nmol m−3 and 1.3±0.5 nmol m−3 in PM2.5 and PM10, respectively. A comparison of the seasonal data (Average with ±1σ) suggests that PInorg concentrations in PM10 are not significantly different (P-value>0.05) during March–April 2006 and January 2009 (Table 1). Despite this similarity in the abundance pattern, evidence for different sources of PInorg is discernible from the analytical data. The abundance of water-soluble calcium (Ca2+) and PInorg in two size fractions (PM2.5, PM10), collected during January 2009, suggests their dominant occurrence in the coarse mode. The mass ratio of PM2.5 to PM10 for Ca2+ and PInorg is computed for each sample and then averaged for all samples. The corresponding ratio centres on 0.31±0.18 and 0.25±0.14 for Ca2+ and PInorg, respectively, indicating their dominant contribution from mineral dust.

The contribution of PInorg from mineral dust has been investigated based on the chemical data presented in Tables 1 and 2. The mass concentration of (with nss-/≥90%) accounts for ~67 and 60% of ΣWSIC in March–April 2006 and January 2009, respectively (where WSIC refers to water-soluble inorganic constituents and taken as sum of concentrations of all cations and anions). During these two sampling campaigns in the BoB, the equivalent ratio of and is far less than one (Table 2), suggesting chemical uptake of excess acid by mineral dust. This is also evident from the significant (P-value<0.05) linear correlation between nss-Ca2+ and nss- (Fig. 2a; regression statistics are presented in Table 3) and among aerosol Ca and nss-Ca2+ (Fig. 2b; see Table 3). The dominant contribution of nss- in the MABL of BoB is clearly evident from the steep slope of its linear regression lines with nss-Ca2+ for data from January 2009 (m=23; R2=0.75; P<0.05) and March–April 2006 (m=2.3; R2=0.62; P<0.05; Fig. 2a; see Table 3). This is in sharp contrast to the lower slope of linear regression for data from the ARS (m=0.83; R2=0.67; Fig. 2a; see Table 3). It is noteworthy that the solubility of aerosol Ca (defined as nss-Ca2+/Ca) in March–April 2006 is relatively high compared with that in January 2009 (Tables 2 and 3). Another notable observation relates to significantly high (P value<0.05; t statistic=5.89) Ca2+ concentration (range: 4–36 nmol m−3; Av: 17±9 nmol m−3) during March–April 2006 compared to January 2009 (range: 0.4–13 nmol m−3; Av: 6±3 nmol m−3). Furthermore, on average, mineral dust concentration (assessed based on measured Al concentration and assuming Al content of 8.0% in the mineral dust) accounts for ~33 and 50% of the aerosol mass in January 2009 and March–April 2006, respectively. The relatively high solubility (Av: 92±5%) of aerosol Ca together with high abundance of mineral dust (Av: 49% of particulate mass) during March–April 2006 is attributed to differences in the source regions of mineral dust and chemical processing during the long-range transport.

Fig. 2.   

The scatter plots between (a) nss-Ca2+ and nss- and (b) aerosol Ca and nss-Ca2+ suggest a reactive uptake of acidic species (H2SO4) by the mineral dust during long-range atmospheric transport. The dominant contribution of nss- in the MABL of Bay of Bengal (BoB) is clearly evident from the steep slope of the linear regression line compared to that in the Arabian Sea (ARS). Data for BoB (March–April 2006) and ARS (April–May 2006) are adopted from Kumar et al. (2008).

On the basis of the above discussion, we argue in favour of the seasonal trend and significant role of chemical processing of mineral dust during March–April 2006 compared to samples collected during January 2009. These differences in the chemical composition of aerosols are attributed to the varying type of mineral dust over the BoB during the two seasons. The AMBTs, during March–April 2006, were from desert regions (Thar Desert) of western India, whereas air masses during January 2009 over the BoB show dominant influence from the Indo-Gangetic Plain. The elemental ratio of Ca/Al in aerosols associated with these air masses averages around 0.81±0.16 for March–April 2006 (Kumar et al., 2008) and 0.35±0.14 for January 2009 (Kumar et al., 2010). The relatively high ratio of Ca/Al is consistent with that reported over the ARS, attributable to the source of dust from the desert region in western India (Yadav and Rajamani, 2004; Rastogi and Sarin, 2009). In contrast, the lower ratio of Ca/Al over BoB (during January 2009) is largely dictated by the outflow from the Indo-Gangetic Plain (Ca/Al=0.23±0.11, as measured at downwind site in the Gangetic Plain) (unpublished data; manuscript under preparation). A significant correlation between nss-Ca2+ and PInorg (R2=0.64; n=23; P-value<0.001; Fig. 3a) for the data from March to April 2006 suggests their common source from the mineral dust. The contribution of dust-derived phosphorous has been traced by the linear relation (P<0.05) between nss-Ca2+ and PInorg (Herut et al., 1999a; Markaki et al., 2003). These studies have shown that the air masses originating from the Arabian Peninsula exhibit a linear relation among nss-Ca2+ and total inorganic phosphorous (Markaki et al., 2003). Higher abundance of P and Ca in the desert dust (from Thar Desert) has been documented by Yadav and Rajamani (2004). These desert soils are reported to be rich in Ca which is reflected from the characteristic high elemental ratio (0.73±0.10) of Ca/Al (Yadav and Rajamani, 2004; Rastogi and Sarin, 2006; Kumar and Sarin, 2009). The reactive uptake of acidic species can lead to enhancement in the solubility of Ca-rich minerals such as carbonates, apatite (Ca3(PO4)2), and is represented by following equations:

Fig. 3.   

Scatter plots: (a) data plotted with symbol () provide evidence for the acid processing of mineral dust and, hence, enhanced supply of water-soluble PInorg during long-range transport to the Bay of Bengal (BoB). Evidence for supply of PInorg from anthropogenic sources is shown with the dotted line plotted through symbol (), with steep slope and high PInorg/nss-Ca2+ ratio. For comparison, data from the Arabian Sea (ARS) (), with high abundance of dust, represent exceptionally lower PInorg/nss-Ca2+ ratios. The scatter plot shown in (b) also exhibits steep slope, with higher abundance of anthropogenic nss-K+ (relative to nss-Ca2+) during January 2009 compared to March–April 2006 (BoB) and April–May 2006 (ARS).

Therefore, the observed linear relation between nss-Ca2+ and PInorg during March–April 2006 (Fig. 3a) over the BoB is attributed to their common source (as evident from the AMBTs originating from the Thar Desert; Fig. 1). In contrast, relatively high concentrations of PInorg associated with extremely low concentrations of Ca2+ during January 2009 (Fig. 3a) suggest the dominance of anthropogenic sources (emissions from biomass burning and use of fertilisers).

3.3. PInorg over ARS

The mean (±1σ) concentration of PInorg in aerosol samples collected from the ARS during April–May 2006 (Cruise SK-223B) corresponds to 0.3±0.1 nmol m−3.

A comparison of PInorg abundances shows significant difference (one-way ANOVA; F(2,72)=39.9; P value<0.05) between the cruises (March–May 2006 and January 2009). These analyses suggest that concentration of PInorg over the BoB is significantly high compared to that over the ARS (April–May 2006). However, concentrations were not significantly different between January 2009 and March–April 2006 (Table 1). Relatively high concentration of PInorg observed over the BoB is due to the predominance of anthropogenic sources and acid processing of mineral dust (Fig. 3a). A significant difference in the PInorg/nss-Ca2+ ratio in aerosols over the ARS and BoB is noteworthy (Table 2). The PInorg/nss-Ca2+ ratio varied from 0.01 to 0.11 (Av: 0.03±0.02) over the ARS during April–May 2006. In contrast, PInorg/nss-Ca2+ ratio averages around 0.09±0.05 (range: 0.04–0.23) over the BoB during March–April 2006, and the ratio is one order of magnitude higher (Av: 0.54±0.54, range: 0.12–2.4) during January 2009 (Table 2). Relatively high PInorg/nss-Ca2+ ratio over the BoB during January 2009 is attributed to enhanced supply of PInorg from fertilisers and biomass burning emissions with relatively low Ca2+ content. The reactive uptake of acidic species (nss-) by mineral dust is also evident from the mass ratio of water-soluble Ca2+ to total aerosol Ca (Fig. 2b) varying from 0.63 to 1.05 (Av: 0.95±0.14), with average nss-Ca2+/Ca2+ ratio centring on 0.89±0.12 (Table 2). However, abundance of nss- in the MABL of ARS is significantly lower than that over the BoB (Fig. 2a; one-way anova, F(2, 70)=16.8; P value<0.05).

3.4. Anthropogenic source of PInorg vis-à-vis mineral dust

A characteristic high abundances of anthropogenic constituents (nss-K+ and nss-) within the MABL of BoB, during January 2009, are evident based on relatively high ratios of nss-/nss-Ca2+ (Fig. 2a) and nss-K+/nss-Ca2+ (Fig. 3b). These characteristic ratios in January 2009 over the BoB are relatively high compared to those observed during March–April 2006 and over the ARS (April–May 2006). The difference in the slopes of the regression lines (Figs. 2a and 3b) reflects the seasonal variability in the chemical composition of aerosols associated with the outflow from the Indo-Gangetic Plain (from December to February) and from the desert regions during March–May. Also, a statistical test has been performed in order to ascertain differences in the ratios of nss-K+/nss-Ca2+, nss-/nss-Ca2+ and PInorg/nss-Ca2+ for the data from January 2009, March–April 2006 (BoB); April–May 2006 (ARS). ANOVA analyses (with post hoc test) suggest that PInorg concentration and ionic ratios (nss-K+/nss-Ca2+, nss-/nss-Ca2+ and PInorg/nss-Ca2+) are significantly different (P<0.05) for seasonal data from all cruises (January 2009, March–April 2006 and April–May 2006). A significant difference is evident between PInorg/nss-Ca2+ for winter (January 2009) and spring intermonsoon (March–April 2006, April–May 2006) cruises [F(2, 71)=15.782, P-value<0.05]. Likewise, significant difference (P-value<0.05) is observed for nss-K+/nss-Ca2+ [F(2,71)=11.56; P<0.05] and nss-/nss-Ca2+ [F(2, 71)=50.6; P<0.05] (Table 2, data from January 2009, March–April 2006 and April–May 2006).

The relatively high concentrations of anthropogenic constituents (nss-& nss-K+) together with the low abundance of nss-Ca2+ within the MABL of BoB, during the wintertime (January–March), is a characteristic feature of the outflow from the Indo-Gangetic Plain (Kumar et al., 2010; Sarin et al., 2011; Srinivas et al., 2011). The high ratios of nss-/nss-Ca2+, nss-K+/nss-Ca2+ and PInorg/nss-Ca2+ during January 2009 (Table 2) are attributed to the dominance of pollution sources. In contrast, the relative decrease in the source strength of the continental outflow during the spring intermonsoon (March–April), together with higher abundance of water-soluble Ca2+, is a dominant cause for the low ratios of nss-/nss-Ca2+, nss-K+/nss-Ca2+ and PInorg/nss-Ca2+ (Figs. 2a, Figs. 3a, Figs. 3b). The high concentration of Ca2+ mainly results from seasonal shift in the source regions of mineral dust during March–April 2006, as evident from the AMBTs which originates from the Thar Desert. The enhanced solubility of aerosol Ca and the linear relation of PInorg with nss-Ca2+ during March–April 2006 suggest that significant fraction of water-soluble inorganic P is derived from the chemical processing of mineral dust.

The mass concentration of PInorg in the MABL of BoB is significantly higher than that derived from the mineral dust as its only source. A significant linear relationship of PInorg and nss-K+ (P-value<0.05) suggests fertilisers and/or emissions from biomass burning as their common anthropogenic source. A recent study by Singh and Singh (2008) has shown significant increase in the use of fertilisers (rich in N, P and K) in the Indo-Gangetic Plain during the past few decades. Therefore, atmospheric transport of dust from disturbed soils and agricultural fields could serve as a dominant source of PInorg. Furthermore, emissions from biomass burning (burning of agricultural crop residue) has been suggested as a significant source of aerosol P, similar to aerosol K+ (Mahowald et al., 2005, 2008 and references therein; Izquierdo et al., 2012). Thus, a linear correlation between nss-K+ and PInorg is attributed to both the emissions from biomass burning and fertilisers. The PInorg/EC ratio in the fine mode aerosols over the BoB, during January 2009, averages about 0.003± 0.002 (also evident from the slope of regression line between EC and PInorg; m=0.001; R2=0.54; figure not shown in this article). A similar ratio of PInorg to BC (0.0029± 0.0001) has been reported for the emissions from biomass burning over the Amazonian region (Mahowald et al., 2008). We, therefore, invoke significant contribution of K+ and PInorg from fertilisers associated with the large-scale agricultural activity and biomass burning (burning crop residue) in the Indo-Gangetic Plain.

The phosphorous contribution from dust (PDust) is subtracted from the total PInorg in order to obtain the anthropogenic component (PAnth). In Fig. 3a, three distinct relationships among PInorg and nss-Ca2+ are identified. The line with steep slope (high PInorg and low nss-Ca2+) is dominated by the anthropogenic phosphorous (PAnth). The line with flat slope (Fig. 3a; low PInorg and high nss-Ca2+) for the ARS is dominated mainly by PInorg derived from the mineral dust. The third relationship with an intermediate slope for the data from the BoB (during March–April 2006) is dictated by varying contribution from dust and anthropogenic sources.

The mass concentrations of PInorg, PDust and PAnth and the corresponding ratios of PDust/PInorg, PAnth/PInorg and PInorg/nss-Ca2+ within the MABL of BoB and ARS are summarised in Table 2. Over the BoB, PAnth contributes significantly, varying from 61 to 94% (Av: 77±10%) during March–April 2006 and from 24 to 96% (Av. 76±18%) during January 2009. It is implicit in this approach that the mobilisation of PInorg by acid processing of mineral dust is also a part of PAnth. In contrast, PInorg over the ARS is mainly dominated by desert dust (PDust=~70%). The spatio-temporal variability of percentage contribution of Panth/PInorg to the study region is depicted in Fig. 4. From this figure, it is inferred that the contribution of PInorg from anthropogenic sources (viz., emissions from biomass burning and fertilisers, etc.) dominates over the BoB during the continental outflow.

Fig. 4.   

Spatio-temporal variability in the concentration of anthropogenic water-soluble inorganic phosphorous (Panth) to PInorg over the Bay of Bengal and the Arabian Sea.

3.5. Atmospheric dry-deposition of PInorg

The dry-deposition flux is estimated by using the following equation.

where Vdry and Cmeas refer to dry-deposition velocity and measured mass concentration, respectively. For the flux computation, we have used a deposition velocity of 2.0 cm s−1. A similar dry-deposition velocity has been used by several studies reported in the literature to estimate deposition fluxes (Duce et al., 1991; Herut et al., 1999b; Markaki et al., 2003; Baker et al., 2006; Chen et al., 2006, 2007). On the basis of experimental data and modelled results, deposition velocity of 0.1 and 2.0 cm s−1 has been suggested for fine and coarse mode aerosols by Spokes et al. (2000). Due to the dominant occurrence of PInorg in PM10 (as evident from the mass ratio of PM2.5/PM10, Av: 0.25), our experimental data suggest an association of PInorg in the coarse mode. Therefore, Vd of 2.0 cm s−1 is used for estimating the deposition flux in this study.

The dry-deposition flux of PInorg to the BoB ranges from 1.1 to 3.6 (Av:1.9±0.6) µmol P m−2 d−1 during March–April 2006 and from 0.5 to 4.8 (Av: 2.3±0.9) µmol P m−2 d−1 during January 2009 (assuming dry-deposition velocity of PInorg as 2.0 cm sec−1), whereas to the ARS it varies from 0.3 to 0.9 (Av: 0.6±0.2) µmol P m−2 d−1 during April–May 2006. These fluxes have been compared with the model-based estimates (Mahowald et al., 2008; Okin et al., 2011). The atmospheric deposition of PInorg to the BoB (this study) is ~0.022 Tg P yr−1 (0.024 for January 2009 and 0.019 Tg P yr−1 for March–April 2006) and to the ARS is ~0.014 Tg P yr−1 for April–May 2006 (areas of BoB and ARS are taken as 2.2×1012 and 4.93×1012 m2, respectively, and dry-deposition is integrated only for 150 d when continental outflow is dominant). Thus, the total dry-deposition flux of PInorg (~0.036 Tg P yr−1) over these two oceanic regions can be considered as representative of the annual deposition of PInorg to the northern Indian Ocean (NIO). A comparison of dry-deposition of PInorg to the northern Indian Ocean (this study) with the model-based estimates (reported in the literature) is presented in Table 4. We have also compared the PInorg flux to NIO with that from other oceanic regions. These estimates are of comparable magnitude with the model-based results for the NIO (0.04 Tg P yr−1 by Okin et al. (2011) and 0.014 Tg P yr−1 by Mahowald et al. (2008)).

3.6. Comparison with riverine supply

The global riverine input of PInorg, via major rivers, is 0.8–1.4 Tg P yr−1 (Compton et al., 2000) and the freshwater influx from the global rivers is about 37.4×1012 m3 yr−1 (Martin and Whitfield, 1983; Kumar et al., 1996). It has been reported that Bay receives approximately 1.625×1012 m3 yr−1 of freshwater (Sarin et al., 1990; Prasanna Kumar et al., 2004). Thus, the riverine input of PInorg to the BoB is estimated to be 35–61 Gg P yr−1. Since PInorg abundance is relatively high during January 2009 compared to that in March–April 2006 (Spring intermonsoon), it is relevant to compare the atmospheric input with the riverine supply in order to estimate the maximum contribution via air-sea deposition to the BoB. The concentration of PInorg in MABL of the BoB, during January 2009, ranged from 0.32 to 2.76 nmol m−3. Therefore, dry deposition of PInorg to the BoB ranges from 0.5 to 4.8 µmol P m−2 d−1 (5–50 Gg P yr−1). A comparison of these estimates suggests that the atmospheric deposition of PInorg is of comparable magnitude to its supply via rivers, and that the atmosphere is a significant source of PInorg to surface waters of the BoB.

4. Conclusion and implications

The spatio-temporal variability in the concentration of water-soluble inorganic phosphorous (PInorg) has been studied over the BoB and the ARS during the continental outflow (January–April). Our study provides the first field evidence for the acid processing of mineral dust during atmospheric transport and, hence, enhanced solubility of PInorg. We suggest that high concentrations of sulphate persisting in the continental outflow from the Indo-Gangetic Plain provide most conducive environment for acid mobilisation of phosphorous (and other nutrients) from dust-laden air masses. Nevertheless, contribution from anthropogenic sources (fertilisers and emissions from biomass burning) is overwhelming and account for nearly 75% of PInorg. In contrast, PInorg concentrations are 4 to 5 times lower over the ARS and anthropogenic fraction contributes no more than ~30% of PInorg. The atmospheric dry-deposition of phosphorous (0.4–3.8 Gmol P yr−1) is comparable to its supply via rivers (1.1–2.0 Gmol P yr−1) draining into the BoB. Furthermore, a comparison of atmospheric dry-deposition of PInorg (0.5–4.8 µmol P m−2 d−1 during January 2009) with water column productivity (99–566 mg C m−2 d−1; Gauns et al., 2005) suggests that Aeolian input can support up to ~10% of in situ primary production. In the present-day scenario of growing anthropogenic activities, these results have implications to significant impact on the biogeochemistry of coastal waters downwind of pollution sources. Our estimate for the dry-deposition of PInorg to the NIO is of comparable magnitude with the model-based results.


We thank ISRO-GBP Programme Office (Bangaluru, India) for a partial funding support during the course of this study. We also thank chief scientists (Drs Krishnamoorthy and C. B. S. Dutt) and crew members of the ORV Sagar Kanya for logistic help during the cruises to the BoB and ARS. The analytical help provided by Dr R. Rengarajan is thankfully acknowledged. The authors would like to thank the reviewers for their critical comments and suggestions.


  1. Alappattu , D , Subrahamanyam , D , Kunhikrishnan , P , Ramachandran , R , Somayaji , K . and co-authors . 2008 . Spatio-temporal variability of surface-layer turbulent fluxes over the Bay of Bengal and Arabian Sea during the ICARB field experiment . Boundary-Layer Meteorology . 126 : 297 – 309 . .  

  2. Anderson L. D , Faul K. L , Paytan A . Phosphorus associations in aerosols: what can they tell us about P bioavailability? . Mar. Chem . 2010 ; 120 : 44 – 56 . .  

  3. Badarinath , K. V. S , Kharol , S. K , Kaskaoutis , D. G , Sharma , A. R , Ramaswamy , V . and co-authors . 2010 . Long-range transport of dust aerosols over the Arabian Sea and Indian region – a case study using satellite data and ground-based measurements . Global Planet. Change . 72 : 164 – 181 . .  

  4. Baker A. R , Jickells T. D , Witt M , Linge K. L . Trends in the solubility of iron, aluminium, manganese and phosphorus in aerosol collected over the Atlantic Ocean . Mar. Chem . 2006 ; 98 : 43 – 58 . .  

  5. Chen , Y , Mills , S , Street , J , Golan , D , Post , A . and co-authors . 2007 . Estimates of atmospheric dry deposition and associated input of nutrients to Gulf of Aqaba seawater . J. Geophys. Res . 112 : D04309 . .  

  6. Chen Y , Street J , Paytan A . Comparison between pure-water- and seawater-soluble nutrient concentrations of aerosols from the Gulf of Aqaba . Mar. Chem . 2006 ; 101 : 141 – 152 . .  

  7. Chester R . Marine Geochemistry2nd ed . Blackwell Publishers : New York , 2002 ; 506 .  

  8. Compton , J , Mallinson , D , Glenn , C. R , Filippelli , G , Follmi , K . and co-authors . 2000 . Variations in the global phosphorous cycle . Marine Authigenesis: From Global to Microbial . SEPM Special Publication No. 66, ISBN 1-56576-064-6 , pp. 21 – 33 .  

  9. Dey , S , Tripathi , S. N , Singh , R. P and Holben , B. N . 2004 . Influence of dust storms on the aerosol optical properties over the Indo-Gangetic basin . J. Geophys. Res . 109 : D20211 – 20224 ., .  

  10. Draxler , R. R . 2002 . HYSPLIT-4 user's guide . NOAA Tech Memo . ERL ARL-230 , 35 .  

  11. Duce , R. A , Liss , P. S , Merrill , J. T , Atlas , E. L , Buat-Menard , P . and co-authors . 1991 . The atmospheric input of trace species to the world ocean . Global Biogeochem. Cycles . 5 : 193 – 259 . .  

  12. Fang T. H . Phosphorus speciation and budget of the East China Sea . Cont. Shelf Res . 2004 ; 24 : 1285 – 1299 . .  

  13. Gauns , M , Madhupratap , M , Ramaiah , N , Jyothibabu , R , Fernandes , V . and co-authors . 2005 . Comparative accounts of biological productivity characteristics and estimates of carbon fluxes in the Arabian Sea and the Bay of Bengal . Deep Sea Res. Part II . 52 : 2003 – 2017 . .  

  14. Ginoux , P , Chin , M , Tegen , I , Prospero , J , Holben , B . and co-authors . 2001 . Global simulation of dust in the troposphere: model description and assessment . J. Geophys. Res . 106 ( D17 ): 20255 – 20273 . .  

  15. Guieu C , Loÿe-Pilot M. D , Ridame C , Thomas C . Chemical characterization of the Saharan dust end-member: some biogeochemical implications for the western Mediterranean Sea . J. Geophys. Res . 2002 ; 107 : 4258 . .  

  16. Herman , J. R , Bhartia , P. K , Torres , O , Hsu , C , Seftor , C . and co-authors . 1997 . Global distribution of UV-absorbing aerosols from Nimbus-7/TOMS data . J. Geophys. Res . 102 : 911 – 922 . .  

  17. Herut B , Krom M. D , Pan G , Mortimer R . Atmospheric input of nitrogen and phosphorus to the Southeast Mediterranean: sources, fluxes, and possible impact . Limnol. Oceanogr . 1999a ; 44 ( 7 ): 1683 – 1692 . .  

  18. Herut B , Zohary T , Robarts R. D , Kress N . Adsorption of dissolved phosphate onto loess particles in surface and deep Eastern Mediterranean water . Mar. Chem . 1999b ; 64 : 253 – 265 . .  

  19. Izquierdo , R , Benítez-Nelson , C. R , Masqué , P , Castillo , S , Alastuey , A . and co-authors . 2012 . Atmospheric phosphorus deposition in a near-coastal rural site in the NE Iberian Peninsula and its role in marine productivity . Atmos. Environ . 49 : 361 – 370 . .  

  20. Jahnke R. A . The phosphorus cycle . Earth system science: From biogeochemical cycles to global change . Jacobson M. C . Charlson R. J . Rodhe H . Orians G. H . Academic Press : London , 2000 ; 360 – 376 .  

  21. Jickells , T. D , An , Z. S , Andersen , K. K , Baker , A. R , Bergametti , G . and co-authors . 2005 . Global iron connections between desert dust, ocean biogeochemistry, and climate . Science . 308 : 67 – 71 . .  

  22. Keene W. C , Pszenny A. P , Gallloway J. N , Hawley M. E . Sea salt correction and interpretation of constituent ratios in marine precipitation . J. Geophys. Res . 1986 ; 91 : 6647 – 6658 . .  

  23. Krom M. D , Herut B , Mantoura R. F. C . Nutrient budget for the Eastern Mediterranean: implications for phosphorus limitation . Limnol. Oceanogr . 2004 ; 49 : 1582 . .  

  24. Kumar M. D , Naqvi S. W. A , George M. D , Jayakumar D. A . A sink for atmospheric carbon dioxide in the northeast Indian Ocean . J. Geophys. Res . 1996 ; 101 : 18121 – 18125 . .  

  25. Kumar A , Sarin M. M . Mineral aerosols from western India: temporal variability of coarse and fine atmospheric dust and elemental characteristics . Atmos. Environ . 2009 ; 43 : 4005 – 4013 . .  

  26. Kumar A , Sarin M. M , Srinivas B . Aerosol iron solubility over Bay of Bengal: role of anthropogenic sources and chemical processing . Mar. Chem . 2010 ; 121 : 167 – 175 . .  

  27. Kumar A , Sudheer A. K , Sarin M. M . Chemical characteristics of aerosols in MABL of Bay of Bengal and Arabian Sea during spring inter-monsoon: a comparative study . J. Earth Syst. Sci . 2008 ; 117 ( S1 ): 325 – 332 . .  

  28. Lelieveld , J , Crutzen , P. J , Ramanathan , V , Andreae , M. O , Brenninkmeijer , C. A. M . and co-authors . 2001 . The Indian Ocean experiment: widespread air pollution from South and Southeast Asia . Science . 291 : 1031 – 1036 . .  

  29. Li Q. P , Hansell D. A . Intercomparison and coupling of magnesium-induced co-precipitation and long-path liquid-waveguide capillary cell techniques for trace analysis of phosphate in seawater . Anal. Chim. Acta . 2008 ; 611 : 68 – 72 . .  

  30. Luo , C , Mahowald , N and del Corral , J . 2003 . Sensitivity study of meteorological parameters on mineral aerosol mobilization, transport and distribution . J. Geophys. Res . 108 ( D15 ): 4447 . .  

  31. Mahowald , N. M , Artaxo , P , Baker , A. R , Jickells , T. D , Okin , G. S . and co-authors . 2005 . Impacts of biomass burning emissions and land use change on Amazonian atmospheric phosphorus cycling and deposition . Global Biogeochem. Cycles . 19 : GB4030 . .  

  32. Mahowald , N , Jickells , T. D , Baker , A. R , Artaxo , P , Benitez-Nelson , C. R . and co-authors . 2008 . Global distribution of atmospheric phosphorus sources, concentrations and deposition rates, and anthropogenic impacts . Global Biogeochem. Cycles . 22 : GB4026-4045 . .  

  33. Markaki , Z , Oikonomou , K , Kocak , M , Kouvarakis , G , Chaniotaki , A . and co-authors . 2003 . Atmospheric deposition of inorganic phosphorus in the Levantine Basin, eastern Mediterranean: spatial and temporal variability and its role in seawater productivity . Limnol. Oceanogr . 48 : 1557 – 1568 . .  

  34. Martin J. M , Whitefield M . The significance of the river input of chemical elements to the ocean . Trace metals in sea water . Wong C. S . Boyle E . Bruland K. W . Burton J. D . Goldberg E. D . Plenum : New York , 1983 ; 265 – 296 .  

  35. McLennan , S. M . 2001 . Relationships between the trace element composition of sedimentary rocks and upper continental crust . Geochem. Geophys. Geosyst . 2 : ISSN: 1525–2027, article no: 2000GC000109 . .  

  36. Middleton N. J . A geography of dust storms in South-West Asia . J. Climatol . 1986 ; 6 : 183 – 196 . .  

  37. Middleton N. J . Climatic controls on the frequency, magnitude and distribution of dust storms: examples from India/Pakistan, Mauritania and Mongolia . Paleoclimatology and Paleometeorology: Modern and Past Patterns of Global Atmospheric Transport . Leinen M . Sarntheined M . Kluwer Academic : Norwell MA , 1989 ; 97 – 132 .  

  38. Mills M. M , Ridame C , Davey M , La Roche J , Geider R. J . Iron and phosphorus co-limit nitrogen fixation in the eastern tropical North Atlantic . Nature . 2004 ; 429 : 292 – 294 . .  

  39. Moutin , T , Karl , D. M , Duhamel , S , Rimmelin , P , Raimbault , P . and co-authors . 2008 . Phosphate availability and the ultimate control of new nitrogen input by nitrogen fixation in the tropical Pacific Ocean . Biogeosciences . 5 : 95 – 109 . .  

  40. Nenes , A , Krom , M. D , Mihalopoulos , N , Van Cappellen , P ., Shi , Z . and co-authors . 2011 . Atmospheric acidification of mineral aerosols: a source of bioavailable phosphorus for the oceans . Atmos. Chem. Phys . 11 : 6265 – 6272 . .  

  41. Okin , G. S , Baker , A. R , Tegen , I , Mahowald , N. M , Dentener , F. J . and co-authors . 2011 . Impacts of atmospheric nutrient deposition on marine productivity: roles of nitrogen, phosphorus, and iron . Global Biogeochem. Cycles . 25 : GB2022 . .  

  42. Paytan A , McLaughlin K . The oceanic phosphorous cycle . Chem. Rev . 2007 ; 107 : 563 – 576 . .  

  43. Prasanna Kumar , S , Narvekar , J , Kumar , A , Shaji , C , Anand , P . and co-authors . 2004 . Intrusion of the Bay of Bengal water into the Arabian Sea during winter monsoon and associated chemical and biological response . Geophys. Res. Lett . 31 : L15304 . .  

  44. Prospero , J. M . 1996 . The atmospheric transport of particles to the ocean . In : Particle Flux in the Ocean . V. Ittekkot , P. Schäfer , S. Honjo and P. J. Depetris . John Wiley & Sons Ltd. : New York , vol. 57 of SCOPE , pp. 19 – 52 .  

  45. Prospero , J. M , Ginoux , P , Torres , O , Nicholson , S. E and Gill , T. E . 2002 . Environmental characterization of global sources of atmospheric soil dust identified with the Nimbus 7 Total ozone Mapping Spectrometer (TOMS) absorbing aerosol product . Rev. Geophys . 40 : 1002 . .  

  46. Raj , P. E , Saha , S. K , Sonbawne , S. M , Deshpande , S. M , Devara , P. C. S . and co-authors . 2008 . Lidar observation of aerosol stratification in the lower troposphere over Pune during pre-monsoon season of 2006 . J. Earth Syst. Sci . 117 : 293 – 302 . .  

  47. Ramanathan , V , Crutzen , P. J , Lelieveld , J , Mitra , A. P , Althausen , D . and co-authors . 2001 . Indian Ocean experiment: an integrated analysis of the climate forcing and effects of the great Indo-Asian haze . J. Geophys. Res . 106 : 28371 – 28398 . .  

  48. Rastogi N , Sarin M. M . Chemistry of aerosols over a semi-arid region: evidence for acid neutralization by mineral dust . Geophys. Res. Lett . 2006 ; 33 : L23815 . .  

  49. Rastogi N , Sarin M. M . Quantitative chemical composition and characteristics of aerosols over western India: one-year record of temporal variability . Atmos. Environ . 2009 ; 43 : 3481 – 3488 . .  

  50. Ridame C , Guieu C . Saharan input of phosphate to the oligotrophic water of the open western Mediterranean Sea . Limnol. Oceanogr . 2002 ; 47 ( 3 ): 856 – 869 . .  

  51. Saliba N. A , Chamseddine A . Uptake of acid pollutants by mineral dust and their effect on aerosol solubility . Atmos. Environ . 2012 ; 46 : 256 – 263 . .  

  52. Sanudo-Wilhelmy , S. A , Kustka , A. B , Gobler , C. J , Hutchins , D. A , Yang , M . and co-authors . 2001 . Phosphorus limitation of nitrogen fixation by Trichodesmium in the central Atlantic Ocean . Nature . 411 : 66 – 69 . .  

  53. Sarin M. M , Krishnaswami S , Somayajulu B. L. K , Moore W. S . Chemistry of uranium, thorium, and radium isotopes in the Ganga–Brahmaputra river system: weathering processes and fluxes to the Bay of Bengal . Geochim. Cosmochim. Acta . 1990 ; 54 : 1387 – 1396 . .  

  54. Sarin , M , Kumar , A , Srinivas , B , Sudheer , A and Rastogi , N . 2011 . Anthropogenic sulphate aerosols and large Cl-deficit in marine atmospheric boundary layer of tropical Bay of Bengal . J. Atmos. Chem . 66 : 1–10 .  

  55. Schlesinger W . Biogeochemistry: An Analysis of Global Change2nd ed . Elsevier : New York , 1997 ; 588 .  

  56. Siefert R. L , Johansen A. M , Hoffmann M. R . Chemical characterization of ambient aerosol collected during the southwest monsoon and intermonsoon seasons over the Arabian Sea: Labile-Fe(II) and other trace metals . J. Geophys. Res . 1999 ; 104 : 3511 – 3526 . .  

  57. Singh B , Singh Y . Reactive nitrogen in Indian agriculture: inputs, use efficiency and leakages . Curr. Sci . 2008 ; 94 : 1382 .  

  58. Spokes L. J , Yeatman S. G , Cornell S. E , Jickells T. D . Nitrogen deposition to the eastern Atlantic Ocean. The importance of south-easterly flow . Tellus B . 2000 ; 52 : 37 – 49 . .  

  59. Srinivas , B , Sarin , M and Kumar , A . 2011 . Impact of anthropogenic sources on aerosol iron solubility over the Bay of Bengal and the Arabian Sea . Biogeochem . . 1 – 12 .  

  60. Subrahamanyam , D. B , Kumar , N. V. P. K , Dutt , C. B. S , Anurose , T. J , Kunhikrishnan , P. K . and co-authors . 2011 . Characterization of air–sea interaction processes over the Bay of Bengal during the winter phase of ICARB field experiment . Atmos. Res . 99 : 97 – 111 . .  

  61. Taylor S. R , McLennan S. M . The Continental Crust: Its Composition and Evolution . Blackwell : Oxford , 1985  

  62. Tegen , I , Werner , M , Harrison , S. P and Kohfeld , K. E . 2004 . Relative importance of climate and land use in determining present and future global soil dust emission . Geophys. Res. Lett . 31 : L05105 . .  

  63. Tindale N. W , Pease P. P . Aerosols over the Arabian Sea: atmospheric transport pathways and concentrations of dust and sea salt . Deep Sea Res. Part II . 1999 ; 46 : 1577 – 1595 . .  

  64. Tyrrell T . The relative influences of nitrogen and phosphorus on oceanic primary production . Nature . 1999 ; 400 : 525 – 531 . .  

  65. Wolf G. T . On the nature of nitrate in coarse continental aerosols . Atmos. Environ . 1984 ; 18 : 977 – 981 . .  

  66. Yadav S , Rajamani V . Geochemistry of aerosols of northwestern part of India adjoining the Thar desert . Geochim. Cosmochim. Acta . 2004 ; 68 : 1975 – 1988 . .  

  67. Zender , C. S , Bian , H and Newman , D . 2003 . Mineral dust entrainment and deposition (DEAD) model: description and 1990s dust climatology . J. Geophys. Res . 108 : 4416 – 4435 . .  

  68. Zhang J.-Z , Chi J . Automated analysis of nanomolar concentrations of phosphate in natural waters with liquid waveguide . Environ. Sci. Technol . 2002 ; 36 : 1048 – 1053 . .  

comments powered by Disqus