Article
Use of Phosphate Oxygen Isotopes for Identifying Atmospheric‑P Sources: A Case Study at Lake Kinneret Avner Gross,*,† Ami Nishri,‡ and Alon Angert† †
The Institute of Earth Sciences, The Hebrew University of Jerusalem, Israel Kinneret Limnological Laboratory, KLL, Israel Oceanographic and Limnological Institute, IOLR, Migdal, Israel
‡
ABSTRACT: The input of phosphorus (P) through atmospheric deposition can be a major source of P to fresh water bodies and may strongly affect their biogeochemistry. In Lake Kinneret (LK), northern Israel, dust deposition provides a significant fraction of the bioavailable P input. Here, we demonstrate that the oxygen isotopic composition of resin-extractable inorganic phosphate (δ18OP) in dust particles can be used to identify the phosphate source. Samples of soils with both natural vegetation and agricultural cover were collected upwind of LK and found to have distinct δ18OP value ranges (17.4−18.2‰ and 19.3−22.1‰, respectively). The δ18OP values for dust, collected continuously over LK during June 2011 to March 2012, were in the same range as agricultural soils. The dust concentration in the air decreased from the dry to the wet season and was correlated with a decrease in P concentration in air, yet no correlation was found between these parameters and dust δ18OP. Dust deposited during short-term desert dust events was characterized by a combination of high δ18OP values ranging from 22.2‰ to 22.7‰ and high concentrations of dust in the air. The data we present demonstrates a new application of δ18OP measurements for direct estimation of dust-P sources to lakes, as well as the potential for tracing dust-P on larger scales. surface environmental conditions (T < 80 °C), and in the short time scales relevant to this study is broken only by enzymemediated reactions. These biotic reactions produce an exchange between oxygen in water and oxygen in phosphate, a process which is accompanied by isotopic fractionation. In contrast, abiotic processes bring about insignificant exchange and only weakly fractionate phosphate oxygen isotopes.11−13 Hence, we can assume that no isotopic fractionation is associated with abiotic adsorption of phosphate to fine soil particles and therefore that the dust δ18OP value reflects the actual source soil value. The utilization of δ18OP as a tracer of various phosphate sources to lakes, rivers, estuaries, and oceans has been found useful when the source isotopic signature has not been erased by P biological cycling after reaching the sink reservoir.14−17 To overcome this confounding factor of possible alteration of δ18OP values after the dust was deposited in LK, dust was collected prior to its deposition in the lake; thus, its δ18OP retained the isotopic signature of the source. The dust contains few P pools, which can be operationally defined by the extraction method (HCl-extractable, NaOHextractable, anion-exchange-resin-extractable, etc.) For being useful as a tracer, the selected P pool should have different δ18OP values at the different possible dust sources. Here, we decided to focus on the resin-P pool, considered to be a form of labile inorganic P, which is lightly adsorbed to the outer
1. INTRODUCTION Studies of atmospheric phosphorus (P) provide evidence for the importance of airborne particles in delivering P to oceans, fresh water bodies, and terrestrial ecosystems.1−3 These airborne particles (hereafter “dust”), derived from soils, consist mainly of fine particles. The significant role of dust particles in P transport is probably due to their highly specific surface area that contributes to their high P sorption capacity.4 Dust-P inputs sustain marine ecosystem productivity in certain areas and may strongly impact terrestrial ecosystem biogeochemistry in other areas.5,6 In fresh water bodies, excessive P input has the potential to accelerate eutrophication. Lake Kinneret (LK; the Sea of Galilee, Figure 1) in northern Israel is the only large body of fresh water in Israel and provides about one-quarter of Israel’s water supply. It also supplies water to the Hashemite Kingdom of Jordan. Maintaining water quality is therefore of prime importance; hence, nutrient loads in the Amud Stream and Jordan River that feed the lake and the concentrations in the lake itself have been monitored since 1969, and factors affecting the lake biological and chemical composition have been studied intensively.7−9 However, the effects of dust deposition on the lake and its water properties have only been considered recently by Ganor and Foner10 that showed that it comprises 40% of the lake annual P supply. Here, we developed an approach based on using phosphate-stable oxygen isotope (δ18OP) to trace the sources of P in dust. This approach is based on comparing the δ18OP values in the dust particles collected at the lake shore to those values found in local soils upwind of the lake. The ability to use the δ18OP as a conservative tracer is based on the chemical stability of the P−O bond, which under most © 2013 American Chemical Society
Received: Revised: Accepted: Published: 2721
July 2, 2012 February 10, 2013 February 18, 2013 February 18, 2013 dx.doi.org/10.1021/es305306k | Environ. Sci. Technol. 2013, 47, 2721−2727
Environmental Science & Technology
Article
Figure 1. Detailed map of the 11 sampling sites west of Lake Kinneret. The dust sampling station in Tabgha is marked with a red “S”. The topographic contour lines are drawn with 100 m intervals.
Table 1. Resin-P Concentration and Its δ 18Op Values from Commercial Fertilizers (1.1 and 1.2) and Soils Upwind to Lake Kinnereta sample name
site and crop
cover
date of sampling
resin-P δ18Op (‰)
resin-P conc (μg PO4 g soil−1)
coordinate
−
−
22.8 ± 0.3
−
−
−
−
21.8 ± 0.3
−
−
2
liquid fertilizer from “Deshanim” solid fertilizer from “Deshanim” watermelon, Beit Netofa
field crop
July 2011
19.3 ± 0.7
55.5 ± 0.6
3
cereal, Beit Netofa
field crop
July 2011
20.7 ± 0.3
160.8 ± 1.8
4
cereal, Ginosar
field crop
July 2011
21.3 ± 0.6
26.8 ± 0.3
5
lychee orchards, Ginosar
July 2011
22.1 ± 0.2
57.9 ± 0.6
6
melon, Ginosar
agricultural plantation field crop
July 2011
19.6 ± 0.2
76.5 ± 0.8
7
cereal, Ginosar
field crop
July 2011
20.4 ± 0.4
43.4 ± 0.5
8
banana orchard, Ginosar
Sept 2011
21.9 ± 0.2
139.7 ± 1.5
9
cereal, Ginosar
agricultural plantation field crop
Sept 2011
20.2 ± 0.4
129.4 ± 1.4
10
Beit Keseht forest
natural vegetation
July 2011
18.2 ± 0.4
12.5 ± 0.1
11
Beit Keshet forest
natural vegetation
Sept 2011
17.4 ± 0.4
11.6 ± 0.1
S
dust collection station, Tabgha
−
June 2011 to March 2012
−
−
32°47′35.76 N, 35°19′70.21 E 32°48′56.80 N, 35°23′52.08 E 32°51′03.73 N, 35°28′58.72 E 32°50′58.80 N, 35°28′23.59 E 32°50′53.00 N, 35°31′35.02 E 32°52′04.00 N, 35°31′47.57 E 32°51′58.61 N, 35°34′57.56 E 32°52′05.67 N, 35°31′68.59 E 32°46′56.30 N, 35°16′67.49 E 32°46′56.30 N, 35°16′67.49 E 32°51′55.97 N, 35°32′53.24 E
1.1 1.2
a
Soils location, crop type cover, and their sampling time are also designated. Statistically significant values (p < 0.01) were found for samples taken from agricultural fertilized soils (2−9) and natural vegetation soils (10 and 11).
2722
dx.doi.org/10.1021/es305306k | Environ. Sci. Technol. 2013, 47, 2721−2727
Environmental Science & Technology
Article
Table 2. Dust and P Concentration in Air, the Resin-P Concentration and δ18OP Values Based on Dust from Filters Pooled into 14 Groups from June 2011 to March 2012 group
sampling date
resin-P conc (μg PO4 g dust−1)
air dust conc (μg dust m−3 air)
P conc in air (ng P m−3 air)
1 2 3 4 5 6 7 8 9 10 11 12
30.6.11−24.7.11 24.7.11−9.8.11 11.8.11−28.8.11 29.8.11−17.9.11 20.9.11−10.10.11 11.10.11−25.10.11 26.10.11−15.11.11 15.11.11−15.12.11 15.12.11−10.1.12 10.1.12−7.2.12 12.2−16.3.12 desert 27−29.9.11 desert 7−8.2.12 desert 16−19.3.12
932 ± 10.3 683 ± 7.5 892 ± 9.8 1137 ± 12.5 813 ± 8.9 1031 ± 11.3 1320 ± 14.5 1143 ± 12.5 479 ± 5.3 1026 ± 11.3 926 ± 10.2 1350 ± 14.9
67.9 126.1 97.1 90.5 89.7 51.9 44.2 25.3 49.2 35.5 24.5 162.6
63.4 86.2 86.6 102.9 72.9 53.5 58.3 28.9 23.6 36.5 22.6 219.5
20.7 20.9 19.5 20.6 19.8 21.4 20.3 19.9 21.8 20.9 20.4 22.1
761 ± 8.4
191.7
145.8
22.6 ± 0.8
418 ± 4.6
135.5
56.7
22.4 ± 0.2
13 14
δ18OP (‰) ± ± ± ± ± ± ± ± ± ± ± ±
1.8 1.6 0.2 0.4 0.7 1.5 1.2 0.3 0.9 1.0 0.7 0.6
northwestern shore of LK, on the roof of the Kinneret Data Center in Tabgha, Israel (Figure 1). The particles were collected on a single filter paper (Whatman no. 42) that was weighed and placed on a high volume air sampler (ECOTECH, Knoxfield, Australia, model Hivol3000) with a volumetric flow rate of 60 m3 h−1. The air sampler was positioned at a height of ∼5 m above the ground to decrease the contribution of particles from a close proximity to the sampling station. Each filter was replaced every 2−3 sampling days, air-dried at 60 °C, and weighed again. For dust-P δ18OP analysis, filters were pooled in chronological order until they reached the total amount of collected dust sufficient for isotopic analysis resulting in 14 groups (Table 2). All filters from each specific time period were pooled together for extraction. During the nine months of this study, there were three short-term events where desert dust was transported over a long range. The dust collected during these events, which occurred on 27−29 September 2011, 7−8 February 2012, and 13−19 March 2012 was analyzed separately. All three events were evident in MODIS images (NASA/GSFC, Rapid response), which shows a plume of dust originating in the Arab desert in northern Saudi Arabia and southern Jordan and traveling to LK from the southeast. By visually checking MODIS images for the entire study period, we verified that these were the only major events. 2.2. P-Fertilizer Inputs. Data on land use and timing and techniques used for fertilization was collected by interviewing local farmers. For agricultural plantations, such as banana or lychee, the P-fertilizer was dissolved in water and dispersed by an irrigation system once a week all year round. For field crops, such as melons and cereal, fertilizer was added as solid Ca(H2PO4)2 or NH4 phosphate, at specific intervals, usually during the wet season. Samples of the dissolved P-fertilizer used by Ginosar Valley farmers (sample 1.1, Table 1) and a sample of the solid P-fertilizer (sample 1.2, Table 1) purchased from ICL Fertilizers (Beer Sheva, 84100, Israel) authorized distributor were analyzed. 2.3. Analytical Methods. In the current study, we have focused on the resin-extractable phosphate (resin-P) and measured its oxygen isotopic ratio (δ18OP) in soil and dust samples and in commercial fertilizers. Angert et al.21 reported the phosphate concentration in dry (and rewet) soils in similar
surfaces of solid particles and is readily available for uptake by micro-organisms in soils.18 The resin-P pool is likely to be also available for aquatic micro-organisms, but to the best of our knowledge, this was never tested directly. A preliminary test showed that the HCl-extractable pool does not discriminate between LK dust-P sources, and for this reason, this pool was not used in the current study. However, this pool as well as other P pools may be useful for tracing dust-P in other sites. The dust sampling strategy for this research was based on previous studies at LK. These studies found that dust-fall on LK was mainly from local sources, 1−20 km from the lake.10,19 The dominant wind in the area during summer and early autumn is a westerly sea breeze originating from the Israeli Mediterranean coast, moving toward the lake.20 This dominating wind regime is occasionally interrupted in spring and fall by southeasterly desert dust storms (Red Sea Trough) and by strong easterly wind events (“Sharqiya”) during the fall and winter. In this study, we focused on local dust sources west of LK, including intensively fertilized agricultural lands and natural vegetation sites.
2. METHODS 2.1. Study Sites. 2.1.1. Soils. Soil samples were collected from 10 agricultural sites and 2 natural vegetation sites in the lower Galilee, west of LK, Israel (Table 1, Figure 1). Agricultural soils sampled in Beit Netofa Valley (sites 2 and 3) are alluvial vertisols. Agricultural soils from Ginosar Valley (sites 4−9) vary between terra rossa (equivalent to chromatic luvisols in the FAO classification system) and basalt derived vertisols. Natural (unfertilized) terra rossa soils (sites 10 and 11) were sampled in the Beit Keshet Forest: a 2500 ha mixed natural oak forest and planted pine forest, 300−400 m above sea level. The soils were sampled during the dry season, from July−September 2011. All of the sites experience similar soil temperatures (dry season mean ∼28 °C). Two samples per plot were taken from the upper soil profile (5 cm deep) and pooled. Soil samples were always taken at places with no vegetation cover. Soil samples were dried at 40 °C for 24 h, ground and sieved to 2 mm, and stored at 4 °C until analysis. 2.1.2. Dust Sampling. Dust samples were collected continuously from 30 June 2011 to 19 March 2012, on the 2723
dx.doi.org/10.1021/es305306k | Environ. Sci. Technol. 2013, 47, 2721−2727
Environmental Science & Technology
Article
Figure 2. δ18OP of resin-P (in permil vs VSMOW) from various agricultural fertilized soils west of Lake Kinneret, averaged over two soil samples taken from the same site. The error bars represent the standard deviation between two soil samples from the same site. (The numbers refer to the samples as presented in Table 1.)
Figure 3. δ18OP (in permil vs VSMOW) of dust collected continuously from June 2011 to March 2012 at Tabgha. The value range of agricultural soils is marked in white background and for natural vegetation soils in brown background. Results are averaged over three Ag3PO4 replicates from the same sample, and the error bars represent one standard deviation.
determinations, three replicates of ∼0.3 mg of silver phosphate of the same sample were packed in silver capsules and introduced into a high-temperature pyrolysis unit (HT-EA), where they were converted to CO in the presence of glassy carbon.25 The HT-EA was interfaced in continuous flow mode, through a gas chromatograph column, to an isotope ratio mass spectrometer (Sercon 20-20). All isotopic values were given in the delta-notation versus VSMOW. The standard deviation between three replicates of the same sample range from 0.2‰ to 1.4‰, with one exceptional data point with standard deviation of 1.8‰. All measurements were performed against two Ag3PO4 lab standards, which were calibrated against the following Ag3PO4 standards (isotopic signatures in parentheses): TU-1 (21.1‰) and TU-2 (5.4‰),25 UMCS-1 (32.6‰) and UMCS-2 (19.4‰),26 and against the IAEA-601 benzoic acid standard (23.3‰).27
sites in Israel was not higher than in the wet soils, indicating that, in contrast to other soils,22 in these soils, there is no significant release of phosphate from microbial cell lysis during rewetting. This was not tested for the dust samples, but we have no reason to assume the relative contribution of the phosphate released from the microbial biomass in dust will be any different than in soils. The resin-P δ18OP measurements were made by a method recently developed in our laboratory23 and described here briefly: the resin-P was first extracted from the soil or dust samples by shaking 100 g of soil or 8−10 filters (from adjacent sampling days that represent an entire time period as shown in Figures 3 and 4) in deionized water with anion-exchange membranes (BDH-55164). Phosphate was then eluted from the membranes by shaking them in 0.2 M HNO3 overnight. Concentrations of the extracted phosphate were determined colorimetrically by the method of Murphy and Riley,24 with an average difference between replicates of the same sample of 1%. Organic matter was removed from the extract by shaking it overnight with 20 mL of Superlite DAX-8 resin (Supelco/ Sigma−Aldrich). The phosphate was then extracted and purified by the precipitation of cerium phosphate. The precipitate was dissolved in nitric acid, the cerium was removed by a cation-exchange resin, and finally, the phosphate was precipitated as silver phosphate. For isotopic composition
3. RESULTS Table 1 summarizes the δ18OP values measured in commercial P-fertilizers and in agricultural and natural soils. The two Pfertilizers produced in Israel had δ18OP values of 21.8−22.8‰. Resin-P δ18OP (hereafter δ18OP) for all sites sampled west of LK (sites 2−11) varied over a range of 17.4−22.1‰. The 2724
dx.doi.org/10.1021/es305306k | Environ. Sci. Technol. 2013, 47, 2721−2727
Environmental Science & Technology
Article
Figure 4. Dust concentration in air (brown) and P concentration in air (black) are plotted according to their sampling time. Analytical error bars are smaller than symbol sizes.
Figure 5. Desert dust (brown triangles) and local dust (black squares) are characterized by distinct signatures, consists of dust concentrations in air and its δ18OP.
highest δ18OP values were observed at sites 5 and 8 (21.9− 22.1‰), lychee and banana plantations. The lowest values were measured in natural vegetation soils at sites 10 and 11 (17.4− 18.2‰). The δ18OP values for agricultural soils ranged from 19.3‰ to 22.1‰ and had statistically significant higher values (p < 0.01) than those of the natural vegetation soils (17.4− 18.2‰). The isotopic values found in agricultural soils were plotted versus time passed since they were last fertilized with Pfertilizer (Figure 2). The soil resin-P concentrations ranged from 11 to 160 μg P g−1 soil. The highest value was in agricultural site 3, and the lowest was in natural vegetation sites 10 and 11. As expected, agricultural soils showed higher resin-P concentrations; nevertheless, no significant correlation between the resin-P concentration and its δ18OP values (Table 1) was observed, probably since fertilization intervals and the amount of fertilizer added varied with sites and crops. This lack of correlation indicates that the δ18OP tracer provides additional information to that contained in the concentration data. The dust samples were chronologically subdivided into 14 groups. The dust resin-P concentrations and its δ18OP values were determined (Table 2). The dust-P δ18OP values ranged from 19.5‰ to 21.8‰ (Figure 3). The lowest value was observed in the period 11−28 August 2011, and the highest value was between 15 December 2011 and 10 January 2012. The air dust concentrations ranged from 24.5 to 126.1 μg dust m−3 air with the higher values observed in the dry season. The highest value was observed between 24 July 2011 and 9 August 2011, and the lowest was between 12 February 2012 and 16 March 2012. In Figure 4, higher air dust concentrations corresponded with higher resin-P concentrations in the air, and
these values show a gradual decrease from a dry season average of 77.6 ng P m−3air to a wet season average of 34.0 ng P m−3 air. The dust collected during the three short-term desert dust events is characterized by high δ18OP values (22.1−22.6‰) and high values of air dust concentration (135.5−191.7 μg dust m−3 air), as compared to the values found for dust of local origin (Figure 5).
4. DISCUSSION 4.1. Phosphate Oxygen Isotopes of Soils Upwind of LK. For reliable discrimination between atmospheric P sources, the δ18OP values of these sources must be unique. This was the case in our study as local natural and agricultural soils upwind of LK show a significantly distinct range of δ18OP values (Table 1). In the following section, the reasons for the distinct values in the two soil types are explained. 4.1.1. δ18Op Values in Natural Vegetation Soils. The δ18OP values found in natural vegetation soils (17.4−18.2‰) are similar to values found in previous studies conducted in Israel by Angert et al.21,28 This indicates that the natural vegetation soils from Beit Keshet forest were not affected by contamination with dust from the neighboring agricultural soils in the lower Beit Netofa valley. We assume that, in this forest, the dust fluxes were limited due to the elevated location and dust captured by trees in the forest periphery. Angert et al.21,28 reported δ18OP average values of 16−19‰ in a series of natural vegetation soils along a climate gradient in Israel, including terra rossa (chromatic luvisols) and basaltic-derived vertisols soils west and east of LK. According to these studies, a biologically mediated isotopic equilibration between oxygen in 2725
dx.doi.org/10.1021/es305306k | Environ. Sci. Technol. 2013, 47, 2721−2727
Environmental Science & Technology
Article
phosphate and soil water is the process that governs δ18OP values in soils in Israel21,28 and elsewhere.29 The phosphate equilibrium fractionation is known to be temperature dependent:11,30,31
samples, and improving our methods for handling such small samples is under-way. In future studies, we also intend to perform dust particle size separation to estimate the contribution of different particles size classes to the total dust δ18OP values. The dust resin-P concentrations were significantly higher than the values found in soils (Tables 1 and 2). These values were in the same range as soluble P concentrations for dust collected over the Gulf of Aqaba.34 As observed for the soils, the dust resin-P concentration showed no correlation with its δ18OP values, indicating that the isotopic values supply additional information. Furthermore, no seasonal trend was observed in the resin-P concentrations in the dust, although a decrease in dust air concentration was observed from the dry to the wet season and was followed by an associated decrease in the P concentration in air from the dry season average to the wet season average (Figure 4). This trend is expected as the wet soils produce less dust. Seasonal trends in the air dust concentration were interrupted by three short-term desert dust events, which approached from the southeast. The dust δ18OP values during these three independent events showed only small variation and ranged from 22.1‰ to 22.6‰ (Table 2). These values reflect the expected isotopic composition of soils from the Arab desert, where the majority of the P is derived from sedimentary phosphorite rocks.32,35 As the P-fertilizers used in Israel are produced from analogous phosphorite rocks, the δ18OP values found in desert dust resembles the values found in several of the local agricultural soils. The air dust concentration measured during these events was significantly higher than the common concentrations found during the study (Table 2). Therefore, as can be seen in Figure 5, we suggest a combination of air dust concentration and its δ18OP values can be used to generate a distinct signature for dust-P arriving from local and desert sources. As these desert dust events are fairly short and distinct in their timing, we assume most of the dust-P deposited in LK with heavy δ18OP values is from local agricultural sources. We suggest the approach presented here, using an integration of traditional measurements of dust and P concentrations together with δ18OP values, is of great value for tracing dust-P, directly from source to sink in the local scale. More study is needed to estimate the potential of this application for tracing atmospheric-P sources in the regional and global scales.
δ18OP = δ18Owater + (111.4 − T )/4.3 (where T is the temperature in °C)
(1)
This equilibrium fractionation stems from a reaction catalyzed by the intracellular enzyme pyrophosphatase that completely exchanges oxygen in phosphate with oxygen in water.31 In soils, this equilibration is assumed to be controlled by the soil microbiota. 4.1.2. Resin-P Concentration and Its δ18OP Values in Agricultural Soils. Because agricultural soils are being intensively fertilized, the soil resin-P δ18OP values are expected to mainly reflect that of the P-fertilizer. The Israeli fertilizer industry produces phosphate fertilizers from sedimentary phosphorites, located in the Rotem Plains of the Negev, which have δ18OP values of 21.5−23‰32 (after adjustment to the revised value of the NBS120b standard33). These values are in agreement with the δ18OP values we found for P-fertilizers in local use (21.8−22.8‰, Table 1). Indeed, agricultural soils that were fertilized up to 1−2 months prior to sampling showed δ18OP values in the range 21.3−22.1‰. However, those values decrease linearly as more time passed from the fertilization point (Figure 2). A reasonable explanation for these findings would be as follows. Close to fertilization, the soil δ18OP values represent that of the P-fertilizer, but as time passes, the ongoing microbial turnover of the applied P accompanied by isotopic equilibrium fractionation gradually changes the δ18OP to the equilibrium value. Differences in the δ18OP observed for agricultural soil under different crop types such as field crops and plantations (Table 1) are probably derived from differing fertilization regimes. The weekly fertilization of banana and lychee plantations continuously replenishes the resin-P pool, and the soil retains the P-fertilizer isotopic signature. In contrast, when the fertilizer is added over longer intervals, the added P has sufficient time to undergo biological turnover before a new addition of the fertilizer replenishes the resin-P pool. 4.2. Air Dust Concentrations, Oxygen Isotopes, and Resin-P Concentrations. The collected dust δ18OP values measured during June 2011 to March 2012, varied in the range 19.5−21.8‰, (Figure 3). These values are higher than the values found in natural soils (16−19‰) but correspond closely with the range we observed for local agricultural soils (19.3− 22.1‰). These findings point to the major contribution of dust-P originating from local agricultural soils to the bioavailable dust-P deposited on LK during the study period. Using a simple isotopic mass balance calculation with two end members, one for natural vegetation soils with δ18OP values of 17.8‰ and another of ∼22.0‰ for agricultural soils that were recently fertilized, implies the dust-P fraction from agricultural sources contributes on a monthly basis between ∼45% and ∼95% of the total dust-P collected at the sampling station on LK shore and ∼65% of the dust-P deposited during the entire the study period. It should be also noted that, since the dust particles in this study were collected only at a single site, the observed values may not represent the dust supply to the entire LK. Three of the 14 dust samples showed large differences between replicates of the same sample (Figure 3). This occurred in the smallest
■
AUTHOR INFORMATION
Corresponding Author
*Phone: 972-2-6586516; fax: 972-2-5662581; e-mail: avner.
[email protected]. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS This study was funded by ISF grant #870/08 and the Israel Water Authority (Grant #13/2012). A.G. was funded by the Israel Ministry of Science and Technology (MOST). We would like to thank Y. Kolodny for his comments on an earlier version of the manuscript and Boaz Hilman for his GIS support.
■
REFERENCES
(1) Mahowald, N. M.; Engelstaedter, S.; Luo, C.; Sealy, A.; Artaxo, P.; Benitez-Nelson, C.; Bonnet, S.; Chen, Y.; Chuang, P. Y.; Cohen, D. D.; Dulac, F.; Herut, B.; Johansen, A. M.; Kubilay, N.; Losno, R.; Maenhaut, W.; Paytan, A.; Prospero, J. A.; Shank, L. M.; Siefert, R. L.
2726
dx.doi.org/10.1021/es305306k | Environ. Sci. Technol. 2013, 47, 2721−2727
Environmental Science & Technology
Article
Atmospheric iron deposition: Global distribution, variability, and human perturbations. Annu. Rev. Mar. Sci. 2009, 1, 245−278. (2) Mahowald, N.; Jickells, T. D.; Baker, A. R.; Artaxo, P.; BenitezNelson, C. R.; Bergametti, G.; Bond, T. C.; Chen, Y.; Cohen, D. D.; Herut, B.; Kubilay, N.; Losno, R.; Luo, C.; Maenhaut, W.; McGee, K. A.; Okin, G. S.; Siefert, R. L.; Tsukuda, S. Global distribution of atmospheric phosphorus sources, concentrations and deposition rates, and anthropogenic impacts. Global Biogeochem. Cycles 2008, 22 (4), GB4026. (3) Okin, G. S.; Baker, A. R.; Tegen, I.; Mahowald, N. M.; Dentener, F. J.; Duce, R. A.; Galloway, J. N.; Hunter, K.; Kanakidou, M.; Kubilay, N.; Prospero, J. M.; Sarin, M.; Surapipith, V.; Uematsu, M.; Zhu, T. Impacts of atmospheric nutrient deposition on marine productivity: Roles of nitrogen, phosphorus, and iron. Global Biogeochem. Cycles 2011, 25 (2), GB2022. (4) Dong, A.; Simsiman, G. V.; Chesters, G. Particle-size distribution and phosphorus levels in soil, sediment, and urban dust and dirt samples from the Menomonee River Watershed, Wisconsin, USA. Water Res. 1983, 17 (5), 569−577. (5) Pett-Ridge, J. C. Contributions of dust to phosphorus cycling in tropical forests of the Luquillo Mountains, Puerto Rico. Biogeochemistry 2009, 94 (1), 63−80. (6) Okin, G. S.; Mahowald, N.; Chadwick, O. A.; Artaxo, P. Impact of desert dust on the biogeochemistry of phosphorus in terrestrial ecosystems. Global Biogeochem. Cycles 2004, 18 (2), GB002145. (7) Serruya, C. Lake Kinneret; B.V Publisher: Boston, 1978. (8) Walline, P. D.; Pisanty, S.; Gophen, M.; Berman, T. The ecosystems of Lake Kinneret, Israel. In Trophic Models of Aquatic Ecosystems; Christensen, V., Pauly, D., Eds.; The International Center for Living Aquatic Resources Management, MCPO and the International Council for the Exploration of the Sea: Manila, Philippines, 1993; pp 103−109. (9) Hambright, K. D.; Zohary, T.; Eckert, W.; Schwartz, S. S.; Schelske, C. L.; Laird, K. R.; Leavitt, P. R. Exploitation and destabilization of a warm, freshwater ecosystem through engineered hydrological change. Ecol. Appl. 2008, 18 (7), 1591−1603. (10) Foner, H. A.; Ganor, E.; Gravenhorst, G. The chemical composition and sources of the bulk deposition on Lake Kinneret (The Sea of Galilee), Israel. J. Arid Environ. 2009, 73 (1), 40−47. (11) Kolodny, Y.; Luz, B.; Navon, O. Oxygen isotope variations in phosphate of biogenic apatites, I. Fish bone apatiterechecking the rules of the game. Earth Planet. Sci. Lett. 1983, 64 (3), 398−404. (12) Liang, Y.; Blake, R. E. Oxygen isotope fractionation between apatite and aqueous-phase phosphate: 20−45 °C. Chem. Geol. 2007, 238 (1−2), 121−133. (13) Jaisi, D. P.; Blake, R. E.; Kukkadapu, R. K. Fractionation of oxygen isotopes in phosphate during its interactions with iron oxides. Geochim. Cosmochim. Acta 2010, 74 (4), 1309−1319. (14) Markel, D.; Kolodny, Y.; Luz, B.; Nishri, A. Phosphorus cycling and phosphorus sources in Lake Kinneret: Tracing by oxygen isotopes in phosphate. Isr. J. Earth Sci. 1994, 43 (3−4), 165−178. (15) Colman, A. S. The oxygen isotope composition of dissolved inorganic phosphate and the marine phosphorus cycle. Ph.D. Thesis, Yale University, 2002. (16) Elsbury, K. E.; Paytan, A.; Ostrom, N. E.; Kendall, C.; Young, M. B.; McLaughlin, K.; Rollog, M. E.; Watson, S. Using oxygen isotopes of phosphate to trace phosphorus sources and cycling in Lake Erie. Environ. Sci. Technol. 2009, 43 (9), 3108−3114. (17) Young, M. B.; McLaughlin, K.; Kendall, C.; Stringfellow, W.; Rollog, M.; Elsbury, K.; Donald, E.; Paytan, A. Characterizing the oxygen isotopic composition of phosphate sources to aquatic ecosystems. Environ. Sci. Technol. 2009, 43 (14), 5190−5196. (18) Hedley, M. J.; Stewart, J. W. B.; Chauhan, B. S. Changes in inorganic and organic soil-phosphorus fractions induced by cultivation practices and by laboratory incubations. Soil Sci. Soc. Am. J. 1982, 46 (5), 970−976. (19) Ganor, E.; Foner, H. A.; Gravenhorst, G. The amount and nature of the dustfall on Lake Kinneret (the Sea of Galilee), Israel: Flux and fractionation. Atmos. Environ. 2003, 37 (30), 4301−4315.
(20) Bitan, A. Lake Kinneret (Sea of Galilee) and its exceptional wind system. Boundary-Layer Meteorol. 1981, 21 (4), 477−487. (21) Angert, A.; Weiner, T.; Mazeh, S.; Tamburini, F.; Frossard, E.; Bernasconi, S. M.; Sternberg, M. Seasonal variability of soil phosphate stable oxygen isotopes in rainfall manipulation experiments. Geochim. Cosmochim. Acta 2011, 75 (15), 4216−4227. (22) Turner, B. L.; Haygarth, P. M. Biogeochemistry: Phosphorus solubilization in rewetted soils. Nature 2001, 411 (6835), 258−258. (23) Weiner, T.; Mazeh, S.; Tamburini, F.; Frossard, E.; Bernasconi, S. M.; Chiti, T.; Angert, A. A method for analyzing the δ18O of resinextractable soil inorganic phosphate. Rapid Commun. Mass Spectrom. 2011, 25 (5), 624−628. (24) Murphy, J.; Riley, J. P. A modified single solution method for the determination of phosphate in natural waters. Anal. Chim. Acta 1962, 27, 31−36. (25) Vennemann, T. W.; Fricke, H. C.; Blake, R. E.; O’Neil, J. R.; Colman, A. Oxygen isotope analysis of phosphates: A comparison of techniques for analysis of Ag3PO4. Chem. Geol. 2002, 185 (3−4), 321− 336. (26) Halas, S.; Skrzypek, G.; Meier-Augenstein, W.; Pelc, A.; Kemp, H. F. Inter-laboratory calibration of new silver orthophosphate comparison materials for the stable oxygen isotope analysis of phosphates. Rapid Commun. Mass Spectrom. 2011, 25 (5), 579−584. (27) Coplen, T. B.; Brand, W. A.; Gehre, M.; Groning, M.; Meijer, H. A. J.; Toman, B.; Verkouteren, R. M. New guidelines for delta C-13 measurements. Anal. Chem. 2006, 78 (7), 2439−2441. (28) Angert, A.; Weiner, T.; Mazeh, S.; Sternberg, M. Soil phosphate stable oxygen isotopes across rainfall and bedrock gradients. Environ. Sci. Technol. 2012, 46 (4), 2156−2162. (29) Tamburini, F.; Pfahler, V.; Bunemann, E. K.; Guelland, K.; Bernasconi, S. M.; Frossard, E. Oxygen isotopes unravel the role of microorganisms in phosphate cycling in soils. Environ. Sci. Technol. 2012, 46 (11), 5956−5962. (30) Longinelli, A.; Nuti, S. Oxygen isotope measurements of phosphate from fish teeth and bones. Earth Planet. Sci. Lett. 1973, 20 (3), 337−340. (31) Blake, R. E.; O’Neil, J. R.; Surkov, A. V. Biogeochemical cycling of phosphorus: Insights from oxygen isotope effects of phosphoenzymes. Am. J. Sci. 2005, 305 (6−8), 596−620. (32) Shemesh, A.; Kolodny, Y.; Luz, B. Oxygen isotope variations in phosphate of biogenic apatites 0.2. Phosphorite rocks. Earth Planet. Sci. Lett. 1983, 64 (3), 405−416. (33) Pucéat, E.; Joachimski, M. M.; Bouilloux, A.; Monna, F.; Bonin, A.; Motreuil, S.; Morinière, P.; Hénard, S.; Mourin, J.; Dera, G.; Quesne, D. Revised phosphate−water fractionation equation reassessing paleotemperatures derived from biogenic apatite. Earth Planet. Sci. Lett. 2010, 298 (1−2), 135−142. (34) Chen, Y.; Mills, S.; Street, J.; Golan, D.; Post, A.; Jacobson, M.; Paytan, A. Estimates of atmospheric dry deposition and associated input of nutrients to Gulf of Aqaba seawater. J. Geophys. Res.: Atmos. 2007, 112 (D4), D04309. (35) Shemesh, A.; Kolodny, Y.; Luz, B. Isotope geochemistry of oxygen and carbon in phosphate and carbonate of phosphorite francolite. Geochim. Cosmochim. Acta 1988, 52 (11), 2565−2572.
2727
dx.doi.org/10.1021/es305306k | Environ. Sci. Technol. 2013, 47, 2721−2727
