GEOPHYSICAL RESEARCH LETTERS, VOL. **, NO. *, **.*029/2001GL013882, 2002

Geochemistry of atmospheric aerosols generated from

lava-seawater interactions

Francis J. Sansone,1 Claudia R. Benitez-Nelson,1 Joseph A. Resing,2 Eric H. DeCarlo,1

Sue M. Vink,1 Jacqueline A. Heath,1 and Barry J. Huebert1

Received 1 August 2001; revised 20 December 2001; accepted 1 April 2002; published 14 May 2002.

[1] Trace elements were measured in the aerosol plume produced extinction, heightening the importance of understanding the mech-

anisms and geochemical effects of these large events.

by lava-seawater interactions along the shoreline of Kilauea

[3] There are few data available on the direct lava-seawater

volcano, Hawaii. Plume concentrations were normalized relative

interactions that occur during upper ocean LIP emplacement.

to Hawaiian basalt composition and showed a linear log-log co-

However, lava from Kilauea volcano (19 200N, 155 000W) on

variation with their emanation coefficient (an indicator of element

the island of Hawaii has been flowing into the ocean almost

volatility). Normalized aerosol concentrations also consistently

continuously since 1986 [Mattox and Mangan, 1997], providing

covaried with corresponding normalized concentrations in dilute a convenient means for studying such processes. Although the

fumarolic gas from Kilauea volcano and fumarolic gas condensates physical and geochemical effects of this lava entry on the adjacent

from Kudryavy and Merapi volcanoes, despite different ocean have been previously reported [Sansone and Resing, 1995;

mechanisms of element volatilization. Conservatively estimated Resing and Sansone, 1999, in press], the composition of the

regional ocean deposition rates of Cu, Cd, Ni, Pb, Mn, Zn, Fe and aerosol ( steam ) plume associated with the lava entry has not

previously been determined.

P were >50 times background rates. Thus, upper ocean volcanism

may be an important source of both toxic and nutrient elements to

the surrounding ocean. It appears unlikely, however, that shallow

ocean volcanism can exert a significant impact on the global

2. Methods

ecosystem, even during massive lava emplacements. INDEX

[4] On the day of sample collection (16 July 2000) lava tubes

TERMS: 8409 Volcanology: Atmospheric effects (0370); 4825

released lava into the surface ocean either at the steep edge of the

Oceanography: Biological and Chemical: Geochemistry; 4875

shoreline lava bench or underwater with little contact with the

Oceanography: Biological and Chemical: Trace elements; 1630

atmosphere before water contact. The resulting boiling of seawater

Global Change: Impact phenomena

produced a visible atmospheric aerosol plume that was transported

parallel to the coastline by the winds on that day. Subaerial volcanic

inputs to the aerosol plume were minimized by both the wind

1. Introduction

direction and the small amount of subaerial lava extrusion along

[2] Significant inputs to the atmosphere are known to occur from the shoreline.

terrestrial volcanism [e.g., Lantzy and Mackenzie, 1979] and may [5] Plume aerosols were sampled using an active mesh collector

play a role in modulating climate by injecting large quantities of (impactor) consisting of an air pump that pulled 151 L min 1 of air

particles and magmatic gases into the atmosphere, potentially through 0.18 m2 of tightly folded plastic screen (0.5 mm opening)

affecting the earth s albedo and influencing the global biogeochem- located in the top of a 4-L wide mouth polyethylene bottle. The

ical cycling of many elements [e.g., Caldeira and Rampino, 1991]. collector was suspended below a Hughes 500D helicopter using

Sub-aqueous volcanism at the ocean surface also has the potential to 60 m of synthetic fiber rope; aerosol entered through four 1-cm

introduce a variety of environmentally important chemical species to diameter holes in the bottle wall, coalesced on the mesh, and was

the atmosphere through lava-water interactions; this process was collected in the bottom of the bottle. The buoyant portion of the

likely important in the geological past when near-surface ocean aerosol plume was sampled for 22 min, $15 m downwind of the

volcanism was more common. For example, volcanism near the lava entry and at an altitude of $10 m above sea level. Visual

ocean surface was a predominant process during the emplacement of observation showed that turbulence from the helicopter s rotor did

large igneous provinces (LIPs) [e.g., Coffin and Eldholm, 1994] and not disturb the aerosol plume in the vicinity of the sampler. The

during the opening of the current oceanic basins [e.g., White and collection efficiency of the sampler was close to 100%, as

McKenzie, 1989]. LIPs have volumes of up to tens of million km3 of evidenced by the absence of visible liquid or solid inside the

mafic rock, with mean emplacement rates of up to tens of km3 y 1 tubing connecting the outlet of the mesh with the air pump after

[e.g., Coffin and Eldholm, 1994]. In addition, LIP generation sampling. Prior to use, all apparatus were washed, soaked in 10%

occurred during periods of rapid species extinction, such as the HCl, triply rinsed with 18-M

deionized water and stored in

67 MA Cretaceous-Tertiary boundary and the 248 MA Permian cleaned polyethylene bags. After collection, samples were trans-

ferred to 30-mL polypropylene bottles.

[6] Cl, SO4, Mg2+ and K+ were determined ( 5%) by ion

2

chromatography (IC); Na ( 7%) by ICP-AES; Ca by both IC and

ICP-AES; Fe ( 2%) and Al ( 8%) by direct-injection flow

injection analysis (FIA) [Resing, 1997]; Hg ( 5%) by cold vapor

1

Department of Oceanography, School of Ocean and Earth Science and

atomic absorption spectroscopy; NO + NO, NH4 and Si ( 0.03

+

3 2

Technology, University of Hawaii, Honolulu, Hawaii, USA.

mM) by AutoAnalyzer; arsenate ( 5%) and soluble reactive P

2

Pacific Marine Environmental Laboratory, Joint Institute for the Study

(SRP) ( 2%) by spectrophotometry [Hansen and Koroleff, 1999];

of Atmosphere and Ocean, National Oceanic and Atmospheric Adminis-

and total dissolved P (TDP) ( 150 nM) by acid persulfate oxida-

tration, Seattle, Washington, USA.

tion [Hansen and Koroleff, 1999]. Other trace elements were

determined ( 0.2 6%) by FIA-ICP-MS [Resing, 1997]. pH was

Copyright 2002 by the American Geophysical Union.

measured ( 0.5) in the field with pHydrion indicating pH paper.

0094-8276/02/2001GL013882$05.00

49 - 1

49 - 2 SANSONE ET AL.: LAVA-GENERATED MARINE AEROSOLS

Table 1. Concenterations of Trace Elements in the Aerosol Plume, Estimated Aerosol-plume Atmospheric Input Rates, Estimated

Aerosol-plume and Tropical North Pacific Background Ocean Deposition Rates, and the Aerosol-plume/Background Deposition Rate

Ratios. Dashes Indicate Data Not Available

Even

distribution Dispersion Background Even distribution Dispersion

Aerosol-plume

Aerosol model model ocean model aerosol- model

atmospheric

plume Seawater deposition deposition deposition plume/ aerosol-plume

input rate rated

conc conc rate rate background /background

(mmol/kg)a (kg/y)b (mg/m3lava)c (g/m2/y) (mg/m2/y) (mg/m2/y)

(mg/kg) deposition rate deposition rate

2

1.0 10

Si 172**-****-**-*** 440

1 10 3

Fe 130**-****-**-** 340 5.6 16000 61

3 10 2

Al 732*-***-**-** 190-**-****-**

SRP 1480 150 8.6 10 37 0.7 14000 53

NPPe 820 84 4.8 5.6 18 1.8 3100 10

2.3 100

TDP 2300 234 13.4 16 55 2.5 6400 22

8.7 101

Sr 401*-***-**-** 100

6 10 3

Zn 184*-***-**-**-** 0.67 19000 72

4 10 3

Cu 1170 120 6.7 8.0 30 0.089 90000 340

5 10 3

Mn 275 28 1.6 1.9 7.1 0.09 21000 79

1 10 5

Pb 241 24 1.4 1.6 6.2 0.07 23000 89

8 10 3

Ni 142 14 0.82 0.93 3.7 0.023 40000 160

2.3 10 2

V 99 10 0.57 0.67 2.6 0.078 8600 33

1.4 100

Rb 98 10 0.56 0.67 2.5

1.0 10 1

Ba 58 5.9 0.34 0.39 1.5

4 10 3

Cr 46 4.7 0.27 0.31 1.2

7 10 4

Cd 34 3.5 0.20 0.23 0.88 0.0035 66000 250

1.1 10 1

Mo 28 2.8 0.16 0.19 0.72

3 10 5

Co 18 1.8 0.10 0.12 0.45

2 10 5

Ce 5.5 0.56 0.032 0.037 0.14

2.3 10 2

As 5.2 0.53 0.030 0.035 0.14 0.0056 6200 25

3 10 5

La 2.9 0.29 0.017 0.019 0.074

2.5 10 5

Ag 2.4 0.24 0.014 0.016 0.062

5 10 6

Hg 2.2 0.22 0.013 0.015 0.057 0.006 2500 10

2 10 5

Nd 1.8 0.18 0.010 0.012 0.047

1.5 10 4

Y 1.5 0.15 0.0085 0.010 0.038

1.3 10 2

U 0.90 0.091 0.0052 0.006 0.023

3 101

NO3+NO2-N 10 4;

elements with lower emanation coefficients fall noticeably above

the line.

[11] Although the extent and distribution of deposition from the

Kilauea aerosol plume remains uncertain, sulfate aerosols from

Kilauea s terrestrial vents have been detected over 1000 km away

[Porter and Clarke, 1997]. As a result, we attempted to model the

atmospheric plume deposition rates to the adjacent ocean, which is

potentially affected by inputs of nutrients such as Fe and P, and

toxic trace elements such a Pb and Cu.

[12] Using visual analysis of video images of the aerosol

plume during our sampling, we estimate the plume height to be

$300 m and the plume production rate to be $730 m3 s 1. The

Kilauea lava production rate was 320,000 m3 d 1, with $10

20% entering the ocean at the Waha ula lava entry where we Figure 2. Concentrations of trace elements in the lava-entry

sampled (J. Kauahikaua, pers. comm., 2001). These values allow aerosol (corrected for seawater contributions and normalized to

atmospheric input rates to be calculated for each trace element on their concentrations in Kilauea basalt) vs. their emanation

both an absolute and a lava-normalized basis (Table 1); these are coefficients. The dashed line (log y = 0.423 + 0.934 log x,

r2 = 0.91) is the linear regression for elements with emanation

likely to be minimum estimates, as it was not possible to keep the

coefficients >1 10 4. The overlapped symbols plotted below Mn

sampler located in the core of the aerosol plume at all times

during sampling. are Fe and Ce.

49 - 4 SANSONE ET AL.: LAVA-GENERATED MARINE AEROSOLS

arguments suggest that, at least on a regional scale, volcanic Graham, W. F., and R. A. Duce, Atmospheric pathways of the phosphorus

cycle, Geochim. Cosmochim. Acta, 43, 1195 1208, 1979.

aerosol deposition can have a significant effect on upper ocean

Graham, W. F., and R. A. Duce, Atmospheric input of phosphorus to remote

biogeochemistry. However, it is not clear at this point what the net

tropical islands, Pac. Sci., 35, 241 255, 1981.

effect of these different inputs is on upper ocean biota, as the Hansen, H. P., and F. Koroleff, Determination of nutrients, in Methods of

aerosol plume deposition includes both nutrients (e.g., Si, Fe, P) Seawater Analysis, 3rd Ed., edited by K. Grasshoff, M. Ehrherd, and K.

and potentially toxic elements (e.g., Cu, Cd, Pb). Note that sea- Kremling, pp. 159 228, Verlag Chemie, Weinheim, 1999.

surface lava-seawater interactions also produce large amounts of Johnson, D. L., Simultaneous determination of arsenate and phosphate in

natural waters, Environ. Sci. Tech., 5, 411 414, 1971.

macroscopic airborne particulate matter that is deposited very close

Lambert, G., M. F. Le Cloarec, B. Ardouin, and J. C. Le Roulley, Vocanic

to the lava entry [Resing, 1997]. This hyaloclastic glass is highly

emission of radionuclides and magma dynamics, Earth Planet. Sci. Lett.,

soluble [Sicks, 1975], and its dissolution may also contribute 76, 185 192, 1985/1986.

significantly to such oceanic processes as Fe fertilization Lantzy, R. J., and F. T. Mackenzie, Atmospheric trace metals: Global cycles

[Martin et al., 1991]. and assessment of man s impact, Geochim. Cosmochim. Acta, 43, 511

525, 1979.

[15] Our results can be used to examine whether large scale

Libes, S. M., An Introduction to Marine Biogeochemistry, Wiley & Sons,

upper ocean lava extrusions, such as those known to occur during

N.Y., 1992.

the formation of LIPs, are capable of injecting significant amounts

Li, Y. H., A Compendium of Geochemistry, Princeton Univ. Press, Prince-

of environmentally sensitive elements into the atmosphere. By ton, 2000.

assuming geochemical dynamics similar to the Kilauea lava entry Mackenzie, F. T., L. M. Ver, C. Sabine, and M. Lane, C, N, P, S global

(Table 1), a LIP lava extrusion of 500 km3 (approximately the size biogeochemical cycles and modeling of global change, in Interactions of

C, N, P and S Biogeochemical Cycles and Global Change, edited by R.

of the 26.5 ka Oruanui eruption, the largest known wet eruption,

Wolast, F. T. Mackenzie, and L. Chou, pp. 1 61, Springer-Verlag, Berlin,

Wilson [2000]) would be expected to inject 40 Gg of Fe, 20 Gg of

1993.

P, and 0.7 Gg of Pb into the atmosphere. These values represent Martin, J. M., R. M. Gordon, and S. E. Fitzwater, The case for iron, Limnol.

only 0.1%, 0.5%, and 10%, respectively, of the current annual Oceanogr., 36, 1793 1802, 1991.

global nonanthropogenic inputs of these elements to the atmos- Mattox, T. N., and M. T. Mangan, Littoral hydrovolcanic explosions: A

phere [Lantzy and Mackenzie, 1979; Mackenzie et al., 1993]. case study of lava-seawater interaction at Kilauea Volcano, Bull. Volcano.

Geotherm. Res.., 75, 1 17, 1997.

Hence, the data suggest that atmospheric inputs to the global

Olmez, I., D. L. Finnegan, and W. H. Zoller, Iridium emissions from

ecosystem from shallow ocean volcanism are small, although

Kilauea Volcano, J. Geophys. Res., 91(B1), 653 663, 1986.

local/regional effects may be quite pronounced. It remains to be Porter, J. N., and A. D. Clarke, Aerosol size distribution models based on in

determined whether regional plume deposition is a net positive or situ measurements, J. Geophys. Res., 102, 6035 6045, 1997.

negative factor on oceanic primary production. Resing, J. A., The chemistry of lava-seawater interactions at the shoreline of

Kilauea Volcano, Hawaii, Ph.D. dissertation, Univ. Hawaii, Honolulu,

[16] The Kilauea lava flowing into the ocean is degassed of

1997.

most of its CO2, H2, H2O, and sulfur-gases. These gases may act as

Resing, J. A., and F. J. Sansone, The chemistry of lava-seawater interac-

carrier phases for the volatile trace elements [Gerlach, 1989], and

tions: The generation of acidity, Geochim. Cosmochim. Acta, 63, 2183

their absence from the lava entering the ocean at Kilauea most 2198, 1999.

likely results in a lower flux of volatile elements from the lava to Resing, J. A., and F. J. Sansone, The chemistry of lava-seawater interactions

the atmosphere. However, this is not the case for eruptions that II. The elemental signature, Geochim. Cosmochim. Acta, in press.

Rubin, K., Degassing of metals and metalloids from erupting seamount and

occur where volcanoes breach the sea surface. Our estimates,

mid-ocean ridge volcanoes: Observations and predictions, Geochim. Cos-

therefore, are very conservative.

mochim. Acta, 61, 3525 3542, 1997.

Sansone, F. J., and J. A. Resing, Hydrography and geochemistry of sea-

surface hydrothermal plumes resulting from Hawaiian coastal volcanism,

[17] Acknowledgments. We thank David Okita for skillful piloting J. Geophys. Res., 100, 13,555 13,569, 1995.

during sampling; Liangzhong Zhuang for IC analyses; Media Arts, Inc. for Sicks, G., The Kinetics of Silica Dissolution from Volcanic Glass in the

video-recording the sampling operations; Telu Li for discussions of the Marine Environment, M.S. thesis, Univ. Hawaii, Honolulu, 1975.

data; and Jim Kauahikaua for information on the Kilauea lava flow rates Symonds, R. B., W. I. Rose, M. H. Reed, F. E. Lichte, and D. L. Finnegan,

and for wide ranging discussions. Reviews by Yuri Taran and Todd Hinkley Volatilization, transport and sublimation of metallic and non-metallic

resulted in substantial improvements to this paper. This research was elements in high temperature gases at Merapi Volcano, Indonesia, Geo-

supported by the JASON Foundation for Education and the Andrew W. chim. Cosmochim. Acta, 51, 2083 2101, 1987.

Mellon Foundation. SOEST Contrib. No. 5946, PMEL Contrib. No. 2249, Taran, Y. A., J. W. Hedenquist, M. A. Korzhinsky, S. I. Tkachenko, and K. I.

JISAO Contrib. No. 894. Shmulovich, Geochemistry of magmatic gases from Kudryavy volcano,

Iturup, Kuril Islands, Geochim. Cosmochim. Acta, 59, 1749 1761, 1995.

White, R., and D. McKenzie, Magmatism at Rift zones: The generation of

References volcanic continental margins and flood basalts, J. Geophys. Res., 94,

7685 7729, 1989.

Arimoto, R., R. A. Duce, B. J. Ray, and C. K. Unni, Atmospheric trace

Williams, R. M., A model for the dry deposition of particles to natural

elements at Enewetak Atoll: 2. Transport to the ocean by wet and dry

surface water, Atm. Environ., 16, 1933 1938, 1982.

deposition, J. Geophys. Res., 90, 2391 2408, 1985.

Wilson, C. J. N., The 36.5 ka Oranui eruption, New Zealand, 1: Life and

Caldeira, K., and M. R. Rampino, The mid-Cretaceous super plume, carbon

times of the largest known wet eruption (abstract), Eos. Trans. AGU,

dioxide, and global warming, Geophys. Res. Lett., 18, 987 990, 1991.

81(48), Fall Mtg. Suppl., F1312, 2000.

Coffin, M. F., and O. Eldholm, Large igneous provinces: Crustal structure,

dimensions, and external consequences, Rev Geophys., 32, 1 36, 1994.

Duce, R. A., et al, The atmospheric input of trace species to the world

ocean, Global Biogeochem. Cycles, 5, 193 259, 1991.

Gerlach, T. M., Degassing of carbon dioxide from basaltic magma at F. J. Sansone, C. R. Benitez-Nelson, E. H. DeCarlo, S. M. Vink, J. A.

spreading centers: II. Mid-oceanic ridge basalts, J. Volc. Geotherm. Heath, and B. J. Huebert, Dept. of Oceanography, Univ. of Hawaii, 1000

Res., 39, 221 232, 1989. Pope Rd., Honolulu, HI 96822, USA. (abqc7k@r.postjobfree.com)

Govindaraju, K., 1994 Compilation of working values and description for J. A. Resing, Pacific Marine Environmental Laboratory, NOAA, 7600

383 geostandards, Geost. Newslett., 18(spec. issue), 1 158, 1994. Sand Point Way NE, Seattle, WA 98115, USA.

Copyright 2002 by the American Geophysical Union.

Contact this candidate