Journal of Petrology | Volume 41 | Number 3 | Pages 357-362 | 2000
© Oxford University Press 2000
The Merzbacher & Eggler (1984) Geohygrometer: a Cautionary Note on its Suitability for High-K Suites
1GEOFORSCHUNGSZENTRUM POTSDAM, PB 4.2, TELEGRAFENBERG B124, 14473 POTSDAM, GERMANY
2DEPARTMENT OF GEOGRAPHY, GEOLOGY AND ANTHROPOLOGY, INDIANA STATE UNIVERSITY, TERRE HAUTE, IN 47809, USA
Received November 13, 1998; Revised typescript accepted September 2, 1999
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONCALCULATION OF THE PROJECTION...LIMITS OF APPLICABILITY OF...CONCLUSIONSREFERENCES |
|---|
The use of the popular Merzbacher & Eggler (1984, Geology 12, 587–590)
experimental geohygrometer for calcalkaline andesites and dacites is critically evaluated and two pitfalls are found. First, calculation of the correct projection parameters is problematic because two endmember calculation schemes are found in the literature; Baker & Eggler (1983, Journal of Volcanology and Geothermal Research 18, 387–404) and Walker et al. (1979, Contributions to Mineralogy and Petrology 70, 111–125). Although related, these two schemes have crucial differences that can result in very different projection parameters for the plagioclase component. This is most crucial for high-K compositions; H2O contents estimated using the Walker et al. (1979) scheme can be as much as 100% higher than those estimated using the Baker & Eggler (1983) projection. Incorrect projection parameter calculation has led to overestimation of water contents in high-K andesites and dacites from the Central Andes. Second, for medium-K and high-K andesitic–dacitic compositions water contents derived using the Merzbacher & Eggler (1984) geohygrometer deviate considerably from water contents estimated using other methods. Experimental data from the literature, and our studies of water contents inferred from melt inclusions and plagioclase–melt equilibrium for dacites from the Altiplano–Puna Volcanic Complex of the Central Andes indicate that the Merzbacher & Eggler (1984) geohygrometer should not be applied to compositions with K2O >1·9 wt %, as originally calibrated. KEY WORDS: magmatic volatiles; pre-eruptive water; experimental geohygrometer; high-K suites; melt inclusion
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONCALCULATION OF THE PROJECTION...LIMITS OF APPLICABILITY OF...CONCLUSIONSREFERENCES |
|---|
The importance of magmatic water in the evolution and eruption of magmas is well established and it is considered the most important volatile by far in volcanic processes involving silicic magmas. It strongly controls phase equilibria (e.g. Naney, 1983
; Moore & Carmichael, 1998), physical properties of magma and melt (e.g. Shaw, 1972; Eichelberger, 1995), and eruptive style and violence (e.g. Wilson, 1980; Gardner et al., 1995). It is not surprising therefore that considerable research effort has been expended to constrain pre-eruptive volatile contents in magmas. This is generally based on glass and melt inclusion studies (e.g. Newman et al., 1988; Anderson et al., 1989) or more commonly using experimental studies of phase assemblages for different compositions conducted under controlled parameters such as pressure, temperature and water fugacity (Naney, 1983; Merzbacher & Eggler, 1984; Housh & Luhr, 1991). One popular approach has been the Merzbacher & Eggler (1984) geohygrometer, which has been used and discussed widely (e.g. Gerlach et al., 1988; Grunder & Mahood, 1988; Rutherford & Devine, 1988; Luhr, 1992; Feeley & Davidson, 1994). It is based on the equilibrium of hydrous melts with plagioclase, orthopyroxene and clinopyroxene. The stability of these minerals is a function of pressure and volatile content, in particular H2O; CO2 is regarded as being of minor importance. The Merzbacher & Eggler (1984) geohygrometer was calibrated using glass compositions in experimental charges in which hydrous melt was in equilibrium with plagioclase, orthopyroxene and clinopyroxene (in addition with ilmenite or magnetite). Starting materials for experiments were similar to andesites and dacites of Mount St Helens (western USA) and Paricutin (Mexico) (see their paper for experimental details). The major oxide compositions of run product glasses were plotted in a ternary projection involving the endmember compositions plagioclase (pl), orthopyroxene (opx) and quartz + orthoclase (qz + or). Magmatic water abundances can be estimated by the extrapolated orthopyroxene–plagioclase cotectic lines in the projection plane (Fig. 1), contoured for H2O contents between 0 and 4 wt % in the melt. The method yields an estimate for bulk water contents for melts in the limited pressure range between 1 and 5 kbar. The Merzbacher & Eggler (1984) geohygrometer was successfully applied to the Mount St Helens dacites as the original estimates of 4–5 wt % water for bulk magmas at liquidus temperatures agree with more recent studies (Rutherford & Devine, 1988; Housh & Luhr, 1991; Gardner et al., 1995).
|
As part of a continuing study of the Upper Tertiary to Recent silicic volcanism in the Central Volcanic Zone of the Andes, we have explored different methods of constraining pre-eruptive volatile contents of the high-K andesites and dacites that dominate this volcanic province (e.g. de Silva, 1989). In doing so, we evaluated the Merzbacher & Eggler (1984) geohygrometer. It seemed appropriate for our studies, because the andesites contain the relevant mineral assemblage of plagioclase, orthopyroxene, clinopyroxene and magnetite. Although the dacites bear additional hornblende, biotite and quartz, they probably represent liquid compositions that were once in equilibrium with plagioclase and pyroxene. The Merzbacher & Eggler (1984) geohygrometer had also been used in a recent major petrological study in the region, that of Ollagüe Volcano by Feeley & Davidson (1994), who inferred water abundances for high-K Andean andesites and dacites in a range from 3 to 5 wt %.
We encountered two major pitfalls. First, the projection scheme can easily be (and has been) incorrectly calculated because of ambiguous endmember calculations in the literature. This has resulted in overestimation of the H2O contents. Second, the H2O contents inferred from the Merzbacher & Eggler (1984) geohygrometer (when calculated correctly) are much lower than those determined by other methods. Here we discuss these pitfalls and caution against the use of this geohygrometer for high-K suites.
|
CALCULATION OF THE PROJECTION PARAMETERS |
|---|
|
TOPABSTRACTINTRODUCTIONCALCULATION OF THE PROJECTION...LIMITS OF APPLICABILITY OF...CONCLUSIONSREFERENCES |
|---|
Critical to correct usage of the Merzbacher & Eggler (1984)
geohygrometer is the projection scheme. However, this is not straightforward because, in attempting to unravel the endmember calculation, the reader of Merzbacher & Eggler (1984) is referred to the paper by Baker & Eggler (1983), who presented formulae for transformation of major elemental compositions to pl, qtz + or and ol. Baker & Eggler (1983) used a projection scheme that itself is based on the study by Walker et al. (1979), but the formulae given there are slightly different from those of Baker & Eggler (1983). Another potential complication arises because in the caption for fig. 3 of Merzbacher & Eggler (1984) qtz + or is given as an endmember for the projection, whereas in the figure the lower right corner is labelled as qtz–or. Although this may merely be a typographical error, it may be another source of confusion. The correct calculation procedure is:
- calculate the molar FeO and Fe2O3 ratio according to Sack et al. (1980).
- Convert oxide percentages to molar proportions by dividing by molecular weights for oxides.
- Sum the molar proportions to endmembers using the equations [modified from Baker & Eggler (1983)]:
- pl = Al2O3 + Na2O – K2O
- opx = MgO + MnO + FeO – Fe2O3 + Al2O3 – CaO – Na2O – K2O
- qtz + or = SiO2 –5Na2O – 3K2O – CaO – MgO – MnO – FeO + Fe2O3 – Al2O3
- pl = Al2O3 + Na2O – K2O
- Divide each endmember by sum of pl, opx and qtz + or, and normalize to 100.
The formulae for endmember calculation according to Walker et al. (1979) are given in the footnotes to Table 1. The difference in the pl component calculation should be noted. For high-K suites, this difference results in the pl value being considerably enhanced if the Walker et al. (1979) scheme is used. When plotted in the plane of projection this results in a concomitant increase in the H2O content. This is less problematic for low-K to medium-K suites as K2O is relatively small.
|
To illustrate the problem, in Table 1 we have calculated projection parameters using the Walker et al. (1979) and Baker & Eggler (1983) schemes for a range of samples: an example given by Merzbacher & Eggler (1984), an experimental glass composition from a recent study on hydrous phase equilibria in andesites (Moore & Carmichael, 1998), one La Celosa high-K dacite sample from Ollagüe (Feeley & Davidson, 1994), and three samples from our study. This exercise illustrates that the two calculation schemes result in only a minor difference for low-K and medium-K samples (K80-25 and PEM12-4 in Table 1). However, there is a significant difference in the plotting parameters for the high-K samples of La Celosa (e.g. OLA9004; Table 1), other Ollagüe samples (from Feeley & Davidson, 1994), as well as our data from the Central Andes (e.g. PED samples; Table 1).
We have recalculated the Ollagüe data of Feeley & Davidson (1994) and two examples are plotted in Fig. 1a. The recalculation yields only 2% H2O using the Merzbacher & Eggler (1984) geohygrometer compared with the originally inferred water content of 4% from fig. 7 of Feeley & Davidson (1994). The purpose of this example is not to criticize the excellent work carried out at Ollagüe, but to highlight the problems that can result from the ambiguity of the projection parameter calculation. Use of the Walker et al. (1979) scheme leads to a drastic overestimation of the water contents that can be inferred from the Merzbacher & Eggler (1984) geohygrometer for high-K suites.
|
LIMITS OF APPLICABILITY OF THE MERZBACHER & EGGLER (1984) GEOHYGROMETER
|
|---|
|
TOPABSTRACTINTRODUCTIONCALCULATION OF THE PROJECTION...LIMITS OF APPLICABILITY OF...CONCLUSIONSREFERENCES |
|---|
We also evaluated the applicability of the Merzbacher & Eggler (1984)
geohygrometer by expanding the range of compositions to higher K contents compared with the original calibration, in which the K2O contents ranged between 1·4 and 1·9 wt % (all values given normalized to 100 wt % anhydrous). In doing this, we included recently published experimental data on intermediate-K calcalkaline compositions (K2O = 1·7–2·5 wt %, Martel et al., 1998; Moore & Carmichael, 1998) and our unpublished data on Central Andean high-K andesites and dacites (K2O = 2·4–3·6 wt %) from the Pleistocene Purico ignimbrite (Schmitt, 1999). In addition, two experiments on dry banakitic compositions (0 wt % water) with very high K2O contents (5·4 and 6·5 wt %) have been included (Meen, 1987).
In the experimental studies melt H2O contents varied from 0·0 to 5·8 wt % and were determined by various techniques (see original works for details). For our samples, H2O contents in glassy melt inclusions from plagioclase and quartz phenocrysts have been determined by secondary ionization mass spectrometry (SIMS) and IR techniques (Fourier transform IR (FTIR) analysis). Analytical procedures have been described elsewhere (Ihinger et al., 1994). Where compositional data for plagioclase were available, we also applied the Housh & Luhr (1991) plagioclase–melt equilibrium method. This allows estimation of melt H2O contents by two independent calibrations based on the exchange of anorthite and albite component between plagioclase and melt. Although the An and Ab calibrations of the Housh & Luhr (1991) plagioclase–melt equilibrium fail to converge for the andesite and dacites from Purico, the water estimates agree in the range of error with the water contents determined by melt inclusion analysis (Table 1). Our work indicates bulk water contents (phenocrysts taken into account) in a range between 2 (andesite) and 3·6 wt % (rhyodacite).
The primary rule in applying the pseudoternary projection as used in the Merzbacher & Eggler (1984) geohygrometer is that rocks represent liquid compositions in equilibrium with orthopyroxene, clinopyroxene, plagioclase and magnetite–ilmenite. Presence of hornblende is not necessarily preclusive if hornblende is a late crystallizing mineral that did not affect the liquid composition by fractionation. All the natural and experimental compositions we have used represent melt compositions supposed to have equilibrated with the required mineral phases. Quartz, hornblende and biotite present in Purico dacites are late crystallizing phases, and the dacitic magma represents a liquid that was in equilibrium with plagioclase and pyroxene in its early stages (Schmitt, 1999). All pressures in the experiments and inferred for natural samples are below 
3 kbar, therefore pressure should not affect the mineral phase stabilities.
We projected the compositional data of experimental and natural samples onto the Merzbacher & Eggler (1984) ternary plot (Fig. 1b). Contrary to the concordance of water contents estimated for the Mount St Helens dacite using the Merzbacher & Eggler (1984) geohygrometer and other techniques, we find major discrepancies between water contents estimated using the Merzbacher & Eggler (1984) geohygrometer when compared with the other techniques for compositions with higher K contents (Table 1). When projected on the Merzbacher & Eggler (1984) geohygrometer, most of the hydrous experiments with intermediate-K compositions yield considerably lower water contents compared with the values given by Martel et al. (1998) and Moore & Carmichael (1998). Furthermore, the melt water contents inferred by the Merzbacher & Eggler (1984) geohygrometer are considerably lower also for the Central Andean high-K rocks compared with the melt inclusion and plagioclase–melt equilibrium data. An inferred magmatic H2O abundance of <1 wt % seems unrealistically low for the amphibole- and biotite-bearing dacites, suggesting that the Merzbacher & Eggler (1984) geohygrometer estimates are too low.
The experiments with banakitic compositions at 0 wt % water plot far off the 0 contour in Fig. 1b. This indicates that the cotectic lines in the Merzbacher & Eggler (1984) projection do not apply to this shoshonite–banakite suite.
Merzbacher & Eggler (1984) excluded high-K compositions in their compilation of water estimates for silicic volcanics, and those workers may never have intended their hygrometer for use with high-K compositions. Nevertheless, it has been applied for high-K suites (e.g. Feeley & Davidson, 1994). Given the problems we have outlined above, we urge caution in calculation of the projection parameters and we do not recommend the use of the Merzbacher & Eggler (1984) geohygrometer for compositions with K2O > 1·9 wt %, which is beyond the maximum of the original calibration.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONCALCULATION OF THE PROJECTION...LIMITS OF APPLICABILITY OF...CONCLUSIONSREFERENCES |
|---|
The application of the Merzbacher & Eggler (1984)
geohygrometer requires a careful and appropriate projection of the bulk-rock or glass composition. The Baker & Eggler (1983) scheme modified to include opx should be used instead of the Walker et al. (1979) scheme. Use of an inappropriate projection scheme will result in a gross overestimation of water contents in high-K andesites and dacites. However, even if the projection scheme is correct, the Merzbacher & Eggler (1984) geohygrometer may not be suitable to use for estimating water contents of volcanic rocks with K2O >1·9 wt % because these estimates are discordant with water contents estimated by other methods. Unrealistically low water abundances were obtained for high-K magmas from the Central Andes.
|
ACKNOWLEDGEMENTS |
|---|
This study was part of a Ph.D. thesis by A.K.S. under supervision of Rolf Emmermann, University of Giessen. The help, encouragement and invaluable advice of Bob Trumbull and Jan Lindsay during the tenure of this project is gratefully acknowledged. Ion probe work was carried out during a 6 month stay in Tempe, Arizona, and was supported by a DAAD grant. Special thanks go to Rick Hervig at Arizona State University for his hospitality and help in ion probe analysis. A.K.S. would like to express thanks to everyone at the Department of Geology at Arizona State University, in particular Ed Bailey, who conducted rehomogenization experiments on melt inclusions. Help and advice in electron microprobe and FTIR analysis from Dieter Rhede and Monika Koch-Müller (GeoForschungsZentrum Potsdam) was greatly appreciated. This contribution has benefited immensely from reviews by D. Eggler, A. Grunder and G. Wörner. Valuable editorial advice from Marjorie Wilson is acknowledged.
|
FOOTNOTES |
|---|
*Corresponding author. Telephone: +49(0)331 288 1468. Fax: +49(0)331 288 1474. e-mail: axelk{at}gfz-potsdam.de
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONCALCULATION OF THE PROJECTION...LIMITS OF APPLICABILITY OF...CONCLUSIONSREFERENCES |
|---|
Anderson, A. T., Jr, Newman, S., Williams, S. N., Druitt, T. H., Skirius, C. & Stolper, E. (1989). H2O, CO2, Cl, and gas in Plinian and ash-flow Bishop rhyolite. Geology 17, 221–225.
Baker, D. R. & Eggler, D. H. (1983). Fractionation paths of Atka (Aleutians) high-alumina basalts: constraints from phase relations. Journal of Volcanology and Geothermal Research 18, 387–404.[Web of Science]
de Silva S. L. (1989). Geochronology and stratigraphy of the ignimbrites from the 21°30'S to 23°30'S portion of the central Andes of Northern Chile. Journal of Volcanology and Geothermal Research 37, 93–131.[Web of Science]
Eichelberger, J. C. (1995). Silicic volcanism: ascent of viscous magmas from crustal reservoirs. Annual Review of Earth and Planetary Sciences 23, 41–63.[Web of Science]
Feeley, T. C. & Davidson, J. P. (1994). Petrology of calc-alkaline lavas at Volcán Ollagüe and the origin of compositional diversity at central Andean stratovolcanoes. Journal of Petrology 35, 1295–1340.
Gardner, J. E., Rutherford, M., Carey, S. & Sigurdsson, H. (1995). Experimental constraints on pre-eruptive water contents and changing magma storage prior to explosive eruptions of Mount St Helens volcano. Bulletin of Volcanology 57, 1–17.[Web of Science]
Gerlach, D. C., Frey, F. A., Moreno-Roa, H. & Lopez-Escobar, L. (1988). Recent volcanism in the Puyehue–Cordon Caulle region, Southern Andes, Chile (40·5°S): petrogenesis of evolved lavas. Journal of Petrology 29, 333–382.
Grunder, A. L. & Mahood, G. A. (1988). Physical and chemical models of zoned silicic magmas: the Loma Seca Tuff and Calabazos Caldera, Southern Andes. Journal of Petrology 29, 831–867.
Housh, T. B. & Luhr, J. F. (1991). Plagioclase–melt equilibria in hydrous systems. American Mineralogist 76, 477–492.[Abstract]
Ihinger, P. D., Hervig, R. L. & McMillan, P. F. (1994). Analytical methods for volatiles in glasses. In: Carroll, M. R. & Holloway, J. R. (eds) Volatiles in Magmas. Mineralogical Society of America, Reviews in Mineralogy 30, 67–121.[Abstract]
Luhr, J. L. (1992). Slab-derived fluids and partial melting in subduction zones: insights from two contrasting Mexican volcanoes (Colima and Ceboruco). Journal of Volcanology and Geothermal Research 54, 1–18.[Web of Science]
Martel, C., Pichavant, M., Bourdier, J.-L., Traineau, H., Holtz, F. & Scaillet, B. (1998). Magma storage conditions and control of eruption regime in silicic volcanoes: experimental evidence from Mt. Pelée. Earth and Planetary Science Letters 156, 89–99.[Web of Science]
Meen, J. K. (1987). Formation of shoshonites from calcalkaline basalt magmas: geochemical and experimental constraints from the type locality. Contributions to Mineralogy and Petrology 97, 333–351.[Web of Science]
Merzbacher, C. & Eggler, D. H. (1984). A magmatic geohygrometer: application to Mount St. Helens and other dacitic magmas. Geology 12, 587–590.
Moore, G. & Carmichael, I. S. E. (1998). The hydrous phase equilibria (to 3 kbars) of an andesite and basaltic andesite from Western Mexico: constraints on water content and conditions of phenocryst growth. American Mineralogist 83, 36–42.[Abstract]
Moore, G., Vennemann, T. & Carmichael, I. S. E. (1998). An empirical model for the solubility of H2O in magma to 3 kilobars. American Mineralogist 83, 36–42.[Abstract]
Naney, M. T. (1983). Phase equilibria of rock-forming ferromagnesian silicates in granitic systems. American Journal of Science 283, 993–1033.
Newman, S., Epstein, S. & Stolper, E. (1988). Water, carbon dioxide, and hydrogen isotopes in glasses from the ca. 1340 AD eruption of Mono Craters, California: constraints on degassing phenomena and initial volatile content. Journal of Volcanology and Geothermal Research 35, 75–96.[Web of Science]
Rutherford, M. J. & Devine, J. D. (1988). The May 18, 1980, eruption of Mount St. Helens. 3. Stability and chemistry of amphibole in the magma chamber. Journal of Geophysical Research 93(B10), 11949–11959.
Sack, R. O., Carmichael, I. S. E., Rivers, M. & Ghiorso, M. S. (1980). Ferric–ferrous equilibria in natural silicate liquids at 1 bar. Contributions to Mineralogy and Petrology 75, 369–376.
Schmitt, A. K. (1999). Melt generation and magma chamber processes in the Purico complex and implications for ignimbrite formation in the Central Andes. Ph.D. thesis, University of Giessen, Germany.
Shaw, H. R. (1972). Viscosities of magmatic silicate liquids: an empirical method of prediction. American Journal of Science 272, 870 – 893.
Walker, D., Shibata, T. & DeLong, S. E. (1979). Abyssal tholeiites from the Oceanographer fracture zone. (II). Phase equilibria and mixing. Contributions to Mineralogy and Petrology 70, 111–125.[Web of Science]
Wilson, L. (1980). Relationships between pressure, volatile content, and ejecta velocity in three types of volcanic explosions. Journal of Volcanology and Geothermal Research 8, 297–313.[Web of Science]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||