The El Chichón stratospheric cloud: Solid particulates and settling rates

The El Chichón stratospheric cloud: Solid particulates and settling rates

Journal o f Volcanology and Geothermal Research, 23 (1984) 125--146 Elsevier Science Publishers B.V., Amsterdam -- Printed in The Netherlands 125 TH...

2MB Sizes 2 Downloads 4 Views

Journal o f Volcanology and Geothermal Research, 23 (1984) 125--146 Elsevier Science Publishers B.V., Amsterdam -- Printed in The Netherlands

125

THE EL CHICHON STRATOSPHERIC CLOUD: S O L I D PARTICULATES

AND SETTLING RATES

1AN D.R. MACKINNON 1, JAMES L. GOODING :, DAVID S. McKAY 3 and UEL S. CLANTON: 1Microbeam Inc., Mail Code SN4; ~Planetary Materials Branch, Mail Code SN2; 3Experimental Planetology Branch, Mail Code SN4; N A S A Johnson Space Center. Houston, T X 77058 (U.S.A.)

(Received April 26, 1984)

ABSTRACT Mackinnon, I.D.R., Gooding, J.L., McKay, D.S. and Clanton, U.S., 1984. The E1 Chich6n stratospheric cloud: solid particulates and settling rates. In: J.F. Luhr and J.C. Varekamp (Editors), E1 Chich6n Volcano, Chiapas, Mexico. J. Volcanol. Geotherm. Res., 23: 125--146. Sampling of the E1 Chich6n stratospheric cloud in early May and in late July, 1982, showed that a significant proportion of the cloud consisted of solid particles between 2 um and 40 pm size. In addition, many particles may have been part of larger aggregates or clusters that ranged in size from <10 ~m to >50 am. The majority of individual grains were angular aluminosilicate glass shards with various amounts of smaller, adhering particles. Surface features on individual grains include sulfuric acid droplets and larger (0.5 pm to 1 urn) sulfate gel droplets with various amounts of Na, Mg, Ca and Fe. The sulfate gels probably formed by the interaction of sulfur-rich gases and solid particles within the cloud soon after eruption. Ca-sulfate laths may have formed by condensation within the plume during eruption, or alternatively, at a later stage by the reaction of sulfuric acid aerosols with ash fragments within the stratospheric cloud. A Wilson-Huang formulation for the settling rate of individual particles qualitatively agrees with the observed particlesize distribution for a period at least four months after injection of material into the stratosphere. This result emphasizes the importance of particle shape in controlling the settling rate of volcanic ash from the stratosphere.

INTRODUCTION C l i m a t i c e f f e c t s a s s o c i a t e d w i t h t h e 1 9 8 2 e r u p t i o n s o f E1 C h i c h 6 n in C h i a p a s , M e x i c o , are c o n s i d e r e d t o b e a m o n g t h e m o s t s i g n i f i c a n t o f this c e n t u r y ( M a t s o n , 1 9 8 4 , t h i s issue; R a m p i n o a n d Self, 1 9 8 4 ) . S a t e l l i t e d a t a (e.g. T h o m a s e t al., 1 9 8 3 ; M a t s o n , 1 9 8 4 , t h i s issue) s h o w e d t h a t t h e M a r c h 2 8 e r u p t i o n , t h e s e c o n d o f t w o e r u p t i o n s o n A p r i l 3, a n d t h e A p r i l 4 e r u p t i o n all p e n e t r a t e d t h e t r o p o p a u s e at 16 k m a n d i n j e c t e d m a t e r i a l i n t o t h e s t r a t o s p h e r e . T h e A p r i l 4 e r u p t i o n was r e s p o n s i b l e f o r t h e n e a r l y c o h e r e n t globe-circling stratospheric cloud which formed soon after the eruption

0377-0273/84/$03.00

© 1984 Elsevier Science Publishers B.V.

126 (Matson, 1984, this issue). During the initial stages of cloud formation, strongly layered regions of particles were detected between 16 km and 26 km altitude (Coulson et al., 1982). With time, the top of the cloud rose from 26 km to between 33 km and 35 km altitude, leaving at least 17 km of cloud depth over much of the northern hemisphere (Coulsen et al., 1982). In May, the cloud was primarily confined between 10°N and 30°N latitudes, although considerable particle density was observed outside the primary cloud area (Gooding et al., 1983 and references therein). By August, instrumentation on the Solar Mesosphere Explorer Satellite detected the presence of the cloud from 60°S to 40°N (West et al., 1982). Stratospheric sampling of the E1 Chichdn cloud included three missions by a WB-57F aircraft flown over North and Central America in support of the Department of Energy's Project Airstream. During these missions, flown by NASA/Johnson Space Center in May, July, and October, 1982~ inertial impaction collectors provided by the Cosmic Dust Program successfully sampled particulates from the E1 Chichdn cloud. Gooding et al. (1983) summarized collection histories, particle abundances and particle-size distributions of samples from all three missions. In this paper, we report details of morphologies, compositions, and settling characteristics for typical samples from the May and July collections. October samples are not discussed here as they were dominated by liquid droplets (apparently sulfuric acid) and contained fewer ash particles than the May a n d J u l y samples (Gooding et al., 19831. EXPERIMENTAL Sample collection utilized retractable, silicone-oil-coated, inertial impaction collectors m o u n t e d on pylons under each aircraft wing. Impaction collectors were exposed to the stratosphere at designated locations at altitudes between 16.8 km and 19.2 km (55,000--63,000 ft). Collectors were processed both before and after flight in the Cosmic Dust Program's Class-100 clean room. Samples discussed in this paper are from collection surfaces (flags) W7036 and W7038 (collected May 7, 1982, over 30°--49°N latitude) and from W7045 (collected July 21, 1982, over 60--9 ° N latitude). Although significant numbers of sub-micrometer particles were found among collected samples (commonly adhering to larger particles), the lower size threshold of these inertial impaction collectors is estimated to be approximately 2 /am. This estimate is based upon the study by Brownlee et al. (1976) in which similar collectors were flown on a U2 aircraft. Consequently, size distributions for particles ~ 2 t~m in our samples are probably not representative of the stratospheric cloud. In principle, there is no upper size limit to the particle-collection process. Individual particles were retrieved from collection flags using a micromanipulator, mounted on suitable substrates and then cleaned of silicone oil using various hexane-rinsing techniques. Details of these laboratory procedures are given by Clanton et al. (1982) and Gooding et al. (1983).

127 Samples were examined using a Wild M5A optical microscope, a JEOL 35CF scanning electron microscope (SEM), and a JEOL 100CX analytical electron microscope (AEM). Samples were initially uncoated (e.g., with Au/ Pd) when examined with the electron microscopes in order to minimize the possibility of artifacts in SEM images. However, all SEM images in this publication were recorded at 25, 35, or 100 kV on samples with a Au/Pd conductive coating. Elemental analyses were obtained with Princeton Gamma-Tech energy-dispersive spectrometers (EDS) attached to both the SEM and the AEM. Spectra were obtained at 25 kV and 35 kV using the SEM and at 100 kV using the AEM. DUST CHARACTERISTICS Size and distribution

Optical inspection of impaction collectors indicated that the collection surfaces from May were dominated by aggregates or clusters of ash particles set in a dilute matrix of individual particles (Fig. 1). The optical photo in Fig. 1 was taken after the collection surface had been transferred from the WB57F assembly and prior to any processing in the laboratory. The clarity of the image is degraded by the presence of a 25-mm-thick lucite container within which the collection surface is stored. The clusters may have formed by electrostatic attraction and may represent a primary mode of ash occurrence in the E1 Chich6n stratospheric cloud. The clusters appear to be porous aggregates of a few tens to a few hundreds of smaller particles. Clusters were too fragile to be removed intact from the viscous silicone oil and disaggregated into their smaller constituents when samples were removed for analysis. However, the clusters in these stratospheric samples may be similar to the volcanic-ash clusters previously observed in air-fall tephra from the May 18, 1980 eruption of Mount St. Helens (Sorem, 1982). Varekamp et al. (1984, this issue) describe clusters in fresh ashes from E1 Chich6n as well. Two lines of evidence suggest that the ash clusters were primary forms in the stratosphere, rather than artifacts produced during or after sample collection. Firstly, clusters were observed in all May and July samples, although total particle abundances were up to 100 times greater in May than July. If the clusters were the result of collection-induced "clumping," the smaller particle abundances in July should have inhibited cluster development. However, observations of July samples showed that clusters were still abundant, though characteristically smaller than the May samples. Indeed, ash clusters were also observed in October samples that accumulated even smaller total numbers of particles. Secondly, flow of silicone oil on impactor flags apparently destroys clusters as readily as it might create them. There is a tendency for a deployed flag to experience a slow net flow of silicone oil from the center to its edges. However, many clusters have been observed near the centers of all flags examined. In addition, ~ l c m - w i d e areas around the edges of

Fig. 1. Binocular transmitted light micrograph of collection flag W7038 before removal of samples for analysis. The quality of the image is degraded by the presence of a 25-ramthick lucite shield between the collection surface and the microscope objective lens. However, the abundance of unusual particle clusters distributed randomly about the collection surface is readily observed. all flags d i s p l a y e d clear evidence f o r progressive disintegration of clusters into individual shards w i t h increasing p r o x i m i t y t o w a r d flag edges. Occasional f l o w - b a n d e d zones of c o n c e n t r a t e d shards included " a r t i f a c t " clusters o f large size Cup to 500 p m ) a n d e l o n g a t e d fabric e l e m e n t s t h a t can be distinguished f r o m the smaller, m o r e r a n d o m l y s t r u c t u r e d " p r i m a r y " clusters. O n l y such " p r i m a r y " clusters were i n c l u d e d in t h e p r e s e n t d a t a c o m p i l a t i o n . Figure 2 is a c u m u l a t i v e p l o t o f particle sizes for p r i m a r y clusters a n d for individual particles. M e a s u r e m e n t of 484 p r i m a r y clusters n e a r t h e c e n t e r o f W 7 0 3 8 i n d i c a t e d t h a t ~ 5 0 % o f the clusters were larger t h a n 10 p m in size, w h e r e a s clusters ) 3 0 p m a c c o u n t for ~ 10% o f the cluster d i s t r i b u t i o n on t h e c o l l e c t i o n surface (Fig. 2). Clusters o f particles were less a b u n d a n t o n W 7 0 4 5 (flown a l m o s t t h r e e m o n t h s a f t e r W 7 0 3 8 ) a n d w e r e smaller t h a n t h o s e o n W7038. This v a r i a t i o n in a b u n d a n c e and size o f particle clusters b e t w e e n s a m p l i n g missions is p r o b a b l y a result o f decreasing particle d e n s i t y w i t h i n the c l o u d due to particle settling. H o w e v e r , c l o u d h e t e r o g e n e i t y m a y also c o n t r i b u t e to these d i f f e r e n c e s in particle size a n d a b u n d a n c e , as d i f f e r e n t

129 latitudes are represented by the May and July samples described here. Individual ash particles were observed on all collection surfaces flown between April, 1982 and March, 1983, although particle abundance generally decreased with time. An increase in the abundance of liquid droplets was observed among samples from later missions. In the May samples, individual particle sizes overlap cluster sizes to some extent between 10 p m and 20 pm. This overlap is probably due to an inherently greater size range for individual particles in May and to the resolution limit of the optical microscope used for identification of clusters. However, the majority of individal particles ( - 8 5 % ) in May are less than 10 pm in size. The average sizes of individual particles from July were smaller than in May; approximately 80 to 90% of July particles were < 5 pm (Fig. 2; Gooding et al., 1983). 100 99.99

PARTICLE SIZE (~m) 10 5

50

IIIII

I

I

I

I

Illll

I I

I

,I

1

0.5

I

I

0.3 I

I

1982 EL CHICHON

STRATOSPHERIC ASH SAMPLES

99.9 _

(PARTICLE MAXIMUM-DIMENSION DATA)

'7

99

95 I-

z

300 INDIVIDUAL PARTICLES, W7036

9o

uJ

~"

uJ Q. m uJ

80

484

"PRIMARY"

m

6o

z

40

< .J

20

=E uJ > )..

=E

u

10 5 INDIVIDUAL PARTICLES, W7045

2 1

0.1

0.01

I 4

I 5

I 6

I 7

I 8

I 9

I 10

1 11

12

PARTICLE SIZE (PHI UNITS)

Fig. 2. Cumulative plot of particle sizes for clusters and individual particles collected in May (W7036 and W7038) and for individual particles from July (W7045).

130

Morphology and mineralogy Low magnification SEM images of particles indicate that morphologies and mineralogical compositions are similar among all samples examined. The majority of grains are angular glass shards with various amounts of adhering particulate material. Figures 3a and 3b are low-magnification SEM images of particles from clusters in samples W7036 and W7045, respectively. These photos illustrate the wide variety of particle shapes observed in stratospheric cloud samples. Although spherical or hemispherical particles are rare or absent in our samples, other unusual shapes are present, and include angular blocky grains, thin plates, and filaments. Aspect ratios for some thin filaments in May samples are as high as 14 to 1. The thin filaments shown in Fig. 4a are usually composed of Ca and S (see spectrum in Fig. 4) and are interpreted to be calcium sulfate. EDS analysis does not allow a distinction between anhydrite and gypsum. Many Ca-sulfate filaments are electronoptically transparent to 100 kV electrons. A number of larger, silicate laths (aspect ratio 8:1 ) were observed in July samples (Fig. 5a). Typical examples of angular, blocky silicate-ash particles from July samples are shown in Fig. 5b. The presence of small, adhering particles on the silicate lath in Fig. 5a suggests that the lath originated as a fragment of primary ejecta, but underwent substantial mixing with smaller particles within the volcanic plume. The smaller (<0.1 ~m) adhering particles were probably formed by explosive fragmentation of larger crystals with well80% of the ash particles; Fig. 6). These aluminosilicate glass shards were rapidly quenched from melt during the eruption. Minor phases show similar abundance patterns in both May and July samples, although there may be a significantly higher proportion of plagioclase in May samples. Neither amphibole nor silica were found in July samples. Minor phases in both samples include plagioclase, Ca-pyroxene, Ca-sulfate, K-feldspar, and Fe-Ti oxides. In contrast, plagioclase is a major phase in pumices of the three major pyroclastic-fall units of El Chich6n while K-feldspar and quartz are not present (Luhr et al., 1984, this issue). Using the AEM, grains previously identified as an aluminosilicate glass (using the SEM) show electron
13]

Fig. 3. Low-magnification SEM images of typical particles from the stratospheric cloud after transfer from the collection flag. a. Particles from flag W7036, in early May. b. Particles from flag W7045, in late July. Note the absence of spherical or rounded particles.

132

808

C8

EL CHICHO'N ASH [email protected]@KV

COUNTS

f J

[email protected]

ENERGY IKEV)

10.0~

Fig, 4. A n e x a m p l e of t h i n Ca-sulfate laths collected f r o m t h e May flights. SEM image t a k e n at 100 k V ; n o t e t h a t holes in t h e N u c l e o p o r e filter s u b s t r a t e can be imaged t h r o u g h t h e t h i n l a t h ( a r r o w e d ) . A n EDS s p e c t r u m f r o m the t h i n l a t h m a r k e d w i t h a r r o w in t h e SEM image is given b e l o w t h e image.

138

Fig. 5. Examples of grains collected from flag W7045, during July. a. A relatively thick (>0.01 urn) silicate lath with many smaller, adhering particulates, b. Typical examples of euhedral, blocky silicate grains.

sity of surface " b u m p s " or droplets. Bulk X-ray elemental analysis of each grain gave a characteristic glass-shard composition-- (K, Ca, Na) aluminosilicate (e.g., Fig. 8), although detailed elemental analysis using a small beam size (<0.005 pro) in the AEM indicates that there is a higher concentration of sulfur in the region of the surface droplets (Fig. 7). Samples from July also show surface droplets, though with a lower density per individual grain. Figure 8 gives an example of an unusually large (~1 /~m) droplet on a grain with the bulk composition shown in Fig. 9a. An EDS spectrum from the region of the droplet is given in Fig. 8 and shows an enhancement of the sulfur peak relative to that in Fig. 9a. A normalized, subtracted spectrum (i.e., a 90

-

MODAL

80

ANALYSIS

/

EL C H I C H O N ASH M O U N T W7036B

70 60

(179 GRAINS)

-

-

z 50 w u m 40 -

W

w

a.

30 20

-

z

,,,

,~

g

.

-

o

¢r

-I

0

~

p.

W

3 o

10

-

I I

0

[

1

I

-

i



b 90 MODAL

80

ANALYSIS

EL CHICH(~N ASH M O U N T W7045A

(98 GRAINS)

70 6O t-

z so

o ~ 4o

uJ

Q.

wJ

30 ..j

20 10

O

~

O

(D

-r

er

Z

<

G

"J

Fig. 6. Modal analyses for b o t h missions o b t a i n e d b y SEM and EDS analysis: a. Mineralogy for flag W7036, collected in May. b. Mineralogy for flag W7045, in July. Particles s h o w n as " q u a r t z " are actually grains o f an u n i d e n t i f i e d silica mineral.

135

Fig. 7. SEM images of silicate shards from flag W7036 showing unusual surface " b u m p s " or droplets (arrows). EDS analyses of the droplets show high concentrations of sulfur relative to the bulk grain.

] 36

GBS~ ~ I

EL CHICHON ASH 188KV

I

I

COUNTS

~.~

~,~R~Y IK~vl

:~'~

Fig, 8. A high m a g n i f i c a t i o n SEM image of a sulfate gel d r o p l e t o n a typical ash grain. T h e EDS s p e c t r u m is f r o m the region m a r k e d " a " in t h e SEM image. N o t e t h e higher i n t e n s i t y of t h e sulfur peak relative to the analysis s h o w n in Fig. 9a.

137

droplet spectrum minus bulk grain spectrum) in Fig. 9b indicates that the surface droplet contains sulphur and sodium. Similar analyses of other grains in both May and July samples indicate that, in general, all droplets contain S and various amounts of Na, Mg, K, and Ca. The surface features shown in Figs. 7 and 8 are quite stable under electron-beam bombardment, with or without a conductive surface coating. However, the large droplets degrade with time (after approximately 15 min of electron-beam irradiation at 35 kV) by developing a wrinkled appearance and the eventual loss of spherical shape. The rate and nature of electronbeam degradation for the large droplets were considerably different to that observed for sulfuric acid droplets (see discussion, below). The morphology and chemistry of the droplets suggest that they are sulfate gels formed by 8869EL CHICHONRSH I~IKV

a

S! COUNTS

6.8

~cy

{KEV)

lg.8

I.

h 1NI~'

Im



~LJ

6.11



L.L

. . . . . . . . . . . . .

FI.IEI~Y IKEVl

L~L 111.8

Fig. 9. EDS spectra for the grain and droplet shown in Fig. 8. a. Composition of the bulk grain at the region marked " b " in Fig. 8. b. Normalized, subtracted spectrum showing the presence of Na and S in the droplet shown in Fig. 8.

13a; the interaction of sulfur-rich gases and the solid substrate. We expect that gels formed on feldspars should be alkali-rich, whereas those formed on amphiboles, pyroxenes or other minor phases should be Mg-, Ca- or Fe-rich. In general, this is the distribution of cations within gels formed on mineral surfaces. The abundance of sulfur gels in May samples, relative to July samples, may be due to special circumstances that prevailed in the initial volcanic plume. The newly formed plume may have been characterized by a combination of high sulfur dioxide and water vapor concentrations at relatively elevated temperatures which favored rapid reaction between glass or mineral fragments and the volcanic gases. As the plume spread, dilution and cooling of the cloud may have quenched these reactions such that further gel formation was delayed until the production of sulfuric acid aerosol by later photochemical processes (Hofmann and Rosen, 1983). Alternatively, high concentrations of vapors may have formed heterogeneously within particular regions of the cloud (viz., cloud layering, Coulsen et al., 1982). Thus, the relative abundance of sulfur gels in May and July samples may be due to cloud heterogeneity. Another type of surface " b u m p " was observed on samples prepared for AEM study using a less vigorous washing of particles after their removal from the collection surface. These smaller (<0.25 pm), less-stable droplets are shown in Fig. 10 (arrowed). Electron-beam irradiation for less than 5 min (at 100 kV) resulted in loss of the hemispherical morphology and an X-ray signal similar to substrate background. Remnants of the droplets remain, as shown in Fig. 10b, and infrequently indicate trace amounts of cations (e.g., Si) in EDS spectra. These types of surface features (seen on both the AEM substrate and ash grains) appear similar to sulfuric acid droplets described by Rose et al., (1980). Trace amounts of adhering silicate particles may also be associated with sulfuric acid droplets. Sulfuric acid droplets are c o m m o n l y observed in the aerosol layer in the stratosphere (Turco et al., 1982) and are intimately associated with the E1 Chichdn cloud (Oberbeck et al., 1983). Sulfuric acid droplets are expected to survive the gentle washing procedures used in the Cosmic Dust Laboratory because hexane and sulfuric acid are immiscible. Acid-ash reactions

The presence of thin, Ca-sulfate laths within the ash cloud and within ash deposits (Varekamp et al., 1984, this issue) leads to some speculation about the stage of the volcanic eruption at which they may have formed. In order to test whether Ca-sulfate laths could form by reaction with moderate concentrations of acid, samples of Mount St. Helens ash were placed in an open petri dish with a few ml of 0.1 N H2SO4. Mount St. Helens ash is relatively free of sulfur (Fruchter et al., 1980) and thin laths were not observed in ash prior to reaction with H2SO4. An SEM image of Mount St. Helens ash after reaction with 0.1 N H2SO4 for 24 hours is shown in Fig. 11. This view is typical of the acid-reacted ash

139

Fig. 10. E x a m p l e s o f sulfuric acid d r o p l e t s (arrows) o n b o t h ash particles a n d substrate. Figures 10a a n d 1 0 b are o f t h e same region, b u t p h o t o g r a p h e d 5 m i n u t e s apart. Sulfuric acid d r o p l e t s are u n s t a b l e in t h e e l e c t r o n b e a m a f t e r a b o u t 5 m i n u t e s at 100 kV.

I i0

Fig. 11. SEM image of Mount St. Helens ash after reaction with 0.1 N H2SO4 for 24 hours. A high magnification view clearly showing the presence of laths on the surface of ash particles,

and indicates that thin laths have f or m ed by reaction of Mount St. Helens ash with H2SO4. EDS analysis of the lath shown in Fig. 11 gives a high concentration of Ca and S. Thus, it is suggested that m oderat e concentrations of sulfuric acid will react with ash particles (over relatively short periods of time) to form sulfate laths. This mechanism for the growth of non-pyroclastic particles may be important during later stages of cloud formation when the c o n c e n t r a t i o n of sulfate aerosols is high (Rosen and H of m ann, 1983) relative to dust concentration. However, even during the early stages of the E1 Chich6n eruption, sulfate concentrations may have been sufficiently high (Mroz et al., 1983) t hat reaction of pyroclastic particles with sulfuric acid (formed from conversion of SO2 gases) produced some types of thin, lath-like sulfate grains within the stratospheric cloud.

141

DUST SETTLING RATES

Relatively large particles (>5 pm diameter) were abundant in samples of the E1 Chich6n cloud that were collected as late as four to six months after the initial eruption (Gooding et al., 1983). Furthermore, the presence of solid particles in the stratospheric cloud in January, 1983 was recently reported by Hofmann and Rosen (1983). The abundance and relatively large size of particles collected during May and July are somewhat surprising in the light of theoretical models of stratospheric circulation (e.g., Toon and Farlow, 1982; Oberbeck et al., 1983) and deserve some attention. The retention of solid particulate matter in the stratosphere can be attributed to a number of factors which include: (a) slow settling of irregularly shaped particles, or clusters of particles; (b) height and location of the eruptive column (Settle, 1978; Self, 1982); (c) growth of particles with time from condensation nuclei (Toon and Farlow, 1982) or by chemical reaction within the cloud (Hofmann and Rosen, 1983); and (d) turbulent mixing in the stratosphere and vertical mixing across the tropopause (Reiter, 1975; Dewan, 1981). Limitations of time and space restrict the analysis of all these factors associated with particle dynamics in the stratosphere. However, a preliminary assessment of settling-rate models and the influence of particle shape will be given in this section. Settling rates for regularly shaped particles can be calculated using the Stokes equation or modifications of same for particles with sizes less than the mean free path of air molecules (Rosinski and Pierrard, 1964). The terminal settling velocity, T, for a given particle can be calculated using the Stokes equation: T =gd2a ( o -

p)/18~?

(1)

In eq. (1), g is gravitational acceleration, p and V are atmospheric density and viscosity, respectively, e is the particle density, and d a is the diameter of the spherical particle. Although the Stokes equation assumes a spherical particle, our observations show that this assumption is not valid for the E1 Chich6n stratospheric-ash particles (e.g., Figs. 3, 4, 5, and 7). Wilson and Huang (1979) have derived an empirical equation for the settling velocity of an irregularly shaped particle, including terms for the shape factor, the coefficient of drag and Reynolds number for a given particle. The shape factor, F, is calculated for three principal dimensions of a particle, a > b > c, by the equation: F = (b + c ) / 2 a

(2)

The shape factor is included in a complex set of coefficients for the quadratic equation: AT 2 + BT + C = O

(3)

Solution of this quadratic then gives an empirical estimate of the terminal

1-12

settling velocity, T, for a given particle of effective size d a = (a + b + c ) / 3 . The reader is referred to the paper by Wilson and Huang (1979) for furt her details on the calculation of T. Although Wilson and Huang (1979) derived their empirical equation using particles with characteristic sizes between 30 p m and 500 pm, there does not appear to be a fundamental restriction against applying the equation to particles with characteristic sizes between 1 t~m and 100 ~m. Indeed, the assumptions are probably no different than those implicit in the application of Stokes' equation to the same problem. Because we restrict our calculations to particles > 1 pm in size, Cunningham (slip) corrections are not significant for either the Stokes or Wilson-Huang equations. Using both the Stokes and Wilson-Huang equations (assuming F = 0.45), model settling rates were calculated for a particle of density 2.5 g cm -3 and an effective size of d a = 10 t~m. These calculations are summarized in Table 1. A value o f F = 0.45 for the shape factor was based upon SEM observations o f ash particles from both May and July samples. The material density of 2.5 g c m -3 is a reasonable value for glass of rhyolitic-andesitic composition. F r o m Table 1, it is apparent that Wilson-Huang settling rates are approximately a factor of two lower than Stokes settling rates. The difference between the two model settling rates could be further enhanced by assuming a smaller value of F (e.g., F = 0.2 would be appropriate for some shards observed), t h e r e b y decreasing the calculated Wilson-Huang rate while leaving the Stokes rate unchanged. However, the following discussion utilizes results for F = 0.45 and, therefore, represents a conservative estimate of the discrepancy between the Wilson-Huang and Stokes predictions for settling times. TABLE I Terminal settling velocities Altitude (kin)

Tw - h ( c m s 1)

T~ ( c m s 1)

0 10 20 30

0.38 0.53 0.52 0,49

0.74 1.02 1.01 0.94

w h e r e d a = 10 # m , a = 2.5 g c m -3, F = 0.45 a n d g, 0, a n d ~ v a r y w i t h a l t i t u d e . T w _ h a n d T s are s e t t l i n g v e l o c i t i e s c a l c u l a t e d u s i n g t h e W i l s o n a n d H u a n g ( 1 9 7 9 ) a n d S t o k e s e q u a t i o n s , r e s p e c t i v e l y . See t e x t f o r d e t a i l s o f c a l c u l a t i o n s .

The utility of the Wilson-Huang settling-rate formulation can be illustrated by comparing calculated settling rates with those inferred from collected samples. Accordingly, Fig. 12 summarizes model settling time as a function o f b o th size and material density of ash particles. From evidence cited earlier in this paper, the E1 Chichdn volcanic cloud of April 4, 1982 is believed to

143 have r e a c h e d an a l t i t u d e b e t w e e n 26 k m a n d 35 k m . T h u s , 26 k m is a conservative e s t i m a t e o f t h e ceiling f r o m w h i c h ash particles w o u l d fall as t h e y s e t t l e d t o w a r d 18 k m ; the n o m i n a l a l t i t u d e a t w h i c h o u r s a m p l e s were coll e c t e d . T h e W i l s o n - H u a n g e q u a t i o n has b e e n used to p r e d i c t (in Fig. 12) resid e n c e t i m e s o f ash particles at or a b o v e o u r s a m p l e c o l l e c t i o n altitude.

lOOOl , OA-YSAFTERJ \ \ ERUPTION ~ \ \

o,

\

~, \

\

\

,, ! \ \ ,

,oo,o. .

.

_23

,o

\

_

-

.

.

.

\

\

.,6--~.~,~-.~-~\ !

.lg • cm "3

da -

\

2a a*b+c

TW.H,T 20',m

\\ \

\

\

.

M,S,,~ON~ I

I\\

i

,,

\

\\

\

\ \

\\\\

\

\

\

5 10 50 CHARACTERISTIC PARTICLE DIAMETER,da (urn)

100

Fig. 12. A plot of characteristic particle diameter (in micrometers) versus settling time (in days) calculated using the Wilson-Huang method for particles with F = 0.45 and at various particle densities (see text for details). Missions 1 and 2 refer to samples collected in May and July, respectively, and are described in this paper. Using Fig. 12, it is p r e d i c t e d t h a t 33 d a y s a f t e r e r u p t i o n ( c o r r e s p o n d i n g to s a m p l e s W 7 0 3 6 a n d W 7 0 3 8 ) , particles having d a > 7.4 /~m s h o u l d have s e t t l e d b e l o w 18 k m . In c o n t r a s t , t h e S t o k e s m e t h o d implies t h a t o n l y particles smaller t h a n 5.2 p m w o u l d still be at or a b o v e 18 k m at t h a t t i m e . Bec a u s e t h e r e is a r e l a t i o n s h i p b e t w e e n particle m a x i m u m d i m e n s i o n , a, and e f f e c t i v e size, da, a Wilson-Huang value o f d a = 7.4 p m implies t h a t particles w i t h m a x i m u m d i m e n s i o n as large as 12 p m c o u l d still exist at or a b o v e 18 k m at the t i m e W 7 0 3 6 was collected. H o w e v e r , in a s e p a r a t e s t u d y , it was d e t e r m i n e d t h a t 31% o f all particles in the May s a m p l e s h a d effective sizes > 5 p m and t h a t 3% o f t h e particles w e r e > 2 0 # m in e f f e c t i v e size ( G o o d i n g et al., 1 9 8 3 ) . O u r SEM size d a t a f o r 300 individual particles f r o m W 7 0 3 6 gave a m e d i a n value o f 7.5 p m f o r particle m a x i m u m d i m e n s i o n (Fig. 2) a n d c o n f i r m e d t h e p r e s e n c e o f particles u p to 20 p m in m a x i m u m d i m e n s i o n . Figure 12 also s h o w s t h a t 108 d a y s a f t e r e r u p t i o n ( c o r r e s p o n d i n g to s a m p l e

l t4

W7045), the largest ash particle at or above 18 km should have possessed an effective size of d a < 4 t~m, corresponding to a m a x i m u m dimension of a < 7 um. However, the Stokes model would not have predicted the presence of particles larger than 2.9 pm at or above 18 km at that time. The study by Gooding et al. (1983) showed that 18% of all particles on W7045 had effective sizes > 5 pm and 1% were > 1 5 pm. Our SEM size data for 300 particles from W7045 gave a median value for particle m a x i m u m dimension of 2.1 pm (Fig. 2) and confirmed the existence of particles up to 7 ~m in maximum dimension. Although neither the Stokes nor the Wilson-Huang settling times adequately explain the observed longevity of large particles in the E1 Chichdn cloud, the Wilson-Huang predictions are qualitatively more concordant with our observations than are the Stokes predictions. In general, particles collected at 18 km in May and July, 1982 were larger than would be predicted by either the Stokes or Wilson-Huang models. This apparent settling rate " a n o m a l y " might be explained by a number of factors including a higher cloud ceiling (i.e. substantially higher than 26 km, as assumed in our calculations) or turbulent mixing of the cloud in the stratosphere. However, an additional factor that deserves consideration is the presence of low-density ash-particle clusters in the stratosphere. For example, a cluster of ash shards with an aggregate porosity of 60% would have an effective material density of only 1.0 g cm -3. Therefore, 33 days after eruption, clusters up to 18 ~m in effective size {corresponding to 28 pm in maximum dimension for F = 0.45) should still exist at or above 18 km altitude (Fig. 12). A number of unconstrained variables prevents us from further evaluating the importance of cluster formation to explain the observed settling rates of E1 Chichdn ash particles (relative to turbulent mixing etc.). However, we agree with Sorem's (1982) contention that the significance of ash-particle clusters in volcanic clouds may be greater than is currently recognized. Using the Wilson-Huang model for settling rates, independent of cluster formation, particles with < 2 pm effective sizes should still be present in the stratosphere up to a year after the eruption of E1 Chichdn. Recent analyses by Hofmann and Rosen {1983) using large aerosol counters indicate that decay of large particles (r - 1 t~m) had not begun at the end of January, 1983. Some of those larger particles may be due to growth within the cloud after the formation of new particles had ceased (Hofmann and Rosen, 1983). We do not propose that all particles remaining in the cloud after one year would be from the April--May eruptions of E1 Chich6n, but suggest that at least some of the original ash particles could persist above 18 km for one year or longer. SUMMARY Analyses of stratospheric dust collections made in May and July after the March/April, 1982 eruptions of E1 Chich6n indicate that the particulate material consisted of aluminosilicate glass shards with minor amounts of

145 feldspars, p y r o x e n e s , amphiboles, and Fe-Ti oxides. O t h e r t y p e s of solid grains m a y include r e a c t i o n p r o d u c t s f o r m e d within the p l u m e or stratospheric cloud, and small f r a g m e n t e d grains adhering to larger particles. Many particles m a y have o c c u r r e d as clusters or aggregates o f individual grains within the s t r a t o s p h e r i c cloud. T h e presence of small, sulfate gels o n the surfaces o f m a n y ash particles indicates t h a t chemical reactions b e t w e e n sulfurrich gases and solid particles p r o b a b l y o c c u r r e d w i t h i n the cloud. Calculated particle-settling rates using the Wilson and H u a n g ( 1 9 7 9 ) f o r m u l a agree m o r e closely with observed particle sizes t h r o u g h t i m e t h a n rates calculated with the S t o k e s e q u a t i o n . This result indicates t h a t particle shape is an i m p o r t a n t f a c t o r in c o n t r o l l i n g the settling o f volcanic ash f r o m a stratospheric cloud. ACKNOWLEDGEMENTS T e c h n i c a l assistance within the Curatorial Facility was p r o v i d e d b y B e t t y Gabel and J a c k Warren, and within the E l e c t r o n M i c r o s c o p e Facility at JSC by G e o r g A n n Nace and Fleur Rietmeijer. S u p p o r t f r o m m e m b e r s o f the N A S A - 9 2 8 flight and g r o u n d crews is also gratefully a c k n o w l e d g e d . Useful discussion and e n c o u r a g e m e n t were p r o v i d e d b y C h u c k Wood, D o u g Blanchard and Frans Rietmeijer. P r e s e n t a t i o n o f this w o r k has b e n e f i t t e d greatly f r o m the efforts o f t w o a n o n y m o u s reviewers and f r o m suggestions b y J o o p V a r e k a m p and Jim Luhr. S u p p o r t for this w o r k was p r o v i d e d b y the NASA Office o f Space Science and Applications t h r o u g h the Cosmic Dust Program and b y R T O P 152-02-40-26 to DSM.

REFERENCES Brownlee, D.A., Ferry, G.V. and Tomandl, D., 1976. Stratospheric aluminum oxide. Science, 191: 1270--1271. Clanton, U.S., Dardano, C.B., Gabel, E.M., Gooding, J.L., Isaacs, A.M., Mackinnon, I.D.R., McKay, D.S., Nace, G.A. and Warren, J.L., 1982. Cosmic Dust Catalog, Vol. 1. Curatorial Branch Publication No. 59, NASA Johnson Space Center. Coulson, K.E., Defoor, T.E. and Deluisi, J., 1982. Lidar and optical polarization measurements of stratospheric cloud in Hawaii. EOS, Trans. Am. Geophys. Union, 63: 897. Dewan, E.M., 1981. Turbulent vertical transport due to thin intermittent mixing layers in the stratosphere and other stable fluids. Science, 211: 1041--1042. Fruchter, J.S., Robertson, D.E., Evans, J.C., Olsen, K.B., Lepel, E.A., Laul, J.C., Abel, K.H., Sanders, R.W., Jackson, P.O., Wogman, N.S., Perkins, R.W., Van Tuyl, H.H., Beauchamp, R.H., Shade, J.W., Daniel, J.L., Erikson, R.L., Sehmel, G.A., Lee, R.N., Robinson, A.V., Moss, O.R., Briant, J.K. and Cannon, W.C., 1980. Mount St. Helens ash from the 18 May 1980 eruption: Chemical, physical, mineralogical and biological properties. Science, 209: 1116--1125. Gooding, J.L., Clanton, U.S., Gabel, E.M. and Warren, J.L., 1983. E1 Chich6n volcanic ash in the stratosphere: Particle abundances and size distributions after the 1982 eruption. Geophys. Res. Lett., 10: 1033--1036. Hofmann, D.J. and Rosen, J.M., 1983. Baloonborne particle counter observations of the E1 Chich6n aerosol layers in the 0.01--1.8 m radius range. Int. Assoc. Meteorol. Atmos. Phys., 571.

146 Luhr, J.F., Carmichael, I.S.E. and Varekamp, J.C., 1984. The 1982 eruptions of El Chichdn Volcano, Chiapas, Mexico: mineralogy and petrology of the anhydrite-bearing pumices. J. Volcanol. Geotherm. Res., 23: 69--108. Matson, M., 1984. The 1982 El Chich6n Volcano eruptions: a satellite perspective. J. Volcanol. Geotherm. Res., 23: 1---10. Mroz, E.J., Sedtacek, W.A. and Mason, A.S., 1983. In situ measurements of stratospheric sulfate following the 1982 E1 ChichOn eruptions. Int. Assoc. Meteorol. Atmos. Phys., 572. Oberbeck, V.R., Danielson, E.F., Snetsinger, K.G. and Ferry, G.V., 1983. Effect of the eruption of E1 Chich6n on stratospheric aerosol size and composition. Geophys. Res. Lett., 10: 1021--1024. Oskarsson, N., 1980. The interaction between volcanic gases and tephra: Fluorine adhering to tephra of the 1970 Hekla eruption. J. Volcanol. Geotherm. Res., 8: 251--266. Rampino, M.R. and Self, S,, 1984. The atmospheric effects of El Chich6n. Sci. Am., 250: 48 57. Reiter, E.R., 1975. Stratospheric-tropospheric exchange processes. Rev. Geophys. Space Phys., 13: 459--474. Rose, W.I., Chuan, R.L., Cadle, R.D. and Woods, D.C., 1980. Small particles in volcanic eruption clouds. Am. J. Sci., 280: 671--696. Rosen, J.M. and Hofmann, D.J., 1983. Measurement of the sulfuric acid weight percent in the stratospheric aerosol from the E1 Chich6n eruption. Int. Assoc. Meteorol. Atmos. Phys., 576. Rosinski, J. and Pierrard, J.M., 1964. Condensation produets of meteor vapors and their connection with noetilueent clouds and rainfall anomalies. J. Atmos. Terr. Phys., 26: 51--66. Self, S., 1982. A comparison of the 1982 E1 Chich6n eruption with the historic reeord of explosive voleanie activity. EOS, Trans. Am. Geophys. Union, 6 3 : 1 1 2 7 . Settle, M., 1978. Volcanic eruption clouds and the thermal power output of explosive eruptions. J. Volcanol. Geotherm. Res., 3: 309--324. Sorem, R.K., 1982. Volcanic ash clusters: Tephra rafts and scavengers. J. Volcanol. Geotherm. Res., 13: 63--71. Thomas, G.E., Jakosky, B.M., West, R.A. and Sanders, R.A., 1983. Satellite limb-scanning thermal infrared observations of the E1 Chiehdn stratospheric aerosol: first results. Geophys. Res. Lett., 10: 9 9 7 - 1 0 0 0 . Toon, O.B. and Farlow, N.H., 1982. Particles above the tropopause: Measurements and models of stratospheric aerosols, meteoritic debris, nacreous clouds and noetilucent clouds. Annu. Rev. Earth Planet. Sei., 9: 19--58. Turco, R.P., Whitten, R.C. and Toon, O.B., 1982. Stratospheric aerosols: Observation and theory. Rev. Geophys. Space Phys., 20: 233--279. Varekamp, J.C., Luhr, J.F. and Prestegaard, K.L., 1984. The 1982 eruptions of El Chich6n Volcano (Chiapas, Mexico): charaeter of the eruptions, ash-fall deposits, and gasphase. J. Voleanol. Geotherm. Res., 23: 39--68. West:, R.A., Thomas, R.J., Thomas, G.E. and Barth, C.A., 1982. Satellite observations of scattered sunlight from the E1 Chieh6n aerosol cloud. EOS, Trans. Am. Geophys. Union, 63: 898. Wilson, L. and Huang, J.C., 1979. The influence of shape on the atmospllerie settling velocity of volcanic ash particles. Earth Ptane~. Sci. Lett., 4 4 : 3 1 1 - 3 2 4 .