Journal of Volcanology and Geothermal Research, 21 (1984) 25--44 25

Elsevier Science Publishers B.V., Amsterdam - - P r i n t e d in The Netherlands

GEOCHEMISTRY A N D EVOLUTION OF THE FUEGO VOLCANIC

COMPLEX, GUATEMALA

CRAIG A. CHESNER and WILLIAM I. ROSE, Jr.

Michigan Technological University, Houghton, MI 49931 (U.S.A.)
(Received August 11, 1982; revised and accepted July 2, 1983)

ABSTRACT

Chesner, C.A. and Rose, W.I., Jr., 1984. Geochemistry and evolution of the Fuego vol-
canic complex, Guatemala. J. Volcanol. Geotherm. Res., 21: 25--44.

Four closely spaced vents along a fissure make up the Fuego and Acatenango volcanic
centers in western Guatemala. The Fuego complex is composed of the Fuego and Meseta
vents, but historic activity has consisted exclusively of high-A1203 basalts from the Fuego
vent. The Meseta vent is inactive and deeply exposed. Prehistoric lavas from Fuego and
Meseta are generally more silicic than historic Fuego lavas, but all the rocks form a single
coherent geochemical variation pattern. Major element chemistry of these rocks is consis-
tent with plagioclase, olivine, augite, and magnetite (POAM) fractionating from high-
A1203 basalt. Separate batches of magma can be recognized from trace-element data
throughout the history of the Fuego complex. This suggests that closed-system, POAM
fractionation of distinct magma bodies occurs at Fuego. Trace-element data requires that
deep fractionation of olivine, clinopyroxene, and perhaps magnetite from primary olivine
tholeiite occurs before arrival of new magma into the shallow (8--16 km) magma chamber
at Fuego. Migration of activity from Meseta to Fuego along the fissure is correlated with
the change towards more mafic compositions at Fuego. The shift of the vents may have
resulted in shorter repose periods and less time for fractionation before eruption. A mini-
mum age of 17,000 years was required to build the Fuego complex.
The andesitic rocks from the adjacent, larger composite volcanoes of Acatenango and
Agua have higher incompatible element concentrations, different incompatible element
ratios, and lower CaO, Na20, and A120 ~ contents than Fuego's lavas. We believe the mag-
matic evolution of Acatenango and Agua is much more complex than Fuego.

INTRODUCTION

Fuego volcano is located within the WNW-trending central Guatemalan

volcanic chain of Central America at 14°29'N, 90°53tW (Fig. 1). A detailed
description of Fuego's regional and tectonic setting is given by Rose et al.
(1978). Fuego is the most southern of four vents which are aligned in a
north--south direction. These vents from north to south are Yepocapa
(3880 m), Acatenango (3976 m), Meseta {3600 m), and Fuego (3763 m)
(Fig. 2). Yepocapa and Acatenango appear to be related vents in the Aca-
tenango complex, whereas the Fuego complex is composed of the Fuego and
Meseta vents. A 3000-m saddle separates the two volcanic complexes. Four-

0377-0273/84/$03.00 © 1984 Elsevier Science Publishers B.V.

26
I
92°W 90" ,,.../ 88 °
_ i6ON/ . . . . . . ' ~ CaribbeanSea
//

,, , GUATEMALA
'7 ° ~;0oo, ,,'J HONDURAS
I O ~ &CATENANGO
° ° ~(:/~'AGUA .(/
FUEGO 0 ~-"~-.. L
-14' ~ .-; ~ALVAE)'131~.-,-
0 0 ~

Pacific Oco ~ , ,

Fig. 1. Location map of the Guatemalan volcanic chain.

0 V'z I 2 3 4 5km V
Fig. 2. Topographic map of the Fuego and Acatenango volcanic complexes. Locations of
Quebrada Barranca Honda and Quebrada Playa Trinidad are indicated. (Taken from Alo-
tenango and Chimaltenango 1:50,000 quadrangle maps prepared by the National Geo-
graphical Institute of Guatemala).
 27

teen kilometers to the east lies Agua volcano, a large composite volcano
which has not erupted in historic time. Although Fuego has historically been
in a very active state, Acatenango has been rather quiet, erupting only thrice
in recent times (Simkin et al., 1981).
In 1974 Fuego recorded its most voluminous eruption since at least 1932
(Deger, 1932). During the 1974 eruption 0.1 km 3 of high-A120~ basalt was
erupted from a zoned magma body which developed by fractionation in a
matter of months (Rose et al., 1978). Magma is inferred to have risen into a
dike-like conduit from a deeper magma body as a result of tidal influences.
Anderson (in press) has studied zoning in plagioclase from this eruption and
also concludes that tidal influences affected the movement of magma in a
dike-like feeder system. Research conducted by Rose et al. (1982) on the gas
budget of the 1974 eruption indicates that a magma body of about 5 times
the volume of erupted material was present beneath Fuego before the erup-
tion. This estimate is derived from calculations utilizing measured degassing
of S during the major eruptive event and S and Cl fluxes measured during
passive degassing following the eruption. Martin and Rose {1981)expanded
upon these studies by evaluating the historic eruptive activity of Fuego.
Their research revealed that since 1932 lavas have become more mafic, repose
periods have decreased, and eruption rates have increased. Clusters of his-
toric eruptions last 20--70 years and occur at 80--170 year intervals. The
latter was suggested to represent the interval of arrival for batches of magma
into a shallow magma chamber.
Despite the various studies already completed on Fuego, none has dealt
with the prehistoric activity of the volcanic center. The purpose of this paper
is to expand the knowledge of Fuego by characterizing its prehistoric activ-
ity. The overall evolution of the Fuego complex is analyzed. Finally, a dis-
cussion of magma evolution patterns at Fuego, Acatenango, and Agua is pre-
sented.

SAMPLING PROCEDURE

A deep exposure in the Meseta vent faces east, revealing m a n y sequential

lava flows. This exposure provides an ideal situation to sample and study the
prehistoric lavas of the Fuego complex. Due to excessive steepness of this
exposure only 14 lava flows were accessible for sampling. These 14 flows are
broken into two separate groups. A sequence of four lava flows which repre-
sents the youngest available extrusions from Meseta was sampled from the
east face of Meseta along the ridge to Fuego (Fig. 3). Ten sequential lava
flows which are stratigraphically older than the previously mentioned four
were also sampled. These flows are located on the southern face of the east--
west-trending Meseta ridge (Fig. 3). Judging from the topographic overlap of
Meseta on Acatenango's flank, this group of samples probably represents
about 10% of Meseta's history.
Interbedded mudflows (Hebberger et al., 1977) and block-and-ash flows
28

FSl
-

'aa~es,

~'[ ' Fuego Acalenango

ddL:NV " "/ ~.

J ..

j/f ~ J N
t J~-JJ

Fig. 3. View l o o k i n g west i n t o t h e e x p o s e d p o r t i o n o f t h e Meseta cone. Lava flows w h i c h

were s a m p l e d for this s t u d y are i n d i c a t e d . L o c a t i o n of this e x p o s u r e is s h o w n o n t h e N--S
cross-sectional view o f t h e f o u r vents.

(Davies et al., 1978) which originate from the steep slopes of Fuego and
Meseta are found in the lower stretches of barrancas which drain the Fuego
complex. Samples of rocks were collected from stream-bed exposures of
these deposits which contain diverse populations of the older lavas of Fuego
and Meseta.
The numerous inaccessible lava flows of Meseta along with some Fuego
flows outcrop in a large amphitheater which is drained almost exclusively
by Quebrada Barranca Honda (Fig. 2). A large number of samples were col-
lected from the lower reaches of this barranca. Thirty-two samples repre-
senting the full hand specimen diversity were then chosen for further study.
In Quebrada Playa Trinidad which drains the southern flank of Fuego (Fig.2)
a variety of 13 samples were collected. In all, 60 samples from the field were
selected for laboratory study. These represent the earlier activity of Fuego
and Meseta. Together with the 100+ samples already studied from Fuego's
historic activity, a comprehensive sampling is now available for synthesis.

DATA AND ANALYTICAL METHODS

Thin sections of 60 samples were examined for mineralogy and textural

relations. Modal analyses of phenocrysts phases (>0.1 mm) based on 1000
counts were determined for the 14 stratigraphic samples from Meseta. Of the
60 samples, 49 were chosen for analysis of 9 major elements and 13 trace
elements. Analyses were obtained by X-ray fluorescence techniques using a
Phillips Model 1400 AXS X-ray Spectrometer. Samples were ratioed to a
standard (Leake et al., 1970). F o r t y international geochemical reference
standards were used to calibrate the runs (Flanagan, 1973; Myers et al.,
1976). Precision for major elements by this technique is generally below 1%
whereas trace-element precision is usually below 10%.
 29

PETROGRAPHY

The 60 thin sections examined include basalts, basaltic andesites, and an-
desites. In general all eruptive types are porphyritic with complexly zoned,
inclusion-rich plagioclase being the most abundant phenocryst phase (Ander-
son, in press). Mafic minerals in Fuego rocks c o m m o n l y occur in glomero-
porphyritic clumps. In many instances olivines were noted to have thin clino-
pyroxene reaction rims. An inverse relation between abundance of orthopy-
roxene and olivine/clinopyroxene is obvious throughout the suite. Modal
analyses of the 14 stratigraphic samples show this relation quite well. The re-
sults from these analyses are summarized in Fig. 4. Magnetite always occurs
as a phenocryst and groundmass phase. Apatite is the only common acces-
sory mineral. A diversity of groundmass textures occur in Fuego's rocks.
Most of these textures range from hypocrystalline to holocrystalline varie-
ties. Mineral compositions for phases in the 1974 eruption of Fuego were:
plagioclase (Angs_ so), olivine (Fo76-66), augite (En4sFsi3Wo4~), and magnetite
(4.3--15% TiO~) (Rose et al., 1978). These mineral compositions probably
reflect those in older Fuego rocks because all of Fuego's rocks are similar in
texture and bulk chemistry.

100

"- _

Batch 4 Batch 3 ~,-ff~' P)ag ~ " ~

Batch 2 Batch 1

10

PHASE %

/
/
0.1 i i J i i i ~ i
14 13 12 11 10 9 8 7 6 5 4 3 2 1
oldest SEQUENCE youngest

Fig. 4, S u m m a r y o f m o d a l a n a l y s e s d e t e r m i n e d f o r t h e 14 s e q u e n t i a l M e s e t a lava flows.

Sample n u m b e r s refer to " F S " s a m p l e s . E r r o r bars derived f r o m V a n Der Plas a n d T o b i
(1965). A b b r e v i a t i o n s are d e f i n e d as follows: g m = g r o u n d m a s s , plag = plagioclase, ol =
olivine, opx = o r t h o p y r o x e n e , cpx = clinopyroxene, and mag = magnetite.
30

Table I summarizes the mineralogy and textures exhibited by the rock

types examined. A few notable exceptions should be mentioned. Two sam-
ples of basaltic andesite contained a faintly pleochroic biotite, probably
phlogopite, occurring in mafic clots consisting of olivine, o r t h o p y r o x e n e ,
and magnetite. Rose et al. (1978) reports the presence of amphibole as in-
clusions in plagioclase from the 1974 eruption of Fuego. These occurrences
are consistent with T ype III, H20 rich rocks of Sakuyama's (in press) classi-
fication o f volcanic arcs according to mineralogy. Also, a pink to brown pleo-
chroic apatite of micro-phenocryst size was observed in m any of the ande-
sites.
TABLE I

Summary of petrography for different rock types in the Fuego suite

Basalt Basaltic Andesite Dike rocks Gabbroic

andesite inclusion
Typical mode plag-33% plag-36% plag-30% plag-65% plag-68%
ol-2.1% ol-2.8% opx-3.6% opx-20% opx-15%
cpx-l% cpx-l.6% mag-0.6% mag-10% mag-10%
mag-1.1% mag-0.7% cpx-0.2% cpx-5% cpx-7%
gm-62.8% opx-2.4% gm-65.6%
gm-56.5%
Gm. texture intersertal intersertal pilotaxitic intergranular
hyaloophitic hyaloophitic hyalopilitic
intergranular pilotaxitic
Gin. rain. plag plag plag plag
mag cpx opx opx
cpx mag cpx mag
-+glass _+glass -+glass

CHEMISTRY

In order to create a more complete data set previous analyses of Fuego

were also used. These analyses are mostly of rocks erupted in the 1970's
(Rose et al., 1973, 1978; Martin, 1979). Analyses of lavas from the eruptions
of 1932 (Deger, 1932}, 1962 (Pushkar, 1966), and 1966 (Stoiber and Rose,
1970) were also used. Nearly all of these rocks are basaltic in composition.
The older rocks of Fuego are generally m ore silicic than the recent eruption
products. This relation is shown in the Harker variation diagrams in Fig. 5.
Fuego eruptives range between calc-alkaline basalts and andesites when
Irvine and Baragar's (1971) classification is used. Table II gives representa-
tive analyses showing the range of chemical variation at Fuego from basalt
to andesite. When plotted against SIO2, selected trace elements show distinct
trends (Fig. 6). Incompatible trends are observed for Zr, Rb, and Ba, where-
as compatible trends are exhibited by Sc and V. St ront i um appears to have
no correlations with SIO2.
 31

~4

0 ~ O'~ ~ - O~ ~o e'~ ~0 ~ - I

c~
<
e.,

o
t.-,

~ ~ ~ 0
o ~ ~ ~ 0
<
d

e~ ~ 4 N ~ 4 N d d d e

o~oS~N~4~dd
~ V

~ ° ~ ° ~ 1 ~

0 •

~ V ~ ~ ~ V L~

~00 0 ~
32

i i i J i i i , i i i L i L 1

io!
FeO% .- "~..
6:

2.0

K20%
T.O-

0.5
~ .~.;.:'... ÷

47 55 65
SiO 2 %

Fig. 5. S e l e c t e d Harker variation diagrams for the F u e g o rock suite• Circled fields indicate
analyses o f Fuego's historical eruptions• Triangles are w e i g h t e d average parent c o m p o s i -
tion o f the 1 9 7 4 eruption• Bars give a measure (+_ l o ) o f the precision for each element.
L I I I I I I I I L I

50- - 800
Sc •_-..
I0 ~
"•-'.•...:-" . V
200-
• • .. . "..•.• 50
Zr
~00 - -50

,.,.~:::.,r.
Rb
iO
7O0

Ba 800

, .'. ;"
Sr
• • •,.'~ .." OQ
4OO
.::~-
600
, i i , i , , , , , , , , , , , , , , , , ,

50 56 62 50 56 62
SiO 2 SiO 2
Fig. 6. Selected trace e l e m e n t variation diagrams p l o t t e d against SiO 2 for the F u e g o suite.
Bars give a measure (_+ l o ) o f the precision for each element•

Figure 7 represents selected major and trace elements plotted against

stratigraphic position of the 14 sequential Meseta lava flows. Although no
smooth trends can be observed, groupings of samples appear to occur• Sam-
ples 1--4, 5--6, 7--9, and 10--14 form distinct groups.
In order to compare the chemistry of the Fuego complex with that of the
Acatenango complex and Agua, several rock samples from these volcanoes
were also studied• These samples generally represent the youngest material
erupted from these volcanoes. Most are summit lavas and a few represent
flank lava flows. Some representative analyses of these rocks are given in
Table II. Figure 8 shows a few composite variation diagrams of Fuego, Aca-
tenango, and Agua. Noticeable differences occur among the two complexes
and Agua. Acatenango has higher K20 and Rb concentrations than those of
Fuego at similar SiO2 contents• Agua has even higher concentrations of these
elements than Acatenango. Furthermore, the K/Rb ratios of the Acatenango
and Agua lavas are distinctly lower than for Fuego (Table II). Izalco, a young
 33

~ l l ' l n l l l l l l l f I I I I I I Ill I I ~
I !•ol
125'
• o. , o ~ •. • -9
K20 % I
• • l • o• CaO %
0.75" o- -7
125 k

Zn •O il50zr
0 0 0 • • 0 I •
-100
75
• O •••
2oo: e•

5OO
V
I00- •
• ee
• •
eo
q
4 0 0 Ba
-- I
i
i700 • ••e o •
Mn i800 Sr
1500 / 700
.Q •, ,• •1 ~ , ,., ,., , , . . . . ', w P', ,

14
SEQUENCE SEQUENCE
Fig. 7. Selected major and trace elements plotted against stratigraphy for 14 sequential
flows from Meseta. This sequence goes from the oldest flow (14) to youngest (1).

50
Rb 3 2, A
20 ...~... - .:.,~.:...':.

2 Q c,.-~'% ,~

1.0" . ":,~-:.'. "."

.P
'1 n r: i I ~ , , ~ i i i 1

47 55 64
Si02 %

Fig. 8. Selected variation diagrams for Fuego, Aeatenango and Agua. Fuego rocks indi-
cated by "+", Aeatenango samples shown with "A", and Agua rocks denoted by "G".

cinder c o n e in E1 Salvador and Santa Ana, a large c o m p o s i t e c o n e adjacent

to Izalco s h o w similar relations: i n c o m p a t i b l e elements at Santa Ana are
higher than those at Izalco (Carr and Pontier, 1 9 8 1 ) . Similar major element
trends are n o t e d at A t i t l i n and Tolim~n in Guatemala (Rose et al., 1 9 8 0 ) .
These v o l c a n o e s also lie adjacent to one another along a n o r t h - - s o u t h trend.
Atitl~in is the southern vent, as is F u e g o , and has higher Na:O, C a • , A1203
and lower K 2 0 than its neighbor Tolimfin, which has similar trends to Aca-
tenango.

INTERPRETATIONS

Fractional crystallization and magma batches

In this section, evidence which suggests that distinct batches of magma

can be recognized in Fuego's early lava sequences is presented.
34

Least-squares modelling of major elements in fractional crystallization

In o r d e r to test fractional crystallization at the F u e g o c o m p l e x , least-
squares mixing (Wright and D o h e r t y , 1 9 7 0 ) and Rayleigh (Shaw, 1 9 7 0 ) cal-
culations were p e r f o r m e d on the 14 stratigraphically related samples f r o m
the Meseta vent. The weighted average for the w h o l e e r u p t i o n of F u e g o in
1974, as d e t e r m i n e d b y Rose et al. ( 1 9 7 8 ) p r o v e d to be the best p a r e n t c o m -
p o s i t i o n for the Meseta lavas. Its bulk c o m p o s i t i o n is p l o t t e d in Fig. 5 w h i c h
shows t h a t it falls within the g e o c h e m i c a l variation p a t t e r n of F u e g o and is
t h e r e f o r e a reasonable parent. Based on p e t r o g r a p h y the p r o b a b l e fraction-
ating phases were plagioclase, olivine, c l i n o p y r o x e n e , and magnetite. Gill
( 1 9 8 1 ) designates this t y p e o f f r a c t i o n a t i o n as POAM and regards it as the
m o s t viable crystal f r a c t i o n a t i o n m o d e l for andesites. The chemical c o m p o -
sitions o f these phases were o b t a i n e d b y m i c r o p r o b e analysis on the 1 9 7 4
e r u p t i o n p r o d u c t s of Fuego. Modes o f this e r u p t i o n range as follows:
g r o u n d m a s s : 40--81%, plagioclase: 15--51%, olivine: 3.6--12.6%, p y r o x e n e :
0.8--1.6%, and m a g n e t i t e : 2.6--3.9% (Rose et al., 1978). These m o d e s are
similar to t h o s e of the older F u e g o rocks (Table I). Results f r o m the mixing
calculations yield small errors w h i c h is consistent with a f r a c t i o n a t i o n m o d e l
for p r o d u c i n g these lavas.

Rayleigh calculations of trace-element fractional crystallization

The observed t r a c e - e l e m e n t c o n c e n t r a t i o n s in Meseta lavas c o u l d n o t
be d u p l i c a t e d in Rayleigh f r a c t i o n a t i o n calculations using the trace-ele-
m e n t c o n c e n t r a t i o n s in the 1 9 7 4 F u e g o lava as a p a r e n t (Table III). If
the t r a c e - e l e m e n t c o n c e n t r a t i o n s are allowed to vary, g o o d a g r e e m e n t
is f o u n d , b u t no single p a r e n t c o u l d p r o d u c e all of the observed d a u g h t e r
c o n c e n t r a t i o n s . T h e r e f o r e , an alternate h y p o t h e s i s involving multiple p a r e n t
m a g m a s with distinct t r a c e - e l e m e n t c o n c e n t r a t i o n s was pursued.

TABLE III

Illustration of the inadequacy of the Fuego 1974 parent in Rayleigh calculations con-
trasted with the good agreement between observed and calculated concentrations when
an individual parent composition is chosen for each batch (values in ppm. except RO)

Sr Zr Y Rb Ba Ni Sc V %K~O

Observed (FS-13) 773 101 20 18 365 17 20 173 0.90

Calculated with 599 98 -- 12 510 15 16 -- 0.90
Fuego '74 parent
Calculated with 777 98 18 16 366 16 19 173 0.90
new parent
Fuego '74 parent 540 84.7 -- 10.5 432 28.7 25.5 -- 0.77
New parent 700 84 16 14 310 30 30 400 0.77
(Batch 4)
 35

8OO
!
p p m Sr
a~. ~ ~
70C

highest %Ptag'Iract"
60G

i i
10 20 30

Fig. 9. Diagram indicating a field (diagonal lines) in which samples in a magma batch with
a Sr parent of 605 ppm should fall. The field is defined by lines which correspond to the
extreme cases of plagioclase fractionation at a distribution coefficient of 1.5. Points rep-
resent concentrations of Sr in another magma batch. Total fractionation (%) refers to the
sum of all fractionating phases. Plagioclase fractionation (%) refers to the percent of the
total fractionation which is plagioclase.

T h e e l e m e n t - s t r a t i g r a p h y plots (Fig. 7) suggest certain groupings o f sam-

ples which could represent separate batches o f magma. Martin and Rose
( 1 9 8 1 ) have suggested this to explain the 2 0 - - 7 0 year clusters o f historic
activity at Fuego. G r o u p s o f samples are best delineated b y s t r o n t i u m con-
c e n t r a t i o n s and we p r o p o s e t h a t this should be true. The reasoning behind
this is t h a t Sr is effectively b u f f e r e d in the f r a c t i o n a t i n g system. Its distribu-
tion c o e f f i c i e n t in calcic plagioclase is usually b e t w e e n 1.5 and 2.5 (Drake
and Weill, 1975) and is virtually 0 in the o t h e r f r a c t i o n a t i n g phases (Gill,
1981). Mass-balance calculations suggest t h a t plagioclase consists of 1/4 t o
1/2 o f the f r a c t i o n a t i n g phases in each sample. T h e r e f o r e , Sr a b u n d a n c e
should show little change with f r a c t i o n a t i o n (Fig. 9). The u p p e r and lower
curves show the limits o f Sr a b u n d a n c e using D = 1.5 for the range o f pro-
p o r t i o n s o f plagioclase f r a c t i o n a t e d in one batch. S t r o n t i u m values for an-
o t h e r b a t c h of lavas are p l o t t e d on this diagram, and do not fall near the
d e f i n e d field, and require a d i f f e r e n t p a r e n t c o n c e n t r a t i o n or a d i f f e r e n t
d i f f e r e n t i a t i o n m e c h a n i s m . Preliminary e x a m i n a t i o n o f the data o f Carr and
P o n t i e r ( 1 9 8 1 ) f o r Izalco indicates t h a t Sr m a y be a m a g m a b a t c h indicator
t h e r e also.
With this a p p r o a c h , Rayleigh f r a c t i o n a t i o n was a t t e m p t e d again but this
t i m e using a d i f f e r e n t p a r e n t for each group o f samples. These groups are
" B a t c h 1 " samples 1--4, " B a t c h 2 " samples 5 and 6, " B a t c h 3 " samples
7--9, and " B a t c h 4 " samples 1 0 - - 1 4 f r o m Fig. 7. The same d i s t r i b u t i o n co-
efficients were used for all samples. This t i m e trace e l e m e n t f r a c t i o n a t i o n
w o r k e d very well, an e x a m p l e is given in Table III.

Identification o f individual batches

In o r d e r to b e t t e r e x e m p l i f y d i f f e r e n t m a g m a batches, selected trace ele-
m e n t s were p l o t t e d against Sr in Fig. 10. Batch 5 in these plots are f o u r rep-
36

P
4o

Ni ~'
I0 ~ ~
i

150~
i
J 11

50
600 700 800
Sr

Fig. 10. Zirconium and Ni vs Sr variation diagrams including calculated parent composi-
tions (A, B, etc.). Numbers indicate different batches and correspond to their parents
A - l , etc.

resentative average analyses of the Fuego 1974 eruption (Rose et al., 1978).
These plots show dramatically the separation of different magma batches.
Two distinct trends can be observed in these plots. In the Zr plot, one
trend is defined by batch number and parent letter, such as A-1-1-1 and has
a fairly steep positive slope. Parent letters A-B-C-D-E exhibit the other trend
which has a gentle positive slope.
Compatible element trends differ from those of Zr. For example, in Fig.
10, Ni gives a negative sloping trend for batch A-1-1-1. The parental trend,
A-B-C-D-E is relatively constant and has no slope. Trends within each batch
for both Zr and Ni are interpretable as fractionation trends which would
develop in the shallow magma chamber below the volcano.
Trends of parental magmas cannot be explained quite so easily. The com-
patible element concentrations in Fuego lavas are much too low for any mag-
ma simply derived by partial melting of garnet peridotite (Ringwood, 1974;
Condie and Hayslip, 1975). Derivation from eclogite is also excluded because
compatible and incompatible element concentrations require very different
degrees of partial melting (Gill, 1974). The Zr parental trend (Fig. 10) is
similar to a partial melting trend but evidently is not a primary trend. There
is no consistent time-related sense in Fuego's "parental" trends.

Evidence for early, deep fractionation

To explain the high concentration of incompatible elements, constant,
very low Ni and Sc, and highly variable V contents observed in Fuego's
parental magmas, fractionation of olivine, clinopyroxene, and magnetite in
a deep magma reservoir is suggested. The variable V contents may be ex-
plained by variable magnetite fractionation from the primary magma. Con-
stant Ni and Sc values are more difficult to explain. To produce similar Ni
and Sc concentrations in successive batches it is necessary that equivalent
amounts of olivine and clinopyroxene be fraetionated from each batch. Pos-
 37

sibly, regular amounts of olivine and clinopyroxene are fractionated from

each batch before it can rise to a high level magma chamber.
The trace-element data therefore suggests that the parental magmas are
not primary magmas. They may represent successive derivatives which came
from a deep magma chamber after fractionating olivine, clinopyroxene, and
magnetite from a primary magma. Such a deep magma reservoir may be situ-
ated at the base of the crust. Carr {1984) and Rose et al. (1977, 1978)
have cited evidence for primary fractionation operating at the base of the
crust in Central America. Stolper (1980), Walker et al. (1979), and Gill
(1974) also give evidence suggesting this process.

Additional evidence for magma batches at Fuego

The fractionation trend with time for each batch is different (Fig. 11).
Batches 1 and 3 show trends of decreasing fractionation through time. This
kind of fractionation trend may represent a magma batch which after a peri-
od of fractionation erupted almost continuously. The opposite trend can be
observed for batches 2 and 4. These batches become more fractionated with
time. This trend could represent eruptions during continuous fractionation
of a batch of magma. Woodruff et al. (1979) have shown that both types of
trends exist at other volcanoes in Central America.
Evidence for distinct batches of magma is suggested by the modal miner-
alogies (Fig. 4) which show some subtle differences between batches. Modal
differences can be seen between batches 1 and 2, and 2 and 3. Another pe-
trographic variation among batches is the presence of phlogopite only in
batch 4.

40 Batch 2 ,J
./
v
z
Q 30

T
}-

Z
0
p-
t9
< 2O
~ atch 3

co
¢,r
u_

0 I t 111 /
O I ¢ ! I I I I i
14 13 12 1 9 S 7 6 5 4 3 2
oldest SEQUENCE youngest

Fig. 11. Plot s h o w i n g f r a c t i o n a t i o n t r e n d s w i t h i n b a t c h e s a n d t h e i r r e l a t i o n s h i p to stratig-

r a p h y a n d SiO 2.
38

The existence of distinct magma batches, each with distinct chemical

trends, fractionation histories, and petrography implies that the Fuego com-
plex is dominated by closed-system behavior.
A summary of evidence for separate magma batches at the Meseta vent is:
(1) Grouping of major and trace elements when plotted against stratig-
raphy, especially Sr.
(2) Necessity for use of different parental trace element compositions in
Rayleigh calculations.
(3) Fractionation trends within batches.
(4) Coherent trends within batches.
(5) Petrographic differences among batches.
(6) Consistency with the historic pattern of activity at Fuego (Martin and
Rose, 1981).

Relation o f vents and their magma chambers

Rose et al. (1978), Martin and Rose (1981), and Anderson (in press) have
given evidence which suggests that Fuego is probably fed by a shallow, dike-

FUEG M ACATENANGO

/
~V~POAM-~J~--~ ~ Isolated Pookets of Crustolly
~ , c l c M~ts
~ t C o o l ...... [~ ;~ ~.- ~:~ ~ "..
Wall Rock warmed ~ ~'~ "~
Cylinder , , ~ - f. " ~">1 ",~
Surrounding ~: ~- ~ ~ - ~~
~. Path of C ~ o f "~.z"
NOT TO /" / Ascending ¢ ":] (
;CALE \ / Magma ["~ ~ \ Partly or cc3
\/ ~ Wholly Solidified
~' Granitic Pluton

?
'~ Zone of Crustal Interaction ~- .... -~
SAnd/Or Melting ~ ".,
--- J i
_- V__L_ f .... ,
CRUST 35 km __ ~ - ~ ~ ~ ~r- . . . . . . . . . . /

Basalt \ / N / N / \ ,."
ii, ,,x, o,,+mog
• • I f ~ractionated

igrattng Mogmatic Axis

Fig. 12. Cross-sectional view of Fuego and Acatenango complexes and their possible
plumbing systems.
 39

like magma chamber (0.1 km3). This is supplied by a deeper, larger parental
magma chamber (0.5 km 3) {Rose et al., 1982). Acatenango and Yepocapa
are reported to have been connected by a I km long fissure (Mooser et al.,
1958). This is supported by observations of the 1972 eruption of Acate-
nango. The five active vents during that eruption were located in the saddle
between Acatenango and Yepocapa along a north--south axis (Smith. Inst.
CSLP, 1972, Event Card No. 1513). Rose has observed a north--south
trending-fissure in the summit crater of Fuego (Martin, 1979). The alignment
of the four vents and observed fissures strongly suggest that Meseta must also
have been fed by a dike-like magma body.
Due to the similarity in lava chemistry of Meseta and Fuego t h e y probably
have the same parental magma chamber. This chamber is relatively small
(0.5 km ~) and behaves as a closed system. Chemical differences between the
Fuego and Acatenango complexes suggest that Acatenango has a separate
parental magma chamber. There is no necessity for these chambers to be at
different depths in the crust, although they may be. Chemical variations be-
tween the two complexes arise within these parental magma chambers. Both
complexes erupt when magma rises into dike-like secondary magma cham-
bers or feeders. A third, primary magma chamber may exist at the base of
the crust (35 km). Before buoyant rise to the parental magma chamber oli-
vine, clinopyroxene, and possibly magnetite are fractionated from the mag-
ma (Rose et al., 1977; Grant et al., 1983; Carr, in press). Figure 12 is an il-
lustration of how these magma chambers are perceived to exist.

Evolution of Fuego

Historic activity at the Fuego vent has been mostly basaltic in composi-
tion, and has been shown to become more mafic with time. The present
study shows that lava compositions at the Meseta vent are more silicic than
the historic lavas of Fuego. It thus appears that the Fuego complex has be-
come more mafic with time. This is in contrast with evolutionary patterns
at many other composite volcanoes which become more silicic with time.
Some examples of these are: Mashu volcano, Japan (Katsui et al., 1975),
Santa Maria, Guatemala (Rose et al., 1977), Bouqueron, E1 Salvador (Fair-
brothers et al., 1978), Atitl~n, Guatemala (Woodruff et al., 1979), and Izalco
and Santa Ana, E1 Salvador (Carr and Pontier, 1981). Mayon volcano in the
Phillippines shows a more mafic trend with time, similar to that of Fuego.
Newhall (1979) attributes this trend to changes in the source .rocks towards
more refractory compositions.

Changing geometry of Fuego's magma chamber?

It is proposed here that Fuego's evolution towards more mafic composi-
tions is primarily a result of changing conditions of the parental magma
chamber. During the time that Meseta was building its cone repose periods
were longer, possibly because of the inhibition of the cone on activity as its
40

size increased, than those observed at the Fuego vent. Reposes at the current
Fuego vent are on the order of m o n t h s or a few years during eruption of
each magma batch (Rose et al., 1978; Martin and Rose, 1981), while those at
Meseta were probably tens of years. The longer reposes at Meseta had the ef-
fect o f allowing more silicic lavas to evolve. Although reposes were longer,
the system remained closed.
In order to explain this difference in repose periods the growth of the
c o mp lex must be examined first. The Meseta vent and lavas are clearly older
than the Fuego vent and its lavas. Therefore, it is suggested that while Meseta
was building its cone the Fuego vent did not exist. Meseta continued to grow
until a large eruption accompanied by collapse dest royed the t op of the cone
creating the present exposures. Such an event m ay have caused the dike sys-
tem under Meseta to be c om e unsuitable for passage of new magma. As a re-
sult, activity shifted along the fissure to the present site of the Fuego vent.
Eggers (1971) shows that such vent migration after eruption/collapse has oc-
curred at Pacaya in Guatemala. Associated with migration of activity along
the fissure the magma chamber may have taken on a new g e o m e t r y and the
new Fuego vent was now situated on the flank of Meseta. This may have re-
duced the inhibiting effects of cone height and volume on eruptions. As a re-
sult o f these changes, repose periods were shortened and lava compositions
became mo r e mafic.

Increased magma generation rate?

A n o t h e r possible explanation for shorter reposes at Fuego than at Meseta
might be higher rates of magma generation. This would result in m ore fre-
q u en t batching and thus shorter reposes and m ore mafic lavas. Higher rates
o f magma generation could be closely related to higher subduction rates. A
problem with this hypothesis is that all the volcanoes in the volcanic front
should exhibit increasing levels of activity but t h e y do not.

Age o f the Fuego system

Assuming the m a x i m u m time of 170 years between clusters o f historic ac-
tivity to represent the time between batch arrivals-throughout Fuego's his-
t o r y (Martin and Rose, 1981), an age for the complex can be estimated. Con-
sidering the Meseta vent first, the exposed section represents approximately
25% o f its history. Selecting 4 lava flows as an average for each batch and
counting +50 lava flows in the exposed section, an age estimate of approxi-
mately 8500 years is calculated for Meseta. Fuego appears to have a similar
volume to that which Meseta had and would also be about 8500 years old.
Combining these ages, the Fuego complex might be about 17,000 years old.
If Fuego has a shorter batching period than Meseta due to different rates of
magma generation, Meseta could represent a longer time period. If this is so,
17,000 years is a m i ni m um age for the complex.
Martin and Rose (1981) calculated the age o f Fuego in a slightly different
manner. Th ey assumed an overall eruption rate, and det erm i ned that at t hat
 41

rate it would take 13,000 years to produce the observed volume of the
Fuego complex. Rose et al. (1977) estimated an age of 30,000 years for
Santa Maria volcano based upon paleomagnetic correlations. Ashes from
Agua have been found interbedded with the Pinos Altos tephra in south-
eastern Guatemala. This tephra has been dated by C 14 methods at 23,000
years (Peterson, 1980). Therefore, the proposed minimum age of 17,000
years for the Fuego complex does not seem unreasonable.

Differences in magma evolution patterns at Fuego compared to Agua and

Acatenango

What causes chemical variation in andesitic rocks at Agua and Acatenango

to diverge from Fuego trends? We think it reflects fundamental differences
in the petrological processes occurring in the shallow magma bodies. Two
hypotheses may explain the data: (1) a different kind of crystal fractiona-
tion may operate at Agua and Acatenango, in which larger, open-system
magma bodies, (O'Hara, 1977), build up incompatible elements, change in-
compatible element ratios and because of low pH:O, have plagioclase-poor
crystallization (Yoder and Tilley, 1962) and (2) the andesitic rocks of Agua
and Acatenango may reflect the admixture of silicic crustally derived melts as
part of a magma mixing or assimilation process which operates simultane-
ously with POAM fractionation, but is absent at Fuego (Fig. 12). This ad-
mixture although relatively minor, is consistent with the chemical divergence
described above.
It is beyond the scope of this paper to discuss the evidence which bears
on these hypotheses, but we favor the second. We view the Fuego system, by
contrast to Acatenango and Agua, to have closed system shallow bodies in
which POAM fractionation occurs without much contamination or magma
mixing. At Agua and Acatenango conditions like this do not apply.
Summary of evidence for a magma chamber dominated by closed-system
behavior beneath Fuego includes:
(1) Identification of several unique magma batches which retain their orig-
inal identity.
(2) The absence of high incompatible element and low CaO, A120~, and
Na20 contents which are observed at Acatenango and Agua and are indica-
tive of more complex magmatic evolution processes.
(3) The differentiation and cooling pattern of plagioclase suggests that
Fuego's magma b o d y has a dike-like nature. Such configuration would pro-
mote rapid cooling and solidification of magma before arrival of a new
batch.

CONCLUSIONS

(1) Fuego's older lavas form a chemically coherent pattern of geochemical

variation with the historic lavas.
42

(2) O l d e r r o c k s are generally m o r e silicic including higher p r o p o r t i o n s o f

basaltic andesite and andesite. R e c e n t r o c k s are high-A12Os basalts.
(3) T h r o u g h o u t its h i s t o r y , s e p a r a t e b a t c h e s o f m a g m a can be r e c o g n i z e d
b y t r a c e e l e m e n t d i f f e r e n c e s . This suggests t h a t o p e n s y s t e m f r a c t i o n a t i o n
is n o t o c c u r r i n g at F u e g o .
(4) E v i d e n c e f o r POAM f r a c t i o n a t i o n w i t h i n F u e g o m a g m a b a t c h e s is
strong.
(5) B e f o r e arrival in the shallow m a g m a b o d i e s u n d e r F u e g o and t h e adja-
c e n t s t r a t o v o l c a n o e s , t h e m a g m a p r o b a b l y u n d e r g o e s f r a c t i o n a t i o n at t h e
base of t h e crust, involving olivine, c l i n o p y r o x e n e , and possibly m a g n e t i t e .
(6) A m i n i m u m age for the F u e g o c o m p l e x based u p o n t h e timing o f
b a t c h arrival and t h e n u m b e r o f e r u p t i o n s p e r b a t c h is 1 7 , 0 0 0 years.
(7) A d j a c e n t larger and older andesitic s t r a t o v o l c a n o e s have higher i n c o m -
p a t i b l e e l e m e n t c o n c e n t r a t i o n s and d i f f e r e n t i n c o m p a t i b l e e l e m e n t ratios
t h a n F u e g o . L o w e r CaO, A1203 and N a 2 0 also are f o u n d in t h e r o c k s o f t h e s e
o l d e r cones.

ACKNOWLEDGEMENTS

F u n d i n g f o r this s t u d y was p r o v i d e d b y N S F G r a n t DES 7 8 - 0 1 1 9 0 . Dr.

E d w a r d E r b and C h e v r o n R e s o u r c e s also deserve t h a n k s f o r t h e i r p a r t in
f u n d i n g t h e trip to G u a t e m a l a . Dr. T h e o d o r e B o r n h o r s t ' s assistance w i t h
the field w o r k was greatly a p p r e c i a t e d . Credit f o r p r e p a r a t i o n o f s u p e r b thin
sections goes to B o b M c C a r t h y . Dr. Douglas M c D o w e l l ' s criticisms and sug-
gestions while reviewing this t e x t w e r e very helpful. D r a f t i n g o f figures was
d o n e b y D e b b i e McDowell.

REFERENCES

Anderson, A.T., in press. Probable relations between plagioelase zoning and magma dy-
namics, Fuego volcano, Guatemala. Am. Mineral.
Carr, M.J., in press. Crustal influence on magmatic diversity along the Central American
volcanic front.
Carr, M.J. and Pontier, N.K., 1981. Evolution of a young composite cone towards a ma-
ture central vent. Izalco and Santa Ana volcanoes in E1 Salvador, Central America. J.
Volcanol. Geotherm. Res., 11: 277--292.
Condie, K.C. and Hayslip, D.L., 1975. Young bimodal volcanism at Medicine Lake Vol-
canic Center, northern California. Geochim. Cosmochim. Acta, 39: 1165--1178.
Davies, D.K., Quearry, M.W. and Bonis, S.B., 1978. Glowing avalanches from the 1974
eruption of the volcano Fuego, Guatemala. Geol. Soc. Am. Bull., 89: 369--384.
Deger, E., 1932. Der Ausbruch des Fuego in Guatemala am 21 Januar 1932 und die
chemische Zusammensetzung seiner Auseurfmaterialien. Chem. Erde, 7: 291--297.
Drake, M.J. and Weill, D.F., 1975. Partition of Sr, Ba, Ca, Y, Eu 2÷, Eu 3÷, and other REE
between plagioclase feldspar and magmatic liquid: an experimental study. Geochim.
Cosmochim. Acta, 39: 689--712.
Eggers, A.A., 1971. The geology and petrology of the Amatitlan Quadrangle, Guatemala.
Ph.D. thesis, Dartmouth College (unpubl.).
 43

Fairbrothers, G.E., Carr, M.J. and Mayfield, D.G., 1978. Temporal magmatic variation at
Boqueron volcano, E1 Salvador. Contrib. Mineral. Petrol., 67: 1--9.
Flanagan, F.J., 1973. 1972 values for international geochemical reference standards.
Geochim. Cosmochim. Acta, 37: 1181--1200.
Gill, J.B., 1974. Rote of underthrust oceanic crust in the genesis of a Fijian calc-alkaline
suite. Contrib. Mineral. Petrol., 42: 29--45.
Gill, J.B., 1981. Orogenic Andesites and Plate Tectonics. Springer Verlag, New York,
N.Y., 390 pp.
Grant, N.K., Rose, W.I., Jr. and Fultz, L.A., 1983. Open-system fractionation in basalts
and basaltic andesites from Guatemala. EOS, Trans. Am. Geophys. Union, 64 (18):
330.
Hebberger, J.J., Davies, D.K. and Bonis, S.B., 1977. Recent lahars, Fuego volcano, Guate-
mala. Geol. Soc. Am. Abstr. Prog., 9 (5): 603--604.
Irvine, T.N. and Baragar, W.R.A., 1971. A guide to the chemical classification of the com-
mon volcanic rocks. Can. J. Earth Sci., 8: 523--548.
Katsui, Y., Ando, S. and Inaba, K., 1975. Formation and magmatic evolution of Mashu
volcano, East Hokkaido, Japan. J. Fac. Sci. Hokkaido Univ., Ser. IV, 16: 533--552.
Leake, B.E., Hendry, G.L., Kemp, A., Plant, A.G., Harvey, P.K., Wilson, J.R., Coats, J.S.,
Aucott, J.W., Lunel, T. and Howarth, R.J., 1969. The chemical analysis of rock pow-
ders by automatic X-ray fluorescence. Chem, Geol., 5: 7--86.
Martin, D.P., 1979. The historic activity of Fuego volcano, Guatemala: Constraints on
the subsurface magma bodies and processes therein. M.S. thesis, Mich. Tech. Univ.
(unpubl.).
Martin, D.P. and Rose, W.I., 1981. Behavior patterns of Fuego volcano, Guatemala. J.
Volcanol. Geotherm. Res., 10: 67--81.
Mooser, F., Meyer-Abich, H. and McBirney, A.R., 1958. Catalogue of the Active Vol-
canoes of the World. Part VI. IAVCEI, Naples, 146 pp.
Myers, A.T., Havens, R.G., Connor, J.J., Conklin, N.M. and Rose, H.L.J., Jr., 1976. Glass
reference standards for trace element analysis for geologic materials. U.S. Geol. Surv.,
Prof. Pap. 1013, 29 pp.
Newhall, C.G., 1979. Temporal variations in the lavas of Mayon volcano, Philippines. J.
Volcanol. Geotherm. Res., 6: 61--83.
O'Hara, M.J., 1977. Geochemical evolution during fractional crystallization of a periodi-
cally refilled magma chamber. Nature, 266: 503--507.
Peterson, P.S., 1980. Tephra of the Laguna de Ayarza calderas of southeastern Guatemala
and its correlation to units of the Guatemalan Highlands. M.S. thesis, Mich. Tech.
Univ. (unpubl.).
Pushkar, P,, 1966. The isotopic composition of strontium in volcanic rocks from island
arcs. Ph.D. Diss., Univ. Calif., San Diego.
Ringwood, A.E., 1974. The petrological evolutions of island arc systems. J. Geol. Soc.
London, 130, 183--204.
Rose, W.I., Jr., Bonis, S., Stoiber, R.E., Keller, M. and Bickford, T., 1973. Studies of vol-
canic ash from two recent Central American eruptions. Bull. Volcanol., 37: 338--364.
Rose, W.I., Jr., Grant. N.K., Hahn, G.A., Lange, I.M., Powell, J.L., Easter, J. and DeGraff,
J.M., 1977. The evolution of Santa Maria volcano, Guatemala. J. Geol., 85: 63--87.
Rose, W,I., Jr., Anderson, A.T., Bonis, S. and Woodruff, L., 1978. The October 1974
basaltic tephra, Fuego volcano, Guatemala: description and history of the magma
body. J. Volcanol. Geotherm. Res., 4: 3--53.
Rose, W.I., Jr., Penfield, G.T., Drexler, J.W. and Larson, P.B., 1980. Geochemistry of the
andesite flank lavas of three composite cones within the Atitlan Cauldron, Guatemala.
Bull. Volcanol., 43: 131--153.
Rose, W.I., Jr., Stoiber, R.E. and Malinconico, L.L., 1982. Eruptive gas compositions and
fluxes of explosive volcanoes: budget of S and C1 emitted from Fuego volcano, Gua-
temala. In: R.S. Thorpe (Editor), Orogenic Andesites. John Wiley, New York, N.Y.,
pp. 669--676.
44

Sakuyama, M., in press. Phenocryst assemblages and H:O contents in circum-Pacific arc
magmas.
Shaw, D.M., 1970. Trace element fractionation during anatexis. Geochim. Cosmochim.
Acta, 34: 237--243.
Simkin, T., Siebert, L., McClelland, L., Bridge, D., Newhall, C. and Latter, J.T., 1981.
Volcanoes of the World. Hutchinson-Ross, Stroudsburg, Pa., 232 pp.
Stoiber, R.E. and Rose, W.I., Jr., 1970. Geochemistry of Central American volcanic gas
condensates. Geol. Soc. Am. Bull., 81: 2891--2912.
Stolper, E., 1980. A phase diagram for mid-ocean ridge basalts: preliminary results and
implications for petrogenesis. Contrib. Mineral. Petrol., 74: 13--27.
Van Der Plas, L. and Tobi, A.C., 1965. Chart for judging the reliability of point counting
results. Am. J. Sci., 263: 87--90.
Walker, D., Shibata, T. and DeLong, S.E., 1979. Abyssal tholeiites from the Oceano-
grapher Fracture Zone II. Phase equilibria and mixing. Contrib. Mineral. Petrol., 70:
111--125.
Wright, T.L. and Doherty, P.C., 1970. A linear programming and least squares computer
method for solving petrologic mixing problems. Geol. Soc. Am. Bull., 81 : 1995--2008.
Woodruff, L.G., Rose, W.I., Jr. and Rigot, W., 1979. Contrasting fractionation patterns
for sequential magmas from two calc-alkaline volcanoes in Central America. J. Vol-
canol. Geotherm. Res., 6: 217--240.
Yoder, H.S., Jr. and Tilley, C.E., 1962. Origin of basalt magmas: an experimental study
of natural and synthetic rock systems. J. Petrol., 3: 342--532.