Results Of Weekly Chemical And Isotopic Monitoring Of
Selected Springs In Norris Geyser Basin, Yellowstone National
Park During June-September, 1995.

By R. O. Fournier1, U. Weltman2, D. Counce3, L. D. White1, and C. J. Janik1

Open File Report 02-344

ABSTRACT

Each year at Norris Geyser Basin,
generally in August or September, a
widespread hydrothermal "disturbance"
occurs that is characterized by simultaneous
changes in the discharge characteristics of
many springs, particularly in the Back Basin.
During the summer season of 1995, water
samples from eight widely distributed hot
springs and geysers at Norris were collected
each week and analyzed to determine whether
chemical and isotopic changes also occurred
in the thermal waters at the time of the
disturbance. In addition, Beryl Spring in
Gibbon Canyon, 5.8 km southwest of Norris
Geyser Basin, was included in the
monitoring program.
Waters discharged by four of the
monitored hot springs and geysers appear to
issue from relatively deep reservoirs where
temperatures are at least 270 ¡C and possibly
higher than 300 ¡C. At the time of, and for
several days after, the onset of the 1995
disturbance, the normally neutral-chloride
waters discharged by these four features all
picked up an acid-sulfate component and
became isotopically heavier. The acid-sulfate
component appears to be similar in
composition to some waters discharged in
100 Spring Plain that issue from subsurface
regions where temperatures are in the range
170Ð210 ¡C. However, the two monitored
springs that discharge acid-chloride-sulfate
waters in the 100 Spring Plain region did not
show any significant chemical or isotopic
response to the annual disturbance. Beryl
Spring, and two neutral-chloride hot springs
at Norris that appear to draw their water from
reservoirs where temperatures are 250 ¡C or
less, also did not show any significant
chemical or isotopic response to the annual
disturbance.
After the start of the annual disturbance,
chloride concentrations in water sampled
from Double Bulger Geyser in the Back
Basin increased from about 800 ppm to about
1500 ppm, nearly twice as high as any
previously reported chloride concentration in
a thermal water at Yellowstone. The isotopic
composition of that water precludes an origin
of the high chloride by evaporation at
atmospheric pressure. One way to account
for the unique chemical and isotopic
composition of this highly concentrated water is by recirculation of water that had gone
through one cycle of adiabatic cooling during
upflow (decompressional boiling) back down
into the hydrothermal system, where it is reheated
to greater than 220 ¡C. This
previously boiled water then undergoes
additional cycles of decompressional boiling
during subsequent upflow. Another way the
unique chemical and isotopic composition of
Double Bulger water might evolve is by
excess boiling in the formation that results
from a decrease in fluid pressure within the
channels of upflow.
The annual disturbance at Norris Geyser
Basin generally appears to be triggered by a
cyclic up and down movement of the boilingpoint
curve within the hydrothermal system
in response to changes in the potentiometric
surface of the cold water that is adjacent to,
and interconnected with, that hydrothermal
system. Annual disturbance phenomena that
are easily recognized at Norris Geyser Basin
may not be easily recognized elsewhere in
Yellowstone National Park because (1) the
neutral-chloride waters at Norris ascend
directly from higher-temperature and higherpressure
reservoirs (270 to >300 ¡C at Norris
compared to 180Ð215 ¡C at Upper and Lower
Geyser Basins) that are capable of producing
massive amounts of high-pressure steam, and
(2) the clay that makes hot spring and geyser
waters become turbid at Norris, heralding the
start of the disturbance, comes from acid
altered rocks that are widely distributed at
intermediate depths at Norris, and that are
rare in other geyser basins.

INTRODUCTION
At Norris Geyser Basin, Yellowstone
National Park (YNP), an annual
hydrothermal "disturbance" occurs, generally
in late August or September, that is
characterized by increased turbidity in many
hot springs, increased discharge of water and
steam, extreme fluctuations in temperatures
of pools, and sometimes small hydrothermal
explosions (White and others, 1988).
Chemical changes at the time of the annual
disturbance also have been observed in
waters collected from Cistern Spring in the
Back Basin of Norris (Fig. 1) (Fournier and
others, 1986). Long-term monthly
collections and chemical analyses of water
from Cistern Spring during 1976Ð1985 showed a yearly cycle in which there was a
general increase in concentrations of all
dissolved constituents over the course of each
summer. This is indicative of increased
boiling leading up to the disturbance
(Fournier and others, 1986; 1992). In
contrast, immediately following the onset of
these disturbances, Cistern Spring waters
have lower chloride and pH, and increased
sulfate, indicative of subsurface mixing of
neutral-chloride and acid-sulfate waters at the
time of the disturbance and for several weeks
thereafter (Fournier and others, 1986; 1992).
Prior to our 1995 study only one other
hydrothermal feature at Norris (recently
informally named "Wistful" Geyser) was
known to behave in a similar manner. It
normally issues water that has pH Å 7.0, Cl
Å 675 mg/kg, and SO4 Å 25 mg/kg. By
chance it was sampled on the first day of the
1990 annual disturbance. That sample had
pH = 2.82, Cl = 527 mg/kg, and SO4 = 133
mg/kg (Fournier and others, 1994). Four
days later its pH had increased to 5.36, its Cl
was 634 mg/kg, and SO4 was 99 mg/kg
(Fournier and others, 1994). These results
clearly show that subsurface mixing of
neutral-chloride and acid-sulfate waters
occurs in at least two regions of Norris at the
time of the disturbance. The present study
was undertaken to determine the extent of
subsurface mixing and discharge of different
thermal waters during these disturbances, and
the duration of "mixed" discharge.

MONITORED HYDROTHERMAL FEATURES

The approximate locations of the eight
hydrothermal features that were monitored at
Norris Geyser Basin are shown in Figure 1,
along with the locations of some other wellknown
springs and geysers. In addition, one
spring (Beryl Spring) that is located in
Gibbon Canyon about 5.8 km to the
southwest of Norris Geyser Basin was
included in the monitoring program to see
whether the annual disturbance extends to
that region of the park. We limited our
monitoring activities to a total of nine springs
because of constraints on the time that could
be devoted to sample collection activities, and
because of a limited budget to carry out the
supporting chemical and isotopic analyses. Eight springs at Norris were selected to
achieve as widespread a geographic
distribution as possible. A major limiting
factor in their selection was the distribution of
flowing springs at the start of the sampling
period in 1995. In the Porcelain Terrace
region we monitored "Wistful" Geyser,
Perpetual Spouter, and "Carnegie" Spring,
the latter issuing at the site of the second
Carnegie drill hole (Fenner, 1936); in the
Back Basin we monitored Cistern Spring,
Double Bulger Geyser, and Porkchop
Geyser; in the 100 Spring Plain we
monitored Sulfur Dust Spring and "Black"
Spring. Quotation marks indicate that a
particular spring or geyser has recently been
given an informal name that has not yet been
officially recognized by the National Park
Service.
The nine hydrothermal features selected
for monitoring were divided into four groups
as follows:
Group 1 a. Beryl Spring
b. Cistern Spring
Group 2 a. Perpetual Spouter
b. ÒBlackÓ Spring
c. Sulfur Dust Spring
Group 3 a. ÒCarnegieÓ Spring at the
Carnegie drill site #2
b. ÒWistfulÓ Geyser
Group 4 a. Cistern Spring
b. Double Bulger Geyser
c. Porkchop Geyser
Each of the springs in a given group were
sampled on the same day, and the different
groups generally were sampled on four
successive days. Three days then passed
with no sampling. Note that Cistern Spring
is in two groups, so it was sampled twice a
week. This was done because we intended to
use its compositional variations as a key
indicator of the onset and progress of the
annual disturbance. The dates of sample
collection are given in Table 1.

COLLECTION AND ANALYSIS PROCEDURES

Waters for use in all the chemical
analyses, except for silica, were collected
without filtration into 500 ml plastic bottles.
For silica analyses, 10 ml of spring water
were pipetted into 80 ml of silica free water
that had previously been measured into a 125ml plastic bottle. Standard analytical
procedures were used for the chemical
analyses (Trujillo, and others, 1987).
Samples for isotopic analyses were collected
in 60 ml glass bottles, with the water
completely filling the bottles.
Values for pH were determined in the
field using a portable pH meter and
combination glass electrode calibrated before
each measurement with pH 4, pH 7, and pH
9 buffer solutions. The field pH readings for
the thermal waters were made with the
combination electrode immersed in the hot
spring and geyser pools. Except for the
interval from June 22 (Julian day 173) to July
21 (Julian day 202), a thermister probe that
also served to make automatic temperature
corrections for pH measurements was used to
measure the temperatures of the hydrothermal
features. On June 22 (Julian day 173) the
thermister probe for the pH meter was
damaged and a replacement was not obtained
until July 22 (Julian day 203). Therefore,
during that time interval, a different
thermister meter and probe, separate from the
pH meter, was used to measure water
temperature. This necessitated making
manual temperature corrections for the pH
measurements. Subsequent calibration
indicated that the interim thermister
equipment gave readings in the field about
2¡C lower than other temperature probes.
On July 15 (Julian day 196), the
combination pH electrode that had been used
up to that time was accidentally broken; it
was replaced with a new electrode on July 16
(Julian day 197). It will be shown
subsequently that some of the field pH
readings were anomalously high (especially
for Sulfur Dust Spring) between the time that
the pH meter temperature probe broke on
June 22 (Julian day 173), and the time when
a new pH electrode was installed on July 16
(Julian day 197). Coincident with installation
of the new combination pH electrode the
anomalously high field pH readings returned
to normal readings. We suspect that the
original pH electrode began to malfunction at
about the time that the pH meter temperature
probe was damaged on June 22 (Julian day
173), and that unreliably high pH readings
were obtained thereafter from that electrode,
especially for the more acid waters.
However, because we now have no way toverify this, in Figs. 2aÐ10a all our field pH
data are plotted even though we believe that
only the data before June 22 (Julian day 173)
and after July 16 (Julian day 197) are likely
to be reliable. The interval in doubt is
bracketed by vertical dashed lines on these
figures and pH is indicated by small
diamonds.
The recorded temperatures of the springs
and geysers were measured in the deepest
and hottest parts of their pools that could be
accessed using a temperature probe with a 1-
meter reach. For Beryl Spring, Cistern
Spring, and Porkchop Geyser, the deepest,
and probably hottest, parts of the pools could
not be reached with our equipment.
However, care was taken always to measure
temperature at the same location within each
pool so that variations are likely to be
significant even if the recorded temperature
was less than the maximum temperature.
Note that at the elevation of Norris Geyser
Basin dilute water boils at about 92 ¡C.

RESULTS AND DISCUSSION

The results of the chemical and isotopic
analyses are given in Table 1 and summarized
in Figs. 2Ð10. The onset of the hydrothermal
event that we set out to study, the 1995
annual disturbance at Norris Geyser Basin,
occurred relatively early in that year, on July
29 (Julian day 210). It was marked by the
usual widespread increase in hydrothermal
activity and the appearance of turbidity in
many spring waters.
Two of the monitored springs, "Black"
Spring and Sulfur Dust Spring, usually
discharge acid-chloride-sulfate waters, typical
of the 100 Spring Plain region and Porcelain
Basin. The other monitored hydrothermal
features usually discharge near neutral,
chloride-rich and sulfate-poor waters. Four
of these hydrothermal features, Cistern
Spring, Double Bulger Geyser, Porkchop
Geyser, and "Wistful" Geyser, apparently
draw most of their water from a relatively
deep reservoir at Norris where temperatures
are at least 270 ¡C, and possibly as high as
310 ¡C according to Fournier and others
(1976), Fournier (1989), and Fournier and
others (1994). The three remaining
monitored hydrothermal features, Beryl
Spring, "Carnegie" Spring, and Perpetual
Spouter, apparently draw most of their water from cooler reservoirs where temperatures
are 250 ¡C or less (Table 2). The estimation
of reservoir temperatures is discussed below.

Estimation Of Reservoir Temperatures

Subsurface reservoir temperatures at
Norris Geyser Basin previously have been
estimated using a variety of chemical
geothermometers and mixing models
(Fournier et. al. 1976; Fournier, 1989;
Fournier and others 1991; Fournier and
others 1994). We have used the silica and
Na/K data of Table 1 to estimate last
temperatures of water-rock equilibration
(presumably reservoir temperatures) that are
shown in Table 2.
The silica (quartz) geothermometer
(Fournier and Rowe, 1966; Fournier and
Potter, 1982) works best for near neutral
waters issuing from reservoirs where
temperatures are in the range 180 ¡C to about
230 ¡C (Fournier, 1985). Where reservoir
temperatures are above 230 ¡C, some
amorphous silica is likely to precipitate
underground during upflow from the
reservoir to the surface. In this case, quartz
geothermometry gives too low an estimated
reservoir temperature. An exception is for
acid waters because polymerization and
precipitation of amorphous silica are inhibited
when pH is less than about 4 (Fournier
1985). When using silica geothermometry
to estimate a reservoir temperature it is
necessary to make assumptions regarding
which silica mineral is controlling dissolved
silica (usually quartz is assumed when the
temperature is > 180 ¡C), and whether there
has been mainly conductive cooling or mainly
cooling by adiabatic decompressional boiling
during ascent (Fournier, 1985, Fournier,
1991). Table 2 shows the range of possible
results with different assumptions.
Many different Na/K geothermometers
have been proposed for hydrothermal
systems (Fournier, 1979; 1981; 1991 and
references therein). The most widely ones
now in use are by Truesdell (1976), Fournier
(1979), Arn—rsson (1983), and Giggenbach
(1988). Which of these works best in a
given area appears to depend on the mineral
suite and the structural states of the alkali
feldspars that are present in the reservoir
rocks (Fournier, 1991). Results using theGiggenbach (1988), Fournier (1979), and
Truesdell (1976) Na/K geothermometers are
given in Table 2. The Arn—rsson (1983)
geothermometer gives results that are very
similar to those of the Truesdell (1976)
geothermometer.
Figure 11 is a plot of temperatures
estimated using the Giggenbach (1988) Na/K
geothermometer versus temperatures
estimated using the quartz geothermometer
(Fournier and Potter, 1982). The straight
line shows where there is agreement between
temperatures estimated by the two
geothermometers. Only waters from Beryl
Spring and "Carnegie" Spring, which are
assumed to have cooled adiabatically, lie on
this line. Adiabatic cooling for waters
discharged by these springs is reasonable, as
they discharge a lot of steam with their
waters. All the other spring and geyser
waters yield silica temperatures that are
significantly lower than the Na/K
temperatures. The estimated temperature of
the reservoir supplying water to Perpetual
Spouter is about the same as that of the
"Carnegie" Spring reservoir, based on the
quartz geothermometer. However, Na/K
geothermometry puts the Perpetual Spouter
reservoir at about 210 ¡C using the Truesdell
(1976) equation and at about 250 ¡C using
either the Fournier (1979) equation or the
Giggenbach (1988) equation (Table 2 and
Fig. 11). We conclude that silica is likely to
have precipitated during upflow and prefer
the 250 ¡C estimated reservoir temperature
because of the excellent agreement of the
Giggenbach (1988) Na/K geothermometer
with the silica geothermometer for Beryl
Spring and "Carnegie" Spring waters.
Similarly, silica is likely to have precipitated
from the Cistern Spring, Double Bulger
Geyser, Porkchop Geyser, and "Wistful"
Geyser waters before reaching the surface.
Again, we prefer the estimated reservoir
temperatures obtained using the Giggenbach
(1988) Na/K geothermometer. The Na/K
geothermometer of Fournier (1979) gives
temperatures which are much higher than the
maximum likely temperature of the reservoir
at Norris, estimated using enthalpy-chloride
relations (Fournier, 1989).
The waters issuing from Sulfur Dust
Spring have silica temperatures (quartz
conductive) of about 200 ¡C and those issuing from "Black" Spring have silica
temperatures ranging up to 205 ¡C (quartz
adiabatic) or to about 230 ¡C (quartz
conductive). In contrast, Na/K gives
temperatures in excess of 300 ¡C (Table 2 and
Fig. 11). These are acid waters, so
amorphous silica is not likely to have
precipitated during upflow. We conclude that
the reservoir temperatures for these two
springs are likely to be about equal to the
silica temperatures, and that Na/K gives
estimated reservoir temperatures that are
much too high, as is usual for acid waters
that are equilibrating with clay minerals in the
absence of feldspars (Fournier, 1991). Note
that the lower-temperature acid-chloridesulfate
waters at Norris have Na/K ratios
similar to those in the high-temperature, nearneutral,
chloride-rich waters. Therefore,
subsurface mixing of these two water types
during upflow will markedly affect Cl/SO4
ratios while Na/K ratios remain almost
unchanged.

Geochemical Response To The
Annual disturbance
The chemical and isotopic response
exhibited by hot-spring and geyser waters to
the annual disturbance ranged from
insignificant for the initially acid-chloridesulfate
waters discharged by Sulfur Dust
Spring and "Black" Spring (Figs. 2 and 3) in
the 100 Spring Plain area, to very
pronounced for the generally near-neutral
waters that issue from the high-temperature
reservoirs feeding Cistern Spring, Double
Bulger Geyser, Porkchop Geyser, and
"Wistful" Geyser (Figs. 4Ð7). Near-neutral
waters issuing from the lower-temperature
reservoirs feeding "Carnegie" Spring,
Perpetual Spouter, and Beryl Spring (Figs.
8Ð10) showed little response to the
disturbance.
The large upward spike in pH exhibited
by Sulfur Dust Spring over the period June
23ÐJuly 14 (Julian days 174Ð195) shown in
Figure 2a is contrary to what would be
expected from the nearly constant Cl and SO4
concentrations measured in the waters
collected at the same time from this spring
(Fig. 2b). It also is contrary to the pH
measurements made in the laboratory (Fig. 2a
and Table 1). As discussed above, we believe that the anomalously high field pH
measurements are in error. However, an
argument can be made that the high field pH
readings for Sulfur Dust Spring might be
accurate (not the result of a faulty electrode
reading). During the period that anomalously
high pH readings were made in Sulfur Dust's
pool there was no detected problem with the
3-buffer calibration of the electrode before
each pH reading. Also, with the exception of
two anomalously high pH readings in
"Black" Spring (Fig. 3a), and one
anomalously high reading in Cistern Spring
(discussed below), consistently "reasonable"
pH readings were made in other thermal
pools on the same day, and over the same
time span that anomalously high pH readings
were made for Sulfur Dust waters. If the
anomalous high field pH readings for Sulfur
Dust and "Black" are real (not the fault of a
malfunctioning electrode) the pH of the
samples will have had to change (become
lower) in the sample bottles after collection.
One way for this to occur would be for the
deep Sulfur Dust Spring water (and "Black"
also) to carry a nearly constant concentration
of colloidal sulfur which usually is
completely converted to sulfuric acid by
bacteria activity before reaching the surface
pool. If, for some unexplained reason, for a
period of about three weeks, this colloidal
sulfur was not converted to sulfuric acid
underground, it could have been converted to
sulfuric acid in the sample bottle by bacterial
activity between the time of collection and the
time of laboratory analyses because the water
samples were not filtered. Again, we stress
that we do not advocate this latter explanation
because the return to "normal" pH readings in
both Sulfur Dust Spring and "Black" Spring
was coincidental with the replacement of the
original pH electrode with a new electrode.
Like the Sulfur Dust results, Cl, SO4
and pH in "Black" Spring's waters,
measured in the laboratory, remain essentially
constant even though some pH readings were
anomalously high (Fig. 3). Again, we
believe that the best explanation for the pH
anomaly is a malfunctioning pH electrode that
was used with the portable pH meter. It is
puzzling, however, why the upward pH
spike for Sulfur Dust Spring was so much
greater and started sooner than that for "Black" Spring. Sulfur Dust Spring
normally has a slightly lower pH and much
lower temperature than "Black" Spring.
Perhaps the higher temperature of "Black"
Spring allowed a faulty pH electrode to come
closer to equilibrium during the relatively
short time of the in-pool measurement than
occurred for Sulfur Dust Spring.
The waters from "Black" Spring are
slightly lower in Cl and are isotopically
heavier than those collected from Sulfur Dust
Spring (Table 1 and Figs. 2d, 3d, and 12).
The isotopically heavy nature of "Black"
Spring relative to Sulfur Dust can be
explained in part by different degrees of
mixing of isotopically light and Cl-rich
waters with isotopically heavy and Cl poor
waters, and in part by a greater degree of
evaporative steam loss from "Black" Spring.
A relatively high degree of evaporation from
"Black" Spring, compared to Sulfur Dust
Spring, seems reasonable because of its
much greater surface area, higher temperature
(compare Figs. 2a and 3a), and, during our
sampling campaign, very low rate of
discharge. Prior to evaporation "Black"
Spring waters would have contained even
less Cl than our reported values. Our data
show that, in general, the thermal waters at
Norris which contain less Cl generally have
heavier ¥D values.
Water collected from the Y-12 drill hole
probably provides the best indication of the
chemical and isotopic composition of the
deepest and hottest fluid beneath the Norris
Geyser Basin. It has ¥D Å -148 %o , ¥18O Å
-16.5 %o , and Cl Å 548 ppm (unpublished
data, U. S. Geological Survey files, sample
YJ93-17, collected July 29, 1993). During
upflow from the deep reservoir, this water
apparently mixes with water in shallower
reservoirs, generally becoming more dilute in
Cl and isotopically heavier. Sulfur Dust
Spring water is about 6 %o heavier in ¥D
and about 1.5 %o heavier in ¥18O than the
Y-12 water. On a plot of ¥18O versus ¥D,
Sulfur Dust Spring water has about the same
¥D as local meteoric water, but is about 4 %o
heavier in ¥18O (Fig. 12a). In contrast, the
¥D of "Black" Spring water is about 17 %heavier than local meteoric water (about 23
%o heavier than Y-12) while its ¥18O is
about 8 %o heavier than local meteoric water
(about 6 %o heavier than Y-12) (Fig. 12b).
For comparison, in Figure 12 the data point
labeled "Vermilion Spg" shows a typical
isotopic composition of a low-chloride, acidsulfate
water in a steam-heated, nonoverflowing
pool.
Cistern Spring (Fig. 4) exhibited a slow
decrease in pH and Cl/SO4 for about 15 days
prior to the start of the annual disturbance
(Figs. 4a and 4c). This trend of changing
composition abruptly became much more
pronounced on the day of the disturbance and
continued for 30 days thereafter. Similar
precipitous changes in Cl/SO4 and pH had
been noted at the times of other annual
disturbances (Fournier and others, 1986).
There also was a small and short-lived
mixing event 20 days prior to the start of the
disturbance (day 190), during which Cistern
Spring's water exhibited decreased Cl and
increased SO4 (Fig. 4b). This event was
accompanied by a decrease in Na/K (Fig.
4c). We expected to see an accompanying
decrease in pH at the time of this small event,
but instead recorded an anomalously high pH
reading (Fig. 4a). As noted above, we
suspect that this high pH on day 190 may be
erroneous, the result of a malfunctioning
electrode.
Isotopic changes in Cistern Spring's
waters generally parallel the variations in Cl
and SO4. Both ¥D and ¥18O become heavier
as Cl decreases and SO4 increases (compare
Fig. 4d with 4b and 4c). This clearly shows
that the sulfate-rich component of the mixed
water is isotopically relatively heavy. We
suspect that the upward spike in ¥18O on day
176 (Fig. 4d) is not real because it is not
accompanied by any other chemical anomaly.
In contrast, the small ¥18O spike on day 190
probably is real because it is accompanied by
distinct ¥D, Cl/SO4 and Na/K anomalies
(Fig. 4c). The decreasing trend in ¥D from
day 173 to 180 is not accompanied by other
major chemical changes, and, at this time, we
can offer no explanation for this anomaly. On a diagram showing ¥18O versus ¥D
for all Norris waters, Cistern Spring waters
show considerable variation, but all plot with
the heavier waters (Fig. 12c). These isotopic
variations probably result in part from
different amounts of mixing of different
waters, and in part from different degrees of
boiling with steam separation during upflow.
Double Bulger Geyser's waters
exhibited dramatic and unexpected changes
over the monitored period. Prior to the
annual disturbance chloride values were very
irregular, ranging up and down from a
minimum of 665 ppm to a maximum of 849
ppm (Fig. 5b and Table 1). Other dissolved
constituents, including Br, Na, K, and SO4,
show similar peaks and valleys when plotted
versus Julian day (Fig. 13), so we conclude
that those ups and downs in Cl concentration
were real, and probably resulted from
different degrees of steam separation during
flow to the surface. However, there was a
general trend of decreasing Cl/SO4 prior to
the annual disturbance (Fig. 5c), so some
subsurface mixing of different waters was
occurring. On the day of the annual
disturbance there was an increase in Cl and
Cl/SO4.(Figs. 5b and 5c). This is contrary
to what was observed in the Cistern Spring
and "Wistful" Geyser waters (Figs. 4b and
4c and Figs. 7b and 7c). Shortly after the
start of the annual disturbance, the water level
in Double Bulger dropped to a level too low
to sample, although discharge of steam
continued, and upward splashing of water
from deep within the vent could be seen. By
eleven days after the start of the annual
disturbance the water level once again was
high enough to sample. By then Double
Bulger Geyser's Cl had attained a value of
1444 ppm, which was about twice as high as
any previously recorded in a Yellowstone
thermal water (extensive tabulations of
chemical analyses are given in Gooch and
Whitfield, 1888; Allen, and Day, 1935;
Rowe and others, 1973; Thompson and
others, 1975; Thompson and Yadav, 1979;
and Thompson and DeMonge, 1996).
However, at the same time the Cl/SO4 ratio
and pH both decreased (Figs. 5a and 5c),
indicating addition of an H2SO4 component
to the water. The addition of an H2SO4 component after the start of the disturbance
also is evident in Figure 13, which shows a
dramatic increase in SO4 while the major
cations, Na and K, continue to reflect mainly
variations in Cl and Br.
Figures 14 and 15 are plots of ¥D and
¥18O respectively versus Cl for Double
Bulger Geyser's waters compared with data
for the other springs collected in 1995. Also
shown are data for Echinus Geyser which
discharges the largest volume of acid-sulfatechloride
water of any feature at Norris. For
comparison, Y-12 drill hole water
(condensed total flow) and local meteoric
water also are plotted in Figures 14 and 15.
In these figures the labeled diamonds show
expected compositions of the residual liquid
at atmospheric pressure after adiabatic
decompressional boiling of water fed by
reservoirs at 300 ¡C, 325 ¡C and 350 ¡C.
For computational purposes, the isotopic
composition and Cl concentration of the
water in the deep reservoir before boiling
were assumed to be similar to that of the
condensed water-steam mixture discharged
from the Y-12 drill hole. The above two
figures show that the isotopic compositions
and Cl concentrations in the waters
discharged by Double Bulger Geyser 11 and
16 days after the start of the annual
disturbance cannot be explained by a simple
adiabatic decompressional model with singlestep
steam loss at atmospheric pressure. The
isotopic changes accompanying single-step
steam loss at atmospheric pressure were
calculated using methods outlined in
Truesdell and others (1977). Figures 16 and
17 are similar plots of ¥D and ¥18O
respectively versus SO4. The Cl and SO4
versus isotopic composition plots all show
widely ranging anionic concentrations while
the isotopic compositions show relatively
little variation. These observations preclude
evaporation at atmospheric pressure of
standing water in a non-discharging pool as a
way to explain the high concentration of Cl
that was obtained after the start of the annual
disturbance because that process would
produce residual water that becomes
progressively heavier as Cl and SO4
concentrations increase.

One way to explain the data would be by
a refluxing of some of the cooled and
relatively dense boiled water back
underground where it would again be heated
to high temperature. This would occur
during the non-discharging period just after
the annual disturbance, when the water level
in the vent was too low to sample. The
previously boiled water, which had counterflowed
downward, then undergoes additional
decompressional boiling during subsequent
upflow. In this model, the refluxed water
must be re-heated to greater than 220 ¡C in
order to minimize the partitioning of lighter
isotopes into the vapor phase during
subsequent "reboiling". Note that above 220
¡C, boiling causes the residual liquid water to
become lighter in ¥D and only slightly
heavier in ¥18O (Truesdell and others,
1977).
Another way to attain the anomalously
high concentrations of non-volatile dissolved
constituents, and the isotopically heavy
water, would be by non-adiabatic evaporation
utilizing excess heat in the reservoir rock.
For example, if the discharge up the main
channel of flow occurred at a faster rate than
fluid could migrate into that relatively open
channel from the surrounding rock, there
would have been a drop in fluid pressure at
all depths within, and immediately adjacent
to, the main channel of upflow. This would
result in boiling-point temperatures in the
neutral-chloride and acid-chloride-sulfate
reservoirs becoming less than the prevailing
reservoir temperatures. Boiling of water "in
place" would then occur in the reservoirs,
producing steam in excess of that obtained by
adiabatic decompressional boiling, until rock
temperatures declined to a point where rock
temperature-fluid pressure conditions were
again compatible with a new, adiabatic
"boiling-point" curve.
According to the above models, Double
Bulger Geyser's waters should have become
isotopically heavier in ¥18O, and perhaps
lighter in ¥D (if boiled in place above 220
¡C), as the Cl concentration increased.
However, the first sample taken after the start
of the annual disturbance was lighter in
¥18O, but heavier in ¥D relative to the preceding sample (Fig. 5d). But, it was not
lighter in ¥18O relative to the composition of
Y-12 water (Fig. 15). We suggest that soon
after the start of the annual disturbance the
main source of the discharged water shifted
slightly to produce fluid isotopically more
closely related to Y-12 water. Subsequently,
as subsurface pressures declined further, a
SO4-rich water with a still different isotopic
composition began to enter the discharge
channel and mix in various proportions with
the Cl-rich water (Fig. 18). We suspect that
SO4-rich water enters the channel of upflow
at a shallower depth than the Cl-rich water
entry.
On a diagram showing ¥18O versus ¥D
for all Norris waters, Double Bulger
Geyser's waters, like Cistern Spring's
waters, plot with the isotopically heavier
waters (Fig. 12d). However, most (but not
all) of Double Bulger Geyser's waters plot
along the right side of the data cluster, while
most (but not all) of Cistern Spring's waters
plot toward the left side of the data cluster
(Fig. 12c).
On a plot of Cl versus SO4, most waters
collected from Double Bulger Geyser before
the start of the annual disturbance fall along
line B, which is radial from the origin, and
which passes slightly above the Y-12 water
data point (Fig. 18). However, the first two
samples of water collected from Double
Bulger Geyser have the lowest SO4
concentrations and plot close to line A, which
passes through the Y-12 water data point.
We note that the two data points that lie
closest to line A also have the lightest ¥D
values of any of the waters collected from
Double Bulger Geyser, and also are among
the three lightest in ¥18O (Table 1). The only
Double Bulger Geyser sample that is lighter
in ¥18O was the first sample collected after
the start of the disturbance (Table 1 and Fig.
18). Thereafter, the ¥18O of waters sampled
from Double Bulger Geyser became
progressively heavier as SO4 increased (Fig.
18). For comparison, Cl-SO4 relations for
waters from Cistern Spring and Porkchop
Geyser also are shown in Figure 18.

In striking contrast to the Double Bulger
Geyser waters, variations in composition of
Cistern Spring waters in the summer of 1995
lie along a single mixing trend with one endmember
water having about 550 ppm Cl and
60 ppm SO4, and the other about 350 ppm Cl
and 110 ppm SO4 (Fig. 18). There is no
sign in the Cistern Spring waters of the
evaporative concentration of non-volatile
components that is exhibited by Double
Bulger Geyser waters. On the other hand,
most of the compositional variations in the
Porkchop Geyser waters seem to be related to
differences in the degree of vapor or steam
loss from water that is compositionally
similar to Y-12 water (the data points plot
near line A in Fig. 18). However, there is no
indication in the Porkchop Geyser waters of
the high degree of high-temperature
evaporation that appears to be occurring in
the Double Bulger Geyser waters.
Porkchop Geyser waters exhibited a
small increase in temperature and Cl/SO4, a
small decrease in Na/K, and a significant
change in isotopic composition (¥D and ¥18O
became lighter) leading up to the annual
disturbance (Fig. 6). After the disturbance
there was a decrease in Cl/SO4 and pH,
indicating that a small increase in an acidsulfate
component was being incorporated
into the discharge. The water also became
isotopically heavier. At the same time the Cl
concentration and Na/K ratio increased, and
discharge temperature decreased. The
increase in Cl concentration and trend toward
an isotopically heavier water after the start of
the annual disturbance were probably mainly
the result of a significant decrease in the rate
of discharge. The slower rate of discharge
allowed evaporation from the surface of the
pool to affect water compositions. However,
on a graph of ¥18O versus ¥D the Porkchop
Geyser waters show only about a 2 %o
range in ¥18O and about an 8 %o range in
¥D (Fig. 19a), suggesting that atmospheric
evaporation effects were not extreme.
Previously, Fournier and others (1991)
proposed a model in which two reservoirs of
alkaline-chloride water at different
temperatures contribute water to Porkchop
Geyser 's discharge. A decrease in Na/K andincrease in the enthalpy of the discharged
water-steam mixture over a span of several
years resulted from drawing a larger
proportion of water from the deeper and
hotter reservoir. Our 1995 data do not show
the extreme variations in Na/K reported by
Fournier and others (1991) (Fig. 20).
Apparently most of the water discharged by
Porkchop Geyser in the summer of 1995
came from the deeper and hotter reservoir,
with the proportion of deeper and hotter
water becoming even larger (more K-rich
water) leading up to the annual disturbance.
After the annual disturbance the proportion of
the deep component again decreased slightly.
"Wistful" Geyser showed little chemical
or isotopic variation leading up the annual
disturbance (Fig. 7). On the day of the
disturbance there was an immediate drop in
pH and Cl/SO4 (Figs.7a and 7c). Cl
continued to decrease and SO4 to increase for
another 10 days, reaching respective minima
and maxima on day 220 (Fig. 7b). By day
240 the concentration of Cl was larger, and
the Cl/SO4 was about the same as prior to the
start of the annual disturbance. The pH
change was as expected, generally decreasing
in proportion with increasing sulfate (Figs.
7a and 7b). As was the case for Cistern
Spring waters, in the "Wistful" Geyser
waters both ¥D and ¥18O became heavier as
Cl decreased and SO4 increased (compare
Fig. 7d with 7b). On the other hand, the
Na/K data showed little variation except for
two anomalously low values on days 218 and
220 that also had the highest SO4 values
(compare Fig. 7c with 7b).
Isotopically, "Wistful" Geyser waters
became heavier in both ¥D and ¥18O as
Cl/SO4 decreased (Figs. 7c and 7d). On a
diagram showing ¥18O versus ¥D for all
Norris waters, "Wistful" waters exhibit
almost the full range in isotopic compositions
found in these waters (Fig. 19b). Prior to the
start of the annual disturbance all the data plot
with the isotopically lighter waters (Fig.
19b). In contrast, during the disturbance and
for a few days thereafter, "Wistful" waters
plot with the isotopically heavier waters,
generally trending toward the isotopic
composition of Vermilion Spring, a typical low-chloride, acid-sulfate, steam-heated,
non-overflowing pool. Like the results for
Double Bulger Geyser, the data points for
"Wistful" Geyser plot at the right side of the
data cluster. There is a general mixing trend
that extends from the Y-12 composition
toward an isotopically heavy acid-chloridesulfate
water.
The response of "Carnegie" Spring
water to the annual disturbance was very
slight. There was a general trend of
increasing temperature, Cl, and Cl/SO4 prior
to the start of the disturbance (Figs. 8a, 8b
and 8c). These trends reversed immediately
after the start of the disturbance. The
relatively large variations in Na/K ratios (Fig.
8c) are more apparent than real because the
"Carnegie" Spring waters contain relatively
little K. Therefore, small analytical
differences of less than 1 ppm (within the
analytical uncertainty) result in relatively large
differences in Na/K. Compared to predisturbance
values, ¥D became slightly
lighter after the disturbance, while ¥18O
became slightly heavier.
The "Carnegie" Spring waters have ¥D
values close to that of the local meteoric
water, and ¥18O values about 1.5 %o heavier
than Y-12 water (Fig. 19c). Even though the
"Carnegie" waters apparently issue from a
relatively low-temperature reservoir at about
190¡C (Table 2 and Fig. 11) they appear to
have attained isotopic as well as chemical
equilibration with the reservoir rock.
However, the ¥D values close to that of the
local meteoric water may be coincidental.
Isotopic exchange with rock may have
occurred mainly in a deeper and hotter
reservoir before the water moved up into the
shallower and cooler 190¡C reservoir, where
chemical, but not isotopic, re-equilibration
occurred.
Perpetual Spouter waters behaved very
similarly to the "Carnegie" Spring waters in
that there was very little isotopic and chemical
response to the annual disturbance. Relative
to the pre-disturbance composition, ¥D
recovered to slightly lighter values after the
disturbance while ¥18O remained nearly
constant. Like the "Carnegie" Spring waters, Perpetual Spouter waters have ¥D values
close to those of the local meteoric water, and
¥18O values about 1.5 %o heavier than Y-12
water (Fig. 19d). Na/K and Cl/SO4
variations probably reflect the analytical
uncertainty in the measurements.
Over the course of the monitoring period
Beryl Spring showed minor chemical and
¥18O variations, but significant ¥D changes
both before and after the disturbance (Fig.
10). However, like "Carnegie" Spring
waters, Beryl Spring waters issue from a
relatively low temperature reservoir
(discussed above). This reservoir is
probably relatively shallow and may not be
sensitive to changes that trigger annual
disturbance responses in the deep, hightemperature
reservoirs beneath Norris Geyser
Basin.

TRIGGERING OF THE ANNUAL
DISTURBANCE
Fournier and others (1986) noted that
several factors may be important in triggering
the annual disturbance, including self-sealing
by mineral deposition, clogging of channels
by clay suspended in flowing waters, large
geyser eruptions, seasonal changes in water
table, deformation and fracturing induced by
tectonic forces, and hydrofracturing. Of
these, for most years seasonal variations in
fluid pressure within the hydrothermal
system are probably most important.
Fournier (1983) pointed out that fluid
pressures measured at the bottoms of shut-in
wells drilled in thermal areas of YNP (White
and others, 1975) generally are close to the
pressures that would be exerted by overlying
columns of cold water. This strongly
suggests that there are relatively open
hydraulic connections between the
hydrothermal systems and the surrounding
low-temperature groundwater systems.
Accordingly, a possible explanation of the
annual disturbance involves changes in
reservoir pressure brought about by changes
in the potentiometric surfaces of the regional
cold water systems.
Temperature-depth curves in the central
part of the hydrothermal system, where
upward convective flow rates are greatest,
come to define boiling-point curvesappropriate for the fluid pressure profile.
While fluid pressures within the
hydrothermal "plumbing system" remain high
(controlled by the cold-water system adjacent
to the Norris hydrothermal system), waters in
shallower, relatively low-pressure reservoirs
are essentially held in check (unable to enter
the channels of upflow and mix with the
waters of deeper origin) because there is no
pressure gradient to drive this entry.
However, this situation is very unstable
because a drop in pressure in the cold-water
part of the system results in a drop in
pressure in the hydraulically interconnected
hot-water part of the system. In turn, this
decrease in pressure requires that boiling
temperatures at given depths decrease. But,
rock temperatures do not decrease as rapidly
as does the boiling-point curve. Liquid water
vigorously flashes to steam (non-adiabatic
evaporation at high pressure), energized by
the heat stored in rock. The boiling "in
place" and temporary increase in pressure
within relatively shallow, acid-water
reservoirs (rich in clay minerals) cause
"muddy" water and steam to be ejected from
these reservoirs into the main channels of
upflow. The annual disturbance is initiated.
The immediate, but temporary upward surge
in pressure at the top of the system is
followed shortly thereafter by a decline in
pressure deeper in the system. This is
because water from the deep reservoirs does
not enter the channels of upflow at as fast a
rate as water is propelled upward by
expansion of "excess" steam, and because the
increased proportion of steam to water
lightens the "load" exerted by the overlying
column of fluid. Where temperatures in
shallow reservoirs are less than the prevailing
boiling temperature, the general decline in
pressure within the hydrothermal system
allows waters in these initially lower-pressure
reservoirs to enter the main channels of
upflow without violent steam propulsion.
There they mix with waters ascending from
the deeper reservoirs, producing the mixed
waters seen in some Norris springs and
geysers.
The annual disturbance generally does
not occur in the spring of the year because
then the water table (potentiometric surface)
is generally high as a result of the melting of
thick covers of snow. At this time fluidpressure at given depths, both outside and
inside the hydrothermal system, is higher
than average. This situation allows the
boiling point curve to rise slightly, and there
is a heating of rock adjacent to the main
channels of convective upflow in the heart of
the hydrothermal system. As the summer
progresses, the water table drops slightly, the
rate of recharge of cold water decreases,
groundwater temperature adjacent to the
hydrothermal system increases slightly, and
the density of the pressure-controlling
column of "cold" water decreases. The
boiling-point within the hydrothermal system
drops in response to these changes and, in
some places, previously heated rock gives up
that heat by causing excess evaporative
boiling (flashing to steam) of ascending
fluids, as discussed above.
Beneath Double Bulger, evaporative
boiling is probably occurring simultaneously
in the shallower acid-chloride-sulfate and the
deeper neutral-chloride reservoirs as a result
of a drop in fluid pressure along the flow
path. Elsewhere, as at Cistern Spring and
"Wistful" Geyser, the main effect of a slight
drop in fluid pressure along the flow path
seems to be to allow larger proportions of
waters from shallower reservoirs (the acidchloride-
sulfate waters) to enter the channels
of upflow and mix with the waters of deeper
origin.
Annual disturbance phenomena may not
occur in other geyser basins in YNP because
the thermal waters in those basins are fed by
shallower, lower-temperature reservoirs
(180Ð215 ¡C at Lower and Upper Geyser
Basins compared to 270Ð310 ¡C at Norris
according to Fournier and others (1976) and
Fournier (1989). It is also possible that
annual disturbance phenomena do occur in
other Geyser Basins, but have not been
recognized because widespread shallow
reservoirs of acid waters are not present in
those basins to furnish the clay which makes
hot spring and geyser waters murky at Norris
Geyser Basin, heralding the onset of the
disturbance there.
SUMMARY AND CONCLUSIONS
At the time of the annual disturbance
there is widespread subsurface mixing of
shallow, acid-chloride-sulfate water
(reservoir temperature probably in the rangeabout 180Ð220 ¡C) with near-neutral,
chloride-rich water ascending from highertemperature
(270 ¡C to >300 ¡C), and
presumably deeper, reservoirs beneath Norris
Geyser Basin. In contrast, hot springs that
discharge near-neutral, chloride-rich waters
that have equilibrated with rock in reservoirs
where temperatures are 250 ¡C or less show
little chemical response to the annual
disturbance. Acid-sulfate-chloride water,
similar in composition to waters discharged
in Porcelain Basin and 100 Spring Plain,
underlies much of the Back Basin, but is
discharged there only after mixing with the
neutral-chloride water. Generally, this
occurs only for short periods of time after
annual disturbances.
The large volume of acid-chloride-sulfate
water discharged by Echinus Geyser in the
Back Basin is distinctly different chemically
and isotopically compared with the Porcelain
Basin and 100 Spring Plain acid-chloridesulfate
waters (Figs. 14Ð17). The low
chloride and isotopically relatively light
Echinus Geyser-type waters do not appear to
be mixing in significant amounts with the
neutral-chloride waters in the Back Basin.
In contrast with the acid-sulfate-chloride
waters discharged by Echinus Geyser, most
of the acid-chloride-sulfate waters at Norris
are isotopically relatively heavy in both ¥D
and ¥18O (Figs. 14Ð18). To date, all of the
waters collected from non-discharging,
steam-heated, acid-sulfate pools at Norris and
elsewhere in YNP have been found to be
isotopically heavy (for example, the
Vermilion Spring pool shown in Figs. 12 and
19). Some of this type of water is likely to
percolate into the ground and mix with other
subsurface waters.
The extremely high concentrations of Cl
and SO4 in waters collected from Double
Bulger Geyser after the start of the 1995
annual disturbance are probably the result of
excess boiling (more steam loss than can be
accounted for by adiabatic decompressional
boiling of a "once-through" fluid). Excess
boiling may have occurred as a result of
counter flow of previously boiled fluid back
underground where it was re-heated to over
220 ¡C. This water then boiled adiabatically a
second time during subsequent convective
upflow; the cycle may have been repeatedmore than once. Alternatively, and more
likely, a drop in fluid pressure all along the
channel of upflow resulted in excess boiling
of water in neutral-chloride and acid-chloridesulfate
reservoirs that supply water to Double
Bulger Geyser.
Triggering of the annual disturbance
probably is caused mainly by annual changes
in elevation of the potentiometric surface of
the cold-water system that surrounds the
Norris hydrothermal system, although
seismicity may be important sometimes. In
the spring the water table is high and rate of
downward percolation of cold water adjacent
to the Norris hydrothermal system is greatest.
At this time fluid pressures within the
interconnected cold-and hot-water systems
are highest. While high fluid pressure
prevails along paths or channels of upward
hydrothermal convective fluid flow, the flow
of acid-chloride-sulfate waters from relatively
shallow, low-pressure reservoirs into the
main channels of upflow is prevented
because there is no pressure gradient driving
that entry. Generally, by the end of the
summer months, there is a decrease in the
elevation of the local potentiometric surface
(the water table) sufficient to result in a
decrease in fluid pressure within that
hydrothermal system. In reservoirs at
various depths excess evaporative boiling is
induced which initially produces an upward
pressure surge at shallow depths, followed
by a rapid decline in pressure deeper in the
hydrothermal system. The drop in pressure
allows acid-chloride-sulfate waters to enter
the channels of upflow and mix with the
neutral-chloride waters of deeper origin.
Annual disturbance phenomena that are
easily recognized at Norris Geyser Basin may
not be easily recognized elsewhere in
Yellowstone National Park because (1) the
neutral-chloride waters that are fed from
relatively shallow, and lower-temperature
reservoirs in other geyser basins may not
react as vigorously in their steam production
as do the waters in the very high-temperature
Norris reservoirs in response to small
pressure changes outside of the hydrothermal
system, and (2) the clay that makes hot
spring and geyser waters become turbid at
Norris, heralding the start of the disturbance,
comes from acid altered rocks that are widelydistributed at intermediate depths at Norris,
and that are rare in other geyser basins.
Acknowledgments
We wish to thank the U.S. National
Park Service for allowing collection of the
samples. We also wish to thank Anne E.
Gartner of the U.S.G.S. for her great help in
keeping track of the samples and expediting
their distribution to the laboratories that
carried out the chemical and isotopic
analyses.
REFERENCES
Allen, E.T. and Day, A.L., 1935, Hot
springs of the Yellowstone National Park:
Carnegie Institution of Washington
Publication 466, 525 p.
Arn—rsson, S., 1983, Chemical equilibria in
Icelandic geothermal systems -
Implications for chemical
geothermometry investigations:
Geothermics, vol. 12, p. 119-128.
Fenner, C.N., 1936, Bore-hole investigation
in Yellowstone Park: Journal of
Geology, vol. 44, p. 225Ð315.
Fournier, R.O., 1979, A revised equation for
the Na/K geothermometer. Geothermal.
Resources Council Transactions, vol. 3,
p. 221Ð224.
Fournier, R.O., 1981, Application of water
geochemistry to geothermal exploration
and reservoir engineering, in Ryback, L.,
and Muffler, L. J. P., eds., Geothermal
Systems: Principles and Case Histories:
New York, Wiley, p. 109Ð143.
Fournier, R.O., 1983, Active hydrothermal
systems as analogues of fossil systems,
in The role of heat in the development of
energy and mineral resources in the
Northern Basin and Range province:
Geothermal Resources Council Special
Report No. 13, p. 263Ð284.
Fournier, R.O., 1985, The behavior of silica
in hydrothermal solutions, in Berger,
B.R., and Bethke, P.M., eds., Geology
and geochemistry of epithermal systems:
Reviews in Economic Geology, vol. 2,
p. 45Ð61.
Fournier, R.O., 1989, Geochemistry and
dynamics of the Yellowstone National
Park hydrothermal system, Wyoming: Annual Review of Earth and Planetary
Sciences, vol. 17, p. 13Ð53.
Fournier, R.O., 1991, Water
geothermometers applied to geothermal
energy, Chapter 2 in D'Amore, Franco,
coordinator, Application of Geochemistry
in Geothermal Reservoir Development:
Rome, Italy, UNITAR/UNDP Centre on
Small Energy Resources, p. 37Ð69.
Fournier, R.O., Kennedy, B.M., Aoki, M.,
and Thompson, J.M., 1994, Correlation
of gold in siliceous sinters with 3He/4He
in hot spring waters of Yellowstone
National Park: Geochimica et
Cosmochimica Acta, vol. 58, p.
5401Ð5419.
Fournier, R.O., and Potter, R.W., II, 1982,
A revised and expanded silica (quartz)
geothermometer: Geothermal Resources
Council Bull., vol. 11, no. 10, p. 3Ð12.
Fournier, R.O., and Rowe, J.J., 1966,
Estimation of underground temperatures
from the silica content of water from hot
springs and wet-steam wells: American
Journal of Science., vol. 264, p.
685Ð697.
Fournier, R.O., Thompson, J.M.,
Cunningham, C.G., and Hutchinson,
R.A., 1991, Conditions leading to a
recent small hydrothermal explosion at
Yellowstone National Park: Geological
Society of America Bulletin, vol. 103, p.
1114Ð1120.
Fournier, R.O., Thompson, J.M., and
Hutchinson, R.A., 1986, Fluctuations in
composition of Cistern Spring, Norris
Geyser Basin, Yellowstone National
Park, WyomingÑvariable boiling and
mixing 1962Ð1985: Extended Abstracts,
Fifth International Symposium on
WaterÐRock Interaction, Reykjav’k,
Iceland, International Association of
Geochemistry and Cosmochemistry, p.
206Ð209.
Fournier R. O., Thompson J. M., and
Hutchinson R. A., 1992, The
geochemistry of hot spring waters at
Norris Geyser Basin, Yellowstone
National Park, USA, in Kharaka, Y. K. ,
and Maest, A. S., eds., Proceedings of
the 7th International Symposium. on
Water-Rock Interactions, Rotterdam, Balkema, ISBN 90 5410 0753, vol. 2, p.
1289Ð1292.
Fournier, R.O., White, D.E., and Truesdell,
A.H., 1976, Convective heat flow in
Yellowstone National Park, in
Proceedings, Second UN Symposium on
the Development and Use of Geothermal
Resources: Washington, D.C., U.S.
Government Printing Office, vol. I, p.
731Ð739.
Giggenbach, W. F., 1988, Geothermal
solute equilibria. Derivation of Na-KMg-
Ca geoindicators: Geochimica et
Cosmochimica Acta, vol. 52, p.
2749Ð2765.
Gooch, F.A., and Whitfield, J.E., 1888,
Analyses of waters of the Yellowstone
National Park, with an account of the
methods of analysis employed: U.S.
Geological Survey Bulletin 47, 84 p.
Rowe, J.J., Fournier, R.O., and Morey,
G.W., 1973, Chemical analyses of
thermal waters in Yellowstone National
Park, Wyoming, 1960Ð1965: U.S.
Geological Survey Bulletin 1303, 31 p.
Thompson, J.M., and DeMonge, J. M.,
1996, Chemical analyses of hot springs,
pools, and geysers from Yellowstone
National Park, Wyoming, and vicinity,
1980-1993: U.S. Geological Survey
Open-File Report 96-68, 66 p.
Thompson, J.M., Presser, T.S., Barnes,
R.B., and Bird, D.B., 1975, Chemical
analyses of the waters of Yellowstone
National Park, Wyoming, from
1965Ð1973: U.S. Geological Survey
Open-File Report 75-25, 59 p.
Thompson, J.M., and Yadav, S., 1979.
Chemical analyses of waters from
geysers, hot springs and pools in
Yellowstone National Park, Wyoming,
from 1974Ð1978: U.S. Geological
Survey Open-File Report 79-704, 49 p.
Truesdell, A. H., 1976, Summary of section
III geochemical techniques in exploration,
in Proceedings, Second UN Symposium
on the Development and Use of
Geothermal Resources: Washington,
D.C., U.S. Gov. Printing Office, vol. I,
p. liii-lxxxix.
Truesdell A. H., Nathenson M., and Rye,
R.O, 1977, The effects of subsurface
boiling and dilution on the isotopic
compositions of Yellowstone thermalwaters: Journal of Geophysical
Research, vol. 82, p. 3694Ð3704.
Trujillo, P.E., Counce, D., Grigsby, C.O.,
Goff, F., and Shevenell, L., 1987,
Chemical analysis and sampling
techniques for geothermal fluids and
gases at the Fenton Hill Laboratory: Los
Alamos National Laboratory Publication
LA-11006-MS, 84 p.
White, D.E., Fournier, R.O., Muffler,
L.J.P., and Truesdell, A.H., 1975,
Physical results of research drilling in
thermal areas of Yellowstone National
Park, Wyoming: U.S. Geological Survey
Professional Paper 892, 70 p.
White, D.E., Hutchinson, R.A., and Keith,
T.E.C., 1988, The geology and
remarkable thermal activity of Norris
Geyser Basin, Yellowstone National
Park, Wyoming: U.S. Geological
Survey Professional Paper 1456, 84 p.

Figure Captions
Figure 1. Sketch map of Norris Geyser Basin, Yellowstone National Park, showing
topographic features, locations of features that were monitored, U.S. Geological
Survey drill holes YÐ9 and YÐ12, and other selected hot springs and geysers.
Contours show equal altitude in feet.
Figure 2. Chemical and isotopic variations versus Julian day for Sulfur Dust Spring.
a) temperature and pH versus Julian day. b) Cl and SO4 versus Julian day.
c) Na/K and Cl/SO4 versus Julian day. d) ¥D and ¥18O versus Julian day.
Figure 3. Chemical and isotopic variations versus Julian day for "Black" Spring.
a) temperature and pH versus Julian day. b) Cl and SO4 versus Julian day.
c) Na/K and Cl/SO4 versus Julian day. d) ¥D and ¥18O versus Julian day.
Figure 4. Chemical and isotopic variations versus Julian day for Cistern Spring.
a) temperature and pH versus Julian day. b) Cl and SO4 versus Julian day.
c) Na/K and Cl/SO4 versus Julian day. d) ¥D and ¥18O versus Julian day.
Figure 5. Chemical and isotopic variations versus Julian day for Double Bulger Geyser.
a) temperature and pH versus Julian day. b) Cl and SO4 versus Julian day.
c) Na/K and Cl/SO4 versus Julian day. d) ¥D and ¥18O versus Julian day.
Figure 6. Chemical and isotopic variations versus Julian day for Porkchop Geyser.
a) temperature and pH versus Julian day. b) Cl and SO4 versus Julian day.
c) Na/K and Cl/SO4 versus Julian day. d) ¥D and ¥18O versus Julian day.
Figure 7. Chemical and isotopic variations versus Julian day for "Wistful" Geyser.
a) temperature and pH versus Julian day. b) Cl and SO4 versus Julian day.
c) Na/K and Cl/SO4 versus Julian day. d) ¥D and ¥18O versus Julian day.
Figure 8. Chemical and isotopic variations versus Julian day for "Carnegie" Spring.
a) temperature and pH versus Julian day. b) Cl and SO4 versus Julian day.
c) Na/K and Cl/SO4 versus Julian day. d) ¥D and ¥18O versus Julian day.
Figure 9. Chemical and isotopic variations versus Julian day for Perpetual Spouter.
a) temperature and pH versus Julian day. b) Cl and SO4 versus Julian day.
c) Na/K and Cl/SO4 versus Julian day. d) ¥D and ¥18O versus Julian day.
Figure 10. Chemical and isotopic variations versus Julian day for Beryl Spring.
a) temperature and pH versus Julian day. b) Cl and SO4 versus Julian day.
c) Na/K and Cl/SO4 versus Julian day. d) ¥D and ¥18O versus Julian day.
Figure 11. Temperatures of reservoirs estimated using the Na/K geothermometer of
Giggenbach (1988) compared with estimated reservoir temperatures using the silica
(quartz) geothermometer of Fournier and Potter (1982). Circles plot at silica
temperatures estimated assuming adiabatic cooling and x's plot at silica
temperatures estimated assuming conductive cooling.
Figure 12. ¥D versus ¥18O for all Norris waters compared with a) Sulfur Dust Spring
waters, b) "Black" Spring waters, c) Cistern Spring waters, and d) Double Bulger
Geyser waters.
Figure 13. Variations in Cl, Na, K, SO4, and Br versus Julian day for Double Bulger
Geyser waters.
Figure 14. ¥D and Cl relations in Double Bulger Geyser Sulfur Dust Spring, Echinus
Geyser, and Y-12 waters compared with all other data for Norris waters.
Diamonds show calculated compositions of Y-12 waters after adiabatic
decompressional boiling from the indicated initial reservoir temperatures with single
stage steam separation at atmospheric pressure.
Figure 15. ¥18O and Cl relations in Double Bulger Geyser , Sulfur Dust Spring, Echinus
Geyser, and Y-12 waters compared with all other data for Norris waters.
Diamonds show calculated compositions of Y-12 waters after adiabatic
decompressional boiling from the indicated initial reservoir temperatures with single
stage steam separation at atmospheric pressure.
Figure 16. ¥D and SO4 relations in Double Bulger Geyser , Sulfur Dust Spring, Echinus
Geyser, and Y-12 waters compared with all other data for Norris waters.
Figure 17. ¥18O and SO4 relations in Double Bulger Geyser , Sulfur Dust Spring, Echinus
Geyser, and Y-12 waters compared with all other data for Norris waters.
Figure 18. SO4 versus Cl for Double Bulger Geyser , Cistern Spring, and Porkchop
Geyser waters. ¥18O values are shown for selected waters. See text for
discussion.
Figure 19. ¥D versus ¥18O for all Norris waters compared with a) Porkchop Geyser
waters, b) "Wistful" Geyser waters, c) "Carnegie" Spring waters, and d) Perpetual
Spouter waters.
Figure 20. K versus Na for Porkchop Geyser waters. Waters collected prior to 1985 all
plot below the dashed line.
Table 1. Results of chemical and isotopic analyses
Table 2. Estimated reservoir temperatures using silica and Na/K geothermometers