PHELIMINAHY HESUL T5 Of I)({ILUNG ANI) TESTING
IN THE PUNA GEOTHERMAL SYSTEM, HAWAII
J. L lovenlttl and W. L [)IOller
Diamond Shamroc" Thermal Power Company
333l Mendocino Avenue, Suite 120
Santa Rosa, California 95401
ABSTRAC'r
lJiulliolld Shuilirock Thermal Power Company, operator for Ihe Puna Geothermal Venture has dr illed
and tested two geothermal wells in the Puna geothermal sy!>tem, Hawaii. The wells were drilled to
depths of 7290 and 0005 feet and completed with
9-S/B inch prodoction casing to about 4200 feel
and 7 inch perforated liner to bollom. Preliminary
shorl term testing hus demonstrated 100% saturatt.:d steal" product iOIl (1185 Btu/lb) at certain
wellhead pressures. Cyclic flow with a minimum
55% steom quality has also been observed. Initial
estimate of non-condensible gas concentration is
0.2% by weight with an average H S concentration
2
of 1100 ppm by weight. The well flowing charocteristics appear to stem Irolll a high temperature
(greOler than 650 0 F) two-phose liquid-dominated
geolhefillal rescrvoir which contains a variable
slewn fnlctiofl. Product:d steam quality variation
at the surface in this fidd is attributed to both
Illechanicol and reservoir loclors.
The u<,olhefillul ~yslell1 uppeors tu lie within the
cast I{ilt Zone ul I<iloueu Volcano. It is a ulind
systelll Illuslwd by oVerlyillg groumJwalcrs alld effectively sealed by impermeable roci<s. lJusaltic
ll1aUllla stored in the rift ;lOne is postulated as the
heat source. The high tt:mperoture portion of the
geothermal reservoir begins at about 4000 ft:et and
extends to un unlmown depth.
The wells ure
interpreted to huve encountered multiple producing
horizolls and interzonal flow is evident. Saturation
conditions prevail below thl:! 9-5/8 inch production
ca~ing. The cri tical point of waler may bl:! surp()ssed at depth in this system.
Initiol lest resulb arc sufficiently encourcll)ing to
worrant on oddilional well by the Puna Geothennal
Venture to identily reserv0ir and well characteristics for electrical power plant consideration which
includes we/lfil:!ld developmcnt costs.
II~Tf(ODUCT ION
Tile PUIlU ~"ulhennal system is luc()led within tile
lower Ea:;t Hift Zone of I<ilouea Volcano on the
island of Hawaii (Figure I). Kilauea is one of the
world's most active volcanos. The Puna GeotherIllal Venture (PGV) wells are stepouts to the successful HGP-A well drilled jointly by the US De-
partment ot Energy, Slale of liuwuii um! University of Hawaii.
HGP-A was cOlTlplet.:d in 1976 to a total depth 01
61150 feet with a boltomhole ternperature of 676°F.
The well has been producing 110,000 Ills/hr at 166
psig wellhead pressure wilh approximately 4]%
stearn quality (Thomas, 1983) to a 3 MW turbine
generator in nearly continuous operation since
May, 1982. The well resul ts have been summarized
by Kihara et al (1977), Chen et 01 (1978), Stone and
Fan (1978)) nlolnos (1982), and Waibel (1983).
HGP-A demonstrated the feasibility of generating
electricul power frolll a Hawaii volcanic rt:sourct.:.
It is the objective of PGV to develop geothermal
energy in Hawaii lor reliable electric power production ()t a marketuble price. PGV consists of
Diwnond Shwnrock Thermal Power COinpony,
Dillingham Geothermal, Illc. Ulld Amlac Enerejy,
Inc. Competitor drilling in the region has reportedly found high . botlomhole telTlperutures but production hus not been demonstruted.
Hepurlcd Ilcrein ure SOllie hilJloIights 01 tile IJCV
well field activity. Thesl: dolO are inlt:'Jra"''' to
formulatt: a conct:ptuul model of the Puna GtOthermal System. We hope thot this overview will
assist the work being done in Ihe devdoJl,nenl 01
high temperature, two-phose geothermal syslt:rns
for electrical production.
GENERAL SETTING
East liift Zone of I(iluuea is one 01 the conduits lor
the laterul miuration of bu~altic 1110'.11110 frUin Ihe
holding chambl:!r bent:ath the volcano's sUfnlni t coldera. It is manifested at the surface as a linear
belt, 2 - 3 km wide, consistillg of linenr (mc/ open
fissures, foults, SIIiUl/ grabens, pi t crllto:rs, COlles
ond venls lor nurlierOlJS eruptiolls. III I/Ie lower
portion of the rift zone eruptions hove occurred as
recent Iy os 1740, 1840, 1955, 1960 and 1961. The
r i f t zone is a cons I ruet ional ridge some 150 - 1500
leet above the adjoining terrain tlirouuhuul its
length except in its lowerrllost portion where tht:
ridge disuppears into a series of grabens and sputter deposits (Moore, 1983). The successful drilling
actitivy to dote lies in the transition area froln the
constructional ridge to the more subdued topography. Underlying the surfoce expression 01 the
East Rifl Zone is a much brooder (5-15 mile) dike
complex inferrt:d on the basis of gravity and magnetic duta (Furumoto, 1978).
This complt:x is
thought to con~i~t of an aggregate of closely
spaced parallel to sub-parallel, vertical to steeply
dipping dil<es whose top is in general about 7600
feet below the surface.
The dikes intrude a
sequence of Mauna Loa and Kilauea lava flows.
This complex is reported (F urumoto 1978) to be
locally above the Curie Point (IOOO'F) and in places
may even approach the melting point of basalt
(about 1900·F). Petrologic studies of lavas in the
rift zone indicate the presence of differentiated
tholeii te which strongly suggests the existence of
subsidiary magma chambers. The Puna geothermal
system overlies one such orca (Moore, 1983). Sufficient heat to drive a geothermal system is clearly
indicated.
The Puna system is considered a blind geothermal
prospect because no surface mani festations exist in
the area except for several hot springs discharging
along the coos t some two mi les to the south and
for isolated steam vents within the rift zone which
are associated with recently active fissures. The
lock of surface- lIlanifestations, in spite of the
tremendous heat flux potential created by the dike
complex, is at tributed to a vigorous, cool groundwater sy~telll which "hydraulically fllasks" the reservoir and to a relatively impermeable seal.
Annual rainfall in this area is about 120 inches
which immediately infiltrates into the ground; virtually no standing water bodies exist. Groundwater
residence time is reported to be on the order of
years (Kroopnick et 01, 1978). Addi tionally, very
recent lava flows could easily cover any surface
evidence of a geothermal system.
Hl.!view of public dOlnain information which has
been rl.!cently sumrnarizt:!d by Thomas 1984, indicated that the Puna system may be localized in the
eastnarthea:;t trending rift zone at the structural
intersection with 0 north-northwest trending transver:;e faul t (Figure I). The successful results of
the HGP-A and PGV wells support this hypothesis.
PGV WELL RESUL TS
8005 feet on 2 April 1982. fhe type of drilling
fluid used wa, as in KS-I except that at depths
greater than 6800 feet a very light saltgel mud was
continuously utilized to Ii ft cullings to the surface.
Other than severe lost circulation problems in the
upper portions of both wellbores (in the highly
permeable subaerial basalts), no significant drilling
problems were encountered. Although not realized
until after flow testing, cementing of the casing
strings was apparently less than optimum in the
shallow lost circulation zones. Both wells are
completed with two ANSI 900-series wellhead
gates. The 9-5/8 inch production casing in KS-I
and KS-2 was run from surface to 4072 and 4209
feet, respectively. An uncemented 7-inch perforated liner was stood in the 8-3/4 inch (KS-I) and
8-1/2 inch (KS-2) wellbores.
Li tholoyy, Mineral AI teration and Lo:; t Circulal ion
Zones
The wells penetrated a sequence of subaerial basalt:>, subaqueou:> basalts, basalt tephra and wbvolcanic intrusive;, which increa~e in frequency
with depth. Preliminary minerological alteration
studies (ColumbiCi Geoscil.!nce, 1982) reveal a variable and complicated section.
Three specific
types of alteration are recognized: (I) deuteric
(vapor phase), (2) contact metamorphic, and (3)
hydrothl.!rrnal. The latter two processe:; totally
overprint the deuteric 01 teration below 2500 feet.
Discriminotion between contact metamorphic effects and hydrothermal alteration is difficult since
both processes alter the rock to a low grade
chlorite-albite gr<:enschist faci<:s.
The original
rock texture is largely obliterated by alteration
below 4500-4700 feet. Barring the deuteric alteration assemblage, the wells can be characterized
as follows:
0- 3000':
Virtually free of alteration
3000-4000':
Localized moderate alteration (3060% of cuttings altered) with rare
highly altered zones (greater than
60% of cuttings altered).
11000'- TO:
Cornmon rnoder(lte alteruliofl wilh
localized zones of highly altered
rock interspersed wi ttl occas~ional
zones of fresh, unaltered rock.
Cornnlents
The PGV :;Iepout wei!:;, Kapoho 5tate-1 (1<5-1) and
Kapoho 5tute-2 0<5-2) were drilled in the rift neur
the trace of the 1955 fissure eruption (Figure I).
In general, the ca:;ing programs, lithologies and
mineral alteration suites, lost circulation zones,
available temperature/pres:;ure surveys and flow
test results are very comparable in both wells.
However, the lot ter two data set:; are not complete
for well. The apparent comrnonality between these
wells has allowed as a first approximation, data set
intercharlgt:! for interpretational purposes.
Drl.!!l!..l9.. md Comple!ion
K5-1 was completed on 12 Novernber 1981 to a
total depth of 7290 feet after 65 days of drilling
operations. The top hole was drilled with mud;
water was used a:; the principal drilling fluid in the
reservoir section. K5-2 was dri lied almost irnmec.iately after KS-I, without any appreciable intervening flow tests. With only 56 days of drilling
operations, K5-2 was completed at total depth of
This al teral ion rnineralogy suite is comparable to
that reported by Waibel (/983) for HGP-A where it
begins a thousand feet shallower. liowever, unaltered intervals below 4000 fed (reservoir section) in the PGV wells are in marked contra~t to
the HGP-A which shows fairly complete alteration
from 2000 feet to total depth. The~1.! zones of
fresh rock are thought to represent, at least in
part, recent int rusives most likely related to thl.!
1955 fissure eruption (Figure I) which have not had
enough time to become hydrothermally or thermally altered (A. Waibel, pers. communication,
1984). They could also represent large blocks of
impermeoble host rocks.
Detailed pt:!trological
studies of the wellbore cull ings are to be conducted.
Lo~t circululion u~ UII inJiculor ot pennt:ubility can
be subdivided into four discrete zones: (I) O-ISOO
feet: total loss; (2) 1500 - 2000 feel: illtennittent
and minor lossesj (3) 2500 - 4000 feet: virtually no
loss; and (4) 4000 feet - total depth: intermi ttent
and generally minor losses. The only significant
lost circulation zone in the reservoir section occurs
below 7000 feel.
Since idt!ntification of hydrotht:rrnal alteration is
not unequivocal, the degree of alteration rather
than a specific mint!rul or mineral assemblage has
been used in conjunction with lost circulation
"one~, flow te~t und other dota to identify a
minimum of three production zones. These production zones can occur in intra-flow boundaries,
fractured flows (and dike!!") and intrusive contacts.
Fracturing is principally fault induced.
Figure 2 illustrutes It:lIlperalure/pressure profiles
for the pev wellS' and HGP-A. Hydraulic masking
i:i evident trolll the surface to 1500 - 2000 feet
(tap of busal water is about 600 feet). Below this
one a rdatively impermeable caprock extending to
2250 feet in HGP-A and to 4200 feet in KS-2 is
suggested by the extremely high conductive gradient. Howt!vt!r, while a caprock is thought to
exist above the reservoir based on the absence of
significant lost circulation and mineral alteration,
close examination of the data strongly suggests
that the observed wellbore temperature/pressure
profiles are dominated by circulation in the perforated liner interval (i.e., interzonal flow dt!scribed
below).
A maximulll telnperolurt! and pre~sure of 648'F and
2120 psig were meosured in KS-2 ut a depth of
5500 feet wht!re the survey tool was blocked presumably by rock bridges. A pressure gradient af
0.248 psi/ft in the perforated liner interval indicates boiling conditions and a liquid-dominated
geothermal system. The open hole interval is
es~entially at saturation both .-£!:Jar to and imm:
edlate~afler the first series of flow tests. Extrapolation of these doto suggests that supercritical
fluids could be present at depth in this system.
This is consistent with the magmatic heat source
postulated to exi~t in the rift.
KS-I show~ u relut ively isothermal gradient in its
perforated liner intt!rval. A maximum temperature
of 643'F was recorded at 6400 feet where the
survt!y tool wos blocked presumubly by rock
bridges. While pressure data has not been obtained
in this well saturation condi tions are assumed
based on flow test results given below.
The
elevated tCI'nperatures/pressures in KS-2 are attributed to its greater depth and possibly additional
production zones beneath the total depth of KS-I.
Similarily, the higher temperature/pressure and
saturation conditions observed in the PGV wells
relative to HGP-A is interpreted to result primarily frolll a production zone(s) below total depth
of HGP-A. Additionally, Rudman and Epp (1983)
have shown that proximity to a hot dike can have a
significant effect on a well's temperature profile.
HGP-A from
The greater lateral distance of
the recent 1955 fissure eruplioll may ubo be u
foclor.
As staled (jbove, interLonul flow is operalive in
open-hole interval in both PGV wellbores. Higher
temperature/pressure fluids enter the wellbore uelow 7000 feet and exit through a permeable horizon
just below the casing shoe. As this situal ion
proceeds in the shut-in mode, the perforated liner
interval is brought to saturation condi tions. This
upfJow disguises the true wdlbore temperature/pressurt! profile in these wells. A gas column
collects below the wt:llhead which drives the liquid
level in the wellbore down to the first pt!nneable
horLwn prewmably at the base of the 9-5/8 inch
casing. At this point gas does not collect in the
wellbort! but exits through this permeability zone
(Grant, 1979).
When this occurs the wellhead
pressure is stable and is referred to by us as the
equilibrium shut-in wellhead pressure; i I attained
1435 psig in KS-I.
Post Completion Probh.!l)1s (Jnd f"!~,!~!'Je~~Fe~ult~
Tht! drilling and completion of tht:se wdls were
relatively straightforward. Flow testing proved to
be altogetht!r different. Hostile wellbore conditions in the wells led to several different types of
wellbore obstructions (e.g., rock bridges, lost survey tools, a 270 foot fish in KS-J). Thermal
cycling has occurred with a telnperature flux of
4 7S'F, in the shallow wellbore, between shut-in and
flow conditions. The accumulated gas column has
allowed H2S stress corrosion in the upper portions
of the weTlbore where the vigorous groundwater
regime is active. Atmo~pheric venting of the wells
to keep the well bores hot and free of gas was not
an environmentally acceptable opt ion because of
the high levels of H2S that would be discharged.
Flow testing was interrupted oy wellhead noise
levels approaching 120 decibels, high H S concen2
trations and porticulate matter eroding surface
equipment. To reduce the noise levels a rockfilled,
pit muffler (15 x 15 x 15 feet) was constructed to
lower effluent venting velocity. An H S abate2
ment system was integrated into the surface test
equipment 10 lowt!r emitted H S levels to within
regulatory standards and permit conditions. Continuous flaw eliminated the particulate matter
problem.
Ini tial lihort term (hourl» vertical venlill~ of 1<5-2
indicated saturated to super iteait:d slearn production in less than one hour flow time. The well
displayed the followillg gelleral charCJcterislics
upon opening:
(I) gas cap is dischorgedj (2)
dischurge of the liquid column uccompanied by a
high temperature/pressure transient (maxirnum
levels were 546'F and Il l54 psig on a 4 inch line)
and abundant iron-sulfide (pyrite?) and other particulate matlt::q and (3) discharge of saturated
steam followed by superheated steam.
Each well underwent four flow tests. The initiol
three tests conducted on a 10" blooic line, pitmuffler and H S abatement system yielded con2
flicting results.
A steel vessel separator was
brought in to accurately measure the sttlom and
liquid phases. A five day test on KS-2 indicated
100% ~teUlIl ~uturated flow at u rate of 33000
Ibs/hr and 173 p~ig wellhead pressure. At lower
wellhead pressures liquid from an identified leak in
the 9-5/8 inch casing became entrained in the
flow. This leak developed some time after the
initial vertical venting. A thirteen day test of 1<5I indicated essentially 100% saturated steam at a
flow rate of 72000 Ibs/hr and 120 psig wellhead
pressure (1-1. Dykstra, written communication,
1983). Total non-condensible gas concentration is
about 0.2% by weight comparable to HGP-A
(Thomas, 1982). H S concentrations are about
2
1100 ppm by Weight. F lashing in the reservior is
c1eurly tukinv pluce.
Figure 3 illustrates some of the KS-I data obtained
during the last 200 hours of almost uninterrupted
flow.
Cyclic flow is generally present at all
wellhead pressures but is most prolninent at wellheod pressures greoter thon 170 psig, where stearn
<juali ty drops frolll aboul I (JO% to about 55%.
Liquid lIow rute incrt!a~es but steurn flow role does
not chul1!Je appreciubly. Cyclic flow with a much
~horter period IoIJld lero to trace liquid occurs at
wellheod p((~sst)res less Iholl 170 psig. This phenomenoll wo~ ubo visuolly observed by one of the
aUlhors while the well wus discharging to atrnosphere. The flow went frorn being slighlly wet to
dry to :.uperheoted on undefint:d number of tirnes
during a one minute interval. For all practical
purposes, however, Ihe well flowed dry saturated
stearn at this tirne.
Grant et 01 (1979) has allributed cyclic flow in
wells to communication between multiple producing lanes significantly separated in depth from
each other. The authors agree with this interpretation but the above data would also suggest another. During the KS-I test a 270 foot fish was
present in its 7-inch liner. It would seem that the
fish would be a significant obstacle to flow from
production zones below it. Qualitatively, while
flow in the annulus area is taking place, it is not
considered sufficient to account for the liquid flow
rate observed. Only one productive ,wne is thought
to exist above the fish.
Either on addi tional
unrecogniLed production lone(s) is present above
the fish or varying wellhead pressure can couse
changes in the flashing characteristics in a single
producing lone. Further work is needed to resolve
this point. Another phenomenon indicated by the
KS-I dalo is thot initial totol stearn flow rate is
somewhat propurt ional to shut-in time. This is
interpreted as indicating that the fluid producing
perrneable zones oround the well bore are lorger
than the smallest conduit feeding them. Thus,
when the well is shut-in, the high permeabillity
zone around the wellbore undergoes recharge.
Aolh PGV wells have been suspended with deep
drilluble cement plugs placed in the 1700-2500 foot
interval of the production casing. This is a safety
precaution to ovoid high shut-in wellhead pressures
and H S stress corrosion.
2
DISCUSSION
The PGV well doto are complicated by a variety of
factors. The wellbore temperature/pressure profiles do not provide any detail on the in-situ
re~ervoir seclion.
However, soturation i:, considered to be the general reservoir thermodynamic
state. Upon opening the well to flow, the ensuing
pressure drop in Ihe wellbore in this very high
temperature environment allows flashing in the
resersoir. Production of superheated steam, 100%
saturated steam, or two-phase flow can toke place
as a function of wellhead pressure. A dry stearn
flow has been observed in a very short period of
time in these wells. Comparable behavior has not
been previously reported in any other two-phose
liquid-dominated systern. A possible explanation
for this phenomenon may be found in Dunn and
Hardee (1981). They have shown thaI the heal
transfer rates drarnatically increase in a permeable
medium in the vicinity of the critical point. The
role (if any) of interzonal flow on production of
100% soturated steam f low at the surface along
with impact of 0 fish on the KS-I test dolo is not
clcorly idenlified. Unreporled duta suggesls the
shallowest production Lonc in K5-1 may produce 0
high quality Iluid ill the absence of inteuonal flow.
Tht: producing behovior of tlte PGV wells "oric>
frolll HGP-A located sOlne y~ to Yl mile to the
south-southwest (Figure 2). While the pruduction
choracteristics of KS-I are markedly different
from HGP-A at wellhead pre~sl)res lower Ihon 170
psig, they becolne comparable at higher wellhead
pressures (Figure 4).
The 57% liquid fraction
produced in HGP-A is attributed to: (J) its perforated liner interval between 4000 and 2900 feet
allowing shallower lower temperature and pressure
fluid production; (2) its completion near the edge
of the field; (3) its shallower total depth; (4) the
natural variability in 0 two-phase geothermal
system (discussed below); (5) its lateral distonce
f rom the 1955 fissure; or (6) some cornbinat ion of
the above.
Figure 5 present~ u conceptual model of the Puna
Geothermal system. The Puna reservoir lopped by
three productive wells is considered to be a very
high temperature, two-phose liquid-dorninated
system with 0 vorying stealf) fraction confined
within the rif t. The reservoir is kept in i Is state by
the very high heat f low within the ri ft lone, by on
effective impermeable rock ~eal, and the lack of
significant venting to the surface. There is no
evidence of a vapor cop. Consequently, verticol
permeability is thought to be slllall relalive 10
horiLontal permeability. A conden:.ale loyer does
not seern to be present ci ther.
lieot flow is
thought to be primarily in the vertical direction
but lateral contributions from dike swarms ure also
possible. The vigorous, cool groundwater system
acts in conjunction with the caprock to keep the
system effectively hidden at the surfoce. Where
the cap is locally broken by structure leakage
would be relatively minor and transient in nalure
as the leaking ~tructure self-~eals due to hydrothermal alteration. This system is significantly
overpressurized relative to conventional boiling
point curve versus depth relotionships (J. Hebein,
personal communication, 1984) for reasons cited
above. The impact of a supercritical fluid (if it
does exist) on the system is not treat cd.
CONCLUSIONS
The PGV wells ~how many of the charocteristics
described by Stt!fans~on and Steingrimsson (1982)
for wells topping two-phose reservoirs. It is not un
common to find wells in high temperature, volcanic
systems (e.g. Krafla, Tongonan, etc.) to exhibit
enthaplies above that for hot liquid water. The
PGV welldata are the first we are aware of in the
literature that indicates early, dry stearn production from such a reservoir. These data provide
some definition of the factors that influence steam
quo Ii ty at the wellhead in two-phase geothermal
reservoir: (I) well completion interval (e.g. total
depth and ca~ing gt!ornetry); (2) number of producing zones opened to the wellborej (3) the thermodynamic stott! of these producing zones (temperature, pressure and stearn fraction); (4) the permeobility of these zones; and (5)
the relative
penneabilities to stearn and water of these zones.
The high stearn quality of the produced fluids in
the PGV wells has enhanced the feasibility of
geothermal electrical power production In the East
Rift Zone of I<ilauea Volcano. If the proposed
system model is verified by future well results we
would then expect the reservoir under exploitation
to become in a very short period of time a typical
vapor-dominated system not unlike The Geyers in
the U.S. and Karnojang in Indonesia. We speculate
that the Puna system may be analogous to the
paleo "Geysers" geothermal system (two-phase
liquid-dominated) before it became the vapardominated ~ystem observed today.
Wellbore complicotions (e.g. hostile environment,
obstructions, interzonal flow, etc) have prohibited
complete definition of the wells and reservoir
performance both with time and varying wellhead
pressures. Technical challenges and higher costs
are indicated both for exploration and development
wells.
A revised well program with upgroded
casing, cementing and completion procedures has
been prepared to best fit the unique characteristics
of the Puna geotherrnal system. An improved well
te~t program to minimize flow interruptions and
maximize data recovery has been similarily designed. A third well will be drilled before mid1985.
Acknowl l!dgeml!nts
We thank the Puna Geotherrnol Venture for allowing tht! release of this paper. A special thanks is
extended to two recent employees of Thermal
Power Company: Messrs. J. Hebein ond K. Goyal,
whose helpful commentaries and critiques have
assisted the preparation of this document.
R£fERENCES CITED
Chen,
6.H., Kihara, D.H., Yuen, P.C., and
P. K., 1970, Well test re~ulh from
HGP-A: GRC Trons., 2, p. 99-102.
Takahu~ki,
Columbia Geo~cience, 1902, Lithology and alteration of drill holes Kapoho State III and Jf.J.., Puna
District. Hawaii: Unpubli~h"d report prepored for
Thermal Power Company.
Dunn, J.C. olld I-Jordec, H.C., 19111, Sup"rconvectin\l geolhe(rJlul zones, J. Volcanol. Geotherm
Res., II, p. 189-201
furumoto, S. D., 1978, Natur" of Ih" magma
conduit under the East liiit Zane of Kilauea Volcano, HawaiI: Bull Valcanol, 41-4, p. 435-453.
Gront, M. A., 1979, Int"rpretatian of dawnhole
measurt:ments in gealh"rmal wells: Report 88,
Applied Matht:matics Division, Deportment Scientific & Industriol Ht:st:arch, Wellinglon, New
Zeolond.
Grant, M.A., 1982, On Ihe exi~tence of two-pha,,,
fluid in good camlllunicotian with liquid water.
Prac"edings of tht: sixth workshop an geothermal
r"servoir engineering, Stanford Univtlrsity.
Grant, M. A., BixltlY, P. F. and Syms, M. C., 1979,
In,;tobility in w,,11 pt>ffarrnonce: GRC Trans., J, p.
275-278.
Kihura, D.; Chen, U.j Yuen, P. and Tokahashi, P.i
1971, Summary of HGPA well testing: Proceedings
of third workshap on guotherrnol reservoir engine"ring, Stanford Univ"rsity, p. 138-144.
Kroopnick, P. M., Buddernllicr, R. W" and Thomas,
D., 1978 Hydrology ond g"O chemistry 01 a
Hawaiian gtlothermal systdm: HGP-AI Hawaii
Institute of Gt:o!*lyslcs R"port HIG-7B-o.
Moor", R., 1983, Di~tribution of diflcr"ntiated
tholeiitic bosahs on th" lower East liift Zone of
Kilautla Volcano, Hawaii:
A possible guide to
gt:otht:rmal tlxploratian: Geology.
Rudman, A. J. ond Epp, D., 1983, Conduction
modds af the temperature distribution in tht: East
Rift Zone of KilaUt:a Volcanol J. Volcano I Geothermo Res., 16, p. IB9-204.
Stt:fan,;:.on, V. ond Stt:ingrimssoll, B., 1982, Production charactt:rlstics 01 wt:lls tapping two phast:
resurvoirs at Krafla CIld Namafjall: Proceedings of
the slXfh workshup on gdoth"rmol re~ervoir engineering, StanfOfd Uniwnily.
Stone, C. and Fan, P., 1978, Hydrothtmnol alterolion of basalt from Hawaii Gt:olhermal Project
Well-A, Kilauea, Hawaiil Gt:alogy, 0, p. 401-404.
Thornos, D., 1982, A sumrnory of Iht: chemical
charactt:rislics of the HGP-A wt:lI, Puno, Hawaiil
Procueding of Ihe t:ighth workshop on gcotht!rmal
rd.drvolr <;ngineerin9,
Stanford Univt:r~ity, p
209-313.
Thomas, D, 1983, Stolus summary of the HGP-A
generollon focllitYI
19U1: GHC Trans., 7, p.
75-80.
Thomas, D., 1704, Geotherfllol re,ources mSl!Ssment In Hawaii, final report: U.S.D.O.E. Heport
DOE/SF /lOBI-I-
n.
Waibel, A, 1903, A review of Ihe hydrolht:rrnol
mineralolO<JY of Hawaii G.wlherrnal Project
Well-A, Kilauea, Hawaii: GRC Trans, 7, p.
205-210.
LEGEND
1
•
N
'
L}]
,,
PHODUCTlv[ wELLS
"'*-
OR'!' W(L LS
~
19!1!:l LAVA f-LOW!)
+
I/OLCANIC PEAK
\
f<IFT ZONE
"
... -:.... ,/
~
o
Surface Expression of East Rift Zone
I
Fjgur~
I:
Locat ion mop
for
---------
2 miles
the Puna geothermol
f itdd.
fl.'AlPENAf/)HE (OF}
,~
~
~
<roo
: 401l'
>roo
6000
•
'''''''
I
11000
S.l.lUIU,tlOH
S~I~'''_
1''''1)110'''' I"" .. WGl,d hu4
.., IIIIU~'14 ,,""',,
th,.,·,,· till
TO •• oo~'
...,
4~,1
.,111
.4 t.r
"'.'h~I\I'
fEIoIft.IUfl.Mf
~5·1""~'IQfI ... c~.U;1I
.:. JU.f't.MATUfi.f
'00
1000
Ijoo
,(000
PR£$$//RC (pII,)
~"oo
,Jooo
'"00
Vigure 2: Temperature/pressure
profile for KS-l,KS-2, and IIGP-A.
Production casing and liner profile
for each well is shown. All data was
measured prior to any significant flow
testing. In December, 1979, the upper
portion of the HGP-A 7 inch slotted
liner was replaced with solid 7 inch
casing to 2900 feet .
,(}() t!
~
,..,
'0
w
~
l((.(lIO
[
"
1m
.........
.. .:::1
t.
60
70
dO
!11tI£/Mllls)
""
~
L[G(NO
I
•
~ '"
... ~ .. u h ..... · ......
~'H
.. u~. II."
I.l
• • ~u., P, . . . . . .
•
t,~
... u ••
"
II ...
nCJ -~J-Oo -J;-Q~-~~W
100·--*/J;----i-;.00
,,}O
1t(;'LlHt-.40
Figure 3: Kapoho State
separator flow test data for the
period 50 to 100 hours. Note
periods of cyclic flow.
PIU!>~.IHt.- Ip~''1J
Figure 4: Comparative flow tCHt
result plot for HGP-A (open symbol)
and' KS-l (closed symbol).
~-------
EXPRESSION
SURFACE
EAST RIFT ZONE
OF~_____
-1
ZONE OF VIGOROUS
GROUNOWATER FLOWS
IMPERMEABLE SEALH'I'DftOTII(f(UAL AlT£RATIOh
stALING fl.!.JLT --~~
HI"H Otl.SITY Of D1Kt:S wlTlUtf
kif T ZOlll INCREASING Iii
FNfQUfNCY WITH llEPTH
A fEw
.).Rl
lu .... OEtoSlfY
1)1.
0f
ONLY
SHOWII fOR CLAklTY
lJiK(S
"-"]'D'V'~;I~;; --;,rn:rr --lj1f
l.tol"GIN~
\
SECONDARY MAGMA CHAMBER(?)
)
In nlTm mn mnrrmfll
DIKE
~
f h
filjure S: Conceptual model 0
' the r.i, 'IIe zone
sea I' 'I.H'I~ougeg~:h~~II~e
1IIlJletllleUu
°
',e
•
o -
:.
}'
~
~
Pun
_ _ _ _ _- -
~otheflrtOl
SCALES tHGHLY
~y~ltJn.
Section i) nUrH'JJl 10 the Ire~ld
., I conf ined t'xcepl in oreos 01 ctoH-fuult ,nq.
;: 1~~'OtJY
0 9,
~~:t~~
COMPLEX
and hydrolhelfl.:ll 01 'erol ion.
EXAG6EfiATEO
ERRATA
Preliminary results of dri lling and testing
System, Hawaii: J. L. lovenitti and W. L. D'Olier
Paper Page
Paragraph
Line
in
the
Puna
Correction*
7
4
... belt, I - 2 miles wide, ...
2
2
4
... some three miles ...
2
3
2
... Thomas 119841 ...
2
4
9
... for each well ...
2
7
2
... tephra.l....
3
3
5
... zone a relatively ...
3
7
6
... tools, and a ...
3
8
4
... a rock-filled ...
3
10
2
... conducted with a ...
12
... concentration is about ...
4
Geothermal
4
5
9
... r eservoi r ...
4
7
6
... in this state ...
4
7
13
... to be
4
... well are shown ... data were measured ...
6
Figure 2
*Correction is underlined.
present~