Borehole packer tests at multiple depths resolve distinct ...afisher/CVpubs/pubs/BeckerFisher2008_JGR.pdfBorehole packer tests at multiple depths resolve distinct hydrologic intervals

Borehole packer tests at multiple depths resolve distinct hydrologic

intervals in 3.5-Ma upper oceanic crust on the eastern flank of

Juan de Fuca Ridge

K. Becker1 and A. T. Fisher2

Received 18 October 2007; revised 17 April 2008; accepted 6 May 2008; published 24 July 2008.

[1] Single-hole hydrogeologic experiments were conducted with a drillstring packer inIntegrated Ocean Drilling Program Hole 1301B, which penetrates 318 m into 3.5-Ma upperoceanic basement on the eastern flank of the Juan de Fuca Ridge. Seven constant-rateinjection tests were conducted in open hole, with the packer inflated at three depths in thedeepest 166 m of the basement section. The primary pressure data recorded in the testintervals required correction for local variations in seafloor pressure (mainly tides), as wellas for changes in the ‘‘baseline’’ pressures during individual tests, probably owing todownhole flow of ocean bottom water induced by drilling. The corrected pressure recordswere fit to standard, idealized models of borehole response during pumping, leading toestimated basement permeabilities within the upper 318 m of basaltic crust near Hole 1301Bof the order of 1 to 3� 10�12 m2. The upper 207m of basement around Hole 1301B appearsto be more transmissive than the 111 m below this depth. Permeability within the upperinterval may be as high as 5� 10�12 m2, and permeability within a 30-m-thick zone betweenpacker setting depths may be as great as 2 � 10�11 m2. Comparison of the results to packerand thermal data from uppermost basement in nearby Holes 1026B and 1027C suggeststhat the most transmissive part of the upper crust in this area may be located not immediatelyadjacent to the sediment-basement transition but deeper in the section.

Citation: Becker, K., and A. T. Fisher (2008), Borehole packer tests at multiple depths resolve distinct hydrologic intervals in 3.5-Ma

upper oceanic crust on the eastern flank of Juan de Fuca Ridge, J. Geophys. Res., 113, B07105, doi:10.1029/2007JB005446.

1. Introduction

[2] Considerable effort has been dedicated in recent yearsto resolving the dynamics and impacts of fluid flow onthe evolution of the oceanic lithosphere [e.g., Davis andElderfield, 2004;Fisher, 2005;Ge et al., 2002]. An importantpart of this effort has included scientific ocean drilling tocollect rock, fluid, and microbiological samples, log and testin situ properties within active seafloor hydrothermal systems,and deploy long-term monitoring instrumentation. Some ofthis work has occurred at and very close to seafloor spreadingcenters, but the majority of scientific drilling to investigateoceanic crustal hydrogeology has focused on ridge flanks insedimented young crust (1–7 Ma) away the immediateinfluence of magmatic activity. This included IntegratedOcean Drilling Program (IODP) Expedition 301, part of amultidisciplinary program designed to: evaluate the formation-scale hydrogeology of young oceanic crust; determine howfluid pathways are distributed within an active hydrothermalsystem; and elucidate relations between fluid circulation,alteration, microbiology, and seismic properties. The complete

IODP experimental program will include a second expeditiontentatively planned to sail in 2010 (subject to IODP sched-uling), followed by long-term monitoring and cross-hole testsfacilitated with submersible and remotely operated vehicleexpeditions extending an additional 4–5 years.[3] Expedition 301 worked mainly at Site 1301 (Figure 1),

at the eastern end of a 80-km-long drilling and observatorytransect established on an off-axis flowline during OceanDrilling Program (ODP) Leg 168. The primary goal ofExpedition 301 was to drill and core two new basementboreholes, collect rock and fluid samples, log the basementsection, test crustal hydrogeological properties, and emplacelong-term observatories to be used for future experiments[Fisher et al., 2005a]. In this study we present results ofhydrogeological tests conducted using a drillstring packer toassess near-borehole permeability in the upper oceanic crustin Hole 1301B (Figure 1). In a related study, an inadvertentcross-hole experiment between Sites 1301 and 1027 isinterpreted to assess crustal hydrogeologic properties at alarger scale, and to draw inferences regarding anisotropy inthese properties [Fisher et al., 2008].

2. Geological Setting and Previous Work

2.1. Eastern Flank of the Juan de Fuca Ridge

[4] The Endeavour segment of the Juan de Fuca Ridge(JFR) spreads at a medium rate, generating ocean crust with

JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 113, B07105, doi:10.1029/2007JB005446, 2008ClickHere

for

FullArticle

1Division of Marine Geology and Geophysics, Rosenstiel School ofMarine and Atmospheric Sciences, University ofMiami,Miami, Florida, USA.

2Earth and Planetary Sciences Department, Institute for Geophysics andPlanetary Physics, University of California, Santa Cruz, California, USA.

Copyright 2008 by the American Geophysical Union.0148-0227/08/2007JB005446$09.00

B07105 1 of 12

http://dx.doi.org/10.1029/2007JB005446

features common to many ridge flanks: extrusive igneousbasement overlain by sediments, abyssal hill topography,high-angle faulting, and basement outcrops [e.g., Davis andCurrie, 1993; Davis and Lister, 1977; Kappel and Ryan,1986; Karsten et al., 1986]. The eastern flank of the JFR inthis area is covered by hemipelagic mud and turbidites thatflowed from the nearby continental margin into CascadiaBasin during the Pleistocene, resulting in rapid burial ofbasement rocks at an unusually young age [Davis et al.,1992; Underwood et al., 2005].[5] In 1996, ODP Leg 168 drilled a transect of eight sites

east of the JFR, in 0.8–3.6 Ma crust extending 20 to 100 kmfrom the active ridge [Davis et al., 1997a]. Borehole packerexperiments were conducted in three holes (Holes 1024C,1026B, 1027C) along the transect [Becker and Fisher,2000], and these and one additional hole (Hole 1025C)were sealed with long-term, subseafloor observatories(‘‘CORKs’’) [Davis and Becker, 2002]. The focus of Leg168 experiments was on lateral variations in thermal andpressure conditions and hydrogeologic properties within theupper tens of meters of oceanic crust. IODP Expedition 301returned to this area, drilled deeper into basement, con-ducted hydrogeologic experiments, and deployed additionalCORK systems [Fisher et al., 2005a]. Expedition 301focused on properties and processes extending deeper intobasement than did Leg 168, and helped to establish part of athree-dimensional observatory network to be used for long-term, cross-hole experiments.[6] Thermal and pore fluid chemical data from the

western end of the ODP Leg 168 drilling transect indicatethat basement rocks in this area are chilled by rapidlyflowing hydrothermal fluids that recharge and dischargethrough exposed basement to the west of Sites 1024 and1025, and through outcrops to the north and south of thetransect, where sediment cover is incomplete [Davis et al.,1992; Wheat and Mottl, 1994; Hutnak et al., 2006]. Sites1026, 1027, and 1031 are located farther to the east, where

the sediment cover is nearly continuous and hydrologicconditions in basement are much more isolated. Neverthe-less, vigorous fluid convection continues there within per-meable basement, largely homogenizing temperatures at thesediment-basement contact, and hydrothermal dischargeseeps through isolated outcrops, sending warm, highlyaltered formation fluid into the overlying ocean [Davisand Becker, 2002; Mottl et al., 1998; Thomson et al.,1995; Wheat et al., 2004].

2.2. Results From Previous Packer and ThermalFlowmeter Experiments

[7] Packer experiments were completed during ODP Leg168 in Holes 1024C, 1026B, and 1027C in 1.2, 3.5, and3.6 Ma seafloor, respectively [Becker and Fisher, 2000].These holes extended as much as 57 m below the ‘‘sedi-ment-basement’’ interface, but because of limited drillinginto basement (Hole 1024C), poor hole conditions (Hole1026B), and the presence of a sill and an underlyingsediment layer (Hole 1027C), these tests were restricted tothe upper 20–25 m of the crustal basement aquifer. Likesimilar tests conducted elsewhere in the upper crust, the Leg168 packer experiments included a mix of slug tests andconstant-rate injection tests, followed by shut-in and pres-sure recovery (experimental methods are discussed below),with no individual test lasting longer than 30 min. The Leg168 tests were conducted mainly by setting the packer incasing above basement, so that properties and conditionswithin the short open basement intervals were evaluated; anadditional set of tests was completed in Hole 1027C bysetting the packer in open hole, to isolate the upper part ofthe basement aquifer below. Results of these experiments[Becker and Fisher, 2000] indicated a change in thepermeability of the uppermost tens of meters of the base-ment aquifer as a function of crustal age: permeability wasgreatest around Hole 1024C (3 � 10�11 m2, 1.2 Ma), loweraround Hole 1026B (1 to 2 � 10�13 m2, 3.5 Ma), and lowerstill around Hole 1027C (4 � 10�14 m2, 3.6 Ma).

Figure 1. Site maps. (a) Regional map showing major tectonic features and locations of holes on theeastern Juan de Fuca flank with observatories and permeability experiments. (b) Detailed bathymetricmap showing locations of Hole 1301B, where single-hole packer experiments were conducted, andnearby Holes 1026B and 1027C. Contour labels indicate seafloor depth in meters.

B07105 BECKER AND FISHER: HYDROLOGIC INTERVALS IN OCEANIC CRUST

2 of 12

B07105

[8] Additional estimates of permeability within upper-most basement around these three holes and Hole 1025Cwere provided by thermal flowmeter experiments [Fisher etal., 1997; Becker and Davis, 2003]. During these experi-ments, fluid flow rates into or out of basement wereestimated on the basis of deviations from conductive con-ditions within a cased borehole that penetrated throughoverlying sediments into permeable basement. Many scien-tific boreholes drilled into the ocean crust and left unsealedhave drawn bottom seawater into the formation, becausethese operations result in the imposition of a cold (higherdensity) column of drilling seawater in a borehole sur-

rounded by warmer (lower-density) formation fluid [e.g.,Becker et al., 1983; Morin et al., 1992]. If there is a naturalformation overpressure of sufficient magnitude, the inducedflow down an open basement hole may spontaneouslyreverse, resulting in sustained upflow from the formationinto the ocean. This actually occurred in Hole 1026B withina few days of Leg 168 drilling [Fisher et al., 1997] and inHole 896A in the eastern equatorial Pacific [Becker et al.,2004]. The thermally determined flow rate up or down an openbasement borehole, in combination with the measured orestimated pressure differential between the formation and theopen hole, was used to calculate the permeability of basement

Figure 2. Summary of basement and borehole characteristics in Hole 1301B. (a) Basement recovery,borehole size, formation bulk density, and depths where drillstring packer was inflated duringhydrogeologic testing (data from Shipboard Scientific Party [2005a]). Basement cores were collectedfrom �355 to 583 mbsf (�90 to 318 m subbasement (msb)), with recovery indicated by black intervalsnext to depth column. Borehole diameter was measured with mechanical caliper during geophysicallogging. Note enlarged intervals above 470 mbsf (220 msb). Bulk density log (line) and analyses of coresamples (dots) show evidence for considerable porosity in uppermost basement, and a layered basementstructure below 470 mbsf (220 msb), with alternating lower- and higher-porosity intervals. Bands at 417,442, and 472 mbsf indicate packer setting depths. (b) Borehole configuration, drawn with consistentvertical scale. Hole 1301B is cased through 265 m of sediments and the upper �100 m of basement,but the innermost casing was not sealed and has a gap near the sediment-basement interface, as discussedelsewhere [Fisher et al., 2008]. Packer experiments were run with the packer set in open hole, at depthsindicated.


3 of 12

B07105

surrounding the borehole [Becker et al., 1983]. This methodyielded estimated basement permeability values of 10�12 to10�10 m2 within the uppermost crust, 0.5–1.0 orders ofmagnitude higher than determined from packer tests, but withvalues varying inversely with crustal age, like the packerexperiments [Becker and Davis, 2003; Fisher et al., 1997].

3. Experimental Operations and Data Processingand Analysis

3.1. Experimental Methods

[9] Packer experiments in IODP Hole 1301B were con-ducted using the same system and methods that had beenused on Leg 168: an inflatable drillstring packer, downholepressure gauges, and test procedures that are described indetail in numerous earlier studies [e.g., Becker, 1996;Fisher, 1998; Becker and Fisher, 2000]. The packer canbe inflated within the cased part of a borehole, or in theopen formation provided the hole diameter is sufficientlysmall and regular. The go-devil that enables inflation alsocarries two autonomous pressure-temperature gauges thatrecord conditions inside the sealed zone beneath the packerduring the experiments. Immediately prior to Expedition301, the downhole gauges (‘‘ERPG-300’’ gauges made byGeophysical Research Corporation, Tulsa OK) had beenserviced and recalibrated by the manufacturer. The gaugeswere set to record digital pressure and temperature values at10.8-s intervals throughout the experiments.[10] Two kinds of pumping tests have been completed

using a drillstring packer during scientific drilling in theoceanic crust: slug tests and injection tests. Both kinds oftests involve pumping fluids into a sealed borehole after thepacker is set. During a slug test, a small quantity of fluid ispumped rapidly, causing an abrupt rise in pressure, and thepressure decay after pumping is monitored [Bredehoeft andPapadopulos, 1980]. These kinds of tests work best in lowto moderate transmissive formations. Constant-rate injectiontests work better in more transmissive formations andinvolve pumping fluid at a known rate for an extended timeand monitoring fluid pressure rise during this time [Theis,1935; Cooper and Jacob, 1946; Horner, 1951]. One im-portant limitation in packer experiments run during scien-tific ocean drilling is that formation fluid pressure ismonitored in the pumping well. This testing configurationlimits determination of formation storage properties, andresults in estimates of formation transmissivity in the regionimmediately surrounding the well [Fisher, 1998]. A secondlimitation is that the time demands on the expensive drill-ship have generally limited individual test periods to 30 min

or less; our injection tests were conducted for twice as long,but this is still short as compared with typical test periods ofdays on land. Additional limitations of this testing config-uration are discussed later.[11] Packer experiments were actually attempted during

IODP Expedition 301 in both Holes 1301A and 1301B[Shipboard Scientific Party, 2005b]. Poor hole conditionsprecluded attempting to inflate the packer element in openbasement in Hole 1301A, so the packer was inflated incasing. However, those poor hole conditions had alsoprevented establishment of a complete cement seal aroundthe base of the casing in which the packer was set. The lackof a seafloor casing seal (discussed by Fisher et al. [2008])allowed fluid pumped into Hole 1301A to flow up theannulus between nested casing strings and into the ocean, sono valid estimates of formation hydrological propertiescould be derived. The casing in Hole 1301B was alsoincompletely sealed, but good hole conditions allowed thepacker to be inflated in open hole below the casing. Onlypacker test operations and observations from the open holesection of Hole 1301B are presented in this study.

3.2. Selection of Packer Inflation Seats in Hole 1301b

[12] Hole 1301B was drilled through 265 m of sedimentand 318 m into basement (Figure 2). Extremely fastbasement drilling and oversize hole conditions [Bartetzkoand Fisher, 2008; Shipboard Scientific Party, 2005a,2005b] suggest that rubbly, unstable hole conditions extendthrough the uppermost �200 m of basement. Packer settingdepths in Hole 1301B (Table 1) were selected wherebasement rocks were relatively massive and the hole diam-eter was sufficiently small for successful element inflation,at 472, 442, and 417 mbsf (207, 177, and 152 m subbase-ment, msb, respectively; see Figure 2). The shallowest anddeepest packer inflation depths also corresponded to infla-tion depths for casing packers used later during installationof the Hole 1301B CORK observatory [Fisher et al.,2005b]. Because of the poor hole conditions, the inflation/test sequence did not include setting the packer within theuppermost�150 m of oceanic crust that has been found to bevery permeable in similar experiments elsewhere [e.g.,Anderson and Zoback, 1982; Becker and Fisher, 2000;Larson et al., 1993]. However, as discussed below, basementproperties throughout the uppermost 318 m of basementmay have been tested during experiments in Hole 1301B.[13] The deepest packer setting depth was located where

there is a distinct change in the character of basementgeophysical logs (Figure 2a). Above 470 mbsf (205 msb)the borehole is highly irregular in diameter and contains

Table 1. Packer Test Specifications in Hole 1301B

Test ID Top of Interval,a mbsf Top of Interval,a msb Open Hole Interval,b m Pumping Rate, L/s

1 472 207 111 3.72 472 207 111 6.73 472 207 111 10.04 442 177 141 5.05 442 177 141 10.06 417 152 166 5.07 417 152 166 10.0aTop of interval is the depth at which packer element was set when inflated; mbsf, meters below seafloor; msb, meters subbasement.bOpen hole interval is distance between packer set depth and total depth of hole. Total depth of Hole 1301B is 583 mbsf or 318 msb.


4 of 12

B07105

several intervals 10–50m thick having low bulk density (andelectrical resistivity and P-wave velocity) [Bartetzko andFisher, 2008]. Below 470 mbsf, the borehole diameter ismuchmore consistent with the drill-bit diameter, and zones oflow bulk density are thinner and more widely spaced. Thischange in geophysical properties corresponds to the bound-ary between massive and pillow basalt units 6 and 7,respectively [Shipboard Scientific Party, 2005b], and corre-lates to similar lithostratigraphy seen in other young crustalsites [e.g., Bartetzko et al., 2001; Becker, 1996].

3.3. Pressure Observations, Corrections,and Processing

[14] Figure 3 shows the downhole pressure and temper-ature records from the entire packer test sequence in Hole1301B, with operational details explained in the caption.The pressure data required several processing steps prior toquantitative interpretation. First, local variations in oceanpressure at the seafloor had to be removed. These variationsresult mainly from ocean tides, but also comprise atmo-spheric pressure changes having a range of frequencies. A

Figure 3. (a) Complete downhole gauge pressure (solid line) and temperature (dashed line) recordduring packer experiments in Hole 1301B. Individual injection test responses are shown in greater detailin Figure 4. At the first setting depth of 472 mbsf, cold hydrostatic pressure in the hole was monitored for30 min (H) before actual packer inflation (I). At each inflation depth, packer inflation produces anuncontrolled pulse in the pressure record as a sliding sleeve is shifted simultaneously to lock the packerinflated and to open a passageway from the pressurized drillstring to the formation isolated below thepacker element. A controlled slug test (S) was attempted only at the first inflation depth. This wasfollowed by three injection tests there at the rates shown and two injection tests at each of the next twoinflation depths. For each injection test, fluid was pumped for 60 min, followed by 60 min of pressurerecovery. At each depth, after packer deflation (D), there was additional monitoring of the cold boreholehydrostatic pressure baseline (H) before the packer was moved up the hole to the next depth. Temperaturerecord shows suppression of borehole temperature by previous pumping into the formation duringdrilling, coring, and other basement operations, as well as pumping during packer experiments. Note thatthe borehole fluid temperature rises during the recovery period following the first five injection tests(and particularly during the second set of tests at 442 mbsf), but there is little thermal recovery followingthe last two tests (at 417 mbsf). (b) Pressure variation (primarily ocean tides) recorded at the seafloor atHole 1027C, 2.4 km to the ENE of Hole 1301B, during the time of packer experiments. Because Hole1301B was open to these seafloor tidal variations until packer inflation, the pressure records werecorrected for these local variations in seafloor pressure prior to further processing for formationhydrogeological properties.


5 of 12

B07105

record of seafloor pressure during the time of packer testingwas recovered after the drilling expedition from nearbyHole 1027C [Fisher et al., 2008]. The amplitude of seafloorpressure variations was 5–10 kPa, approaching the samemagnitude as the pressure changes resulting from injectiontesting (Figures 3 and 4). As Hole 1301B was open toseafloor pressures for days immediately prior to the packertests, seafloor pressure variations from Site 1027 wereremoved from individual packer test records prior to pro-cessing.[15] A greater challenge to interpretation of the packer

test records resulted from transient effects of pressureperturbations related to drilling and other operations.Throughout the drilling operations, cold seawater waspumped down the drillstring at high rates (10–100 L/s).Seafloor and borehole thermal data from Site 1301, andnearby Sites 1026 and 1027, indicate that in situ temper-atures at the sediment-basement interface in this region areremarkably uniform at 60–65�C, despite considerable base-ment relief, as a result of vigorous local convection withinthe upper oceanic crust [Davis and Becker, 2002; Davis etal., 1992; Hutnak et al., 2006; Shipboard Scientific Party,2005b]. The drilling fluid (mostly surface seawater) gener-ally is cooled to nearly bottom water temperature (1.8�C inthe case of Site 1301), and it is much denser than ambientformation fluid. While it is not known how much of this

fluid penetrates into basement and how much returns theannulus between the drillstring and casing, the imposition ofa cold column of borehole fluids during basement opera-tions in Holes 1301A and 1301B likely allowed andinduced cold seawater to flow into the formation for severalweeks prior to the packer testing, causing a significantpressure and temperature perturbation that probably extend-ed deeper into the formation than tested by the packerexperiments. The radial variation of temperatures aroundthe borehole is not known, but temperatures were assumedto be close to the value of ocean bottom water owing to thelarge volumes of cold fluid drawn into the formation duringthe weeks of downhole flow and also pumped into theformation during the packer tests.[16] Careful examination of downhole pressure and tem-

perature records during packer testing illustrates the influ-ence of these distinct processes (Figures 3 and 4). Prior tosetting the packer at the first test depth (472 mbsf), the coldhydrostatic pressure in the borehole was 31.87 MPa (labeledH on Figure 3a). Borehole fluid temperatures were close to2�C during this time, because cold bottom seawater hadflowed so freely into the open hole. The downhole pressuredropped abruptly by 40 kPa when the packer was inflated,illustrating the pressure perturbation caused by boreholeoperations and the flow of cold seawater into the formation,but the pressure baseline rose during the subsequent hour

Figure 4. Expanded pressure records during injection tests in Hole 1301B. (a–c) Tests 1–3 at 472 mbsf.(d and e) Tests 4 and 5 at 442 mbsf. (f and g) Tests 6 and 7 at 417 mbsf. Each plot has been corrected forseafloor pressure variations (Figure 3b), has the same relative horizontal scale, and shows raw pressuredata (small dots) and pressure records filtered with a 2.4-min boxcar (solid line) during a 2-hour timeinterval. Filtered records were used for processing, as shown in Figure 5. Dashed lines in each plot showapparent trend of the baseline pressure during individual experiments as a result of pumping of cold waterinto a warm basement and subsequent flow and/or recovery, as discussed in the text.


6 of 12

B07105

because sealing of the borehole gave the formation anopportunity to move toward reestablishment of the naturaloverpressure. The borehole fluid temperature rose slightlyduring this time, but dropped during each subsequentpumping event (Figure 3a). Pressure rose during eachinjection test, but the ‘‘baseline’’ pressure after each testwas often lower than that observed at the start of each test(Figure 4), mainly because of the injection of cold, denseseawater into the formation. The difference between initialand final basement pressures was most consistent during thefirst three packer tests (472 mbsf; see Figures 4a–4c).[17] The greatest pressure changes before and after indi-

vidual tests were observed when the packer was set at442 mbsf (tests 4 and 5; see Figures 3, 4d, and 4e). Thesetests were preceded and followed by steep rises in bothborehole fluid pressure and temperature, suggesting that atthis setting depth the packer element was isolating aparticularly transmissive, producing zone that was morecapable of overcoming the cold thermal perturbation thanthe deeper section. Curiously, the baseline pressure hadessentially stabilized by the time the packer was set at417 mbsf (tests 6 and 7; see Figures 4f and 4g).[18] Given the uncertainties in thermal conditions in the

formation surrounding Hole 1301B before and duringpacker experiments, and in the amount and rates of fluidflow into (and potentially out of) the formation at all timesexcept during injection tests themselves, it is not possible tocorrect pressure records with confidence for these multiple,superimposed influences. Instead, we bracket a reasonablerange of pressure corrections during data processing byconsidering two end-member cases. First, we process indi-vidual injection test records by referencing pressures duringeach test to the baseline pressure before and after each test(Pinitial and Pfinal, respectively). Second, we apply a baselinecorrection to individual injection test records by assumingthat the background pressure varied linearly (Plinear) duringeach test. This approach allows the data to be treated in aself-consistent way. Later sealed-borehole experiments willeventually allow hydrogeologic testing to be completedafter the disturbance associated with drilling has dissipated[Fisher et al., 2005b].

3.4. Analytical Models and Limitations

[19] The pressure response in a well resulting frompumping at a constant rate can be interpreted using anequation having the form: DP = f (Q, t, T), where Q = fluidpumping rate, t = time, and T = formation transmissivity.Transmissivity is the product of aquifer thickness andhydraulic conductivity, K, where the latter is related topermeability by k = Km/rg, where m is the viscosity andrg is the specific weight of the fluid. Seawater viscosity isknown to be strongly temperature-dependent, but we usedthe value at 2�C on the basis of the assumption that fluids inthe zone of investigation were kept close to bottom-watervalues by the pumped and induced flow into the formation.Additional parameters that can influence the pressure re-sponse in a pumping well include formation heterogeneityand anisotropy, well-skin effects, leakage through confininglayers, and partial penetration of the well into the aquifer.We have insufficient information to resolve the potentialinfluence of the first of these parameters, but variations inapparent formation transmissivity and permeability with

depth, shown later, suggest that there is heterogeneouslayering of the crust around Hole 1301B. Particularly withinhighly transmissive formations, partial penetration generallyhas a modest influence on transmissivity values inferredfrom single-hole tests [Halford et al., 2006]. The issue ofaquifer anisotropy is addressed in a related study [Fisheret al., 2008].[20] Individual constant-rate injection tests were inter-

preted by referencing measured pressures to a range ofbaseline conditions (Pinitial, Pfinal, Plinear), as discussedearlier. Pressure-time data were compared to analyticalsolutions of a conservation of mass equation, in which fluidflows radially away from a pumping well [Theis, 1935].This model was derived on the basis of numerous assump-tions and idealizations, including: a flat, tabular aquifer ofinfinite extent and constant thickness; isotropic and homo-geneous formation properties; equilibrium conditions priorto the start of injection; a constant rate of pumping; andlaminar flow conditions at the borehole wall and within theaquifer (i.e., Darcy’s law applies). We recognize that someof these conditions may not have been met, but the resultsof these calculations are useful nevertheless, particularlywhen compared to results from similar tests conductedelsewhere in the crust. We also explored fitting packer testsresults to other models, including an approximation to thefull transient solution [Cooper and Jacob, 1946], a steadystate model based on the pressure offset from baseline[Snow, 1968], and a transient model including a leakyconfining layer [Hantush, 1960], but the resulting estimatesfor formation permeability were not significantly differentfrom those based on the Theis [1935] model. Fits to theanalytical solution reported herein were made using anonlinear, least squares approach, with the first 200–500 sof data omitted from each test (because the earliest time datagenerally yielded a poor fit to the analytical model).[21] Although we tested multiple overlapping intervals of

the crust, we assume that the properties reported applyequally to the entire tested interval. In principle, estimatesof formation transmissivity based on application of theTheis [1935] model do not depend on a priori knowledgeof aquifer thickness (except that the aquifer is presumedto be fully penetrated by the well). However, the mostpermeable part of the formation surrounding Hole 1301Bis likely much thinner in total than the complete openhole interval, comprising zones of breccia, rubble, pillows,and flows within which fractures are most concentrated.Given an inferred transmissivity value for a testedinterval, the bulk permeability scales linearly with theaggregate thickness of the basement zone(s) through whichmost of the flow occurs. Permeabilities reported herein areaverage or equivalent porous-medium values, denoted asbulk permeabilities.

4. Results and Discussion

4.1. Interpretation of Constant-Flow Tests

[22] Results of individual constant-rate injection tests aresummarized in Table 2, and fits of the analytical model toseveral tests are shown in Figure 5. In general, we foundlittle difference between transmissivities inferred on thebasis of selecting either the initial or final baseline pressureas a reference, and data from most of the tests resulted in a


7 of 12

B07105

good fit to the idealized analytical model. For consistency,our preferred interpretations are those based on a constantoffset from the final baseline pressure determined immedi-ately after testing (Pfinal in Table 2). Application of a linearbaseline correction usually resulted in calculation of aformation transmissivity (and bulk permeability) that wassomewhat lower than that determined without a linearcorrection, because the linear correction resulted in a greaterpressure rise during the test.[23] Data from test 1 (472 mbsf) did not fit as well to the

Theis [1935] model as did other tests because test 1 pressuresleveled off after 200–300 s, then dropped by several kPaafter 1000 s and rose again after 2000 s (Figure 5a). Test 1data could be fit somewhat better to a leaky aquifer model[Hantush, 1960], but this interpretation is inconsistent withthat from subsequent tests at the same depth that show noevidence for leakage. Because of this inconsistency, resultsfrom test 1 were not included in the mean permeabilityvalue calculated from tests at 472 mbsf. However, the bulkpermeability calculated using a leaky aquifer model fromtest 1 is very close to that inferred from tests 2 and 3 usingthe conventional Theis [1935] model. All remaining tests

were reasonably well fit with the Theis [1935] model. Tests4 and 7, at 442 and 417 mbsf, were not processed using alinear baseline correction because data from these testsindicated stable baseline conditions (Figures 4d, 4f, 5d,and 5f).[24] As shown in Table 3, geometric mean values from

packer tests at 472 mbsf are T = 0.0034 m2/s and k = 1.7 �10�12 m2 (standard deviation = 2.3 � 10�13 m2), whereasmean values for tests at 442 and 417 mbsf are about twice asgreat, T = 0.0064 m2/s and k = 3.2 � 10�12 m2 (sd = 5.1 �10�13 m2). Transmissivity and bulk permeability valuesinferred from the last four tests (at depths of 442 and417 mbsf) are very similar, differing little as a function ofmeasurement depth. In contrast, transmissivities inferredfrom tests at the initial setting depth (472 mbsf) areconsistently lower than those from later (shallower) tests.This observation suggests that there is a significant differ-ence in formation hydrogeologic properties above andbelow 472 mbsf (207 msb), corresponding to the observedchange in the character of geophysical logs (Figure 2). If weassume that the properties measured during the final fourtests (setting depths of 442 and 417 mbsf) apply mainly to

Figure 5. Fits of filtered pressure data to analytical Theis [1935] model of aquifer response to pumping.(a–c) Tests 1–3 at 472 mbsf. (d and e) Tests 4 and 5 at 442 mbsf. (f and g) Tests 6 and 7 at 417 mbsf.Plots show possible model fits on the basis of using the initial or final baseline pressure as a reference(circles with solid line and squares with dashed line), and from making a linear correction for theobserved baseline shift during each test (crosses, dotted line). Observational data have been filtered(Figure 4), and only every fourth filtered value is shown for clarity. In each plot the preferred analyticalmodel is shown with a thicker line corresponding to the transmissivity and bulk permeability valuesshown in boldface in Table 2. Unlike all other packer test records, the record from test 1 levels off after200–300 s; the pressure drops after 1000 s. This pattern prevents achieving a good fit with the Theis[1935] model. For this reason, results from this test are not included in calculated mean values based ontests at 472 mbsf. See discussion of alternative models in text. The baseline pressure was consistentbefore and after tests 4 and 7, so no linear baseline correction was made for these tests.


8 of 12

B07105

the uppermost 207 m of basement, bulk permeability withinthis interval is closer to k = 5 � 10�12 m2. If the differencein transmissivity determined during shallower and deepertests is applied to the 30-m interval between 442 and472 mbsf (177–207 msb), the bulk permeability of thiszone is k = 2 � 10�11 m2. Attributing most of thetransmissivity seen during tests 4–5 to this zone is consis-tent with the observation that borehole pressure and fluidtemperatures rose rapidly before and after these tests, butnot when the packer was set deeper or shallower.

4.2. Comparison With Global Data

[25] Results from packer testing in Hole 1301B arecompared to results of packer and thermal (flowmeter)experiments globally in the oceanic crust (Figure 6). Ingeneral the global data set shows a decrease in bulkpermeability with depth into basement. Values higher than10�13 m2 are generally restricted to the upper 300 m ofbasement, although the global data set from deeper crustallevels is sparse. Tests of smaller depth intervals in upperbasement tend to yield higher permeabilities, suggestingthat some of these narrow intervals may span zones with thegreatest permeability. This interpretation appears to beconsistent with field studies and data compilations from

aquifers and crustal systems on land [e.g., Brace, 1984;Paillet et al., 1987; Black, 1990; Clauser, 1992; Ingebritsenand Manning, 1999].[26] Prior to the completion of packer tests in Hole

1301B, the bulk permeability from Hole 839B seemed tobe anomalously high (Figure 6). This hole penetrates 2.2 Mabackarc seafloor of the Lau Basin [Bruns and Lavoie,1994]. As with tests in Hole 1301B, the depth extent anddistribution of the most permeable part of the crust sur-rounding Hole 839B are unknown. The bulk permeabilityinferred from packer experiments in Hole 1301B (whendistributed evenly to the depth of the hole) is an order ofmagnitude greater than that inferred from packer experi-ments within the upper 25–50 m of Hole 1026B, 1 km tothe north, but is about the same as that inferred from thermallogs in that hole [Becker and Fisher, 2000; Becker andDavis, 2003] (Figures 1b and 6). Possible explanations forthis observation are that: (1) basement is more permeable atSite 1301 than at Site 1026, (2) basement is more permeableat 152–318 msb than it is close to the sediment-basementinterface, or (3) the liner and rubble at the base of Hole1026B limited communication with the formation duringshort-duration packer tests, causing these to underestimateactual formation properties. Bulk permeability in Hole1301B is also greater than that determined with either apacker or thermal log from Hole 1027C, 2.4 km to the eastwhere basement rocks are covered by nearly twice as muchsediment and the crust is 0.1 Ma older.[27] Packer experiments tend to yield bulk permeability

values that are 0.5–1.0 orders of magnitude lower thanthermal (flowmeter) logs in the same holes, and both ofthese experimental methods tend to yield bulk permeabilityestimates 2–4 orders of magnitude lower than those inferredfrom numerical modeling and analysis of responses to tidalpressure variations and tectonic events [e.g., Becker andDavis, 2003; Davis et al., 1997b, 2000, 2001; Fisher, 1998,2005; Spinelli and Fisher, 2004]. This difference in calcu-lated properties is generally attributed to the heterogeneousnature of permeability within the oceanic crust and thecharacteristic scales and assumptions inherent in the differ-ent methods used. Earlier studies also suggested that theremay be a consistent decrease in bulk permeability of theupper crust with age, perhaps resulting from progressivesealing of fluid pathways [Fisher and Becker, 2000; Beckerand Fisher, 2000; Becker and Davis, 2003; Fisher, 2005].[28] New packer test results from Hole 1301B are some-

what inconsistent with the patterns defined by earlier

Table 2. Summary of Results of Individual Packer Tests from

Hole 1301Ba

Test ID Preferenceb T, m2/s K, m/s k Single Test,c m2

1d Pinitial 0.0066 2.1 � 10�5 3.2 � 10�12

Pfinal 0.0057 1.8 � 10�5 2.8 �10�12

Plinear 0.0040 1.3 � 10�5 2.0 �10�12

2 Pinitial 0.0054 1.7 � 10�5 2.6 � 10�12

Pfinal 0.0048 1.5 � 10�5 2.3 � 10�12

Plinear 0.0031 9.9 � 10�6 1.6 � 10�12

3 Pinitial 0.0025 7.9 � 10�6 1.2 � 10�12

Pfinal 0.0024 7.6 � 10�6 1.2 � 10�12

Plinear 0.0034 1.1 � 10�5 1.7 � 10�12

4e Pfinal 0.0065 2.1 � 10�5 3.2 � 10�12

5 Pinitial 0.0098 3.1 � 10�5 4.8 � 10�12

Pfinal 0.0077 2.4 �10�5 3.8 � 10�12

Plinear 0.0047 1.5 � 10�5 2.3 � 10�12

6 Pinitial 0.0070 2.2 � 10�5 3.4 � 10�12

Pfinal 0.0064 2.0 � 10�5 3.2 � 10�12

Plinear 0.0052 1.6 � 10�5 2.6 � 10�12

7e Pfinal 0.0064 2.0 �10�5 3.1 � 10�12

aBoldface indicates preferred interpretations. Abbreviations are asfollows: Preference, how pressure changes during the test were referenced;Pinitial, initial pressure at start of test was subtracted from test pressures;Pfinal, final pressure at end of test was subtracted from test pressures; Plinear,baseline pressure during test defined by a straight line connecting initial andfinal pressures; T, transmissivity; K, hydraulic conductivity, k, permeability.

bExamples of how pressure changes during the test were referenced areshown in Figure 4.

cConversions calculated using seawater viscosity at 2�C, 0.0016 Pa-s.Single test values are based on individual test results using associatedPreference values. No preferred value is given for test 1, results of which donot match well with Theis’s [1935] model used to interpret other tests.

dThe analytical Theis [1935] model provides a relatively poor fit to test 1data using any of the three pressure references listed (Figure 6a). The dataare better fit using a constant-offset baseline (Pinitial or Pfinal) and a leakyaquifer model [Hantush, 1960], which indicates a bulk permeability verysimilar to other tests at this depth. For consistency, results from this test areomitted from means calculated for the packer tests at 472 mbsf.

eThese tests were processed with a constant-offset pressure reference(Pinitial = Pfinal) on the basis of the observed consistency of pretest andposttest baseline pressures (Figures 5a and 5d).

Table 3. Estimated Interval Transmissivities and Permeabilities in

Hole 1301B, Calculated as Geometric Means of Preferred Results

From Individual Tests Shown in Bold in Table 2a

Interval, msb T, m2/s k Interval, m2 k SD,a m2

207–318 0.0034 1.7 � 10�12 8.1 � 10�13

152–318 0.0064 3.2 � 10�12 3.1 � 10�13

177–207b 0.0033 2 � 10�11 N/AaAbbreviations are as follows: T, transmissivities; k, permeabilities; SD,

standard deviation.bProperties calculated on the basis of the difference in transmissivities

calculated between tests run at 472 mbsf (207 msb) and those run at 442and 417 mbsf (177 and 152 msb). These properties are inferred to apply to a30-m interval between 472 and 442 mbsf (207 and 177 msb).


9 of 12

B07105

studies. First, bulk permeabilities estimated from the latestpacker tests are higher than seen at most other sites atsimilar depths into basement. In addition, bulk permeabil-ities in Hole 1301B and are similar to or greater than thoseestimated from thermal logs in nearby Hole 1026B [Fisher

et al., 1997; Becker and Davis, 2003], although the lattertests lasted much longer (days rather than an hour) andshould have sampled a much larger volume of the formationaround the borehole. In addition, the highest permeabilitymeasured in Hole 1301B may be located mainly within a

Figure 6. Comparison between near-borehole permeabilities calculated from packer experiments inHole 1301B and those determined from packer and single-hole thermal (flow) experiments in other holesin the ocean crust; labeled P and T, respectively. Data from other studies and sites are compiled fromFisher [1998], Becker and Fisher [2000], Becker and Davis [2003], Harris and Higgins [2008], andreferences therein. Borehole packer and flow experiments from other sites indicate permeability values of10�14 to 10�10 m2 in the uppermost 200–400 m of basement, with the greatest inferred permeabilities inthe uppermost crust and within thin intervals of several young crustal sections. Hole 1301Bpermeabilities are shown for the depth intervals between the packer element and the base of the hole(longer rectangles), and between packer depths (smaller rectangle). Hole 1301B packer permeabilities are1–2 orders of magnitude higher than seen at comparable depths at most other sites, with the exception ofdata from Hole 839B in the Lau Basin [Bruns and Lavoie, 1994]. Hole 1301B packer data indicate higherpermeabilities than detected with similar tests shallower in the crust in nearby Holes 1026B and 1027C(Figure 1). In addition, Hole 1301B data suggest that permeability may be higher above 207 msb (472mbsf) than it is deeper in the hole, and the highest permeability may be concentrated within a 30-m-thickinterval. Poor hole conditions shallower within the crust prevented testing of these depths, but on thebasis of observations elsewhere and the rubbly and fast-drilling hole conditions shallower in the section[Fisher et al., 2005a], it seems likely that near-borehole permeabilities at shallower depths around Site1301 are as great or greater than those shown.


10 of 12

B07105

�30-m-thick zone extending to �200 m into basement(Figure 6). Thus it seems likely that the crustal rocks testedaround Hole 1301B are more permeable than those tested atshallower depths around Holes 1026B or 1027C, and thatpermeability within the crust surrounding Hole 1301B isheterogeneously distributed.

4.3. Limitations and the Need for Multihole,Multidepth, Multidirectional Testing

[29] Packer tests completed in IODP Hole 1301B are ofrelatively good quality, lasted longer than many earlier tests,and provide new insights concerning the magnitude anddepth distribution of crustal hydrogeologic properties.Nevertheless, these tests are limited in several importantways. First, data from single-hole tests provides informa-tion only on transmissive properties, although storage(compressive) properties are also important for dynamic fluidflow processes such as convection, coupled fluid-heat-solutetransport, and transient responses to tectonic events. Second,even injection tests that are 60 min long, 2–3 times longerthan earlier packer experiments in the oceanic crust, are shortcompared to tests that are run commonly in continentalaquifers. Many injection tests on land last 1–3 days orlonger, and data collected during the first 60 min ofpumping is often neglected or the response is attributed toproperties mainly at or near the borehole wall [Dawson andIstok, 1991]. Third, the superposition of pressure andthermal perturbations associated with constant-rate fluidinjection and the response to and recovery from drillingdisturbances introduces uncertainty in data interpretation.An assessment of several possible data reference andcorrection options suggests that this uncertainty is modest,but it cannot be eliminated with available data. Theselimitations will be addressed when the borehole networkaround Site 1301 is complete, and long-term, cross-holeexperiments can be run using multiple, sealed observationwells [Fisher et al., 2005b], hopefully beginning in Summer2010. These long-term tests will also provide additionalinformation concerning the nature of permeability anisotro-py and the depth distribution of hydrogeologic propertieswithin the upper oceanic crust.

5. Summary and Conclusions

[30] Single-hole hydrogeologic experiments were con-ducted with a drillstring packer in IODP Hole 1301B.Seven-hour-long constant-rate injection tests were con-ducted in open hole with the packer inflated at three depthswithin the upper 152–318 m of basaltic oceanic crust.Pressure records required correction for local variations inseafloor pressure (mainly tides), and additional correctionswere made to account for changes in the ‘‘baseline’’pressure during individual tests. Pressure records were fitto standard, idealized models of borehole response duringpumping, leading to estimated basement permeabilities ofthe upper 318 m of basaltic crust near Hole 1301B of theorder of 1 to 3 � 10�12 m2. The upper 207 m of basementaround Hole 1301B appears to be more transmissive thanthe 111 m below this depth. Permeability within the formerinterval may be as high as 5 � 10�12 m2, and permeabilitywithin a 30-m-thick zone between packer setting depthsmay be as great as 2 � 10�11 m2. Comparison of packer

data from Hole 1301B to packer and thermal data fromHoles 1026B and 1027C suggests that most transmissivepart of the upper crust in this area may not be locatedadjacent to the sediment-basement transition.

[31] Acknowledgments. We thank the technicians, officers, and crewof the J. Resolution for assistance in running packer experiments duringIODP Expedition 301. This research used additional data provided by theIODP and was supported by NSF grant OCE0400471 and Joint Oceano-graphic Institutions, Inc., project T301A8 (K.B.); NSF grants OCE-0326699and OCE–0550713; and JOI Projects T301A7 and T301B7 (A.T.F.).Careful reviews by S. Ingebritsen and an anonymous reviewer greatlyimproved this manuscript.

ReferencesAnderson, R. N., and M. D. Zoback (1982), Permeability, underpressuresand convection in the oceanic crust near the Costa Rica Rift, easternequatorial Pacific, J. Geophys. Res., 87, 2860–2868.

Bartetzko, A., and A. T. Fisher (2008), Physical properties of young(3.5 Ma) oceanic crust from the eastern flank of the Juan de Fuca Ridge:Comparison of wireline and core measurements with global data,J. Geophys. Res., 113, B05105, doi:10.1029/2007JB005268.

Bartetzko, A., P. Pezard, D. Goldberg, Y.-F. Sun, and K. Becker (2001),Volcanic stratigraphy of DSDP/ODP Hole 395A: An interpretation usingwell-logging data, Mar. Geophys. Res., 22, 111–127, doi:10.1023/A:1010359128574.

Becker, K. (1996), Permeability measurements in Hole 896A and implica-tions for the lateral variability of upper crustal permeability at Sites 504and 896, Proc. Ocean Drill. Program Sci. Results, 148, 353–363.

Becker, K., and E. Davis (2003), New evidence for age variation and scaleeffects of permeabilities of young oceanic crust from borehole thermaland pressure measurements, Earth Planet. Sci. Lett., 201, 499–508.

Becker, K., and A. T. Fisher (2000), Permeability of upper oceanic base-ment on the eastern flank of the Endeavor Ridge determined with drill-string packer experiments, J. Geophys. Res., 105, 897–912.

Becker, K., M. Langseth, R. P. Von Herzen, and R. Anderson (1983), Deepcrustal geothermal measurements, Hole 504B,Costa Rica Rift, J. Geo-phys. Res., 88, 3447–3457.

Becker, K., E. Davis, F. N. Speiss, and C. P. de Moustier (2004), Tempera-ture and video logs from the upper oceanic crust, Holes 504B and 896A,Costa Rica Rift flank: Implications for the permeability of upper oceaniccrust, Earth Planet. Sci. Lett., 222, 881–896.

Black, J. H. (1990), Focused packer testing: A structured approach to savetime and money, in Field Testing in Environmental Engineering, editedby F. G. Bell et al., pp. 313–321, Geol. Soc., London.

Brace, W. F. (1984), Permeability of crystalline rocks: new in-situ measure-ments, J. Geophys. Res., 89, 4327–4330.

Bredehoeft, J. D., and S. S. Papadopulos (1980), A method for determiningthe hydraulic properties of tight formations,Water Resour. Res., 16, 233–238.

Bruns, T., and D. Lavoie (1994), Bulk permeability of young backarc basaltin the Lau Basin from a downhole packer experiment (Hole 839B), Proc.Ocean Drill. Program Sci. Results, 135, 805–816.

Clauser, C. (1992), Permeability of crystalline rocks, Eos Trans. AmericanGeophysical Union, 73(233), 237–238.

Cooper, H. H., Jr., and C. E. Jacob (1946), A generalized graphical methodfor evaluating formation constraints and summarizing well-field history,Eos Trans. AGU, 27, 526–534.

Davis, E. E., and K. Becker (2002), Observations of natural-state fluidpressures and temperatures in young oceanic crust and inferences regard-ing hydrothermal circulation, Earth Planet. Sci. Lett., 204, 231–248,doi:10.1016/S0012-821X(02)00982-2.

Davis, E. E., and R. G. Currie (1993), Geophysical observations of thenorthern Juan de Fuca Ridge system: Lessons in seafloor spreading,Can. J. Earth Sci., 30, 278–300.

Davis, E. E., and H. Elderfield (2004), Hydrogeology of the Oceanic Litho-sphere, 706 pp., Cambridge Univ. Press, New York.

Davis, E. E., and C. R. B. Lister (1977), Heat flow measured over the Juande Fuca Ridge: Evidence for widespread hydrothermal circulation in ahighly heat-transportive crust, J. Geophys. Res., 82, 4845–4860.

Davis, E. E., et al. (1992), FlankFlux: An experiment to study the nature ofhydrothermal circulation in young oceanic crust, Can. J. Earth Sci., 29,925–952.

Davis, E. E., A. T. Fisher, and J. Firth (Eds.) (1997a), Proceedings of theOcean Drilling Program, Initial Reports, vol. 168, 470 pp., Ocean Drill.Program, College Station, Tex.

Davis, E. E., K.Wang, J. He, D. S. Chapman, H. Villinger, andA. Rosenberger(1997b), An unequivocal case for high Nusselt-number hydrothermal


11 of 12

B07105

convection in sediment-buried igneous oceanic crust, Earth Planet. Sci.Lett., 146, 137–150, doi:10.1016/S0012-821X(96)00212-9.

Davis, E. E., K. Wang, K. Becker, and R. E. Thompson (2000), Formation-scale hydraulic and mechanical properties of oceanic crust inferred frompore-pressure response to periodic seafloor loading, J. Geophys. Res.,105, 13,423–13,435.

Davis, E. E., W. Wang, R. E. Thomson, K. Becker, and J. F. Cassidy (2001),An episode of seafloor spreading and associated plate deformation in-ferred from crustal fluid pressure transients, J. Geophys. Res., 106,21,953–21,963.

Dawson, K. J., and J. D. Istok (1991), Aquifer Testing, Design and Analysisof Pumping and Slug Tests, 344 pp., Lewis Publ., Chelsea, Mich.

Fisher, A. T. (1998), Permeability within basaltic oceanic crust, Rev. Geo-phys., 36, 143–182.

Fisher, A. T. (2005), Marine hydrogeology: Future prospects for majoradvances, Hydrol. J., 13, 69–97, doi:10.1007/s10040-004-0400-y.

Fisher, A. T., and K. Becker (2000), Channelized fluid flow in oceanic crustreconciles heat-flow and permeability data, Nature, 403, 71 – 74,doi:10.1038/47463.

Fisher, A. T., K. Becker, and E. E. Davis (1997), The permeability of youngoceanic crust east of Juan de Fuca Ridge determined using boreholethermal measurements, Geophys. Res. Lett., 24, 1311–1314.

Fisher, A. T., T. Urabe, and A. Klaus (Eds.) (2005a), Proceedings of theIntegrated Ocean Drilling Program, Expedition Reports, vol. 301,doi:10.2204/iodp.proc.301.2005, Integrated Ocean Drill. Program, Col-lege Station, Tex.

Fisher, A. T., et al. (2005b), Scientific and technical design and deploymentof long-term, subseafloor observatories for hydrogeologic and relatedexperiments, Proc. Integrated Ocean Drill. Program, Exped. Rep., 301,doi:10.2204/iodp.proc.301.103.2005.

Fisher, A. T., E. E. Davis, and K. Becker (2008), Borehole-to-boreholehydrologic response across 2.4 km in the upper oceanic crust: implica-tions for crustal-scale properties, J. Geophys. Res., 113, B07105,doi:10.1029/2007JB005447.

Ge, S., et al. (2002), Hydrogeology Program Planning Group Final Report,Joint Oceanogr. Inst. for Deep Earth Sampling, Washington, D. C.

Halford, K. J., W. D. Weight, and R. P. Schreiber (2006), Interpretation oftransmissivity estimates from single-well pumping aquifer tests, GroundWater, 44(3), 467–471, doi:10.1111/j.1745-6584.2005.00151.x.

Hantush, M. S. (1960), Modification of the theory of leaky aquifers,J. Geophys. Res., 65, 3713–3725.

Harris, R. N., and S. M. Higgins (2008), A permeability estimate in 56 Macrust at ODP Hole 642E, Vøring Plateau, Earth Planet. Sci. Lett., 267,378–385, doi:10.1016/j.epsl.2007.11.055.

Horner, D. R. (1951), Pressure build-up in wells, Proc. World Pet. Congr.,3, 501–521.

Hutnak, M., A. T. Fisher, L. Zuhlsdorff, V. Spiess, P. H. Stauffer, and C. W.Gable (2006), Hydrothermal recharge and discharge guided by basementoutcrops on 0.7 – 3.6 Ma seafloor east of the Juan de Fuca Ridge:Observations and numerical models, Geochem. Geophys. Geosyst., 7,Q07O02, doi:10.1029/2006GC001242.

Ingebritsen, S., and C. E. Manning (1999), Geological implications of apermeability-depth curve for the continental crust, Geology, 27, 1107–1110, doi:10.1130/0091-7613(1999)027<1107:GIOAPD>2.3.CO;2.

Kappel, E. S., and W. B. F. Ryan (1986), Volcanic episodicity and a non-steady state rift valley along the northeast pacific spreading centers:evidence from sea marc I, J. Geophys. Res., 91, 13,925–13,940.

Karsten, J. L., S. R. Hammond, E. E. Davis, and R. G. Currie (1986),Detailed geomorphology and neotectonics of the Endeavour segment,Juan de Fuca Ridge: New results from Seabeam swath mapping, Geol.Soc. Am. Bull., 97, 213 –221, doi:10.1130/0016-7606(1986)97<213:DGANOT>2.0.CO;2.

Larson, R. L., A. T. Fisher, R. Jarrard, and K. Becker (1993), Highlypermeable and layered Jurassic oceanic crust in the western Pacific, EarthPlanet, Sci. Lett., 119, 71–83, doi:10.1016/0012-821X(93)90007-V.

Morin, R. H., A. E. Hess, and K. Becker (1992), In situ measurements offluid flow in DSDP Holes 395A and 534A: results from the Dianautprogram, Geophys. Res. Lett., 19, 509–512.

Mottl, M. J., et al. (1998), Warm springs discovered on 3.5 Ma oceaniccrust, eastern flank of the Juan de Fuca Ridge, Geology, 26, 51–54,doi:10.1130/0091-7613(1998)026<0051:WSDOMO>2.3.CO;2.

Paillet, F. L., A. E. Hess, C. H. Cheng, and E. Hardin (1987), Character-ization of fracture permeability with high-resolution vertical flow mea-surements during borehole pumping, Ground Water, 25, 28 – 40,doi:10.1111/j.1745-6584.1987.tb02113.x.

Shipboard Scientific Party (2005a), Expedition 301 summary, Proc. Inte-grated Ocean Drill. Program, Exped. Rep., 301, doi:10.2204/iodp.proc.301.101.2005.

Shipboard Scientific Party (2005b), Site 1301, Proc. Integrated OceanDrill. Program, Exped. Rep., 301, doi:10.2204/iodp.proc.301.101.2005.

Snow, D. T. (1968), Rock fracture spacings, openings, and porosities, J. SoilMech. Found. Div. Am. Soc. Civ. Eng., 94, 73–91.

Spinelli, G. A., and A. T. Fisher (2004), Hydrothermal circulation withintopographically rough basaltic basement on the Juan de Fuca Ridge flank,Geochem. Geophys. Geosyst., 5, Q02001, doi:10.1029/2003GC000616.

Theis, C. V. (1935), The lowering of the piezometer surface and the rate anddischarge of a well using ground-water storage, Eos Trans. AGU, 16,519–524.

Thomson, R. E., E. E. Davis, and B. J. Burd (1995), Hydrothermal ventingand geothermal heating in Cascadia Basin, J. Geophys. Res., 100, 6121–6141.

Underwood, M., K. D. Hoke, A. T. Fisher, E. G. Giambalvo, E. E. Davis,and L. Zuhlsdorff (2005), Provenance, stratigraphic architecture, andhydrogeologic effects of turbidites in northwestern Cascadia Basin,Pacific Ocean, J. Sediment. Res., 75(1), 149 – 164, doi:10.2110/jsr.2005.012.

Wheat, C. G., and M. J. Mottl (1994), Hydrothermal circulation, Juan deFuca Ridge eastern flank: factors controlling basement water composi-tion, J. Geophys. Res., 99, 3067–3080.

Wheat, C. G., M. J. Mottl, A. T. Fisher, D. Kadko, E. E. Davis,and E. Baker (2004), Heat flow through a basaltic outcrop on a sedimen-ted young ridge flank, Geochem. Geophys. Geosyst., 5, Q12006,doi:10.1029/2004GC000700.

��K. Becker, Division of Marine Geology and Geophysics, Rosenstiel

School of Marine and Atmospheric Sciences, University of Miami, 4600Rickenbacker Causeway, Miami, FL 33149, USA. ([email protected])A. T. Fisher, Earth and Planetary Sciences Department, Institute for

Geophysics and Planetary Physics, University of California, Santa Cruz,CA 95064, USA.


12 of 12

B07105

Documents

Borehole packer tests at multiple depths resolve distinct ...afisher/CVpubs/pubs/BeckerFisher2008_JGR.pdfBorehole packer tests at multiple depths resolve distinct hydrologic intervals