ML20031D785
| ML20031D785 | |
| Person / Time | |
|---|---|
| Site: | Wolf Creek  |
| Issue date: | 10/09/1981 |
| From: | Koester G KANSAS GAS & ELECTRIC CO. |
| To: | Harold Denton Office of Nuclear Reactor Regulation |
| References | |
| KMLNRC-81-119, NUDOCS 8110140184 | |
| Download: ML20031D785 (20) | |
Text
,..
KANSAS GAS AND ELECTRIC COAfPANY o,+,. o j
. n, e n.s OLFNN L D. O f 5 Y E R
..<..w.,
g...
October 9, 1981 Mr. Harold R.
Denton, Director
<b' Office of Nuclear Reactor Regulacion
[
I Nuclear Regulatory Commission g
L
~1 OCT1 3 '38I,,k Warhington, D.C.
20555
/
KMLNRC 81-119 h, U.S.$%
7 Re:
Docket No. STN 50-482 b/j Ref:
Letter dated 5/12/81 from RLTedesco, NRC to GLKoester, KG&E g
Subj. Liquid Pathway Class 9 Accident Analysis
Dear Mr. Denton:
The referenced letter requested additional information concerning the Wolf Creek Generating Station, ' snit No. 1 Environmental Report (Operating License S tace). As part of that request, KGAE was asked to perform a liquid pathway class 9 accident analysis assessment (Question 240.13).
Dames and Moore has performed tha requested analysis for KG6E.
Their
'r 3 report describing the analysis procedure and parametric values used is attached.
It is concluded in the analysis that essentially no dose to the public would be experienced as a result of a liquid pathway release from a postulated core melt. accident.
The Wolf Creek Fnvironmental Report will be revised to reflect the results of this analysis.
Yours very truly, pWY GLK:gm Attach cc:
Dr. Gordon Edison (2)
Mr. Thomas Vandel Division of Project Management Resident NRC Inspector Office of Nuclear Peactor Regulation Box 311 U.S. Nuclear Regulatory Coamission Burlington, KS 66839 Washington, D.C.
20555 8110140184 811009 OJ POR ADOCK 05000482 V
O PDR 5\\\\
701 N Marker - Wehrta. stansas - MA:I Att<4ress' PO. Bou 208 ; Wich.ta, k ansas 6??Of - Telephone Area cme (3161 26 f-6451
.t OATil OF AFFIRMATIOf1 STATE OF KAIISAS
)
) SS:
COU JTY OF SEDGWICK )
I, Glenn L.
Koester, of lawful age, being duly s-rn upon oath, do depose, state and affirm that I am Vice President - fluclear of Kansas Gas and Electric Company, Wichita, Kansas, that I have signed the foregoing letter of transmittal, know the contents thereof, and that all.~tatements contained therein are true.
KNJSAS GAS Af1D ELECTRIC COMPA!JY ATTEST:
h,f;"JL)
$ g(g g 9
By
'Gienn L. Koester' g
V Vlec President - fluclear W.B.
Walker, Secretary STATE OF KN15AS
)
) SS:
COU;1TY OF SEDGWICK )
BE IT RE!CI1BERED that on this 9 tl1 day of _Qcto_bm 1981
, be fore me, Evelyn L.
Fry, a TJotary, personally appeared Glenn L. Koester, Vice President - fluclear of Kansas Gas and Electric Company, Wichita, Kansas, who is personally known to me and who executed the foregoing instrument, and he duly acknowledged the execution of the same for and on behalf of and as the act and deed of said corporation.
If1 WITf1ESS WilEREOF, I have hereunto set my hand and a f fixed my seal the
,......djf,te and year above written.
,.O,,, g
./.e, K.'.
~.
[
g,CTile),
/ 'zh v V i
e *-=
i
_.velyn id Fry,110t.a O C O.
,* e:-
My Commission expires on August 15, 1985.
i
FINAL REPORT CLASS 9 ACCIDENT LIQUID PATHWAY ASSESSMENT WOLF CREEK GENERATING STATION
1.0 INTRODUCTION
The purposa of this analysis is to provide a response to a question posed by the Nuclear Regulatory Commission (NRC) to Eansas Gas and Electric Company (KG&E) regarding the Environmental Report (ER) for KG&E's Wolf Creek Generating Station. The question posed by the NRC is:
240.13 Calculate the radiological consequences of a liquid pathway (ER) release from a postulated core melt accident. The analysis (7.1) should assume, unless otherwise justified, that there has been a penetration of the reactor basemat by the molten i
core mass, and that a substantial portion of radioactively contaminated sump water was released to the ground.
Doses should be compared to those calculated in the Liquid Path-way Generic Study (NUREG-0440, 1978). Provide a summary of your analysis procedures and the values of parameters used (such as permeabilities, gradients, populations effected, water use). It is suggested that meetings with the staff of the Hydrologic Engineering Section be arranged so that we may share with you the body of information necessary to perform this analysis.
Several meetings were held between representatives of KG&E, Dames & Moore, and the NRC regarding this and other questions. An informal meecing was held on April 21, 1981 at the Wolf Creek site regarding preliminary discussion of the question. Another meeting, which was open to the public, was held the following day in the National Guard Armory in Burlington, Kansas. At the meeting the NRC posed the question as stated above. On behalf of KG&E, Dames & Moore gave a preliminary partial answer, the substance of which is as follows:
"In order to provide c complete responsive answer to your question, we would need more details concerning the assump-tions and parameters involved in the postulated core melt.
We concur with your suggestion that our technical people meet with ye r technical people in order to share information required to make the analysis.
In the meantime, I can provide you with a preliminary partial answer indicating the order of magnitude of travel times expected for radionuclides entering the ground water as a result of a core melt.
1
The ground water gradient would be from the reactor site toward the cooling lake. There are no water supply wells along this ground water pathway. Thus, no ground water used for public supply would be endangered.
Because of the expected low hydraulic gradieta, the low permeability of the rock formations, and the retardation of Sr-90 and C -137 in the subsurface materials, the expected s
travel times for S -90 and Cs-137 are very much greater t
than the travel times for S -90 and C -137 postulated in r
3 the " Liquid Pathway Generic Study".
Most of the radio-isotopes are expected to be decayed before reaching the cooling lake. Radio-isotopes which do reach the lake are expected to be greatly reduced in concentration initially by radioactive decay and subsequer.tly by dilution in the lake water.
Resulting concentrations of radio-isotopes in the lake would depend upon the initial volumes and concentrations, which have not yet been specified".
Following this meeting arrangements were made for KG&E and Dames & Moore to meet with the NRC technical personnel in Washington, D. C.
At that meeting, held on May 8, 1981, further discussions were held regarding the NRC Question (240.13). At that meeting the NRC indicated that they were not expecting a complex computer model of a ground water flow system. The ERC indicated that hand calculations using standard analytical techniques would provide the level of sophisti-cation that they were seeking. The NRC also indicated that dispersion could be neglected in the ground water transport cr.lculations. The NRC then provided KG&E and Dames & bbore with a number of reference documents for the purpose of providing background data and assumptions for the liquid pathway study.
The documents provided by the NRC for this purpose were:
NUREG-0440 Liquid Pathway Generic Study, February 1980.
NUREC/CR-0912, Vol.1 and 2, UCRL-52719, Vol.1 and 2, Geoscience Data Base Handbook for Modeling and Nuclear Waste Repository, January 1981.
Harris, V. A. ; Yang, Ting-Yea; and Arkentien, J. S. ; decident Mitigation-Slurry Wall Barriers.
Masnik, M. T.; Estimates of Fisheries Harvest Downstream of the Virgil C.
Summer Nuclear Power Station, Oct 1980.
i I
2
Masnik, M. T.; Estimates of Fisheries Harvest Downstream of the Grand Gulf Nuclear Station; May 1981.
NUREG-0534 Supplement; Draft' Environme$ttal Statement related to the operation of the Virgil C. Summer Nuclear Station Unit No.1, Docket No. 50-395 South Carolina Electric & Gas Company, November 1980.
Environmental Impacts of a Radiological Accident at San Onofre on Fisheries.
Nicholson, T.; Three Mile Island Nuclear Station Unit 2 Ground Water Studies, April 19, 1979, Docket No. 50-320.
Plant Accident Section for Environmental Impact Statement, Waterford Unit 3, Docket No. 50-382.
Plant Accident Section for Final Environmental. Statement San Onofre Units 2 and 3 and Responses to Comments on DES Docket No. 50-361/
362.
Plant Accident Section for Environment Impact Statement Fermi Unit 2, Docket No. 50-341.
Preliminary Assessment of the Consequences of Class 9 Accident at the Salem Nuclear Generating Station, Docket Nos. 50-272 and 50-311.
Class 9, Liquid Pathway Assessment Input to the Comanche Peak FES.
Comparison of ERIE Site with Liquid Pathway Generic Study Great Lakes Site by Richard B. Codell and Myron H. Fliegel.
Class 9, Liquid Pathway Assessment Input to Fermi EIS.
Environmental Statement Input for San Onofre Units 2 and 3, Supplement.
Sandia Draft Liquid Pathway Study, Appendix F.
Three untitled rough drafts of documents.
2.0 SCOPE OF WORK 2.1 Review of Reference Documents A review was made of each of the documents received from the NRC listed in Section 1.0 above. In addition, pertinent sections of the 3
. - =.
i Wolf Creek Generating Station FSAR Add 2ndum, Sectica 2.4 through 2.5 and the Wolf Creek Generating Station ER(OLS), Sections 2.4 through 13.0 were reviewed and referenced for background data to be used as a basis for response to the NRC question (240.13) posed above. No field work was undertaken other than a site visit made on April 21, 1981 to the Wolf Creek Generating Station.
2.2 Analyses and Results It is postulated (Reactor Safety Study WASH-1400) that the core meltdown could penetrate downward through the basemat into the underlying rock formations forming a cylindrical shaped rubble zone 70 feet in diameter and 50 feet deep. The elevation of the bottom of the basemat is given as 1064 feet in Table 2.5-53 of the FSAR. Thus, the zone affected by the meltdown would be between Elevation 1064 and Elevation
'314.
The formations affected by the meltdown would be, in descending order:
Heumader Shale Member Plattsmouth Limestone Member.
Undifferentiated Heebner Shale, Leavenworth Limestone, and Snyderville Shale Members Toronto Limestone Member Unnamed Lawrence Shale Member Initially, ground water outside the containment structure would leak into the containment structure until the water level inside the containment structure is equal to the piezometric level of the ground water outside the structure. This is directly analagous to a large diameter dug well which had been instantaneously pumped dry after which the pump had been turned off thereby allowing the water level to rise in the well. Af ter this state of equilibrium is reached, it would then be possible for water from the sumps containing radio-isotopes to leak downward through the rupture in the basemat into the j
radioactive debris zone cylinder in the bedrock under the basemat.
In addition, ground water flowing under the basemat would be expected to pass through the radioactive debris of the core melt i
thereby leaching radio-isotopes from the debris.
4
4 Ground water flew from the meltdown site could flow through any of the geological formations encountered by the meltdown as listed above. The direction of ground water flow would be expected to be toward the cooling water lake. Bec<use of the position of the reactor building on a peninsula, ground water from the core melt could flow radially toward the cooling water lake in directions which range from westerly to southerly to easterly. Which direction it would take is not known, but for conservatism it is assumed it will take the shortest 1
path.
The ground water would take the path of least resistance which would be along the formation having the highest permeability. According to the permeability tests performed on the site, the highest permeability
-0 measurement (2 x 10 cm/sec) was in the Plattsmouth Limestone. These permeability values are tabulated in Table 2.4-34 of the FSAR.
For the sake of conservatism, it is assumed that the piezometric.
surface for all of the geologic formations penetrated will initially be at or close to the land surface (Elevation 1099 feet). Whereas the elevation of the basemat is 1064, there would be a maximum 35 feet head differential between the outside ' ground water and any sump water within the reactor building. Thus, rupture of the basemat by meltdown and penetration of the underlying formation would cause ground water to enter the reactor building until the water level inside the building rises to the level of the ground water outside the building. While the water level inside the building is rising, the ground water leve. outside the building would be declining in a manner similar to the way the ground water level declines in response to pumping a well. The ground j
water would be entering the reactor building through the rupture-in the basemat af ter having passed through the highly permeable rubble in the penetration zone underlying the basemat. This inflow of ground water would bring in radioactive isotopes leached from the rubble zone. This entering ground water would mix with radioactive water assumed to be present in the sumps in the reactor building.
f 1
5
-_, _._ __.,. - _, ~... _ -
_. _. J Ji, _ _ _ -
- -. _ ~ ~.
i From a theoretical standpoint, the radioactive water in the s
ground under the reactor building would not be escaping from the vicinity of the reactor building until after the water level'inside the reactor building comes u,p to an elevation equal to or greater than the ground water level outside the building. The time required for this equilibrica to occur is finite and realistically can not be considered to be instantaneous. However, for the basis of our liquid pathway analysis, i
j the fill time for the reactor building can be neglected if it is small i
j compared to the length of time for the travel time for radioactive ground water to travel from the reactor site toward the nearest surface I
water body. For this basis we would arbitrarily consider a fill-up time amounting to ten percent or more of the expected travel time for the radioisotopes to be significant. A fill-up time of less than ten percent would be considered insignificant and would thus be neglected in the overall liquid pathway assessment. Neglecting this fill-up time would also support the conservative approach taken in this analysis.
The time required for ground water to fillup the reactor containment building as a result of a sudden breach in the basemat caused by a meltdown can be estimated by means of using a slug test analysis. A slug test is a method of determining the transmissivity and storage coefficient of an aquifer. hor injecting a slug of water into a well which is screened in the aquifer to be tested. An alternate method is to remove a slug of water from the well instantaneously by bailing. After injecting (or bailing) the slug of water in the well, the water levels in the well are measured while the water level is returning to its normal level.
=
By appropriate plotting of the water level recovery curves on semi-logarithmic paper, the recovery curves can be matched against a set of published curves whereby the transmissivity and storage coefficient can be calculated. Because the values of transmissivity and storage coefficient are already known, the inverse problem can be worked to calculate the time recovery curve for the water level if the reactor building and the underlying rubble zone are treated as a large well being subjected to a slug test wherein a slug of water has been instantly removed from the well at the time of meltdown.
6
The published curves all converge at 100 percent recovery.
However, the curves can be differentiated up to about 95 percent recovery.
For conservatism we will assume 95 percent to be equivalent to total recovery.
With reference to the well diagram in the Appendix, the well casing is considered analagous to the reactor containment building with a radius of r, the rubble zone underlying the reactor containnant e
building is considered anala;,ous to the screened portion of a well with a radius of r, H is the initial water level difference in the s
o "well", and H is the water level difference at time t.
The value of Ho would be the difference between the elevation of the basemat (Elevation 1064) and the elevation of the static water level outside the reactor building (Elevation 1099).
The values used in the calculations are:
rc = 2438 cm rs = 1050 cm S = assumed storage coefficient of.12,*
8
-2 thus,p=
S=
(.12) = 2.23 x 10 23
-4 Kh = 2 x 10 cm/see permeability.for Plattsmouth Limestone d = 1600 cm thickness of entire penetration zone
-1 T = K d = 3.2 x 10 cm2/see transmissivity.
h
-2 From graph of p = 10 curve at H/H =.05 o
Tt/r
=8 c
)
8 l0-1 = 1.48 x 10 t=
sec
- Value obtained from FSAR Section 2.4.13.3.3, Page 2.4-67.
It is recognized (1) that the porosity determination in the FSAR was made for the Heumader shale on the basis of bulk density measurements, the porosity measurements do not account for fractures, and (2) that (3) that flow through the Plattsmough limestone could probably be primarily by flow through fractures.
We are, therefore, assuming that the average void ratio of fractures in the Plattsmouth limestone is equivalent to the porosity measure-ments in the Heumader shale. Because of expected variations in fracture patterns, the ground water velocity could be expected to deviate locally from the average value calculated in the analysis.
7 n-
Thus, t = 4.7 years for the water level in the breached reactor building to come to 95 percent of the static ground water level outside the reactor building. This is a conservative estimate. If a smaller storage coefficient is used in the estimate and if only the actual thickness of the Plattsmouth Limestone is used re.aer than the totti thickness of sediments penetrated (most of which are shale with
~0 a permeability of about 10 cm/sec) the time required for water to fill the reactor vessel room would be considerably longer.
For the sake of being conservative in the overall calculations for the liquid pathway, and to be consistent with the analysis scheme used in the LPGS NUREG-0440, which assumes an instantaneous release, the length of time required to fill the reactor room will be ignored.
For the purposes of the liquid pathway analysis, it will be assumed that the radioactive water inside the containment building will leak into the radioactive rubble zone underlying the basemat.
This leakage is ordinarily expected to take a relatively long time, especially if the water level inside the building is not significantly higher than the ground water level outside the building. In order for all of the radioactive water to leak out of the building, it would have to be replaced by clean water while the leakage is taking place.
For conservatism, it is assumed that the leakage of radioactive water into the underlying rubble zone will take place immediately after the meltdown.
It is assumed that the primary ground water flow path.
will be through the zone of greatest permeability which, according to field and laboratory tests, is the Plattsmouth Limestone with a maximum measured permeability of 2 x 10~ cm/sec.
It is not known in which direction ground water passing through the rubble zone would flow in order to reach the Plattsmouth Limestone outcrop zone the cooling water lake. However, for conservatism it is assumed that it will take a path which has the shortest distance from the reactor building to the outcrop zone which is about 2600 feet (79,200 cm) in a southeasterly direction.
8
The hydraulic gradfent (I) would be(Gr und water level uder 1
(Distance from reactor to Reactor) - (Normal operating level of cooling ~1ake),'1099-1087
= 0.00462 outcrop zone in cooling lake) 2600 With an effective porosity of 0.12 the ground water approach velocity is 1
V=
n 2 x 10~
ec (0.00462)
-0 cm
= 7.7 x 10 cm/sec The time required for ground water to travel 79,200 cm would 0
thus be 1.03 x 10 see or 326 years.
The half life of S -90 is 28 years and the half life of Cs-137 r
is 30 years. From these half lives we can calculate what the activity of these isotopes would be in 326 years in terms of a decimal fraction of the initial activity. Activityt = Activity to exp (-kt).
Where k =.693/ half life However, the radio-isotopes of Sr-50 and Cs-137 are retarded in the soil and travel slower through the aquifer than the ground water in which they are dissolved. For conservatism, we will assume retardation factors of at least 9.3 for Sr-90 and 83 Cs-137. These are the values j
used in the Liquid Pathway Generic Study. Retardation factors for
{
limestone would actually be greater than these values.
With these retardation factors the travel time would be 3,031 7
years for Sr-90 and 27,000 years for Cs-137.
Source terms for these two isotopes. are assumed to be 24 6 curies) and 100 percent percent of initial inventory for Sr-90 (6.1 x 10 i
of initial inventory for C -137 (8.6 x 106 curies).
(Maximum values from s
Table A-8 of LPGS NUREG-0440, and as specified by NRC technical staff I
during May 8,1981 meeting in Washington, D.C.)'
On this basis the decimal fraction activity of Sr -90 after.3031 years is:
,693
-33
~
exp (
3031) = 2.6 x 10 Likewise, the decimal fraction activity of Cs -137 after 27,000 years is:
~* 9
}
exp (
27,000) = less than 1 x 10 "
3 1
Thus, the total quantity of Sr -90 reaching the cooling water lake.after 3031 years is:
6.1 x 100 (.24) (2.6 x 10-33) = 3. 8 x 10-curies 9
The total quantity of Cs -137 reaching the cooling water lake after 27,000 years is less than:
8.6 x 106 (1.00) (1 x 10 ") which is less than 8.6 x 10-93 curies.
On this basis, it is estimated that virtually all of Sr -90 and Cs -137 will have decayed before having reached the cooling water lake. The travel times for these isotopes are very much longer tha.a those estimated in the LPCS land based river case.
The previous analysis was based on the assumption that the cooling water lake would be at its design operating level. If the cooling water lake level were to be lowered below the Plattstucuth Limestone cutcrop elevation, and if the ground water piezometric elevation were to remain at Elevation 1099, the hydraulic gradient in the aquifer would be increased, thereby allowing the ground water and the radio-isotopes to travel somewhat more rapidly toward the cooling water lake.
For this case, the piazometric level at the discharge end of the aquifer would be the elevation of the aquifer, approximately 09 Elevation 1060. The gradient would be
= 0.015.
2600 The approach velocity now would be:.
2 x 10 0.015)
-5
= 2.5 x 10 cm/see The time required for ground water to travel 79,200 cm would be 3.2 x 10 sec (100 years).
Using the conservative retardation factors of at least 9.3 for Sr -90 and 83 for Cs - 137 gives travel times of 930 years for Sr - 90 and 8300 years for Cs - 137 to reach the cooling water lake.
The decimal fraction activity of Sr -90 remaining after 930 years is:
-10 exp (~
930) = 1.0 x 10 8
The decimal fraction activity of Cs - 137 remainir.g after 8300 years is:
~
-84 exp ( ~ 0 8300) = 5.4 x 10 The total quantity of Sr -90 reaching the outcrop in the cooling water lake area would be:
6.1 x 106 (.24) (1.0 x 10-10) = 1.5 x 10-4 curies 10
The total quantity of Cs - 137 reaching the outcrop in the cooling water lake area would be:
8.6 x 100 (1.00) (5.4 x 10-84) = 4.6 x 10
~
curies The travel times for these isotopes are much greater than the travel times estimated in the LPGS land based river case.
Virtually all of the activity will have decayed before reaching the outcrop in the cooling water lake, assuming that the water level in the lake is below the Plattsmouth Limestone outcrop.
2.3 Interdiction There are several viable options regarding interdiction of the radio-isotopes in the ground water following a core meltdown. These viable options are as follows:
1.
Pump water out of the reactor building, thereby causing it to act as a large diameter well.
In this manner radio-isotopes in the ground water under or near the reactor would be drawn into the reactor building and recovered. The cone of depression thereby created in the various formations penetrated by the meltdown would minimize the migration of radioactive ground water away from the site. Because of the slow travel rates for the ground water in this area, sufficient time, on the order of several years, would be available to initiate interdiction action.
2.
Because of the extremely long times required for the radio-isotopes to move through the ground, it appears that a feasible option is to have no interdiction action. There are no ground water users between the plant and the cooling water lake. By the time the radio-isotopes reach the lake they would be reduced in activity to a very low level. Of course, if this option were to be taken, it would be necessary to install an observation well network around the plant to monitor the movement of the ground water and radio-isotopes in the ground water.
f 11
The possibility of constructing a slurry trench cut-off wall around the reactor site was investigated. However, the expected effective
-4 permeability of a slurry trench cut-off wall is in the range of 10
-0 to 10 cm/sec, which is in the same order of magnitude as the existing rocks in the site area. Thus, a slurry trench would not be expected to affect the ground water velocity significantly.
3.0
SUMMARY
AND CONCLUSIONS For a Class 9 accident at the Wolf Creek site, the travel times for Sr-90 and Cs-137 (930 and 8300 years, respectively) via the liquid pathway are calculated to be much greater than the travel times estimated in the Liquid Pathway Generic Study (NUREG-0440) for the land based river case. Virtually all of the activity will have decayed before reaching the Wolf Creek cooling lake (or the Plattsmouth Limestone outcrop in the absence of the cooling lake.) Because of the extremely small amount of radiation reaching the cooling lake, there would be essentially no dose rate of radioactivity to the public. The population he would, therefoce, be very much smaller for the Wolf Creek site then it is for the generic case in NUREG-0440.
In addition, there are several options regarding interdiction which could *a exercised to further n.inimize the liquid pathway impact.
I
+
l 12 i
9 8
e e
e APPENDIX f
a
- r t
-*T---
y w,
y-9
-w wye-oy,-
y---
gw-
.,7-
--7 7
y w
u
-p+
g - - - - - - -
e
=
--a r-m=
+-
y-
w n
4 b
E a
b b' EO!Ogy E
"#i>
C. W. FETTER, JR.
pf Universi9 of Wisconsin-Oshkosh P b?
t t..
E t
I t
~e rp i
e:
g Charles E. Merrill Publishing Company I
3 A Be!! & Ho'.wli Company
' dney Y
Columbus To*onto London Si
' ^'}
1 i
2
+
YS*.
cp. -
mad
~.} $;
-/iJh,!f
~
CROUNDWATER FLOW TO WELLS 9e.f,
$ h' The data are plotted in Figure 8.7. The drawdown per log cycle is 8.8 feet and r is 460 feet. Find the values of T and S.
o scree k
(s 523Q radiu
.!P.
y,f '
T = ath, - h) withe u.I initia; 528 x 400 T.
eleva
=
1 8.8
' ^'T
= 24,000 gal / day /ft
?'$Y Tt 9
i S = 4790q, s-3 b
,24,000 x 200
- m. i.
4790 x 460'
- i. '
= 0.0047 a' F. '
l 20 g
a
!:l w
14 l
I
- ,s 1.
16 i
4 8, '
- k a- - - -- --- - - - -l-- 7 i
3
..e p
'2 i
.$ j -g,
$ 10
,hn - hl = a 8 8
i
. + N3{ 8 f,
r.* SP8 if,/j!
2 j
O
' }
,,,..i r,o -
r: Guai p
->f [g Of a Ct
?
I r*
1 to suo t000 p
tf Papado-Distance un v:
ss,I.
~fdf] _
rlCURE 3.7.
Variar:on of the Jacob method of solution of pumpin;; test data for a fully
/{7g j
confined aquifer. Drawdown is p! cited as a function of d <tance to observation well on tion (H y
semilegarithmic paper.
scale,.
U
"[-* ? t
}
8.3.4 SLUG TESTS g l at-where f i?.th in some field investigations, the practicing hydrogeologist may be working with Iup low-permeability materials. This is especially likely in site studies for areas of 6 MN 3
potential waste storage or disposal. The soil materials in these areas may have J"d
- ii' ?.h a conductivity too small to conduct a pumping test. An altemative me: hod of
.} ;k, testing involves either injecting into or withdrawing a slug ci water of known w t volume from the well. The rate at which the water level rises or fails is con-
'f..
trciled by the formation characteristics. This is known as a slug test (7,8,9).
form in I' $.
s3 i."
ik 270 4, ilt.
- l i
4 oe.
,.0,
'.. s 3.
r
'd.g,,'
2-l 8.3 WELL HN DPAULICS IN A COMPLETELY CONFINED AREALLY EXTENSIVE AQUIFER i- !
l
- ii
[
f...f:'; k it A well is drilled and ca>ed abose the confined aquifer. A well
,]
6 l
screen or open ho'e penetrates (Fe aquifer (Figure 8.8). The well casing has a radius, r, and the well screen has a radius, r,. Immediately after injection or j ;/7E j.
withdrawal, the water les el in the well has an elevation, Ho, above or below the J.
1 initial head. As the water level rises or falls, the difference. H, in water-level 3Y elevat.on bets. ten that at time t and that at the original head is measured.
[h,k ij
.g.
v Water lesel ammediately
'ft D i
j
/ airer eniection s
/
(,Ql'-
I ge 1
Water level at time r
, Head in aquifer 7;! ' g j
l
- at time r
.t..
g.,
i h.r to
...a s
t 4.*
j
~ \\initi I head
. ~IY in aqu;ter
./l. p g
-- We:1 casing tg
[.',
p}.
s',:wv:m';o v
':'/;rav:rn;m' mymx:
, :,. i l.
-}3 @. h
-l C. -
i Weil screen or I
I i.
l' wall of open hole' 3
t-I Ar,urer I g
.3
.$. h as,
- 1
.p.
l ll.
f
.I l
- yg- }-
l 6 f J.t -
' 1:.JJ. ' ' i,.' </,w:'? ; n?>;: N,
' i;wlM:/4nON RN fy, i
F f GU'tE 3.8.
Wi4' ir i w. ;ch a solun:e. V. of water is suddenly injected for a slug test d
J or a confir < d aqu.te-SOURCE: H H. Cooper Jr, J. D. Bredehoeft, and I. S.
~ ?d Pa;ut'ept.lm. W.t:er bouc es Research. I (19M):263-6(3.
i
.iF &
e 11 *D A
A p!ot of the ratio or the measured head to the head after injec.
.7 {
. n.i tica (H:H ) is rn.* a, a function of time. The ratio H/H is on the arithmetid
/'k N o
scale, and tirr.e 6 on the 'opiithmic sca!c of semilogarithmic paper:
' }q A.?c-0l}~
H:H = F(v1.p)
(8 16) a u he -
.jf1
..w>..,.
c
.c;
'th 77 m 7gjfl (g.j7)
. ot UCd d. (4 we
- O
2 g = r,2Lr (8 18)
.j:,g.J.
wn fir < 4) :s a 'urmtion, the values of which are given in tabulatcd
, jt, l.
in-
.,y,
form :n Appem:r..'.9,. TI e tabulatal vaiues are plotted as a series of type j Y b..e [IA e
a*
i M%..*
271 N
3 j t.'
3.w.
.5r. $NoN\\iYo* nee NNoEN$nW5DSU'at i ~ n ir 5^
Ht a
ls
.s s. tr
.s
, A'p&rM Sv--ge qp$ip %@e.Weq'gbyQJ/{v,yjrp.*pWWm pwG.;L,c c/pg,g4gA shifwamb maa dasMMmb2s2n%y*
r e
g veg p
' k?M) 1 4.v%
F % $'O:Q. ';t 4-y g).[ay,?y CROU.NOWATER FLOW TO WELLS
. yn n..-4 ~
d '..
curves in Figure 8.9. The type curves should be en semilogarithniic paper with prc
,l t
the same scale as field data.
tati
-7 An.
s su j' y,J g.t
\\i\\$
"*{'
Ob fN dT
- 1' os rf.
'g f.h un ihf.'
I bMg)h
\\
de?
& 9'
\\
g\\
\\a -ry gy,'%*
- g,.
'V l
\\
,y
.e k.
os 6 1 i-ei i
b, g g
_\\
l r.
.+h. k '
c as h
- h.\\,
t
- q..
~
.,. \\
3??,...
\\
.84,'Ml(;"
3 o4
-??W.
N
',Qe17 ^
e- \\
0'
~
}O
?!*QX.%
G 4%4 M:.
- t. \\ '::
,,\\
4
%\\
\\\\
IN$ddC
'l6h
~
(\\
\\\\
M)d)u,fpt-c,\\
,i Ys p\\h
- +i pt
$'# %f '5' "hQk'
.Q)Lt.,ph..-;..
1
'~
to to-'
ir3 to-'
i to s o-
- f h Tr. ;y
?/f[iN I
FIGURE 8.9.
T.pe curves for slug test in a v. ell of finite ciiameter. SOURCE:
S. 5.
Papacopulor.. J. D. Bredehocit, and H. H. Cooper, Jr., nater Recurces Researcl7, 9
,$;P hjg.hci
'I.~
t19731.1037-89.
in 13
. T The field-data curve is placed over the type curves with the
- ), e5g [;
f arithmetic axis coincident. That is, the value of H/H = 1 for the field data lies o
'I 2 on the horicontal axis of 1.0 cn the type curve. The data are matched to the f
h (fh type curve ( ), which has the sarne curvature. The vertical time-asis, t, which i
,yg,7,J. -
oserlays the vertical axis for 7t/r; = 1.0 is selected. The transmissivity is found hkbh I*
l rDiMP 1.Or.
T=
(8 19) g.'.-Q,$p k'.
Yh'Y
- gpj egt
The value of storativity can be found from the value of the
,Qili'c
- p. curse for the field data. Since
= (r{/rl)S, i
" r. ;:-r, P (d. 9,WNi 5 = (rl )/rj f3-20) 7,[j'.",y fj,
For sma!I values of, however, the curves are ofte.. very similar: therefore, in h*O matchir g the field data, the question of which u value to use is often GLJtw encountered. The use of this method to estirrate storativity shou!d be ap-w.4.c ;
f.hifti 272 p' ji i
a e e
F
- ],1 b
'd 8.3 urLL HYDRAULICS IN A CCMPLETELY CONFINED AREALLY EXTENSIVE AQUlFER j[
4
.4
,er with proached with caution. Likewise, the value of T that is determined is represen.
5 '
tative only of the formation in the immu!iate vicinity of the test hole.
')
Several other types of slu,; tests can be run. One useful proce-3<
dure can be applied to both confined and unconfined aquifers and used for
,'S ci:her partially penetrating or fully penetrating wells (10). This procedure is M
based oa a bail-down-type test, in which the well has a volume of water 4
suddenly remosed by bailing. The rate of nse of water in the well is then
- .ff.
j observed. Auger holes can also be used for slug tests (11). These are shallow, unlined, cyhndrical holes, augered by hand or machine for the purpose of
- @6 determining transmissivity.
GP1 EXAMPLE PROBLEM l.
1
}
A casing with a radius of 7.6 ant: meters is installed through a g4-
- .a confining layer. A screen with a radius of 5.1 centimeters is installed
}[
t in a formation with a thickr'ess of 5 meters. A slug of water is
, d.Y injected, raising the water lesel 0.42 meter. The decline of the head
- j$
is gven in Table 8.2. Find the salues of 7, K, and S.
l M.(
TABLE 8.2
'.0 5-Tirne asect H tr.:)
H1f d.
n l
2
- 0. 3 /
0 89 l
5 0 34 0.81
.,g a
10 0 27 0.65 g,.
21 0.18 0.43
- 4 46 0.09 0.21 hjp 55.
70 0.05 0.11
!rth
.,g f,, 9 1GO O 02 0.06
,.n s
- A plot of H/ll, as a function of f is made on semi!oganthm;c paper. It is overhin on the type curve (Figure 8.10).
ith th" l.lf[
ata lies At the axis for Tt/r2,1.0, the value r, is 13 seccnds.
' '[A.
l to the
- s *.,y
. hid1 T = ' O
'0*Ib'm
. =
= 4.44 cm /sec l.d5k i found f
13 SCC g
i K = isb
. ei 't ':
= 3 85 (m'.'sec/500 cm
- .'I'y (d-19) j g x gg.,
of the The g-curse is 10". Wi:h r, e-7.6 centimeters and r, = 5.1 cen-iM timeters, we find (8 20)
S = ie f)'t,'
' @~
c,,g, in
- - i l O " < 7.6 ').'f 3.1)#
,l G.
e:
gllg p
=,
- 4 10-'
'.9 a
!:e ap-
!.'th,h
! D" M
273
& 3,
, b,I.
b s a a m w w wj/'
.,k k NY h1N-h hh-(.
w
- s ~~.
e a
e O
J
~
m 4
ll
-1 w
.Jg
.)
b.
i 9
CROUNDWATER FLOW TO WELLS
...;g '
!i 0
i l
l no a'
jI O
h l
l 0.9 e
3' e
l oi w
ii o.7 I
t o6 y,-
l hO' t.
i i
04 i
.a i
i I,
I 1
{
o.3 l
s-it i
3 *- !
l o:
f.- W i l
'lf l
'l o.s l
o in in niin sixu y,
Time < eci
.-[
4-FIGURE 8.10. Field-data plot oiH/H as a function of time for a slug-test analysis.
.N o
v f.
- '{
E
'E i
c' o
A q
s s
Oo FLOW IN A SEMICONFINED AQU!FER g:
{l
[.
Most confined aquifcrs are not toa!!y isolated from sources of vertical recharge.
4 k**,,
Semipervious layers, either above or below :he aquifer, can leak water into the 1
'l aquifer if the direction of the hydraulic gradient is favorable (Figure 8.11).
.{
.y._
' 4j
- f,
- _ _ _ _. w.,,cr ic.ble - - -
?
f W
i th.
unscre. r ed aquirt, y,jg.
.i. ;
{ l ',
tedkV LOnlining IJyt4 h'
l( *,$ *,
s is M-y p,,
b K.S
.T t
)
Aqu:fer I
u'gc zu,nar -
t a
'u f!GURE 3.11. Fully penetrating wc!! in an aquifer merlain by a semipermeab!c con-g, fining laver.
at-t.
- 1. 5 3k l.
e
< i:
,l 274
!h
,