Hydraulic Conductivity of MinnesotaMinnesota
Confined Glacial Aquifers
R G SouleR. G. SouleMinnesota Department of Health
R. J. BarnesUniversity of Minnesota
OverviewOverview• Well by well comparison of pumping test and
ifi it t t ti t f h d lispecific capacity test estimates of hydraulic conductivity (K).
• Explore the additional information afforded by using specific capacity test K values.
• Consider the limitations of these tests.
Pumping and Specific Capacity Tests“Gold Standard” N = 239 N = 70,773
Pumping and Specific Capacity Test Pairs
• Must have the same unique well number*• Must have the same unique well number*,
• Must have verified location*,
• Must be sufficient well construction and geologic information*,
• Must be analyzed in a similar fashion.
*County Well Index (CWI) Minnesota Geological Survey
Estimating Specific Capacity Test Transmissivity
Required Well Information
• Screened interval (L)Soil
r w
Screened interval (L)
• Radius (rw)
A if hi k (b)
Sand
Sandy Clay • Aquifer thickness (b)Sandy Clay
Clay
Sand &L Sand & Gravel
L b
Clay
Aquifer Thickness Problem
Soil• Many wells do not have l fi i l
Sand
Sandy Clay
lower confining layer information.
Sandy Clay
Clay
• Option 1: Use the observed “minimum” thickness.
Sand and Gravel
• Option 2: Estimate from the known thicknesses.
“b” or minimum estimate of “b”
Clay
Estimating Aquifer Thickness
400
Frequency of Known Aquifer ThicknessTotal = 8,000
300
350
200
250
eque
ncy
100
150
Fre
Average of
0
50
0 50 100 150 200 250 300
Average of Thickness > 20 ft. = 44 ft.
Minimum0 50 100 150 200 250 300
Aquifer Thickness [ft.]
MinimumThickness 20 ft.
Tests Requirements
Pumping TestsS F d l (USGS) St t (MDH DNR &– Sources: Federal (USGS), State (MDH, DNR & MPCA), consultants, well drillers and publications.
– Analysts chosen Transmissivity
Specific Capacity
Analysts chosen Transmissivity.– Cooper‐Jacob or Theis methods.
Specific Capacity
– Source: CWI
Pumping rate (Q) N=100 Pairs– Pumping rate (Q)
– Duration (t)Drawdown (s)
N=100 Pairs
– Drawdown (s)
The Comparison SetAVERAGE WELL INFORMATION
YEAR BUILT 1985.4CASING DIAMETER [INCH] 11.9
WELL USE PERCENTPUBLIC WATER SUPPLY 60%
DEPTH [FT.] 174.5SCREEN LENGTH [FT.] 29.3KNOWN AQUIFER THICKNESS [FT.] (N=56) 50.5MINIMUM AQUIFER THICKNESS [FT.] (N=44) 50.0
IRRIGATION & INDUSTRIAL 35%PUMP OUT 3%TEST & MONITORING 2%
AVERAGE TEST INFORMATION Pumping Test Specific CapacityDURATION [HOURS] 24 16DISCHARGE [GPM] 605 664TESTS WITH OBSERVATION WELLS [% ] 76% 0%TESTS WITH OBSERVATION WELLS [% ] 76% 0%ESTIMATED STORATIVITY 4.5E‐03 1.5E‐03
1600
Specific Capacity and Pumping Test ComparisonKnown Aquifer Thickness
1400
1600
Specific Capacity K (K sc)
Pumping Test K (K pt)
1000
1200
Test
R2 = 0.50
= 0.97
800
rom Pum
ping
T R 0.50
31 Overestimates 25 Underestimates
400
600
K fr
0
200
0 200 400 600 800 1000 1200 1400 1600
K from Specific Capacity Test [ft. /day]
1600
Specific Capacity and Pumping Test ComparisonMinimum Aquifer Thickness
1400
1600
Specific Capacity K (K sc)
Pumping Test K (K pt)
1000
1200
Test
R2 = 0.48= 0.72
800
rom Pum
ping
19 Overestimates 25 Underestimates
400
600
K fr
0
200
0 200 400 600 800 1000 1200 1400 1600
K from Specific Capacity Test [ft./day]
1600
Specific Capacity and Pumping Test ComparisonEstimated Aquifer Thickness
1400
1600
Specific Capacity K (K sc)
Pumping Test K (K pt)
1000
1200
Test R2 = 0.49
= 1.03
800
rom Pum
ping
27 Overestimates 21 Underestimates
400
600
K fr
0
200
0 200 400 600 800 1000 1200 1400 1600
K from Specific Capacity Test [ft./day]
1600
Specific Capacity and Pumping Test ComparisonKnown H Estimated H
1400Specific Capacity K (K sc)
Pumping Test K (K pt)
1000
1200
Test R2 = 0.50
= 0.99
800
rom
Pumping
58 Overestimates 46 Underestimates
400
600K fr
0
200
00 200 400 600 800 1000 1200 1400 1600
K from Specific Capacity Test [ft./day]
Additional Information from Specific Capacity Tests
• Apply the same methods to estimate K for theApply the same methods to estimate K for the remaining 70,673 specific capacity tests.
• Compare the local specific capacity tests to• Compare the local specific capacity tests to the pumping tests.
E l i l l i f K l• Evaluate spatial correlation of K values.
100%All Specific Capacity Tests Pumping Tests
Frequency Distributions of K
• Both test types have lognormal frequency distribution.80%
90%
100%
distribution.
• Not the same populations.
60%
70%
80%
cy of K
[%]
40%
50%
60%
ve Frequ
enc
TestsMedian K [ft./day]
100 Pumping Test 13270 000 S ifi
• Heuristic for 50% 20%
30%
40%
Cumulativ 70,000 Specific
Capacity Tests27
range:– Lower = mean/2
– Upper = mean * 2.50%
10%
20%
0%1 10 100 1000
K [ft./day]
Spatial Correlation of Pumping and Specific Capacity KS ifi C it T t N li b t
1500 m
Specific Capacity Tests Normalize between pumping test wells:
Specific Capacity K (K sc)
500 m
1000 m
500 m
1000 m
Excluding the Ksc of the Pumping Test Well
p p y ( sc)
Pumping Test K (K pt)
500 m500 m
PumpingFrom (ft.) To (ft.)
MeanKsp/Kpt
Pumping Test Well
Test0 500
500 1000
1000 1500
116%92%
49%1000 1500……. ……. …….
Spatial Correlation of Pumping and Specific Capacity TestsSpecific Capacity K / Pumping Test K 95% Confidence Boundary
100%
120%
Test K
p p y / p g y
Ksp > Kpt within 500 m. of pumping test wells
80%
100%
/ Pu
mping
Te
Ksp / Kpt ≈ 90% between500 and 1000 m.
60%
Capa
city K /
Beyond 1000 m
20%
40%
age Specific Beyond 1000 m
Ksp / Kp ≈ 35 to 45 %.
0%
20%
0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000
Avera
0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000
Distance from Pumping Test Well [m]
Spatial Correlation of Specific Capacity TestsSource of Variability Amounty
Noise 41%
Small Scale Spatial Variability 20%
Large Scale Spatial Variability 39%
Confined Glacial AquifersPermeability [ft./day]
Conclusions for Confined Glacial Aquifers• A pumping test is more valuable than a specific capacity test.capacity test.
• Specific capacity K values are consistent with pumping test K values.p p g
• Specific capacity K values are “noisy”.
• Hydraulic conductivity is spatially related.Hydraulic conductivity is spatially related.– Change exponentially over a few kilometers.
– Change linearly over hundreds of kilometers.
• Pumping tests in this data set are located in higher hydraulic conductivity zones.