Municipal Well Data

Where available, I used 2020 numbers
Where multiple samples were taken, I used the TA 'mod 537' results.
For wells without 2020 data, results came from the 2019 PFAS monitoring report:

well

analyte:

PFBA

PFBS

PFPeA

PFPeS

PFHxA

PFHxS

PFHpA

PFHpS

PFOA

PFOS

FOSA

SumAll

Sum-PFBA

6


1.40

1.10

0.77

0.74

0.93

4.20

0.29

0.00

0.82

0.47

1.60

12.32

9.32

7


0.60

0.00

0.00

0.00

0.00

0.75

0.00

0.00

1.00

0.00

2.40

4.75

1.75

8


1.10

0.00

0.67

0.00

0.75

0.93

0.31

0.00

1.10

1.50

2.00

8.36

5.26

9


37.00

0.92

1.00

0.26

0.82

1.40

0.35

0.00

1.20

0.68

3.10

46.73

6.63

11


4.10

0.48

0.73

0.00

0.53

1.70

0.26

0.00

1.00

0.75

1.20

10.75

5.45

13


1.80

1.10

1.60

0.00

1.90

2.60

0.52

0.00

1.40

0.54

2.40

13.86

9.66

14


3.90

1.70

2.00

0.41

2.20

5.00

0.70

0.00

1.80

0.76

2.00

20.47

14.57

15


2.94

3.22

5.75

2.92

6.15

20.06

2.43

0.02

5.68

5.84

0.00

55.02

52.08

16


1.60

0.85

1.20

0.00

1.10

2.90

0.50

0.00

1.60

1.80

1.70

13.25

9.95

17


0.85

0.00

0.00

0.00

0.00

0.77

0.00

0.00

1.00

0.71

2.50

5.83

2.48

18


1.00

0.00

0.49

0.00

0.00

0.46

0.00

0.00

0.80

0.53

2.60

5.88

2.28

19


0.70

0.00

0.00

0.00

0.00

0.30

0.00

0.00

0.00

0.00

2.10

3.1

0.3

20


0.48

0.00

0.00

0.00

0.00

0.25

0.00

0.00

0.00

0.00

1.80

2.53

0.25

23


16.96

1.18

3.07

0.64

2.99

7.26

1.18

0.00

2.37

4.16

2.47

42.27

22.84

24


0.64

0.00

0.00

0.00

0.00

0.28

0.00

0.00

0.00

0.52

2.90

4.34

0.8

25


0.45

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.82

0.62

3.20

5.09

1.44

26


0.90

0.25

0.42

0.00

0.00

0.92

0.00

0.00

0.79

0.99

1.80

6.07

3.37

27


1.20

0.62

0.93

0.00

0.87

1.80

0.30

0.00

1.20

0.55

2.20

9.67

6.27

28


0.71

0.00

0.00

0.00

0.00

0.26

0.00

0.00

0.00

0.00

2.60

3.57

0.26

29


1.20

0.00

0.00

0.00

0.00

0.43

0.00

0.00

0.78

0.00

3.60

6.01

1.21

30


0.61

0.00

0.00

0.00

0.00

0.30

0.00

0.00

0.80

0.51

3.70

5.92

1.61

31


0.41

0.00

0.00

0.00

0.00

0.26

0.00

0.00

0.00

0.00

4.40

5.07

0.26


For reference see the well map.

PFAS testing map

Well Comparison

The PFBA results for the three most contaminated wells (highlighted in red) varied by an order of magnitude, making PFBA an outlier. I decided to omit it from the analysis for the time being (PFBA is discussed elsewhere), to prevent it from 'wagging the dog' in my comparison of well results.
The numbers were normalized so as to reduce the effect of distance from the source. I summed the numbers for the non-PFBA analytes for each well, then used the sum as a divisor. The results were plotted using both net and pie chart types.
Note the well order has been changed to group them according to which analytes were detected. Well numbers are color coded to match the water utility's contamination levels as depicted in the PFAS testing well map (ref above).
The three most contaminated wells are at the top, followed by wells 6, 14, 13 & 27 (sorted by increasing PFOS) for which PFHxA still accounts for 10-20%. Wells 18 & 25 have no PFHxA or PFHpA. The remaining wells also have little or no PFPeA or PFBS.

well (rank*)

PFBS

PFPeA

PFPeS

PFHxA

PFHxS

PFHpA

PFHpS

PFOA

PFOS

Net

Pie

9

(7)

0.139

0.151

0.039

0.124

0.211

0.053

0.000

0.181

0.103

15

(1)

0.062

0.110

0.056

0.118

0.385

0.047

0.000

0.109

0.112

23

(2)

0.052

0.134

0.028

0.131

0.318

0.052

0.000

0.104

0.182













well

PFBS

PFPeA

PFPeS

PFHxA

PFHxS

PFHpA

PFHpS

PFOA

PFOS



6

(6)

0.118

0.083

0.079

0.100

0.451

0.031

0.000

0.088

0.050

14

(3)

0.117

0.137

0.028

0.151

0.343

0.048

0.000

0.124

0.052

13

(5)

0.114

0.166

0.000

0.197

0.269

0.054

0.000

0.145

0.056

27

(8)

0.099

0.148

0.000

0.139

0.287

0.048

0.000

0.191

0.088

11

(9)

0.088

0.134

0.000

0.097

0.312

0.048

0.000

0.183

0.138

16

(4)

0.085

0.121

0.000

0.111

0.291

0.050

0.000

0.161

0.181

8

(10)

0.000

0.127

0.000

0.143

0.177

0.059

0.000

0.209

0.285





































well

PFBS

PFPeA

PFPeS

PFHxA

PFHxS

PFHpA

PFHpS

PFOA

PFOS



26

(11)

0.074

0.125

0.000

0.000

0.273

0.000

0.000

0.234

0.294

18

(13)

0.000

0.215

0.000

0.000

0.202

0.000

0.000

0.351

0.232













well

PFBS

PFPeA

PFPeS

PFHxA

PFHxS

PFHpA

PFHpS

PFOA

PFOS



7

(14)

0.000

0.000

0.000

0.000

0.429

0.000

0.000

0.571

0.000

17

(12)

0.000

0.000

0.000

0.000

0.310

0.000

0.000

0.403

0.286

24

(18)

0.000

0.000

0.000

0.000

0.350

0.000

0.000

0.000

0.650

25

(16)

0.000

0.000

0.000

0.000

0.000

0.000

0.000

0.569

0.431

29

(17)

0.000

0.000

0.000

0.000

0.355

0.000

0.000

0.645

0.000

30

(15)

0.000

0.000

0.000

0.000

0.186

0.000

0.000

0.497

0.317

*ranked by sum of analytes listed at top of table (PFBA excluded as noted above)





Discussion

When looking at the charts I found it helpful to compare them with a similar site investigation at Wurthsmith AFB in Michigan.

Wurtsmith plume
          composition map
Within the most contaminated area:
It is worth noting that the wide variation in PFOS levels is mostly due to its strong adsorption to soil and rock, which retards its movement within the groundwater. The level you measure depends strongly on your position within the plume.
For Madison wells, except for well #8, the well-to-well variations fall well within what would be expected based on the Wurthsmith AFB example until you get to well 17, where the sum of values for the analytes listed in the above table drops below 3 ppt. (See far right column of the table below with wells sorted in order of increasing sum.) At that point, levels are low enough that you would expect the less prominent analytes to start dropping below the limit of detection and thus 'falling out' of the chart.
















Another example using charts to illustrate composition of groundwater is the ITRC case studies: Figure 15-2. Radial diagrams illustrating PFSA trends at an AFFF release site.
PFHxS was generally the biggest constituent, followed by PFOS and PFBS. Note that this is groundwater in and very near the source.
Fig 15.1 shows levels in soil. They are not expected to be comparable directly to groundwater, but are a useful illustration of how levels can be expected to vary within a site.
A broader but less specific comparison is available from the EWG top 100 most contaminated AFFF site list.

Composition data alone cannot pinpoint the geographic source of the PFAS in Madison wells. However, it appears to be consistent with AFFF. Curiously, this seems to apply not only to wells on the east side, but also to five wells which are west of the isthmus: 14; 16; 6; 27; 26. Well 13 is also in an odd location, upstream of Truax relative to the prevailing groundwater flow.

One non-point source consistent with AFFF is the Madison Fire Department. The city website has a searchable database of fire calls going back to 4/23/2015. By selecting incident type ‘vehicle’ and expanding the range to 10 miles I got 34 pages of results @ 10 incidents per page. 340 incidents / 6 yrs = 56 vehicle incidents/yr. In four randomly selected result pages there were four incident summaries in which foam is mentioned, or ten percent. So we could roughly estimate 5 to 6 foam applications per year in the city. This is hardly a mass balance but it makes sense from a qualitative standpoint. The potential influence of surface water transport on the west side is discussed here.

In the 12/21 site investigation workplan, section 10.9.4.3 City of Madison Municipal Wells shows a list of Madison wells near Truax. The second (in terms of PFOS + PFOA) and third (in terms of total PFAS) most contaminated wells are missing: 9 & 23. (See the well map near the top of this document.) This may have been done due to distance from the source. However, PFAS heavily contaminated Starkweather Creek, which carried it into Lake Monona. Surface transport is discussed in detail here.

Local efforts to use groundwater models to identify sources also underestimate the extent of PFAS plumes.

As the developer of the TOP assay has pointed out, "Despite broad background PFAS contamination, there are a finite number of PFAS point sources."

Given the depth of these municipal wells, the level of contamination and what is known about expected plume extents, it seems likely that the Truax area is at least partly responsible for at least the three most contaminated wells (in terms of total PFAS).