The 2001 crash plume consists mainly of PFHxS and PFOS (charts in table row 1 below) whereas the area southwest of Del Ray Avenue has significant amounts of these compounts but also has a large amount of PFOA (charts in row 4 below). The source of the additional PFOA is apparently the terminal apron (charts in row 5).
![]()
![]()
![]()
![]()
![]()
![]()
![]()
![]()
![]()
![]()
The main plume from the 2001 crash has a parallel ‘shadow’ plume just to its southwest, as depicted in 20210407_43_Interim_SIR.pdf PDF file p. 79 Fig 27 (below). Its composition is depicted in the charts in table row 2 above. Like the composition in the Del Ray Ave area, it is a mix of the contributions from the 2001 crash site plus the terminal apron, with a greater emphasis on the crash site than that seen in the Del Ray Ave area.
![]()
The area near the 1970 crash site also appears to receive a significant contribution from the high-PFOA terminal apron (see chart table row 3 above).
The composition of the terminal apron PFAS deserves special comment. According to ITRC “Legacy fluorotelomer AFFF were manufactured and sold in the United States from the 1970s until 2016 and encompass all other brands of AFFF besides 3M Light Water or their licensed products (Schultz et al. 2006[57]). Although they are not made with PFOA, they contain polyfluorinated precursors (Backe, Day, and Field 2013[12]) (Place and Field 2012[312]) that are known to degrade to PFCAs, including PFOA (Weiner et al. 2013[882]) (Harding-Marjanovic et al. 2015[304]).”
The terminal apron results also show significant levels of 6:2 FTS which suggests the additional presence of newer C6 fluorotelomers which also degrade to PFCAs, including PFBA but not PFOA. The terminal apron is the likely source zone for the PFBA which appeared at significant levels in the Del Ray Ave area, to a lesser degree in the main residential plume from the 2001 crash, and at intermediate levels in the ‘shadow’ plume to its southwest. PFBA acts as a tracer and at significant distances from the source zone tends to overwhelm the other analytes, so I followed common practice by removing it from the charts in order to make them easier to interpret.
6:2 FTS is mobile in the groundwater. It’s presence in the source zone but not elsewhere could be partially due to the extra exposure to oxygen in standing water that accumulates in the western infiltration area during heavy runoff events.
I grouped the private well data by using the maps from 20210804_43_PW_Sample_Results.pdf Fig 3c (Del Ray Ave), Fig 3d and Fig 3b.
2001 crash main plume extending to Dawson Ave interchange includes addresses on :
Bainbridge St samples 338, 341, 342 (also 343 on 1st Ave West), 110-2, 110-1, 111, 229, 230
1st Ave East; 251, 252, 255, 256, 197, 200, 202
2nd Ave East 214, 215, 218, 219, 220, 186, 188, 189
2001 crash SW ‘shadow’ plume has addresses on:
1st Ave West: samples 384, 385, 386, 387, 388, 389, 328, 332, 335
Bainbridge: 329, 330
Del Ray Ave has addresses on:
Del Ray (east side): samples 84, 88, 89, 91, 94, 95, 96
Del Ray (west side): 79, 1049, 1050, 1051
Grand Ave (east side): 83
Grand Ave (west side): 47, 49, 72, 1047, 1048, 1193
11-1, 27, 30, 412, 1045
samples 72-0 2713 Grand Street and 255-0 2545 1st Ave E have multiple problems so they were dropped from the analysis
Lake Shore Drive near 1970 crash site has samples:
1004, 1016, 1024, 1027, 1425, 1433, 1435, 1437, 1438, 2317
Most of the information regarding water movement in the airport is in a stormwater management memo.
20210407_43_Interim_SIR.pdf PDF p. 119 Attachment D Airport Stormwater Drainage Memo 3/11/2021
PDF file p. 122 contains Fig 2: Airport Overview - South Airport
![]()
As mentioned in the summary there are two areas in the southern part of the airport where runoff infiltrates:
(1) West infiltration area: serving the area west of the main runway and clearly visible as a dark oval between the Del Ray Avenue area to the west and the junction of the main north/south and NE/SW runways to the east.
(2) East infiltration area: serving the area east of the main runway near the intersection of Fanta Reed Rd and Dawson Ave, indicated by the small red arrow pointing NNE from Fanta Reed Road.
The west infiltration area receives runoff from the terminal apron which mostly drains to the grassy area along its eastern edge. Some of the runoff passes through culverts under the main taxiways serving the apron. It apparently also receives runoff from the storm sewer under Airport Road. Groundwater mounding would tend to push some of the contamination westward toward Del Ray Ave.
There is a third infiltration area at the north end of the airport served by the same Airport Road storm sewer seen in Fig 2 above.
![]()
It is described by the interim site investigation report:
14.2.1.2 Terminal, Parking & Airport Drive
Stormwater runoff on Airport Drive, the terminal and associated parking area either infiltrates to the groundwater or is carried by storm sewer to a large stormwater retention and settling pond, located at the north end of the airport, adjacent to the bay formed by the extension of Runway 18 to Bell Island. The pond, in turn, discharges at an outfall into the bay. Lenz noted, that during high flow conditions a portion of the stormwater will discharge via a storm sewer pipe, just south of the terminal, into the internally drained airfield.
Although Airport Drive, the terminal and associated parking areas are not considered potential source areas, should stormwater in these areas contain PFAS, it would infiltrate to the groundwater or be carried by storm sewer north to the stormwater retention pond and then to the bay formed by the extension of Runway 18 to Bell Island. In the event of leakage from the storm sewer, PFAS contamination would reach the groundwater flow which is flowing southeasterly in that location.
I have not seen evidence of any attempt to sample either the west infiltration area which receives runoff directly from the terminal apron or the northern stormwater retention pond which receives water from an area at the southwest corner of the terminal apron. This is despite their being mentioned as potential source areas in the interim site investigation report.
9.2.7 Areas of Stormwater Infiltration
Areas of stormwater infiltration for the internally drained portions of the airport, particularly adjacent to areas of frequent or heavy airport and tenant operations, are considered potential source areas. See the attached March 16, 2021 memorandum regarding Airport Stormwater Drainage21 from Bernard Lenz, P.E. – La Crosse Utility Manager.Indeed, the interim report only attempts to explain the spread of high-PFOA contamination by referring to background levels in the adjoining river:
Based on concentrations observed in up-gradient wells along Lake Onalaska, it is believed that at least a portion of the PFAS detections can be attributed to “background” levels from Lake Onalaska/Mississippi River in groundwater recharge of the sand-and-gravel aquifer.
Potential discharge of relatively high-level PFAS contamination from the storm sewer is especially concerning given that its outfall goes directly to a major groundwater recharge area for the island, seen as a red band along the northern portion of the island in Fig 11b from the interim site investigation report.
The east infiltration area receives runoff from a storm sewer that runs lengthwise under the large SW/NE apron in front of the fire station. Zooming in on Google Maps satellite view shows dark circles along this line, presumably sewer grates. It apparently dumps out to a culvert opening visible in the mottled area where the apron bends to north/south. In the drainage map above see the tip of the large red arrow showing the direction of flow through the apron storm sewer.
Surface flow proceeds south entering a culvert under the taxiway which serves the south end of the main runway. Outflow from this culvert (along with that from another serving the grassy area adjoining the runway intersection) enters a narrow dark strip leading SSW to a low area stretching NNW/SSE toward Fanta Reed Road just west of its intersection with Fanta Reed Pl. In the drainage map above the small red arrow shows the direction of flow through a culvert under the road which empties into the low area.
Several culverts under this part of Fanta Reed road appear to align with residential streets. It is not immediately clear whether this infiltration area is also receiving runoff from the pavement in the residential develpment south of the road. However, the USGS topographic viewer indicates a drop of 5 meters as you proceed NE across this development toward the infiltration area.
Mounding in this area would tend to push contamination from the 2001 crash toward the WNW. This would explain the sulfonates in the Del Ray Ave area as well as the composition of the ‘shadow’ plume SW of the main plume from the 2001 crash site.
Lee Donohue provided photos taken 5/5/2023 of water standing in this area. She commented: “This water has been there since last week although we've barely had rain in a week.” It was still there in subsequent photos taken on 5/18.
I suspect that elevation of the water table due to the spring flood – the highest since 2001 – has effectively reduced the gradient of the groundwater mound and slowed lateral movement, causing it to take longer than usual for the surface water to recede. According to the Minnesota Stormwater Manual, a shallow water table is a risk factor for mounding:
- Thin vadose zone. Mounding is more likely to occur in areas with a shallow groundwater table (thin vadose zone).
![]()
There is strong evidence in the PFAS sampling results that transport is much more complex than the groundwater model suggests. The sampling results map shows that in the Del Ray Avenue area contamination levels are highest near the airport boundary and decline with distance to the west.
In their 10/7/2020 status update (20201007_43_Status_Update_SIR.pdf) the consultants said:
Well 26 was sampled for PFAs compounds under the UCMR in 2014, and none were detected. Given the strong and consistent southeasterly groundwater gradient across the site, OSG does not consider Well 26 as a potential receptor.This well was sampled the following March and tested positive.
Their reasoning relies on a USGS/ WGNHS groundwater model created for two overlapping areas that combine to cover approximately 30 miles by 30 miles. The model developers have this to say about the model limitations:
As is the case with all ground-water-flow models, the LLC and Pool 8 models are a simplification of the “real world” ground-water system, and have corresponding limitations in model precision and how the model can be used. For example, the MODFLOW model discretization (node) is 500 feet by 500 feet. As a result of this discretization, the conditions within the node (ground-water level, ground-water flow) are reduced to one average value for the entire node. Therefore, even though the resolution of the model grid is relatively high, the model is not suitable for analysis of site-specific problems or issues....UW Geoscience Professor Michael Cardiff suggested that he believes the groundwater model used by the consultants does not account for recharge on French Island and that water movement may be influenced by private wells west of Airport Road and by water table mounding (footnote below).
Because of these inherent simplifying assumptions, the ground-water-flow model cannot simulate small-scale complexities of the ground-water-flow system.
Such highly dynamic details are especially difficult to analyze given that, according to the authors, "As currently implemented, the models simulate a steady-state ground-water system; that is, ground-water levels are not changing with time. "
USGS/WGNHS scientists were also involved in development of the Dane County Groundwater Model, where they felt the need to extend the steady state model so it could simulate non steady state conditions.
Water table mounding - especially in the presence of spring flood events - would call for a transient analysis.Transient models are needed to simulate field data where head or flow changes with time. Simulating transient groundwater flow was an important objective of the modeling project as many societal questions cannot be adequately addressed using a steady-state model.
While the consultants included a storm sewer management map, they chose to ignore obvious surface transport routes which lead to stormwater retention/infiltration basins. The consultants mention these as potential sources but I have seen no evidence that they were sampled and analyzed. Groundwater mounding in these areas would result in further distribution of any contaminants, potentially in directions contrary to the prevailing steady state groundwater flow. Furthermore, the appearance of multiple parallel plumes in the residential wells southeast of the 2001 crash site suggests the involvment of episodic/transient events.
Via email Feb 2, 2023, 11:32 AM:
“my understanding is that this model does not include any recharge on the island (precipitation, which will tend to “mound” groundwater and push it away from the center of the island). Similarly, depending on the number and strength of groundwater pumping by residents, that may also impact the flowfield relative to what’s presented here, and cause movement more to west.”
Michael Cardiff (he/him/his) Associate Professor
Department of Geoscience University of Wisconsin-Madison
Site Investigation Workplan (SIWP) Received (fee) 20191101_135_SIWP.pdf
The volatile organic compound (VOC) release for which the City of La Crosse was responsible was closed by WDNR on May 5, 2010. Based on the information that has been submitted to WDNR regarding this site, we believe that this newly reported PFAS contamination is related to firefighting foam that was used at the same fire training burn pits which were the source of VOC contamination in municipal wells 23 and 24.
In the December 2003 Request for Site Closure3, the City’s then environmental consultant for the site, RMT, described the history of the site thus:
The La Crosse Municipal Airport is located in. the northern portion of French Island in La Crosse, Wisconsin (Figure 1). Two of the City's production wells (23H and 24H), located on the eastern side of French Island, were shut down as of November 1994, except for emergency back-up use, because of the presence of chlorinated volatile organic compounds (VOCs) and xylenes in samples from the wells. The source of the contamination was identified as the two former test burn pits that were located approximately 3,000 feet northwest of the municipal wells.In May 1994, RMT, Inc. (RMT), was retained by the City to conduct the activities necessary to obtain the WDNR' s approval of a closure plan for the hazardous waste units. RMT performed additional investigations during the summer of 1994 and presented the Additional Investigation Activities, Treatability Studies, and Remedial Options Analysis (1994).