Donate

Goodman Community Center

Kipp PCBs continue to pollute area along Isthmus bike path, even after multiple remediations

Kipp PCBs continue to pollute area along Isthmus bike path, even after multiple remediations
 madison.com

The company expects to find a solution soon, but critics say government regulators have failed to properly investigate the troubled industrial site. More…

(A diagram of PCB levels found along Kipp’s storm drainage pipe is below. See the full PCB report by Kipp’s consultants here.)

 

High PCB levels

Digg thisShare on FacebookShare on Google+Email this to someonePrint this pageShare on RedditShare on TumblrTweet about this on TwitterPin on Pinterest

PCB Excavation Along Bike Path next to Madison Kipp–Enjoy Photos!

PCB Excavation Along Bike Path next to Madison Kipp–Enjoy Photos!

A previous post shared details about the excavation of soils contaminated with high levels of PCBs, from October 6-9, 2015, along the bike path between Madison Kipp Corporation and the Goodman Center.

Here’s what the excavation looked like…enjoy! (sorry about the strange sizes, we had some formatting problems…)

Excavation begins, October 6, on the east end of the raingarden…

Earth First Advanced Waste drops off a PCB dumpster across from the splash pad…(Yes, Earth First came to Kipp!!) 

The bike path is a busy place! Neighbors sit at the splash pad bench and enjoy the show…

Children pick vegetables near the excavation (red PCB dumpster behind them)…

A neighborhood mom whose children are playing at Goodman tells Earth First Advanced Waste they are “killing the earth” and does a ritual to “take the land back” and “release the bad energy”

A dust monitor is in place (good step in the right direction….but is this monitor to assess dust exposures to the excavation workers? Or people on the bike path? Either way, it seems to be the wrong kind of monitor and in the wrong place….).

Dust barriers put up on 2nd day of excavation (Good, finally!! But won’t some fine dust go up and over them? )

A “Best Waste Solutions” truck behind the excavation spews smoke (oops!)…

Are there really no PCBs on the right side of this fence, next to the bike path?? This area has not been tested. Hmmm…..

Yellow “caution” tape was put up around dust barriers on October 8

The dust monitor concentration on the outside of the excavation dust fence read 19 µg/m3 at 10 am on October 8 (time-weighted average of 12 µg/m3)…

Soils with the highest PCB levels, along about an 80 foot stretch across from splash pad, were excavated right to the edge of the bike path, including the jogging path…

A couple hotspots in the raingarden, excavated on October 6, were re-seeded and covered with hay…

A PCB hotspot in the grassy area along the far eastern end of the bike path was “capped” with a few inches of clean soil (and later fenced off)…

 

Digg thisShare on FacebookShare on Google+Email this to someonePrint this pageShare on RedditShare on TumblrTweet about this on TwitterPin on Pinterest

More PCBs on city bike path to be excavated October 6-9; City of Madison says no warning signs needed

More PCBs on city bike path to be excavated October 6-9; City of Madison says no warning signs needed

In June 2015, PCBs (polychlorinated biphenyls) of up to 680 ppm—orders of magnitude above the residential (0.2 ppm) and industrial (0.7 ppm) direct contact standards—were found in soils next to the highly used Capital City bike path and just across from the Goodman Community Center’s new splash pad. The source of the PCBs, which likely have been in soils along the bike path for decades, is Madison Kipp Corporation. Test results are depicted on this map; the full report is here.[1]

These high PCB levels were found much closer to the city bike path and adjacent jogging/walking path than those found in previous sampling in the city raingarden that began in spring 2014 (which uncovered PCB levels up to 550 and 1020 ppm–see here and here)[2] Soils along the bike path also contain other toxic contaminants, including heavy metals, PCE (tetrachloroethylene), PAHs (polycyclic aromatic hydrocarbons) and more.

A large volume of scientific studies shows that long term PCB exposures are associated with neurological, endocrine, immune system, and a variety other health problems. In February 2013, the International Agency for Research on Cancer (IARC) classified PCBs as carcinogenic to humans (Group 1)[IARC, 2013]. More information about the health effects of PCBs is here and here. Also, PCBs are usually contaminated with dioxins and furans, which are even more toxic than PCBs.

The June PCB findings were not shared with the public and, as of Sept. 22, 2015, were not posted on the DNR website.[3] No signs have ever been posted along the bike path about this significant contamination.

Neighborhood residents, frustrated by the refusal of city and state officials to post warning signs along the bike path, posted their own signs in mid-September 2015 (see photo below). The signs were taken down a day after they were posted. Who took them down? Madison Kipp? The city?

These highly contaminated soils will be excavated October 6-9, 2015. Soils with PCBs at such high levels would in most cases be excavated immediately, but city and state officials decided not to excavate while the splash pad was open, because they know excavation will disrupt contaminated soils, releasing PCBs into surrounding air and onto nearby soils and pavement—exposing toddlers and children playing at the splash pad just feet away, with mothers (often holding babies) looking on.[4]

Even though they decided to leave the soils with high PCB levels in place all summer, neither DNR nor City of Madison officials felt that signs alerting people to the high levels of PCBs along the bike path were warranted during this time. They fenced part of the PCB contaminated area off, but left the jogging path open.[5]

City staff said they do not plan to post any signs before or during the upcoming excavation in October, assuring us they will make sure the area is “secure.” How will the area be “secured” to prevent the release of PCB contaminated dusts into surrounding air, soils, and bike path during the three-day (or more) excavation?[6]

The photograph below shows the dust clouds released when soils were excavated for the construction of the Goodman splash pad last summer/fall; soils under the splash pad are known to have high levels of PAHs (polycyclic aromatic hydrocarbons), heavy metals, and numerous other toxic contaminants (likely including PCBs, though they haven’t been tested)–see here. Children were playing beneath this dust cloud when this photo was taken.  Nothing was done to by city or state officials, or Goodman Center leaders, to prevent this.

Though the city plans to communicate about the upcoming excavation with the select group of people on the neighborhood association listserve, most parents of children using the splash pad and bike path, and/or whose children play daily at the Goodman Community Center, including many minority, low-income parents from the Darbo Worthington area, are not on this listserve and will continue to be unaware of high PCB levels right next to where their children play. Most people walking, jogging, biking, walking pets, and pushing strollers along the bike path are also not on this listerve, and are therefore in the dark about what they could be collecting on their shoes and bike/stroller wheels—and what their pets might be ingesting and gathering on their fur—and bringing into their homes.

Why are city officials unwilling to post warning signs, though they definitely have the authority to do so on their own property? What is the harm? Signs could be made with almost no cost to the city.

Perhaps city officials do not want the public to be aware of how contaminated this highly-used city property is? Or perhaps the city is protecting Madison Kipp Corporation, the source of the PCBs?

Maybe the City of Madison is worried about its own liabilities? If so, isn’t it sad that city government’s concerns about its liabilities are a higher priority than protecting public health? Doesn’t the city have an ethical responsibility to let people know about high levels of toxic contaminants on public property, so they can choose to avoid contaminated areas if they want to?

Apparently the City of Madison does not think so.

[1] Full lab reports here.

[2] The raingarden PCB saga goes back many years. See here and here for more about the raingarden PCB history.

[3] In November 2014, a DNR website post said “clean-up activities for the rain garden along the bike path have been completed” and “Soil sampling confirmed that no further action was necessary” on city property. Obviously, this is not the case, but the DNR has chosen not to post any further updates for the public (as of Sept. 22 2015).

[4] This begs the question: Why didn’t they test and excavate these PCBs before opening the splash pad?

[5] DNR officials said that even though PCBs were found just 4 inches down right next to the jogging path, there is no chance of exposures to those on the path. However, surface soils on the jogging path were not tested at all so this is just speculation.

[6] Why don’t they block off the bike path during the excavation?

 

Digg thisShare on FacebookShare on Google+Email this to someonePrint this pageShare on RedditShare on TumblrTweet about this on TwitterPin on Pinterest

Madison-Kipp Clean Air Act violations: MEJO response to Kipp CEO

Madison-Kipp Clean Air Act violations: MEJO response to Kipp CEO

Madison-Kipp Corp President/CEO Tony Koblinski responded to our post regarding the EPA consent decree for his company’s Clean Air Act violations. Here is our response:

Dear Mr. Koblinski: We offer the following responses to your comments in hopes of providing some clarity as to why we are dissatisfied with the resolution of the EPA’s Notice of Violation (NOV) to Madison-Kipp Corp.

Koblinski: “This NOV is not a new matter.  It was issued to us September 12, 2012.”  

MEJO Response: Right—these violations are not new. They go back many years. For more on Kipp’s air pollution issues since 1990, and the community’s struggle to address them, see here and here.

The EPA violations occurred from 2007 (and before) through at least 2013. Kipp’s lawyers negotiated with EPA for 2.5 years after the initial violation was sent to them on Sept. 12, 2012.

In 2008, a DNR air pollution compliance engineer notified Kipp that they were using the wrong emissions factors, underestimating the stack discharges and violating their permit. However, after this engineer passed away and another one was assigned, the DNR continued to allow Kipp to use the wrong emissions factors for years, even after the 2012 NOV from EPA was issued. In 2013, DNR found Kipp in full compliance with air regulations, even while the company was still using the wrong emissions factors. DNR South Central Air Management Supervisor Tom Roushar explained in fall 2013 (about a year after the EPA NOV was issued) that it was acceptable that Kipp continued to use the 2001 emissions factors through 2013 even though their permit specified that they should use the higher 2007 emissions factors. He felt that EPA’s allegations were unsubstantiated (though oddly, he claimed not to have seen the NOV).

Koblinski: “It does not allege that we violated our emissions limits, but rather that we did not have adequate controls on our data recording, emissions factors, record keeping, calibration and plant signage.  These are technical issues which we have taken very seriously, but at no time have we exceeded our permitted emissions limits.”  

MEJO Response: Kipp’s estimated levels of air emissions, a key basis on which agencies assess regulatory compliance, are almost entirely based on the company’s own data recording, records keeping, and/or the emission factors they use. Because Kipp kept bad (or no) records on important processes that affect emission levels, and/or used old, incorrect emissions factors, inaccurate emission estimates were reported to the DNR, EPA, and the public for over five years. These are not just minor “administrative” or “technical” issues.

In 2012, Representative Chris Taylor asked the DNR to require Kipp to measure the actual emissions from its stacks. The DNR responded by saying that the extensive recordkeeping required of Kipp was superior to testing emissions. Shortly thereafter, the EPA issued its notice of violation because Kipp was neither conducting its required recordkeeping nor filing correct reports.

Here’s one example of the consequences of using an incorrect emissions factor. Because Kipp used a 2001 emissions factor for chlorine that was 5.8 times lower than its 2007 permit required it to use, chlorine wasn’t reported at all on several air emissions inventories between 2007 and 2012 even though it should have been; incorrect chlorine emissions factors produced inaccurate chlorine emissions estimates that were lower than DNR reporting thresholds. Stack tests done in 2014 as part of the recent NOV showed that Kipp was emitting almost 12 times more chlorine than the 2001 emissions factor they used from 2007 to 2013 predicted they would emit.

Kipp also used 2001 emissions factors for dioxins through at least 2012, even though a more appropriate stack test in 2007 showed dioxin/furan emission levels orders of magnitude above the levels found in 2001 tests. Dioxins (and closely related compounds, called furans) are highly toxic at extremely minute levels—much, much more toxic than chlorine and even the tetrachoroethylene (PCE) Kipp workers dumped onto the ground for years. Even though Kipp stack tests in 2001, 2003, and 2007 showed that Kipp emitted dioxins/furans (and emissions factors increased with each test), none of Kipp’s air inventories have ever reported dioxin/furan emission levels, because using the old emissions factor kept the estimated emission levels under the DNR’s reporting limit. DNR was either oblivious to Kipp’s method of keeping their emission off the public air inventories, or decided that this was acceptable.

Kipp was also found in violation by EPA for not maintaining accurate records of its “Hazardous Air Pollutant” (HAP) emissions, which include a number of other highly toxic chemicals emitted from the factory.

Again, these record-keeping and reporting issues are not minor technical matters. Given the decades of community concerns about dioxins, chlorine, and other hazardous air pollutant emissions from Kipp, and hundreds of complaints to government agencies about strong chemical odors and health effects in the neighborhood, these reporting issues are very problematic, whether or not emission levels exceeded regulatory limits (which studies show, for many of the contaminants emitted Kipp, are too high to adequately protect public health).

Considering the discovery of extremely high levels of contaminants in soil and groundwater at Kipp, is it any surprise that Kipp did not maintain the records needed to show it complied with air pollution control laws? Pollutant levels reported to the DNR for air inventories are the only way citizens have to know what hazardous pollutants Kipp is emitting and at what levels. Kipp knows that and has worked hard to keep these chemicals off air inventories, in part, by using incorrect emissions factors. According to DNR, in 2007/2008, industries lobbied hard to not be required to report emissions below reporting limits (as they had been before). DNR allowed this.

Violations led to increases in actual emissions

Some of Kipp’s violations likely led to actual higher emissions of hazardous air pollutants, not just incorrect emissions estimates. The lubricant used to make aluminum castings is evaporated and partially burned, then exhausted through the roof. Rather than actually capturing and controlling the die lubricant emissions, diluting the die lubricant is the method required by DNR for reducing emissions from the die casting process. One of the EPA violations against Kipp was not diluting the die lubricants as much as required in its permit, resulting in higher VOC and other toxic emissions from die casting.

Other violations included not recording how much they diluted the die lubricant and/or not calibrating their die lubricant mixing equipment correctly—in other words, not following the emissions control method they are required by the permit to follow. These die casting violations are not minor technical issues. The “waxy/oily/burnt” and “metallic” smells neighbors have complained of for years are primarily die casting emissions. According to EPA, the top eight Toxic Release Inventory (TRI) air releases reported from aluminum die casting industries in the U.S. are: aluminum (fume or dust), trichloroethylene, tetrachloroethylene, zinc (fume or dust), copper, hexachloroethane, glycol ethers, and zinc compounds.

Actual emissions from Kipp’s die casting processes have been measured only sporadically and incompletely over the years inside the plant—and not for many years. In 1994, an OSHA inspector measured “oil mists” and “release agents” from die casting processes made up of metals and numerous organic compounds, including: aliphatic hydrocarbons, aliphatic alcohols, acetic acid, organic acids, butyrated hydroxyl toluene, long chain aromatic compounds, fatty acid methyl esters, propylene glycol, hydroxytriethylamine, methyl styrene, 1-2-2—methoxy-1-methyl-ethoxy-1-methylethoxy-2- propanol, and several “unidentified compounds.” The “condensate of mold release agent” contained “50% gray metallic flakes” made up of lead, aluminum, zinc, copper, and iron as well as “small particles, oily or greasy substances” and “brown particles.”

The 2001 “Madison Kipp Corp Exposure Assessment,” done by the Madison Public Health Department listed the following emissions measured above Kipp’s die casting machines in 1998: 1, 1, 1 trichloroethane, benzene, toluene, 1, 2 butadiene, hexane, ethanol, acetone, and several “unknown alkanes or alkenes.” Other reports and studies indicate that aluminum die casting emissions may also contain other very toxic chlorinated compounds such as chlorinated paraffins, PCBs, and dioxins/furans (as far as we know, these compounds have never been tested for in Kipp’s die casting emissions, inside the plant or from the stacks/vents).

During the 2011 inspection by EPA, the inspector noted “hazy air” in the die casting areas in both the Atwood and Fair Oaks facilities. This “haze” is the “oil mist” or die cast emissions, made up of many of the hazardous air pollutants listed above. Workers inhale this mist. It goes out open doors and windows and into roof vents and stacks uncontrolled. That’s why the odors are so strong in the neighborhood on some days. Again, Kipp has never characterized the chemical components of their die cast stack emissions or reported them to the DNR, nor has DNR required them to do so.

Prior to 1995, the “oily mist” from die casting was primarily vented through the doors and windows into adjacent backyards. At that time, after hundreds of neighborhood complaints, Kipp installed many roof vents sending the die casting fumes and the waxy odor further through the neighborhood. Only in 2007 did DNR recognize that Kipp was violating the 1971 air quality standards and required taller stacks rather than simple roof vents, in order to disperse the die casting emissions higher and further away. Current emissions factors for die casting are based on stack tests done in 1994 (that did not assess chemical components of the emissions).

Die casting emissions deposit on Kipp’s side walls and roofs, are washed off during rain/snow, and go into the storm drain system and eventually to the raingarden and/or Starkweather Creek and then Lake Monona. Kipp doesn’t test the chemicals in its stormwater runoff, so we don’t know what contaminants are going into the raingarden and the already highly impaired Starkweather Creek and Lake Monona.

Koblinski: “Under the direction of the EPA we completed new stack testing in May of 2014 and further verified that we operate within a fraction of our allowable emissions levels.”

MEJO Response: In the May 2014 EPA tests, the Atwood aluminum furnace stack emissions were tested for chlorine, hydrochloric acid, PM (particulate matter) and PM-10 (particulate matter ten microns or less in size). The Fair Oaks stacks were tested only for PM/PM-10. The smaller, more toxic components of particulate matter (PM 2.5) were not assessed at stacks from either facility. Dioxins/furans were not tested. Aluminum salts were not tested. No other hazardous pollutants were tested. Die casting emissions were not tested. Therefore, we have no idea whether Kipp was “within a fraction of allowable emission levels” for small particulates and many of the other hazardous air pollutants they emit.

The allowable emissions levels in Kipp’s expired air pollution permit are no basis to judge if the surrounding neighborhood is safe. The discharge limits were established to meet the old air quality standards. All the stacks on Kipp’s roof were designed to comply with the 1971 air standard for PM. The DNR’s own analysis shows current permit limits are not sufficient to comply with modern air quality standards for fine particles or PM2.5. Die casting emissions factors are over two decades old.

The community has asked repeatedly over the years that Kipp more completely characterize die casting emissions and update die casting emissions factors. Given that several of Kipp’s EPA violations had to do with issues that would affect die casting emissions, it is problematic that die casting emissions were not part of the stack testing done for the NOV. We don’t know why, but Kipp’s lawyers likely negotiated them out of the agreement if they were ever on the table for consideration. Further, originally the EPA planned to test dioxins/furans to resolve this Notice of Violation. However, Kipp’s legal team talked them out of doing dioxin/furan tests some time in 2014 (based on MEJO communications with EPA).

Koblinski: “I approached the Goodman Center to see if they had an energy efficiency project that I could help them with and ultimately agreed to pay $80,000 towards their needed chiller upgrade…”  

MEJO Response: We think improving energy efficiency is very important, but this Goodman Supplemental Environmental Project (SEP) does nothing to reduce exposures to Kipp’s toxic pollution (as described above) among the children who play at Goodman just feet from the factory. Also, it makes it even more likely that Goodman leaders will not raise any questions about Kipp’s pollution. Madison-Kipp, its wealthy owner Reed Coleman, and/or the several foundations to which Mr. Coleman funnels his wealth, have supported the Goodman Community Center with funding and other types of support since it was located next to Kipp and before that when it was the Atwood Community Center. This is one reason the Goodman Center leaders do not raise concerns about Kipp’s pollution, and dismiss or ridicule those in the community who do raise these concerns. The Goodman SEP is, in effect, more hush money to Goodman. What a boon for Kipp—congratulations!

Meanwhile, the SEPs Kipp agreed to didn’t consider in any way the concerns and suggestions of many people in the Kipp neighborhood for over two decades. Since the 1990s, hundreds of Kipp neighbors have asked for more monitoring, better emissions and noise controls, and a variety of other actions to reduce or eliminate the noise, odors, and toxic pollution from Kipp. In most cases, their concerns have been dismissed or ignored by Kipp (and DNR).  

Koblinski: “The violations indicated in the NOV did not “expose residents and children at the center to harmful pollutants” as they were administrative in nature and we have always operated well within the limits of our permits (in fact less than 4% of our total allowable).

MEJO Response: As outlined above, the violations were much more than just “administrative” and likely did result in higher emissions of some hazardous air pollutants, which continue to be emitted from the factory at unknown levels. Close neighbors of Kipp, including many children, and children playing at the Goodman Center, are among the most exposed and the most vulnerable to these hazardous pollutants released from stacks and vents just feet away from where they live and play.  

Koblinski: “The EPA did not make a “surprise, unannounced visit to our facility,” rather the inaccuracies in our reporting were discovered through the standard annual reporting that we do to the EPA and subsequent information exchanges throughout 2012.”

MEJO Response: This EPA document about the Jan. 12, 2011 inspection of Kipp describes the inspection as “unannounced.”  We don’t think EPA would lie about this.  

Koblinski: Lastly, I would use this opportunity to indicate that we are progressing well on a number of fronts as we take responsibility for the unintended environmental impacts caused over previous decades.  

I won’t list all of the actions taken and completed as those are documented through various reports submitted to the WDNR and posted to their site.  Work still ahead of us includes:

– Some additional soil remediation (removal) work remains to be done in the storm water ‘bio-basin’ (Rain Garden) and adjacent bike path.  

– We will be bringing the Ground water Extraction and Treatment System GETS on line later this month to pump and filter PCE impacted groundwater.

– We are working with WDNR and EPA to resolve remaining PCB issues under our plant.  

We will, of course, continue to regularly monitor and verify our network of wells and probes to ensure there is no risk to public health.  

MEJO Response: Please share with the neighborhood and public the most recent documents about these activities—especially those regarding the continued removal of PCB contamination from the raingarden/bike path and also under the factory. What work remains to be done and why? Please share any available data.

A representative from the Madison/Dane Co Board of Health recently reported to the Madison Committee on the Environment that all the PCBs at Kipp have been remediated. Obviously, this is incorrect. Recent documents about these activities are not posted on the DNR website. People in the neighborhood, and the public officials who serve them, need to know about ongoing remediation of highly toxic soils going on right in the middle of where they live, work, and play (children are playing at a splash pad just feet from these PCB remediations!).

Sincerely,  

Tony Koblinski

President/CEO

Madison-Kipp Corp.

tkoblinski@madison-kipp.com

MEJO Response: Thanks, Mr. Koblinski! Drop us an email any time if you have further questions or comments.

Sincerely,

MEJO

info@mejo.us


Digg thisShare on FacebookShare on Google+Email this to someonePrint this pageShare on RedditShare on TumblrTweet about this on TwitterPin on Pinterest

EPA fines Madison-Kipp for air violations; Goodman Center gets A/C in the deal!

EPA fines Madison-Kipp for air violations; Goodman Center gets A/C in the deal!
[ABOVE] The view of Madison-Kipp Corp. from the Goodman Community Center; PCBs are being removed from the rain garden in the foreground

The U.S. Environmental Protection Agency has fined Madison-Kipp Corporation for Clean Air Act violations from 2007 through 2012. While it could have been fined $37,500  per day for each violation, Kipp got away with paying a $35,000 fine, and agreeing to install energy-efficient windows in its factory and to pay for new air conditioning at the adjacent Goodman Community Center!

That’s right: a community center gets new air conditioning because a factory repeatedly violated the Clean Air Act. [This is not an Onion headline!]

From the EPA Administrative Order: “On behalf of the Goodman Community Center, a not-for-profit organization located in Madison, Wisconsin….respondent [Kipp] must…provide funding to replace, in whole or in part, a faulty and inefficient tandem chiller with a more energy-efficient unit…Respondent must spend at least $80,000 towards this….”

So Clean Air Act violations occurred for five years, affecting the Atwood neighborhood, exposing residents and children at the center to harmful pollutants, and the “penalty” is air conditioning for the center. WTF!?!

All the while, the Wisconsin Department of Natural Resources didn’t notice these Kipp air violations. If not for concerned residents asking for it, the EPA never would have made a surprise, unannounced visit to the Kipp factory and uncovered the problems.

The Clean Air Act violations agreements can be  can be found here:

EPA Administrative Order re: Madison-Kipp air violations (March 31, 2015)

EPA Administrative Order re: Madison-Kipp air violations (April 6, 2015)

 

 

 

Digg thisShare on FacebookShare on Google+Email this to someonePrint this pageShare on RedditShare on TumblrTweet about this on TwitterPin on Pinterest

City to Lease Rain Garden to Polluter

City to Lease Rain Garden to Polluter

(High resolution version of above graphic here)

SUMMARY: The City of Madison wishes to extend a lease with Madison-Kipp Corp. through 2023 that will allow Kipp to control the rain garden it has polluted, continue to use part of the north parking lot (which it has also contaminated), and will give full credit towards its lease costs for a wooden fence that allegedly blocks sound. Meanwhile, people who live next to Kipp say that noise from the factory has been louder than ever, and often is the worst in the middle of the night. [1]

Given Kipp’s long history of polluting Starkweather Creek and Lake Monona through its unmonitored discharges into the City storm water drains (including one in the rain garden, where most of Kipp’s contaminated stormwater discharges are released), it’s curious that the City would now like to give Kipp control over this land.

DETAILS: The City of Madison has a long relationship with Madison-Kipp Corp. The city owns the land under the Kipp Fair Oaks factory and under the north parking lot at the Kipp Waubesa factory. Now the City wants to lease its heavily-polluted rain garden to Kipp for free (zero rent because of the full credit for the “sound blocking” fence).

Runoff from nearly the entire Kipp Waubesa factory property has been going into a City stormwater drain for many decades (a century?). In 2006, the City built the ill-fated rain garden where this pipe discharges, despite existing documentation of extensive PCE contamination throughout the Kipp site. This area was already leased to Kipp, so the City amended its lease, took back the land and gave Kipp a $900 lease credit. From the 2009 amended lease:

During the time of the recent construction of the new Goodman Community Center at 149 Waubesa Street, the City of Madison constructed a storm water management surface water retention facility adjacent to the East Rail Corridor Bike Path. The facility was constructed in a portion of the leased premises that the City leased to the Madison-Kipp Corporation (MKC) beginning in 1998. In order to correct the situation, it is necessary to amend the lease to remove that portion of the leased premises that the City reoccupied. The portion contains approximately 2,100 square feet and its removal equates to a permanent reduction of the annual rent of $900.

In 2012 and 2013, several toxic contaminants, including PCE, PCBs, PAHs and metals were discovered in rain garden soils (see the main graphic above here).  Soil PCB levels in some spots were many order of magnitude above DNR standards for direct contact. Several rounds of excavation have been done there, but the big question still remains: where did all this pollution come from?

MEJO has learned of a never-before disclosed Kipp factory drainage system that goes under the factory and leads to the stormwater drain and ditch (leading to Starkweather Creek, which eventually drains into Lake Monona). Historically, PCBs, PCEs, PAHs, dioxins, metals, and other unknown (unmeasured) chemicals from Kipp processes have flowed into this drainage system.

The City does not know the full extent of this drainage system and has not investigated what chemicals are currently draining into it from Kipp’s air vents/stacks and ongoing remediation projects–or what might be entering it from under the Kipp factory.

It has finally been made public that Kipp had a secret trench in its factory that drained into the City stormwater drain. EPA is now working to get Kipp to remediate the unbelievably high levels of PCBs under the factory. FYI: EPA still hasn’t settled with Kipp over its air pollution permit notice of violation. Perhaps the City should investigate the toxins going into the rain garden before it hands it over to Kipp.

City "rain garden" between Kipp and Goodman

City “rain garden” between Kipp and Goodman

Kipp has polluted this City land for a century (along with Kipp’s City-owned parking lot).  Why should the City allow Kipp to have control of public land that it has treated so badly?  Is there any evidence that Kipp can be a “good steward”of public lands? Since this land is adjacent to the City bike path and a community center, the City might be wise to keep control of it rather than allow a lessee with such a bad track record control it. Instead, this public rain garden will now become Kipp’s private property.And people who live along Waubesa and Marquette Streets, as well as those families who use the Goodman Community Center, not to mention the parents of all the children who will use the new splash pad next year (right next to the rain garden!), should know about this and have a chance to tell District 6 Alder Marsha Rummel and the City if they agree with the wisdom of giving Kipp control of more public land. There should be a public neighborhood meeting at the Goodman Center before the Madison City Council allows Kipp to lease the rain garden.

Addendum: The “sound blocking” wooden fence was supposed to be on the north boundary too (along the bike path, between Kipp and the Goodman Community Center. This wasn’t built. Here is the proposed location per the 2009 amended lease.

[1] Kipp and government agencies attribute this increased noise to the testing and construction of the groundwater remediation system. However, many people living on Marquette Street have been experiencing increased noises from Kipp since before this construction started. Also, the noise often goes all night long and people say they cannot sleep. Why does Kipp need to do this noisy work all night long?

Digg thisShare on FacebookShare on Google+Email this to someonePrint this pageShare on RedditShare on TumblrTweet about this on TwitterPin on Pinterest

PHMDC Assures Public Goodman Splash Pad is Perfectly Safe Using Incorrect Numbers

PHMDC Assures Public Goodman Splash Pad is Perfectly Safe Using Incorrect Numbers

(After this story was posted, PHMDC corrected their numbers in the document linked to below)

On August 26 the Wisconsin DNR released a public update including an “Evaluation of the Potential Health Concerns Associated with the Construction of the Goodman Center Splash Pad” by Public Health Madison Dane County (PHMDC).

Unfortunately, the “Residual Contaminant Level” (RCL) table in the document is riddled with incorrect numbers and miscalculations. The RCLs listed are outdated and even some of the old numbers are incorrect (or have mixed up units–e.g., the old RCL for benzo(a)pyrene was .0088 mg/kg, not 8.8 mg/kg; other RCL numbers are also incorrect). Numbers in the “estimated increase in disease risk” column are miscalculated.

The table shows that contaminants levels found in soils at the Goodman Center were large orders of magnitude higher than DNR direct contact soil RCLs, but PHMDC concludes that “patrons of the water activity”—in other words, “small children”—will not be exposed to contaminated soils, dust, and sediment during “normal operation” of the splash pad. What about when the splash pad isn’t operating “normally?” What might “abnormal operation” of the splash pad entail? The document also asserts that the splash pad water tank, buried in contaminated soil and at times submerged in groundwater, will be “impermeable.” What?!?! Everybody knows that underground storage tanks always leak, sooner or later….

To be continued…

 

 

Digg thisShare on FacebookShare on Google+Email this to someonePrint this pageShare on RedditShare on TumblrTweet about this on TwitterPin on Pinterest

More Alarming PCB Hotspots Along City Bike Path Next to Goodman Center; Public in the Dark

More Alarming PCB Hotspots Along City Bike Path Next to Goodman Center; Public in the Dark

Kipp’s Sludge Garden Saga continues… an Arcadis (Kipp’s consultants) report dated August 6, 2014 includes data generated in April and May of this year showing the highest PCB hotspots found to date–550 ppm and 1020 ppm–from soils excavated at the far west edge of the raingarden (see pg. 13 of above report), right next to where a Kipp storm drain pipe emptied contaminants into the raingarden area for decades. These levels are far higher than in previous rounds of excavation in the raingarden, which also revealed many PCB hotspots.

The highest PCB level found in recent excavations, 1020 ppm, is 1457 times the DNR’s industrial direct contact residual contaminant level (RCL) of 0.7 ppm and 5100 times the non-industrial (residential) RCL of 0.2 ppm.[1],[2]

Oddly, the Arcadis report does not mention the sources of the PCBs. Moreover, it does not state the depth of these samples, making it very difficult to assess what the past and current sources of the PCBs might be, how and where the contaminants might travel over time, and how likely it is that people will be exposed to them now or in the future in different contexts.

Though some soils in the raingarden area with the highest detections of PCB have been removed, soils in much of the area could not be excavated because of buried utility lines, and the full extent of the contamination remains unknown. Contaminated runoff from the northern part of the Kipp property continues to discharge from the stormwater pipe that drains into the raingarden, as Kipp removes/repaves parking lots and constructs the groundwater extraction system. Soils in the northern parking lot area, where most of this work is occurring, are among the most contaminated soils on the site.

The City of Madison land on which these hotspots were found is next to a heavily used city bike path along which many children bike and play every day (see here and here) and just across from the ill-advised Goodman Center children’s splash pad, currently under construction. Throughout the excavations, no signs have ever been posted to let people know what is going on and to keep children and pets away.

To date, this PCB data has not been shared with the public by Kipp, DNR, or another agency. It is not posted on the DNR website and it is not in the Hawthorne Library.[3] Why not?

 

[1] Though government agencies are using the industrial RCL, given that this is public land feet away from a public bike path and a community center, the lower, more protective non-industrial (residential) RCL should be used.

[2] Four samples analyzed just last week to the west of these hotspots under Kipp’s driveway (leased from the City of Madison) showed much lower levels of PCBs, but consultant reports again do not include the depths of the samples, or even maps of their locations, making it difficult to assess the importance of these findings.

[3] Several years ago, people in the Kipp neighborhood asked DNR to place hard copies of critical Kipp and government documents relevant to Kipp’s soil and groundwater pollution (including consultant reports as well as DNR and other agency reports and communications, etc) in the Hawthorne Library for people who do not have internet access. Although there are several large Arcadis reports in the library, the recent raingarden data is not there and to date none of important Kipp-related DNR and other agency documents have ever been placed there other than a few very old DNR reports on Kipp’s site investigations.

 

Digg thisShare on FacebookShare on Google+Email this to someonePrint this pageShare on RedditShare on TumblrTweet about this on TwitterPin on Pinterest

A “Healthy and Safe Place” for “Good Clean Fun”?

A “Healthy and Safe Place” for “Good Clean Fun”?

Is this a safe place for children to frolic in a splash pad? Here are some of the toxic contaminants and safety risks not visible in the above photo…

A post in today’s Cap Times online raises important questions about whether the children’s splash pad planned at the Goodman Community Center will be a “healthy and safe place” for “good clean fun,” as the center’s Executive Director Becky Steinhoff assured us it will be in the May/June issue of the Eastside News.

Adding to questions raised in previous posts, there is a glaring omission in considerations about the safety of this location for a splash pad. Trucks carrying chlorine, toxic chemicals, and hazardous wastes to and from Kipp—and spewing hazardous diesel fumes—will rumble just feet past the children’s splash pad many times a day, every day.

Aluminum die casting is inherently toxic and dangerous, and Kipp has a lousy health and safety record. As we outlined in a previous post, between 1998 and Feb. 2014, the Madison Fire Department/EMS (emergency medical services) made 172 calls to Kipp for fires, explosions, accidents, and/or worker health problems and injuries.[1]

Kipp uses, stores, and transports chlorine, in addition to numerous other highly toxic chemicals. Chlorine is a highly toxic gas that can cause severe health problems (including death) to people exposed to high enough levels. Severe health effects from chlorine inhalation can occur within minutes—well before HazMat teams can get to the scene. The highest potential for such a release near Kipp is during the transport of chlorine; for details, see Kipp’s “Hazardous Materials Incident Initial Response Guidelines.” . With all the trucks going in and out of Kipp for well installation, remediation, etc., the chances of such an accident have likely increased in recent years.[2]

What will happen if there is a chlorine truck accident involving a chlorine release next to the splash pad while children are playing there? What plan does Goodman Community Center have in place for such an incident at its facility? Not long ago, community members asked Goodman Center staff this question and they had never heard of such a plan; they were clearly unprepared. How will the center staff handle such a chlorine or other toxic material accident next to a crowded splash pad? How will kids at an outdoor splash pad “shelter-in-place” if a chlorine spill happens 50 feet away?

[1] The actual number of fires, accidents and injuries is probably higher than this, since Kipp likely tries to avoid calling the Fire Department unless the accidents reach a certain severity level.

[2] Emergency Planning and Community-Right-To-Know (EPCRA)

MKC uses, stores, transports, and releases many highly hazardous chemicals, and is required to follow Emergency Planning and Community-Right-To-Know laws—see here for federal regulations and here for Dane County. Hazardous chemicals at Kipp include chlorine, propane, sodium hydroxide, nitrogen, fluorides, molten aluminum, fuel oils, lubricants and many more. The use, storage, and transport of chlorine are among the most potentially dangerous situations at and around Kipp. According to the “Hazardous Materials Incident Initial Response Guidelines” for Madison Kipp, “the worst-case scenario would involve chlorine being released directly to the outside of the facility during transport.”

Chlorine’s health effects: Permissible Exposure Level–1 ppm; detectable odor threshold–over 1 ppm; 3-5 ppm–slight irritation of the nose and upper respiratory tract; 5-8 ppm–irritation of the respiratory tract and eyes; 10 ppm– immediately dangerous to life and health; 15-20 ppm–immediate severe irritation of the respiratory tract, intense coughing and choking; 30 ppm–shortness of breath, chest pain, possibly nausea and vomiting; 40-60 ppm–development of chemical bronchitis and fluid in the lungs, chemical pneumonia. Prolonged exposure over 50 ppm will cause unconsciousness and death.

Two types of “vulnerability zones” for chlorine accidents at Kipp were modeled in Kipp’s Haz Mat Guideline document. One is the area in which chlorine levels could reach 1/10 of the IDLH (immediately dangerous to life and health)—or the level that “poses a threat of exposure to airborne contaminants when that exposure is likely to cause death or immediate or delayed permanent adverse health effects or prevent escape from such an environment.” The 1/10 factor is added to protect especially vulnerable people, such as those with respiratory disease or illness. The IDLH for chlorine is 10ppm so the 1/10 IDLH is 1ppm. An alternative vulnerability zone based on ERPG-2 (Emergency Response Planning Guidelines), or “the maximum airborne concentration below which it is believed that nearly all individuals could be exposed for up to one hour without experiencing or developing irreversible or other serious health effects or symptoms which could impair an individual’s ability to take protective action.” The ERPG-2 for chlorine is 3 ppm.

Vulnerability zones for chlorine releases are: 3.1 miles at 1 ppm (1/10 IDLH), 1.9 miles at 3 ppm (ERPG-2) “in the worst case scenario, dangerous or deadly levels of chlorine contamination may reach a distance of between .1 and .25 miles downwind of the source within a matter of minutes following the initial release.” Later, the document states that “The lead time for a HAZMAT incident could be from 15-40 minutes. As a result, this short time may not allow for a safe evacuation…An evacuation under these circumstances may expose the population to dangerous toxic chemicals and the decision may be made to shelter-in-place.” The document then lists 11 “shelter-in-place” recommended instructions.

**********

Do you live in the Kipp Neighborhood? If there is a chlorine accident at Kipp or an accident involving a chlorine truck on a street in the neighborhood:

-How will you know? Is there any system in place to immediately notify you and others in the neighborhood?

-Do you/your family know what the “shelter-in-place” steps are? Do staff at Goodman Community Center/Lowell School?

-Is the community prepared? Are people near Kipp even aware that the potential exists for such an accident?

If you live in the Kipp neighborhood, contact your elected officials, public health agencies, Madison Fire Department and OSHA, with questions. Ask them to look into whether Kipp is following EPCRA laws. Ask them to initiate a public meeting to help prepare residents, schools, and community centers for a chlorine and/or other hazardous chemical accident at or around Kipp. Always include your street address when contacting your political representatives.

Madison Alder Marsha Rummel: 608-772-4555, district6@cityofmadison.com

Senator Fred Risser: (608)266-1627, sen.risser@legis.wisconsin.gov

Representative Chris Taylor: (608) 266-5342, Rep.Taylor@legis.wisconsin.gov

John Hausbeck (Epidemiologist, Madison Dane County Public Health): 608.243.0331, JHausbeck@publichealthmdc.com

Henry Nehls-Lowe (Epidemiologist, Department of Health Services): 608-266-3479, Henry.NehlsLowe@dhs.wisconsin.gov

Dave Bursack, Dane County Local Emergency Planning Committee: 608-266-9051,

Madison Fire Department: 608-266-4420, fire@cityofmadison.com

Occupational Safety and Health Administration (Madison Office): (608) 441-5388

 

Digg thisShare on FacebookShare on Google+Email this to someonePrint this pageShare on RedditShare on TumblrTweet about this on TwitterPin on Pinterest

“Forgotten” Kipp Toxic Hotspot Bubbles to the Surface Next to Goodman Community Center

“Forgotten” Kipp Toxic Hotspot Bubbles to the Surface Next to Goodman Community Center

Now that the neighborhood class action lawsuit against Kipp is settled [1] and Madison-Kipp has paid a total of $7.2 million dollars to homeowners near the factory, Kipp and government agencies would like all questions about risks to people living, working, and playing on properties adjacent to the factory to just go away.[2] With the Goodman Community Center about to build a children’s splash pad less than 50 feet from one of Kipp’s most toxic areas, unanswered questions about risks to people who spend time at the center, especially children and pregnant mothers, are more relevant than ever.

Just how toxic is the northern part of the Kipp property and the adjacent city property? Here’s a visual summary. Where did these high levels of contaminants come from? Do they pose potential risks to people at the Goodman Center? In a memo on April 30, 2013 (pg. 4) Dr. Lorne Everett, expert in the class action lawsuit, noted:

Considering the proximity of the Goodman Community Center (immediately north of Madison-Kipp) to high on-site PCE concentrations in soil vapor and given the fact that high levels of groundwater contamination are found even farther north, it appears that groundwater and/or soil vapor under the Goodman Community Center are impacted with Madison-Kipp’s contamination. It would be prudent to conduct soil vapor and subslab sampling at the Goodman Center but the Arcadis report completely omits a discussion of this off-site threat.”

Since the class action lawsuit was settled, even more data has been released that indicate testing at Goodman would be “prudent”—especially before allowing children to frolic at a splash pad just feet away.

Instead, Kipp and government agencies seem to be digging their heads deeper and deeper in the sand. Read on…

High Soil Vapor Results along City of Madison Bike Path

The most recent publicly available data from the northern boundary of Kipp’s property and on the City of Madison property between Kipp and the Goodman Center begs for more testing of groundwater, soils, and vapors on the Goodman property. Three rounds of soil vapor tests (two in 2012, one in April 2013) just south of the bike path on City of Madison property—only about 20-30 feet from where the splash pad will be—have shown increasing levels of vapors in the soil only about 7-8 feet down. In April 2013, the soil vapor probe directly across from the splash pad area (VP-6) showed alarmingly high vapor levels: PCE: 19,000 µg/m3; TCE: 6,100 µg/m3; cis 1, 2 DCE: 8,200 µg/m3; trans-1, 2 DCE: 320 µg/m3; Vinyl chloride: 340 µg/m3.

Shhhhh….Kipp Doesn’t Want Anyone to Notice…

Madison-Kipp seems to be purposely trying to downplay (or hide?) these important results on how contaminated the north side of their property is and how their operations have contaminated the City of Madison property adjacent to it. The most recent soil vapor data was part of the 2013 Kipp Annual Report not released till April 2014.[3] The very high VP-6 results were tucked away at the end of a huge table in a separate document (p. 152) that was not included with the actual Annual Report document, and now doesn’t seem to be easily accessible anywhere. Results were reported in ppbv (parts per billion by volume), which looks much lower than the more standard reporting units of µg/m3 (micrograms per cubic meter).

Oddly, Kipp consultants used the less protective, non-residential screening levels with an attenuation factor of 0.001 for deep soil gas to indoor air (vs residential attenuation factor of 0.01), which allowed these very high vapors levels not to be flagged as exceeding the “Wisconsin Residential Deep Soil Gas Calculated Screening Levels.”[4] Using the deep soil gas, non-residential attenuation factors and screening levels in this context is questionable, for a number of reasons (and depend on who is the focus of the risk assessment—see discussion below). The groundwater at this location tends to be very shallow; in spring 2014 the groundwater was only a few feet down and all the 7-8 feet deep vapor probes on the bike path were under water. Also, the probes are on public property (City of Madison), not Kipp’s property.[5],[6]

Further, the text of the Annual Report doesn’t mention these high vapor probe results and the fact that vapor levels at VP-5 and VP-6 have increased significantly in recent tests. Instead, it states disingenuously that “In general, the 2013 data are consistent with previous vapor probe data.” In fact, the 2013 VP-5 and VP-6 levels are significantly higher than the previous tests. For instance, vapor probe 6 levels for just PCE went from 63 ppbv (413 µg/m3) on 3/3/2012 to 190 ppbv (1245 µg/m3) on 10/26/12 to 2900 ppbv (19,000 µg/m3) on 4/29/13—increasing by 46 times from the first to the third test. PCE breakdown products TCE and vinyl chloride, which are even more toxic than PCE, also increased significantly at this probe and at VP5. The obscure placement of these results, use of deep soil gas attenuation factors and non-residential screening levels, use of ppbv instead of ug/m3 in reports, and dishonest discounting of them in the Annual Report seem to be attempts to make it more likely that the few people who ever see find these results buried in reports will skim over them without notice.

What Potential Exposures are Being Assessed? Exposures to Workers, Children at Goodman, or Anyone At All?

The kind of attenuation factors and screening levels risk assessors choose typically depend on whether nearby buildings (into which vapors might enter) are industries, businesses, homes, schools, community centers, etc.—which in turn determines what kinds of people might be exposed to vapors and to what extent.

If the goal is to assess vapor risks to workers in the Kipp factory, non-residential attenuation factors and screening levels would be more defendable (though still problematic).[7] However, the original reason for the probes along the bike path, as far as we understand (based on our conversations with Henry Nehls-Lowe at Department of Health Services),[8] was to explore, moving outward from Kipp, how far vapors extend away from Kipp towards Goodman Community Center. The more protective residential attenuation factors and screening levels are typically used for schools, daycares, and community centers such as Goodman. In other words, if the goal is to assess potential risks to people at the Goodman Center building, the residential attenuation factor of 0.01 should be used—which would put these levels well over the residential screening levels, indicating that more testing should be done.

Why are the vapor probes along the bike path in the first place?

While the levels found at vapor probes along the bike path raise big red flags about the Goodman property, because they suggest that there is a strong contaminant source nearby, they do not tell us whether or not vapor intrusion might be occurring inside the Goodman Community Center building.

EPA’s OSWER Draft Guidance for Evaluating the Vapor Intrusion to Indoor Air Pathway from Groundwater and Soils (Subsurface Vapor Intrusion Guidance (OSWER is the Office of Solid Waste and Emergency Response) [9] recommends that buildings be assessed for vapor intrusion when they are “within approximately 100 ft laterally or vertically of known or interpolated soil gas or groundwater contaminants” and the “contamination occurs in the unsaturated zone and/or the uppermost saturated zone.” (p. 16).   The EPA OSWER guidelines also highlight that “It is important to consider whether significant preferential pathways could allow vapors to migrate more than 100 feet laterally. For the purposes of this guidance, a “significant” preferential pathway is a naturally occurring or anthropogenic subsurface pathway that is expected to have a high gas permeability and be of sufficient volume and proximity to a building so that it may be reasonably anticipated to influence vapor intrusion into the building. Examples include fractures, macropores, utility conduits, and subsurface drains that intersect vapor sources or vapor migration pathways.”(emphasis in original) Building conditions, weather, water table fluctuations, other buildings, barometric pressure, and all sorts of other factors can also influence vapor intrusion in buildings. In some contexts buildings further away from deeper groundwater sources can have more vapor intrusion problems than those closer to sources.[10] Give all of this, repeated tests in the subslab of the Goodman building (which is within 100 feet of contamination on Kipp and city property) in several locations in the building over time (in different seasons) would be the more appropriate way to assess, or rule out, vapor intrusion and potential exposures to people there.[11]

Kipp, Wisconsin Department of Natural Resources (DNR), and Department of Health Services (DHS) have repeatedly claimed that the groundwater contamination under Goodman is too deep to cause vapor intrusion. However, to date, as far as we know, just two groundwater samples have been taken on the Goodman site– from only one shallow depth (7-17 feet) from a well far from the Goodman building on the east side of the property. The August 2013 sample found PCE at 1.4 µg/L, and the October 2013 sample was less than 0.17 ug/L PCE. Based on this, Kipp’s CEO Tony Koblinski assured the public in his March 19, 2014 presentation (Part 2, 15-16 minutes in) that they had verified, with a “no detect” that “there isn’t any PCE in the shallow groundwater that would cause a soil vapor concern” at Goodman. These limited groundwater results, at one depth, hundreds of feet from the Goodman building, are nearly meaningless as far as assessing vapor intrusion there.

Also, as far as we know, Kipp has not investigated any potential preferential pathways, such as sanitary sewers and storm sewers, electric and gas lines, etc., all around and in between Kipp and the Goodman Center that could have routed shallow level contamination offsite, causing soil and groundwater contamination and/or vapor intrusion along these conduits. Mapping these preferential pathways is supposed to be one of the very first steps in developing a “Conceptual Site Model” (CSM) in the initial stages of site investigations for VOC contaminated sites (as MEJO and our collaborators pointed out to the DNR/DHS/PHMDC/Water Utility and city engineering team when we met with them in February 2012).

Do government agencies really think enough is known to rule out vapor intrusion at the Goodman Community Center, from contaminated groundwater and/or shallow soils on Kipp’s northern boundary, or both? Interestingly, DNR’s vapor intrusion expert Terry Evanson, in an April 15 2013 memo reviewing Kipp’s Site Investigation report, stated that: “The vapor pathway, including all buildings at risk, both on and off-site, needs to be addressed in the SI” (site investigation) (highlights added). Yet as of spring 2014, even with the planned splash pad, neither Kipp nor DNR have any intention of testing soils, soil gas or vapors anywhere near or in the Goodman building—or groundwater beneath it. Just as troublingly, soils on the site, which had significant contamination due to past industrial activities, haven’t been tested for contaminant levels since before the site was remediated (no confirmation/post-remediation soil testing was done by the consultants).[12],[13]

We are not surprised that Kipp is denying that there could be problems at Goodman related to their operations. We are not surprised Kipp is ignoring evidence, scientific studies, and government guidances on assessing pollution and health risks. We are not surprised that Kipp is not very concerned about environmental laws or exposing children to toxic pollution. But why do some government agencies seem to be going along with this approach as well? Perhaps it is because Kipp’s law firm Michael Best and Friedrich advised them to stay away from the City of Madison public property adjacent to Kipp as well as the Goodman Center?[14] Perhaps all these entities are worried about the messy political can of worms a more thorough investigation of Goodman site might uncover, because Kipp has supported Goodman Center financially for years and City of Madison property is in between Goodman and Kipp? Perhaps they don’t want anyone to uncover the evidence that Kipp lubrication division, KLS Lubriquip, which used PCE, operated and/or stored materials at the Goodman site in the past (see our previous article)?

What’s the Vapor Source for High Levels Along Bike Path?

The increasing contaminant trends at the two vapor probes (VP-5 and VP-6) along the bike path strongly indicate that there is a high contaminant source nearby. What might it be? Contaminated groundwater was just inches below Vapor Probe 6 when the April 2013 vapor tests were done, so contaminant vapors may have come from the groundwater itself. Though shallow VOC levels at the nearest water table well (Well 1) in 2012-2013 tests were not particularly high, in levels at Well13, just feet to the south of that, have been consistently the highest at the sight (though the shallowest measurements are 44 feet and the water table is well above that). However, Kipp has not shared all the data about groundwater results near these vapor probes. Groundwater data from “temporary wells” on the Kipp site in 2012, never shared with the public, showed the following levels at B51, at the northeast corner of the Kipp building, not far from VP-6: 1500 µg /L PCE, 500 µg /L TCE, 4078 ug/L cis/trans DCE, 890 ug/L vinyl chloride. These results are particularly concerning because of the very high levels of PCE breakdown products (TCE and vinyl chloride), which are considerably more toxic than PCE.[15] It is not clear how deep these tests were.

In addition to groundwater, there is another likely source of the contaminated vapors at VP-5 and VP- 6—highly contaminated soils along the former storm sewer that runs from the contaminant hotspot on the northern edge of the building, along the northern edge of the property, and empties into the former raingarden. The end of the pipe that emptied into the raingarden, uncovered in the most recent excavation of the raingarden (see photo above), emptied much of the highly contaminated wastewater from Kipp’s processes into this area for many decades. Several other storm sewer pipes—most likely including some from the contaminated areas further south on the Kipp site—drain into this pipe.

Where did the contamination that went into the ditch and storm sewer come from? According to the 2012 deposition of former Kipp “environmental manager” Jim Lenz during the recent class action lawsuit, there were storm drains all over Kipp’s northern parking lot. Apparently Kipp built storm drains purposely routed north, which joined together into one pipe that emptied into the rain garden area (which then routes runoff to the northeast corner of the property). The contamination entering this “contaminant watershed” came from leaking above ground PCE tanks that drained into the ditch, vapor degreasers along the east side of the plant that vented outside (condensing contaminants onto the soil), and other contaminated areas on the site. Some PCE and various other wastes were purposely dumped outside by Kipp workers, and according to Mr. Lenz one dumping area was off the northern corner of the building (near the northern end of the ditch that connected to the storm sewer pipe). After the parking lot was paved, the oils on the parking lot, which included PCBs, PCE, and other contaminants, ran off to the north over the surface and via underground storm sewer drains.[16] There is evidence that Kipp employees continued to dump chlorinated compounds and other toxic wastes into storm and sanitary sewers through the 1990s if not longer.

Kipp Waste Ditch and Storm Sewer Were Focus of Investigations in the 1990s

Though most recent documents cite 1993-94 as the first discovery of Kipp’s groundwater and soil contaminant problems, engineering consultants (and likely DNR and Kipp as well) knew Kipp might have a groundwater problem when testing at the Kupfer Ironworks site in1986 found some chlorinated compounds in groundwater that were attributed to Kipp operations.[17] It isn’t clear what happened after this initial discovery of chlorinated compounds at the Kupfer site, but in 1993 PCE was found at the Brassworks property and again attributed to Kipp. In 1994, the DNR asked Kipp to formally investigate the contamination, after which a series of investigations by Kipp consultants Dames & Moore was launched.

Dames and Moore’s investigations focused on the northern ditch and storm sewer through the 1990s, after which these sources abruptly disappeared from reports. Dames and Moore noted in one of its first reports, on Dec. 14 1994, that “It is probable that the chlorinated VOCs discovered in the soils at the Madison-Kipp site is (or was), within the “watershed” which contributes to the ditch.” (italics added). Reports clearly diagramed the location of the ditch and storm sewer (e.g., 1996 Dames & Moore Kipp report, March 1997 Dames and Moore report and May 1997 Dames and Moore report; downloading documents can take some time).

All the consultant reports throughout the 1990s outline extremely high levels of soil, soil gas, and groundwater CVOCs (chlorinated volatile organic compounds) as well as benzene, xylene, toluene, and napthalene contamination throughout the ditch and storm sewer area. Reports are clear that the ditch and sewer route were significant contaminant sources that had already impacted groundwater. For instance, the March 20, 1996 Dames & Moore report says: “…soil sample analyses reveal a source area for chlorinated VOCs” in an area “coincident with a former drainage ditch, which extended from the building at the south end (upgradient, adjacent to the former above ground PCE tank), northward until it terminated at the northeast corner of the building. At the terminus, a pipe transferred runoff from the ditch to the storm sewer.” Interestingly, according to this report, the PCE tank “was taken out of service several years ago, the ditch was filled in and the area paved in 1995”[18] (italics added).

The 1995 document reported extremely high levels of CVOCs in soil gas down to 18-20 feet.[19] Consultants admitted that this ditch/storm sewer area is a “source for the transport of chlorinated compounds to the water table” and that “contaminants have migrated into the sandstone bedrock.” However, consultants proposed to do a pilot study on potential remedial options and initiate groundwater extraction and containment.[20]

The March 18 1997 Dames & Moore document reported that the extraction well (EW-1, which appears to be near where MW 13 is currently) along the sewer drain, 10-20 feet off the northeast tip of the factory, had 150,000 µg/L PCE. [21] Again, the report highlights the storm sewer as the source of contaminants. The May 30, 1997 Dames & Moore document, which didn’t mention EW-1,reported stunningly high levels of PCE in shallow soils (2-4 feet) at GP-9, a boring just off the northeast corner of the building: 6,440,000 µg/kg PCE; 126,000 µg/kg TCE; 30,900 µg/kg cis-1,2, dichloroethylene, as well as high levels of benzene compounds, xylenes, toluene, and naphthalene. They also found high levels of most of these compounds 8-10 feet down at this location. Just feet away from that, “beneath a drain pipe which is covered with thick concrete,” where a catch basin pipe joined with the storm sewer pipe, and further along the storm sewer drain, very high levels of CVOCs were found in soils.

Consultants dropped plans for excavating and propose a site specific standard of 1000 ug/kg for soils…

Despite these high levels, consultants concluded that “complete soil remediation by means of excavation” was “technically and economically infeasible” because “the depth of impacted soil (18 to 20 feet) would result in the need for areally (sic) extensive excavation or significant shoring requirements for soil stability” and “utilities and other obstructions would likely prohibit either of these options.” Instead, Kipp and its consultants proposed “a modified excavation” of 8-10 feet of soils in two areas, combined with soil vapor extraction, to remediate soils impacted above “site-specific residual contaminant levels” (RCL). Consultants came up with  “site-specific” groundwater and soil RCLs based on equations (from a 1982 paper) and some problematic assumptions about “natural attenuation,” plume size at that time, and groundwater depth. Based on this, consultants recommended a site specific RCL for soils of 1000 µg/kg (or 1 mg/kg)—over 200 times the DNR’s soil to groundwater pathway “residual contaminant level” (RCL) of 4.54 µg/L for PCE in soils.

Kipp’s Northern Hotspot Was Never Adequately Remediated (but after 2000 it disappears from reports)

Oddly, consultant reports after 2000 lack any discussions of the ditch/storm sewer route or any other preferential pathways.[22] Dames and Moore’s 1999 and 2000 reports discounted soil vapor extraction as not feasible “due to the low permeabilities of the soil unit containing the highest concentration of PCE.”[23] Further, the consultants concluded that the previously planned excavation of the two hotspots was “unworkable” because of proximity to the 100-year old building, buried utilities, and because the most contaminated areas were in high levels of vehicle traffic.

Instead, they decided to inject an oxidizing reagent (BiOx) that would “react with the VOCs to chemically degrade them to harmless by-products.” Several bioremediation treatments were done in 1998 and 1999, and verification samples were taken in 1999. The March 21, 2000 report highlights a significant reduction in PCE concentration, but after the BiOx treatments, several verification samples still exceeded the 1000 µg/kg site specific RCL for PCE significantly in the northern hotspot area. The GP-9 area off the northernmost corner of the building had 329,000 ug/kg after remediation, 329 times Kipp’s site specific RCL and over 72,000 times the DNR’s soil to groundwater pathway RCL. PCE breakdown products were also present in significant quantities after the treatment, as were petroleum constituents.

Regardless, after the 2000 report, testing along the storm sewer route seems to have stopped. The ditch and storm sewer were no longer included in maps and diagrams in reports, and there are no further discussions of this major source area on the northern part of the site. During the 2000s, reports occasionally referred to a “former source area” or “previously defined source areas,” and a couple mentioned successful remediation there, but they did not identify or map where they were, or report any data from the storm sewer area.

All attention instead shifted to another source area (the former degreaser vent area) along the eastern side of the plant. This source was significant, and important given the proximity of homes, but the disappearance of any explicit discussions about the ditch and storm sewer was sudden and likely not an accident. Though the BiOx treatment had reduced the levels of contaminants in some particular hotspot areas, levels remained high in soils throughout that area and very high in the groundwater beneath it.[24]

Why did maps and discussions of the storm sewer source disappear? Interrogations in the class action lawsuit deposition of Jim Lenz, an engineer who worked at Kipp during this time, suggest that Kipp’s law firm Michael Best and Friedrich may have instructed the consultants to stop testing along the storm sewer route at some point.[25] Our speculation is that they did not want attention brought to the fact that contamination had likely spread offsite to the north and northeast—including onto and beneath homes, the city bike path property, and the Goodman Center site.

Zoom forward to 2014: Kipp northern hotspot still there (shhhh)…

Whatever caused the shift in focus away from the northern hotspot area and storm sewer over a decade ago, it worked. Government agency staff either never knew about it (because they didn’t bother to read older reports?), or they forgot about it, and/or they were pressured by Kipp and its law firm not to investigate or discuss it. Other than a couple of the larger Arcadis site investigation reports that review the consultant reports from the 1990s, thousands of pages of Arcadis consultant documents avoid explicitly discussing this critical hotspot area and preferential pathway. None of the numerous maps in Arcadis reports diagram the location of the storm sewer and the other storm drain pipes in the northern parking lot that connect to it. Even the reports outlining contamination and remediation of the raingarden, which received wastes from this storm sewer for decades, do not diagram or discuss it. In fact, the 2013 Site Investigation report, absurdly, says “The rain garden captures precipitation runoff from an adjoining bike path”—with no mention of the storm sewer, PCBs coming off Kipp’s parking lot, etc.[26]

As the earlier consultant reports did, Arcadis Site Investigation reports in 2012 and 2013 review the reports in the 1990s that discuss the storm sewer route, but then disingenuously give the impression that the BiOx remediation adequately reduced contaminant levels along the storm sewer route—e.g., the May 2012 report notes: “Enhanced biodegradation was implemented at the two former vapor degreaser vent areas and the former drainage ditch to address impacted soil. Post-remediation sampling indicated that the remedy was successful at decreasing VOC concentrations.” The 2013 report is even more misleading (and in fact, outright incorrect), noting that after remediation, “The data showed that concentrations of PCE were reduced to below the Site-specific residual contaminant level (RCL) of 1 mg/kg established for the remedy.” In fact, several borings throughout the northern area had contaminant levels hundreds of times above 1 mg/kg (1000 ug/kg) after remediation and thousands of times above the DNR’s soil to groundwater RCLs.

Given the above, it is not surprising that significant levels of contaminants remain in soils along the storm sewer route. Moreover, the groundwater in the northern part of the site, in the area where the ditch drainage pipe and older maps suggest that the storm sewer pipes come together, is where the most contaminated wells on the Kipp site are to this day—see the recent plume maps released by Arcadis. Groundwater well MW-13 has significantly higher levels of PCE than any of the other wells (see pg. 16 of above document). Most troubling, groundwater and soil data from the northern area suggest that the BiOx used in the remediation in the late 1990s, rather than breaking the PCE into “non-toxic byproducts” (as consultants claimed it would in their reports), broke PCE down into its far more toxic degradation byproducts trichloroethylene and vinyl chloride. The two northernmost shallow wells near the storm sewer route—MW-1 and MW-13—continue to have very high levels of PCE and its more toxic breakdown products TCE and vinyl chloride.

These results are highly relevant to considering risks to people at the Goodman Center property, much less than 100 feet away from Kipp. Yet, city/county and state public health agencies, and the DNR, have apparently given the stamp of approval to Goodman Center to build a splash pad only about 20-30 feet away from Kipp’s northern hotspot, without even requiring the testing they are required to do according to their “Cap Maintenance Plan.”

Why are they burying their heads in the sand (errr….toxic sludge)?

Questions, comments, corrections, additions? We welcome them–please send to info@mejo.us. THANKS!!

 

[1] Also see here

[2] In previous stories, we have raised questions about risks to children at the Goodman Community Center just north of Kipp and whether Goodman Center leaders followed DNR laws—see here and here and here. Government officials have repeatedly assured us and Goodman owners that these kids are not exposed to Kipp’s pollution, and therefore are not at risk.

[3] This data wasn’t released until April 2014, a full year after the samples were gathered

[4] In the last revisions of the DNR’s NR 700 regulations (which include vapor intrusion regulations and risk assessments), the state decided to adopt standard/screening levels for VOCs 10 times higher, or 10 times less protective of health, than the previous levels.

[5] Bike path probes are 7-8 feet deep and in the wet spring of 2013, groundwater was 8-10 feet down. Spring 2013 vapor probe data was only available for VP6, just inches above the groundwater, but the others (VP3-VP5) were under water.

[6] This raises further questions about who is actually legally responsible for this situation at this point, and any effects it might have on Goodman property.

[7] However, industrial screening levels for vapor intrusion are completely unprotective of workers based on scientific literature. These levels are highly influenced by industry and corporate lobbying and standards allow workers to be exposed to very high levels of contaminant vapors that are likely to cause health problems over the long term.

[8] Nehls-Lowe is the vapor intrusion expert at the DHS who is the lead person making decisions about vapor intrusion and public health risks at Madison-Kipp and other contaminated sites all over the state. In several conversations with him since 2011 about how DHS is assessing vapor intrusion risks at the site, he explained this step by step approach to assessing the potential for vapor intrusion off the Kipp property, moving slowly outward from Kipp. Mr. Nehls-Lowe does not think contaminated groundwater is a source of vapor intrusion in buildings anywhere off the Kipp site because clay soils act as barriers. He does not think that the EPA conceptual site model document (link below) is relevant to the Kipp situation. He does not believe, based on the existing data, that Kipp’s contamination has caused contaminant vapor intrusion into nearby homes—but that any vapors found in these homes are likely from household products.

[9] Also, see here and here

[10] Like EPA guidances, DNR vapor intrusion guidances say that testing the subslab of buildings is much more relevant to understanding risks to people in buildings than testing outdoor soil vapors.

[11] See EPA’s 2012 Conceptual Site Model Scenarios for the Vapor Intrusion Pathway.

[12] In recent years, the Goodman Center has done a couple spot soil tests for raised bed gardens along the bike path where soils were removed and replaced before remediation.

[13] Even if Kipp groundwater/soil contamination didn’t exist, children and other people at the Goodman site are exposed to the factory’s significant toxic air pollution (dioxins, chlorine, hydrochloric acid, PAHs, metals, particulates, oil mists from die casting, etc), much of which is uncontrolled and unmonitored. Also, industries formerly at the Goodman site are known to have used chlorinated compounds, including PCE and others. The site was remediated, but consultant documents in the DNR database show it has significant levels of contaminants, including lead, arsenic, and PAHs, remaining in soils below the two feet of new soil added after the contaminated soils were removed. Excavation for the splash pad will remove soils many feet below the 2 feet of “clean” soils placed on top of the contaminated soil. Yet as of May 2014, the DNR had no intention of asking Goodman to test the soils before, during, or after the splash pad construction, even though it is required by DNR’s regulations. Apparently nobody, even the DNR, minds that they didn’t follow these regulations last time they excavated, so this time around the laws don’t seem to matter either.

[14] When the vapor probes along the bike path were first proposed in early 2012, Michael Best attorney David Crass wrote the following in a March 14, 2012 email to Steve Tinker, Mark Giesfeldt, Patrick Stevens, Matt Moroney, Michael Schmoller, Henry Nehls-Lowe, Thomas Dawson, and Linda Hanefeld: “Installation of soil vapor probes is not recommended on the public bicycle path north of the Site for several reasons. First, the public bicycle path is located over former railroad tracks and a railroad right-of-way. It is highly likely that chemicals associated with these uses will be present and negatively interfere with the sampling results. Second, historic PCE use has been documented at the Goodman Center property, which will lead to questions regarding origins of any results and will infect any future decisionmaking regarding the results. Third, soil vapor sampling is only useful for evaluating (or estimating vapor intrusion) if the data are collected near a building. As Terry Evanson has said time and again, soil vapor data is of little consequence in the environment unless it is nearby buildings which can then be of concern. The collection of soil vapor samples in an open area will not provide any additional information on the potential migration of CVOCs into nearby buildings.”

[15] We located it in an open records request.

[16] Kipp consultant documents in the 1990s say this area wasn’t paved till 1995, though other documents say the parking lot was paved as early as the 1970s.

[17] Not surprisingly, many documents written by Kipp or their consultants throughout the years and recently have implied that this contamination was from past Kupfer operations (Kipp is trying to blame someone else).[18] So Kipp decided to pave this over after discovering really high levels of contaminants there? Also, other documents say the parking lot was paved much earlier, in the 70s. And, when was “several years ago”?

[19] Chemicals like toluene are known to degrade materials used in sewer drain pipes—causing contaminants to leach through pipes into the soils (and then groundwater) beneath the pipes.

[20] It is not clear what happened during this time, but presumably Kipp began to extract groundwater, which is why an extraction well (EW-1) appeared on their maps in 1997.

[21] EW-1 is on a map in the 1997 report, but no others after that, other than one mention in the 2005 report. The purpose and depth of this well remain unclear. Was Kipp extracting groundwater from this well from 1997 to 2005 or longer?

[22] DNR only recently started asking Kipp to assess preferential pathways—but so far we see no evidence that Kipp consultants are doing so even now (or if they are, the data is not being shared publicly).

[23] This is quite ironic given that currently (2014) soil vapor extraction is being used to remediate contaminated soils at Kipp, and consultants are implying that it is effective (though there are no data yet to verify this claim).

[24] Interestingly, there was one mention of the northern hotspot in the March 25, 2005 consultant report. Well EW-1 (the extraction well, off the northernmost tip of the plant in the remediated hotspot), was mentioned as having “elevated concentrations of CVOCs” (chlorinated volatile organic compounds). Ironically, it says “unlike other locations where the constituents are typically PCE an TCE, samples from EW-1 yielded elevated concentrations of other daughter products, including cis-1, 2, DCE and vinyl chloride, suggesting that natural degradation is occurring.”

[25] See page 220 of Jim Lenz deposition (pg. 55 of the PDF).

[26] Here’s the full quote from the Feb.12-Jan.13 Site Investigation: “There is a feature identified as a “rain garden” located adjacent to the northeast property line (Figure 2-3). The rain garden is a demonstration project completed by the city of Madison to illustrate how runoff of precipitation in an urban setting can be reduced through the use of vegetated areas. The rain garden captures precipitation runoff from an adjoining bike path. While not a habitat for sensitive species, this area was identified as part of the investigation scoping because rain gardens are designed to retain storm water and facilitate infiltration to groundwater. Care was taken during the investigation to reduce the potential for runoff contacting investigative activities.”

Digg thisShare on FacebookShare on Google+Email this to someonePrint this pageShare on RedditShare on TumblrTweet about this on TwitterPin on Pinterest