PATEA FREEZING WORKS – SITE INVESTIGATION & FIRE

Shane Reynolds Scientific Officer Taranaki Regional Council Phone: 06 765 7127 Email: [email protected]

Introduction

A freezing works was established on the present site, on the banks of an estuary in Patea, in 1881. It grew to become the life blood of the district, employing some 900 people at the height of processing. In the early 1980’s a national over processing capacity became apparent and with it’s reputation for industrial stoppages, Patea was one of the first works to close, ceasing operations in 1982. The effect on the town was devastating and as the years went by, the derelict and decaying works did little to improve the outlook for the residents that remained.

In the South Taranaki District Council’s [STDC] consultation process for community and district plans, the issue of the derelict works regularly came up in both public meetings and submissions. There was a general feeling of dissatisfaction within the community of Patea always being referred to as, “the town with the old freezing works”. This was the catalyst for STDC approaching the Taranaki Regional Council [Council] in April 2007 and requesting assistance to investigate potential contamination of the site. With a view to its ultimate remediation an hence to submit an application to the Contaminated Sites Remediation Fund [CSRF], administered by Ministry for the Environment [MfE]. This would become the largest contaminated site investigation ever undertaken by the Council.

Preliminary Site Inspection

A desktop study was conducted to gather as much information as possible about the site and a preliminary inspection was made. The following issues were identified as requiring further investigation and were the basis of a “scope of work”:

 Underground fuel storage tanks – remove three identified tanks and quantify any residual contamination.  Boiler fuel oil storage - bund and product lines to the boilerhouse  Fill areas – trench, log and sample  Groundwater - existing wells, install monitoring wells  Asbestos – options for disposal  Electrical and mechanical equipment – PCBs and tank sludges  Reservoirs – sample water and sludge  Fellmongery – paint mix tanks  Stormwater and surface water – sample to determine quality and impacts on estuary  Miscellaneous issues – coal tar seal, and dust/debris in buildings

- 1 - Based on the information gathered, a list of the possible contaminants associated with historical activities at the works was developed and is shown in the table below.

HAIL Activity Issue Potential contaminants 6 Building materials Roofing and cladding Asbestos Paint Lead Galvanising Zinc 36 Fuel storage Underground storage tanks and spills TPH, BTEX, PAH 41 Rail shunting Coal ash disposal Heavy metals, PAH Refuelling Hydrocarbons 46 Fellmongery Wool stripping, liming, neutralizing and pickling Chromium, manganese, copper, ammonia, sulphides, acids, sodium hydroxide, lime, formaldehyde, solvents, cyanide 49 Waste disposal Solids to fill areas, liquids to estuary Various Boilers and engine Coal ash/clinker disposal Heavy metals, PAH room Lagging Asbestos Electrical equipment Capacitors and transformers PCBs Cool stores Refrigeration plant Ammonia or ammonium hydroxide (if contacted water) HAIL: Hazardous Activities and Industries List TPH: total petroleum hydrocarbons BTEX: benzene, toluene, ethylene, xylenes PAH: polycyclic aromatic hydrocarbons PCBs: polychlorinated biphenyls

An overall assessment of contamination at the site was then possible and is summarised in the table below.

Contaminant Extent of Contamination Possible Exposure Routes TPH Site inspection confirmed the likely presence of 3 UST’s, age Soil/Dermal BTEX and condition unknown, extent of contamination difficult to Groundwater/Ecological risk gauge. PAH Asbestos Site inspection and previous sampling have confirmed Airborne/Inhalation extensive contamination. Heavy metals On site coal and incinerator ash disposal confirmed by past Soil PAH employee and identified at surface on site inspection, Surface water/Ecological risk extensive contamination of soil in fill areas is likely. Groundwater/Ecological risk

A report and CSRF application were submitted to MfE and funding was approved for costs associated with a detailed site investigation.

- 2 - The fire and detailed site investigation

Progress on the investigation was significantly delayed by a fire which broke out in the early hours of Waitangi Day, 6 February 2008. The fire raged through most of the main body of buildings on the site. The Freezing and Cold Storage Buildings (predominantly timber) in the centre of the site were completely destroyed, the Slaughter House to the east was significantly damaged and a number of buildings west of the Boiler House were also completely destroyed.

The fire meant that further site work had to be postponed until (i) the fire was extinguished, including actions undertaken by the Fire Service to gain access to hotspots buried beneath fallen debris; (ii) the resultant asbestos risk was dealt with to the satisfaction of the Department of Labour [DoL], to ensure that staff could continue working around the site in safety; and (iii) the scope and scale of the site investigations were reviewed to take account of the changed nature and characteristics of the site, particularly the likelihood that asbestos fibres had been released across the site due to shattering and fragmentation of asbestos sheeting and disintegration of asbestos lagging on pipework during the fire and consequent fire-fighting activities.

The objectives of the detailed investigation, were to:

 determine whether potential contaminants identified in the preliminary report were present,  establish the locations and extent of such contamination,  assess the potential for off-site effects, and  assist in the development of appropriate options for site remediation.

Completed works

Underground fuel storage tanks Contractors removed two tanks, the pits were inspected and sampled by an environmental consultant who recommended that a small amount of bedding material was also removed. At a third site no tank was found, although concrete slabs, pipework and tank straps indicated the location was correct. Excavation soon revealed what appeared to be extensive petrol residues in the soil, and approximately 100 m3 of soil was consequently removed. This was stockpiled on site along with the contaminated bedding material. It was regularly turned by excavator to aid in degradation of the hydrocarbons. Almost three weeks later, approximately 20 samples were collected from the surface of the stockpiled soil and composited for heavy metal TPH, BTEX and PAH analysis.

Boiler fuel storage The boiler fuel oil tanks had previously been removed, inspection of the tank bund showed no obvious signs of soil contamination. At one point the pipework to the boiler house was rusted through and oil had leaked onto a concrete pad. This was soaked up with bedding material from the tank pits and added to the stockpile of contaminated soil.

- 3 - Fill areas An 85 m long trench had been excavated parallel to the estuary to capture water run off from fighting the fire. Inspection of the trench spoil confirmed that extensive dumping of building rubble had occurred in this area. A band of dark material approximately 100 mm thick was present at a depth of around 1 m, along almost the entire length of the trench. Possibly deposits from the dumping of coal ash/clinker from the boilers, samples of this material were collected at 10 m intervals. Thirteen test pits were excavated on terraces in the south east and south west corners of the site and samples were collected from any areas of interest. Overall a total of 18 samples of fill material were collected for heavy metal and PAH analysis.

Groundwater Eight existing bores on the site were inspected, samples were collected from two and analysed for heavy metals. They were considered unsuitable for use as monitoring wells due to their depth, as potential groundwater contamination was more likely to be affecting shallow groundwater. The groundwater level in four, was surveyed in order to model the groundwater flow direction. Using this information, one background and four monitoring wells were installed and one sample was collected from each well to be analysed for heavy metals, PAH and TPH.

Asbestos A proposal was obtained from a specialist contractor for the removal of asbestos, demolition of buildings and reinstatement of the site. Seven samples of dust/debris were collected from the floor of safely accessible buildings across the site and analysed for asbestos.

Electrical and mechanical equipment The Council was alerted to the location of a number of transformers containing PCBs. These were removed and disposed of, following appropriate procedures, however some PCB equipment could not be safely removed for disposal and remains on site. A number of tanks and vessels were identified and visual inspection indicated that most have corroded to the point where it is reasonable to assume that they contain no liquid residues, however the presence of solid chemical residues cannot be discounted.

Reservoirs The main reservoir was supplied from the Patea River when the works was operational. It was pumped out of water during the fire, exposing a minimal amount of sludge in the bottom. The sludge would have consisted of silts and sediments from river water, so no samples were taken. A smaller reservoir partly beneath the south west corner of the Boiler House, contained water and a sample of the sediment on the bottom was collected for heavy metal, TPH and PAH analysis.

Fellmongery – paint mix area Paint is a chemical depilatory applied to sheep skins to dissolve the fibre root allowing for easy removal of the wool. The active ingredient is sodium sulphide which is dissolved in water and thickened with a hydrated lime slurry. These ingredients react to form sodium

- 4 - hydrosulphide and caustic soda in the final product which is highly alkaline. Two soil samples were collected from this area and tested for pH, to determine impacts on soil.

Stormwater and surface water All water drains south, from the site to the estuary and then into the Patea River. The only visible source of surface water is from springs originating at the base of cliffs on the eastern boundary. Water sampling was timed to coincide with the irrigation operation, where water from the Patea River was being irrigated over fire damaged areas of the site to minimise the risk of asbestos becoming airborne. This resulted in a large volume of run off and made it possible to sample from a number of locations which otherwise may not have been discharging due to the extremely dry weather conditions prevailing at the time. In addition it would ensure an accurate assessment of any offsite effects from the irrigation operation. A total of 10 samples were collected from drains and surface flow, including background samples from the Patea River and analysed for heavy metals, TPH and PAH.

Miscellaneous issues Nine samples of road seal from across the site were collected to determine whether coal tar was present in the tar/bitumen chip binder. These were subject to PAH analysis. Seven dust/debris samples were collected from the floor of selected buildings and analysed for heavy metals and asbestos, to indicate the level of risk associated with any cleanup works.

Investigation results and discussion

Soil Results were assessed against guideline values for commercial/industrial land use. No results exceeded the values for heavy metals, from the Guideline on the Investigation Levels for Soil and Groundwater (NEPC 1999). The results for PAH, TPH and BTEX were conservatively assessed against values in the Guidelines for Assessing and Managing Petroleum Hydrocarbon Contaminated Sites in New Zealand (MfE, 1999) for the soil type SAND, which has the most stringent values of the different soil types. No results exceeded these values, it should be noted that many samples were actually fill material and not soil. The samples collected from the paint mixing area were tested for pH and indicate that the soil is slightly alkaline.

Water Both surface water and groundwater (assumed) discharge from the site to the estuary, there is not a significant risk to human health and so the results were compared to the Australian & NZ Guidelines for Fresh & Marine Water Quality (ANZECC, 2000 ), trigger values for 95% species protection. This is a conservative approach as there is a large dilution factor/regular flushing of the receiving environment from tidal movement. In general, most groundwater results complied with the guideline values (3 slight exceedances). The two existing bores that were sampled met the criteria for heavy metals in the Drinking Water Standards for New Zealand (MoH, 2005). All surface water samples exceeded the guideline values for at least one metal. It is expected that the concentrations of these metals should reduce once the site has been cleaned up and they indicate minimal risk to the aquatic environment of the estuary.

- 5 - It was noted that background samples from the Patea River had elevated concentrations of some metals, most likely due to the high level of suspended solids in the river.

Other The results for heavy metal analysis on the dust/debris floor samples, although not of soil, were conservatively assessed against the commercial/industrial land use, health investigation levels for soil, in the Guideline on the Investigation Levels for Soil and Groundwater (NEPC, 1999). A high copper result was possibly due to the presence of some copper wire in the sample from outside the switch room. Elevated levels of lead are undoubtedly due to the amount of flaked-off paint that was present in many of the samples. It is expected that these materials (both the dusts/debris, and their sources) would be disposed of as part of any cleanup of the on-site materials. If correct procedures are followed then workers’ exposure during cleanup should be low and the detected concentrations should be no cause for concern. However, this material will need to be disposed of appropriately.

The PAH results for samples of road seal were exceptionally low (even for bitumen based seal) and rule out the presence of coal tar. Results for the Boiler House Reservoir sample are well below the guideline values for sand at less than 1m in the Guidelines for Assessing and Managing Petroleum Hydrocarbon Contaminated Sites in New Zealand (MfE, 1999). Consequently, if this reservoir was drained the sludge would be acceptable for disposal to land on the site.

Asbestos Of the seven dust/debris floor samples collected, no asbestos was detected in two and chrysotile asbestos was detected in five samples. A visual inspection of the Boiler House had already revealed the presence of amosite asbestos (typically used in older lagging) confirmed by these results. Amosite asbestos was also detected in the Fellmongery Paint Mixing room, which indicates that asbestos lagging may also have been used in this area. (Note: most if not all, of the lagging on the site consists of rockwool, a glass-based insulation material containing no asbestos.) The results indicate that the distribution of asbestos is widespread across the site buildings.

Actions on site

Irrigation of the site Immediately following the fire and departure of the Fire Service, irrigation of the fire affected areas was undertaken to extinguish any residual ‘hotspots’ and then to minimise the risk of asbestos becoming airborne. To achieve this, water was pumped from the Patea River to a rotating irrigator, which was regularly moved to ensure that all fire damaged areas remained damp. This operation continued for approximately two weeks until a polymer binder was applied, providing an encapsulating membrane which is preventing any asbestos from being released into the air. An interceptor trench approximately 85m long was excavated on the downslope side of the site, adjacent to the estuary, to collect surface run-off. A bunded collection pond was also constructed to capture run-off discharging from the site through the main stormwater drain.

- 6 - Cleanup of McColls The fire completely destroyed all except one of the buildings on Noel McColl’s property (the area immediately west of the main factory). Only the Storage building remained standing, although severely damaged. The fire damage resulted in widespread asbestos contamination of the area. Following the gaining of the appropriate approvals from DoL, a cleanup was conducted by Noel McColl under the supervision of a certified asbestos expert. This involved the demolition of severely fire damaged buildings, the removal of steel for recycling, and scraping up of all debris (capturing all asbestos), including material deposited on carriage ways in the area. This material was placed in the Main Reservoir, which was pumped out of water during the fire. The Council believes the reservoir to be impervious and structurally sound (based on its observed water retention) and therefore a suitable structure for long term containment.

Air monitoring The Council established 12 sites around the works and within the town for air quality monitoring, to establish whether there was any risk from airborne asbestos fibres. Samples were collected by drawing a measured quantity of air, through a membrane filter by means of a battery powered sampling pump. The filters were transformed from an opaque membrane into a transparent optically homogenous specimen. The fibres were then sized and counted using a phase contrast microscope. Results are then expressed as fibres per milliliter of air [f/ml], calculated from the number of fibres on the filter and the measured volume of air sampled. The initial sampling run was conducted during application of the polymer. To date a further six sampling runs have been completed and all results are for an asbestos concentration of <0.01 f/ml. This confirms the efficacy of the polymer as an interim control measure for protecting the health of onsite personnel and residents alike. Equipment and analysis was provided by Capital Environmental Services [CES] in Lower Hutt.

Ash sampling Twelve samples of ash were collected in safely accessible areas where the fire had destroyed the Freezing and Cold Storage Buildings, and analysed for asbestos content. In six of the samples no asbestos was detected, the remaining six had varying levels of chrysotile asbestos (up to 20%). Chrysotile asbestos forms approximately 20% content in the fibre cement cladding used extensively throughout the site. No amosite asbestos was detected in any of the samples.

Application of polymer Tests of air and ash samples in the first couple of days following the fire indicated there had been no asbestos contamination in residential areas, but there remained the potential threat of release of asbestos from the substantial debris and ash left after the fire, now even more exposed to the elements than previously. Much of the intact fibrolite had been either shattered by the fire’s heat or damaged during the fire-fighting efforts. After investigating the options, the Council settled on a non-toxic acrylic resin called AW95 supplied by RST Environmental Solutions Ltd, to bind and settle dust particles to help prevent any possibility of airborne asbestos contamination.

- 7 - There were a number of challenges to overcome, the buildings are expansive and fire damage meant that personnel could not enter them or access across roof framing. Fire debris and ash lay in deep piles around the site, while cladding and wrapping were still draped over much of the building skeleton. It was necessary that the fire was completely extinguished, with no deep-seated hot spots remaining. A digger was used to drag scattered debris back to the building line, and to expose the interior of the larger piles. The product was applied as a 10% solution by a mobile, truck mounted, water cannon with a range of some 40-50 metres. Operators also used a hose reel to spray through doorways and windows, ensuring maximum possible coverage of fire damaged areas. The operation took one full day and was managed in accordance with DoL guidelines for asbestos sites.

The interceptor trench prevented run-off into the adjacent tidal estuary. Air quality monitoring conducted during and after the application operation confirmed that there was no release of asbestos fibres at any time. The polymer remained sticky to the touch for a couple of weeks after application, adhering to hard surfaces and soaking into the fine material and forming a thick spongy layer that has bound ash and other fire debris. The integrity of this membrane is inspected on a regular basis and further polymer applied where necessary. The polymer degrades upon exposure to sunlight and is considered to have a life expectancy of less than twelve months.

Actions in town

Evacuation of residents A decision was made by the Fire Service and Public Health to evacuate up to 300 residents because of concerns that there may be asbestos fibres in the smoke and dust from the fire, which the southeasterly wind was blowing across the town. Transport was offered to those that required it, to an emergency welfare centre set up in Hawera, 25 km away. Later that evening, residents were advised that it was safe to return home as the fire was under control and smoke and ash were no longer affecting the town.

Dust Sampling Public Health staff with assistance from CES, coordinated the collection of 22 dust samples and 3 air samples from throughout the area. Dust samples were also collected from the Marae and Red Cross volunteer overalls in the days immediately following the fire. Dust samples were analysed by CES using polarised light microscopy and no asbestos was detected in dust or air samples. A further five dust samples collected from inside a Patea house, organised by an insurance company, and once more no asbestos was detected.

Washing down/inspections STDC organised the Fire Service to wash down the outside of buildings affected by the fallout of dust and ash from the fire. This was carried out over the weekend following the fire. The interior of houses close to the fire were inspected for sanitary conditions, the only issue was the smoke smell and not fire debris.

- 8 - Overall conclusions

Water discharging from the site has concentrations of metals which exceed the guideline values for 95% protection of aquatic ecosystems. However this is a conservative assessment, as the guidelines relate to receiving water and not concentrations in discharges, which in any case should reduce once the site has been cleaned up. Furthermore, a survey of the estuary mudflats bordering the site by a Council marine ecologist has shown healthy numbers of plants, birds, crabs and snails typically found in this type of environment. Discharges from the site are not considered to be having an adverse effect on the estuary or the Patea River. With the exception of asbestos, the concentrations of all on site contaminants are below the relevant guideline values. The risk from asbestos is currently minimised by the restriction of access to the site and the polymer binder applied to fire damaged areas. Some electrical equipment containing PCBs remain on the site. The Taranaki Regional Council considers that any further work on the site focus on the safe removal and disposal of asbestos and electrical equipment containing PCBs.

The Civil Defence Emergency Management Group, emergency services, Public Health Unit and a number of other agencies were involved in the initial response to the fire and subsequent actions in the days that followed. Overall, it was felt that the event had been well managed but as always, there was also an opportunity to make some improvements and be better equipped to respond to future events.

What next ?

On 4 July 2008 the Government announced that it had allocated $1.5 million towards the cost of cleaning up the site. STDC are currently working to resolve issues over the ownership of part of the site. A consultative group has decided upon a “cleanfields” remediation option which will involve the removal of all contaminated material. It is hoped that cleanup works will begin prior to Christmas this year.

References Taranaki Regional Council, 2007, Patea Freezing Works – Preliminary Site Inspection Report Taranaki Regional Council, 2008, Patea Freezing Works – Detailed Site Investigation Report

- 9 -