20190397_Calcium Magnesium Acetate De-Icing Agent - New Zealand Transport Agency_20180810

Calcium Magnesium Acetate
De-icing Agent
A review of environmental effects
monitoring in New Zealand 1998-2009

Report prepared for the New Zea/and Transport Agency by
Opus Internationa/ Consultants Ltd
Calcium Magnesium Acetate De-
icing Agent:
A review of environmental effects monitoring in
New Zealand 1998-2009
Prepared By
John Turner
Principal Ecologist
�2i
�
Reviewed By ------------------------------
Keith Hamill
Principal Environmental Scientist
Date: 16th April 2010
R ef e re n ce : 2-64500. 80
Status: Final

CMA De-icing Agent: A review of environmental effects monitoring in New Zealand 1998-2009
Acknowledgements
This report was funded by the New Zealand Transport Agency. Thanks to Alan Burkett
(NZTA) for the original idea of preparing this review and support during its preparation.
Thanks also to Dr Mike Scarsbrook (formerly of Leader of the National Centre for Water
Resources NIWA) for inputs to the scoping of the report; Aslan Wright-Stow (NIWA) for
assistance during its preparation, and Simon Beale of Montgomery Watson for providing
data and commentary on results for the South Island.
16'h April 2010

CMA De-icing Agent: A review of environmental effects monitoring in New Zealand 1998-2009
Executive summary
Icy conditions are experienced on many parts of the Central North Island and South Island
State Highway networks each winter. Icy road surfaces present hazardous conditions for
motorists and are a factor in many accidents which in some cases leads to serious injury or
death. In addition, a number of key sections of highway in the North and South Island's are
also regularly closed because of snow and ice. Traditionally grit has been used in New
Zealand to combat these conditions. However, grit is less effective than salt and other
chemical treatments used overseas. Furthermore, grit can itself introduce a hazard to
motorists, particularly when cornering. As a consequence, in 1996 the New Zealand
Transport Agency (then Transit New Zealand) initiated investigations to find a more
effective tool to assist maintaining highways open and safe for use.
After an extensive review of the overseas literature Calcium Magnesium Acetate (CMA)
was selected as a potential chemical alternative to grit. Overseas studies showed that CMA
was an effective ice treatment, had a low risk of causing significant adverse effects on the
environment, as well as being significantly less corrosive. In October of 1996, NZTA
submitted a resource consent application to permit the first trial use of CMA at selected ice
black spots along the Desert Road, which traverses North Island's Central Plateau. The
success of these trials both in terms of providing effective ice treatment and demonstrating
no adverse environmental effects has eventually led to CMA being consented for used on
many parts of the State Highway through Central North Island and the entire South Island
State Highway Network.
Despite overseas studies indicating a low risk of CMA causing significant adverse
environmental effects, some residual concerns regarding the potential environmental
impacts of CMA in New Zealand remained when its introduction was first being proposed.
In part, this was due to the fact that most predictions regarding the potential environmental
effects of CMA were based on experimental data derived from laboratory testing and field
trials, rather than analysis of data collected from monitoring actual operational situations.
There were also concerns that the effects of CMA could vary between locations given the
variation in climate, vegetation, soils and the characteristics of aquatic environments. As a
consequence of these concerns, CMA has now been subjected to extensive environmental
testing and monitoring in New Zealand since 1998.
Over a period of more than 10 years monitoring of the possible effects of CMA on soils,
vegetation, streams and lakes, in many parts of New Zealand (Waikato Region, Hawke's
Bay Region, Wanganui Region, Canterbury Region, West Coast Region, Otage Region,
Nelson/Tasman/Marlborough Region) has found no adverse environmental effects that
could be attributed to its use as a de-icing agent on the State Highway network. The
monitoring throughout New Zealand has included:
• dissolved oxygen monitoring on 15 steams for periods of up to 5 consecutive years,
• aquatic invertebrate populations in 19 streams for periods of up to 11 consecutive
years,
• benthic algae and heterotrophic growth in 18 streams for periods of up to 11
consecutive years,
16'h April 2010 II

CMA De-icing Agent: A review of environmental effects monitoring in New Zealand 1998-2009
• roadside vegetation health at 21 locations adjacent to sections of highway receiving
CMA in stormwater runoff and potentially affected by spray drift, for up to 8 consecutive
years,
• vegetation species composition at 26 locations adjacent to sections of highway
receiving CMA in stormwater runoff and potentially affected by spray drift, for up to 10
consecutive years,
• soil properties at 27 locations adjacent to sections of highway receiving CMA in
stormwater runoff and potentially affected by spray drift, for up to 10 consecutive years.
These findings confirm the conclusions of McFarland and O'Reilly (1992) who predicted,
after reviewing available laboratory and field trials undertaken up to that time, that "negative
environmental and toxicological impacts are likely to be insignificant in the vast majority of
CMA applications".
Based on a search of overseas published sources it is apparent that direct field monitoring
of environmental effects of CMA under fully operational conditions has been undertaken
more in New Zealand than anywhere else in the World. This monitoring provides a
substantial body of evidence, supported by the overseas laboratory and field trials, that the
risk of significant adverse effects resulting from normal operational use of CMA to roadside
vegetation health, species composition and soil health is negligible. Furthermore, none of
the monitoring undertaken in New Zealand has found any discernible adverse effect on
aquatic life or fauna populations found in various types of waterbodies, and no evidence
has ever been found that it causes a proliferation of benthic algae or heterotrophic growths.
It is therefore also reasonable to conclude that the risk posed by CMA to aquatic
environments and aquatic life, from normal operational use, is negligible for most, if not all,
receiving environments adjacent to New Zealands' State Highway network.
The highest risk scenario of CMA causing significant adverse environmental effects is a
situation where substantial quantities of CMA enter a small enclosed waterbody. However,
experience has shown, from the extensive use of CMA around the New Zealand State
Highway networks and associated environmental assessments, that the incidence of small
waterbodies occurring immediately adjacent to the highway is extremely low. Furthermore,
no situations have been identified where a small waterbody is likely to receive sufficient
CMA to cause a significant adverse effect. Therefore, whilst there is a theoretical risk of
adverse effects on small waterbodies from substantial inputs of CMA, the circumstances
where this actually occurs in a New Zealand context have not been found, despite the
extensive use of CMA over many parts of the State Highway network.
Overall, the environmental monitoring undertaken in New Zealand has confirmed the
original expectations that CMA would prove an effective and environmentally benign tool to
assist with winter maintenance of the State Highway Network. When considering the overall
merits of CMA, the negligible risk posed to most, if not all, receiving environments, needs to
be considered in the context of the substantial environmental advantages CMA has over
sodium chloride (the most extensively used de-icing chemical used overseas), and the
significant economic and safety benefits its use brings to the New Zealand travelling public.
16'h April 2010 III

CMA De-icing Agent: A review of environmental effects monitoring in New Zealand 1998-2009
Contents
Acknowledgements....................................................................................................................... i
Executivesummary ...................................................................................................................... ii
1 Introduction ..........................................................................................................................1
1.1 Background ..................................................................................................................1
1.2 Purpose and scope of report .........................................................................................1
2 Ice and snow on the State Highway Network and impact on motorists ...........................2
2.1 Ice and snow ................................................................................................................ 2
2.2 Road closures ...............................................................................................................3
2.3 Ice related accidents .....................................................................................................3
3 The search for more effective highway winter management tools ...................................4
3.1 Limitations of winter management tools in New Zealand before CMA ...........................4
3.2 Historic international approach to treatment of ice and snow ........................................ 4
3.3 The international search for an alternative to salt .......................................................... 4
3.4 The introduction of CMA into the winter maintenance program in New Zealand............5
4 Attributes and methods application of CMA ......................................................................6
4.1 Properties ..................................................................................................................... 6
4.2 De-icing mechanism ..................................................................................................... 6
4.3 Methods of application .................................................................................................. 7
4.4 Costs ............................................................................................................................ 7
5 Overseas research into the environmental and health effects of CMA ............................8
5.1 Overseas research in the mid-1980's and early 1990's ................................................. 8
5.2 Overseas studies 1990's to the present time ................................................................ 9
5.3 Potential environmental effects based overseas research ..........................................10
5.4 Human health effects ..................................................................................................12
6 Early modelling and testing in New Zealand ....................................................................12
6.1 Perceived risks to the New Zealand environment .......................................................12
6.2 Modelling and testing effects on BOD for Desert Road streams ..................................13
6.3 Monitoring on the Desert Road 1998-2002 .................................................................14
7 Operational use of CMA on the highway network in New Zealand .................................15
7.1 Regions where CMA is consented for use as a de-icing agent ....................................15
7.2 Extent of network where treatment with CMA is permitted ..........................................15
7.3 Basis on which applications have been processed .....................................................15
7.4 Environmental monitoring requirements for CMA ........................................................15
7.5 Consent duration ........................................................................................................16
16'h April 2010 IV

CMA De-icing Agent: A review of environmental effects monitoring in New Zealand 1998-2009
8 Results of Environmental Monitoring in New Zealand 1998-2008 .................................17
8.1 Environmental aspects monitored ...............................................................................17
8.2 Dissolved oxygen ........................................................................................................17
8.3 Aquatic invertebrates ..................................................................................................19
8.4 Benthic algae and heterotrophic growths .................................................................... 21
8.5 Vegetation health ........................................................................................................22
8.6 Vegetation species composition ..................................................................................23
8.7 Soils ............................................................................................................................24
8.8 Ecotoxicology ............................................................................................................. 26
9 Risk assessment ................................................................................................................27
9.1 Introduction ................................................................................................................. 27
9.2 Dissolved oxygen ........................................................................................................29
9.3 Aquatic fauna .............................................................................................................. 30
9.4 Benthic algae and heterotrophic growths .................................................................... 30
9.5 Vegetation health ........................................................................................................32
9.6 Vegetation species composition ..................................................................................32
9.7 Soils ............................................................................................................................33
9.8 Small waterbodies ......................................................................................................34
10 Conclusions ....................................................................................................................... 37
Tables.......................................................................................................................................... 38
References.................................................................................................................................. 51
16'h April 2010 V

CMA De-icing Agent: A review of environmental effects monitoring in New Zealand 1998-2009
1 Introduction
1.1 Background
Icy conditions are experienced on many parts of the Central North Island and South Island
State Highway networks each winter. Icy road surfaces present hazardous conditions for
motorists and are a factor in many accidents which in some cases leads to serious injury or
death. In addition, a number of key sections of highway in the North and South Island's are
also regularly closed because of snow and ice. Traditionally grit has been used in New
Zealand to combat these conditions. Common salt (sodium chloride) was also used until the
early 1980's, when public concerns about vehicle corrosion led to it being discontinued
(Jamieson & Dravitzki, 2006), after which grit became the main treatment agent. However,
grit is less effective than salt and other chemical treatments. Furthermore, grit can itself
introduce a hazard to motorists, particularly when cornering. A proposal to reintroduce salt
as a de-icing agent was rejected by road user and environmental groups (Burkett & Gurr,
2004). As a consequence, in 1996 the New Zealand Transport Agency (then Transit New
Zealand) initiated investigations to find a more effective tool to assist maintaining highways
open and safe for use.
After an extensive review of the overseas literature Calcium Magnesium Acetate (CMA)
was selected as a potential chemical alternative. Overseas studies showed that CMA was
an effective ice treatment, had a low risk of causing significant adverse effects on the
environment, as well as being less corrosive. In October of 1996, NZTA submitted a
resource consent application to permit the first trial use of CMA at selected ice black spots
on the Desert Road, which traverses North Island's Central Plateau. The success of these
trials both in terms of providing effective ice treatment and demonstrating no adverse
environmental effects has eventually led to CMA being used on many parts of the State
Highway through Central North Island and in all South Island regions.
Despite overseas studies indicating a low risk of CMA causing significant adverse
environmental effects, some residual concerns regarding the potential environmental
impacts of CMA in New Zealand remained. In part, this was due to the fact that most
predictions regarding the potential environmental effects of CMA were based on
experimental data derived from laboratory testing and field trials, rather than analysis of
data collected from monitoring actual operational situations. There were also concerns that
the effects of CMA could vary between locations given the variation in climate, vegetation,
soils and the characteristics of aquatic environments. As a consequence of these concerns,
CMA has now been subjected to extensive environmental testing and monitoring in many
parts of New Zealand since 1998.
1.2 Purpose and scope of report
The primary purposes of this report are:
• To summarise the results of environmental testing and monitoring connected with the
use of Calcium Magnesium Acetate (CMA) as a de-icing agent on the New Zealand
State Highway Network between 1998 and 2009, and;
16'h April 2010 1

CMA De-icing Agent: A review of environmental effects monitoring in New Zealand 1998-2009
• To review the risk posed by CMA to the environment based on the results of this testing
and monitoring, as well as international experience.
This review draws on the results of numerous tests and monitoring programs undertaken in
North and South Island New Zealand, as well as international literature.
2 Ice and snow on the State Highway Network and impact on
motorists
2.1 Ice and snow
Ice and snow result in hazardous driving conditions on many sections of the State Highway
Network in North and South Island New Zealand. In the North Island, snow is generally
confined to the Central Plateau and other high elevation sections of highway, while ice may
occur at much lower elevations, particularly sections of highway that remain in heavy shade
for much of the day. In the South Island the formation of ice and occurrence of snow is
more widespread. Due to the higher latitude, winter temperatures in south Island are
generally lower, resulting in snow, as well as ice, at lower altitudes, although snow only
tends to persist at low altitudes for short periods.
The most heavily impacted section of the North Island State Highway Network is the Desert
Road section of SH1 which traverses the Tongariro National Park at an altitude of between
450m and 1074m above sea level. Road closures and ice related accidents are a significant
problem during winter months and present a persistent annual maintenance challenge.
Similarly, sections of State Highway in the South Island are affected by road closures due
to snow each winter and accidents related to both ice and snow are frequent. Furthermore,
the alpine nature of the Central Plateau and a number of parts of the South Island can
result in extremely rapid climatic deterioration. Sometimes these sudden changes result in
unexpected snow falls occurring before road closures can be effected, and this inevitably
result in occasional stranding of motorists.
There are two operational problems that stem from these extreme winter climatic
conditions. These are:
Icing - caused by freezing water, not only on the road surface, but from moisture in the
surrounding air. This problem may be exacerbated in gullies as a consequence of
shadowing and cold air ponding.
Snow — snow and snow drifts render the highway impassable to motorists until it is cleared
by maintenance personnel. While mechanical equipment can in most cases effectively clear
snow the need to use heavy machinery for this purpose often causes compaction of the
lower layers of a snow drift. This can result in the formation of thin layers of ice as snow is
changed to ice under the action of pressure and compaction, chemical action, fluctuations
of temperature and refreezing of inelt water within the snow matrix. This process can lead
to build up of a considerable thickness of ice on the road surface (i.e. up to 2 cm thick)
which sticks to the road surface and can be difficult to remove. Motor vehicles also add to
the compaction effects too.
16'h April 2010 2

CMA De-icing Agent: A review of environmental effects monitoring in New Zealand 1998-2009
2.2 Road closures
Snow and ice result in the closure of a number of New Zealand's State Highways each
year. As the highest altitude road in North Island New Zealand, the Desert Road provides a
particularly good example of this problem. A strategic corridor evaluation prepared for the
section of SH1 between Taupo and the Desert Road summit reports that this section of the
state highway network is a route of national importance, due to the large volumes of freight
traffic and passenger cars using it as a link between Auckland, Waikato and Bay of Plenty
regions to the north of the Central Plateau, with the Wellington and Manawatu regions to
the south. It is also a major route used by tourists (Opus International Consultants, 2009).
The Desert Road is closed each winter due to snow and ice. Table 1 summarises road
closure data between 1993 and 2008. The average number of hours the road was closed
each year during this period was approximately 140 and the average duration of each
individual closure was approximately 17 hours.
The closure of the Desert Road results in significant inconvenience to road users who have
high expectations with regard to availability for use of such a strategically important route.
As well as disruption for users, closure also results in a significant cost to the New Zealand
economy. The strategic corridor evaluation (Opus International Consultants, 2009)
estimates the economic cost of closure of the Desert Road at $20,120 per hour which
equates to an average annual cost resulting from closures of the Desert Road of
approximately $2.78m. Interestingly, further analysis of the data in Table 1 indicates that
the average length of road closure since 1998 when CMA was first used on the Desert
Road has been 12 hours. Given that when the previous 5 years data is also included the
average increases to 17 hours, this suggests that CMA may have contributed to a
significant reduction in the duration of individual road closures.
2.3 Ice related accidents
The presence of ice and snow is a significant contributing factor in many winter accidents in
New Zealand. Many of these accidents result in serious injury and death. Table 2 provides
a summary of crashes in New Zealand between 1980 and 2008 where ice and snow where
a contributing factor (Source: NZTA's Crash Analysis System — CAS). Figures for the
Desert Road over the same period are also given. Nationally, over the 29 year period
shown, there have been a total of 3345 people involved in ice/snow related accidents
resulting 69 deaths and 363 serious injuries. As well as the great suffering for those
involved and their families, the economic cost is estimated at $779m, an average annual
cost of nearly $29m. These statistics provide significant impetus to the search to develop
new tools to improve the efficiency and safety of the State Highway network during the
winter period.
16'h April 2010 3

CMA De-icing Agent: A review of environmental effects monitoring in New Zealand 1998-2009
3 The search for more effective highway winter management tools
3.1 Limitations of winter management tools in New Zealand before CMA
Since the early 1980's when the use of salt was abandoned, and prior to the introduction of
CMA in 1998, the maintenance practices employed to clear snow and ice from the
highways were limited to the use of inechanical methods such as snowploughs, graders,
front end loaders, rotary brooms and the spreading of grit (fine aggregate) on areas of ice to
improve traction. Grit and sand have limited effectiveness as they don't prevent ice
formation and they tend to become bound into the ice over time. These methods were of
limited use for actually clearing of ice and often the length of time that a section of highway
was closed was usually dictated by the time that ice took to thaw by natural processes.
Furthermore, grit build-up has been a factor in a number of crashes which is particularly
problematic when it occurs on corners. It also raises concerns from motorists regarding
damage to paintwork and windscreens, and impacts on the road itself, covering and
abrading road marking paint and reducing the effectiveness of raised pavement markers
(Jamieson & Dravitzki, 2006).
Snow falls can usually be quickly removed and roads usually re-opened in a short time
once snow stops falling. Ice however is a more difficult problem to deal with. For very light
ice, grit or sand applied to the road will provide improved traction and assist the break-up of
the ice. However, with temperatures below freezing point the surface water and grit mixture
will soon re-freeze once traffic numbers decrease, such as at night.
3.2 Historic international approach to treatment of ice and snow
Ice and snow is a problem for road networks in many high latitude and high altitude regions
of the World. The problem is particularly extensive in the northern hemisphere where
extensive parts of road networks occur in high latitude areas. Internationally, history
indicates a trend toward the use of sodium salts supplemented by various mechanical
means for the control of ice and snow (Works Consultancy Service Ltd., 1996). For
example, salt (sodium chloride) has been used in the US since the 1930's, due to it being
cheap and abundant (Frischel, 2001). The extensive use of salt has permitted many roads
to remain open during winter.
However, since the 1960's numerous studies have been conducted on the environmental
and human health effects of de-icers, as well as their corrosion properties (Frischel, 2001).
Sodium choride, the most widely used road salt internationally, has been shown to have
been associated with a variety of negative environmental effects, and is highly corrosive
(Works Consultancy Service Ltd., 1996).
3.3 The international search for an alternative to salt
Concerns about the negative environmental effects of sodium chloride and corrosive
properties led to the US Federal Highway Administration to initiate research to find a
suitable alternative (Horner, 1988) in the late 1970's. In the subsequent decade a number
of studies were undertaken to assess the potential environmental effects (Horner, 1988;
Winters et. al., 1985). Although indicating some caution and recommending safeguards
around its' use, the results of these studies indicated that CMA was likely to pose much less
16'h April 2010 4

CMA De-icing Agent: A review of environmental effects monitoring in New Zealand 1998-2009
risk to the environment than sodium chloride. Furthermore, it was considered much less
corrosive, and was essentially non-toxic to human and mammalian life, although potentially
a mild eye and skin irritant.
3.4 The introduction of CMA into the winter maintenance program in New Zealand
3.4.1 Initial review of options
As a result of the limitations and shortcomings of existing winter management tools for
combating ice and snow NZTA (then Transit New Zealand) initiated investigations to find a
more effective tool to assist maintaining highways open and safe for use in 1996. As part of
the process of identifying a suitable de-icing alternative a comprehensive review of
available options was undertaken. The comparison of alternatives is described in detail in
the Assessment of Environmental Effects (Works Consultancy Services, 1996) that
accompanied the application for resource consent for the first trial use of CMA on the
Desert Road.
The principal chemical options reviewed were:
• Sodium Chloride
• Calcium Chloride
• Damp (Prewetted) Salt
• Calcium Magnesium Acetate
• Urea
• Verglimit
The principal mechanical options reviewed were:
• Abrasives
• Graders
• Rotary Brooms
• Snow Ploughs
• Snow Blowers
• Pavement Heating
3.4.2 Evaluation of options
The evaluation of the alternatives listed above was carried out in two parts. Firstly an initial
scoping exercise was undertaken to identify the best options for detailed evaluation, based
on the following criteria:
• dollar cost of application;
• effects on the natural environment;
• corrosion potential for vehicle and road infrastructure; and
• effectiveness at achieving highway clearance.
16'h April 2010 �J

CMA De-icing Agent: A review of environmental effects monitoring in New Zealand 1998-2009
Based on these criteria sodium chloride (salt) and calcium magnesium acetate (CMA) were
selected for more detailed evaluation. The full analysis of options is given in Works
Consultancy Service Ltd. (1996).
3.4.3 Recommended de-icing option
The preferred de-icing option that resulted from the more detailed analysis of the two
alternatives was CMA. A decision was made to proceed to seek resource consent for the
trial use of CMA at five ice black spots along the Desert Road, for the following reasons:
• CMA was an effective de-icer and had the potential to allow for the easy removal and
clearance of ice from the highway;
• CMA could be used pro-actively as well as re-actively;
• CMA, if applied within certain parameters was likely to avoid and minimise adverse
environmental effects;
• The Benefit/Cost ratio for CMA met acceptable highway maintenance funding criteria
for TNZ;
• CMA had significantly less corrosion potential than sodium chloride with a reported
potential equivalent to tap water;
• Sodium chloride had the lowest dollar cost of application of all alternatives and
therefore was easily fundable. However the overseas literature indicated that it could
have significant adverse environmental effects.
• Conversely CMA had a high application cost but potentially less adverse environmental
effects on water quality and fauna and flora than any of the salts considered.
The trial use of CMA at the 5 selected ice black spots was subsequently consented and
undertaken on the Desert Road between 1998 and 2002, subject to conditions requiring
further environmental testing and monitoring. These trials eventually resulted in the full
operational use of CMA being permitted along the entire length of the Desert Road within
the Waikato Region from 2003, althougth environmental monitoring has continued up to the
present time.
4 Attributes and methods application of CMA
4.1 Properties
Calcium Magnesium Acetate (CMA) is a formulation of dolomite lime and acetic acid
presented in a granular form. The commercial structure of the product is as follows:
CMA = CaxMgy(C2H302)2�X+y� where x= 3 to 4 and y= 6 to 7
Typically a de-icing formulation of CMA is produced in a 3:7 ratio of calcium:magnesium
and is approximately 91 % CMA on a dry basis.
4.2 De-icing mechanism
CMA can be applied to the highway surface in solid (round granules) or in a liquid form
(typically a 25% solution). In a solid form, CMA is highly hydroscopic and absorbs moisture
16'h April 2010 6

CMA De-icing Agent: A review of environmental effects monitoring in New Zealand 1998-2009
from the surrounding air and pavement surface causing it to dissolve and cover the road
surface in a thin layer of CMA solution. The application of liquid CMA simply quickens this
process. Research indicates that CMA is most effective above -7°C.
CMA interferes with the ability of frozen particles to adhere to each other or to the surface
of the road. Because this mechanism does not result in the melting of snow, the best
results are achieved when CMA is applied to the highway surface prior to or in the initial
stages of a snowfall or freezing event. These properties make CMA highly effective for
proactive use and application before freezing occurs, thus preventing ice formation. Unlike
salt-based products, CMA does not form flowing brine, helping to keep the snow drier and
improving surface traction.
The ability of CMA to interfere with the adherence of frozen particles prevents the
accumulation of layers of ice and hardpack snow. Should a layer of snow accumulate on
the pavement, the action of CMA retains it in a loose state allowing effective removal by
way of a plough or grader.
Typically a thin layer of snow and de-icing solution are retained on the surface following
ploughing and treatment. This residual layer continues to guard against the formation of ice
and hardpack snow and the rates of successive applications can be reduced throughout the
remainder of the event. The residual effect also lasts between events providing that there
is no rainfall. Experience has shown that surfaces treated with CMA often exhibit anti-icing
properties due to this residual effect.
4.3 Methods of application
CMA is supplied from the manufacturers in solid granular form. It can be applied in this form
but is often applied in solution. Application in dry form is generally undertaken following a
snow event and is distributed via a unit similar to an agricultural fertiliser spreader. This unit
requires only one pass to cover both lanes.
The liquid form is usually applied to prevent ice formation. It is mixed from CMA granules
dissolved in water at 25% solution. Liquid CMA is applied by a purpose made applicator
that delivers output at the required rate.
Application of solid and liquid forms takes place when wind speeds have subsided.
4.4 Costs
CMA is imported into New Zealand from the US and is an expensive de-icing agent when
compared to others such as sodium chloride. The cost per tonne of the CMA used during
the 2009 winter season was NZ $4265 (excluding GST), although price is highly dependant
on exchange rate fluctuations. The cost of sodium chloride on the international market is
highly variable, depending on country or state, supply and demand, and transport cost.
However, its' cost is generally at least an order of magnitude less per tonne than CMA.
16'h April 2010 7

CMA De-icing Agent: A review of environmental effects monitoring in New Zealand 1998-2009
5 Overseas research into the environmental and health effects of
CMA
5.1 Overseas research in the mid-1980's and early 1990's
A number of studies were undertaken in the US during the 1980's to investigate the
potential environmental effects of using CMA to treat icy roads. One of the earliest was
undertaken by the California Department of Transportation, with results reported in Winters
et. al. (1985). This study was based on a literature review and limited laboratory
investigations. Three years later the US Transportation Research Board published the
results of a comprehensive study based on laboratory tests and controlled field plot
experiments (Horner, 1988). This study involved an investigation of the transport fate of
CMA from the highway, the effects on soil physical properties, biodegradation of CMA in
soils, effects on plants and phytochemicals, effects on surface water quality and effects on
aquatic biota. A key difference between the two studies was that Winters et. al. (1985) used
pure, reagent grade CMA, while Horner (1988) used production grade CMA, which was
expected to be contaminated with by-products of the production process.
During this period a number of smaller more targeted investigations of the potential
environmental effects of CMA were also undertaken; mobilisation of trace metals (Amrhein
et. al. 1992) and effects on small ponds (LaPerriere & Rea, 1989).
In 1992 McFarland and O'Reilly integrated the results of many of these studies to provide
an indication of environmental risk and help guide the operational application of CMA. Their
review concluded the following:
"(1) CMA concentrations used to deice roads have little to no phytotoxic effects on roadside
vegetation;
(2) CMA did not mobilize pre-existing heavy metals from a variety of roadside soils and may
provide some benefit to soil structure;
(3) CMA had little to no toxic effects on aquatic species tested, including vertebrates and
invertebrates;
(4) CMA did not increase algal, periphyton, or phyto-plankton biomass;
(5) CMA is unlikely to cause treatment problems in POTWs (Publically Owned Treatment
Works) receiving expected CMA concentrations in runoff;
(6) CMA is unlikely to have significant negative impacts on receiving water dissolved
oxygen for most application scenarios;
(7) CMA toxicological effects were similar to or less severe than sodium chloride in a full
complement of short-term toxicology tests leading to the conclusion that CMA has low
acute mammalian toxicity and should affect workers health no more than similar use of salt.
16'h April 2010 $

CMA De-icing Agent: A review of environmental effects monitoring in New Zealand 1998-2009
This body of CMA research conducted in a variety of academic and private labs indicates
that negative environmental and toxicological impacts are likely to be insignificant in the
vast majority of CMA applications."
5.2 Overseas studies 1990's to the present time
Although the review undertaken by McFarland and O'Reilly (1992) suggested that there
could be some exceptions to the conclusions that negative environmental and toxicological
impacts were likely to be insignificant, it did provided a high level of confidence that CMA
presented lower risk than salt in terms of environmental impact. CMA also had low
corrosion potential and also presented minimal risk in terms of human health. Since its
subsequent introduction into winter maintenance programs overseas it has also proved an
effective de-icing and anti-icing agent, however, it is relatively costly to produce, up to 40
times the cost of salt (Levelton Consultant Ltd., 2007). This has presumably been an
obstacle to its widespread application in areas where large volumes of de-icing agent are
required, despite the lower associated environmental and corrosion risks. Overseas
therefore, CMA tends to be used in specialist situations e.g. where corrosion is a potential
issue for bridge decks (Levelton Consultant Ltd., 2007).
Given the research pointing to lower environmental impact than salt, which is already
extensively used overseas, and the relatively limited extent of its' use, it is perhaps not
surprising that monitoring of environmental effects of CMA in actual operational situations
has been limited. A study by the U.S. Geological Survey (Tanner & Wood, 2000) of the
effect of CMA on the water quality of Bear Creek, Clackamas County, Oregon is one of few
such studies. This monitoring found that there were no measurable effects of the
application of CMA to Highway 26 on Biological Oxygen Demand (BOD), calcium or
magnesium of Bear Creek and its tributaries, reinforcing the conclusions of the laboratory
investigations and field testing.
Since the mid-1990's a few studies have been undertaken investigating various aspects of
the environmental and toxicological effects of CMA. Toxicology work (Robidoux & Delisle,
2001) presents the results of a comparative ecotoxicological study of three de-icers (sodium
chloride, sodium formiate and calcium magnesium acetate) on macrophyte growth, seed
germination and earthworms. The methodology involved applying control substances and
varying concentrations of de-icer to various soil/substrate mixes contained in petri dishes
and observing effects on the organisms in relation to endpoints including IC10, IC50, LC10,
LC50, NOEC (No Observed Effect Concentration) and LOEC (Lowest Observed Effect
Concentration). The study indicated that CMA was relatively toxic to plants and
earthworms. More recently Joutti et. al. (2003) undertook ecotoxicity investigations that
indicated that CMA had significant effects on plant growth, microbial luminescence and
enzymatic activity. However, results of such studies require substantial care in their
interpretation since conditions under which the toxicological data are gathered are usually
significantly different from those found in the field i.e. organisms are continually exposed to
CMA in confined and closed systems. No toxicity effects have been observed by the
monitoring undertaken in New Zealand indicating that concentrations typically found in the
receiving environments under operational conditions are below levels necessary to cause
significant toxic effects.
16'h April 2010 9

CMA De-icing Agent: A review of environmental effects monitoring in New Zealand 1998-2009
Under more realistic experimental conditions CMA has been shown as per earlier studies to
have minimal impact on plant growth. Akbar, et. al. (2006) undertook a greenhouse study
on road common roadside plants (Festuca rubra, Lolium perenne, Plantago lanceolata and
Trifolium repens) in the UK to determine the relative effects of salt and CMA. No signs of
injury or adverse symptoms were observed in the plants sprayed with CMA and the study
confirmed that CMA was less harmful than salt.
A more recent report concerning of the environmental effects of CMA, based on a review of
literature, is contained in a report published by the Colorado Department of Transportation
(Fischel, 2001). This report generally supported the findings of earlier studies that CMA
poses a low risk to the environment, concluding that aquatic toxicity to fish and
invertebrates is low, and that CMA is not harmful to terrestrial vegetation at concentrations
typically used on highways. Some potential to cause depletion in oxygen in soils as a result
of soil micro-organisms breaking down the acetate ion was noted, but the principal risk
identified was the potential to cause oxygen depletion in rivers, streams and lakes.
However, since in most situations the dilution of de-icer by precipitation and/or meltwaters
from roads entering streams was estimated to range from 100 to 500, depletion was only
likely to occur in small, slow flowing streams and ponds. Further dilution was also expected
to occur in the receiving waterbody.
5.3 Potential environmental effects based overseas research
In the process of evaluating de-icing options for trial in New Zealand the AEE prepared by
Works Consultancy Services Ltd. (1996) reviewed the international studies undertaken up
to that time. From the literature reviewed a number of preliminary conclusions concerning
the main potential environmental risks associated with CMA application on the Desert Road
environment were made:
• CMA biodegrades naturally into its' component parts of calcium, magnesium and
acetate. Magnesium and calcium are secondary plant nutrients and also benefit the soil
structure by increasing its' permeability. This is the result of flocculation of soil particles
caused by calcium and magnesium ions. Acetate is an abundant organic acid and the
acetate ion presents a readily usable source of carbon for micro-organisms and hence
biodegrades naturally (Horner, 1988). These components have no adverse effect on
soil strength or compaction.
• CMA, like other salts, had the potential to affect soil structure, physicochemical
properties and the mobility of trace elements. In "normal" agricultural soils, the pool of
exchangeable cations in soil solution has adequate levels of calcium and magnesium.
In contrast, the soils in the Desert Road region were described as being strongly
leached with low levels of calcium, magnesium, potassium and sodium (Rijkse, W.C.
unpubl. data). They were also low in organic matter, acidic and low in phosphorus.
From this it was concluded that CMA application to road surfaces had the potential to
increase soil fertility (by increasing Ca and Mg levels) and increase pH. However, Ca
and Mg cations tend to displace existing K+ and NH4+ ions from soil colloids and so
could deplete soil K+ levels at a local level. Horner (1988) also pointed out that some
production grade CMA can be contaminated with phosphorus, which could present an
even greater risk of increasing soil fertility. There was therefore some uncertainty
16'h April 2010 1 �

CMA De-icing Agent: A review of environmental effects monitoring in New Zealand 1998-2009
regarding the potential impact of CMA on roadside vegetation patterns. Subsequent
tests on the CMA used in New Zealand have shown that it contains very low levels
phosphorus at approximately 56g/tonne.
• The mobilisation of trace metals by CMA application was raised as an issue by the
literature as it was with NaCI applications. However, the AEE (Works Consultancy
Services, 1996) concluded that metal contamination was not a significant risk with use
of NaCI. Since CMA poses less of a risk than NaCI with respect to metal mobilisation
(Amrhein et. al. 1992), CMA was not expected to cause unacceptable levels of inetals
in the environment.
• The toxicity of CMA in solution to natural plant communities bordering the Desert Road
was unknown. However, tests on plant species overseas (e.g. Horner 1988) suggested
that CMA applications, at the levels expected on the Desert Road, were unlikely to
cause detrimental effects on roadside vegetation through tissue damage. The longer-
term effects of continuing CMA applications had not been tested overseas, but were
considered unlikely to be significantly detrimental to plant growth at the concentrations
likely to be attained in field situations (Horner, 1988). Modification of plant species
composition was a possibility given the perceived potential for improving soil fertility.
• The depletion of dissolved oxygen from the degradation of acetate was considered to
be the major potential water quality effect related to the full-scale use of CMA (Horner &
Brenner, 1992). Several studies had shown that CMA decomposition exerted a
significant BOD on receiving waters with potential consequent adverse effects on fish
and other aquatic life (e.g. Horner, 1988; LaPerriere & Rea, 1989; Eheart et. al. 1993).
LaPerriere & Rea (1989) demonstrated that in concentrations of about 60mg/L in
standing waters, oxyygen depletion could be severe enough to be lethal to salmonid
fishes. However, the studies undertaken were based on worst-case scenario situations
i.e. the receiving waters were either small enclosed water bodies or small streams with
heavy loadings. Fischel (2001) cited MacFarland and O'Reilly (1992) as reporting that
the use of CMA was unlikely to have significant impacts on dissolved oxygen levels in
receiving water for most application scenarios. Nevertheless, this issue was
considered to present the area of greatest environmental risk and uncertainty with
regard to the use of CMA in New Zealand.
• A number of studies reported on the potential impacts of CMA on algal (phytoplankton)
and periphyton biomass (e.g. Horner, 1988; LaPerriere & Rea, 1989). These led to
concerns that CMA could have the potential to cause "sewage fungus" growth, a
common term used to describe a nuisance community of filamentous bacteria and/or
fungi. Such communities can smother the stream bed forming unsightly pink or white,
feathery, cotton-wool like growths. Furthermore, the metabolism of the sewage fungus
can cause rapid oxygen depletion in the water column, particularly within the stream
bed and this together with the general smothering effect can make the habitat
unsuitable for the oxygen sensitive faunas of mountain streams. The water use
standards in the third schedule of the RMA (1991) require that "There shall be no
undesirable biological growths as a result of any discharge of a contaminant into the
water" for water classified to be managed for aquatic ecosystems, fishery, fish
16'h April 2010 11

CMA De-icing Agent: A review of environmental effects monitoring in New Zealand 1998-2009
spawning, contact recreation, irrigation and industrial abstraction purposes. Given the
location of the streams in a National Park and their use for trout spawning, this
standard was considered applicable (Works Consultancy Service Ltd., 1996).
• Standard tests carried out by Horner (1988) showed that toxicity of CMA to aquatic
biota was low.
5.4 Human health effects
The Material Safety Data Sheet (MSDS) and web-based product toxicology information
prepared by one of the principal manufacturers of CMA, Cryotech De-icing Technology, is
included in Appendix 1. Key attributes relevant to human health and environmental impact
summarised in this information are:
• If absorbed through the skin, inhaled or swallowed it is considered practically non-toxic
to internal organs. Acute oral toxicity in rats is greater than 5000mg/kg.
• Draize Scores for eye and skin irritation tests for rabbits indicate that the product is
practically non-irritating and no special eye or skin protection is usually necessary.
• No special respiratory protection is normally required. However, if operating conditions
create high airborne concentrations, the use of an approved respirator is
recommended.
• It is not designated as a hazardous material by the US Federal DOT and its movements
are not restricted under any transport regulations.
This information indicates that CMA is non-hazardous, poses minimal risks to the
environment in quantities typically used for de-icing purposes, and represents negligible
health risk to operators and third parties.
6 Early modelling and testing in New Zealand
6.1 Perceived risks to the New Zealand environment
Whilst, most overseas research indicated that the risk of CMA causing significant
environmental effects was likely to be low, in the quantities used to treat icy roads, some
uncertainties remained. This prompted regional authorities to require testing and monitoring
that was specific to the New Zealand environment as conditions of the consents issued to
permit the use of CMA to treat icy roads.
The key uncertainties were perceived to be:
• Dissolved oxygen depletion in streams and small waterbodies receiving stormwater
runoff from roads treated with CMA;
• Adverse effects on aquatic invertebrates resulting from dissolved oxygen depletion;
• Increases in the abundance of benthic algae and heterotrophic growths in waters
receiving runoff containing CMA;
• Adverse effects on the health of roadside vegetation;
• Changes in species composition in plant communities adjacent to the State Highway;
16'h April 2010 1 2

CMA De-icing Agent: A review of environmental effects monitoring in New Zealand 1998-2009
• Modification of soils and potential fertiliser effects due to presence of phosphorus as an
impurity in CMA supplied from some sources.
Perhaps not surprisingly concerns were particularly high regarding these issues for the
Desert Road, the first section of highway in New Zealand where CMA was trialled. The
Desert Road passes through the Tongariro National Park, a place of high natural value and
a dual World Heritage Area, recognising its importance for Maori cultural and spiritual
associations, as well as its outstanding volcanic features.
6.2 Modelling and testing effects on BOD for Desert Road streams
The potential for CMA to deplete dissolved oxygen levels in receiving waters with its'
consequent potential to cause adverse effects to aquatic life was undoubtedly the issue of
greatest environmental concern surrounding the introduction of CMA. That CMA has a BOD
was demonstrated in laboratory experiments undertaken by Horner (1988) and by other
studies (LaPerriere & Rea, 1989; Eheart et. al. 1993).
While the risk of oxygen depletion for most application scenarios in the field was assessed
to be low due to the effects of dilution and biodegradation before entering the stream, it was
considered prudent to undertaken investigations to confirm these predictions with regard to
the Desert Road streams, particularly given the pristine nature of the environment. The
initial investigation involved a simple Streeter-Phelps modelling exercise undertaken by
NIWA to determine stream response for the five Desert Road trial areas.
Field and laboratory studies confirmed that the Desert Road streams had high water quality.
Background BOD levels were low, DO was at saturation and re-aeration rates were high.
Given the nature of the streams and the sensitivity of their fauna, 80% saturation
concentration was set as the minimal allowable DO level (Works Consultancy Services,
1996).
Results of the simple Streeter-Phelps DO modelling exercise suggested that at `normal'
rates of CMA delivery to streams, the risk of DO depletion causing adverse effects on
aquatic biota would be negligible. However, under worst-case scenarios, the model
indicated that severe oxygen depletion may occur, which would severely impact fish
populations and invertebrate assemblages. The fauna in small streams receiving CMA
runoff could therefore be most at risk.
As a consequence of these results further modelling work was carried out by NIWA to refine
the outputs of the model. In order to do this, inputs to the model needed to be accurately
defined to test whether the original modelling results were realistic. Therefore site specific
information from the Desert Road streams was collected to allow for further modelling under
an expanded range of scenarios.
In large Desert Road streams, such as the Waihohonu and Oturere, modelling results
showed that there was no risk of significant DO reductions occurring under even the most
pessimistic of CMA application scenarios. Furthermore, in the Te Piripiri Stream, and other
medium-sized streams (Mangamate, Mangatawai and Makahikatoa), the risk of significant
DO reductions occurring was deemed low. Under a 6 applications scenario, followed by
short, intense flood, the Te Piripiri Stream DO levels dropped below the 80% saturation
16'h April 2010 13

CMA De-icing Agent: A review of environmental effects monitoring in New Zealand 1998-2009
threshold. However, the risk of DO reductions in Te Piripiri Stream (and other of similar
size) was considered to be minimal so long as the mass of CMA applied did not exceed a
predetermined critical value (known as the critical burden).
The model showed that the Wharepu, and other small (< 0.05 m3 s-' baseflow) streams,
had the greatest risk of significant DO reductions occurring as a result of CMA applications.
These higher risk streams all occur within trial Site 5(Wharepu-Waikato). Calculations of
the mass of CMA required to produce a significant reduction in DO, showed that under
"worst-case" conditions an application rate of 35 gm-2 would produce a 20% reduction in
DO. Therefore so long as the application rate did not exceed this level, the environmental
risk was predicted to be minimal.
Subsequent testing and verification of the model in the field, included a number of
calibration tests to check predictions of effects on DO and assess input fractions and input
duration (Ray, 1998a; Ray, 1998b; Ray, 1999), this resulted in the critical burdens for these
streams being increased significantly. Furthermore, the field testing and the monitoring over
the following 5 years showed that the model's predictive capabilities for streams in the
Desert Road environment are very conservative i.e. effects have not occurred when
predicted. This was considered to be the result of the very low background levels of
bacteria in the near-pristine Desert Road streams, which means that DO is not oxidised in
the way predicted by the model. As an example the calibration tests resulted in the critical
burden for the Wharepu Stream being raised from an initial 28kg to 500kg.
6.3 Monitoring on the Desert Road 1998-2002
The Streeter-Phelps DO modelling exercise undertaken by NIWA was an important first
step and generally confirmed that the risk of CMA causing a significant lowering of
dissolved oxygen in Desert Road streams was low. However, given the high value of the
receiving environment, field monitoring was deemed necessary to confirm the results
predicted by the model under actual operational conditions. Over the next five years (1998-
2002) environmental monitoring, not only of dissolved oxygen levels in streams, but also
aquatic invertebrates, benthic algae and heterotrophic growths, terrestrial vegetation health
and species composition, and soil properties, was undertaken by NIWA and Landcare
Research, for the five trial CMA application sites. At the same time from 2000, the
operational use of CMA was expanded to many other parts of the State Highway Network.
In almost all cases environmental monitoring was undertaken to comply with regional
council consent conditions. The results of this monitoring are described in section 8 and
analysed in section 9 of this report.
16'h April 2010 14

CMA De-icing Agent: A review of environmental effects monitoring in New Zealand 1998-2009
7 Operational use of CMA on the highway network in New Zealand
7.1 Regions where CMA is consented for use as a de-icing agent
Based on the results of the environmental testing and monitoring undertaken on the Desert
Road in the late 1990's, as well as demonstrated effectiveness, the use of CMA was
progressively extended to many other parts of the State Highway network from 2000
onwards. CMA is now consented for use over the entire State Highway Network of South
Island and over ice and snow prone sections of the network in the following North Island
Regions:
• Waikato Region
• Manawatu-Wanganui Region
• Hawke's Bay Region
• Bay of Plenty Region
Table 3 summarises the parts of State Highway Network where the use of CMA has been
consented and gives details of when those consents were issued and current monitoring
requirements, as well as the basis on which the most recent consents were processed.
7.2 Extent of network where treatment with CMA is permitted
The extent of the State Highway network consented for treatment with CMA varies from
region to region, and there is a marked difference in consent coverage between North and
South Island. In the North Island, where CMA was first introduced in 1998, treatment is
generally consented for specific sections of the highway network that most frequently
experience hazardous conditions due to the presence of ice and snow. The exception is the
South-West Plateau network within the territory administered by Horizons Regional Council,
where CMA treatment is permitted on the entire network. This includes substantial sections
of highway that never or rarely require treatment. By contrast, CMA has been permitted for
use on the entire State Highway network in South Island since 2001. These South Island
consents were granted on the basis of the results of overseas research and the initial
monitoring and testing that had been undertaken on the Desert Road up to that time.
7.3 Basis on which applications have been processed
The consents granted in the late 1990's and early 2000's were generally processed on a
notified basis. As monitoring has confirmed that the environmental effects of CMA are
indeed minor, consents and variations of original consents granted since 2006 have mainly
been processed on a non-notified basis. All recent consents and variations to existing
consents issued by South Island regional authorities have been processed on a non-
notified basis with the exception of Environment Canterbury processed on a limited notified
basis.
7.4 Environmental monitoring requirements for CMA
Most Regional Councils' issuing consents permitting the use of CMA have at least initially
required environmental monitoring as conditions of consent. Whilst there was a reasonable
level of acceptance that CMA posed low environmental risk, there were still some residual
16'h April 2010 15

CMA De-icing Agent: A review of environmental effects monitoring in New Zealand 1998-2009
concerns regarding unforeseen environmental effects, particularly relating to its' potential to
lower DO levels in water. Concerns were further compounded by the fact that most
overseas research was based on laboratory experiments and field plot tests, and there was
little monitoring of environmental effects in relation to operational use of CMA. The
possibility was also raised that results of testing in one environment might not transfer
directly to another.
Since the introduction of CMA in New Zealand a substantial amount of environmental
monitoring has been undertaken in many regions. As will be described in section 8 of this
report, extensive monitoring in New Zealand has found no significant environmental effects
that can be linked to the use of CMA as a de-icing agent. As a consequence, most regional
council's have progressively reduced the monitoring requirements since it was first
introduced. Most recently, some regional council's have completely removed the
requirement for environmental monitoring on the basis that the effects of CMA are
considered to be no more than minor; Environment Southland in 2006 and Otago Regional
Council in 2009.
Even today, nearly 25 years after the first environmental investigation were undertaken,
there appears to be little published environmental monitoring data relating to the operational
use of CMA overseas. On this basis, it is possible that the environmental effects of CMA
have now been monitored more extensively in New Zealand than anywhere else in the
World. One possible explanation for this is that in Northern latitudes salt (NaCI) is used
extensively to treat icy roads. Various early studies investigating the likely environmental
effects of CMA indicated that CMA was likely to pose less environmental risk to the
environment than salt. Hence, the use of CMA use brings immediate environmental benefits
over the existing NaCI use. However, introduction of CMA has tended to be selective and
not widespread due to it significantly higher cost associated with using CMA and the more
extensive ice and snow problems experienced in these countries.
7.5 Consent duration
There is significant variation between regions as to the duration of consents issued to
permit the use of CMA. Several of the currently operative consents permitting the discharge
of CMA were issued for a term of 10 years; Environment Waikato, Environment Canterbury,
Horizons Regional Council, Hawke's Bay Regional Council and Otago Regional Council.
However, a number of regional councils have issued consents for a longer period; West
Coast Regional Council (15 years), Environment Bay of Plenty (15 years), Nelson City
Council (20 years), Tasman District Council (20 years), Marlborough District Council (20
years) and Environment Southland (25 years). These longer term consents have been
issued since 2006 which reflects the growing acceptance that the effects of CMA are likely
to be no more than minor.
16'h April 2010 16

CMA De-icing Agent: A review of environmental effects monitoring in New Zealand 1998-2009
8 Results of Environmental Monitoring in New Zealand 1998-2008
8.1 Environmental aspects monitored
Between 1998 and 2009 monitoring and testing connected to the use of CMA in New
Zealand has focussed to a greater or lesser extent, depending on the territorial authority, on
the following aspects of the environment:
• Dissolved oxygen
• Aquatic invertebrates
• Benthic algae and heterotrophic growths
• Vegetation health
• Species composition
• Soil properties
There has also been ecotoxicology work undertaken in connection with the consent issued
by Otago Regional Council (Thompson, 2002).
8.2 Dissolved oxygen
8.2.1 Dissolved oxygen monitoring North Island
Dissolved oxygen monitoring has been undertaken in 4 different North Island streams for up
to 5 consecutive years (Table 4). Remote DO probes and temperature loggers recorded
dissolved oxygen levels continuously during the winters of 1999 to 2002 at 15 minute
intervals on the Wharepu, the Te Piripiri and Mangamate streams on the Desert Road
(SH1). Recording was undertaken 50m upstream of SH1 and as close to the confluence of
the streams with the Tongariro River as possible. Although short-term increases Total
Organic Carbon (TOC) after rain indicated CMA was entering the streams, no discernible
effects on dissolved oxygen that could be attributed to CMA were recorded during this 5
year period. Dissolved oxygen levels remained above 90% saturation levels except for
2000 when levels remained above 85% saturation (Opus International Consultants, 2002;
Ray & Scarsbrook, 2002).
Similarly, continuous monitoring of dissolved oxygen levels on the Kuratau Stream (SH41)
during the period 2005 to 2009 has also has also found no effects on dissolved oxygen
levels that could be attributed to CMA, although results of the 2009 winter season have still
to be reported. Dissolved oxygen levels have proved highly variable on this stream with
dissolved oxygen levels as low as 40% saturation. However, the high variability could not
be linked to CMA and was observed upstream of the highway as well as downstream
(Wright-Stow, 2009a).
8.2.2 Dissolved oxygen monitoring South Island
Otago Regional Council initially required dissolved oxygen monitoring in 5 streams
commencing in 2000. This was subsequently reduced to 2 streams in 2003 (Table 4). The
monitoring included spot readings of inter-gravel dissolved oxygen levels and temperature
20m upstream and 20m downstream of the discharge point in each stream. Continuous
16'h April 2010 1 7

CMA De-icing Agent: A review of environmental effects monitoring in New Zealand 1998-2009
recording of dissolved oxygen levels was also logged every 10 minutes by a smart sensor
installed in the Pass Burn over a 6 week period during the winter of 2003. None of the
dissolved oxygen monitoring found discernible effects related to CMA application and runoff
events (Beale, 2003; Transit New Zealand, 2003).
Dissolved oxygen as well as temperature, conductivity and pH was measured using a hand-
held multi-parameter sensor (YSI Inc.), pre-winter, post winter and on four occasions during
winter at five locations within the Tasman/Nelson/Marlborough Region. In contrast with the
monitoring method in Otago, DO readings were taken within the flow at the stream bed
rather than within the gravel substrate. This change in method resulted from
inconsistencies in early trials with inter-gravel sampling and difficulties encountered with the
rocky nature of the streams monitored. All measurements taken showed consistently high
dissolved oxygen levels, and were generally greater than 90%, with all above 80%
saturation. Small amounts of variation noted between upstream and downstream sites were
within the accepted range for sampling variability, indicating no trends that were attributable
to the action of CMA in road runoff (Shearer & Sneddon, 2005; Sneddon & Clark, 2007).
The original consents issued by Environment Canterbury and West Coast Regional Council
in 2001 required environmental monitoring at 6 and 4 sites respectively. However, lack of
CMA being applied has meant that only two sites, the Craigieburn Stream and Longslip
Creek within the Canterbury Region have been monitored for DO concentrations on a
regular basis. Dissolved oxygen was measured using a hand-held multi-parameter sensor
(YSI Inc.), pre-winter, post winter and on several occasions during winter at control and
treatment sites. DO levels remained consistently high in all cases (>8 mg/L) and no
significant adverse effects were observed (Beale, 2004; Kingett Mitchell Ltd., 2004).
8.2.3 Dissolved oxygen modelling Lake Pearson, Lake Lyndon and Craigieburn
Stream
Due to the lack of cold winters between 2001 and 2006, and relatively low usage of CMA
Transit New Zealand agreed with Environment Canterbury to undertake modelling
exercises to further assess the risk posed by CMA to Lake Lyndon and an unnamed
tributary, and to Lake Pearson and its' tributary the Craigieburn Stream. The Streeter-
Phelps Model was used to assess the potential for dissolved oxygen depletion on these
waterbodies. In applying the model it was conservatively assumed that all CMA was
washed into the Craigieburn Stream within 15 minutes, resulting in minimal dilution. This is
highly unlikely as a significant proportion of the CMA is likely to be immobilised before ever
reaching the stream. DO in the Craigieburn Stream is typically at saturation, approximately
12g/m3 during the winter months. The minimum DO predicted due to a single CMA
application event was 9.9g/m3 which is above the 30% reduction in DO limit set by the RMA
(in this case 9.6 g/m3). However, it is also important to recognise that this minimum would
only be reached 28km from the application point and therefore could never occur in this
system as Lake Pearson was only 3km downstream (Campbell, 2006).
Further modelling of multiple CMA application scenarios for the Craigieburn Stream used a
100g/m2 application of CMA all washed into the stream within a 15 minute rain event as the
basis for calculation. This calculation determined that there could be a slight depression in
DO levels, however even such cases of extreme levels of application DO would still remain
16'h April 2010 1 $

CMA De-icing Agent: A review of environmental effects monitoring in New Zealand 1998-2009
at or above 80% saturation. Temperature sensitivity analysis did identify the potential for
DO to drop to 70% if water temperature increased to 10 °C. However, winter temperatures
above 7°C were considered highly unlikely for this particular stream and therefore risk of
adverse effects on biota were considered minimal (Campbell, 2006).
Oxygen depletion modelling for Lake Pearson and Lake Lyndon, and an unnamed tributary
of Lake Lyndon, indicated that even under "worst case scenarios" of multiple applications
during a cold winter there was unlikely to be any significant depletion in dissolved oxygen
concentrations (Campbell, 2006).
8.2.4 Lowland stream DO test: Paritikona Stream
Most CMA is applied in upland environments where water temperature in streams tends to
be less than 10 °C and re-aeration potential is high due to higher stream gradients. While
the extensive monitoring that has been undertaken in these environments showed no
effect, concerns still remained regarding warmer, lower gradient, lowland streams. Since
CMA application in lowland environments tends to be more localised and less frequent,
difficulties arose finding a site where a CMA response was possible i.e. low volume, low
gradient and likely to receive substantial direct discharge, for monitoring purposes. In order
to assess the risk to lowland streams a test was developed to monitor dissolved oxygen
levels in a stream following the direct input of CMA.
DO was monitored in the Paritikona Stream in a test which simulated "worst-case" and
"extreme worst-case" discharge scenarios, in single and multiple applications situations.
Quantities of CMA equivalent to application rates of 10g/m2 and 30g/m2 applied to a 500m
section of highway were discharged directly to the stream over periods of less than 30
minutes. The discharges based on the 10g/m2 and 30g/m2 application rates were
undertaken consecutively less than 2 hours apart, thereby effectively creating a multiple
application discharge scenario. Despite the passage of CMA downstream being clearly
measurable via conductivity spikes, DO concentrations did not vary from the diurnal
changes observed at a control site (Wright-Stow et. al., 2008). It was therefore concluded
that direct discharge of CMA to the stream had no measurable impact on dissolved oxygen
levels.
8.3 Aquatic invertebrates
8.3.1 Aquatic invertebrate monitoring North Island
Aquatic invertebrate monitoring has been carried out on the Wharepu, the Te Piripiri and
Mangamate streams on the Desert Road (SH1) during the winters 1999 to 2004 and
subsequently on the Te Piripiri, Mangamate and Kuratau (SH41) streams, during the
winters 2005 to 2009 to comply with Environment Waikato consent conditions (Table 5).
Typically these assessments have involved upstream and downstream sampling and use of
several invertebrate community metrics to detect impairment; taxa richness; numbers of
Ephemeroptera, Plecoptera and Tricoptera (EPT) taxa; %EPT individuals,
Macroinvertebrate Community Index (MCI) and Community Loss Index (CLI). Between
1999 and 2002 the sampling was undertaken monthly, subsequently sampling was
undertaken pre and post season only. In 10 years of monitoring no effects that could be
linked to the use of CMA have been observed (Opus International Consultants, 2002;
16'h April 2010 19

CMA De-icing Agent: A review of environmental effects monitoring in New Zealand 1998-2009
Wright-Stow, 2009a). Results of the 2009 winter season have still be analysed and
reported.
Similar Aquatic invertebrate sampling was undertaken pre and post winter for the Horizons
Regional Council consent at two sites. The consent required sampling at four sites however
CMA was only applied to the sections of highway adjacent to the Makatote and Hihitahi
streams (Table 5). The assessment was based on taxa richness, MCI, QMCI, %EPT (taxa),
%EPT (individuals), community loss index (CLI, relative to upstream reference site) and
invertebrate density calculations (Wright-Stow, 2008b). No significant changes in aquatic
invertebrates communities attributable to CMA were observed over the two years of
monitoring.
In 2009 a very small tributary of the Mangapapa Stream which intersects a treated section
of SH5 (Taupo-Napier) near Te Haroto was monitored pre and post season for macro-
invertebrates. The monitoring undertaken on this stream has particular significance as it is
the smallest stream sampled in North Island with a width ranging between 0.5 and 1.Om. In
addition to this it is a rare example of a small stream receiving runoff from a curbed and
channelled stormwater system with channelized discharge direct to the stream. These
factors mean that this stream has the potential to receive high concentrations of CMA.
Despite CMA being applied to the highway during 2009, no differences were found in the
metrics calculated that were attributable to CMA, indeed the MCI increased post winter at
the downstream site, and post winter %EPT scores were essentially the same at both
upstream and downstream sites (Wright-Stow, 2009b).
8.3.2 Aquatic invertebrate monitoring South Island
Otago Regional Council initially required benthic invertebrate monitoring in 5 streams
commencing in 2000 (Table 5). This was subsequently reduced to 2 streams in 2003; Pass
Burn and Manuka Stream. Sampling involved three replicate kicknet samples from
sampling locations. Analysis involved was based on MCI and semi-quantitative macro-
invertebrate community index scores (SQMCI). The monitoring found no evidence of effects
on benthic invertebrates related to the application and discharge of CMA to the streams
monitored. Slight differences between control and treatment sites observed in Pass Burn
and Manuka Stream were extremely small and considered to be unlikely to be biologically
meaningful (Beale, 2003; Transit New Zealand, 2003).
Macro-invertebrate samples were collected pre-winter, post winter and twice during winter
at five locations within the Tasman/Nelson/Marlborough Region between 2004 and 2006
(Table 5). Samples were analysed to provide MCI and SQMCI scores, and %EPT. While
some statistically significant differences in MCI and SQMCI values between upstream and
downstream sites were observed there were no consistent trends that could be linked to the
presence of CMA in road runoff. For example in 2006, MCI values fell over time and were
significantly lower than upstream during and after winter, therefore raising the potential of a
CMA effect. However, this site received relatively few CMA applications and MCI values
both upstream and downstream were indicative of a clean environment throughout the
winter period. In another unnamed stream a downward trend in MCI scores was also
observed between upstream and downstream sites, however the observed trend began
before application of CMA commenced. The monitoring reports comment that if a CMA
16'h April 2010 2�

CMA De-icing Agent: A review of environmental effects monitoring in New Zealand 1998-2009
related effect does exist, it is likely to be short-term and very difficult to separate from
natural variation between reaches (Shearer & Sneddon, 2005; Sneddon & Clark, 2007).
Similar monitoring of macro-invertebrates has also been undertaken in the Canterbury
Region. Monitoring of macro-invertebrates was required at a total of 6 locations in the
Canterbury Region and 4 locations in the West Coast Region as conditions of consents
issued by the respective regional councils in 2001. However, due to the low incidence of
CMA application meaningful monitoring was only undertaken in the Craigieburn Stream and
Longslip Creek in the Canterbury Region. Observations were made on several occasions
during the winter, as well as pre and post season. The monitoring found no evidence of
adverse effects on macro-invertebrate populations (Beale, 2004; Kingett Mitchell Ltd.,
2004), and monitoring was discontinued in 2006.
8.4 Benthic algae and heterotrophic growths
8.4.1 Benthic algae and heterotrophic growth monitoring North Island
Benthic algae and heterotrophic growths were assessed on the Wharepu, the Te Piripiri
and Mangamate streams on the Desert Road (SH1) during the winters 1999 to 2004 and
subsequently on the Te Piripiri, Mangamate and Kuratau (SH41), streams during the
winters 2005 to 2009 to comply with Environment Waikato consent conditions (Table 6).
These assessments involved visual assessments of 10 randomly selected cobbles per
reach at upstream and downstream sampling sites. Between 1999 and 2002 the sampling
was undertaken monthly, subsequently sampling was undertaken pre and post season. In
10 years of monitoring no effects that could be linked to the use of CMA have been
observed (Ray & Scarsbrook, 2000; Opus International Consultants, 2002; Wright-Stow,
2009a). Results of the 2009 winter season have still be analysed and reported.
Periphyton biomass monitoring was also undertaken during the summer months for the
Horizons Regional Council consent on the Makatote and Hihitahi streams. The method
used assessed mean algal biomass as measured by chlorophyll a(mg/m2) (Wright-Stow,
2008b). No significant changes in chlorophyll a concentrations were observed over the two
years of monitoring.
8.4.2 Benthic algae and heterotrophic growth monitoring South Island
Otago Regional Council initially required benthic algae and heterotrophic growths
monitoring in 5 streams commencing in 2000. This was subsequently reduced to 2 streams
in 2003 (Table 6). Sampling involved qualitative visual assessments to determine gross
changes in the composition and/or density of algal and heterotrophic growths, and included
photographic recording pre and post winter season. No discernible differences were
observed between upstream and downstream sampling locations in any of the streams
sampled during the years they were being monitored (Beale, 2003; Transit New Zealand,
2003).
Gross changes in benthic algae and heterotrophic growths were monitored by estimating
percentage cover in a 500mm quadrat, pre-winter, post winter and on four occasions during
winter at five locations within the Tasman/Nelson/Marlborough Region (Table 6). At no point
during any of the monitoring were heterotrophic growths or excessive slimes indentified in
16'h April 2010 21

CMA De-icing Agent: A review of environmental effects monitoring in New Zealand 1998-2009
any of the streams and similarly no significant growths of filamentous algae were observed
(Shearer & Sneddon, 2005; Sneddon & Clark, 2007). Observations made in this study show
no discernible effect of CMA on benthic algae and heterotrophic growths.
Similar monitoring of macro-invertebrates has also been undertaken in the Canterbury
Region. Monitoring of benthic algae and heterotrophic growths was required at a total of 6
locations in the Canterbury Region and 4 locations in the West Coast Region as conditions
of consents issued by the respective regional councils in 2001. However, due to the low
incidence of CMA application meaningful monitoring was only undertaken in the
Craigieburn Stream and Longslip Creek in the Canterbury Region. Observations were
made on several occasions during the winter, as well as pre and post season. The
monitoring found no discernible differences in algal coverage between control and
treatment sites, and no excessive heterotrophic growths and slimes were observed at any
of the sites on any occasion (Beale, 2004; Kingett Mitchell Ltd., 2004). Monitoring was
discontinued in 2006.
8.5 Vegetation health
8.5.1 Vegetation health monitoring North Island
Prior to the first application of CMA in 1998, permanent monitoring plots adjacent to
sections of highway treated with CMA, and also adjacent to control sections of highway
where no CMA was applied, were established in four representative plant communities
along the Desert Road; kanuka forest, beech forest, fernland and tussock grassland (Table
7). Selected plants were monitored for classic symptoms of chemical injury; yellowing, foliar
necrosis, defoliation, and crown dieback. During the period 1998 to 2002 monitoring was
undertaken prior to the first application of CMA, after the first application of CMA and then
each month until 2 months after the last application of CMA in compliance with Environment
Waikato consent conditions. Reporting on the results of the 2002 winter season Smale and
Fitzgerald (2003a) recommended that in the absence of any obvious signs of chemical
injury over the five years of monitoring, that monitoring discontinue.
Subsequently, between the years 2005 and 2009, vegetation health has been monitored in
upland conifer/broadleaved forest and montane conifer plant communities to comply with
Environment Waikato consent conditions to extend the use of CMA to other parts of the
network. Smale and Fitzgerald (2008) concluded that no changes in plant health that could
be related to the use of CMA were observable in these two plant communities and that
these conclusions were similar to those drawn in 2005, 2006 and 2007. Results of the 2009
winter season have still to be analysed and reported.
8.5.2 Vegetation health monitoring South Island
The first consent permitting the application of CMA issued by Otago Regional Council in
July 2000 required vegetation health to be monitored at 5 locations (Table 7). This was
subsequently reduced to 2 locations (Pass Burn-Lindis and Leith Saddle) by a subsequent
consent issued in April 2005. Pre and post winter season qualitative visual assessments of
5m transects established at right angles to the highway over up to 8 seasons of monitoring
found no evidence of die-back or foliar burning in native vegetation. Pass Burn-Lindis and
Leith Saddle (Beale, 2003; New Zealand Transport Agency, 2009).
16'h April 2010 22

CMA De-icing Agent: A review of environmental effects monitoring in New Zealand 1998-2009
Vegetation health was monitored pre and post winter season at 8 locations between 2003
and 2006 in Tasman/Nelson/Marlborough Region (Table 7). Monitoring involved
comparisons based on semi-quantitative visual assessments and photographs at treatment
and control sites. This was undertaken pre and post the CMA treatment season along 10m
transects established perpendicular to the highway. The surveys found that while
vegetation health was almost universally good, its physical state varied in response to
roadside maintenance practices principally mowing, but also herbicide treatment. This
made separation of the effects of CMA from the effects of roadside maintenance practices
difficult. However, apart from impacts from obvious causes related to road maintenance
practices there was no evidence that CMA causing adverse effects on the health of
roadside vegetation. Healthy spring growth was also noted in all cases. With roadside
maintenance practices having a major impact on vegetation it was concluded that CMA
would have to have a substantial and sustained impact for its effects to be observable
(Shearer & Sneddon, 2005; Sneddon & Clark, 2007).
A similar approach to monitoring vegetation health has been used at a number of sites in
the Canterbury and West Coast regions since 2001 (Table 7). However, in the case of
these sites a shorter transect length of up to 5m was used (Beale, 2004; Beale, 2008).
Beale (2008) concludes that visual observations and photographic records indicate no sign
of leaf burn or dieback and that after seven years of observation, potential effects of CMA
are no greater than can be attributed to currently accepted winter management practices.
Overall there is no evidence of any effects on vegetation health that could be attributed to
the use of CMA.
8.6 Vegetation species composition
8.6.1 Vegetation species composition monitoring North Island
Prior to the first application of CMA in 1998, permanent monitoring plots adjacent to
sections of highway treated with CMA, and also adjacent to control sections of highway
where no CMA was applied, were established in four representative plant communities
along the Desert Road; kanuka/manuka secondary forest, primary beech forest, tangle fern
and red tussock grassland to provide a basis for monitoring species composition changes
that could be attributed to the use of CMA (Table 8). A fifth plant community,
kanuka/manuka scrub, was added to the program in 2003. Monitoring was in the
kanuka/manuka forest, beech forest and kanuka/manuka scrub plots was based on
measuring the diameter at breast height of all trees >2.5cm, counting the number of
seedlings in each plot and estimating percentage ground cover. For the fernland and
grassland communities monitoring assessed percentage species cover. Plots were re-
measured in 2000, 2002 and 2008 (Smale et. al. 2001; Smale & Fitzgerald 2003b; Smale &
Fitzgerald 2009). Analysis of the monitoring results in 2008 found only one pattern of
change that could be potentially related to CMA treatment, small increases in red tussock in
high treatment plots verses declines in low treatment and reference plots. However, the
report concluded that such a link was unlikely, given a plot distance from the highway of
15m (Smale & Fitzgerald 2009). Furthermore, soil monitoring found little evidence of
changes in the soils that could be related to CMA (McLeod & Fraser, 2009).
16'h April 2010 23

CMA De-icing Agent: A review of environmental effects monitoring in New Zealand 1998-2009
Subsequently, in 2005 monitoring plots were also established in three other vegetation
community types; upland conifer/broadleaved forest, montane conifer plant and upland
wetland, to comply with Environment Waikato consent conditions to extend the use of CMA
to other parts of the network. These plots were scheduled to be re-measured every 5 years.
The results of the first re-measurement in 2009 are not yet available and therefore no
assessment of effects on these plant communities has been carried out.
8.6.2 Vegetation species composition monitoring South Island
The first consent permitting the application of CMA issued by Otago Regional Council in
July 2000 required vegetation species composition to be monitored at 5 locations (Table 8).
This was subsequently reduced to 2 locations (Pass Burn-Lindis and Leith Saddle) by a
subsequent consent issued in April 2005. Pre and post winter season photographic
recording of transects established adjacent to the road margin found no evidence of
enhanced weed growth. This was supported by consistently low soil nutrient levels in soil
samples taken from the same locations (Beale, 2003; New Zealand Transport Agency,
2009).
Vegetation was monitored pre and post winter season at 8 locations between 2003 and
2006 in Tasman/Nelson/Marlborough Region (Table 8). Monitoring involved comparisons
based on semi-quantitative visual assessments and photographs at treatment and control
sites. This was undertaken pre and post the CMA treatment season along 10m transects
established perpendicular to the highway. The surveys found no changes in vegetation
indicative of a fertilising influence from CMA (Shearer & Sneddon, 2005; Sneddon & Clark,
2007).
A similar approach to monitoring vegetation health has been used at a number of sites in
the Canterbury and West Coast regions since 2001 (Table 8). However, in the case of
these sites a shorter transect length of up to 5m was used (Beale, 2004; Beale, 2008).
Beale (2008) concludes that based on visual observations and photographic records there
are no differences between weed or grass growth when comparing control and treatment
plots that indicate a fertilising effect of CMA. Overall there is no evidence of any effects on
vegetation species composition that could be attributed to the use of CMA.
8.7 Soils
8.7.1 Soils monitoring North Island
At the same time that vegetation species composition plots were being established on the
Desert Road prior to the first application of CMA in 1998, soils sampling was undertaken in
the same vegetation types (Table 9). Composite soil samples (20 cores) were collected at
0-10 cm and 10-20 cm depths from 3 points in each of the eight permanent vegetation plots
at about 10m intervals along the transects. Sampling was undertaken at quarterly intervals
initially, but from 2000 onwards this was change to pre and post winter season, with up to
two samplings during the season. Samples were taken from both control and treatment
sites. Samples were then analysed for pH, electrical conductivity, cation exchange capacity,
base saturation and exchangeable calcium, magnesium, potassium and sodium (Rijkse;
1999, 2000a, 2000b, 2001 & 2002).
16'h April 2010 24

CMA De-icing Agent: A review of environmental effects monitoring in New Zealand 1998-2009
During 1999 to 2001 no large increases in measured parameters that could be attributed to
the use of CMA were observed. Generally values of exchangeable calcium, magnesium,
potassium and sodium remained low to very low. Minor increases in pH, exchangeable
calcium, magnesium, potassium and sodium immediately (3 days) after CMA applications
were found in 2000 and increased levels of exchangeable calcium in 2001. However, these
increases only occurred adjacent to the road and levels had returned to normal within 1
month of the final CMA application (Rijkse; 1999, 2000a, 2000b, 2001 & 2002).
Following the move to full operational use of CMA the requirement for soil sampling was
reduced to reflect the lack of observed effects during the trial period. Re-measurement of
the same soils variables with the addition of bulk density in kanuka forest and fernland,
concluded that although there were slight differences in exchangeable cations within soils
at different sites, there was little evidence to indicate that differences were the result of
CMA application at high treatment sites (McLeod & Fraser, 2009).
Subsequently, in 2005 baseline soil sampling was undertaken in high and low rainfall
locations, Lake Rotoaira-Waituhi-Saddle-Pouakani and Mihi 8km south-west of Reporoa
respectively, in control and treatment sites (Wilde, 2007). This was to comply with
Environment Waikato consent conditions to extend the use of CMA to other parts of the
Central Waikato network. The samples were analysed for the same suit of variables as the
samples taken from the Desert Road. These plots were scheduled to be re-measured every
5 years. The results of the first re-measurement in 2009 are not yet available and therefore
no assessment of effects on these soils has been carried out.
8.7.2 Soils monitoring South Island
The first consent permitting the application of CMA issued by Otago Regional Council in
July 2000 required soils to be monitored at 5 locations (Table 9). This was subsequently
reduced to 2 locations (Pass Burn-Lindis and Leith Saddle) by a subsequent consent
issued in April 2005. Soil monitoring involved collection of pre and post winter season
samples from treatment and control transects. These were analysed for pH, electrical
conductivity, cation exchange capacity, base saturation, calcium (Ca), magnesium (Mg),
potassium (K) and sodium (Na) (Beale, 2003). Soil samples taken from Leith Saddle
showed generally higher post winter/post application concentrations of cations (Ca, Mg, K
and Na) compared to pre-winter/pre-application samples. This seasonal trend tended to
apply to both treatment and control sites. The concentrations of Ca, Mg and K cations also
tended to be higher in post winter samples from treated sites. However, no cumulative
increase in cation concentrations was recorded in the treatment samples during the period
2001 and 2008 (Beale, 2003; New Zealand Transport Agency, 2009). The absence of any
evidence of cumulative effects on soils led to the cessation of soil monitoring in 2009.
Soils were sampled pre and post winter season at 8 locations between 2003 and 2006 in
Tasman/Nelson/Marlborough Region (Table 9). Samples were collected from the 10m
vegetation monitoring transects established perpendicular to the highway, usually at
identified runoff paths. Samples were taken from the top 150mm of the soil profile and
analysed for pH, electrical conductivity, concentrations of Ca, Mg, K and Na ions, CEC, and
base saturation (Shearer & Sneddon, 2005; Sneddon & Clark, 2007). In analysing the data
in 2006, Sneddon and Clarke (2007) indicated that there was a large amount of variability in
16'h April 2010 2�J

CMA De-icing Agent: A review of environmental effects monitoring in New Zealand 1998-2009
soil variables amongst sites and an absence of any discernible trends in the data. They
expected a high degree of heterogeneity in road verge soils and indicated that road
maintenance activities and storm events led to significant differences in soil characteristics
sampled at a single site, concluding that observed changes in soil variables could not be
attributed to any single factor such as CMA in road runoff.
A similar approach to soil sampling has been used at a number of sites in the Canterbury
and West Coast regions since 2001 (Table 9), with analysis undertaken for the same suit of
variables, with the inclusion of volume weight. However, in the case of these sites a shorter
transect length of up to 5m was used (Beale, 2004; Beale, 2008). Beale (2008) states that
variations in cation concentrations occurred in the soil samples collected along treated and
untreated sections of highway. However, they showed little evidence of upward trend in
concentration during the winter period. Concentrations of cations generally remained within
the natural range expected for pastoral and upland soils. Overall Beale (2008) concludes
that there is no evidence of any adverse effects on soils that could be attributed to the use
of CMA.
8.8 Ecotoxicology
A New Zealand study to test the toxicity of CMA to aquatic organisms was undertaken by
Ryder Consulting Ltd in 2002. The study involved conducting a Whole Effluent Toxicity
(WET) test to determine toxic responses of three species of aquatic invertebrate and two
fish species. The results of the test suggested that direct CMA application into an aquatic
environment would have a significant impact on macro-invertebrates and fish life, causing
mortality at relatively low concentrations. In the case of the mayfly Deleatidium LCSo (50%
mortality) occurred in a test solution at 3% of the CMA solution concentration usually
applied to the highway, which typically contains 360g/L of CMA. However, the report noted
that it is important to understand that the laboratory situation is distinct from the field. In the
laboratory tests, the organisms were exposed in standing water, in which precipitate readily
formed a thick stationary layer. In flowing water physical mixing will interfere with
precipitation formation with currents constantly removing precipitation formed (Thompson,
2002). This probably explains why no toxic effects on aquatic life have ever been observed
by field monitoring.
It seems likely therefore that CMA concentrations in most field situations are likely to be
well below critical thresholds for causing toxic effects on aquatic life. Campbell (2006) used
the results of Thompson (2002) to assess CMA toxicity risk to a number of Canterbury
waterbodies; Lake Lyndon and an unnamed tributary, and Lake Pearson and its' tributary
the Craigieburn Stream. The assessment concluded that even under "worst case scenario"
conditions of multiple CMA applications that CMA concentrations would be well below any
concentrations that are toxic to aquatic life (Campbell, 2006). However, the results do
indicate that some caution may need to be exercised where direct runoff of stormwater
containing CMA enters small pools or ponds (Thompson, 2002).
16'h April 2010 26

CMA De-icing Agent: A review of environmental effects monitoring in New Zealand 1998-2009
9 Risk assessment
9.1 Introduction
This section of the report uses the evidence concerning the environmental effects of CMA
accumulated through environmental monitoring and testing of CMA in New Zealand, as well
as overseas experience, to assess the risk of CMA causing significant negative
environmental impact when used to treat icy roads. According to the Environmental Risk
Management Authority in New Zealand (ERMA) risk is measured in terms of the likelihood
of occurrence and magnitude of consequence (ERMA, 2009). Risk can therefore be
expressed as:
Risk = Consequence x Likelihood
The following scale of terms can be used to express likelihood of an effect occurring:
Descriptor Description
Highly improbable Almost certainly not occurring but cannot totally be ruled out
Very unlikely Considered only to occur in very unusual circumstances
Unlikely (occasional) Could occur, but not expected to occur under normal operating
conditions
Likely A good chance that it may occur under normal operating
conditions
Highly likely Almost certain, or expected to occur if all conditions met
Source: ERMA 2009
16'h April 2010 27

CMA De-icing Agent: A review of environmental effects monitoring in New Zealand 1998-2009
Similarly adverse consequence can be defined by the following terms:
Descriptor Description
Minimal Highly localised and contained environmental impact, affecting a
few (less than 10) individual members of communities of flora or
fauna, no discernible ecosystem impact.
Minor Localised and contained environmental impact, some local plant
or animal communities temporarily damaged, no discernible
ecosystem impact.
Moderate Measurable long-term damage to local plant and animal
communities, but no obvious spread beyond defined
boundaries, medium term individual ecosystem damage, no
species damage.
Major Long term/irreversible damage to localised ecosystem but no
species loss.
Massive Extensive irreversible ecosystem damage, including species
loss.
Source: ERMA 2009
Using these descriptors a matrix defining level of risk can be constructed as follows:
Magnitude of effect
Likelihood Minimal Minor Moderate Major Massive
Highly improbable A A A B B
Very unlikely A A B B C
Unlikely A B B C C
Likely B B C C D
Highly likely B C C D D
Source: ERMA 2009
Where: A = negligible
B= low C= medium D= high
16'h April 2010 2$

CMA De-icing Agent: A review of environmental effects monitoring in New Zealand 1998-2009
These descriptors and the risk matrix are useful tools for determining level of risk. However,
some caution is needed in their application. The wide range of environmental variables that
come into play when assessing the risk associated with CMA in different situations means
that the overall risk may be negligible when considering CMA application in most situations,
but might be higher when considering certain very specific situations.
9.2 Dissolved oxygen
The biodegradation of CMA into its component parts has been shown by laboratory and
field experiments to result in a biological oxygen demand with the potential to cause
dissolved oxygen depletion in receiving waters (Horner, 1988; LaPerriere & Rea, 1989;
Eheart et. al. 1993). Oxygen depletion was identified as the major risk associated with the
operational use of CMA by early studies (Horner & Brenner, 1992). However, it was
expected that in most operational situations dilution would result in negligible effects
(MacFarland & O'Reilly, 1992).
Dissolved oxygen monitoring has now been undertaken on 15 streams in New Zealand for
periods of up to 5 years. Monitoring has included spot measurements and continuous
logging in some streams. There have been no observed effects that could be attributed to
the use of CMA. Furthermore, direct application of CMA to a lowland stream at case worst
case scenario levels failed to produce any discernible response in DO levels (Wright-Stow
et. al., 2008).
The potential for CMA to cause oxygen depletion has been clearly demonstrated
experimentally. However, the monitoring undertaken in New Zealand shows that there are
very few real field situations were CMA is ever likely to reach concentrations sufficient for it
to cause significant oxygen depletion in receiving waters. The weight of evidence points to
the likelihood of CMA causing a significant adverse effect in most operating situations as
being highly improbable. Furthermore, even if an effect were to occur in flowing water, it is
likely to be no more than minor i.e. localised and temporary, resulting in no long-term
impact on the stream. Consequently, the level of risk associated with using CMA to treat icy
highways in terms of oxygen depletion and associated potential negative effects is
considered negligible for almost all normal operating scenarios. Either there will be
sufficient dilution of CMA in the stormwater and receiving environment, and/or CMA will
biodegrade before it reaches the waterbody.
The evidence indicates there are likely to be only two situations where CMA could cause
significant oxygen depletion in receiving waters. An accidental spillage of CMA into a
waterbody could introduce sufficient of CMA quantities such that a substantial depletion of
DO could occur. However, this is not a normal operating situation and this risk equally
applies to many other substances transported along New Zealand's State Highway network
on a daily basis. Such risks are managed by appropriate containment and transportation
systems, backed-up by spill response procedures.
The other circumstance where CMA could cause significant oxygen depletion is where
substantial amounts of CMA enter a very small waterbody, immediately adjacent to the
highway. The risks surrounding such a scenario are considered further in section 9.8.
16'h April 2010 29

CMA De-icing Agent: A review of environmental effects monitoring in New Zealand 1998-2009
9.3 Aquatic fauna
Aquatic invertebrate populations have been monitored in 19 streams receiving CMA in
stormwater runoff in New Zealand for periods of up to 11 consecutive years. Monitoring has
included pre and post season sampling, and in some cases monthly sampling during the
winter season. No effects have been observed on aquatic invertebrate populations that
could be distinguished from natural variations. The most recent monitoring undertaken in
2009 involved a very small tributary of the Mangapapa Stream (between 0.5 and 1.Om)
which intersects a treated section of SH5 (Taupo-Napier) near Te Haroto. This is the
highest risk scenario monitored in North Island with stormwater from a curbed and
channelled drainage system conveyed via a lined channel directly to the stream. However,
despite the application of CMA no significant effects on the aquatic invertebrate populations
were observed. These findings are consistent with there having been no significant oxygen
depletion events or toxic impacts of CMA, and concur with the predictions of studies
overseas regarding low potential impact on aquatic life in the quantities typically used to
treat highways for icing (Horner, 1988; MacFarland & O'Reilly, 1992).
The weight of evidence points to the likelihood of CMA causing a significant adverse effect
on aquatic fauna, in almost all operating situations, as being highly improbable. Even if an
effect were to occur in flowing water it is likely to be no more than minor i.e. localised and
temporary, resulting in no long-term impact on the stream. Consequently, the level of risk
associated with using CMA to treat icy highways in terms of adverse effects on aquatic is
considered negligible for almost all normal operating scenarios. Either there will be
sufficient dilution of CMA, or CMA will biodegrade before it reaches the waterbody.
Impacts on aquatic organisms, if they were to occur, would most likely be the result of
dissolved oxygen depletion or toxic effects. Many aquatic invertebrate taxa are sensitive to
lowering of dissolved oxygen levels and will disappear from streams if oxygen levels drop
below critical thresholds. Similarly, fish can be adversely impacted by lowering of dissolved
oxygen levels, to different degrees depending on the species. CMA has also been shown to
be toxic to aquatic invertebrates and fish in laboratory tests (Thompson, 2002). However,
the monitoring of streams in New Zealand has found no effect on aquatic invertebrates in
flowing water environments. Invertebrates are monitored as indicators of the health of
aquatic ecosystems, and under most circumstances if the invertebrates are unaffected, it
can be assumed that the fish are similarly unaffected.
The only scenario, where CMA could impact adversely aquatic life on aquatic life is a
scenario where CMA enters a very small poorly flushed waterbody in significant
concentrations e.g. DO depletion, and algae and bacteria stimulation, have measured in
Alaskan ponds of about 2,600m3. The risks surrounding such a scenario are considered
further in section 9.8.
9.4 Benthic algae and heterotrophic growths
Benthic algae and heterotrophic growth monitoring has been undertaken in 18 streams
receiving CMA in stormwater runoff in New Zealand for periods of up to 11 consecutive
years. Monitoring has included pre and post season sampling, and in some cases within
season sampling. There have been no observed effects. These findings are consistent with
16'h April 2010 3�

CMA De-icing Agent: A review of environmental effects monitoring in New Zealand 1998-2009
the predictions of overseas studies which anticipated minimal impact on aquatic life in the
vast majority of operational situations (Horner, 1988; MacFarland & O'Reilly, 1992).
There is a significant body of evidence that indicates that the likelihood of CMA causing a
significant promotion of the growth of benthic algae and heterotrophic growth in operating
situations, as being highly improbable. In the absence of any observed effects from
monitoring, the level of risk associated with using CMA to treat icy highways in terms of
promoting benthic algae and heterotrophic growth is considered negligible. Either there will
be sufficient dilution of CMA or CMA will biodegrade before it reaches the waterbody.
One of the principal original concerns regarding CMA was that it could have a fertilising
effect due to phosphorus impurities, and this amongst other effects could promote algal
growth if it entered streams in sufficient quantities. The CMA used in New Zealand is
supplied by Cryotech based in the US. This CMA contains very low levels of phosphorus
approximately 56g/tonne. This is a very small quantity. Typically CMA application rates
range between 56kg/km and 315kg/km (the latter rate tending only to be applied in a few
locations such as the summit of the Desert Road). However, even assuming the upper end
of the range the quantity of phosphorus being applied per kilometre is 17g. Even if all of this
washed off into a single small stream during a rain event (highly improbable in itself) this is
insufficient to trigger a growth response in benthic algae, particularly given the low
temperatures found in streams during period when CMA is likely to be applied.
To put this into context if the 17g of phosphorus applied to a 1 km length of highway all
washed into a small to medium sized stream flowing at 100L/s during a rain event lasting 1
hour, the average discharge rate to the stream would be 4.7mg/s (calculated from
17g/3600s). After dilution in the stream this would increase average phosphorus levels by
47µg/L during the 1 hour discharge period (calculated as 4.7mg/s divided by 100 L/s). This
is twice the recommended default trigger value given in the ANZECC water quality
guidelines (2000) for total phosphorus in a slightly impaired upland river, which is set at
26µg/L. However, when increased phosphorus loading to the system is average out over a
24 hour period this amounts to an increase of just 2µg/L over the day, which is well below
the trigger threshold. In addition there will be dilution as it is washed off the road (e.g. a
10mm rain event would result in an additional 2 x dilution for a 100 L/s stream). It needs to
be stressed that this is an extreme scenario. Small to medium sized streams typically have
a much smaller contributing road pavement areas than the 1 km used for this example.
Furthermore, not all the CMA would be expected to reach the stream as there would be
retention by roadside vegetation and soils. The calculation assumes the highest CMA
application rate which in practice is used in very few situations. Typical phosphorus
concentrations would be closer to 3g/km rather than 17g/km. It is therefore likely that
phosphorus dilution in real field situations would be significantly higher than these
calculations indicate.
Unlike algae, the concern relating to heterotrophic growths was that the bacteria would
proliferate as a result of breaking down the acetate molecule. The lack of any sign of
heterotrophic growth proliferation is likely to be the result of a sufficient dilution. It likely that
CMA would have to be applied in much higher quantities and with much frequently than
necessary to treat ice on roads before any increase in heterotrophic growths was possible.
16'h April 2010 31

CMA De-icing Agent: A review of environmental effects monitoring in New Zealand 1998-2009
However, even under such circumstances low water temperatures would tend to suppress
growth.
9.5 Vegetation health
Roadside vegetation health has been monitored in New Zealand at 21 locations adjacent to
sections of highway receiving CMA in stormwater runoff and potentially affected by spray
drift, for up to 8 consecutive years. Monitoring has included pre and post season sampling,
and in some cases monthly sampling during the winter season. There have been no
observations made at these monitoring locations that indicate discernible effects relating to
the use of CMA. These findings are consistent with the predictions of early overseas
studies that anticipated no significant effect on roadside vegetation (Horner, 1988;
MacFarland & O'Reilly, 1992). More recent greenhouse tests undertaken by Akbar, et. al.
(2006) in the UK, also support these observations.
Such significant changes in the physical state of roadside vegetation that have been
observed during monitoring have been clearly linked to roadside maintenance practices
such as mowing and herbicide treatment (Shearer & Sneddon, 2005; Sneddon & Clark,
2007). Roadside maintenance practices, which occur to varying degrees throughout the
State Highway Network, are an important consideration when assessing the potential effect
of CMA on roadside vegetation health. Such practices regular bring about significant
changes in the stature and condition of road side vegetation. Against such maintenance
induced changes and also edge effects, such as greater exposure to wind and potential for
weed invasion, the adverse health effects from the use of CMA, were they to occur, would
need to be substantial to be discernible. However, there is no substantial or consistent
evidence that CMA is having any significant impact on roadside vegetation health.
Furthermore, Sneddon and Clark (2007) observed health spring growth in all locations
sampled, indicating that any minor, unobservable effects as may have occurred are likely to
be short-lived.
The evidence provided by this monitoring indicates that the likelihood of CMA causing a
significant adverse health effects in roadside vegetation, as being highly improbable. Any
small effects that may be occurring cannot be separated from other disturbances occurring
within the road margin, and therefore effects are at most minor. Overall, the level of risk to
roadside vegetation health from CMA used to treat icy highways is considered negligible.
9.6 Vegetation species composition
Vegetation species composition has been monitored in New Zealand at 26 locations
adjacent to sections of highway receiving CMA in stormwater runoff and potentially affected
by spray drift, for up to 10 consecutive years. Monitoring has included pre and post season
sampling, and in some cases monthly sampling during the winter season. Most of the
monitoring has involved semi-quantitative visual assessments and comparison of
photographic records, although monitoring of plots on the Desert Road and at the other
Central North Island has involved more quantitative analysis. Monitoring at locations in the
South Island for up to 8 consecutive seasons has found no evidence of enhanced weed
growth which has been supported by consistently low soil nutrient levels in soil samples
taken from the same locations (Beale, 2003; New Zealand Transport Agency, 2009).
16'h April 2010 32

CMA De-icing Agent: A review of environmental effects monitoring in New Zealand 1998-2009
Similarly, monitoring in the North Island has found no evidence of significant changes in
species composition.
One of the original concerns regarding CMA was that it could have a fertilising effect on
roadside plant communities due to phosphorus impurities, promoting weed growth and
bringing about undesirable changes in species composition (Works Consultancy Services,
1996). Information subsequently provided by the US supplier, Cryotech, indicates that the
CMA supplied by them contains very low levels phosphorus approximately 56g/tonne. This
is a very small quantity. Even at extreme treatment levels of 315kg/km (only applied in a
few locations such as the summit of the Desert Road) phosphorus would only be entering
the roadside environment at the rate of 17mg per linear metre per application.
Given the absence of any evidence from observations of vegetation and soils that there has
been any fertilising effect from using CMA over the past 10 years, it is considered highly
unlikely that CMA is causing significant adverse changes in roadside vegetation species
composition. Furthermore, the dynamic nature of the road edge environment means that
minor changes resulting from the use of CMA are likely to be masked by other much higher
impact activities such as roadside maintenance activities. Most CMA applied to the highway
enters the environment immediately adjacent to the highway and invariably this
environment has already been modified to a greater or lesser extent due to the presence of
the road, even in locations that would otherwise be considered natural. In the context of this
modification and ongoing disturbance, and considering all the available evidence, any
effects of CMA are unlikely to be more than minor, and for the most part are likely to be
undetectable. Overall, the level of risk to roadside vegetation species composition from
CMA used to treat icy highways is considered negligible.
9.7 Soils
Soil properties have been monitored in New Zealand at 27 (2 sites have yet to be re-
measured against original baseline) locations adjacent to sections of highway receiving
CMA in stormwater runoff and potentially affected by spray drift, for up to 10 consecutive
years. Monitoring has included pre and post season sampling, and in some cases monthly
sampling during the winter season. Variables analysed included pH, electrical conductivity,
cation exchange capacity, base saturation, calcium, magnesium, potassium and sodium.
None of the analysis indicated any significant changes or trends in soil characteristics at
any of the locations sampled. Seasonal trends of cation (Ca, Mg, K and Na) concentration
were observed by (Beale, 2003; New Zealand Transport Agency, 2009) and the
concentrations of Ca, Mg and K cations also tended to be higher in post winter samples
from treated sites. However, no cumulative increase in cation concentrations was recorded
in the treatment samples during the period 2001 and 2008
In analysing the data in 2006, Sneddon and Clarke (2007) indicated that there was a large
amount of variability in soil properties amongst sites and an absence of any discernible
trends in the data. However, they commented that they expected a high degree of
heterogeneity in road verge soils and indicated that road maintenance activities and storm
events led to significant differences in soil characteristics sampled at a single site,
concluding that observed changes in soil properties could not be attributed to any single
factor such as CMA in road runoff. This an important issue when assessing the potential
16'h April 2010 33

CMA De-icing Agent: A review of environmental effects monitoring in New Zealand 1998-2009
consequences for the use of CMA on soil properties. The soils most likely to be impacted
by CMA are those immediately adjacent to the road seal. However, invariably these soils
have previously been disturbed by road construction activities and continue to be subjected
to further disturbance by ongoing maintenance activities. Against such maintenance
induced changes and also other edge effects, it is likely to be very difficult to identify
anything other than substantial changes resulting from the use of CMA. No such changes
have been observed. It also raises the question as to the relative importance of minor, most
likely undetectable changes, resulting from CMA application, in the context of much more
substantial past and ongoing disturbances in these roadside soils.
Given the absence of any evidence to indicate that roadside soils have been significantly
altered by using CMA over the past 10 years, it is considered highly unlikely that CMA will
cause significant adverse effects in roadside soils. Furthermore, in the context past
modification and ongoing disturbance, and considering all the available evidence, any
effects of CMA are unlikely to be more than minor, and for the most part are likely to be
undetectable. Overall, the level of environmental risk from modification of roadside soils
resulting from CMA used to treat icy highways is considered negligible.
9.8 Small waterbodies
Extensive monitoring throughout New Zealand has not observed any adverse effects on
aquatic life in flowing waterbodies or any oxygen depletion that was attributable to the use
of CMA. Given that CMA at least has the potential to deplete oxygen in receiving waters
and has been shown to be toxic to aquatic life in sufficient concentrations, this lack of
observed effects must reflect the fact that CMA is not entering and remaining in flowing
streams in sufficient concentration, for a sufficient period of time to allow effects to occur,
given conditions within the receiving environment; temperature, bacteria levels and
potential for re-aeration. Furthermore, given that CMA is fully soluble in water, in flowing
systems it will not persist and will pass through the system, eventually discharging to the
ocean if it does not biodegrade in transit.
It is however at least theoretically possible that CMA could cause adverse impacts on
aquatic life if sufficient concentrations accumulated in a waterbody and persisted for a
period of time. Under such circumstances if the concentrations of CMA were high enough
oxygen depletion could occur and thresholds for inducing toxic impacts on aquatic life might
be exceeded. This risk was identified early on by a number of early studies (Horner, 1988;
LaPerriere & Rea, 1989) and latterly it has been raised again by Sneddon & Clarke, 2007.
However, there has been very little monitoring of the effects of CMA on small waterbodies
in New Zealand. The reason for this lack of monitoring is that very few risk situations have
been identified close to the highway. A study to identify small waterbodies at risk of
receiving appreciable quantities of CMA laden stormwater, found only two such
waterbodies close enough to the highway around the entire South-West Plateau network
(Opus International Consultants, 2008). Only one such waterbody has been identified as
being at risk of receiving CMA inputs adjacent to treated sections of highway around the
Central Waikato State Highway network. In only one of these three cases was there the
potential for direct stormwater runoff from the highway. This was from a short section of
highway (<100m long). Furthermore, the maximum concentration of CMA likely to occur in
this waterbody was estimated to be 7.8mg/L, which assumed that all the CMA from the
16'h April 2010 34

CMA De-icing Agent: A review of environmental effects monitoring in New Zealand 1998-2009
potentially contributing road surface discharged into the pond. This concentration is almost
an order of magnitude less than the 60mg/L estimated by LaPerriere & Rea (1989) to be
needed to cause oxygen depletion severe enough to be lethal to salmonids (Opus
International Consultants, 2008).
For CMA to present a significant risk to a small waterbody a number of conditions need to
be satisfied:
• The CMA needs to be able to reach the waterbody, which means that in most cases it
needs to be very close the highway with clear rapid drainage pathways from the
highway to the waterbody. The greater the distance from the highway and the less
direct the pathway, the less likely CMA will ever reach the waterbody.
• The waterbody needs to be small. Oxygen depletion modelling for Lake Pearson (1.8
km2) and Lake Lyndon (1 km2) indicated that even under "worst case scenarios" of
multiple applications during a cold winter there was unlikely to be any significant
depletion in dissolved oxygen concentrations and no risk of toxic effects on aquatic life
(Campbell, 2006). LaPerriere and Rea (1989) undertook studies on much smaller
waterbodies, on small ponds in Alaska. Their study was based on experiments where
the CMA added directly to test ponds was approximately equivalent to one chemical
application of CMA applied to 0.4km of a typical section of road (one lane) draining
entirely to a small pond. Their study observed a number of effects; calcium elevation
did occur but did not persist into the following summer; acetate from the CMA mixture
was apparently rapidly taken up by aquatic organisms and cycled for several months,
depleting dissolved oxygen in the water to below 4 mg/L; bacteria and algae both
appeared to be stimulated by CMA additions, indicated by higher standing crops in the
treated ponds. Cladocerans were also denser in treated ponds than in control ponds,
probably because food (bacteria and algae) was available. Salmonid fishes, were not
present, but if they had been, they may have been stressed by low dissolved oxygen
caused by CMA. The important point regarding these ponds however is that they were
very small. The size of the pond where a very large drop in DO was observed was
approximately 2600m3. To put this volume into perspective this equates to a pond with
a depth of 1 m having dimensions of approximately 50m x 50m, this is massively
smaller than Lake Pearson or Lake Lyndon.
• The waterbody needs a significant contributing area of road pavement. However, since
small waterbodies tend to have small catchments this would tend to favour small
waterbodies having small contributing areas of road pavement.
• A waterbody needs to support sufficient values to merit concern, if already significantly
compromised by other environmental impacts (e.g. small farm ponds with cattle
access) then CMA is unlikely to have a significant additional impact.
• Very shallow waterbodies supporting significant macrophyte growth might be expected
to be subjected to significant diurnal fluctuations in dissolved oxygen levels, particularly
during the summer months. Such waterbodies are likely to support faunas tolerant of
16'h April 2010 35

CMA De-icing Agent: A review of environmental effects monitoring in New Zealand 1998-2009
DO fluctuations that might result from the presence of CMA, and are unlikely to suffer
significant long-term effects, at least not caused by DO fluctuations.
Experience has shown, from the extensive use of CMA around the New Zealand State
Highway networks and associated environmental assessments, that the incidence of small
waterbodies occurring immediately adjacent to the highway is very low. Furthermore, the
circumstances where direct runoff from the highway pavement to such waterbodies occurs
are extremely rarer. The single situation that has been identified adjacent to the North
Island network where there is the potential for direct runoff has a small contributing road
pavement. As such, even assuming maximum runoff of CMA from the contributing road
surface, the risk of significant impact on dissolved oxygen levels has been assessed as low
(Opus International Consultants, 2008). Despite extensive use of CMA around the South
Island network no high risk situations for small waterbodies have been identified where
CMA is regularly applied.
Overall therefore, whilst there is a theoretical risk of adverse effects on small waterbodies
from substantial inputs of CMA, the circumstances where this actually occurs in a New
Zealand context have not been found. In the unlikely event that a high risk situation is
identified, then the particular risks of the situation need to be considered on their merit. The
most significant matter which needs to be determined is the maximum likely concentration
of CMA likely to accumulate in the waterbody, which is largely dependant on the size of the
contributing road surface relative to the volume of water in the waterbody. However, other
factors such as the degree of flushing due to water throughflow also need to be taken into
account as high turnover of water is likely to substantially to reduce the risk of adverse
effect. The most high risk scenario is a situation where a small waterbody has no outflow
and forms a completely closed system. Based on the work of LaPerriere and Rea (1989)
this is most likely to be a waterbody <3,OOOm3 volume. Such a system receiving substantial
direct runoff from a substantial area of highway surface is however an unlikely proposition.
Small, closed system waterbodies by their nature tend to have small catchments. In the
unlikely event that a high risk is identified, there are measures that could be employed to
mitigate adverse effects. These could include modifications to drainage pathways or
creation of buffer zones.
16'h April 2010 36

CMA De-icing Agent: A review of environmental effects monitoring in New Zealand 1998-2009
10 Conclusions
CMA has proved to be an effective winter maintenance tool in New Zealand in recent years
reducing icing and helping to keep roads open for longer. Its use has now been consented
on all parts of the South Island State Highway network and main parts of the Central North
Island network.
Over a period of more than 10 years extensive monitoring of the possible effects of CMA on
soils, vegetation, streams and lakes, in many parts of New Zealand has found no adverse
environmental effects that could be attributed to its use as a de-icing agent on the State
Highway network. These findings validate the conclusions of McFarland and O'Reilly (1992)
who predicted, after reviewing available laboratory and field trials undertaken up to that
time, that "negative environmental and toxicological impacts are likely to be insignificant in
the vast majority of CMA applications".
There is now a substantial body of evidence from the extensive monitoring undertaken in
New Zealand, and supported by the overseas laboratory and field trials, that the risk of
significant adverse effects resulting from normal operational use of CMA to vegetation
health, species composition and soil health is negligible. Furthermore, none of the
monitoring undertaken in New Zealand has found any discernible adverse effect on aquatic
life or fauna populations, and no evidence has ever been found that it causes a proliferation
of benthic algae or heterotrophic growths. It is therefore also reasonable to conclude that
the risk posed by CMA to aquatic environments and aquatic life, from normal operational
use, is negligible for most, if not all, receiving environments.
The highest risk scenario of CMA causing significant adverse environmental effects is a
situation where substantial quantities of CMA enter a small enclosed waterbody (<3000m3).
However, experience has shown from the extensive use of CMA around the New Zealand
State Highway networks and associated environmental assessments, that the incidence of
small waterbodies occurring immediately adjacent to the highway is very low. Furthermore,
such waterbodies as have been identified along sections of highway where CMA is
regularly applied have been assessed as being at low risk of significant adverse effects
occurring. Consequently, whilst there is a theoretical risk of adverse effects on small
waterbodies from substantial inputs of CMA, the circumstances where this actually occurs
in a New Zealand context have not been found, despite extensive use over many parts of
the State Highway network.
Overall, the environmental monitoring undertaken in New Zealand has confirmed the
original expectations that CMA would prove an effective and environmentally benign tool to
assist with winter maintenance of the State Highway Network. When considering the overall
merits of CMA, the negligible risk posed to most, if not all, receiving environments, needs to
be considered in the context of the substantial environmental advantages CMA has over
sodium chloride (the most extensively used de-icing chemical used overseas), and the
significant economic and safety benefits its use brings to the New Zealand travelling public.
16'h April 2010 37

CMA De-icing Agent: A review of environmental effects monitoring in New Zealand 1998-2009
Tables
..- �- -... . •• ��:
Year Total Number of hrs Total Number of Average Duration
Closures (hrs)
1993 210.00 8 26.25
1994 316.00 9 35.11
1995 396.50 8 49.56
1996 159.50 11 14.50
1997 168.75 11 15.34
1998 26.00 2 13.00
1999 96.00 10 9.60
2000 58.75 5 11.75
2001 59.25 3 19.75
2002 65.00 7 9.29
2003 68.20 5 13.64
2004 160.90 12 13.41
2005 93.00 9 10.33
2006 175.50 14 12.54
2007 30.00 3 10.00
2008 124.40 10 12.44
Average 137.98 Average Duration of 17.28
total/yr (hrs) closure (hrs)
Source: Opus International Consultants, 2009
16'h April 2010 3$

CMA De-icing Agent: A review of environmental effects monitoring in New Zealand 1998-2009
.. - . . -. . . �- -... •:� ��:
All NZ Ice/Snow Crashes I All SH 1 Desert Road Ice/Snow Crashes
injury Total
11 13 26 21362000
11 11 23 16105000
14 18 34 24215000
13 19 35 26954000
8 15 4 30 22952276
5 10 1 16 6068069
20 23 10 54 27205690
7 12 20 12973000
7 19 27 56 23026319
15 34 26 80 40345886
4 16 15 37 14820399
17 33 64 117 36317236
17 44 45 111 44097925
6 20 51 80 23015883
7 51 63 125 32406075
13 36 76 129 36132792
23 55 160 241 45684136
8 35 133 179 28889453
10 15 83 108 13894091
7 22 97 129 25020149
6 17 93 116 10624145
14 69 180 267 39436824
20 35 68 123 19484408
19 58 127 206 30323108
18 67 107 195 34225297
14 41 79 135 19773650
23 86 154 266 42041828
14 87 108 214 42974916
12 55 126 193 18590432
363 1016 1897 3345 $778 m
16th April 2010 39
1 1023000
0 0
2 2046000
1 121000
1 121000
0 0
2 242000
1 4318000
3 1099138
2 242000
1 1023000
3 114207
2 4356069
0 0
4 5615069
6 627069
6 307345
12 614690
3 977138
8 378483
0 0
6 205152
3 797384
5 4324400
11 1163840
4 803003
5 240328
5 290246
3 114246
100 $31 m

CMA De-icing Agent: A review of environmental effects monitoring in New Zealand 1998-2009
.. - . . . -. . � . . . . .
-. . ��•
Issuing Authority Year Year Application Extent of State Highway Environmental
consent current process where CMA use consented monitoring
first consent for latest required
granted granted consent
Environment Waikato 1997 2004 Notified Specified ice black spots Ongoing
around the Central Waikato
State Highway network
Horizons Regional Council 2005 2005 Notified State Highways 1, 4, 41,47, Ongoing
48 & 49 within the South-
West Plateau
Environment Bay of Plenty 2007 2007 Notified SH5 (Taupo-Napier) Not required
Hawkes Bay Regional Council 2008 2008 Non-notified Specified sections of SH2 Ongoing
and SH5
Tasman D. C./Nelson 2001 2006 Non-notified Entire State Highway Network Ongoing
C.C./Marlborough D. C.
Environment Canterbury 2001 2006 Limited Entire State Highway Network Ongoing
notified
West Coast Regional Council 2001 2006 Non-notified Entire State Highway Network Ongoing
No longer
Otago Regional Council 2000 2009 Non-notified Entire State Highway Network required
Environment Southland 2000 2006 Non-notified Entire State Highway Network No longer
required
16'h April 2010 4�

CMA De-icing Agent: A review of environmental effects monitoring in New Zealand 1998-2009
.. - � � . . . -. . .- . . . .- . .
Location Regional Period Significant effects
Authority observed
Mangamate Stream (SH1) Environment 4yrs Nil
Waikato
Te Piripiri Stream (SH1) (1999-2002)
Wharepu Stream (SH1)
Kuratau Stream (SH41) Environment 5yrs Nil
Waikato
(2005-2009)
Pigroot Stream (SH85) Otago 3yrs Nil
Regional
Leith Saddle/Pigeon Flat Stream (SH1) Council (2000-2002)
Kaikorai Stream (SH1)
Pass Burn, Lindis Pass (SH8) Otago 4yrs Nil
Regional
Manuka Stream (SH8) Council (2000-2003)
Little Hope River/Washout Creek (SH6) Tasman 3yrs Nil
Nelson
Upper Motupiko River (SH63) Marlborough (2004-2006)
Region
Collins River (SH6)
Wairau River
Unnamed stream
Craigieburn Stream (SH73) Environment 5yrs Nil
Canterbury
(2002-2006)
16'h April 2010 41

CMA De-icing Agent: A review of environmental effects monitoring in New Zealand 1998-2009
.. - � . .. -. . . . .- . .
Location Regional Period Significant effects observed
Authority
Wharepu Stream (SH1) Environment 6yrs Nil
Waikato
(1999-2004)
Mangamate Stream (SH1) Environment 11 yrs Nil
Waikato
Te Piripiri Stream (SH1) (1999-2009)
Kuratau Stream (SH41) Environment 5yrs Nil
Waikato
(2005-2009)
Makatote River (SH4) Horizons 2 years Nil
Regional
Hihitahi Stream (SH1) Council (2006-2007) NOTE: Monitoring at Taonui Stream
and Peter's Hill limited to pre-season
Taonui Stream (SH49) only as no CMA applied to adjacent
section of highway.
Peter's Hill (SH1)
Pigroot Stream (SH85) Otago 3yrs Nil
Regional
Leith Saddle/Pigeon Flat Council (2000-2002)
Stream (SH1)
Kaikorai Stream (SH1)
Pass Burn, Lindis Pass (SH8) Otago 4yrs Nil
Regional
Manuka Stream (SH8) Council (2000-2003)
Little Hope River/Washout Tasman 3yrs No effect that could be separated from
Creek (SH6) Nelson natural variation.
Marlborough (2004-2006)
Upper Motupiko River (SH63) Region
Collins River (SH6)
Wairau River
Unnamed stream
Craigieburn Stream (SH73) Environment 5yrs Nil
Canterbury
Longslip Creek (SH8) (2002-2006)
16'h April 2010 42

CMA De-icing Agent: A review of environmental effects monitoring in New Zealand 1998-2009
.. - � . .. -. . . . .- . .
Location Regional Period Significant effects observed
Authority
Mangapapa Stream tributary Hawke's Bay 1 yr Nil
(SH5) Regional
Council (2009)
16'h April 2010 43

CMA De-icing Agent: A review of environmental effects monitoring in New Zealand 1998-2009
.. - . � . .- . ..- . - . .. . . . . . .- .
..
Location Regional Period Significant effects observed
Authority
Wharepu Stream (SH1) Environment 6yrs Nil
Waikato
(1999-2004)
Mangamate Stream (SH1) Environment 11 yrs Nil
Waikato
Te Piripiri Stream (SH1) (1999-2004)
Kuratau Stream (SH41) Environment 5yrs Nil
Waikato
(2005-2009)
Makatote River (SH4) Horizons 2 years Nil
Regional
Hihitahi Stream (SH1) Council (2006-2007) NOTE: Monitoring at Taonui Stream
and Peter's Hill limited to pre-season
Taonui Stream (SH49) only as no CMA applied to adjacent
section of highway.
Peter's Hill (SH1)
Pigroot Stream (SH85) Otago Regional 3yrs Nil
Council
Leith Saddle/Pigeon Flat (2000-2002)
Stream (SH1)
Kaikorai Stream (SH1)
Pass Burn, Lindis Pass (SH8) Otago Regional 4yrs Nil
Council
Manuka Stream (SH8) (2000-2003)
Little Hope River/Washout Tasman Nelson 3yrs Nil
Creek (SH6) Marlborough
Region (2004-2006)
Upper Motupiko River (SH63)
Collins River (SH6)
Wairau River
Unnamed stream
Craigieburn Stream (SH73) Environment 5yrs Nil
Canterbury
Longslip Creek (SH8) (2002-2006)
16'h April 2010 44

CMA De-icing Agent: A review of environmental effects monitoring in New Zealand 1998-2009
.. - � . -.- . . . . .- . .
Location Regional Period Significant effects observed
Authority
Kanuka Forest (SH1 — Desert Road) Environment 5yrs Nil
Waikato
Beech Forest (SH1 — Desert Road) (1998-2002)
Fernland (SH1 — Desert Road)
Grassland (SH1 — Desert Road)
Upland conifer/broadleaved forest (SH30 Environment 5 yrs Nil
adj. Pouakani S.R.) Waikato
(2005-2009)
Montane conifer/broadleaved forest (SH 41
—Waituhi Saddle)
Pigroot Stream (SH85) Otago 4yrs Nil
Regional
Kaikorai Stream (SH1) Council (2001-2004)
Manuka Stream (SH8)
Pass Burn, Lindis Pass (SH8) Otago 8yrs Nil
Regional
Leith Saddle (SH1) Council (2001-2008)
Riwaka Valley Road Tasman 3yrs Nil
Nelson
Little Hope River/Washout Creek (SH6) Marlborough (2004-2006)
Region
Upper Motupiko River (SH63)
Whangamoa Hill (SH6)
Collins River (SH6)
Wairau River
Dune Slack Wetland
Unnamed stream
Craigieburn Stream (SH73) Environment 5yrs Nil
Canterbury
Longslip Creek (SH8) (2002-2006)
16'h April 2010 45

CMA De-icing Agent: A review of environmental effects monitoring in New Zealand 1998-2009
..- . -. � . -.- . . . . .- . .
Location Regional Period Significant effects observed
Authority
Craigieburn Stream (SH73) Environment 5yrs Nil
Canterbury
(2002-2006)
Lake Lyndon/Starvation Gully (SH73) Environment 7yrs Nil
Canterbury
Ahuriri Cutting (SH8) (2002-2008)
Rahu Saddle (SH7) West Coast 6yrs Nil
Regional
Council (2002-2007)
Tawhai (SH7) West Coast 3yrs Nil
Regional
Council (2007-2009) 2009 results still to be reported
Scotchman's Creek (SH7) West Coast 2yrs Nil
Regional
Council (2008-2009) 2009 results still to be reported
16'h April 2010 46

CMA De-icing Agent: A review of environmenta/ effects monitoring in New Zealand 1998-2009
.• - : ' • -�- • •- • •• • • • • •- • •
Location Regional Period Significant observed effects
Authority
Kanuka Forest (SH1 — Desert Road) Environment 10yrs Small potential effect on
Waikato tussock grassland identified
(1998-2009) after 10years. However, link to
use of CMA considered
Beech Forest (SH1 — Desert Road) unlikely.
Fernland (SH1 — Desert Road)
Grassland (SH1 — Desert Road)
Upland conifer/broadleaved forest (SH30 Environment 5 yrs Results of the first re-
adj. Pouakani S.R.) Waikato measurement in 2009 still
(2005-2009) under analysis.
Montane conifer/broadleaved forest (SH 41
— Waituhi Saddle)
Upland wetland - (SH47 — Lake Rotoaira
Wetlands)
Pigroot Stream (SH85) Otago 4yrs Nil
Regional
Kaikorai Stream (SH1) Council (2001-2004)
Manuka Stream (SH8)
Pass Burn, Lindis Pass (SH8) Otago 8yrs Nil
Regional
Leith Saddle (SH1) Council (2001-2008)
Riwaka Valley Road Tasman 3yrs Nil
Nelson
Little Hope River/Washout Creek (SH6) Marlborough (2004-2006)
Region
Upper Motupiko River (SH63)
Whangamoa Hill (SH6)
Collins River (SH6)
Wairau River
Dune Slack Wetland
Unnamed stream
16'h April 2010 47

CMA De-icing Agent: A review of environmental effects monitoring in New Zealand 1998-2009
.. - : . -. � . -.- . .- . .. . . . . .- .
.
Location Regional Period Significant effects observed
Authority
Craigieburn Stream (SH73) Environment 5yrs Nil
Canterbury
(2002-2006)
Lake Lyndon/Starvation Gully (SH73) Environment 7yrs Nil
Canterbury
Ahuriri Cutting (SH8) (2002-2008)
Rahu Saddle (SH7) West Coast 6yrs Nil
Regional
Council (2002-2007)
Tawhai (SH7) West Coast 3yrs Nil
Regional
Council (2007-2009) 2009 results still to be reported
Scotchman's Creek (SH7) West Coast 2yrs Nil
Regional
Council (2008-2009) 2009 results still to be reported
16'h April 2010 4$

CMA De-icing Agent: A review of environmental effects monitoring in New Zealand 1998-2009
.. - • � . . . . . .- . .
Location Regional Period Significant effects observed
Authority
Kanuka Forest (SH1 — Desert Road) Environment 5yrs Minor and temporary increases
Waikato in some constituents observed
Beech Forest (SH1 — Desert Road) (1998-2002) in roadside soils following CMA
application. However, no
Fernland (SH1 — Desert Road) significant effects attributable to
Grassland (SH1 — Desert Road) the use of CMA observed.
Kanuka Forest (SH1 — Desert Road) Environment 6yrs No effects attributable to the
Waikato use of CMA observed.
Fernland (SH1 — Desert Road) (2003-2008)
Lake Rotoaira—Waituhi Saddle-Pouakani Environment 5 yrs Plots still to be re-measured
Waikato post 2009 winter season
Mihi — 8 km west of Reporoa (2005-2009) therefore no results available
yet.
Pigroot Stream (SH85) Otago 4yrs Nil
Regional
Kaikorai Stream (SH1) Council (2001-2004)
Manuka Stream (SH8)
Pass Burn, Lindis Pass (SH8) Otago 8yrs Nil
Regional
Leith Saddle (SH1) Council (2001-2008)
Riwaka Valley Road Tasman 3yrs Nil
Nelson
Little Hope River/Washout Creek (SH6) Marlborough (2004-2006)
Region
Upper Motupiko River (SH63)
Whangamoa Hill (SH6)
Collins River (SH6)
Wairau River
Dune Slack Wetland
Unnamed stream
16'h April 2010 49

CMA De-icing Agent: A review of environmental effects monitoring in New Zealand 1998-2009
.. - • . -. � . . . . . .- . .
Location Regional Period Significant effects observed
Authority
Craigieburn Stream (SH73) Environment 5yrs Nil
Canterbury
(2002-2006)
Lake Lyndon/Starvation Gully (SH73) Environment 7yrs Nil
Canterbury
Ahuriri Cutting (SH8) (2002-2008)
Rahu Saddle (SH7) West Coast 6yrs Nil
Regional
Council (2002-2007)
Tawhai (SH7) West Coast 3yrs Nil
Regional
Council (2007-2009) 2009 results still to be reported
Scotchman's Creek (SH7) West Coast 2yrs Nil
Regional
Council (2008-2009) 2009 results still to be reported
16'h April 2010 �J�

CMA De-icing Agent: A review of environmental effects monitoring in New Zealand 1998-2009
References
Akbar, K. F.; Headley, A. D.; Hale, W. H. G.; Athar, M.; 2006. A comparative Study of De-
icing Salts (Sodium Chloride and Calcium Magnesium Acetate) on the Growth of Some
Roadside Plants of England. J. Appl. Sci. Environ. Mgt. March, 2006. Vo1.10 (1) 67-71.
Amrhein, C.S.; Strong, J.E.; Mosher, P.A. 1992. Effect of deicing salts on metal and organic
matter mobilization in roadside soils. Environmental Science and Technology 26: 703-709.
ANZECC; ARMCANZ; 2000. Australia and New Zealand Guidleines for Marine and
Freshwater Quality. Volume 1: The Guidelines. National Water Quality Management
Strategy Paper No 4.
Beale, S. 2003. Application of Calcium Magnesium Acetate (CMA) Monitoring Report 2003.
MWH New Zealand Ltd, December 2003.
Beale, S. 2004
Acetate Annual
December 2004.
Transit New Zealand: Environmental effects of Calcium Magnesium
Report for Canterbury-West Coast Region. MWH New Zealand Ltd,
Beale, S. 2008. NZ Transport Agency: CMA Monitoring Report 2008 Canterbury and West
Coast Regions. MWH New Zealand Ltd, December 2008.
Burkett, A.; Gurr, N. 2004. Icy Road Management with Calcium Magnesium Acetate to Meet
Environmental and Customer Expectations in New Zealand. Transportation Research
Circular E-0063: Snow Removal and Ice Control Technology. Transportation Research
Board, Washington D.C.
Campbell, C. 2006.
Memorandum from
Zealand Ltd.
Environmental Effects of Calcium Magnesium Acetate Deicing. Project
Craig Campbell, Kingett Mitchell Ltd. to Steve Higgs, Transit New
Eheart, J. W.; Ho, Wing-Ho; Gingrich, C. D.; Hauser, R. J.; Thompson, S. R. 1993.
Environmental effects of alternative de-icers: Review and assessment method for Calcium
Magnesium Acetate biochemical oxygen demand applied to lllinois example case.
Transportation Research Record 1387: 93-97.
ERMA New Zealand, 2009. Decision Making
and Evaluating Risks, Costs and Benefits.
March 2009.
A Technical Guide to Identifying, Assessing
Environmental Risk Management Authority,
Fischel, M. 2001. Evaluation of selected de-icers based on a review of the literature.
Colorado Department of Transportation, Reseach Branch. Report No. CDOT-DTD-R-2001-
15.
Horner, R. R. 1988. Environmental Monitoring and Evaluation of Calcium Magnesium
Acetate. Transportation Research Board, NCHRP Report 305, Washington D. C.
16'h April 2010 51

CMA De-icing Agent: A review of environmental effects monitoring in New Zealand 1998-2009
Horner, R. R.; Brenner, M. V. 1992. Effects of calcium magnesium acetate (CMA) on
dissolved oxygen in natural waters. Resources, Conservation and Recycling, 7(1992) 239-
265. Elsevier Science Publishers B. V.
Jamieson, N.; Dravitzki, V. 2006. Management of Skid Resistance under Icy Conditions on
New Zealand Roads. Land Transport New Zealand Research Report 293.
Joutti, A.; Schultz, E.; Pessala P.; Nyst. T.; Hellsten, P. 2003. Ecotoxicity of Alternative de-
icers. Journal of Soils & Sediments 3(4) 269-272.
Kingett Mitchell Ltd. 2004. Effects of CMA use on water quality and ecology of Craigieburn
Stream. Report prepared for MWH New Zealand Ltd. by Kingett Mitchell, December 2004.
LaPerriere, J. D.; Rea, C. L. 1989. Effects of calcium magnesium acetate de-icer on small
ponds in interiorAlaska. Lake and Reservoir Management 5: 49-57.
Levelton Consultant Ltd. 2007. Guidelines for the Selection of Snow and Ice Control
Materials to Mitigate Environmental Impacts. Transportation Research Board, NCHRP
Report 577, Washington D. C.
MacFarland, B. L.; O'Reilly, K. T. 1992. Environmental impact and toxicological
characteristics of calcium magnesium acetate. pp. 194-227. In: Chemical Deicers and the
Environment. Lewis publishers, Boca Rattan, Florida.
McLeod, M.; Fraser, S. 2009. Long-term impacts of CMA on soils beside and near the
Desert Road (SH1). Landcare Research Contract Report: LC0809/098. March 2009.
New Zealand Transport Agency, 2009 . Resource Consent 2005.059 - Application of CMA
in the Otago Region. Application for change or cancellation of resource Consent Conditions
Prepared for NZTA by Simon Beale, MWH New Zealand Ltd, June 2009.
Opus International Consultants, 2002. CMA De-icer Trial: Summary of monitoring results
1998-2001 and recommended future monitoring program. Opus International Consultants,
March 2002.
Opus International Consultants, 2008. South-West Central Plateau: CMA — Small Lentic
Waterbodies Assessment of Effects. Opus International Consultants, March 2008.
Opus International Consultants 2009. SH1 Taupo to the Desert Road Summit: Strategic
Corridor Evaluation. Opus International Consultants, September 2009.
Ray, D. E., 1998a. Desert Road De-icing Tests: Te Piripiri Stream Calibration. NIWA Client
Report: OIC80202/2, October 1998.
Ray, D. E., 1998b. Desert Road De-icing Tests: Wharepu Stream Calibration. NIWA Client
Report: OIC80202/3, December 1998.
Ray, D. E., 1999. Desert Road De-icing Tests: Second Wharepu Stream Calibration. NIWA
Client Report: OIC80202/4, March1999.
16'h April 2010 �J2

CMA De-icing Agent: A review of environmental effects monitoring in New Zealand 1998-2009
Ray, D.; Scarsbrook, M. 2000. Desert Road De-icing Trials: Winter 1999 Monitoring
Resu/ts. NIWA Client Report: OIC00207/1, March 1999 (Revised).
Ray, D.; Scarsbrook, M. 2002. Desert Road De-icing Trial: Winter2002 Monitoring Results.
NIWA Client Report: HAM2002-068, December 2002.
Rijkse, W. 1999. Soil analysis of Central Plateau State Highway plots to monitor chemical
de-icing (1999 Results). Landcare Research.
Rijkse, W. 2000a. Soil analysis of Central Plateau State Highway plots to monitor chemical
de-icing (March 2000 Results). Landcare Research.
Rijkse, W. 2000b. Soil ana/ysis of Central Plateau State Highway plots to monitor chemical
de-icing for 2000. Landcare Research.
Rijkse, W. 2001. Soil analysis of Desert Road monitoring chemical de-icing trials (2001
Results). Landcare Research Contract Report: LC0102/037.
Rijkse, W. 2002. Soil analysis of Desert Road monitoring chemical de-icing trials (2002
Results). Landcare Research Contract Report: LC0102/037.
Robidoux, P. Y& Delisle, C. E. 2001. Ecotoxicology evaluation of three deicers (NaCI,
NaFo, CMA) - effects on terrestrial organisms. Ecotoxicology and Environmental safety 48,
pp128-139.
Thompson, R. 2002. CMA Ecotoxicity Testing. Ryder Consulting Ltd. December 2002.
Scarsbrook, M. R.; Cooke, J. G.; Wilcock, R. J.; 1996. Use of chemical de-icing agents on
the Desert Road (SH1): Assessment of potential environmental effects (AEE). NIWA Ltd
Consultancy report WOR70206/1.
Shearer, K.; Sneddon, R. 2005. Environmental Effects of Calcium Magnesium Acetate:
Annual Report for Tasman/Nelson/Marlborough Region - 2005 monitoring. Cawthron
Report No. 1119, December 2005.
Smale, M. C.; Whaley. K. J.; Fitzgerald, N. B. 2001. Assessment of changes in vegetation
following three years of CMA trials on the Desert Road (State Highway 1). Landcare
Research Contract Report: LC0001 /. April 2001.
Smale, M. C.; Fitzgerald, N. B. 2003a. Roadside vegetation health beside the Desert Road
(State Highway 1) after the application of CMA in 2002. Landcare Research Contract
Report: LCO203/162. February 2003.
Smale, M. C.; Fitzgerald, N. B. 2003b. Changes in vegetation after five years of trials on the
Desert Road (State Highway 1). Landcare Research Contract Report: LCO203/163.
February 2003.
Smale, M. C.; Fitzgerald, N. B. 2008. Roadside vegetation health beside State Highways 30
and 41 in central North Island after application of CMA in 2008. Landcare Research
Contract Report: LC0809/113. November 2008.
16'h April 2010 53

CMA De-icing Agent: A review of environmental effects monitoring in New Zealand 1998-2009
Smale, M. C.; Fitzgerald, N. B. 2009. Changes in vegetation beside the Desert Road (SH1)
after ten years of CMA application. Landcare Research Contract Report: LC0809/114.
January 2009.
Sneddon, R. 2005. Environmental Effects of Calcium Magnesium Acetate: Annual Report
for Tasman/Nelson/Marlborough Region - 2006. Cawthron Report No. 1002, April 2005.
Sneddon, R.; Clark, K. 2007. Environmental Effects of Calcium Magnesium Acetate: Annual
Report for Tasman/Nelson/Marlborough Region - 2006. Cawthron Report No. 1119,
January 2007.
Tanner, D.Q. & Wood, T. M. 1999. The Effects of Calcium Magnesium Acetate (CMA) De-
icing Material on the Water Quality of Bear Creek, Clackamas County, Oregon, 1999. U.S.
Geological Survey. Water-Resources Investigations Report 00-4092.
Transit New Zealand, 2003. NZ Transport Agency Resource Consent 2005.059 -
Application of CMA in the Otago Region. Application for change or cancellation of resource
Consent Conditions. Prepared for TNZ by Simon Beale, MWH New Zealand Ltd, December
2003.
Wilde, H. 2007. Baseline chemistry of some central North Island soils before application of
calcium-magnesium acetate (CMA) as a de-icing agent. Landcare Research May 2007.
Winters, G. R.; Gidley, J.; Hunt, H. 1985. Environmental evaluation of calcium magnesium
acetate (CMA). California Department of Transportation, Sacramento, California,
FHWA/CA/TL-84/03.
Works Consultancy Service Ltd. 1996. Assessment of Environmental Effects - Chemical
De-icing: Application of Chemical De-icers, CMA, Desert Road, SH1 (Volume 1). Works
Consultancy Services, October 1996.
Wright-Stow, A. 2008a. Effects of CMA applied to the South West Central Plateau State
Highway Network on aquatic invertebrates: winter 2007 results. NIWA Client Report:
HAM2008-009, February 2008.
Wright-Stow, A. 2008b. Effects of CMA applied to the South West Central Plateau State
Highway Network on periphyton biomass: summer 2008 results. NIWA Client Report: HAM
2008-025, March 2008.
Wright-Stow, A.; Nagels, J.; Smith, B. 2008. Environmental evaluation of calcium
magnesium acetate (CMA) for highway deicing application on dissolved oxygen in a
lowland stream. NIWA Client Report: HAM2008-200, December 2008.
Wright-Stow, A. 2009a. Effects of CMA applied to the Desert Road and Highway 41 on
benthic growths, aquatic invertebrates and dissolved oxygen: Tenth annual report and
winter2008 results. NIWA Client Report: HAM 2008-195, June 2009.
Wright-Stow, A. 2009b. Effects of CMA applied to areas surrounding the Mangapapa
Stream (SH5) on aquatic invertebrates. NIWA Client Report: HAM 2008-174, December
2009. IN PREP.
16'h April 2010 54