Content uploaded by Shane Orchard
Author content
All content in this area was uploaded by Shane Orchard on Aug 05, 2024
Content may be subject to copyright.
Content uploaded by Shane Orchard
Author content
All content in this area was uploaded by Shane Orchard on Aug 05, 2024
Content may be subject to copyright.
ACCEPTED MANUSCRIPT
Published version: https://doi.org/ 10.1002/aqc.2898
Earthquake-induced habitat migration in a riparian spawning fish
has implications for conservation management
Shane Orchard1,2
Michael J. H. Hickford2
David R. Schiel2
1 Waterways Centre for Freshwater Management, University of Canterbury and Lincoln
University, Christchurch, New Zealand
2 Marine Ecology Research Group, University of Canterbury, Christchurch, New Zealand
Abstract
1. Galaxias maculatus is a riparian spawning fish that supports an important recreational
fishery in New Zealand with spawning habitat requirements strongly structured by
salinity gradients at rivermouths. This study reports changes to the spawning habitat
following a series of large earthquakes that resulted in widespread deformation of
ground surfaces in the vicinity of waterways.
2. Assessments of habitat recovery focused on two rivers systems, the Avon and
Heathcote, with pre-disturbance data available over a 20 year period. Recovery
dynamics were assessed by field survey and mapping of spawning habitat prior to and
on seven occasions after the disturbance event. Riparian land-use and management
patterns were mapped and analysed using overlay methods in a GIS.
3. Habitat migration of up to 2 km occurred in comparison to all previous records and
several anthropogenic land uses have become threats due to changed patterns of co-
occurrence. Incompatible activities now affect more than half of the spawning habitat in
both rivers, particularly in areas managed for flood control purposes and recreational
use.
4. The results are an example of landscape scale responses to salinity and water level
changes driven by tectonic dynamics. These dynamics are not the source of the stress
per se, rather, they have increased exposure to pre-existing stressors.
5. The case illustrates important principles for managing subtle, yet widespread, change.
Adaptive conservation methods and investments in information are priorities for
avoiding management failure following environmental change.
Keywords
Intertidal, estuary, conservation evaluation, fish, urban development, engineering
1. Introduction
1.1 Earthquake recovery context
The Canterbury region of New Zealand was affected by a sequence of major earthquakes in
2010 and 2011. The most devastating of these was a Mw 6.3 earthquake centred beneath the city
of Christchurch that caused widespread damage and loss of life (Quigley et al. 2016). After six
years of recovery activities the process has entered a more strategic phase. The focus is now on
longer term adaptation to environmental and societal change. Important land-use decisions
remain for many geographical areas and with regards to many aspects of the natural and built
environment. Examples relevant to waterway management include responses to water quality,
erosion, flood risk and coastal inundation issues, and the potential re-zoning of large tracts of
riparian and floodplain land. Existing statutory arrangements apply to many of the recovery
activities and identify institutional responsibilities. Due to the scale and impact of the event
bespoke legislation was also created. The organizations involved now include new planning
entities with specific tasks (Regenerate Christchurch 2017) and a wide range of interests across
central, regional, and local government, non-governmental organizations, and local community
groups.
Initially, urgent decisions were made to address risks to property and life, and to reinstate
essential infrastructure. Remaining decisions have the benefit of more time. There is a unique
opportunity to secure benefits through earthquake recovery planning in relation to historical
degradation of natural environments and improved resilience to future events. Natural values in
the affected areas have thus far received less attention, but include traditional cultural uses such
as the wild harvest of food and fibre (Jolly & Ngā Papatipu Rūnanga Working Group 2013;
Lang et al. 2012), risk reduction functions (Orchard 2014), nature-based recreation, and habitat
for many indigenous and migratory species with protected status. However, knowledge gaps are
a barrier to securing benefits through the planning process. Information requirements include
quantifying impacts of the earthquakes and identifying opportunities for future gains.
1.2 Riparian spawning habitat of īnanga
In the present study, the particular focus is Galaxias maculatus, or ‘īnanga’, a riparian spawning
fish. G. maculatus is an amphidromous species currently listed as ‘at risk - declining’ in the
New Zealand Threat Classification System (Dunn et al. 2018). Reversing the decline of īnanga
is addressed in many statutory documents as well as non-statutory plans and it is a priority issue
for Māori. Juvenile fish are harvested in an iconic recreational and culturally important fishery
(McDowall 1984). The harvest of īnanga and other ‘whitebait’ species creates an ongoing
tension between conservation and sustainable use. However, use and non-use interests share the
objective of enhancing īnanga populations. The protection of spawning habitat is an urgent and
practical goal due to a history of degradation associated with land-use changes near lowland
waterways (McDowall 1992; McDowall & Charteris 2006).
Īnanga has a specialized reproductive strategy that is synchronized with the spring tide cycle
and strongly influences the distribution of spawning sites (Burnet 1965). Spawning sites occur
close to the maximum upstream extent of saltwater intrusion and occupy only a narrow
elevation range (Taylor 2002). Eggs are laid in riparian vegetation just below the spring tide
high-water mark and hatch in response to inundation after a 2-4 week development period
(Benzie 1968). The composition and condition of riparian margins at these specific sites is
critical to spawning success (Hickford & Schiel 2011a).
This specificity suggested that earthquake-induced land deformation could affect habitat in
several ways. First, disturbance could reduce the availability or condition of existing spawning
sites, and enduring changes might result from vegetation recovery effects. Second, large scale
impacts were possible due to physico-chemical effects. This was the particular focus of the
study in light of suspected earthquake-driven hydrodynamic changes and the reported
structuring of habitat by salinity (Richardson & Taylor 2002; Taylor 2002). Because there was
no prior salinity baseline available, the focus was on direct detection of changes in the
distribution of spawning sites. By reconstructing a spawning site distribution baseline using data
from previous studies, this comparison was possible for the consideration of earthquake effects.
The objectives of the study were therefore to quantify the pre- and post-quake spawning site
distribution against riparian land uses and evaluate distributional effects to identify management
implications.
2. Methods
2.1 Study area
The two study catchments are the Avon River (Ōtākaro) and Heathcote River (Ōpāwaho)
(Figure 1). These are spring-fed, lowland waterways with small average base flows (approx. 2
and 1 cumecs respectively) originating within the city of Christchurch, New Zealand (White et
al. 2007). The catchments are heavily urbanized, particularly in their upstream reaches. The two
waterways are extensively channelized through the use of bank stabilization engineering and
flow regulation structures including flood-gates. The lower catchments support riparian
saltmarsh areas that contribute to the Avon-Heathcote Estuary / Ihutai (Figure 1). These are
remnants of a larger and relatively mobile ecosystem of coastal hydrological features (Kirk
1979).
Vertical seismic shifts and lateral spread were pronounced in the vicinity of Christchurch
waterways particularly towards the estuary (Hughes et al. 2015). Changes in ground levels in
and around the estuary were in the order of ± 0.5 m with a trend towards uplift in the south and
subsidence in the north (Beaven & Litchfield, 2012). Hydrodynamic modelling of the estuary
showed extensive bathymetric change and an estimated 15% reduction in the estuarine tidal
prism (Measures et al. 2011).
[insert Figure 1 here]
2.2 Pre-earthquake baseline
A literature review was completed to identify pre-quake spawning records augmented with
information from current researchers (M. Taylor, S. McMurtrie, C. Meurk, pers. comm.). This
resulted in a database of 14 technical reports and additional personal communications together
with records from the National Īnanga Spawning Database (www.inangaconservation.nz).
Historical spawning site data were restricted to information associated with observations of eggs
in riparian vegetation. All information was digitized in GIS by identifying coordinates for the
upstream and downstream extent of spatially discrete spawning sites using the original data
sources. Sites were defined as semi-continuous stretches of eggs identified through riparian
vegetation surveys on waterway margins. These locations were identified using the co-
ordinates, maps, photographs and text descriptions provided in technical reports and direct
communication with researchers.
2.3 Post-earthquake studies
A census-style survey methodology was used with the objective of detecting all spawning
occurrences at the catchment scale following the methods of Orchard & Hickford (2018). The
search areas were approximately 4 km reaches in each river (Figure 1). The survey area
extended from the downstream transition to saltmarsh vegetation, which is unsuitable for
spawning (Mitchell & Eldon 1991), to 500 m upstream of the inland limit of saltwater. In the
Avon this included the confluence with a prominent tributary to the north. The saltwater limit
was established using conductivity/temperature loggers (Odyssey, Dataflow Systems Ltd, NZ)
deployed during spring tide sequences and additional spot measurements using a handheld
conductivity/salinity/temperature meter (YSI Model 30, YSI Inc., USA). The survey period
included the peak spawning months (Taylor 2002) over two years. Surveys commenced five
days after the peak tide in the spring tide sequence and followed a set schedule to minimize
temporal confounding effects between months (Table S1). Reaches surveyed later in the
schedule were more sensitive to egg mortality effects due to the time elapsed since spawning.
Results are more likely to underestimate the extent of spawning occurrences in these areas, but
are comparable between months.
The search area was surveyed systematically in the first two months of the study by conducting
three searches for eggs within contiguous 5 m blocks along each riverbank. Each search
involved opening up the vegetation down to ground level at random locations within the block
following a transect line perpendicular to and spanning the high water mark. On subsequent
months, the survey effort was reduced to areas of potential habitat following a habitat
classification system (Orchard & Hickford 2018). Whenever eggs were found, the survey was
extended 50 m either side of the last occurrence to confirm the full extent of the spawning site.
Spawning sites were defined as the area occupied by continuous or semi-continuous patches of
eggs. Upstream and downstream extents were established and the width of the egg band
measured on the centreline of the search transects within the extent of the site (minimum three).
Zero counts were recorded where these occurred such as when the egg patch was not
continuous. Area of occupancy (AOO) was calculated as length x mean width. The total number
of eggs present was calculated by sub-sampling patches. At each width measurement location,
eggs were counted in a 10 x 10 cm quadrat placed in the centre of the egg band. Productivity
was calculated as mean egg density x AOO.
Riparian land uses and management activities were mapped in the field using 0.075 m
resolution post-quake aerial photographs (Land Information New Zealand 2016).
Anthropogenic stressors were identified based on reported incompatibility with īnanga
spawning sites (Hickford & Schiel 2011a, 2011b; Mitchell 1994). Areas affected were
delineated using aerial photographs in the field and digitized for overlay analysis in QGIS
v2.8.18 (QGIS Development Team 2016). Four classes of land use activities were classified as
threats to spawning habitat. These were bank stabilization using engineered structures, invasive
species control, mowing of recreation reserves, and vegetation removal for flood management.
Threats from riverbank engineering were defined on the basis of surfaces devoid of any
vegetation capable of supporting spawning (Mitchell 1994). Examples include retaining walls,
bridge abutments, riprap, and other bank stabilization works. Invasive species control was
classed as a threat where it involved spraying or extensive mechanical clearance (e.g. using
scrub cutters, line trimmers & similar). This recognizes that vegetation suitable for spawning
may take several months to recover following clearance activities (Hickford & Schiel 2014).
Mowing was classed as a threat where it resulted in short grass conditions at the top of the
riverbank in the location of spawning habitat.
3. Results
3.1 Pre-earthquake spawning distribution
Eighteen pre-quake spawning studies spanning a 25 year period were identified, most of which
involved surveys in both catchments. Thirteen of these had quantified spawning in the Avon and
nine in the Heathcote (Table 1). In some years field surveys were conducted that did not find
any spawning and these records are not shown in Table 1. In the Avon, most of the spawning
occurrences have been in the Avondale Road area (Figure 2a) and often found a short distance
upstream from the road bridge on the true right (Table 1). The maximum extent of pre-quake
spawning sites recorded in any one year was 2000 m in 2007. This also represents the maximum
extent of the spawning reach based on all known records.
In the Heathcote, most of the records have been in the vicinity of Opawa Road (Figure 2b).
Although the downstream limit of all records is c. 1 km further downstream this relates to only
two observations of spawning below Opawa Road in the 25 year period (Table 1). However, the
first spawning recorded in the catchment was much further upstream (> 3 km). At the time the
river was under the influence of a floodway, constructed in 1986, that effectively shortened the
length of the river. In 1994 a tidal barrage was installed to reduced saline intrusion and this
resulted in a shift of c. 2km downstream in the upstream limit of spawning (Taylor 2004).
Although these variations in the location of pre-quake sites complicate historical analyses the
location of spawning has been remarkably consistent since 1994 (Table 1) centred on the Opawa
Road site. The maximum extent of pre-quake spawning recorded in any one year was 1050 m in
2004 (Table 1) associated with the discovery of small sites in a reserve ca.1 km upstream of
Opawa Road.
[insert Table 1 here]
3.1 Post-quake studies
Spawning distribution
A total of 85 spawning sites were identified in the 2015 post-quake survey. These were
distributed along 2.4 km of riverbank in the Avon and 2.5 km in the Heathcote. In both rivers
there were marked differences in the spawning distribution in comparison to previous records
(Figure 2). In the Avon, the spawning reach had expanded approximately 250 m upstream and
180 m downstream of the previous extent. In the Heathcote, the changes were more dramatic
with spawning recorded 1.5 km downstream of all previous records (Figure 2a).
The 2016 survey identified 101 spawning sites, some of which represented repeat use of 2015
sites. In the Avon, the upstream and downstream limits were very close to those recorded in
2015. In the Heathcote, the upstream limit was also similar to 2015, but the spawning reach
extended a further 400 m downstream. The furthest downstream sites were found in the final
month of the survey at a distance of nearly 2 km from all pre-quake spawning records and ca.3
km from the previous centre of spawning at Opawa Road (Figure 2b).
[insert Figure 2 here]
Distribution of threats and protected areas
There are three areas managed specifically to protect spawning habitat at well-known sites
(Figure 3). The protection mechanisms include recognition in local authority plans and
implementation of compatible riparian management on the ground. There is also a considerable
reach in the lower Heathcote that is not subject to vegetation clearance for flood or reserves
management purposes. Part of this reach is characterized by tall woody riverbank vegetation and
the remainder is downstream of the tidal barrage where there is less need for channel works
associated with flood management.
Collectively, the four classes of threats affect a large proportion of the study area (Figure 3). In
both rivers, threats from riverbank engineering occupied only a small proportion of the post-
quake spawning extent (Figure 3). Extensive channelization using gravel embankments also
occurs in the Avon. Although the area available for spawning may be reduced by these
structures they were not classified as threats based on observations of spawning if suitable
vegetation co-occurred. Invasive plant species that have historically been the subject of spraying
or mechanical clearance are widespread throughout the study area. In the Avon the major
concern is yellow flag iris (Iris pseudacorus). It is distributed throughout the spawning reach
with the exception of sections engineered with gabion baskets and in Lake Kate Sheppard. This
species is largely absent from the Heathcote and instead reed canary grass (Phalaris
arundinacea) is the major concern and is the dominant canopy species in many areas. In
addition, Glyceria maxima and Rubus fruticosus are present there. There were no major spray
eradication campaigns during the study period despite the severe level of infestation. Decisions
on control will be required in the near future under regional pest management plans. Riparian
mowing occurs in discrete areas in both river systems associated with a network of parks and
reserves (Figure 3). Vegetation control for flood management was conducted on a semi-regular
basis in both rivers using scrub cutters or line trimmers. This work is regularly scheduled
through the Avon study area with the exception of the two areas protected for spawning habitat
(Figure 3a) and in the upper section of the Heathcote study area (Figure 3b).
[insert Figure 3 here]
Area of occupancy and egg production
In 2015, the total area of occupancy (AOO) of spawning habitat was 152.5 m2 in the Avon and
75.4 m2 in the Heathcote as calculated using maximum figures recorded at each site across all
four surveys. Total egg production in 2015 was 11.8 million eggs (Avon 6.9 x 106, Heathcote
4.9 x 106). In 2016, egg production was higher (Avon 13.9 x 106, Heathcote 5.0 x 106) despite
the survey period being reduced to only three months. The AOO was also higher in both rivers
(Avon 472.9 m2, Heathcote 99.1 m2) although average egg densities were lower. The marked
increase in AOO in the Avon was associated with several new large spawning sites that were
not utilized in 2015 in addition to re-use of other sites. In both years, AOO and productivity
were not evenly distributed across the study area and high production was not always correlated
with AOO due to differences in egg densities (Figure 4). Egg densities of >10 eggs cm-2 were
recorded at several sites with the highest being 13.5 cm-2.
[insert Figure 4 here]
Effectiveness of protected areas
In the Avon, the proportion of the AOO occurring in protected areas was 70% in 2015. In 2016
this figure had decreased to only 28% reflecting many new sites discovered in other locations.
In the Heathcote, the proportion of AOO protected was very low (11% and 6% for the two years
respectively) reflecting the discovery of many spawning sites at locations never previously
known for spawning. Egg production was also considerable outside of the protected areas
(Figure 5). In the Avon, the proportion of egg production outside the protected areas was 28%
in 2015 and 38% in 2016 (Figure 5a). In the Heathcote, 82% of egg production occurred outside
of the protected areas in 2015 and 98% in 2016 (Figure 5b). On average across the two years of
post-quake studies, only 4.5% of the spawning reach was protected in the Heathcote and 27.6%
in the Avon (Figure 6). Although the timing of threats was variable in relation to the presence of
eggs, vegetation clearance for reserves and flood management purposes was observed at many
of the unprotected spawning sites after egg deposition had occurred (see Supplementary
Material). Repeat egg surveys at some of these sites after the vegetation clearance indicated
close to 100% egg mortality, consistent with previous studies (Hickford & Schiel 2014).
[insert Figure 5 here]
[insert Figure 6 here]
4. Discussion
4.1 Evidence for īnanga spawning habitat migration
There are several limitations for accurately characterizing the pre-quake spawning baseline.
They include the variable frequency, extent and intensity of historic surveys, and temporal
effects in relation to the peak months of spawning activity, all of which may lead to under-
estimation of the areas utilised. These sources of inaccuracy have a bearing on the identification
of change in relation to the distribution and area of occupancy spawning habitat. Despite this,
the Christchurch waterways have the most extensive record of īnanga spawning for any
catchment in New Zealand in terms of the total number of surveys conducted and the length of
the survey record (Taylor 2002). The relatively consistent results obtained by researchers over
the historical pre-quake period are another important aspect. Additionally, we have taken the
maximum values identified over all records. This produces a precautionary approach in relation
to the area and extent of pre-quake spawning habitat recorded in most years. In both catchments
these maximum values were atypical of the full survey record suggesting that they may over-
estimate the relevant habitat parameters. However, the prevalence of degraded riparian
vegetation in the study area is likely to have caused high egg mortality if spawning occurred in
those areas. This effect reduces the detectability of spawning sites in field surveys (Orchard &
Hickford 2018), and was specifically addressed in the design of post-quake surveys by attention
to the timing of surveys in relation to the estimated date of spawning events. For these reasons
the peak detected over all pre-quake records is considered to be the best estimator of typical
spawning activity over this period.
In the Avon, the majority of historical spawning has been recorded at the Avondale site (Figure
2a). In this vicinity, the spatial extent of spawning steadily increased since discovery of the site in
1989 and was assisted by protection from mowing (Taylor 1999). In 2004, new sites were
identified further downstream in the mainstem, and in 2006 spawning was found at Lake Kate
Sheppard and then regularly thereafter. This is an area of restored riparian margins in a tributary
waterway and lake system located close to the mainstem. In the Heathcote, the pre-quake
distribution shifted downstream in association with construction of the tidal barrage in 1994 to
reduce saltwater intrusion upstream (Taylor 1995, 1998). Subsequently, spawning has been
centred on the Opawa Road site with only two sites have been recorded further downstream in all
known records. Earthquake-induced migration of habitat a further 1.5 km downstream in 2015
and 1.9 km in 2016 represents a major change in spawning habitat distribution.
4.2 Effectiveness of protected areas
A high proportion of īnanga spawning now occurs outside of the areas designated for spawning
site protection. Risk exposure is now greater due to the co-occurrence of habitat with
anthropogenic threats. Earthquake-induced change is not the source of heightened vulnerability
per se. Rather, this is an effect of natural dynamics that have increased exposure to pre-existing
stressors. These activities are now threats that require management to achieve conservation
objectives. Mowing of vegetation within riparian reserves co-occurs with several spawning sites
in both river systems. It is a particular issue where the spring high tide water levels are
sufficient to inundate riparian terraces. These provide locations where spawning habitat may be
relatively expansive in comparison to areas with steeper topography. Vegetation clearance using
scrub bars also occurs on the bank face throughout much of the study area for flood
management purposes with the exception of locations specifically managed for īnanga spawning
(Figure 4). Compared to reserve maintenance activities, vegetation clearance for flood
management affects the upper intertidal zone of the waterway margin. At many locations this
results in a direct overlap with the spawning habitat elevation band. High egg mortality from
mowing and grazing has been previously reported (Hickford & Schiel, 2014). This is believed
to be mostly attributable to UV irradiation or the drying out of eggs (Hickford et al. 2010;
Hickford & Schiel 2011b). Recovery from vegetation clearance can take many months, with the
re-establishment of sufficient cover being a critical factor (Hickford & Schiel 2014). In addition,
these activities may occur after eggs have been laid in vegetation that would otherwise have
been suitable for spawning. This was observed at many of the spawning sites recorded in this
study and is particularly problematic for conservation. Due to the gregarious behavioural
ecology of G. maculatus (Benzie 1968; McDowall 1990), the majority of spawning production
is typically supported by only a few sites in the catchment in each spawning event. This
contributes to the vulnerability of spawning to stochastic events. Anthropogenic threats
affecting these highly productive sites may have a large impact on the total egg production on a
seasonal basis.
4.3 Learning for adaptive management
This case illustrates important principles for managing subtle yet widespread change. The
results demonstrate habitat migration that was not detected by conservation management
practitioners. Pre-disturbance land-use activities had continued without adaptation exposing the
habitat to increased risk despite its apparent expansion. Adaptive management responses are
needed to control anthropogenic stressors in areas that have now become īnanga spawning
habitat. Achieving this requires further work to develop solutions that accommodate other
necessary or desirable waterway management activities in the riparian zone. Although historical
AOO figures are not available, the post-quake studies indicate that in both catchments the
extent of spawning habitat is now greater than all previous records. This is a positive finding
and suggests a potential improvement in the opportunities available for accommodating
incompatible activities through tools such as spatial planning. If these are addressed and
solutions identified, conservation gains could be secured in terms of increasing the area of
protecting habitat and ultimately improved egg production.
This case also provides several important lessons for the wider community in relation to
conservation management following major disturbance events. These include the need to fully
characterise environmental change and consequences for protected species despite that
information acquisition may be difficult. Challenges to overcome include the likelihood that
post-disturbance landscapes will be in a state of transition until a new relatively stable state
becomes re-established. In our case, this was partly addressed by commencing investigative
work four years from the disturbance event but also demanded temporal replication in the post-
quake studies to confirm whether the apparent effects could be related to an enduring post-
disturbance change. These aspects of the study approach may be useful considerations for the
design of other post-disturbance studies, and the importance of baseline measurements is also
highlighted since these are essential for interpreting change. In our case, further work is also
required to establish the cause-effect relationship driving the observed change. Salinity effects
are thought to be the most likely driver of the large-scale catchment position changes in
consideration of the literature suggesting a close relationship between spawning habitat and
saltwater intrusion. As such the disturbance event offers a unique opportunity to test
fundamental knowledge for conservation biology. Taking such opportunities requires a
commitment to mobilize the necessary resources at the required time. However the potential
gains from attention to these ‘natural laboratories’ are considerable especially given that the
circumstances may represent relatively uncommon events that cannot be readily replicated or
otherwise observed.
The implementation of statutory protection mechanisms for the achievement of conservation
objectives adds another dimension to this case. It is important to note that protection of the
post-quake habitat is a legislative requirement. However, conservation policy often suffers from
implementation gaps in practice (Knight et al. 2008) which may result from a lack of attention
to methods that are effective in the societal context (Knight et al. 2010). Dynamic environments
and spatio-temporal variation create additional challenges for the design of conservation
methods that are effective and socially acceptable. Our results illustrate that investments in
information are a pivotal activity that contributes to all of these needs. In addition,
contemporary information must be coupled with appropriate responses to facilitate an adaptive
approach. Our case highlights that further policy-related work may be a necessary aspect of
addressing the societal dimension. In particular a review of current conservation planning
arrangements with a focus on role of protected areas is a practical necessity given that these are
important conservation management tools.
Lastly, the effects described here are an example of landscape-scale responses to infrequent
tectonic dynamics. They have likely been mediated by hydrological and salinity changes
together with smaller-scale effects on ground surfaces in the riparian zone. In the Heathcote in
particular, the magnitude of horizontal shift deserves further investigation and despite the
current unknowns regarding causative factors the opportunity for learning is clear. Post-
earthquake studies present opportunities to evaluate many aspects of socio-ecological systems
for impacts and associated responses. Not only are tectonic events relatively common in
evolutionary time, they may exert similar effects to climate change through influencing water
levels and salinity gradients relative to existing topography (Beavan & Litchfield 2012).
Earthquakes present unique and important opportunities to study vulnerable ecosystems and
provide examples of real-life adaptation in action. In turn, this may assist in developing
methods to achieve conservation objectives and avoid implementation failures in the face of
ongoing change.
Acknowledgements
We thank Mark Taylor, Shelley McMurtrie and Colin Meurk for providing historical records.
We acknowledge the many volunteers and staff of the Waterways Centre for Freshwater
Research and Marine Ecology Research Group who assisted with the post-quake field studies,
and local government staff for information on riparian management activities. Funding was
provided by the Ngāi Tahu Research Centre, Institute of Professional Engineers of New
Zealand Rivers Group, Brian Mason Scientific and Technical Trust, and a New Zealand
Ministry of Business, Innovation and Employment grant (C01X1002) in conjunction with the
National Institute of Water and Atmospheric Research.
References
Beavan, J., & Litchfield, N. J. (2012). Vertical land movement around the New Zealand coastline:
implications for sea-level rise. GNS Science Report 2012/29. Lower Hutt.
Benzie, V. (1968). Some ecological aspects of the spawning behaviour and early development
of the common whitebait Galaxias maculatus attenuatus (Jenyns). Proceedings of the
New Zealand Ecological Society, 15, 31-39.
Burnet, A. M. R. (1965). Observations on the spawning migrations of Galaxias attenuatus. New
Zealand Journal of Science, 8, 79-87.
Dunn, N. R., M., A. R., P., C. G., Crow, S. K., David, B. O., Goodman, J. M., . . . Rolfe, J. R. (2018).
Conservation status of New Zealand freshwater fishes, 2017. Zealand Threat
Classification Series 24. Department of Conservation, Wellington. 11 pp.
Eldon, G. A., Kelly, G. R., Bonnett, M. L., & Taylor, M. J. (1989). Fisheries survey of the
Heathcote River, January 1989. New Zealand Freshwater Fisheries Report, 111, 50.
Hickford, M. J. H., Cagnon, M., & Schiel, D. R. (2010). Predation, vegetation and habitat-specific
survival of terrestrial eggs of a diadromous fish, Galaxias maculatus (Jenyns, 1842).
Journal of Experimental Marine Biology and Ecology, 385(1-2), 66-72.
doi:10.1016/j.jembe.2010.01.010
Hickford, M. J. H., & Schiel, D. R. (2011a). Population sinks resulting from degraded habitats of
an obligate life-history pathway. Oecologia, 166(1), 131-140. doi:10.1007/s00442-010-
1834-7
Hickford, M. J. H., & Schiel, D. R. (2011b). Synergistic interactions within disturbed habitats
between temperature, relative humidity and UVB radiation on egg survival in a
diadromous fish. PLoS ONE, 6(9), e24318. doi:doi:10.1371/journal.pone.0024318
Hickford, M. J. H., & Schiel, D. R. (2014). Experimental rehabilitation of degraded spawning
habitat of a diadromous fish, Galaxias maculatus (Jenyns, 1842) in rural and urban
streams. Restoration Ecology, 22(3), 319-326.
Hughes, M. W., Quigley, M. C., van Ballegooy, S., Deam, B. L., Bradley, B. A., Hart, D. E., &
Measures, R. (2015). The sinking city: earthquakes increase flood hazard in
Christchurch, New Zealand. GSA Today, 25(3-4), 4-10.
Jolly, D., & Ngā Papatipu Rūnanga Working Group. (2013). Mahaanui Iwi Management Plan
2013. Mahaanui Kurataiao Ltd. Ōtautahi Christchurch.
Kirk, R. M. (1979). Dynamics and management of sand beaches in southern Pegasus Bay.
Morris and Wilson Consulting Engineers Limited, Christchurch.
Knight, A. T., Cowling, R. M., Difford, M., & Campbell, B. M. (2010). Mapping human and social
dimensions of conservation opportunity for the scheduling of conservation action on
private land. Conservation Biology, 24(5), 1348-1358. doi:10.1111/j.1523-
1739.2010.01494.x
Knight, A. T., Cowling, R. M., Rouget, M., Balmford, A., Lombard, A. T., & Campbell, B. M.
(2008). Knowing but not doing: selecting priority conservation areas and the research-
implementation gap. Conservation biology : the journal of the Society for Conservation
Biology, 22(3), 610-617. doi:10.1111/j.1523-1739.2008.00914.x
Land Information New Zealand. (2016). Christchurch 0.075m Urban Aerial Photos (2015-16).
Accessed October 3, 2016 from 6993 GeoTIFF sources in NZGD2000 / New Zealand
Transverse Mercator 2000.
Lang, M., Orchard, S., Falwasser, T., Rupene, M., Williams, C., Tirikatene-Nash, N., & Couch, R.
(2012). State of the Takiwā 2012 -Te Ähuatanga o Te Ihutai. Cultural Health
Assessment of the Avon-Heathcote Estuary and its Catchment. Christchurch: Mahaanui
Kurataiao Ltd. 41pp.
McDowall, R. M. (1984). The New Zealand whitebait book. Wellington: Reed.
McDowall, R. M. (1990). New Zealand freshwater fishes: A natural history and guide. Auckland:
Heinemann Reed.
McDowall, R. M. (1992). Particular problems for the conservation of diadromous fish. Aquatic
Conservation: Marine and Freshwater Ecosystems, 2(4), 351-355.
doi:10.1002/aqc.3270020405
McDowall, R. M., & Charteris, S. C. (2006). The possible adaptive advantages of terrestrial egg
deposition in some fluvial diadromous galaxiid fishes (Teleostei : Galaxiidae). Fish and
Fisheries, 7(3), 153-164. doi:10.1111/j.1467-2979.2006.00217.x
Measures, R., Hicks, M. D., Shankar, U., Bind, J., Arnold, J., & Zeldis, J. (2011). Mapping
earthquake induced topographical change and liquefaction in the Avon-Heathcote
Estuary. Environment Canterbury Report No. U11/13. Christchurch: Environment
Canterbury. 28pp.
Meurk, C. D. (1989). Vegetation of the inanga spawning grounds on the banks of the lower
Avon River. Botany Division DSIR. Christchurch.
Mitchell, C. P. (1994). Whitebait spawning ground management. New Zealand Department of
Conservation, Science and Research Series, 69, 1-23.
Mitchell, C. P., & Eldon, G. A. (1991). How to locate and protect whitebait spawning grounds.
Rotorua: Freshwater Fisheries Centre., 49pp.
Orchard, S. (2014). Potential roles for coastal protected areas in disaster risk reduction and
climate change adaptation: a case study of dune management in Christchurch, New
Zealand. In R. Murti & C. Buyck (Eds.), Safe Havens: Protected Areas for Disaster Risk
Reduction and Climate Change Adaptation. Gland, Switzerland: Internation Union for
the Conservation of Nature. pp 83-93.
Orchard, S., & Hickford, M. J. H. (2018). Census survey approach to quantifying īnanga
spawning habitat for conservation and management. New Zealand Journal of Marine
and Freshwater Research, 52(2), 284-294. doi:10.1080/00288330.2017.1392990.
QGIS Development Team. (2016). QGIS Geographic Information System. Open Source
Geospatial Foundation Project. http://qgis.osgeo.org.
Quigley, M. C., Hughes, M. W., Bradley, B. A., van Ballegooy, S., Reid, C., Morgenroth, J., . . .
Pettinga, J. R. (2016). The 2010–2011 Canterbury Earthquake Sequence:
Environmental effects, seismic triggering thresholds and geologic legacy.
Tectonophysics, 672-673, 228-274. doi:10.1016/j.tecto.2016.01.044
Regenerate Christchurch. (2017). Outline for the Ōtākaro / Avon River Corridor Regeneration
Plan. Christchurch: Regenerate Christchurch.
Richardson, J., & Taylor, M. J. (2002). A guide to restoring inanga habitat. NIWA Science and
Technology Series, 50, 1-29.
Taylor, M. J. (1994). Inanga spawning ground enhancement, Heathcote River. National Institute
of Water & Atmospheric Research Ltd. Christchurch.
Taylor, M. J. (1995). Inanga spawning on the Heathcote River. National Institute of Water &
Atmospheric Research Ltd. Christchurch.
Taylor, M. J. (1996). Inanga spawning in the Avon and Heathcote Rivers. National Institute of
Water and Atmospheric Research Ltd. Christchurch.
Taylor, M. J. (1997). Inanga spawning activity on the Avon and Heathcote Rivers. NIWA client
report CHC97/57. National Institute of Water and Atmospheric Research Ltd.
Christchurch.
Taylor, M. J. (1998). Inanga spawning in the Avon and Heathcote Rivers, April 1998. NIWA
client report CHC98/42. National Institute of Water and Atmospheric Research Ltd.
Christchurch.
Taylor, M. J. (1999). Inanga spawning on the Avon and Heathcote Rivers, 1999. National
Institute of Water & Atmospheric Research Ltd. Christchurch.
Taylor, M. J. (2000). Inanga spawning in the city region during the year 2000. National Institute
of Water & Atmospheric Research Ltd. Christchurch.
Taylor, M. J. (2002). The national inanga spawning database: trends and implications for
spawning site management. Science for Conservation 188. Wellington: Department of
Conservation. 37pp.
Taylor, M. J. (2004). Inanga spawning grounds on the Avon and Heathcote rivers. Aquatic
Ecology Ltd. Christchurch.
Taylor, M. J., & Blair, W. (2011). Effects of seismic activity on inaka spawning grounds on city
rivers. Aquatic Ecology Limited. Christchurch.
Taylor, M. J., Buckland, A. R., & Kelly, G. R. (1992). South Island inanga spawning surveys, 1988-
1990. New Zealand Freshwater Fisheries Report No. 133. Ministry of Agriculture and
Fisheries. Christchurch.
Taylor, M. J., & Chapman, E. (2007). Monitoring of fish values; inanga and trout spawning. AEL
Report No. 54. Christchurch: Aquatic Ecology Limited., 22pp.
White, P. A., Goodrich, K., Cave, S., & Minni, G. (2007). Waterways, swamps and vegetation of
Christchurch in 1856 and baseflow discharge in Christchurch city streams. GNS Science
Consultancy Report 2007/103. Taupo.
Table 1. Extent of īnanga spawning habitat utilised in the Avon and Heathcote Rivers over the
period 1989 – 2014 from all known records.
Year
Description
Extent of
spawning† (m)
References
Avon
1989
TRB 40m reach downstream and upstream of Avondale Road
bridge (ARb)
80
Meurk (1989); Taylor et al.
(1992)
1993
TRB 15m reach above ARb
15
Taylor (1996)
1996
TRB 90m reach above and 25m reach below ARb
115
Taylor (1996)
1997
TRB 90m reach above ARb
90
Taylor (1997)
1998
TRB 70m reach above and 20m reach below ARb
90
Taylor (1998)
1999
TRB 250m reach above ARb
250
Taylor (1999)
2000
TRB at ARb
90
Taylor (2000)
2004
TRB from Alloway Street to Orrick Crescent; TLB at Amelia
Rogers Reserve, above and below ARb, and at Corsers Stream
1500
Taylor (2004)
2006
TLB Amelia Rogers Reserve
TLB Lake Kate Sheppard
1070
University of Canterbury
unpubl. data
2007
TRB from ARb to Sharlick Street and in Lake Kate Sheppard
2000
Taylor & Chapman (2007)
2008
TRB above ARb
250
Hickford & Schiel (2014)
2010
TRB above ARb
unknown
Taylor & Main unpubl. data
2011
TRB above ARb
90
Taylor & Blair (2011)
Heathcote
1989
TLB 70m reach downstream and 20m reach upstream of Wilsons
Road bridge, TRB 20m reach downstream of Wilsons Road
bridge
90
Eldon et al. (1989)
1991
TRB 100m reach within King George V Reserve
100
Taylor et al. (1992)
1994
TRB 30m reach below Opawa Road bridge (ORb)
30
Taylor (1994)
1995
TRB 50m reach below ORb
50
Taylor (1995)
1998
TRB 50m reach below ORb
50
Taylor (1998)
1999
TRB from ORb to downstream of rail bridge
70
Taylor (1999)
2002
TRB small patch in King George V Reserve
10
University of Canterbury
unpubl. data
2004
TRB in King George V Reserve, TLB and TRB below ORb
1050
Taylor (2004)
2010
TLB 12m reach adjacent to Woolston Park
12
Taylor & Blair (2011)
† Calculated as the distance between upstream and downstream limits of spawning as measured on the centreline of the mainstem for each
river. Where spawning also occurred in tributaries the location of the confluence was used for this calculation. TRB = true right bank. TLB
= true left bank.
