The Otaio River flow/ ecology memo

The Otaio River: An overview of ecological values, and potential
impacts of a range of minimum flow and allocation regimes.
Duncan Gray
Graeme Clarke
Environment Canterbury are reviewing minimum flow and allocation limits for rivers and streams as
part of the South Canterbury Coastal Streams (SCCS) sub-regional plan process. Previous flow
reviews, for the Variation 9 plan change to the NRRP, recommended minimum flow and allocation
blocks suitable for streams around Waimate (Golder 2012). However, the Otaio River fell outside the
geographic coverage of the Variation 9 plan change. Accordingly, De Joux (2014a) recommended
minimum flows and allocations based on both ecological and groundwater recharge considerations
for the Otaio Catchment. This memo reviews the ecological values of the Otaio River and considers
the likely impacts of current and proposed abstraction on those values.
The Otaio River flows for approximately 47.5 km from its headwaters in the Hunter Hills to the sea.
Upstream of the Otaio Gorge the river runs through steep hill country with a scattering of native
bush remnants and tussock. Below the gorge the Otaio River flows 23 km to the Otaio mouth which
lies 16 km south of Timaru. The mouth is typically closed to the ocean by a gravel bar which retains
a hapua or lagoon.
Ecological values
The ecological values of the Otaio River are exhaustively reviewed by Benn (2011) and little
subsequent information is available. Ecologically, the Otaio River can be divided into two distinct
units. The steeper, permanently, flowing headwaters and the lower gradient, intermittent reaches
below the Otaio Gorge. A river with intermittent flow is one where parts of the bed may be dry for
long periods, but there is also flow for several months of the year (Larned et al. 2010). This is
distinct from an ephemeral river which only flows briefly after rain.
The upper reaches of the Otaio, at Otaio Gorge, have been sampled twice for fish, in 1961 and again
in 2013 (Dave Kelly, ECan). Non-migratory upland bully were the only species identified in 1961, but
in 2013 only brown trout were present. The gorge is known to contain a self-sustaining brown trout
population which takes refuge from low flows in deep pools in the Otaio gorge (Daly 2004). DoC
(1999) assumed that introduced brook char would be present in smaller headwater tributaries. It is
also likely that Canterbury Galaxias, koaro and eels may inhabit the permanently flowing upper
reaches, although further survey work would be required to confirm this.
The lower reaches flow across alluvial gravels, and lose flow to groundwater inland, but regain flow
closer to the coast. Flow permanence is a critical determinant of the ecological value of waterways
(Davey & Kelly 2007; Arscott et al.2010; Larned et al. 2010). Stream drying not only impacts upon
species resident in a river reach, but also prevents migratory species from moving between the
upper and lower reaches. Additionally, Greenwood & McIntosh (2010) found that river drying in the
Selwyn River affected terrestrial communities in the wider floodplain, beyond the wetted area of the
According to the New Zealand Freshwater Fish Database fish communities in the lower Otaio River,
including the hapua, have been surveyed in 1961, 2001, 06, 07 and more recently in 2013 by ECan.
Further information has come from surveys undertaken by DoC (1998, 1999a & b) and Daly (2004).
In addition to introduced brown trout and brook char, nine species of native fish were identified, five
of which are ranked by Allibone et al. (2010) as threatened (Table 1). Furthermore, the
koura/freshwater crayfish, Paranephrops zealandicus, has been found at several sites along the
lower Otaio River in 2006 and 2007 (Figure 1). The freshwater crayfish is listed as threatened, in
gradual decline, by Hitchmough (2007). Koura are tolerant of brief periods of drying and will burrow
into damp substrates until flow returns.
Figure 1. Paranephrops zealandicus the freshwater crayfish. Photograph Philip Lindsay.
Table 1. Fish species known to occur in the Otaio River and tributaries.
Common/Māori name
Species name
Longfin eel/tuna
Bluegill bully
Shortfin eel
Otaio river
Otaio river
Otaio River
Daly 2004
ECan 2013
Upland bully
Common/Māori name
Species name
Common bully
Canterbury galaxias
Brown trout
Salmo trutta
Brook char
Otaio River
and mouth
Upper and
lower Otaio
lower Otaio
Upper and
lower Otaio
the gorge
DoC 1998,
1999a, Daly
Kelly 2013
DoC 1998,
1999a, Daly
DoC 1998,
1999a, Daly
DoC 1998,
At times
Six of the fish species found in the Otaio River are diadromous meaning they must migrate out to sea
to complete their life cycle. Longfin and shortfin eels, koaro, bluegill bully (Figure 2), common bully
and torrentfish all require passage to the sea. The occurrence of these fish in the Otaio River
indicates that the mouth is open regularly enough to allow the inward migration of fish. However,
the extent and condition of freshwater habitat once fish enter the river is critical to the survival of
individuals to reach breeding age. Longfin eels and koaro are known to penetrate a considerable
distance inland and will colonise the headwaters of streams if downstream conditions permit
passage. Flows in the river are critical in providing passage to the headwaters and suitable habitat
for resident fish during time of non-connectedness.
Figure 2. Migratory bluegill bully have been found in the lower reaches of the Otaio River.
Photograph Prof. Angus McIntosh.
While the bulk of the migration between the sea and the river occurs during high spring and autumn
flows (table 2), migration of individuals upstream and downstream within the Otaio River catchment
could occur throughout much of the year.
Table 2. Migration times of diadromous fish found in the Otaio River catchment (Jellyman 2012)
Upstream migration
Downstream migration
Longfin eel
Jul - Oct
Apr - Jun
Shortfin eel
Aug - Oct
Feb - Apr
Apr - Nov
Feb - May
Aug - Nov
Mar - Jun
Bluegill bully
Sep - Dec
Sept - Feb
Common bully
Dec - Feb
Oct - Nov
Brown trout (not obligate diadromous)
Feb - Apr
Nov - Apr
Studies of invertebrates and fish in the Selwyn River, Canterbury, have shown that many species are
able to tolerate short periods of river drying (Davey & Kelly 2007; Arscott et al.2010). However, the
longer the stream bed is dry, the more species are permanently lost. Thus, the duration of drying, as
well as the spatial extent, is of importance to in-stream ecology. Although there is scant information
on the extent and duration of drying in the lower Otaio River, either natural or the effect that
abstraction has on drying, it is likely that ecological values are considerably impacted.
A Block minimum flow and allocation
During periods of low flow, a range of ecological and physico-chemical processes can have a
detrimental effect on the ecological values in a stream. These include loss of usable habitat for
aquatic organisms, increased likelihood of warm water temperatures and associated reductions in
dissolved oxygen concentrations, and increased susceptibility to predation. Drying of rivers results in
a complete loss of aquatic habitat, loss of migration opportunities and loss of invertebrate and fish
recolonisation by downstream drift. Long periods of stream drying can result in a depauperate fish
and aquatic invertebrate assemblage when flow returns, compared to streams that are dry for
shorter periods (Datry et al 2007, Larned et al 2007, Arscott et al 2010). To manage these effects,
minimum flow and allocation limits are required to manage abstraction of water. The minimum flow
is designed to maintain a particular level of ecological protection or resilience to the effects noted
above, and defines a flow below which all abstraction should cease. Above the minimum flow an
allocation block can be defined so that river flows are not ‘flatlined’ or maintained at low levels for
long periods. A range of important ecological processes occur at mid-range flows, and streams are
unlikely to support a diverse aquatic community if sustained at low flows for extended periods.
Critically for the Otaio River, a low minimum flow and high allocation flow regime would increase the
duration and extent of the dry reaches seen in the river above those experienced naturally.
A number of methods can be used to derive a minimum flow. These are (1) historic flow regime, (2)
hydraulic and (3) habitat based methods. Historic flow methods rely on the historic flow of the
waterway, and generally recommend a proportion of a certain flow statistic (mean, median flow etc)
as a minimum flow. Proponents of these methods assert that they maintain a proportion of the flow
that the aquatic communities present are adapted to, and that they eliminate the uncertainties
associated with habitat and hydraulic based methods. Hydraulic flow methods rely on information
from stream surveys to gain information on width, depth and velocity at different flows. This
information can be used to identify a flow range where a parameter (e.g. wetted perimeter) declines
sharply, or perhaps allows a particular stream width to be maintained. Habitat methods allow the
use of information regarding a particular aquatic species’ hydraulic requirements to set minimum
flows that meet or exceed those requirements (Jowett 1997).
In the absence of sufficient data related to the Otaio River to use a hydraulic or habitat based
methodology for deriving a suitable minimum flow, the historic flow regime must be used.
Additionally, the use of habitat or hydraulic methods is problematic in naturally drying waterways.
The Ministry for the Environment (2008) has proposed a National Environmental Standard (NES) for
managing low flows, and suggests a figure of 90% of the naturalised seven day mean annual low flow
(7DMALF) be adopted as a minimum flow for smaller waterways like the Otaio River. The NES figure
of 90% MALF is a default minimum flow for situations where no minimum flow has been previously
set by more river-specific analysis. The proposed minimum flow of 90 L/sec being discussed by the
Otaio group is approximately 83% of the 109 L/sec natural 7DMALF at the gorge. This would indicate
that a slightly smaller amount of habitat would be available for in-stream communities under this
minimum flow than would be available during a naturally occurring dry period. The low flow period,
below 7DMALF, may be considered as a pinch point in the flow regime which defines the ecological
community for the rest of the flow year.
The NES also proposes a default total allocation based on the historic flow regime for catchments
that do not have a defined allocation block, and suggests a default allocation of the greater of 30%
of the MALF or the current total allocation as an interim position. The implication being that this is
the recommendation until such time as river-specific analysis can identify a more appropriate
allocation limit, which may or may not require clawing back from the current allocation. For the
Otaio River 30% of the 7DMALF would equate to approximately 33 L/sec.
A flow sharing regime with a series of steps has been proposed for the Otaio River as outlined in
table 3. It can be seen from table 3 that the allocated volume or rate of water abstraction is
considerably greater than the NES (MfE 2008) default of 30 % of natural 7DMALF (33 L/sec). This A
block allocation is highly likely to result in an increased duration and extent of drying in the lower
river and degradation of water quality in any areas of the stream which remain wet. Furthermore,
the configuration of the proposed A block partial restriction, is likely to result in a large downward
“step change” in mid reach flows once the minimum flow is exceeded at the gorge. This in contrast
to a classical partial restriction that does not permit abstraction until the minimum flow plus the
partial allocation is present in the river and thereby protects the entire minimum flow volume. The
result of the proposed configuration is less water in the river than would occur under natural
7DMALF conditions in the absence of abstraction. The only potential justification for having a less
protective regime than the classical partial restriction above 7DMALF configuration, would involve
making value judgements concerning the economic and social effects of imposing a more protective
regime and/or a smaller allocation block. Such a consideration is beyond the scope of this report.
While it is acknowledged that the river dries naturally in the mid reaches, the provision of a
minimum flow and suitable A block allocation mechanism would ensure some degree of protection
of a proportion of a natural flow regime in the lower reaches of the Otaio River by way of
maintaining groundwater levels, or reducing the rate at which they are depleted.
Table 3. Proposed flow and allocation regime for the Otaio River (based on flows at Otaio Gorge
Flow at Gorge Max abstraction rate (L/sec) Max 7 day volume
Equivalent average 7day rate
91 - 150
178 L/sec
150 L/sec
151 - 200
236 L/sec
200 L/sec
406 L/sec
372 L/sec
The lower reaches and hapua of the Otaio River are assumed to have gone entirely dry on occasion
in the past, but abstraction of water from the catchment has likely resulted in an increase in the
extent and duration of the dry reaches.
The hapua and lower reaches of Canterbury plains hill-fed rivers provide habitat to a diverse
community of fish, invertebrates and birds. Species such as inanga will spend much of their lifecycle
in this area, whereas other species will travel upstream and move inland. In the case of the Otaio
River where the middle reaches are often dry the hapua may provide an important low flow refuge
for fish which can travel upstream when the flow returns. This concept is reinforced by the large
numbers of small diadromous torrent fish found in the lower Otaio River in 2013. The presence of
good quality water in the lower reaches and hapua are critical to the survival of fish and invertebrate
individuals in this area. Low flows will reduce the habitat available in these reaches. Large numbers
of fish in small areas are prone to increased rates of predation and competition between individuals.
Low flows are also prone to rapid solar heating and result in stress to aquatic organisms due to low
dissolved oxygen concentrations. Torrent fish, as their name suggests, prefer swiftly flowing riffle
type habitat. Provision of sufficient flow to maintain this habitat type in the lower reaches of the
Otaio River is important to ensure individuals survive until higher flows return.
Flow and water quality in the lower reaches and hapua are predominantly derived from up-welling
groundwater, except during floods. This groundwater is derived from land surface recharge in the
catchment, but also water lost from the Otaio River as it flows across the inland plains. Thus,
although the middle reaches of the river may be dry, abstraction of shallow groundwater in the
catchment will continue to effect flow and water quality in the lower reaches and hapua. The
precise relationship between flow at the Otaio Gorge, groundwater levels and flow in the lower
Otaio River are poorly understood. However, we know that abstraction of surface water and shallow
groundwater in the catchment will impact upon flows in the lower Otaio River and hapua. As flows
decline there will be a commensurate reduction in the area and suitability of habitat and
degradation of water quality in the lower river.
In summary, from an ecological perspective the A block allocation in the Otaio River is of such a
magnitude that it almost certainly impacts upon the flows, water quality, duration, regularity and
extent of drying in the lower Otaio River and hapua. In addition, the increase in duration, regularity
and extent of drying in the lower river affects the ability of migratory fish species to reach the
permanently flowing headwaters of the river. There is potential for considerable improvement in
the current ecological values of the Otaio River if a minimum flow could be set (in the vicinity of 90%
7DMALF) and if water allocated to the A block could be shifted away from base flow abstraction to
alternative sources such as a B block, which harvests water during higher flows, irrigation schemes
or deep groundwater. If a lower minimum flow (than 90% MALF) was set in order to lessen the
economic impacts on existing users (e.g. 75L/s mentioned by Otaio users group) then the level of
ecological protection afforded will be less (and will decrease the lower the minimum flow is set) but
this would still be better than the current situation with no minimum flow at all. Other information
on the economic and social effects would be needed to make the necessary value judgements to
decide on a lower minimum flow and this is beyond the scope of this report.
B Block minimum flow and allocation
In water short catchments irrigation reliability may be improved, without exacerbating the
magnitude and duration of low flows, by abstracting water from streams during higher flows and
storing that water until required. These flood/flush harvesting water takes, sometimes referred to
as a ‘B block’, require a minimum flow and allocation limit. If the B block minimum flow is too low
and abuts the top of the A block allocation the effect will be to increase the duration of low flows,
because this situation essentially makes the A block greater. Therefore, it is often recommended to
include a ‘gap’ within which medium size flow events are un-abstracted to maintain flow variability,
flush the river and recharge groundwater. Above this ‘gap’ a suitable B block minimum flow and
allocation can be applied with limited theoretical impacts on the ecology of the river.
Golders (2012) suggested that B block abstraction for storage should be allocated using statistics and
flow rules calculated separately between winter and summer. This is because the flow regimes of
small hill-fed coastal streams vary quite distinctly between summer and winter (winter being defined
as April to September). Winter and summer flow stats can be markedly different for low flows in
particular, but the rationale holds for other ecologically relevant flow statistics such as the median or
FRE3 (frequency of flow events in a year exceeding 3X median).
Golder (2012) recommended that a suitable B block minimum flow for SCCS in winter and summer
would be at least the median flow, but preferably more, calculated for that specific season.
Furthermore, an allocation of <30 % of the seasonal median would suitably protect flow variability
and flushing flows. These reccomendations were necessarily generic due to an absence of river
specific analysis and applied a conservative approach for the purposes of ecological protection.
Analogous to an A block, the B block minimum flow and allocation interact. In general as the B block
inimum flow increases, so too can the size of the B block allocation rate while still providing
protection for ecological values. There are ecological limits to the size of the B block allocation,
irrespective of the minimum flow, which are required to ensure processes such as periphyton
flushing, channel morphology maintenance and river mouth opening.
The location of the flow gauging site used to derive the minimum flow and allocation should also be
considered. In rivers with considerable flow losses and gains along their length, like those in the SCCS
area, there will be significant differences in estimates of mean or median flow (and therefore FRE3)
depending on the position of the flow monitoring site within the catchment.
The Otaio River
Using current flow statistics (Adam Martin pers. comm. November 10, 2014) and the generic rules
suggested by Golder (2012) a suitable Winter B block minimum flow for the Otaio at the Gorge
would be;
Winter Median ≥ 320 l/s
And a suitable maximum allocation would be
30 % winter median ≤ 96 l/s
De Joux (2014a) undertook some specific analysis on the Otaio River to assess the effects of a
different B block minimum flow and allocation regime on flow variability in the river. De Joux
(2014a) used ~90 % of the FRE 3 flow statistic as a minimum flow, but note that statistic was based
on the annual median.
De Joux (2014a) refers to Aitchison-Earle (2014) and cites the relationship between groundwater
levels, flow permanence in the middle reaches of the Otaio River and flow at the gorge. According
to Aitchison-Earle (2014) at flows <300 l/s at the Otaio Gorge the middle reaches of the river dry and
groundwater levels in the catchment decline. Presumably, the proposed B block minimum flow
suggested by De Joux (2014) was designed to reduce river drying and allow for groundwater
De Joux (2014a) recommended;
B block minimum of 780 l/s
B block allocation of 500 l/s
De Joux (2014a) provides modelled estimates of the effects of the proposed B block minimum flow
and allocation regime on the occurrence of flushing flows (defined as 3 x the median or FRE3). The
modelling suggests very little impact on flushing flow occurrence which is typical of flashy hill-fed
streams where high flow events often have a considerable magnitude relative to low and median
The De Joux (2014a) B block minimum flow recommendation aligns with that of Golder (2012) in
that is it greater than median flow and would protect flow variability in the Otaio River. The B block
allocation of De Joux (2014a) is greater than would be recommended by Golder (2012) being 156 %
of the winter median flow. However, as the proposed minimum flow is more than twice the winter
median there are unlikely to be negative impacts on in-stream ecology.
If it was proposed to further enlarge the b block allocation we recommend that a further analysis be
undertaken to examine the effect of such a large allocation on channel forming flows. These flood
events are critical for clearing macrophytes and terrestrial vegetation from the river bed as well as
the transport of gravel along the channel. Flows six times the median are required to uproot
macrophytes (Biggs 1996), whereas flow statistics such as the mean annual flood should be used to
assess potential change to river morphology and floodplain vegetation.
The Otaio River has some significant ecological values despite the exacerbation of low flows and
river drying associated with abstraction.
The A block minimum flow and allocation for the Otaio is currently 0 L/sec and 406 L/sec
respectively (De Joux 2014b). This is a very large allocation block likely to cause “flat lining” and/or a
longer period and extent of river bed drying.
A minimum flow of 90 % of 7DMALF at the Otaio Gorge would provide better protection for
ecological values than the current situation. A lower minimum flow or the adoption of a partial
restriction configuration other than the classical structure would provide a lower level of ecological
protection, even if coupled with a reduced allocation. This situation could only be justified by the
economic and social implications of the higher minimum flow and is outside the scope of this report.
While it is clearly acknowledged that the river dries naturally in the mid reaches, the provision of a
minimum flow and reasonable A block allocation would ensure some degree of protection of a
proportion of the natural flow regime in the lower reaches of the Otaio River, by way of attempting
to maintain groundwater levels, or reducing the rate at which they are depleted.
The proposed B block for the Otaio River appears unlikely to cause any significant impact on
ecological values, especially when considered alongside the large A block allocation.
Any transfer of A block abstraction to alternative sources such as B block, irrigation schemes or deep
groundwater will ease the impact of low flows on the lower reaches of the Otaio River and allow a
recovery of the remaining ecological values. Alternatively, an A block allocation mechanism that
allows the minimum flow to achieve its intended purpose would be warranted.
Allibone, R., David, B., Hitchmough, R., Jellyman, D., Ling, N., Ravenscroft, P., Waters, J., 2010.
Conservation status of New Zealand freshwater fish, 2009. New Zealand journal of marine and
freshwater research 44, 129–148.
Arscott, D., Larned, S., Scarsbrook, M., Lambert, P., 2010. Aquatic invertebrate community structure
along an intermittence gradient: Selwyn River, New Zealand. Journal of the North American
Benthological Society 29, 530–545.
Benn, J., 2011. In stream intrinsic values of the Otaio River catchment (Draft Technical Support
Summary No. docDM-739962). Department of Conservation.
Biggs, B.J.F., 1996. Hydraulic habitat of plants in streams. Regulated Rivers-Research & Management
12, 131–144.
Daly, A., 2004. Inventory of instream values for rivers and lakes of Canterbury, New Zealand. (No.
Report U04/13), Environment Canterbury. Christchurch, New Zealand.
Davey, A.J.H., Kelly, D.J., 2007. Fish community responses to drying disturbances in an intermittent
stream: a landscape perspective. Freshw. Biol. 52, 1719–1733.
De Joux R 2014a. Otaio River water users – impacts of “B” allocation on river flows. Nov 4th.
De Joux R 2014b. Otaio River catchment:Draft proposed flow regime. 6 Nov.
DoC 1998. Conservation Resources of the Kainawaru Pastoral Lease, Canterbury Conservancy,
pastoral lease tenure review report to Knight Frank Limited. 7p.
DoC 1999a. Conservation Resource report to Frank Knight Limited on Tenure Review of Mt
Studholme Pastoral Lease. Department of Conservation, Canterbury conservancy, Christchurch 7p
DoC 1999b. Conservation Resources Report on the Mt Cecil Pastoral lease. 11p.
Datry T., Larned S.T. and Scarsbrook M.R., 2007. Responses of hyporheic invertebrate assemblages
to large-scale variation in flow permanence and surface-subsurface exchange. Freshwater Biol., 52,
Golders 2012. Ecological flow review: Waihao River and Wainono lagoon catchment. Report
prepared for Environment Canterbury.
Greenwood, M.J., McIntosh, A.R., 2010. Low river flow alters the biomass and population structure
of a riparian predatory invertebrate. Freshwater Biology doi:10.1111/j.1365-2427.2010.02462.x.
Hitchmough, R., Bull, L., Cromarty, P., 2007. New Zealand threat classification system lists, 2005.
Wellington, Department of Conservation.
Jellyman D. 2012. Fish recruitment into Te Waihora/Lake Ellesmere: a consideration of the
requirements for key species. NIWA client report CHC2011-094 75p.
Jowett, I. G. 1997: Instream flow methods: a comparison of approaches. Regulated rivers 13: 115127.
Larned S.T., Datry T. & Robinson C.T. (2007) Invertebrate and microbial responses to inundation in
an ephemeral river reach in New Zealand: effects of preceding dry periods. Aquatic Sciences:
Research Across Boundaries, 69, 554–567.
Larned, S.T., Datry, T., Arscott, D.B., Tockner, K., 2010. Emerging concepts in temporary-river
ecology. Freshw. Biol. 55, 717–738.
MfE (2008). Proposed National Environmental Standard on Ecological Flows and water levels.
Discussion Document.
South Canterbury/Waitaki Eel Management Committee (SCWEMC). 1999. South Canterbury/Waitaki
Eel Management Plan 109 p.