34
Hydrology of the Mataura River: A constraint on water abstraction at Gore
Hydrology of the Mataura River:
A constraint on water abstraction at Gore
Hydrology of the Mataura River:
A constraint on water abstraction at Gore
Prepared by Opus International Consultants Ltd
Sheryl Paine Wellington Environmental Office
Water Resources Scientist Level 5, Majestic Centre, 100 Willis Street
PO Box 12 003, Thorndon, Wellington 6144
New Zealand
Reviewed by Telephone: +64 4 471 7000
Dr Jack McConchie Facsimile: +64 4 499 3699
Principal water Resources Scientist
Date: July 2012
Reference: 6CM088.00
Status: Draft for review
© Opus International Consultants Ltd 2012
Mataura River at Gore
6CM088.00
July 2012 i
Contents
1 Background .......................................................................................................................... 1
2 Mataura catchment .............................................................................................................. 2
3 Flow regime .......................................................................................................................... 3
4 Cyclic behaviour and trends ............................................................................................... 5
4.1 El Niño – Southern Oscillation ...................................................................................... 6
4.2 Interdecadal Pacific Oscillation ..................................................................................... 7
4.3 Effect of climatic oscillations ......................................................................................... 9
4.4 Summary .................................................................................................................... 15
5 Mataura Water Conservation Order .................................................................................. 15
6 Effects of Climate Change ................................................................................................. 18
7 Groundwater ...................................................................................................................... 20
7.1 Background ................................................................................................................ 20
7.2 Connectivity between groundwater and surface water systems .................................. 22
8 Conclusions ....................................................................................................................... 24
9 References ......................................................................................................................... 25
Mataura River at Gore
6CM088.00
July 2012 ii
Mataura River at Gore
6CM088.00
July 2012 1
1 Background
The current potable water supply for Gore is obtained from shallow groundwater bores at the
Coopers Wells, Jacobstown, and Oldham Street sites. This arrangement reflects the
incremental growth of the water supply system over time and as a result the system has
several constraints (SKM, 2006). These constraints include:
• The configuration of the existing infrastructure, with separate intakes and treatment
facilities located at some distance from demand centres;
• Problems meeting supply requirements during periods of high demand, particularly
from the Jacobstown bores;
• Limited supply security as a result of the shallow unconfined source aquifer and
surrounding land use; and
• A requirement to upgrade existing treatment plants to meet New Zealand Drinking
Water Standards.
Gore District Council manages the reticulated water supplies for Gore and Mataura. Water
demand figures from July 2003 to September 2006 for Gore, and from May 2002 to April
2006 for Mataura, are shown in Table 1.1.
Table 1.1: Water demand in Gore and Mataura.
Gore Mataura Total Average daily demand (m³/day)
4050 1100 5150
Peak daily demand (m³/day)
6050 1600 7650
The existing Gore water supply therefore consists of three groundwater sources with
associated treatment facilities. The existing capacity of the supply (Table 1.2) is currently
constrained by the rate of groundwater abstraction.
Table 1.2: Existing storage capacity and abstraction rates.
Treated water storage (m³)
Maximum capacity (m³/day)
Coopers wells 4000 5600 Jacobstown 1000 2750 Oldham Street 135 1375
Given challenges faced by Gore District Council in the form of increased demand for water
arising out of population growth from lignite projects, the council has opted to put greater
priority on securing a supplementary source of water. Accordingly, a specific project with
dedicated resourcing has been established with the objective of locating and securing an
additional 5,000m³ of water per day. The additional capacity is intended to cater for the extra
Mataura River at Gore
6CM088.00
July 2012 2
consumption expected from an increased resident population, and commercial development
resulting from the proposed lignite projects. It is not intended that this extra capacity will be
utilised to service the lignite projects directly. Obviously the quality of raw water supply and
its distance from existing infrastructure will determine the strategic direction of water
treatment plants and future facilities for the District. This project forms a foundation upon
which the Councils strategy for water supply will be developed.
Therefore, at present the Gore and Mataura communities rely on the shallow unconfined
aquifer for their potable water supply. The behaviour, dynamics, and sustainable yield of this
aquifer are all affected by flows in the adjacent Mataura River. The flow regime of the
Mataura River therefore acts as a major constraint on the current, and potentially any future,
water supply. Quantifying the nature of this constraint is critical to exploring a potential new
water supply to meet the increased demand for potable water.
2 Mataura catchment
The Mataura catchment is the second largest in Southland (after the Waiau) both in terms of
area and flow. It covers an area of 5400km² stretching from its headwaters in the Eyre
Mountains to the Fortrose estuary, east of Invercargill. The Mataura River has one main
tributary, the Waikaia River, which joins the Mataura just east of Riversdale. This tributary
contributes half of the flow of the catchment above the confluence with the Mataura (Figure
2.1). Other large tributaries of the Mataura River include the Brightwater Spring, Eyre Creek
and Roberts Creek in the upper catchment, the Nokomai River, Waimea Stream and
Waikaka Stream in the mid catchment, and the Mokoreta River in the lower catchment.
The catchment has significant water supply values for various communities and industrial
uses, with a reasonably high level of allocation in its middle and lower reaches.
The Mataura and Waikaia Rivers are subject of a National Water Conservation Order which
was promulgated to protect the outstanding fisheries and angling amenity features of the
catchment. This Order restricts the granting of water permits to take water by: requiring that
flows not be reduced beyond a specified limit; prohibits damming of the main stem of the
Mataura and Waikaia Rivers and restricts damming of other tributaries; and places
restrictions on discharge permits to ensure that water quality is maintained.
There are four flow monitoring stations on the main stem of the Mataura River, and a number
of others on its various tributaries (Figure 2.1). This study, however, focuses predominantly
on the flow record from the Mataura at Gore because of its critical control on the potable
water supply. The Mataura at Gore flow monitoring station lies within the middle ‘climate
zone’ of the Mataura catchment. This zone covers the Waimea Plains between Gore and
Lumsden, and is located between the more coastal-dominated climate toward the south
coast and the more sub-alpine conditions in the upper catchment (Hughes et al., 2011).
Mataura River at Gore
6CM088.00
July 2012 3
Figure 2.1: The Mataura catchment and key localities.
3 Flow regime
Flow within the Mataura River is highly variable, mostly because of its alpine headwaters but
also because of the considerable size of the catchment. Over the last 35 years flows in the
Mataura River at Gore have varied from a low of 8m³/s to a maximum flood of 2297m³/s
(Table 3.1 and Figure 3.1). The flow regime is characterised by long periods of low flow
interspersed with high magnitude but low frequency floods. Consequently, the median flow
is significantly less than the mean as it is less affected by these short duration but high
magnitude flood events.
Table 3.1 Summary statistics in m³/s for the Mataura River at Gore (1977-2012).
Min Max Mean Std Dev LQ Median UQ
8.00 2297.00 64.77 65.18 30.95 48.90 77.99
Figure 3.1 shows the flow record for the Mataura River at Gore. While floods are generally
less than 1000m³/s there have been four large events greater than 1500m³/s; the largest with
a peak discharge of approximately 2300m³/s. The last major flood was in November 1999. It
Mataura River at Gore
6CM088.00
July 2012 4
would appear that last 15 years have been characterised by generally lower flows and
reduced flood activity. Floods tend to form relatively rapidly and flows can increase by
2000m³/s in 24 hours. Although large flows generally occur between September and April,
major flood events can occur in any time of the year.
Figure 3.1 Flow record for the Mataura River at Gore (1977-2012).
The flow regime of a river is often summarised by a flow-duration curve which shows the
proportion of time during which flow is equal to or greater than given magnitudes, regardless
of chronological order (Figure 3.2). The overall slope of the flow-duration curve indicates the
flow variability. Because of the large range of flows experienced within the Mataura River,
and therefore the relatively poor resolution at both ends of the distribution, the flow-duration
data can be summarised in tabular form (Table 3.2). The table can be read as; 0% of flows
were over 2297m³/s (the maximum flow recorded), 53% of flows are over 46.8m³/s and 100%
of flows are over 8m³/s (the minimum flow recorded).
The flow regime of the Mataura River at Gore appears to show both a seasonal (i.e. annual)
pattern of variability and also some longer term variability, particularly with respect to flood
activity. During the year winter is characterised by generally higher flows and a greater
number of moderate flood events. However, the largest floods generally occur during spring
and summer. This pattern of flooding is largely controlled by the nature and distribution of
precipitation. During winter much of the precipitation in the upper catchment generally falls
as snow and this delays runoff until the spring. Consequently, during spring floods can be
enhanced by rain-on-snow events which generate greater rates of runoff than would occur
from the rainfall alone. While the seasonal and annual variability can be accommodated
relatively easily within a water supply scheme, managing the longer term variability is more
problematic.
1978 1988 1998 2008
0
500
1000
1500
2000
2500
Flo
w (
m³/
s)
Mataura River at Gore from 18-May-1977 00:15:00 to 1-Jun-2012 00:00:00
Mataura River at Gore
6CM088.00
July 2012 5
Figure 3.2: Flow distribution for the Mataura River at Gore (1977-2012).
Table 3.2: Distribution of flows in the Mataura River at Gore (1977-2012).
0 1 2 3 4 5 6 7 8 9
0 2297.00 312.50 229.86 192.66 169.94 155.73 145.43 137.09 129.94 123.97
10 119.02 114.05 110.01 106.16 102.87 99.87 96.97 94.12 91.83 89.33
20 87.04 85.04 83.07 81.09 79.28 77.99 75.96 74.84 73.00 71.88
30 70.06 68.96 67.63 66.13 65.08 64.03 62.96 61.54 60.20 59.04
40 58.02 57.02 56.03 55.04 54.06 53.09 52.13 51.39 50.98 50.03
50 48.90 47.96 47.02 46.80 45.91 44.99 44.07 43.90 42.99 42.08
60 41.92 41.01 40.09 39.71 39.00 38.08 37.16 37.01 36.09 34.98
70 34.83 33.92 33.00 32.32 31.88 30.95 30.02 29.77 28.95 28.02
80 27.10 26.95 26.04 25.13 24.98 24.07 23.14 22.96 22.05 20.93
90 19.99 19.84 18.90 17.96 17.01 16.03 15.05 14.06 13.05 11.98
100 8.00
4 Cyclic behaviour and trends
As mentioned, the flow regime of the Mataura River indicates some longer term variability
which has the potential to impact on water resource availability. Consequently, the effects of
any trends in the climate which affect runoff, and therefore river flow, must be assessed.
Climatic trends and oscillations have the potential to bias any rainfall or flow record used in
analysis. This is not a major issue with long-term records that include a number of
oscillations e.g. both El Niño and La Niña phases for example. The effects of both phases
0
500
1000
1500
2000
2500
0 20 40 60 80 100
Flo
w (
m³/
s)
Percentage Greater
Mataura River at Gore
6CM088.00
July 2012 6
will be inherent in the record, and their effects will therefore be included in the results of any
statistical analysis. However, when a flow record coincides largely with one or other of these
phases, the data may be biased. The resulting record may reflect either increased or
decreased flows (and consequently water availability) depending on the exact period of the
record.
The presence of cyclic behaviour and periods of either increased or decreased flows has
considerable importance when assessing the reliability of water resources. For example,
sustained periods of reduced flows in the Mataura River would have significant implications
to the volume of water which could be abstracted from the shallow unconfined aquifer, and
the duration of periods when no abstraction may be possible. There are also implications for
the resilience of supply, and potentially technical constraints with regard to water supply
infrastructure. Assessing the nature and significance of climatic variability to flows in the
Mataura River, and consequently potential water supply, is therefore a critical consideration
when assessing the practicality and viability of a new water supply.
4.1 El Niño – Southern Oscillation
The El Niño – Southern Oscillation (ENSO) is a global climate phenomenon that is triggered
by changes in the ocean-atmosphere system in the tropical Pacific. These changes are
measured by the pressure difference between Tahiti and Darwin known as the Southern
Oscillation Index (SOI).
El Niño, the negative phase in the SOI, occurs when the westerly trade winds soften, and
warmer sea surface temperatures (SSTs) occur off the coast of South America. Although
New Zealand is not usually affected as strongly by El Niño conditions as parts of Australia,
there is often still a significant influence. Typically, during El Niño conditions New Zealand
experiences stronger and more frequent westerly winds in summer, and lower SSTs. During
summer months this can lead to higher rainfall in south-western parts of the South Island,
and drought conditions in the east. These conditions also bring more benign weather in the
north and east of the North Island. During winter the wind becomes dominant from the
south, leading to overall colder conditions. El Niño conditions generally bring colder
temperatures. These are more noticeable in the North Island in all but the summer months
(Kidson and Renwick, 2002). Although El Niño has an important influence on New Zealand’s
climate, it accounts for less than 25% of the year to year variance in seasonal rainfall and
temperature at most New Zealand measurement sites. East coast droughts may be common
during El Niños, but they can also happen in non El Niño years (for example, the severe
1988-89 drought). Also, serious east coast droughts do not occur in every El Niño, and the
districts where droughts occur can vary from one El Niño to another. However, the
probabilities of the climate variations discussed above happening in association with El Niño
are sufficient to warrant their consideration. Where the effects of these climatic variations
are considered significant, appropriate management actions and planning can be
implemented.
Mataura River at Gore
6CM088.00
July 2012 7
Alternatively La Niña, the positive SOI phase, occurs when strengthened trade winds and
colder SST in the eastern Pacific extend further west than usual. La Niña years tend to have
a weaker effect on the climate of New Zealand; with more north-easterly winds which bring
more moist, rainy conditions to the north-east parts of the North Island, and reduced rainfall
to the south and south-west of the South Island. Warmer temperatures are typically
experienced over the whole country. Higher rainfall is experienced in the north and eastern
part of the North Island during summer. The south and south-west of the South Island can
experience drier conditions. Above average rainfall occurs in the other seasons in all areas
except on the east coast of both islands which have normal or below average rainfall (Kidson
and Renwick, 2002).
Consequently, the ENSO has the potential to affect the climate of the Mataura catchment,
and as a result the runoff regime and reliability of water supply. The inter-annual ENSO
events vary in strength, can last from several months to several years, and tend to occur
three to seven years apart. Figure 4.1 shows the occurrence of the ENSO events from 1900-
2012. More recently, a La Niña event occurred during the summer of 2010 and 2011.
Figure 4.1: Variation in the SOI index which has been related to changes in the rainfall
regime. (Source: www.cgd.ucar.edu/cas/catalog/climind/soi.html)
4.2 Interdecadal Pacific Oscillation
The Interdecadal Pacific Oscillation (IPO) is a climatic fluctuation in atmospheric and SST in
the Pacific Basin that operates over a time scale of decades. Studies have shown that in
1901 1921 1941 1961 1981 2001
-6
-5
-4
-3
-2
-1
0
1
2
3
4
5
6
SO
I In
dex
SOI - 3 months Fixed Average
Mataura River at Gore
6CM088.00
July 2012 8
some areas of New Zealand there is a strong correlation between heavy rainfall and flooding,
and the IPO phases. This results in successive ‘benign’ and ‘active’ phases in flooding that
occur in conjunction with negative and positive phases of the IPO respectively. A positive
IPO phase persisted from 1922-1945, and again from 1977-1999; while from 1946-1976 the
IPO was in a negative phase. The IPO is currently in a negative phase, and so the incidence
of heavy rainfall is likely to be less than the long-term average (Figure 4.2). Shifts in the IPO
modulate the frequency of occurrence and intensity of El Niño and La Niña phases of the
ENSO. The positive phase is most commonly associated with higher frequency and intensity
of El Niño-like conditions, while the negative phase is associated with a prevalence of La
Niña patterns. For example, more El Niño episodes occurred from 1978 to 1999 than the
previous three decades which saw more La Niña episodes (McKerchar & Henderson, 2003)
(Figure 4.3). El Niño episodes tend to give more rain in the south and west of the country,
and drier conditions in the northeast. La Niña episodes tend to give less rain in the south
and east, and more rain in the north-east.
Figure 4.2: Variation in the IPO phase which has been related to changes in the rainfall
regime. (Source: www.iges.org/c20c/IPO_v2.doc)
1900 1920 1940 1960 1980 2000
-6
-4
-2
0
2
4
6
IPO
Index
IPO - 2 years moving average
Mataura River at Gore
6CM088.00
July 2012 9
Figure 4.3 Relationship of the SOI index with the IPO phase.
Recent climatological studies have demonstrated that the assumption of stationarity (i.e., that
all data are drawn from the same continuous population) may not be valid, at least for annual
rainfall in New Zealand. Compared with the period 1947-1977, consistent rainfall decreases
of up to 8% occurred for the period of 1978-1999 in the north and east of the North Island,
and increases of up to 8% occurred in the west and south of the South Island. These
changes are attributed to shifts in the phase of the IPO.
McKerchar and Henderson (2003) found that under a positive phase of the IPO (i.e., 1978-
1999) many rivers in the South Island showed increased median, and higher maximum,
annual flows. These trends were most noticeable in the south and south-west. It is therefore
possible that the IPO has affected flows in the Mataura River, and its impact is detectable in
the flow record.
4.3 Effect of climatic oscillations
Hughes et al. (2011) found that changes in the SOI exhibited a clear influence on the inter-
annual variation in rainfall in Southland. They used (inverse) SOI values compared with the
12-15 month rainfall departure from the mean at various sites within the Mataura catchment.
Above ‘normal’ rainfall occurred during negative ENSO phases, while below ‘normal’ rainfall
was experienced during positive ENSO phases (Figure 4.4). However, since rainfall is only
one of the factors which affect runoff within a catchment, variations in the SOI and IPO were
also compared to various flow statistics from the Mataura at Gore.
The flow record from the Mataura River at Gore was compared with variations in both the
SOI (Figure 4.5) and the IPO (Figure 4.6). Over the 35 years for which data are available
there is no strong evidence of either the SOI or IPO affecting the overall flow regime of the
1900 1920 1940 1960 1980 2000
-6
-4
-2
0
2
4
6IP
O I
nde
x
SOI - 3 months fixed averageIPO - 2 years moving average
Mataura River at Gore
6CM088.00
July 2012 10
Mataura River. Therefore, while various climatic indices may affect rainfall, and in particular
variability of rainfall, the same signature may not be apparent in the flow regime. This is
likely because runoff in a river is the net result of a wide range interacting factors including
both the climate and the physical characteristics of the catchment.
Figure 4.4: Relationship between rainfall departure at Mandeville and SOI values, (SOI
values are inverted) (Hughes et al., 2011).
Figure 4.5: Comparison of flows in the Mataura River at Gore with variation in the SOI
(1977-2012).
1978 1988 1998 2008
0
400
800
1200
1600
2000
2400
-6
-5
-4
-3
-2
-1
0
1
2
3
4
5
6
Flo
w (
m³/
s)
SO
I In
dex
Mataura River at Gore - Flow (m³/s)SOI - 3 months fixed average
Mataura River at Gore
6CM088.00
July 2012 11
Figure 4.6: Comparison of flows in the Mataura River at Gore with variation in the IPO
(1976-2011).
To further assess any potential link between flows in the Mataura River and climatic
oscillations, the annual average SOI and IPO indices were compared to the annual median
flow (Figure 4.7 and Figure 4.8 respectively); and the annual minima (Figure 4.9 and Figure
4.10 respectively).
There appears to be no strong relationship between variation in the SOI or IPO and changes
in the annual median or minimum flows. However, when the SOI index goes strongly
negative it would appear that there is a greater chance of higher median annual flows. This
pattern would appear to be stronger for the SOI than for the IPO index.
To highlight whether the SOI and IPO indices affect variation around mean conditions, rather
than the flows per se, both indices were compared to the deviation about the long-term
median flow (Figure 4.11 and Figure 4.12).
It would appear that both the SOI and IPO indices have a weak effect on the variation in flow
relative to average conditions. This apparent relationship, however, is more obvious over the
early part of the flow record. Periods when the SOI is strongly negative tend to be
associated with higher than average flow conditions and vice versa. These periods of higher
than average flows are also associated with periods when the IPO is strongly positive.
Periods with negative IPO indices are associated with lower than average flow conditions.
Since about 2000 the relationship between the various climatic indices and flow in the
Mataura River appears to have got weaker. The last 10 years of record is generally
characterised by lower than average flow conditions. It is perhaps not a coincidence that this
period also coincides with a dramatic increase in the abstraction of water for irrigation.
1978 1988 1998 2008
0
400
800
1200
1600
2000
2400
-6
-5
-4
-3
-2
-1
0
1
2
3
4
5
6F
low
(m
³/s)
IPO
Index
Mataura River at Gore - Flow (m³/s)IPO - 1 year moving average
Mataura River at Gore
6CM088.00
July 2012 12
Figure 4.7: Comparison of the median annual flows in the Mataura River with variation in
the SOI (1978-2011).
Figure 4.8: Comparison of the median annual flows in the Mataura River at Gore with
variation in the IPO (1978-2011).
-2.5
-2.0
-1.5
-1.0
-0.5
0.0
0.5
1.0
1.5
2.0
2.50
10
20
30
40
50
60
70
80
90
100
197
8
197
9
198
0
198
1
198
2
198
3
198
4
198
5
198
6
198
7
198
8
198
9
199
0
199
1
199
2
199
3
199
4
199
5
199
6
199
7
199
8
199
9
200
0
200
1
200
2
200
3
200
4
200
5
200
6
200
7
200
8
200
9
201
0
201
1
SO
I In
dex
Med
ian
An
nu
al F
low
(m
³/s)
Median Annual Flow Mean Annual SOI
-4.0
-3.0
-2.0
-1.0
0.0
1.0
2.0
3.0
4.0
5.0
6.0
0
10
20
30
40
50
60
70
80
90
100
197
8
197
9
198
0
198
1
198
2
198
3
198
4
198
5
198
6
198
7
198
8
198
9
199
0
199
1
199
2
199
3
199
4
199
5
199
6
199
7
199
8
199
9
200
0
200
1
200
2
200
3
200
4
200
5
200
6
200
7
200
8
200
9
201
0
201
1
IPO
In
dex
Med
ian
An
nu
al F
low
(m
³/s)
Median Annual Flow Mean Annual IPO
Mataura River at Gore
6CM088.00
July 2012 13
Figure 4.9: Comparison of the annual minima flows in the Mataura River with variation in
the SOI (1978-2011).
Figure 4.10: Comparison of the annual minima flows in the Mataura River with variation in
the IPO (1978-2011).
-4.0
-3.0
-2.0
-1.0
0.0
1.0
2.0
3.00
5
10
15
20
25
30
35
197
8
197
9
198
0
198
1
198
2
198
3
198
4
198
5
198
6
198
7
198
8
198
9
199
0
199
1
199
2
199
3
199
4
199
5
199
6
199
7
199
8
199
9
200
0
200
1
200
2
200
3
200
4
200
5
200
6
200
7
200
8
200
9
201
0
201
1
SO
I In
dex
An
nu
al M
inim
um
Flo
w (
m³/
s)
Annual Minima Mean Annual SOI
-6.0
-4.0
-2.0
0.0
2.0
4.0
6.0
8.0
0
5
10
15
20
25
30
35
197
8
197
9
198
0
198
1
198
2
198
3
198
4
198
5
198
6
198
7
198
8
198
9
199
0
199
1
199
2
199
3
199
4
199
5
199
6
199
7
199
8
199
9
200
0
200
1
200
2
200
3
200
4
200
5
200
6
200
7
200
8
200
9
201
0
201
1
IPO
In
dex
Med
ian
An
nu
al F
low
(m
³/s)
Annual Minima Mean annual IPO
Mataura River at Gore
6CM088.00
July 2012 14
Figure 4.11: Comparison of the deviation from long-term median flows with the variation in
the SOI (1978-2011).
Figure 4.12: Comparison of the deviation from long-term median flows with the variation in
the IPO (1978-2011).
-4
-3
-2
-1
0
1
2
3-30
-20
-10
0
10
20
30
40
197
8
197
9
198
0
198
1
198
2
198
3
198
4
198
5
198
6
198
7
198
8
198
9
199
0
199
1
199
2
199
3
199
4
199
5
199
6
199
7
199
8
199
9
200
0
200
1
200
2
200
3
200
4
200
5
200
6
200
7
200
8
200
9
201
0
201
1
SO
I In
dex
Devia
tio
n fro
m lo
ng
-term
med
ian
(m
³/s)
Flow Deviation SOI Index
-4
-3
-2
-1
0
1
2
3
4
5
-30
-20
-10
0
10
20
30
40
197
8
197
9
198
0
198
1
198
2
198
3
198
4
198
5
198
6
198
7
198
8
198
9
199
0
199
1
199
2
199
3
199
4
199
5
199
6
199
7
199
8
199
9
200
0
200
1
200
2
200
3
200
4
200
5
200
6
200
7
200
8
200
9
201
0
201
1
IPO
In
dex
Devia
tio
n fro
m lo
ng
-term
med
ian
(m
³/s)
Flow Deviation IPO Index
Mataura River at Gore
6CM088.00
July 2012 15
4.4 Summary
Hughes et al. (2011) found a significant link between rainfall within the Mataura catchment
and the SOI and IPO indices. It is important to recognise, however, that while various
climatic indices may affect rainfall, and in particular the variability in rainfall, the same
signature may not be apparent in the flow regime. This is because runoff in a river is the net
result of a wide range interacting factors; including both the climate and the physical
characteristics of the catchment.
The SOI and IPO indices have a weak effect on the variation in flow relative to average
conditions. This relationship, however, is more obvious over the early part of the flow record.
Periods when the SOI is strongly negative tend to be associated with higher than average
flow conditions and vice versa. These periods of higher than average flows are also
associated with periods when the IPO is strongly positive. Periods with negative IPO indices
are associated with lower than average flow conditions. Therefore it would appear that with
respect to the Mataura catchment:
• A positive IPO index is associated with a negative SOI index;
• A positive IPO index is often associated with higher than average flow conditions;
• A negative IPO index is often associated with lower than average flow conditions; and
• While variation in the IPO and SOI indices would appear to be associated with changes
in the average flow conditions, they do not appear to affect the maximum and minimum
flows experienced in any year.
The last 10 years of record is generally characterised by lower than average flow conditions.
It is perhaps not a coincidence that this period also coincides with a dramatic increase in the
abstraction of water for irrigation.
5 Mataura Water Conservation Order
For much of the year there is considerable flow in the Mataura River, however, flows can
drop to low levels for significant periods over summer and autumn; the potential irrigation
season and period of highest water demand.
The Mataura Water Conservation Order (WCO) was drafted to protect the river and maintain
the outstanding recreational fisheries (fish stocks and habitat) values, particularly during
periods of low flow. The provisions of the WCO apply to all the surface water resources
across the entire catchment with the exception of a few small streams. The Mataura WCO
has been interpreted to mean that in-stream flows must be maintained at 95% of their natural
level. Any permit to abstract water from either the Mataura River, or shallow aquifers with a
direct hydraulic connection to the river, must comply with the WCO. This means that
Mataura River at Gore
6CM088.00
July 2012 16
abstractions have to cease when flows at Mataura River at Gore reach particular threshold
levels e.g. 17m³/s.
Because of the size of the Mataura catchment, and the gradual low flow recession, the mean
daily flow and instantaneous flow are very similar in the Mataura River during summer low
flow periods; at least within the accuracy and resolution of the available flow data. Analysis
of the mean daily flows in the Mataura at Gore shows that abstraction is restricted at a flow of
17m³/s operate most years, although with differing degrees of severity (Figure 5.1).
Figure 5.1: Mataura River at Gore periods of abstraction with a WCO cut-off of 17m³/s (1977-
2012).
It would appear that to maintain the necessary minimum flow within the Mataura River at
Gore restrictions on abstractions are imposed almost every year. The number of times
abstraction is restricted, and the duration of the periods of restricted abstraction, both vary on
an annual basis. The longest period of restricted abstraction, assuming a minimum flow of
17m³/s, was in 2001 and lasted for 82 days. In only three years since 1998 have no
restrictions been imposed i.e. 2000, 2005, and 2011. Over the past 5-6 years it would appear
that periods of restriction have been more frequent but of shorter duration.
Table 5.1 shows the frequency with which abstractions were restricted for various durations
i.e. how long the mean daily flow remained below 17m³/s. Thus, on 7 occasions the mean
daily flow went below 17m³/s for only 1-day, and on 8 occasions the flow was below this limit
for 2-days. On two occasions abstraction was restricted for more than 51 days.
Table 5.1 also shows the total number of days each year when abstraction were restricted
because of the low flow provisions in the WCO. The greatest number of abstraction-
restricted days occurred in 2011. Of the total of 82 days when restrictions were in place, 80
occurred consecutively (Figure 5.2).
1978 1988 1998 2008
0
20
40
60
80
100
Mataura River at Gore duration of restricted abstraction (days)
Mataura River at Gore
6CM088.00
July 2012 17
Table 5.1: The frequency and duration of periods of no abstraction, assuming a cut-off at
17m³/s. Also shown are the total number of days when abstraction was
restricted in each year. Years when no restrictions were imposed are not
included.
Days Frequency Year Days Year Days
1 7 1978 47 2008 69
2 8 1981 65 2009 28
3 1 1985 17 2010 42
4 8 1986 12 2012 17
5 5 1988 4
6 3 1989 22
7 4 1990 57
8 7 1991 12
9 2 1995 31
10 1 1998 7
11-15 8 1999 53
16-20 3 2001 82
21-30 5 2002 9
31-40 1 2003 45
41-50 1 2004 57
51-100 2 2006 15
Total 66 2007 24
Figure 5.2: Total duration, and consecutive days of no abstraction each year with an
abstraction cut-off of 17m³/s in the Mataura River at Gore (1978-2012).
0
10
20
30
40
50
60
70
80
90
197
8
197
9
198
0
198
1
198
2
198
3
198
4
198
5
198
6
198
7
198
8
198
9
199
0
199
1
199
2
199
3
199
4
199
5
199
6
199
7
199
8
199
9
200
0
200
1
200
2
200
3
200
4
200
5
200
6
200
7
200
8
200
9
201
0
201
1
201
2
Days w
ith
ou
t ab
str
acti
on
Total Days Consecutive Days
Mataura River at Gore
6CM088.00
July 2012 18
The periods of restricted abstraction were compared with both the IPO and SOI indices to
determine whether these climatic oscillations had a significant effect on potential water
abstraction (Figure 5.3). There appears to be no correlation between the various phases and
strength of the IPO and SOI indices and the number and durations of period of restricted
abstraction.
Figure 5.3: Relationship between periods of no abstraction and the IPO and SOI indices.
6 Effects of Climate Change
Predictions of future climate depend on projections of future concentrations of greenhouse
gases and aerosols, as well as on model assessments of how the global climate system will
respond to these changing concentrations (MfE, 2008). When the results of all the various
models, and all the IPCC emissions scenarios, are considered a wide range of projected
temperature increases are derived for New Zealand. These projected increases range from
0.2–2.0°C by 2040; and 0.7–5.1°C by 2090. The mid-range projections are that New
Zealand temperatures will increase by about 1°C by 2040, and 2°C by 2090, relative to the
temperature in 1990 (MfE, 2008).
Figure 6.1 shows the annual-average pattern of warming over New Zealand. Warming is
projected to be fairly uniform over the country, although slightly greater over the North Island
than the South Island. The winter season in the South Island has the greatest warming,
whereas spring has the least warming of all seasons.
The projected mid-range change in the average annual rainfall has a pattern of increased
rainfall for the Southland region (Figure 6.2). This annual pattern of ‘wetter in the south and
west and drier in the east’ results from the changes in the dominant seasons of winter and
spring.
-4
-3
-2
-1
0
1
2
3
4
5
0
10
20
30
40
50
60
70
80
90
197
8
197
9
198
0
198
1
198
2
198
3
198
4
198
5
198
6
198
7
198
8
198
9
199
0
199
1
199
2
199
3
199
4
199
5
199
6
199
7
199
8
199
9
200
0
200
1
200
2
200
3
200
4
200
5
200
6
200
7
200
8
200
9
201
0
201
1
201
2
Ind
ex
Days w
ith
ou
t W
ate
r
Total Days Consecutive Days SOI Index IPO Index
Mataura River at Gore
6CM088.00
July 2012 19
Figure 6.1: Projections for increases in mean annual temperature by 2040 and 2090
(Ministry for the Environment, 2010).
Figure 6.2: Projections for increases in rainfall by 2040 and 2090 relative to 1990 (Ministry
for the Environment, 2010).
Mataura River at Gore
6CM088.00
July 2012 20
Overall, the projected impacts of this pattern of warmer temperatures will mean an increase
in westerly airflows and rainfall in the Southland region. The increased rainfall may cause an
increase in flow in the Mataura River at Gore, and consequently fewer and shorter periods of
restricted abstraction. However, changes in temperature may also result in changes in the
wind run and a net increase in evapotranspiration. If this was to occur, the net water balance
may be very similar to the current situation.
However, in all but the extreme scenarios any changes in water demand and availability
resulting from climate change are likely to be significantly less than that caused by natural
variability resulting from SOI and IPO variations in atmospheric circulation (MfE, 2008).
The potential impacts of climate change on rainfall, river flow, and irrigation demand were
reviewed comprehensively in Morgan and Evans (2003). Two Global Climate Models
(CSIRO9 and HadCM2) were applied to the Oreti catchment and showed increases in rainfall
of between 1-2% (CSIRO9) and 7-13% (HadCM2). An increase of 1°C in mean monthly
temperatures was also predicted, along with an increase in average annual
evapotranspiration of 4%. The net effect of these changes is predicted to be small to
moderate increases in some river flows (depending on the model) and an increase in long
term average flow (based on HadCM2 model). Overall, the assessment suggested climate
change (until 2050) will only have a small impact on the Southland region, with a slight
reduction in drought frequency and severity. These results are generally consistent with
those predicted and outlined in MfE (2008).
7 Groundwater
7.1 Background
Both shallow and deeper aquifers are known to exist in the vicinity of Gore. The shallow
aquifer is unconfined and occurs within the Quaternary glacial outwash and alluvial terraces
of the Mataura River. This aquifer is referred to as the Knapdale groundwater zone (SKM,
2006). This aquifer is primarily recharged through infiltration of rainfall, but also receives
significant recharge from the Mataura River and its tributaries to which it is hydraulically
connected (SKM, 2006). The Coopers and Jacobstown wells all draw groundwater from
shallow unconfined aquifers (Figure 7.1). As an unconfined, transmissive aquifer, the
Knapdale aquifer is regarded as being vulnerable to potential contamination from land use
activities.
Mataura River at Gore
6CM088.00
July 2012 21
Figure 7.1: Simplified regional geological map (Gusyev et al., 2011).
Mataura River at Gore
6CM088.00
July 2012 22
7.2 Connectivity between groundwater and surface water systems
Extensive interaction exists between the surface water and groundwater resources within the
Mataura catchment (Hughes et al., 2008). Generally, throughout the catchment there are
patterns of losses or gains in streamflow that cannot be accounted for simply by measured
tributary inflows. Hughes et al. (2008) discuss a relatively constant flow loss near the
Riversdale Bridge during low flow conditions. This loss is thought to make a significant
contribution to the water balance of the adjacent Riversdale groundwater zone. Similar
patterns of losses and gains of streamflow have been observed with regard to the Mataura
River and the adjacent Knapdale aquifer in the vicinity of Gore.
Figure 7.2 and Figure 7.3 show the groundwater levels recorded in two of the bores in the
Jacobstown borefield, and the corresponding flow in the Mataura River at Gore. These data
illustrate the close relationship between groundwater levels and river flow; despite the fact
that the groundwater levels are also affected by pumping to meet community water supply
demand. When river flows are high groundwater levels tend to rise, while during periods of
low river flows groundwater levels fall. This type of interaction is typical of that observed in
riparian aquifer systems as a result of flow variations into and out of the groundwater system,
usually in response to changes in the rivers flow and stage (Hughes et al., 2008).
Figure 7.2: The relationship between groundwater levels in the Jacobstown well-field (Well
1) and the flow in the Mataura River at Gore (2005-2012).
2006 2008 2010 2012
0
100
200
300
400
500
600
700
800
900
1000
0
1000
2000
3000
4000
5000
6000
7000
8000
9000
10000
Flo
w (
m³/
s)
Mataura River at GoreJacobstown No1
Mataura River at Gore
6CM088.00
July 2012 23
Figure 7.3: The relationship between groundwater levels in the Jacobstown well-field (Well
3) and the flow in the Mataura River at Gore (2005-2012).
To isolate the groundwater response to river levels from the effects of pumping data from the
nearest Environment Southland monitoring bore was obtained (i.e. bore F45/0569).
Although this bore is located several kilometres up-river from Gore, it is located in a similar
riparian aquifer to that tapped by the Jacobstown and Coopers well-fields. Figure 7.4 shows
the groundwater levels recorded at bore F45/0569 and the flow in the Mataura River at Gore.
There is a close relationship between the groundwater level and flow in the Mataura River.
In general, flood events are associated with a rapid rise in the groundwater while levels
decrease slowly over periods of low flow. In effect, the shallow unconfined groundwater
would appear to largely reflect the hydrograph of the Mataura River but in more subdued and
buffered manner.
There is little doubt therefore that there is a strong hydraulic connection between the Mataura
River and the adjacent shallow unconfined aquifer which is tapped by the Jacobstown and
Coopers well-fields. Because of this strong hydraulic connection, any abstraction from these
well-fields would likely be subject to the same restrictions as surface water permits so as to
comply with the Mataura Water Conservation Order.
2006 2008 2010 2012
0
100
200
300
400
500
600
700
800
900
1000
0
2000
4000
6000
8000
10000
12000
14000
16000
18000
20000F
low
(m
³/s)
Mataura River at GoreJacobstown No3
Mataura River at Gore
6CM088.00
July 2012 24
Figure 7.4: The relationship between groundwater levels at Bore F45/0569 and the flow in
the Mataura River at Gore (January-July 2005).
8 Conclusions
Flow within the Mataura River is highly variable, mostly because of its alpine headwaters but
also because of the considerable size of the catchment. Over the last 35 years flows in the
Mataura River at Gore have varied from a low of 8m³/s to a maximum flood of 2297m³/s.
The flow regime of the Mataura River at Gore shows both seasonal and longer term
variability, particularly with respect to flood activity. Winter is characterised by generally
higher flows and a greater number of moderate flood events, however, the largest floods
generally occur during spring and summer. This seasonal and annual variability in flow can
be accommodated relatively easily within a water supply scheme. Managing the longer term
variability is more problematic.
The presence of cyclic behaviour and periods of either increased or decreased flows has
considerable importance when assessing the reliability of water resources. Sustained
periods of reduced flows in the Mataura River have significant implications to the volume of
water which could be abstracted from the shallow unconfined aquifer, and the duration of
periods when no abstraction may be possible. There are also implications for the resilience
of supply, and potentially technical constraints with regard to water supply infrastructure.
With respect to the Mataura catchment:
• A positive IPO index is associated with a negative SOI index;
Jan-2011 Mar-2011 May-2011
0
100
200
300
400
500
600
86.8
86.9
87.1
87.3
87.5
87.7
87.9
Mataura River at Gore - Flow (m³/s)Bore F45/0569 - Groundwater Level (m)
Mataura River at Gore
6CM088.00
July 2012 25
• A positive IPO index is often associated with higher than average flow conditions;
• A negative IPO index is often associated with lower than average flow conditions; and
• While variation in the IPO and SOI indices would appear to be associated with changes
in the average flow conditions, they do not appear to affect the maximum and minimum
flows experienced in any year. For example, there appears to be no correlation
between the various phases and strength of the IPO and SOI indices and the number
and durations of periods of restricted abstraction.
The last 10 years have been characterised by lower than average flow conditions. It is
perhaps not a coincidence that this period also coincides with a dramatic increase in the
abstraction of water for irrigation.
The potential impacts of climate change on rainfall, river flow, and irrigation demand has
been reviewed comprehensively. Overall, the assessment suggests climate change (until
2050) will only have a small impact on the Southland region with a slight reduction in drought
frequency and severity.
Extensive interaction between the surface water and groundwater resources within the
Mataura catchment means that those factors which affect the flow regime of the Mataura
River are also likely to affect the adjacent shallow groundwater resource. Both the
Jacobstown and Coopers well-fields, which supply Gore’s potable water supply, are located
in the shallow Knapdale aquifer. The strong hydraulic connection between river flow and
shallow groundwater in the vicinity of Gore means that any abstraction from these well-fields
would likely be subject to the same restrictions on abstraction as surface water permits.
These restrictions are to comply with the Mataura Water Conservation Order.
The Mataura Water Conservation Order is therefore likely to act as a major constraint on any
future development of the shallow groundwater resource in the vicinity of Gore. This problem
is compounded by the fact that the potential impact of the WCO is greatest when the demand
for water is likely to be highest i.e. during periods of low flow during summer and early
autumn.
9 References
Gusyev, M.; Moureau-Fournier, M.; Tschritter, C. 2011: Capture zone delineation for Gore District
Council drinking water production wells. GNS Science Consultancy Report 2011/32, February,
2011.
Hughes, B.; Harris, S.; Brown, P. 2011: Mataura catchment strategic water study. Report prepared
for Environment Southland, May 2011.
Kidson, J.W.; Renwick, J.A. 2002: Patterns of convection in the Tropical Pacific and their influence on
New Zealand weather. International journal of climatology 22: 151-174.
Mataura River at Gore
6CM088.00
July 2012 26
McKerchar, A.I.; Henderson, R.D. 2003: Shifts in flood and low-flow regimes in New Zealand due to
interdecadal climate variations. Hydrological sciences 48(4): 637-654.
Ministry for the Environment, 2008: Climate Change Effects and Impacts Assessment: A guidance
manual for local government in New Zealand, 2nd edition.
Ministry for the Environment, 2010: Preparing for future flooding. A guide for local government in New
Zealand.
Morgan, M.; Evans, C. 2003: Southland water resources study – stages 1-3. Report prepared for
Venture Southland, September 2003. Report no. 4597/1.
SKM, 2006: Gore water master planning study. Report prepared for Gore District Council, January
2006.
Mataura River at Gore
6CM088.00
July 2012 27
