, Volume 1, Issue 2, pp 184–195

Development of a Tool to Facilitate Participation of Maori in the Management of Stream and River Health


    • Department of ZoologyUniversity of Otago
  • Gail Tipa
    • Te Runanga o Moeraki
  • Laurel D. Teirney
    • Southern Woman Consultancy
  • Dev K. Niyogi
    • Department of ZoologyUniversity of Otago
    • Department of Biological SciencesUniversity of Missouri-Rolla
Original Contributions

DOI: 10.1007/s10393-004-0006-9

Cite this article as:
Townsend, C.R., Tipa, G., Teirney, L.D. et al. EcoHealth (2004) 1: 184. doi:10.1007/s10393-004-0006-9


A cultural health index (CHI) for streams was developed in a program of collaborative research involving members of Ngai Tahu (an iwi [tribe] within the South Island of New Zealand) and ecologists at Otago University. The aim was to provide a tool for effective participation of Maori in resource management decisions. Five cultural values are of central importance to the nature of the CHI: mauri (spiritual life force), mahinga kai (traditional resource harvesting), kaitiakitanga (guardianship obligation), ki uta ki tai (mountains-to-the-sea holistic philosophy), and wai taonga waters that are treasured). The CHI has three components. Forty-six stream sites in two culturally important river catchments were first classified according to whether there is a traditional association with Maori. The second component assessed the historical and contemporary mahinga kai status of the site, including questions of legal and physical access. The third component was a Cultural Stream Health Measure (CSHM) that encapsulates indicators of catchment, riparian, and instream condition in a manner that is consistent with Maori values. The CSHM was found to be significantly correlated with “western” measures of stream health commonly used in New Zealand (Macroinvertebrate Community Index, Stream Health Monitoring and Assessment Kit), and performed at least as well in encapsulating the relationship between land development and stream health. We describe a multistep process by which other indigenous people can develop a cultural ecosystem health measure, and then use the tool to ensure a substantial role in decision making with the agency in charge.


ecosystem healthcultural health indexresource managementcatchmentriparian


Indigenous People and Resource Management in New Zealand

Maori chiefs and their iw (tribes: see Table 1 for a glossary of Maori words) signed the Treaty of Waitangi with the British Crown in 1840 in part to protect their rights to access, to use and to manage resources of significance to them. However, traditional harvesting of mahinga kai (the cultural practice of gathering food and other materials) was soon restricted and eventually virtually extinguished (Evison, 1993). In the modern era, the responsibilities of the Crown reside in the Government of New Zealand whose resource management agencies have a statutory obligation to “give effect to” or “take account of” the Treaty of Waitangi and its principles and ”to recognise and provide for“ cultural values and practices. However, the Crown has imposed these obligations on resource management agencies without providing the tools with which to meet them.
Table 1.

Glossary of Maori Words

Maori word



Sub-tribe, extended whanau




Specifically appointed guardians responsible for protecting valued taonga


The exercise of guardianship


A respected elder within the tribe

Ki uta ki tai

From the mountains to the sea

Mahinga kai

Food and other resources, and the areas where they are collected


The essential life force or principle: a metaphysical quality inherent in all things both animate and inanimate

Ngai Tahu

An iwi whose tribal territory extends over the lower two-thirds of the South Island of New Zealand and includes Stewart Island and the sub-Antarctic islands. A definition of the tribal area is set out in section 5 of the Te Runanga o Ngai Tahu Act 1996




Teenager, young adult


A local representative group or community system of organization within a defined geographic area




Treasured possessions, both tangible and intangible

Wai taonga

Waters that are valued



The purpose of this article is to describe the development of a tool, based on the beliefs, values, and practices of Ngai Tahu, an iwi in the South Island of New Zealand, to assess the health of streams. We describe how the tool was developed, and outline a general process that other iwi and other indigenous peoples can use to generate their own ecosystem health measures and so facilitate their participation in decision making with relevant agencies.

Maori Perspective on Freshwater and Its Management

The long history of Maori occupation and travel throughout New Zealand has yielded many stream and river sites of significance and has enabled iwi to accumulate knowledge and develop management practices appropriate for the sustainable use of freshwater resources in their tribal areas. Maori knowledge, cultural values, and customary practices, which have been fundamental to the development of the stream health tool, remain valid today and have the potential to complement contemporary management practices and enhance the overall effectiveness of management. Cultural values of significance include: mauri, mahinga kai, kaitiakitanga, wai taonga, and ki uta, ki tai.

Maori believe that the forests, the waters, and all the life supported by them possess a mauri or life force (Marsden, 1992). Through mauri, all things in nature are coherent. The primary management principle for Maori is the protection from desecration of the mauri or life-giving essence of a resource. Maori have become increasingly concerned about threats to the integrity of the waterways on which their survival and their cultural identity depend. The decline in both water quantity and quality has impacted on cultural values and uses of rivers, and puts at risk the mauri which is unable to protect itself against unnatural changes to the environment.

Without mahinga kai, Ngai Tahu would lose a key component of their cultural identity. Mahinga kai encompasses the ability to gain access to the resource (both legal and physical access), the site where gathering occurs, the activity of gathering, the fitness of the resource for cultural usage, and the spiritual and mental well-being associated with being able to continue a cultural practice. Freshwater mahinga ka resources were of crucial historical importance and remain important to contemporary Maori.

Maori have a holistic ki uta ki tai (mountains to the sea) perspective. In the context of freshwater management, it is necessary to consider a catchment in its entirety, from its source, the passage of its waters through a network of tributaries, onto lower floodplains, and to its interface with saltwater in estuaries along the coast.

Kaitiakitanga includes the obligations of the iwi and its members to be custodians, protectors and guardians of their taonga (treasures) and other resources. This approach to environmental management is holistic (Duker, 1994). As kaitiaki, the mandated individuals within an iwi are obliged to protect the interests of future generations and to stress the importance of ancestors to tribal identity. Only by honoring those who have come before can the interests be truly protected of those yet to be. It is this continuum that makes development within the limits of sustainability of key significance to Maori.

Wai taonga refers to waters that are significant to whanau (family), hapu (sub-tribe or extended family), and/or iwi because of their value as places from which resources are customarily gathered, that are ecologically significant (e.g., as breeding or migratory habitats), or where particularly significant species or taonga are located (Ministry for the Environment, 1998).

Development of a Stream Cultural Health Index (CHI) with Three Components

Against this cultural background, and acknowledging the importance of kaitiakitanga, we developed a Cultural Health Index with three components.
  1. 1.

    Traditional status of the stream site: Sites were classified according to whether they have a traditional association with iwi, and whether iwi believe they would use the site in future.

  2. 2.

    Mahinga kai measure: The indicators and data needed to derive this measure were obtained from interviews with kaumatua (respected elders) and iwi resource managers, as well as contemporary assessments of mahinga kai availability.

  3. 3.

    Cultural Stream Health Measure (CSHM): A number of cultural indicators of stream health were derived after interviews with kaumatua and iwi resource managers, and assessed at 46 sites by teams of Maori who were mandated as kaitiaki of the particular sites. We describe how elements of this database were used to provide a measure of stream health that encapsulates limited tangible aspects of mauri and provides a holistic assessment in line with the ki uta ki tai philosophy.


The CSHM is the only element of the CHI with “western” equivalents. If the CSHM is appropriate to its stated purpose for Maori, we recognize that it is not necessary for it to equate in any way to existing “western” measures. However, we expected that dialogue between Maori and resource management agencies would be enhanced if the new measure could be viewed in a broader context. Thus, to place the work in a broader perspective, we compare the CSHM with two stream health measures commonly used in New Zealand (Boothroyd and Stark, 2000). We also use a Geographic Information System (GIS) of contemporary land use to compare the performance of the cultural and western stream health measures in relation to the degree of land development in the stream catchments.


Site Selection

The study was performed in the summer of 2000/2001 (December 2000 to February 2001) in two river catchments of significance to Ngai Tahu, as recognized by the Crown in the Ngai Tahu Claims Settlement Act of 1998. The Taieri and the Kakaunui Rivers are respectively in the territories of Te Runanga Otakou and Te Runanga o Moeraki; each runanga (a local representative group or community system of organization within a defined geographic area) holds significant traditional knowledge about their rivers.

The Taieri River rises in the uplands of Central Otago and meanders through the block mountains of the upper catchment before passing through an incised gorge and crossing the Taieri Plains in the Lower Taieri. The Taieri joins the Waipori River before it passes through another gorge to the sea. The Kakaunui River drains coastal hills further north in the province of Otago, rising in high country and passing predominantly through dry lowlands. Sites were chosen in three size classes: small headwater streams (first and second order streams), medium-sized tributaries of the mainstem (third and fourth order streams), and large major tributaries/mainstem (fifth order or bigger). The sites had a range of land uses in their catchments (native tussock grassland, pasture, native forest, and pine plantation). In the Taieri River, thirty stream sites spanned the above categories, with more than half being of traditional significance (Fig. 1a). Headwater streams were not available for study in the Kakaunui, because most had dried up during the summer period when sampling took place; 8 medium and 8 large streams were sampled, of which 11 were of traditional significance (Fig. 1b).
Figure 1.

Location of sites in (a) the Taieri River and (b) the Kakaunui River. Small sites are shown as triangles, medium sites as circles, and large sites as squares. Solid symbols indicate sites with traditional Maori significance. The positions of the two river catchments in the South Island of New Zealand are shown in the top left.

Three Components of the CHI

To derive the CHI index at a particular stream site, first iwi association was identified, then mahinga kai values were assessed and finally cultural stream health was evaluated. The first two components were assessed mainly in an objective manner but partly subjectively (see next section), whereas the cultural stream health measure required preliminary work and analysis, as described below.

Traditional Association

Sites judged from historical records and kaumatua knowledge to be of traditional significance were assigned the letter “A,” and those without traditional significance a “B.” If runanga members judged they would return to the site in future, the site was assigned the number “1” and, if not, a “0.”

Mahinga Kai Measure

Each of four features (Table 2) was rated on a 1–5 scale (1 is poor and 5 the highest mahinga kai rating) and the mahinga kai score was derived as the average of the four ratings. The first feature is number of mahinga kai species currently present (plant and bird mahinga kai resources associated with the stream or riparian margins were assessed visually by runanga members, and fish and crayfish by electric fishing a 50-m stretch of stream, or in larger sites by netting and electric fishing). The second feature is the percentage of mahinga kai species known historically from a site that are still present there. The third feature is the current status of access to the site (combining issues of legal and physical access). Finally, runanga members judged whether or not they would come to use the site in the future. This fourth feature, concerning future use, recognizes the different perspectives of Maori and non-Maori in the specification of water standards and the definition of water pollution. For example, Maori spiritual values with respect to drinking water include standards and perceptions of pollution that may conflict with scientific measures. A “drinking water standard” may be scientifically defined as carrying contaminants, but at a level that is not toxic. In other words, a certain level of degradation can occur. In contrast, Maori would require drinking water to be protected from physical and spiritual pollution, which requires an absolute prohibition on certain discharge activities (Ministry for the Environment, 1998).
Table 2.

Mahinga kai Measurea


Score 1

Score 2

Score 3

Score 4

Score 5

Current mahinga kai species






Historical species still present






Barriers to access

Both legal and physical barriers

Either legal or physical barriers

No barriers

Would Maori use the site



aEach of four features was rated on a 1–5 scale (1 is poor and 5 is the highest mahinga kai rating) and the mahinga kai score was derived as the average of the four ratings. Dashes represent component scores not used.

Cultural Stream Health Measure

In addition to the many intangible qualities associated with the spiritual presence of a waterway, mauri can in part be tangibly represented by some of its physical and biological characteristics, including water quality, flow, condition of the catchment, riparian zone and streambed, and indigenous riparian and instream vegetation (Ministry for the Environment, 1998). Extensive structured interviews with 28 runanga members produced a list of indicators that iwi use to assess stream health (Tipa, 1999). Of these, 17 were identified as potentially useable indicators that together could comprise an overall measure of stream health. An 18th measure was an overall subjective assessment of how healthy a particular stream appeared to runanga assessors. Overall health has particular significance in developing the CSHM because it provides the yardstick against which to judge the other, mostly more objectively measurable indicators. Three generations were represented in the runanga teams— kaumatua pakeke (adults), and rangatah (teenager, young adult). Each team comprised at least four members, who individually and without discussion recorded their scores for each of the indicators at each of the sites on their river.

The indicators are listed in a series of categories in Table 3. Pearson product-moment correlation coefficients between each indicator and “overall stream health” (for the 46 sites) are also shown (a correlation coefficient of 0.29 or more is significant at the P < 0.05 level of confidence).
Table 3.

Indicators Evaluated for Inclusion in the Cultural Stream Health Measure (CSHM)

Indicators—category and number

Extremes of descriptors (1–5 scale)

Correlation with overall healthb

River health values


Overall stream health

Very unhealthy/very healthy



Fish safe to eat

Completely unsafe/completely safe



Water safe to drink

Completely unsafe/completely safe



Would go fishing

Would not/great place to fish


Catchment scale


Catchment land use

Land and margins grazed by stock/naturala


Riparian river-margin scale


Use of riparian margin

Riparian zone grazed/naturala



Riparian vegetation

Little or none/indigenous riparian vegetation



Indigenous species

Exotic vegetation on adjacent land + margins/indigenous vegetation



Riverbank condition

Banks eroding/stable


Instream physical characteristics


Use of river—modification

Evidence of modification/appears naturala




Sediment covering bed/bed free of sediment



Riverbed condition

Muds and sands/cobbles and gravels



Use of river—takes and discharges

Takes and discharges evident/no evidence


Instream flow


River flow—visible

No movement/movement visible (1 or 5)



River flow—audible

No sound/flow audible (1 or 5)


Instream water quality (WQ)



Foams and or oils evident, nota




Unpleasant odors evident, not




Discoloration, clear


aThese indicators were chosen for inclusion in the CSHM.

bPearson correlation coefficients are shown for the relationships between each indicator and overall stream health.

Every ecosystem health index is a human construct that depends on a subjective value of what is considered healthy (Townsend and Riley, 1999). Thus, the cultural stream health measure derived in this study must encapsulate and be closely related to the overall measure of what runanga members, as kaitiaki, consider healthy. Selecting the indicators that best express stream health from a cultural perspective involved the following steps:
  1. A.

    Identify and remove value judgment indicators that are equivalent to overall stream health and that cannot be consistently measured.

  2. B.

    Determine indicators, from several a priori categories of indicators, that are most strongly correlated with “overall stream health” (catchment category, riparian river-margin category, instream physical category, instream flow category, and instream water quality category).

  3. C.

    Perform a stepwise multiple regression analysis to determine, in a rigorous statistical manner, which variables together account for most of the variation in “overall stream health” at a site.

  4. D.

    Compare the outcome of the multiple regression analysis in step C with the variables selected in step B, and decide on a final set of indicators to include in the cultural stream health component of the CHI. Average the scores for the selected indicators to provide the cultural stream health component at every site.


Comparison of CSHM with Western Measures

We compared CSHM values with two commonly used “western” measures of stream health: the Macroinvertebrate Community Index (MCI) and the Stream Health Monitoring and Assessment Kit (SHMAK).

MCI is widely used by regional authorities and researchers to assess stream health and is based on the presence or absence of certain types of invertebrates on the streambed that differ in their ability to tolerate pollution (Stark, 1985, 1993). Healthy streams have high values of the MCI (120 or above), whereas unhealthy streams have values as low as 80 or less. We employed Ngai Tahu scientists to sample and identify invertebrates from all 46 stream sites. These scientists were not involved in the runanga stream health assessments.

SHMAK was developed as a tool for landowners to monitor the health of streams flowing through their properties, particularly in relation to changes to riparian or catchment land use (Biggs et al., 1998), and has been quite widely adopted by community groups working to improve stream health in both rural and urban settings. SHMAK was developed primarily for comparing stream health at a site through time, and upstream and downstream sections of the same stream, as well as to compare results with neighbors within a catchment or region. It is made up of a habitat component and an invertebrate component (we used level 2* from the protocol described in Biggs et al., 1998). Measurements of flow velocity, pH, temperature, conductivity, clarity, bed composition, deposits, and bank vegetation over a 10-m stream reach, make up the habitat component. The invertebrate component is based on a simplified version of the MCI, incorporating just 17 invertebrate taxa. The outcome of the SHMAK protocol is a categorical rating of very poor, poor, moderate, good, very good, and excellent health (the details of derivation of ratings differ somewhat according to whether the stream bed is stony, silty, etc.). We converted these ratings to numerical values of 1, 2, 3, 4, 5, and 6 for statistical comparisons with the CSHM at the 46 sites.

Relationship between Stream Health Measures and Land Development

We used a Geographic Information System developed at the University of Otago (Arbuckle et al., 1999) to determine “% of developed land” for each stream catchment, defining developed land as the sum of bare ground, urban, pasture, and pasture plus riparian willows. A topical question in stream ecology is the scale at which stream health is influenced by land use (Allan et al., 1997; Townsend et al., 2003). Is stream health influenced more by land use in the entire catchment (all the land draining into the headwaters that feed the site in question) or just in the riparian zone adjacent to the stream? We address this question and check how well the cultural and western measures of stream health encapsulate the land-use effect. We compare the relationships between the cultural stream health measure, MCI, and SHMAK, and land use using three scenarios: percentage of developed land in the entire catchment above each site, in a 100-m-wide riparian strip from the site upstream to its furthest headwaters, and in a 100-m-wide riparian strip extending for just 500 m above each site.


Traditional and Mahinga Kai Components of the CHI

The information on traditional and mahinga kai components of the CHI belongs to the kaitiaki runanga who will use it in discussions about resource management with the appropriate agency. Thus, we cannot present all the information here. Instead, as examples, we provide traditional and mahinga kai scores for five sites (Sites 1, 6, 11, and 21 from the Taieri River; Site 38 from the Kakaunui River), with a brief analysis in each case.

Site 1 (McRaes Creek). Traditional score B-1 (not a traditional site, but runanga members would use it in future); mahinga kai score 2.69 (this is a higher than average score reflecting the reasonable range of mahinga kai species present, the fact that it is not a traditional site so no historic comparison can be made—a value of 1 assigned for this component, that it is accessible but involves a significant walk, and that runanga members would use it in future).

Site 6 (Barbours Stream). Traditional score B-0 (not a traditional site and, given its current state, runanga members would not use it in future); mahinga kai score 1.3 (this is a low score reflecting the absence of mahinga kai species, the fact that it is not a traditional site, that it has poor access, and runanga members would not wish to use it in future).

Site 11 (Owhiro Creek). Traditional score A-0 (a traditional site but, given its overall state, runanga members would not wish to use it in future); mahinga kai score 1.75 (this is a reasonably low score reflecting the absence of mahinga kai species, except eels, a high score for the comparison between historically important and current mahinga kai because eels were traditionally important, a high score for access, but a low score because runanga members would not return).

Site 21 (Taieri River–Ferry Bridge). Traditional score A-1 (a traditional site that members would be happy to continue to use); runangascore 1.65 (this is quite a low score reflecting a good range of mahinga kai species, but the loss of several that were historically important, average access, and the fact that runanga members would use it in future).

Site 38 (Island Stream). Traditional score A-0 (a traditional site but, given its overall state, runanga members would not wish to use it in future); mahinga kai score 2.56 (this is a higher than average score reflecting a limited range of mahinga kaispecies present, the continued presence of historically important eels, good access, but runanga members would not wish to use it in future).

Cultural Stream Health Measure

To determine a reduced set of indicators to incorporate in the CSHM, we followed steps A–D as defined in the Materials and Methods section.

At step A, indicators 2, 3, and 4 (fish safe to eat? water safe to drink? would go fishing?) were found to be very strongly correlated with overall stream health (correlation coefficients greater than 0.80). Note that all four items are value judgments and are essentially measures of the same thing, indicating that the Maori concept of overall stream health is strongly tied with their perceived ability to use natural resources. For the purposes of developing a robust and repeatable index of cultural stream health, indicators 2–4 were not considered further.

Step B identified one indicator from each category in the list in Table 3 as being most strongly correlated with “overall stream health.” In the catchment category, only a single measure was available but it was strongly correlated with “overall stream health.” Two or more indicators were present in each of the other categories. Within a category, it was often the case that indicators were highly correlated with each other and may be considered to be measuring essentially the same thing (e.g., in the water quality category, smell (0.77) and discoloration (0.71) were both highly correlated with pollution). Reducing each category to a single indicator that is most highly correlated with “overall stream health” avoids problems of using several indicators of the same condition. This step yielded five indicators: Catchment scale, catchment land use; Riparian river-margin scale, use of riparian margin; Instream physical characteristics, use of river channel (modification); Instream flow, river flow (visible); Instream water quality, water quality (pollution).

At step C, a full stepwise multiple regression analysis (setting the necessary statistical significance for inclusion of an indicator as P < 0.05) yielded the four indicators below, given in order of importance. Taken together, these indicators accounted for an acceptable 76% of the variation in overall stream health at the sites. The four factors were water quality (pollution), use of riparian margin, use of river channel (modification), and river flow (visible).

Finally (step D), the conclusions from steps B and C are very similar, providing considerable confidence for the choice of indicators to include in the CSHM. The subjective choice of individual factors from five a priori habitat categories was the same as the result from the objective stepwise multiple regression analysis, except that the latter did not identify catchment land use as a significant component of the model. This is no doubt because catchment land use and use of the riparian margin were strongly correlated with each other (0.84). On this basis, we could have decided to omit catchment land use from our cultural measure. However, the holistic view of river systems held by the iwi is such that it was more appropriate to retain catchment land use. The five retained indicators are indicated in a footnote in Table 3.

Deriving the CSHM for each site involved calculating the average score for the different runanga members for each of the five indicators (each on a 1 to 5 scale), before calculating a grand mean for the five indicators together (again on a 1 to 5 scale). We performed a one-way, repeated measures Analysis of Variance (SigmaStat version 2.03) to compare the CSHM scores of rangatahi, pakeke, and kaumatua and found that kaumatua generally arrived at lower scores (mean ± standard error for CSHM scores of kaumatua were 2.70 ± 0.16, for rangatah 2.96 ± 0.15, and for pakeke 2.98 ± 0.14; F = 4.09; degrees of freedom = 2, 45; P = 0.020). This was because kaumatua were generally harsher in their evaluation of water quality and use of the riparian margin, a result that confirms the importance of runanga teams having a consistent representation.

For reference, the five sites described in the previous section had the following CSHM values: McRaes Creek 4.87, Barbours Stream 3.02, Owhiro Creek 1.65, Taieri River–Ferry Bridge 2.72, Island Stream 1.06. The scores for the three components are then brought together to give the CHI for each site:

McRaes Creek: B-1/2.69/4.87

Barbours Stream: B-0/1.3/3.02

Owhiro Creek: A-0/1.75/1.65

Taieri River–Ferry Bridge: A-1/1.65/2.72

Island Stream: A-0/2.56/1.06

Comparison of CSHM with Western Measures

The Cultural Stream Health Measure was significantly correlated with both MCI (0.58) and SHMAK (0.49) (Fig. 2a, b). The relationship between CSHM and MCI was similar for both Taieri and Kakaunui sites (see different symbols in Fig. 2), whereas the relationship between CSHM and SHMAK was much less clear for the Kakaunui. When the SHMAK measure is split into its two components, the SHMAK habitat component was strongly related to CSHM in both rivers, but the SHMAK invertebrate component was poorly correlated with CSHM in the Kakaunui sites (Fig. 2c, d). This is mainly a consequence of the widespread distribution in the Kakaunui River (particularly in larger river sites) of dense populations of larvae of Oxyethira caddisflies, a taxon that is particularly influential in reducing the invertebrate score in the SHMAK index. The relationships between CSHM and “western” measures were much stronger for small- and medium-sized streams than for larger tributary and mainstem sites (data not shown).
Figure 2.

Relationships between “western” measures of stream health and the Cultural Stream Health Measure (CSHM). Fitted lines are derived from linear regression for all sites combined (regression equation and coefficient of determination R2 shown). a: MCI, Macroinvertebrate Community Index; b: SHMAK, Stream Health Monitoring and Assessment Kit; c: SHMAK habitat component; d: SHMAK invertebrate component. Taieri sites are shown as circles and Kakaunui sites as squares.

Relationships between Stream Health Measures and Land Development

There were negative relationships between percentage of developed land at all three scales and each of the stream health measures (as expected, stream health measures declined as percentage of developed land increased) (Table 4). Of all the relationships, the strongest was between CSHM and land development in the local riparian zone (−0.56), with lower, but statistically significant, correlations with development in the entire riparian zone and entire catchment. The MCI also performed well but, in this case, showed a stronger relationship at the larger scales of entire riparian zone and entire catchment (−0.52 and −0.51, respectively) than in the local riparian zone (Table 4). SHMAK performed less well, with statistically insignificant correlations of between 0.24 and 0.28 at the three scales.
Table 4.

Correlation Coefficients between Three Stream Health Measures and Percentage of Land Developed in the Entire Catchment, in the Entire Riparian Zone, and in the Riparian Zone Extending 500 m Upstream of the Site


Percentage developed

Stream health index


Complete riparian

500-m riparian













MCI, Macroinvertebrate Community Index; SHMAK, Stream Health Monitoring and Assessment Kit.


Capturing an Iwi Perspective

The Cultural Health Index for streams is intended to enable Maori to collect data specific to their cultural values, so it was essential to ground it in their beliefs, values, and practices. The design of the CHI, and in particular the decision to have three components, responds to Maori values associated with mahinga kai, mauri, kaitiaki, and the holistic, ki uta ki tai philosophy, as well as wai taonga (through site selection). The CHI has been designed in such a way that it can only be applied by Maori, because the calculation of CHI scores needs to be informed by knowledge held only by Maori. Thus, it represents a means of enabling Maori and resource managers to use traditional data to inform decision-making.

Comparison of the CSHM with Western Measures

Only the third component of the CHI, namely the Cultural Stream Health Measure, has “western” equivalents. To place the cultural index in a broader context, we compared it with two western measures of stream health that are commonly applied in New Zealand. The CSHM was significantly correlated with both the MCI and SHMAK indexes. This indicates that the cultural measure, like its western counterparts, successfully captures aspects of stream health.

The strong relationship between CSHM and MCI is particularly notable because the latter depends on a reductionist approach based only on stream invertebrate community composition. The CSHM, on the other hand, has no invertebrate component but assesses stream health on the basis of a holistic combination of catchment, river margin, and instream characteristics. Streams judged by Maori to be in poor health turned out to possess a set of invertebrate species that are tolerant of poor water quality. In this sense, the cultural measure captures information about stream invertebrates without studying them.

The CSHM was less strongly correlated to the invertebrate component of SHMAK than to the habitat component, which is not surprising given the strongly habitat-based character of the CSHM. Overall, the poorer relationship with SHMAK than with MCI may reflect the fact that SHMAK was not developed as a survey tool, but primarily to compare stream health within a site over time or to compare local streams. The relationships between CSHM and the western measures were poorer for the larger tributary and mainstem sites. This partly reflects past criticism that the MCI performs less well in low-gradient streams with silty, muddy, or weed-covered beds (Boothroyd and Stark, 2000), a description that applies to a few of our larger sites. Similarly, SHMAK is intended to be applied only in streams that are safely wadeable at normal flows (Biggs et al., 1998); again a few of our sites were near the prescribed limits.

Stream health measures are often framed in the context of pressure (human actions in the catchment) and state (resulting biotic and abiotic conditions in the stream) (Fairweather, 1999). Indicators of state (such as the MCI) directly gauge the water quality response to human activity around and in the stream. SHMAK, on the other hand, combines indicators of both pressure (e.g., bank vegetation) and state (e.g., water clarity, invertebrate community composition). The CSHM also combines pressure (catchment, riparian, and bank condition) and state indicators (instream physical and chemical measures), although this was not a consideration when designing the CSHM and the CHI, as the priority was to ensure that these measures reflected the Maori holistic philosophy by placing particular emphasis on the catchment as a whole.

Research in New Zealand (Wilcock, 1986; Rutherford et al., 1987; Cooper and Thomsen, 1988; Quinn et al., 1997; Townsend et al., 1997; Wilcock et al., 1999; Niyogi et al., 2003) and elsewhere (e.g., Carpenter et al., 1998; Hunter et al., 2000) has established that land-use development in stream catchments, such as conversion from native vegetation to agriculture, can influence stream health by changing water chemistry, turbidity, temperature, and the physical nature of the stream bed and banks, with consequences for stream life. Thus, it is of interest to consider the relationship between the Cultural Stream Health Measure and land use in the stream’s catchment, and to compare performance of the cultural and “western” measures in this respect. There were negative relationships between the health measures and percentage of developed land at all three scales assessed (entire catchment, entire riparian zone, local riparian zone). The fact that MCI was most strongly related to land development at the larger scales indicates that invertebrate communities are more strongly influenced by land development in the catchment as a whole. On the other hand, the Cultural Health Index was more strongly related to land development in the riparian zone extending just 500 m upstream. This is not surprising given that the Maori assessments of catchment and riparian condition were performed visually from the stream site itself. We conclude that the CSHM performs as well as MCI and better than SHMAK in encapsulating the relationship between land development and stream health.

CHI As a Diagnostic and Monitoring Tool

The multifactorial nature of the CHI makes it particularly appropriate as a diagnostic and monitoring tool. Thus, a stream site may score poorly because of particular kinds of degradation at particular spatial scales, or because it fails to support mahinga kai species. Even sites in good health may have poor access to mahinga ka and score poorly as a result. Having diagnosed the problems associated with a particular location, management actions can be designed, and it will be obvious if these are effective (because legal or physical access is reinstated, or valued species can again be harvested sustainably, or the condition of the riparian vegetation or riverbank is improved). Moreover, sites with a particularly high priority for restoration management can be identified because of their traditional status or because valuedmahinga ka species are no longer present.

Incorporating the CHI into Resource Management Processes

We envisage the following steps to implement the CHI scores in a meaningful way. First, Maori convention required that the full runanga approve the use of the CHI and confirm the composition of the runanga teams that applied the tool on the Taieri and Kakaunui Rivers. Then the agency that is responsible for water management, the Otago Regional Council in this case, is informed about the CHI. The next step, for each stream site, should involve runanga members interpreting the scores applied to the three CHI components and identifying priority issues. A fourth step, to safeguard traditional knowledge, involves the runanga identifying sensitive information that is to be appropriately protected. Then, runanga members should work with the Otago Regional Council to determine ways of addressing priority issues in each catchment, to set objectives and initiate remedial action. By these means, the relationship between Maori and the agency in charge of resource management would move beyond mere consultation (Taiepa et al., 1997) to an effective expression of Maori participation in management and decision-making.

Extending the CHI Development Process to Other Indigenous People

It is not difficult to envisage extending the process undertaken by Ngai Tahu in two particular rivers to other iwi in New Zealand, or indeed to indigenous people in other parts of the world. Having established a traditional view of ecosystem health and a set of possible indicators, the process of evaluation at a range of local stream sites (or other ecosystems) can take place, followed by a systematic approach to condensing a large set of indicators into a smaller and more manageable set that encapsulates the essence of cultural ecosystem health. Different indigenous peoples are not likely to have precisely the same perspective or come up with the same set of indicators, but the underlying process can be essentially the same. Thus, a different situation may produce fewer or more components than the three adopted in the CHI, and each of these may be encapsulated in quite different indicators.

Collaborative resource management refers to situations where relevant stakeholders are involved in a substantial way with the agency having responsibility for resource management (Borrini-Feyerabend, 1996). The move from top-down to participatory approaches is especially significant for indigenous people, because it acknowledges the importance of understanding their needs, perspectives, and knowledge (IIED, 1994; IUCN, 1997).


Many Maori associate their well-being, both as individuals and as members of whanau, hapu, and iwi, with maintaining the health of the natural environment. One expression of their philosophy states ”If you do not sustain the waterways, the mahinga kai sourced from them, and sites of significance in the wider environment, then you cannot sustain yourself, honour your ancestors, or provide for the children of your children into the future“ (Tipa et al., 2002). Sustainability and the long-term well-being of Maori are very closely linked. The development of the Cultural Health Index is an important step towards enabling Maori to fulfill their obligations as kaitiak to sustainably manage New Zealand’s freshwater resources.


We acknowledge the important contribution of members of Te Runanga Otakou, namely Bill Loper, Terry Broad, Rose Clucas, and Andrea Todd, and of Te Runanga o Moeraki, namely Isbel Williams, Ray Williams, Rua McCallum, Huia McGlinchey, and Kyle Nelson. For oversight of the project and administrative assistance, we sincerely thank Te Runanga o Ngai Tahu and Linda Constable. For supplying GIS information, preparing Figure 1, and providing other technical assistance, we thank Chris Arbuckle. Funding for this work was supplied by New Zealand’s Ministry for the Environment and Foundation for Research, Science and Technology.

Copyright information

© EcoHealth Journal Consortium 2004