Water Resources Management

, Volume 27, Issue 7, pp 2013–2027 | Cite as

Development and Application of the River Values Assessment System for Ranking New Zealand River Values

Article

Abstract

In New Zealand and elsewhere no system has existed for objectively ranking the relative importance of different use (e.g., irrigation and hydro electric power) and non-use (e.g., whitewater kayaking, recreational angling, native birdlife) river values. Development of such a system would provide an opportunity for improved policies and rules around water and river use, development and conservation opportunities, and for understanding tradeoffs when competing and overlapping demands are placed on the same resource. In this paper the River Values Assessment System (RiVAS), a Multi Criteria Analysis based approach, is described and demonstrated by application to the salmonid angling value (and in a more limited way to swimming) in Tasman District rivers of the South Island, New Zealand. The system has 10 steps, and a decision support system which finally helps decide the national, regional or local (or high, medium or low) importance or significance of rivers for particular values. As with any MCA approach there is a wide range of limitations all of which are addressed, and none of which are ultimately fatally detrimental to the system.

Keywords

Multi criteria analysis River values Relative importance rankings New Zealand 

1 Introduction

In New Zealand but also worldwide, river managers are faced frequently with the challenge of deciding priorities for management between competing river values, both instream (e.g., recreational such as angling and kayaking, and environmental such as native birdlife and natural character, and cultural—indigenous peoples’ values) and out-of-stream (e.g., irrigation and hydro power). Judges and hearing panels at many levels of local and central government, the New Zealand Environment Court (or its equivalent in other jurisdictions), and local and central government generally, have for decades been seeking an objective method for ranking the comparative value of rivers for the range of in– and out–of–stream uses, to assist with decision making. No such tool has been available, despite global efforts at implementing Integrated Water Resources Management (see UN-Water 2008 for example).

The large diversity of values to be considered in any river ranking exercise meant Multi Criteria Analysis (MCA) was the obvious tool for development. There has been increasing global interest in the application of MCA approaches to the planning and management of water resources. Hajkowicz and Collins (2007) reviewed 113 MCA applications from 34 countries and drew the following conclusion (p1564): “Water management is typically a multi-objective problem which makes MCA a well-suited decision support tool. The outcomes are often intangible and are measured in a variety of units. MCA has been found to assist with conflict resolution, stakeholder participation and community engagement. It has also been shown to improve the auditability, transparency and analytic rigour of water management decisions”. The list of MCA benefits identified by Hajkowicz and Collins (2007) are the same criteria river managers are looking for in river ranking exercises. And, while internationally, MCA approaches have more recently been used by Hermans et al. (2007) as the basis for collaborative planning in a single catchment in North America, by Mladenović-Ranisavljević et al. (2012) for ranking Danube river water quality, and by Balica et al. (2012) to develop a flood vulnerability index for coastal cities, amongst a range of applications, no one appears to have used the method to develop a standardised tool applicable across the range of river values for ranking river importance.

While globally there has been no standardised ranking method established, historically in New Zealand, Teirney et al. (1982) and Teirney and Richardson (1992) for recreational trout and salmon fisheries, and Egarr and Egarr (1981) for whitewater kayaking, identified lists of rivers and streams for their relative importance for these values. More recently, the relative importance issue was addressed under the Water Programme of Action, part of the Labour Government’s 2003 Sustainable Development Programme of Action. The programme identified the need for the Department of Conservation (DoC) to identify water bodies of national importance (WONI) and a list of water bodies that would protect the full range of freshwater biodiversity values. In a complementary way MfE (2004) listed water bodies important for recreation, and MfE and MAF (2004) produced lists of waters of national importance for: the biodiversity dimension of natural heritage; geodiversity and geothermal features; recreation; irrigation; energy; industry and domestic; and tourism. But, despite much work in this context none of these lists have been produced using a methodology which enables ready and objective comparison between the internal priorities derived. There thus remained no objective framework that clearly identified the criteria upon which importance is determined for specific values, or which allows for comparison between values either at a national or a local scale.

It is within the context outlined above that this paper describes and analyses development of the River Values Assessment System (RiVAS). The paper first summarises the legislative and related policy needs for river value priorities in New Zealand. It then describes the process employed to develop RiVAS. Next it shows by example how the system has been applied. Finally, the discussion and conclusions consider issues that have arisen during the system development, the potential further applicability of the system, key ingredients for an integrated approach and possible future developments.

2 The Legislative and Policy Framework

Within New Zealand’s leading environmental legislation, the Resource Management Act 1991 (RMA), there are multiple contexts for consideration of relative importance, e.g.,
  1. 1)

    National Policy Statements: draft National Policy Statement on Freshwater (e.g., ‘Identify notable values of outstanding freshwater resources’).

     
  2. 2)

    Regulations: National Environmental Standard – Ecological flows draft (the technical process involves an assessment of the relative significance of aquatic values). The existing National Environmental Standard for Plantation Forestry requires identification of nationally significant rivers.

     
  3. 3)

    Orders: Water Conservation Orders (outstanding amenity or intrinsic values, habitat, fishery, wild, scenic or other natural characteristics, scientific or ecological value, recreational value).

     
  4. 4)
    Plans: Regional and District. Typical of the outputs produced for such plans are:
    1. i.

      Schedules in regional plans of water bodies to be managed for specified purposes

       
    2. ii.

      Schedules in regional plans that list values of water bodies

       
    3. iii.

      Water conservation orders that list rivers with outstanding value for specified uses and values.

       
     
Most often the requirements in the list outlined above are water quality and discharge management related, the language used is very variable, and there is little or no connections between regional and national value. Even in the MfE and MAF (2004) ‘Potential Water Bodies of National Importance’ report, which provides lists of rivers across multiple values, there is no unifying methodology and no clear thresholds for defining the different levels of importance. Perhaps the most telling statement from that report is:

“The Water Bodies of National Importance Working Group was tasked with the completion of seven sub-projects. Each would develop a method for identifying water bodies of national importance for that value and a potential list. Most of the projects have developed methods and identified initial lists of water bodies or catchment areas that are of potential national importance” (MfE and MAF 2004: 6).

Unfortunately each of these methods was different and no unifying and overarching methodology was developed. Not surprisingly there is no means for standardising these lists or comparing any one list with any others. Furthermore, little original data were collected and even when it was its application was seldom criteria driven. Given these and many other related limitations there was a clear need to develop a new approach. In the next section the process for developing such an approach is outlined.

3 Research Approach

The research was funded by means of several Envirolink grants (central government research funding applied for by designated regional councils with research provided by a university or other approved provider) sought by Tasman District Council (a unitary (including regional functions) authority located in the north of the South Island). The sequence and relative sizes of the grants was important: the first, a small advice project, enabled holding a small planning workshop to scope the opportunity and associated research and implementation requirements; the second, a medium advice project, enabled a national workshop to be held, definition of values to be worked on, and agreement to begin work on a method and its application to salmonid angling values in Tasman; the third, a tools project, enabled the project in full to be undertaken with the method involving trials on multiple values around selected case study regions of New Zealand; then finally, further medium advice projects which enabled a full roll out of the method to multiple values in Tasman.

Aside from the research funding it was considered vital to integrate key stakeholders into all stages of the research process. To this end a steering group comprising the Tasman District Council champion, Fish and Game Nelson-Marlborough (a demonstrated long-term interest in this issue), Cawthron Institute freshwater fisheries scientist, Wellington Regional Council policy manager (an end user), and the Lincoln University research provider. This group held the small workshop in Nelson where it reviewed existing prioritisation approaches and brainstormed a new approach that could be applied to all values. It soon became clear that Multi Criteria Analysis (MCA) was a tool that could be applied in a problem space which had multiple values and potentially hundreds of ranking criteria to choose from.

Multi Criteria Analysis has existed in a formal sense since the 1970s and is now widely used as a decision support tool in a wide range of contexts. A simple description of MCA is that it is a tool which enables choices to be made between multiple options using multiple criteria to help organise and prioritise diverse subjective and objective data. There are dozens of different forms of MCA (see Hajkowicz and Collins 2007) but perhaps the most widely used, the Weighted Sum Model, was used herein. Its choice was based on its widespread use and its reliance on unitary scales of measurement. However, as with any methodology, it has limitations—these are identified and addressed in the discussion.

Having identified MCA as the preferred approach it was decided to iteratively develop a sample application based on the Tasman salmonid fishery (primarily brown trout, Salmo trutta). Fish and Game New Zealand (the Quasi Autonomous Non Government Organisation responsible for managing mostly introduced fish and game in NZ) have regular approximately 6–7 yearly national surveys of anglers (The National Angler Survey), undertaken under contract by the National Institute of Water and Atmospheric Research (NIWA) (e.g., Unwin 2009). The survey data set is enormous, containing much quantitative information about levels of use by river, and about the relative importance of different rivers and lakes across a range of criteria. No other data set for any other value is considered to be as comprehensive as that existing for angling. The following section outlines the method used and its application to salmonid angling (which then for the other extreme, i.e., data lacking, is contrasted with swimming).

4 The River Values Assessment System (RiVAS)

Unlike salmonid angling most other values have limited (native birds, whitewater kayaking) or sometimes nonexistent (e.g., swimming, natural character) databases. Because of the enormity of this spectrum it was decided to use a ‘best available information’ approach incorporating both objective and subjective data. This approach then necessitated value management by means of an expert panel approach with the selection of each panel consistent with ‘best practice’ principles, e.g., credibility, i.e., known and respected ‘experts’ in the value—such experts would include value practitioners (e.g., irrigated farmers for irrigation, kayakers for whitewater kayaking), relevant scientists/consultants (e.g., a bird ecologist for native birds, a recreation specialist for river swimming, a hydrologist for irrigation), and appropriate policy makers e.g., planner from a regional or district council with an understanding of the value, policy advisor from key stakeholder organisations (e.g., field officer from Fish and Game); an appreciation of the value from a national perspective; a demonstrated record of working within the collaborative approach of an expert panel context; and an understanding of multi criteria approaches. The initial expert panel and subsequent panels for each new value then required a skilled facilitator who was familiar with the relevant value and the method. The expert panel for salmonid angling in Tasman therefore comprised the TDC freshwater policy expert, the Fish and Game Nelson-Marlborough manager, two fisheries scientists (one from Cawthron Institute and the other the research leader for the National Angler Survey from NIWA), a recreational planning consultant as the facilitator (and for this value the overall project leader); by way of contrast the swimming panel comprised the same facilitator and three regional council staff (including planning and water quality staff, with expert local knowledge of swimming sites).

The expert panel was then tasked with applying the agreed approach which essentially comprised of two parts (the first for a national panel; the second for both the initial national panel and then for subsequent regional level applications) and 10 steps, as follows.

Part 1—comprised of steps 1–4, and involved clarifying and prioritising the characteristics that typify or make up the value.
  1. 1)
    Definition of the river value and river segments:
    1. a.

      Recreational fishing is a very broad term and in New Zealand encompasses angling for introduced salmonids and fishing for native species including eel (Anguilla spp.) and whitebait (Galaxiid spp.)—consequently, it was decided to deal with salmonid angling as a separate and easily identifiable value (by contrast swimming is simply defined as river swimming);

       
    2. b.

      Different values have different connections with parts of or the entire river system—for salmonid angling it is mostly with an entire river; for swimming it is with identifiable ‘swimming holes’ or locations or segments which need to be separately identified but then considered representative of a river in terms of its relative value for swimming.

       
     
  2. 2)
    Identification of attributes:
    1. a.

      Each value is comprised of a very wide range of attributes (typically present in clusters of like attributes) ranging from the biophysical, through social, economic and cultural (in the case of indigenous peoples’ values);

       
    2. b.

      A full list of attributes needed to be documented as part of the explicit expert panel approach—33 attributes of salmonid angling, and 34 attributes of swimming. Table 1 demonstrates, using examples from salmonid angling and swimming, how this step and steps (3) and (4) below were applied.

       
    Table 1

    Examples from salmonid angling and swimming showing the progression from value, to attributes of the value, to primary attributes and to indicator measures

    Step 1: value

    Example attribute cluster

    Step 2: example attributes (from relevant cluster)

    Step 3: selected primary attributes from example set

    Step 4: relevant indicator

    Salmonid angling

    Users

    Level of use

    Level of use

    Number of angler days p.a.

    Intensity of use

    Intensity of use

    Mean free reach (MFR) = average distance (in km) an angler would have to travel on an average day before encountering another angler

    Level of commercial use

      

    Origin of New Zealand users

    Origin of New Zealand users

    Mean number of km travelled from home by NZ anglers

    Level of international use

    Level of international use

    Percent overseas anglers (of total number angler days)

    User demographics

    Behaviour of users

    Swimming

    Social

    Level of use

    Level of use

    Number of swimmers on a peak use day

    Travel distance

    Travel distance

    Number of kms travelled by swimmers from previous night’s location

    Perception of safety

    Other users and uses

    Diversity of recreation opportunities

    Physical river features

    Swimming holes

    Swimming holes

    Maximum water depth

    Variable water depth

    Variable water depth

    Morphological variability

    Width of river

    Flow

    Hard/soft river bed bottom

    Natural jump-off features (e.g. large rock)

    Beach

    Pools

    Pool/riffle/run sequences

    Rapids

     
  3. 3)
    Selection of primary attributes:
    1. a.

      For a range of manageability reasons it was decided to select a subset of primary (most important/representative) attributes, with this subset limited to 5–10 such attributes;

       
    2. b.

      For salmonid angling the 10 primary attributes were: Angler days; Intensity of use (mean for each river reach); Travel distance; Overseas anglers; Perception of catch rate; Perception of fish size; Water quality; Perception of scenic attractiveness; Perception wilderness; and Perception of importance.

       
    3. c.

      For swimming the eight primary attributes were: Number of swimmers on a peak use day; Number of kms travelled by swimmers from previous night’s location; Presence of facilities (toilets; camp facilities—designated camping sites, ablution block, signage, etc.); Perception of scenic attractiveness; Maximum water depth; Morphological variability; Compliance with periphyton and cyanobacteria guidelines; and Compliance with horizontal visibility guidelines.

       
     
  4. 4)
    Identification of indicators for each of the primary attributes:
    1. a.

      Indicators act as measures of the primary attribute and for attribution and weighting reasons only one indicator could be chosen for each primary attribute;

       
    2. b.

      All potential indicators were evaluated against SMARTA (i.e., Specific, Measurable, Achievable, Relevant, Timely, Already in use) criteria before being finalised.

       
     
Part 2 of the process involved six further steps, the first two of which are demonstrated in Table 2.
  1. 5)
    Determination of indicator thresholds:
    1. a.

      The system was designed to initially have three levels of relative significance only, i.e., high, medium or low—upon reflection it was realised there were situations where some primary attributes had less than low, i.e., no significance, thus in these cases a ‘no’ significance value was added.

       
    2. b.

      The determination of thresholds in many cases is judgmental, thus being defined by the expert panel. But for some values, e.g., native birdlife, some primary attributes already have defined thresholds of relative significance, e.g., the presence of endangered species is deemed of high significance for native birdlife.

       
    Table 2

    Examples from salmonid angling and swimming showing indicator thresholds of relative significance levels applied to the Motueka River, Tasman District

    Value

    Example primary attributes

    Indicators for example primary attributes

    Step 5: indicator relative significance thresholds (scale: mostly 1–3; sometimes also 0)

    Step 6a: raw indicator data from Motueka River; data source (and reliability)

    Step 6b: Motueka River converted significance score

    Salmonid angling

    Level of use

    Number of angler days p.a.

    High: >5,000 angler days p.a. (score: 3); Medium: 1,000–5,000 angler days p.a. (score: 2); Low: <1,000 angler days p.a. (score: 1)

    3351 angler days

    2

    Data: national angler survey (high)

    Intensity of use

    Mean free reach (MFR) = average distance (km) an angler would have to travel on an average day before encountering another angler

    High: MFR <5 km (score: 3)

    4.8 km

    3

    Medium: MFR 5–20 km (score: 2)

    Data: national angler survey (high)

    Low: MFR >20 km (score: 1)

    Origin of New Zealand users

    Mean number of km travelled from home by NZ anglers

    High: >100 km (score: 3); Medium: 50–100 km (score: 2); Low: <50 km (score: 1)

    23 km

    1

    Data: national angler survey (high)

    Level of international use

    Percent overseas anglers (of total number angler days)

    High: >20 % overseas angler visits (score: 3); Medium: 10–20 % overseas angler visits (score: 2); Low: <10 % overseas angler visits (score: 1); none: no use by overseas anglers (score: 0)

    8 %

    1

    Data: national angler survey (high)

    Swimming

    Level of use

    Number of swimmers on a peak use day (3 = High numbers; 2 = Medium; 1 = Low)

    High (score: 3);

    2

    2

    Medium (score: 2);

    Data: expert panel (medium)

    Low (score: 1)

    Travel distance

    Number of kms travelled from home by swimmers (3= >20 km; 2 = 10–20 km; 1 = <10 km)

    High: >20 km (score: 3); Medium: 10–20 km (score: 2); Low: <20 km (score: 1)

    2

    2

    Data: expert panel (medium)

    Swimming holes

    Maximum water depth (m)(3= >3 m; 2 = 2–3 m; 1 = <2 m)

    High: >3 m (score: 3); Medium: 2–3 m (score: 2); Low: <2 m (score: 1)

    2

    2

    Data: expert panel (medium)

    Variable water depth

    Morphological variability (3 = High variability; 2 = Medium; 1 = Low)

    High (score: 3);

    2

    2

    Medium (score: 2);

    Data: expert panel (medium)

    Low (score: 1)

     
  2. 6)
    Population of indicators with raw data (step 6a) and application of their thresholds (step 6b):
    1. a.

      As noted the raw data can be in the form of measurable scientific biophysical, economic or social data, or could be the subjective agreed judgments of the expert panel.

       
    2. b.

      The conversion of raw data to the predominantly 3-point numeric scale matches the requirement of the Weighted Sum Model MCA method to have numeric scales of measurement.

       
     
Steps 7 and 8 are illustrated using the salmonid angling example (Table 3).
  1. 7)
    Weighting of the primary attributes:
    1. a.

      The weighting is considered a measure of the relative contribution of each primary attribute to the river value—equal value may or may not be given across the primary attribute set.

       
    2. b.

      This step provides an opportunity to test the sensitivity of the system in producing final rankings or relative importance and in all applications where unequal weightings were applied they needed to be properly evaluated.

       
     
  2. 8)
    Determination of river significance for the value:
    1. a.

      The threshold scores are summed according to the weightings applied to each indicator, and the rivers ranked from highest to lowest score (as shown in Table 3 for three different weightings of the perceptions of importance score).

       
    2. b.
      Using the expert panel and defined decision support criteria the rivers are identified as nationally (or highly significant), regionally (of medium significance) or locally (or low significance) importance. From the ranked list from Step 8, the Expert Panel closely examined the rivers, and their primary attribute scores. It was noted that a strong correlation existed between angling and rivers which scored a 3 (high) for the indicator % overseas anglers. Intuitively this made sense—international anglers were likely to target ‘the best’ rivers in New Zealand. Therefore this attribute was chosen as a surrogate attribute. No obvious national trigger attribute presented itself. The following criteria were applied:
      1. i.
        National significance:
        1. 1)

          Criterion 1: % overseas anglers = 3, plus 25 % or more of the other attributes = 3; or

           
        2. 2)

          Criterion 2: 50 % or more of the attributes = 3.

           
         
      2. ii.
        Regional significance:
        1. 1)

          Those rivers in the table not defined as nationally or locally significant.

           
         
      3. iii.
        Local significance:
        1. 1)

          Sole criterion: % overseas anglers < 3, plus maximum of one other attribute = 3.

           
         
       
    Table 3

    Significance assessment calculations for salmonid angling in Tasman District—selected rivers only (the grey shaded column was that finally chosen for ranking purposes)

     
  3. 9)
    Outline other factors that might be relevant to the significance assessment:
    1. a.

      For some values there will be factors that are difficult to take into account in a standardised approach such as that proposed here, e.g., for native birds there are potential criteria and applications related to wetlands of international importance, yet rankings only go to national.

       
     
  4. 10)
    Review the method as applied to the value
    1. a.

      Many assessments dealt with entirely inadequate data sets. In these instances it was important to note this issue and recommend data acquisition priorities for the future.

       
     

5 Discussion

At first glance RiVAS appears simple, robust and to produce useful outcomes and outputs. Amongst its many advantages are: clear criteria for selection of an expert panel; clear criteria for selection of peer reviewers; a thoroughly open and transparent process; explicit elicitation of all attributes, the identification of indicators and the recording of raw data and how it is converted to a numeric scale. Ultimately a defensible set of priorities is elicited which should be of use to water resource managers and users generally. This set of advantages essentially mirrors the benefits of using MCA in water resources management identified by Hajkowicz and Collins (2007, p1574) thus lending further credibility to the approach adopted here. Somewhat surprisingly however, and despite the proliferation globally of Integrated Water Resources Management approaches (e.g., UN-Water 2008), there seems to have been few other efforts at developing ranking systems that traverse the range of water resource values. RiVAS may then provide a tool to help fill this gap, especially in developed countries. And, as shown by Hughey and Booth (2012) it may also assist in identifying an integrated set of water resource indicators that also traverse the range of river values. Some of these indicators and the related importance rankings of rivers have informed council priorities for monitoring river swimming locations for example (see Hughey 2012). River rankings elicited by RiVAS have been used also in establishing regional level policies for rivers and for a range of other purposes (Hughey 2012). Despite these positive features of RiVAS there are potential limitations and more generic issues with RiVAS—these matters are discussed next.

5.1 Limitations

There is no perfect decision support system and all systems have their limitations. Some of the key limitations identified through the course of developing RiVAS and the approaches used to try and minimise their potentially negative influence follow (and are detailed more fully in Hughey et al. 2010a, b).

5.1.1 Expert Panels

Using expert panels and the need for subjective decision-making by them is challenging. The method includes criteria to guide the appointment of panel members (see Hughey et al. 2010a, b), e.g., credibility of each panel member, an appreciation of the value from a national perspective, a demonstrated record of working within the collaborative approach of an expert panel context, and an understanding of how multi criteria approaches function. Hughey et al. (2010a, b) suggested that a national body (perhaps a central government agency) could ‘take up the reins’ and apply the RiVAS in a coordinated manner, thus reducing the potential for expert panel bias. The ‘upside’ from using national and regional level expert panels is the involvement of external stakeholders in the process, something highly valued by the councils which have participated in RiVAS applications (Hughey 2012).

5.1.2 Auto-Correlation

It seems likely, despite the best attempts to reduce this, that relationships between some primary attributes, known technically as auto-correlation will occur. The smaller the list of primary attributes, the less likely this is to occur, but when it does occur, results may be influenced. RiVAS requires 6–10 primary attributes be identified to encompass the various aspects of each river value. The balance between providing an adequate number/diversity of attributes and minimising their auto-correlation is challenging, and some auto-correlation is almost unavoidable. The method separates attributes as far as possible and weighting attributes can be used to explicitly address attributes with, or suspected to have such relationships.

5.1.3 Weighting Attributes

Attributes can be weighted in the RiVAS methodology (i.e., adjusted to recognise their greater ‘contribution’ to explaining the relative importance of the river value). The default in the method is to apply equal weighting to attributes but this may not be correct. For most values there is little data about the relative importance of the attributes and without empirical data, this problem cannot easily be resolved. However, the method does consider and allow for attributes to be weighted, and a sensitivity analysis procedure is recommended.

5.1.4 Thresholds

For some values (e.g., native birdlife and to an extent native fish), criteria already exist to clarify national importance (high significance), and these have been applied where appropriate. Examples of such criteria include definitions of threatened and endangered species and thresholds of nationally important populations. For other values, including swimming, natural character and abstractive uses (such as irrigation), there are no national-level significance criteria so the threshold tests are not so clear. For these values, relevant RMA interpretations have been used, e.g., water bodies defined as outstanding in water conservation orders (WCO) for particular values are accorded nationally important (high significance) status. As there is no consistency in the criteria used between each WCO deliberation the selected thresholds need to be tested and, where necessary, amended as the method is applied within and between councils.

5.1.5 Connectivity Between Rivers

The method involves developing river specific rankings. In some circumstances, a series of rivers in relatively close proximity are attractive because of their proximity, e.g., for salmonid angling or whitewater kayaking. This factor may not be properly included within the methodology. The opportunity exists for additional notes to highlight such situations.

5.1.6 Comparative Grades

In developing the method, ‘raw’ indicator data has been converted to comparative (normally) 1–3 (low to high relative significance) scores which are then aggregated to give a total relative significance or importance score. An alternative system of 1–5 scores could also be used and has been used in limited situations. The 1–3 scoring, however, does adequately differentiate across the range of attributes in most cases. It provides a less complicated approach that also reflects the three grade system in the ultimate ranking (i.e., national, regional and local; or high, medium and low).

5.1.7 Mathematical Issues

MCA type analyses assume that all the values lie in what is effectively our normal mathematical world—that all values lie in a comparable and (effectively) linear ‘space’. This may not always be true—values may lie in a logarithmic or other non-linear spacing, there may be gaps or big jumps between different states of a value, or the differences between states may not even be comparable in an ordinal manner. There is also the ‘apples and oranges’ problem when comparing two different values, in that they may not be comparable within our understanding or interpretation of the world, despite having been scored on a similar numerical scale. Mathematical manipulation of values makes further assumptions about the nature and ordinality of the values, and their comparability. However, because the same approach is applied to all values it is expected the same weakness, or issue, if present will be treated the same way in all circumstances thus reducing the potential negative impact of this issue, across the board.

5.2 Other Issues

To date application of RiVAS has been driven largely by three unitary councils, and while central government support has been provided in kind there has been no explicit policy development endorsing the approach and its further application. Such endorsement would be beneficial if the method is to develop and have credibility. Currently this lack of central government support is particularly problematic around the question of how to treat the final lists of rankings undertaken on a region by region basis. Are these rankings lists of national, regional or local importance for a value or are they of high, medium or low significance. Some argue the former cannot be applied until full national-level coverage for each value is undertaken, while others argue there are legislative (and in places policy) criteria and guidance that can provide the context for such definitions.

For two abstractive values, namely irrigation and hydro electric power the maximum potential future use was evaluated. This approach was taken to enable a degree of like with like comparisons to occur, i.e., most native birdlife rivers (at least for braided rivers) have retained their relatively natural/relative importance status that has existed for many years. The full ‘worth’ of any river for any value is thus in evaluating its full potential—to this end most whitewater kayaking, many (but certainly not all, depending on location) rivers important for native birds, salmonid angling and the like are still exhibiting those values’ full potential. Consequently it was appropriate to consider also abstractive use values from a full potential level. This approach works in many circumstances but not in all, especially where the instream value is now substantially degraded on its full potential, e.g., where a river has been fully or substantially diverted for irrigation and its salmonid angling and native birdlife values are significantly reduced, or because of water pollution its potential worth for swimming is degraded. RiVAS+ is a complementary tool that explores and measures this potential (see Hughey et al. 2011).

6 Conclusions

Development of the River Values Assessment System (RiVAS) using Multi Criteria Analysis is a long-awaited initiative now receiving a great deal of interest amongst councils, central government agencies, environmental and recreational interests and from developers in New Zealand. The system is easily understood, easily applied and appears defensible, but has its limitations and ongoing areas for further development. Despite these issues RiVAS appears to hold much potential and there appears no reason it cannot be applied to many more values in New Zealand and also be applied as an approach to enhancing our understanding of river values and potential tradeoffs between these values internationally.

Notes

Acknowledgments

This work was funded by the Foundation for Research, Science and Technology as part of the project ‘Developing a significance classification framework for water body uses and values’—funding was by means of several grants and I thank the Foundation for support via:

1. Small Advice: 532-TSDC40

2. Medium Advice: 612-TSDC41– Developing a significance classification framework for water body uses and values: Project scoping workshop; 894-TSDC69—Significance assessment of river uses and values in Tasman; 898-TSDC70—Significance assessment of hydro-electric power generation values of NZ rivers

3. Tools: LINX0810—Uses and values of water bodies.

Many council staff assisted with all of the trial applications and to them I extend a sincere thanks. Likewise I thank the members of non government organisations who helped improve the method and specific applications to specific values.

References

  1. Balica SF, Wright NG, van der Meulen (2012) A flood vulnerability index for coastal cities and its use in assessing climate change impacts. Nat Hazards 64:73–105CrossRefGoogle Scholar
  2. Egarr GD, Egarr JH (1981) New Zealand recreational river survey: part 1: methods and conclusions. Water and soil miscellaneous publications no. 13. Ministry of Works and Development, WellingtonGoogle Scholar
  3. Hajkowicz S, Collins K (2007) A review of multiple criteria analysis for water resource planning and management. Water Resour Manag 21:1553–1566CrossRefGoogle Scholar
  4. Hermans C, Erickson J, Noordewier T, Sheldon A, Kline M (2007) Collaborative environmental planning in river management: an application of multicriteria decision analysis in the White River Watershed in Vermont. J Env Manag 84:534–546CrossRefGoogle Scholar
  5. Hughey KFD (2012) RiVAS (and RiVAS+): insights and lessons from 5 years’ experience with the River Values Assessment System. Proceedings of the New Zealand Agricultural and Resource Economics Society Conference, 30–31 August, Nelson, http://ageconsearch.umn.edu/bitstream/136047/2/Huhgey%202012%20complete.pdf. Accessed 23 November 2012
  6. Hughey KFD, Booth KL (2012) Monitoring the state of New Zealand rivers: how the river values assessment system can help. NZ J Mar Freshw Res 46:545–556Google Scholar
  7. Hughey KFD, Booth K, Baker M-A (2010a) River Values Assessment System (RiVAS)—the method. In: the River Values Assessment System: volume 1: overview of the method, guidelines for use and application to recreation values. Hughey KFD, Baker M-A (eds) Land environment and people report no. 24. Lincoln University, New Zealand, pp 47–67Google Scholar
  8. Hughey KFD, Booth K, Harris S, Baker M-A (2010b) A guide to using the River Values Assessment System (RiVAS). In: the River Values Assessment System: volume 1: overview of the method, guidelines for use and application to recreation values. Hughey KFD, Baker M-A (eds) Land environment and people report no. 24. Lincoln University, New Zealand, pp 69–72Google Scholar
  9. Hughey KFD, Sinner J, Booth K (2011) RiVAS and RiVAS+: opportunities for application of a multi criteria river value assessment system approach which considers existing and potential states. Proceedings of the New Zealand Agricultural and Resource Economics Society Conference, 25–26 August, Nelson, http://ageconsearch.umn.edu/bitstream/115406/2/Hughey%20Paper.pdf. Accessed 23 November 2012
  10. MfE and MAF (2004) Water programme of action: potential water bodies of national importance. Technical working paper. Ministry for the Environment, WellingtonGoogle Scholar
  11. Mladenović-Ranisavljević I, Takić L, Vuković M, Nikolić D, Živković N, Milosavljević P (2012) Multi-criteria ranking of the Danube water quality on its course through Serbia. Serbian J Manag 7:299–307CrossRefGoogle Scholar
  12. Teirney LD, Richardson J (1992) Attributes that characterize angling rivers of importance in New Zealand, based on angler use and perceptions. N Am J Fish Manag 12:693–702CrossRefGoogle Scholar
  13. Teirney LD, Unwin MJ, Rowe DK, McDowall RM, Graynoth E (1982) Submission on the draft inventory of wild and scenic rivers of national importance. Fisheries environmental report No 28. Ministry of Agriculture and Fisheries, ChristchurchGoogle Scholar
  14. UN-Water (2008) Status report on integrated water resources management and water efficiency plans. Prepared for the 16th session of the commission on sustainable development. United Nations, New YorkGoogle Scholar
  15. Unwin M (2009) Angler usage of lake and river fisheries managed by Fish & Game New Zealand: results from the 2007/08 National Angling Survey. Prepared for Fish & Game New Zealand, NIWA client report: CHC2009-046. National Institute of Water & Atmospheric Research Ltd, ChristchurchGoogle Scholar

Copyright information

© Springer Science+Business Media Dordrecht 2013

Authors and Affiliations

  1. 1.Department of Environmental ManagementLincoln UniversityLincolnNew Zealand

Personalised recommendations