List of all metadata reports | This report's TOC | Previous Page | Next Page

2.0 Classification Systems and Spatial Frameworks

2.1 What is a classification system?

Most classification systems provide the organisational structure for databases and are the fundamental building blocks for spatial frameworks (i.e. the spatial depiction of geographic areas).

A classification is a systematic arrangement of information, using previously defined terminology or nomenclature.

Classification systems partition and attempt to simplify the complexity of the environment by providing a basis for dividing it into ordered, rational and scientific units for further study or management. They provide a generally objective approach for dividing the environment into units with similar characteristics. Once delineated, these units may contribute to a spatial framework for monitoring and reporting on aspects of the environment.

Classification systems are often hierarchical, providing ascending levels of generalisation. That is how they provide building blocks, the means of aggregating single factor information up for interpretation and reporting within a spatial framework at the local, regional or national scale. (See Figure1).

Figure 1. Relationship between classification systems and spatial frameworks.

E.g. Spatial Unit for Reporting has a relationship with Wetland hydroclasses and terrestrial landform which also has a relationship with Atkinson's Vegetation classification.

Mapping units at each level in a classification system, have a relatively high degree of homogeneity, depicting similar attributes (e.g. plant communities). These units may be repeated many times, often in sites that are not contiguous with each other.

Highly detailed classification systems used to describe and organise single factor data from specific parts of the environment are sometimes termed 'homogeneous'. Each homogenous unit classification is specific to an environmental system, attribute, or type (e.g. soils, vegetation, land).

Homogenous unit classification systems are often used for:

the actual collection of monitoring data (eg. by vegetation types)
providing the framework for site selection.

Examples include:

Atkinson's Vegetation Classification System
New Zealand Soil Classification - Hewitt
UNEP GRID New Zealand Wetland Classification
IUCN Criteria for Threatened Species Categories
Land Use Capability Land Classification

Sometimes the reporting of information also uses an environmental classification system, eg. % of each wetland type remaining that is legally protected.

Classification systems can over time be used for purposes not envisaged when the system was originally developed. In some cases these new uses may be inappropriate.

2.2 What is a Spatial Framework?

Spatial frameworks reflect a combination of environmental attributes across landscapes rather than specific attributes such as soils or vegetation.

Spatial frameworks can be based on composites of homogenous, hierarchical classification systems and often operate at a less detailed (bio-regional) scale. They are generally applied by mapping geographic entities of similar character (such as land systems or ecological districts) at national, regional and district scales. Spatial frameworks can provide the spatial basis and the context for environmental monitoring by aggregating data from homogenous unit classification systems.

Examples include:

ecological regions and districts
land systems classifications and regional data layers developed for specific purposes (e.g. Environment Bay of Plenty proposed Land Monitoring Framework, Taranaki Regional Council Sustainable Land Use Classification).
bio-climatic zones

2.3 Classification Systems, Spatial Frameworks and the EPI Programme

A classification system in the context of the EPI Programme is an organised, descriptive structure for categorising different parts of the environment so as to be able to gather, understand, organise and display information about aspects of that environment in a meaningful way.

Classification systems therefore provide a context and a method for:

determining the structure of any associated databases
assembling spatial frameworks for:
risk analysis - selecting broad areas for monitoring and management action (e.g. Taranaki Regional Council Sustainable Landuse Classification)
design of sampling programmes
selection of monitoring sites
interpretation of results
reporting

Because of the broad range of environments and issues considered in the EPI Programme a number of classification systems may be required. When choosing the most appropriate system(s) a range of factors need to be considered. These include:

the environmental policy goals
issues
the specific indicators
monitoring methodology
how the environmental classification system will be used
relationship between levels in classification systems (see Figure 2)
how independent data layers relate to each other in a spatial framework and for
how independent data layers relate to the indicators being monitored

Figure 2. Classification System Levels

Single Level, eg. Molloy & Davis Priorites for Threatened
Species Conservation Work. Hierarchical Levels, eg. New Zealand Soil Classification - Hewitt.

Many of the most powerful environmental distinctions (differences) occur at the local scale rather than the large (bio-regional) scale. The detection and tracking of change associated with environmentally distinct areas (at the local scale) implies a need for a detailed level of classification, while the tracking of broad scale changes implies a need for the bio-regional context or more general levels of classification. The appropriate level will depend on the specific purpose for which you want the distinctions. It will be important to clearly identify the purpose, and carefully select a classification system(s) which provides the right type of information and has the ability to be integrated with other systems.

It is not always possible to identify the specific scale that a classification system or spatial framework should be mapped at. This is because some systems and frameworks have never been mapped and/or a range of scales could be appropriate.

A major limiting factor is the resolution at which the data is collected. The finer the resolution (i.e. more points on the ground) the more detailed the mapping scale. For example a vegetation classification system operating at the type (most detailed) level will tend to collect data to map boundaries at a fine scale (eg 1:5000). In comparison the coarser the resolution (i.e. less points on the ground) the less detailed the mapping scale. For example a spatial framework such as bioclimatic zones defines boundaries at a coarse level (eg 1:250,000)

For monitoring and reporting Environmental Performance Indicators, it may be necessary to combine classification systems and spatial framework(s) for representing indicator data across related environmental issues.

2.4 Administrative Boundaries

It is likely that much EPI information will be gathered and reported within a framework of administrative/jurisdictional boundaries, e.g. by regional council or territorial authority boundaries.

Generally administrative boundaries do not parallel environmental systems, such as catchments and ecosystems, and by their nature sub-divide many naturally occurring environmental systems. However, most regional council boundaries in New Zealand are generally based on regional water catchments.

Consideration of administrative/jurisdictional boundaries as a spatial framework was not originally included within the scope of this contract. However, it is likely that administrative/jurisdictional boundaries will be an important reporting framework for all local authorities as these reflect the range of environments, issues and geographical units under their management and control.