Methods - how we monitor
Currently this indicator covers 19 estuaries in the Waikato region; 15 on the East Coast (including the southern Firth of Thames) and four on the West Coast. Prioritisation has been given to estuaries most at risk from human activity.
Table 1. Currently surveyed estuaries in the Waikato region, as of 2012
|West Coast||Aotea Harbour|
|Waikato River Estuary*|
|Whaingaroa (Raglan) Harbour|
|East Coast||Colville Bay|
|Firth of Thames**|
|Otama River Mouth|
|Te Kouma Harbour|
* Estuarine vegetation survey covers the area of the river mouth only.
**Estuarine vegetation survey area covers the southern Firth of Thames only and does not include intertidal areas throughout the entire Firth.
It is proposed that intertidal estuarine habitats will be monitored approximately every five years to facilitate effective management. This is consistent with the timeframe proposed in the Ministry for the Environment’s Environmental Performance Indicators Programme and similar international monitoring programmes.
More frequent purpose designed measurements will be undertaken in areas where significant change is occurring, or there are potential adverse effects on habitats.
Initial baseline information on estuarine vegetation for the eight estuaries on the East Coast was based on interpretation of colour aerial photographs from 1995 and 1996, in conjunction with field surveys undertaken from 1997 to 1999. Information for the four West Coast estuaries is based on aerial photographs taken in 2002 and 2003 and field surveys undertaken from 2004 to 2005.
In the future, change in spatial extent and distribution of each habitat type will be estimated by referring back to this baseline information. For some habitats, information on historical extent and distribution can be obtained from older records. Limited availability of suitable photographs and reliable field data, does limit the role of historical information in vegetation trend analysis. However, this information is important to interpret long-term cycles.
Following the initial surveys outlined above, from 2005 to 2010 12 additional surveys were completed on the East Coast (including eight repeat surveys) and four repeat surveys on the West Coast.
A further four repeat surveys were carried out on the West Coast from 2011 to 2012, as well as three additional surveys in previously unmapped estuaries on the East Coast.
Baseline measurement technique
Initial baseline information on intertidal estuarine vegetation was derived from the interpretation of aerial colour photographs taken in 1995 and 1996 (East Coast estuaries) and 2002 and 2003 (West Coast estuaries) at a scale of 1:5,000 or 1:10,000, combined with ground survey.
Aerial photographs of each estuary were mosaiced. A skilled interpreter drew the vegetation boundaries onto transparent acetate film overlaying the mosaics. These boundary lines were ground surveyed to establish their accuracy.
The traced vegetation boundaries were converted to digitised images, and imported into Geographic Information Systems software (GIS). The mapped vegetation was registered to the New Zealand Map Grid by overlaying the vegetation and harbour boundaries over a 1:50,000 map sheet. Known points on the map were used to adjust the positions of the un-referenced data.
Revised measurement technique
The method used in initial surveys whereby a skilled interpreter drew the vegetation boundaries onto transparent acetate film overlaying the mosaics, has been superseded by Geographic Information Systems (GIS) technology. While the new technology allows greater accuracy there is a need to relate the dimensions of the vegetation areas back to the baseline surveys which introduces an element of error. The mapped vegetation has been registered to the New Zealand Map Grid on a 1:50,000 scale.
Estuarine vegetation has been identified within four broad vegetation communities in Waikato region. Examples are given in table 2 below of the possible makeup of these communities.
Table 2. Examples of species makeup for vegetation communities in Waikato region estuaries
|Vegetation community||Habitat type|
|Mangrove – Avicennia marina subsp. australasica||Mangrove|
|Seagrass – Zostera muelleri subsp. novozelandica||Seagrass|
|Salt marsh||The rush community includes, for example, species such as Apodasmia similis (golden jointed rush) and Juncus kraussi subsp. australiensis (sea rush).||Salt marsh|
|The marsh ribbonwood community includes, for example, species such as Plagianthus divaricatus (marsh ribbonwood), Samolus repens (sea primrose), Selliera radicans (remuremu) and Austrostipa stipoides (coastal immorality grass, Buggar grass).||Salt marsh|
|The salt meadow community includes for example, species such as Samolus repens (sea primrose), Selliera radicans (remuremu), Sarcocornia quinqueflora (glasswort), Cotula coronopifolia (bachelor's button), Eleocharis actuta (sharp spike-sedge), Isolepis cernua (slender clubrush) and Triglochin striata (arrow grass).||Salt marsh|
|Invasive species - for example, Paspalum vaginatum (saltwater paspalum), Spartina sp. (cord grass).||Invasive exotic species|
The classification of vegetation types broadly follows the proposed national classification system, developed under a Ministry for the Environment Sustainable Management Fund project.
The New Zealand Coastal Policy Statement 1994, the purpose of which is set out in the Resource Management Act Section 56, states as matters of national importance:
- The preservation of the natural character of the coastal environment (including the coastal marine area), wetlands, and lakes and rivers and their margins, and the protection of them from inappropriate subdivision, use, and development;
- The protection of outstanding natural features and landscapes from inappropriate subdivision, use, and development;
- The protection of areas of significant indigenous vegetation and significant habitats of indigenous fauna;
- The maintenance and enhancement of public access to and along the coastal marine area, lakes, and rivers; and
- The relationship of Māori and their culture and traditions with their ancestral lands, water, sites, waahi tapu, and other taonga.
There are a number of limitations associated with monitoring the extent and distribution of coastal habitats using remote sensing technologies and ground survey.
- The spatial resolution is generally too coarse to discriminate small units of different habitat types. For example, under ideal conditions individual patches of habitat as small as 1 m in diameter are detectable in photography at a scale of 1:24,000, but in practice, minimum habitat sizes recorded are greater than or equal to 0.3 ha (Ferguson and Wood, 1990). Estimates of habitat spatial extent therefore tend to be conservative and are not suitable for detecting short-term or small-scale or local changes.
- The resolution of the imagery is insufficient to discriminate between individual species.
- Locating the boundaries of different habitat types is often difficult and error prone (for example, tracing accuracy may be ± 5 m on well defined boundaries, but may degrade to ± 10 m along poorly defined boundaries). Problems arise in particular where there are mixed or co-dominant habitat types and decisions need to be made concerning the classification of these habitats. For example, a sparse mangrove may sustain an understorey of saltmarsh, and seagrass beds may support recently-germinated mangrove seedlings. For the purpose of monitoring it is necessary to acknowledge the presence of both vegetation types in these areas, which can be identified as a 'mixed' category.
- A number of errors may be introduced through the remote sensing and ground survey process. For example, in drawing boundaries on an image. A 1 mm pen line represents 10 m on the ground at a scale of 1:10,000 and at the same scale a 0.2 mm diameter point covers an area of 2.54 m². In geo-referencing the digitised images to a 1:50,000 scale map, an error in position of 0.5 mm represents a 25 m error on the ground. These errors may be additive.
- Aerial images before Waikato Regional Aerial Photography Service (WRAPS) 2002 have not been geo-rectified. Without this information it is not possible to provide an indication of the locational accuracy (in terms of metres) of the habitat maps. Aerial images from 2002 onwards are digital orthophotos. All images have a resolution of 1m2 or less and a spatial accuracy of ± 3 m at 90 per cent confidence.
- The classification of habitat units, or communities, is potentially a significant source of variation in the mapping process. This is primarily due to the current absence of an acknowledged classification system for coastal ecosystems (this is currently being addressed through a number of MfE Sustainable Management Fund Projects).
Changes to this indicator are planned in the short and long-term. These will include:
Development of a regional 'operational protocol'. This will incorporate 'mapping rules', which will provide a uniform basis for collecting information on habitat extent, distribution and type, from scene to scene and time to time, thereby allowing robust comparison of two or more scenes or times. Protocols for generating, ground survey, classification and controlling the quality of the habitat maps, documenting change in habitat spatial extent and distribution, and the use of historical data, will be produced. An assessment of computer-aided image sampling to map habitat boundaries will also be included.
Expansion of regional coverage of estuaries included in the programme. Sixteen estuarine areas in the region have been included in the scope for the estuarine component of the WRAPS project: the whole of the Firth of Thames, Manaia, Te Kouma, Coromandel, Waikawau, Kennedy Bay, Whangapoua, Whitianga, Tairua, Wharekawa, Whangamata, Otahu, Port Waikato, Raglan, Aotea and Kawhia.
Further development of the classification of coastal habitats in the region and improving the quality and quantity of data extracted from the remote sensing imagery (such as quantitative descriptors and photographic keys). This may, for example, include classification according to morphological types as either 'continuous' or 'patchy' (for example, extensive seagrass beds vs fragmented ones, containing numerous open bare patches); identifying important 'thematic cover classes' for each habitat type (for example, dense seagrass = 70-100 per cent cover; medium/dense seagrass = 30-70 per cent cover; sparse seagrass <30 per cent [e.g., Fyfe et al., 1999]); and an evaluation of the potential for relating functional health/condition status (such as biomass and productivity) of habitats to spectral radiances observed through remote sensing.
Where available, historical aerial photography at similar scales will be analysed in order to quantify historical changes and trends in habitat spatial extent and distribution.