Methods - how we monitor
We currently monitor sediment-dwelling animals at five sites in the Firth of Thames, five sites in Whaingaroa (Raglan) Harbour and five sites in Tairua Harbour (see maps below). Monitoring sites are distributed throughout the main part of each estuary and are located at the mid-intertidal level of sand and mud flats (in areas without seagrass or mangroves).
Monitoring frequency and history
Monitoring is currently undertaken once per year in spring (i.e. between September and November). Monitoring began at the Firth of Thames and Whaingaroa Harbour in 2001 and at Tairua Harbour in 2012.
We currently collect 10 cores (13 cm in diameter and 15 cm deep) from a permanent plot located at each monitoring site. Each plot is approximately 10,000 m2 (100 m x 100 m), and is divided into 10 equal-sized sectors. Within each sector a random location is selected and a core sample taken. Cores taken in adjacent sectors must be at least five metres apart.
We separate animals from the sediment in each sample using a sieve with 500 µm mesh. We then preserve samples in 70 per cent isopropyl alcohol (diluted with tap water) and add a stain (Rose Bengal) that turns all the animals red/pink.
In the laboratory we separate the stained animals from the sediment and other debris, identify and count the animals to the lowest practicable taxonomic level (generally species or species group), and then store the samples in 50 per cent isopropyl alcohol.
This indicator presents a Traits Based Index (TBI) score for each monitoring site in the Firth of Thames, Whaingaroa (Raglan) Harbour and Tairua Harbour. The information is derived from the sediment-dwelling animals identified in the core samples collected from each monitoring site from 2012 to 2017.
The Traits Based Index (TBI), developed by NIWA and Auckland Council, is based on the numbers of types of animals exhibiting particular traits. It has been applied to Waikato region monitoring data following a series of trials and efforts to improve taxonomic resolution to be consistent with Auckland.
All animals identified in samples are first categorised according to traits (e.g., their feeding mode, degree of mobility, body size, etc.) that are reflective of their ability to perform certain ecosystem functions. The index value is then calculated based on the numbers of taxa in seven particular trait groups (groups that have previously been shown to respond to mud and heavy metal contaminants).
Index values range between zero and one (with values near zero indicating highly degraded sites, and values near one indicating more pristine environments). Areas with high numbers of taxa per functional group (and high index values) are thought to have a greater capacity to cope with species losses. This is a component of resilience that contributes to maintenance of ecosystem functions in the face of stress and environmental disturbance. This is what the TBI gauges and, as such, is a meaningful indicator of estuarine health.
There are no formal guidelines or standards available for assessing this data in New Zealand at present. Our approach is based on similar programmes that have been developed by the National Institute of Water and Atmospheric Research (NIWA). Auckland Council uses similar estuary monitoring programmes.
With just six years of data suitable for TBI calculations, trends in estuarine health over time have not yet been analysed statistically. It is also not yet possible to set thresholds for TBI values that constitute good, intermediate and poor functional redundancy. However, in the future we expect to be able to develop this indicator and to detect changes in health over time, should changes be occurring.
Quality control procedures
Qualified and experienced personnel carry out all sorting and taxonomic identification in line with a processing, identification and quality assurance protocol. On each sampling occasion a different person resorts, re-identifies and re-counts at least 10 percent of the samples from each site for quality control. The minimum acceptable sorting efficiency is 95 percent, and the minimum acceptable identification and counting efficiency is 90 percent.