are a biological stream health indicator which reflects the current state and potential changes in, for example, water quality and stream habitat.
To be able to measure stream health by monitoring macroinvertebrates, a common measure known as theis used. This index involves scoring the diversity of observed at a site based on their tolerance to pollution. Those taxa which are characteristic of more unpolluted conditions score more highly than those that dominate polluted streams, and generate a higher MCI score. In this way, higher MCI scores generally indicate better river condition.
If you want to know more about benthic macroinvertebrates have a look at the factsheet.
We estimate current MCI across the whole country by using models, which are based on data collected from hundreds of monitoring sites over a five year period to 2011.
The map below shows that water quality is generally better (i.e., have a higher MCI score) in areas with more, such as on the western coast of the South Island and the central North Island. Areas with the poorest water quality (i.e., lowest MCI scores) include the upper North Island, the Manawatu Plains and lowland Southland.
Analysing the results by land cover classes (see boxplot below) shows that streams in catchments affected by urban areas are significantly worse than those in other land covers, with MCI scores less than 80. Rivers and streams in catchments where predominates are predicted to have the highest stream health, indicated by a high above 119.
Rivers and streams in pasture catchments have the widest range of MCI scores which is shown by the long 'whiskers' in the graph below. This means that streams that run through pastoral land use have both, the lowest and the highest MCI scores in New Zealand. Such variability might reflect the wide range of environments, land-use intensities and management practices associated with pasture land cover. Improved land management practices, such as riparian planting, play an important role in improving MCI scores.
Recent national trends in macroinvertebrate communities are calculated using data from a 10-year period. Usually sampling occurs only once per year over summer. This makes detecting significant trends more difficult than for the other indicators which are sampled monthly, as less data is available.
Overall, no significant trend was detected for any land use class at the majority of monitoring sites shown by the grey colour in the graph below. This doesn’t necessarily mean that there were no trends, but instead, because the sites are only sampled annually, there are not enough data points available to identify significant trends. However, where a change was detected, there were more worsening trends than improving in all land-cover classes. Pasture land cover exhibited the highest proportion of sites showing improvement but also the highest proportion of deteriorating sites.
See a factsheet on how national trends are calculated.