Publications: Peer-reviewed journal articles (by staff)

Seasonal patterns in ecosystem metabolism in five Auckland rivers

1 November, 2006
Cawthron Report 1226. Prepared for Auckland Regional Council.


Auckland Regional Council (ARC) has been collecting continuous records of dissolved oxygen (DO)concentration at a number of sites around the Auckland region since 2003. These data can be used to calculate rates of ecosystem metabolism (the combination of gross primary production and ecosystem respiration) and thus provide a picture of the functional health of these river systems. Continuous datasets such as these enable improved understanding of temporal variability in metabolism rates and the effects of environmental stressors on this variability. This information is important for developing and improving guidelines for interpreting what metabolism measurements mean in terms of ecosystem health.

Data from five rivers were used for this study. The sites were Mahurangi at College (high % native in catchment), Ararimu at Old North Road (forestry/agriculture), Mauku at Aka Aka Road (horticulture/agriculture), Whangapouri at Paerata (urban), and Kaukapakapa at Taylors (agriculture). A 6-12 day period within each season was extracted from the dataset available for each river and used to calculate daily rates of metabolism. Ecosystem metabolism was calculated successfully during all seasons at Ararimu, Mahurangi, Mauku and Whangapouri. However, there were problems with the calculations at Kaukapakapa during summer, autumn and spring that appear to be related to the extremely low reaeration coefficient at this site. Metabolism rates subsequently were estimated at Kaukapakapa assuming that the reaeration coefficient was zero.

Rates of gross primary production (GPP) generally were low at Ararimu, Kaukapakapa and Mahurangi and indicative of healthy conditions. However, rates of GPP at Mauku and Whangapouri were considerably higher and indicative of satisfactory or poor ecosystem health. Rates of ecosystem respiration (ER) at Mahurangi and Kaukapakapa were relatively low and indicative of healthy or satisfactory ecosystem health. In contrast, rates of ER were high at Whangapouri and consistently indicative of poor ecosystem health. Rates of ER at Ararimu and Mauku were in between these extremes and indicative of poor health at times and satisfactory health at other times.

The balance between GPP and ER indicated that these sites generally were relying on organic matter from upstream or the surrounding catchment to support the rates of ER that were recorded. The only exception to this was Mahurangi during spring when rates of GPP were slightly higher than rates of ER and in-stream production may have been supporting the food web. The P/R ratios provided no indication of poor ecosystem health at any of the sites, suggesting that this ratio is not a particularly sensitive measure of stream health.

Of the five sites included in this study, Mahurangi and Kaukapakapa had rates of ecosystem metabolism that were similar to those measured in healthy river ecosystems in other parts of New Zealand and overseas. However, the results suggest that ecosystem health is impaired to some extent at Ararimu and Mauku. Ecosystem health appeared to be particularly poor at Whangapouri, which had consistently the highest rates of GPP and ER. These results were expected for Mahurangi (largest proportion of native forest and likely to be most healthy) and Whangapouri (largest proportion of urban and likely to be least healthy). However, the ranking for the other sites would have been difficult to predict based on land cover percentages.

Continuous monitoring of DO at a site that represents reference condition is recommended. It would be useful also to determine the linkage between functional measures of river ecosystem health and more traditional measures such as water quality and indices of stream invertebrate community composition. The existing guidelines for interpreting metabolism measurements in terms of river health appear to be useful within the Auckland region. However, these guidelines could be revised as more specific information on reference condition within the Auckland region becomes available.