Publications: Research reports and publications

Temporal variability in ecosystem metabolism of rivers in the Manawatu-Whanganui Region – Updated

1 June, 2010
Cawthron Report 1791. Prepared for Horizons Regional Council.


A one-off measurement of the daily change in dissolved oxygen in the lower Manawatu River in November 2007 indicated that this river has very high rates of ecosystem metabolism (primary production and ecosystem respiration) that are indicative of very poor ecosystem health. An assessment of the lower Rangitikei River on the same day suggested this site also had poor ecosystem health due to high rates of ecosystem respiration. Horizons Regional Council has been collecting continuous records of dissolved oxygen (DO) at five sites throughout the region since 2005, which can be used to calculate ecosystem metabolism. The aim of this study was to determine if the concerns raised about these rivers are consistent over time or among sites. This updated report replaces an earlier report (Clapcott & Young 2009) that was based on calculations from raw data which has subsequently been shown to have some measurement errors. These errors have now been addressed and metabolism rates for the five sites recalculated.

We calculated ecosystem metabolism using DO data from five sites over one year (2007). The sites were Manawatu at Hopelands, Manawatu at Teachers College, Mangatainoka at Pahiatua Town Bridge, Rangitikei at Mangaweka and Rangitikei at Onepuhi. The sites varied in their intensity of catchment land use ranging from 39% native vegetation in the catchment for the Rangitikei at Mangaweka to 10% native vegetation for the Manawatu at Hopelands. Dissolved oxygen data generally displayed characteristic daily patterns at most sites. However, we had concerns with the accuracy of the data, at some sites during some seasons where night-time DO values did not drop below 100% saturation prior to dawn. To address these concerns we corrected the data, but uncertainties involved with the correction mean that less confidence can be placed on metabolism values calculated from corrected data.

Rates of gross primary production (GPP) were low in the Rangitikei River at Onepuhi and Mangaweka and indicate good–satisfactory health throughout the year according to broad guidelines on interpretation of these measures. Rates of GPP and ecosystem respiration (ER) in the Manawatu River at Teachers College and Mangatainoka at Pahiatua suggested good–satisfactory health in autumn, winter and spring, but were indicative of poor ecosystem health in summer. In contrast, rates of GPP and ER were consistently high at Manawatu at Hopelands and indicated poor ecosystem health throughout the year. These differences in rates of metabolism among sites reflect differences in land cover with the highest values found at the most modified site (Manawatu at Hopelands) and the lowest values found at the sites with the largest proportion of the upstream catchment in native forest (Rangitikei River at Mangaweka and Onepuhi), although a point-source discharge upstream of Mangaweka appears to have had an effect on the measurements at that site.

The balance between GPP and ER indicated that all five sites were generally relying on some organic matter from upstream or the surrounding catchment to support the recorded rates of ER. However, the ratio of GPP:ER was greater than one at Manawatu at Hopelands during autumn and spring, and at Rangitikei at Onepuhi during spring suggesting that algae probably contribute significantly to the food chain in these rivers at times.

As expected there were distinct seasonal patterns in the ecosystem metabolism measurements with most sites displaying higher values in the warmer months. For example, rates of GPP and ER in the Manawatu River at Teachers College indicated satisfactory–good ecosystem health in autumn, winter and spring, but higher in summer and suggesting poor ecosystem health.

A weak positive relationship was observed between rates of GPP and water clarity reflecting the importance of light availability at the riverbed for algal photosynthesis. There was also an indication that GPP may be positively related with concentrations of dissolved reactive phosphorus, although neither of these relationships were statistically significant with the available sample size.

Continuous monitoring of DO concentration provides the opportunity to calculate rates of ecosystem metabolism that can be used to assess ecosystem health. The accuracy of the metabolism calculations is heavily dependent on the accuracy of the raw DO data. Therefore, we recommend regular checks and calibration of the DO monitoring equipment. As more data becomes available it would be useful to compare results from the Manawatu–Whanganui Region with measurements from similar large rivers to determine if the broad guidelines used in this study to interpret the metabolism measurements are appropriate for large rivers generally. Furthermore, it would be useful to conduct further comparisons of ecosystem metabolism to traditional water quality, biomonitoring or periphyton assessments to determine whether rates of ecosystem metabolism provide suitable surrogate measures of ecosystem health for large river systems where other measures can not be easily undertaken.