Case study 2 – Assessment of Escherichia coli Load Reductions to Achieve Freshwater Objectives in the Rivers of Otago
This study quantifies the Escherichia coli (E. coli) load reductions needed to achieve options for freshwater objectives (FWOs) for human contact in rivers of the Otago region. E. coli concentrations and loads are used to indicate the risk to human health posed by freshwater. Otago Regional Council is required to set objectives for human health using the E. coli indicator, and policies for achieving them, under the National Policy Statement – Freshwater Management (NPS-FM).
The analysis does not consider how E. coli load reductions would be achieved and only aims to inform the Otago Regional Council about the magnitude of the load reductions needed for each option. The load reductions required were assessed for all individual river segments in the study area. The results can be reported at several levels of detail including for individual river segments, individual freshwater management units and the whole study area.
The study estimated the uncertainties associated with all assessments of the reductions in E. coli loads required to achieve nominated FWOs. Uncertainty is unavoidable because the analyses are based on models that are simplifications of reality and because the models are informed by limited data. The uncertainties associated with two key components of the analyses: the estimated E. coli concentrations and loads were quantified and were combined in a Monte Carlo analyses. The Monte Carlo analyses simulated 100 ‘realisations’ of the load reduction calculations, which were used to define the probability distributions of all estimates. The probability distribution describes the range over which the true values of the load reductions are expected to lie and was represented by the 5th and 95th percentiles of the distribution (i.e., the limits of the 90% confidence interval).
The regional E. coli load reduction required to achieve the national bottom lines (i.e., the least stringent acceptable FWO) for the study area was 24% (90% confidence interval 6% to 74%). The uncertainty bounds are very large and it is unlikely that these can be significantly reduced in the short to medium term (i.e., in less than 5 to 10 years). This is because, among other factors, the modelling is dependent on the collection of long-term water quality data and reducing uncertainty would require data for considerably more sites than were available for the study.
This study can help inform the process for deciding on limits to resource use, by providing an assessment of the approximate magnitude of E. coli load reductions needed to achieve several options for FWOs, with a quantified level of confidence and risk associated with each option. However, this report does not consider what kinds of limits on resource might be used to achieve any load reductions, how such limits might be implemented, over what timeframes and with what implications for other values. The NPS-FM requires regional councils to have regard to these and other things when making decisions on setting limits. This report shows that these decisions will ultimately need to be made in the face of uncertainty.