Investigation of relationships between invertebrates and dissolved nutrient concentrations in New Zealand rivers

Report: TR 2022/44

Author: Ton Snelder (LWP Ltd)

A requirement of the National Policy Statement for Freshwater Management 2020 (NPS-FM; NZ Government, 2020) is the identification of appropriate nutrient concentration criteria and the implementation of these criteria through setting limits on resource use via regional plans.

For rivers, the NPS-FM requires regional councils to determine the appropriate nutrient concentration criteria to “achieve a target attribute state for periphyton, any other nutrient attribute, and any attribute that is affected by nutrients” (NPSFM, Section 3.13). However, guidance provided by MFE for implementation of NPS-FM Section 3.13 highlights that a challenge in defining instream nutrient concentration criteria is accounting for the extent to which environmental factors influence the sensitivity of attributes to nutrient enrichment.

Two NPS-FM attributes that can be expected to be affected by nutrients are the macroinvertebrate community index (MCI) and the quantitative macroinvertebrate community index (QMCI). The values of both indices can be understood as surrogate measures of ecosystem health. The NPS-FM defines bands for both indices from A to D that indicate a scale from excellent to unacceptably poor ecosystem health.

Death et al. (2018) and Canning et al. (2021) derived nutrient criteria that pertain to objectives that are expressed as threshold values for these two indices. The criteria were derived by fitting bivariate linear regression models that express the relationship between the indices and nutrient concentrations observed at river monitoring sites across New Zealand. The linear regressions were fitted to all the available data so did not attempt to account for differences in environmental conditions that exist across the national pool of monitoring sites. Because these models do not account for variation in environmental factors that likely influence the sensitivity of invertebrates to nutrient enrichment, the derived “national criteria” incur a risk of not achieving the desired outcomes. The risk extends in two directions, the national criteria may be under-protective in some systems and over-protective in others.

This study undertook two sets of statistical analyses to investigate the criteria derived by Death et al. (2018) and Canning et al. (2021). First, simple bivariate linear regression models were used to assess the strength of the relationships underlying the criteria and the precision of the derived criteria. Second, variance partitioning analysis was also used to quantify the strength of relationships between the invertebrate indices and nutrients while considering the extent to which these relations may be overestimated if other environmental variables are not accounted for.