Q&A: The state of our rivers and lakes

More data is needed to improve monitoring of our lakes and rivers, scientists say. Photo / File

A new report out today shows more worrying trends around New Zealand's freshwater estate. Here, two leading scientists who served on the report's technical advisory group field some questions about the health of our lakes and rivers. They are Professor Richard McDowell, chief scientist of the Our Land and Water National Science Challenge, and Dr Scott Larned, manager of freshwater research at the National Institute of Water and Atmosphere.

There has been a big public focus on "swimmability" but what are some of the other big issues facing our freshwater ecosystems?

Professor Richard McDowell (RM): Swimmability as indicated by E. coli and toxic cyanobacteria in rivers and lakes is only one aspect of water quality.

The current report outlines the state and trend in several other aspects such as nitrogen and phosphorus concentrations (as they relate to periphyton - algae growth), the macroinvertebrate community index (an indicator of general stream health), the cultural health index (an iwi perspective of local stream/estuarine health) and water flows.

Dr Scott Larned (SL): Yes, there are several big issues.

Some of the biggest are wetland loss and degradation, nutrient enrichment in lakes, estuaries and lowland rivers, invasive non-native species (particularly aquatic plants and algae), potential effects of river flow alteration due to diverting flood flows to storage (a potential problem at this point, with inadequate data), potential decline and loss of rare native species, and high contaminant levels (including dissolved metals) in urban streams.

Were there any trends found in the latest data that were of particular concern?

RM: Some important context is that the state (i.e. concentrations of water quality indicators) are still very good when compared internationally; indeed, some by several orders of magnitude.

For trends, we still see more sites showing increasing concentrations of indicators than decreasing, commensurate with the respective pressures outlined in the report.

There are differences in some indicators due to lag-times between intensification and detection caused by different flowpaths.

There was one (pleasant) surprise: the decrease (i.e. improvement) in phosphorus concentrations - nationally.

There are some hypotheses proffered as to why on page 46 [of the report].

SL: There is a real mixture of improving and degrading trends in water quality, and this applies to all of the major land-cover classes (pastoral, urban, natural and plantation forest).

The general trend of greatest concern is the gradual increase in nitrate-nitrogen at many (but not all) pastoral and urban river monitoring sites.

This general pattern may be partly due to the "legacy effect" of nitrate that was leached from land years or decades ago and is now emerging from groundwater into surface water.

It's a concern because it's very difficult to reduce nitrate levels in groundwater, and because the current trend may continue far into the future due to the legacy effect.

Several data gaps have been identified, what areas do you think we need to spend more effort to better understand?

RM: I suspect you'll get a different answer from every member of the technical advisory group.

Hence, this needs to be done objectively and with an analysis of benefit. MfE are endeavouring to do this.

There is a sundry question, which is what can we do better with what we have: I would point to our inability to link cause and effect between pressures (e.g. land use) and response (e.g. water quality indicators).

This is because we only have general non-specific data about landuse (e.g. fertiliser applications and stocking rates), which mean that any land use signature get averaged-out before the stream.

The data is there, but privacy concerns rightly prevent it being used.

The solution may be to anonymise the data such that these links can be made and therefore quantitative solutions found.

SL: Good question.

In general, the freshwater domain report focuses on patterns.

These include patterns in space, like broad differences in water quality in urban, agricultural and native forest areas, and patterns in time, like temporal trends in water quality and fish abundance.

What we need most, in addition to patterns, are cause-and-effect relationships.

We need to identify causes to understand why some water quality variables are getting worse in some areas, and other variables are improving.

The most valuable causal data would be detailed land use data.

These data are not available at national scales, so we rely on land-cover data instead, which doesn't provide cause-and-effect relationships.

To make this need more tangible, here is an example.

There is a widespread improvement under way in river phosphorus concentrations in agricultural areas. This could be due to stock exclusion, better fertiliser management, better effluent management, better erosion control, or some combination.

If we have good information about the causes for the improvements in river phosphorus, we can encourage specific practices, and provide evidence that investments in fencing, effluent treatment, etc. are good investments.

Other big gaps include wetland monitoring, which is minimal at present and mostly focused on the extent of wetlands, not their conditions.

More lakes and groundwater sites need to be monitored to assess national state and trends.

And more rivers in undeveloped catchments (like the DoC estate) need to be monitored to provide baseline conditions.

Those baselines are needed to identify the effects of global change without the confounding effects of local land use.

Baselines are also needed for restoration targets.

More work on the health risks of cyanobacteria is needed.

Cyanobacteria and faecal pathogens are both part of the swimmability proposals, but our knowledge of health risks due to contact recreation near cyanobacteria is pretty minimal.

For water quantity, the biggest gap is measured abstraction data (also called metered takes).

The current analyses are based largely on consented takes, which are potential, not actual.

Other gaps in water quantity reporting concern interactions between surface water and groundwater, including how abstraction of one affects the quantity of the other.

For ecosystems and habits, the biggest gaps include physical habitat monitoring, fish, periphyton and cyanobacteria monitoring, and estuary monitoring.

Regional councils are starting to expand their monitoring programmes to include these areas, but it will be several years before there is adequate data for accurate assessments.