View all publications

Human activities are degrading some ecosystems

New Zealand has a long history of using river controls works, such as stop banks, weirs, channel realignment, and streambank plantings (eg willow trees) to confine rivers to well-defined channels and to protect nearby floodplains from floodwater that could damage infrastructure and houses.

River channelling prevents river migration, makes floodplains available for infrastructure development, increases agricultural efficiency, and improves flood control and security (Catlin et al, 2017). However, river-control works alter the natural character of rivers including their width, slope and depth, number of bends, and bed elevation/depth (Davies & McSaveney, 2011). These changes can erode river banks and increase deposition of sediments downstream (Fuller et al, 2011).

River-control works reduce the ecological connectivity of floodplains (and associated lakes and wetlands) within rivers by altering flood pulses into the floodplain. Flood pulses are an important driver of connectivity within riverine floodplains – flood pulses aid indispersingseeds, establishing plants, cycling nutrients, scouring, depositing sediment, and maintaining species richness (Catlin et al, 2017). This reduction in ecological connectivity of floodplains is particularly important in rivers with low-gradient floodplains and prolonged floods, and less important in rivers with steep gradients and flashy flows (Young et al, 2004).

Planting trees at the river edge or on banks is also a form of modifying physical habitats in rivers. Riparian planting can be an effective way to quickly restore bank stability, improve stream habitat, and control streambank erosion (Phillips & Daly, 2008). Willow trees are commonly planted in New Zealand as they grow faster than native species. However, willow can invade stream channels, causing stream diversions and blockages and disrupting habitats for freshwater species (Phillips & Daly, 2008).

Deposited fine sediment can degrade freshwater habitats

Deposited sediment occurs naturally in the beds of rivers and streams, but too much fine sediment (particles less than two millimetres in size) can severely degrade streambed habitat (Davies-Colley et al, 2015). Excess fine sediment fills up the spaces between cobbles and gravel used by fish and invertebrates and can alter food resources or make them difficult to consume (Davies-Colley et al, 2015). Excess fine sediment can also affect the aesthetic appeal of rivers and streams for human recreation.

Erosion from the land results in the deposition of sediment in rivers. For example, our fast-eroding Southern Alps produce masses of sediment (eg boulders, rocks, cobbles, sand, silt, and clay), most of which is initially displaced by landslides, large gullies (Fuller & Marden, 2008), and earthflows (Glade, 1998). Sediment is then transported downstream by water and gravity; during this time abrasion contributes to the creation of fine sediment. These processes of transport and abrasion have contributed to the formation of our iconic braided river systems and many of our flat land areas (like the Canterbury Plains) (Gray et al, 2016). Because sediment is naturally transported longitudinally through a river network, its state at any given point will be influenced by climate, geology, topography, and current velocity (Knighton, 1998).

Human activities can have a negative impact on the natural sediment cycle by accelerating the delivery of sediment to streams and increasing the quantity of smaller particle sizes. Changes in land cover are one factor that can alter rates of erosion (Dymond et al, 2010). Native forests covered 85 to 90 percent of New Zealand before humans arrived around 1,000 years ago (Quinn & Phillips, 2016). Polynesian settlers cleared the forests with fire, reducing the original forest by half. Subsequent European colonisers further reduced forest cover to make way for agricultural and urban land use (Quinn & Phillips, 2016). Accelerated erosion is still being caused by human activities like earthworks, forest harvesting, livestock farming, and cultivation practices. Basher (2013) suggested that annual sediment loads (the amount of eroded sediment that makes its way into streams) from agriculture may have been declining over the past 30 years. This is based on changes to land cover (eg an increase in plantation forestry and scrubland), but does not take into account variations in weather or other factors that may also be having an effect on sediment loads.

There are not many sites across the country where fine sediment deposition has been observed over the long-term using consistent methods. This makes it challenging to report on deposited fine sediment at a national level (Hicks et al, 2016; Clapcott et al, 2011).

A spatial model was developed using the sediment observations from the New Zealand Freshwater Fish Database. The model estimated an average fine-sediment cover of 29 percent for stream segments (Clapcott et al, 2011). The same model also estimated an average fine sediment cover of 8 percent in the absence of human land use and land-cover change, suggesting a significant increase in deposited fine sediment cover has occurred in New Zealand rivers since human occupation. In future, collecting robust data is needed to validate such estimations, as fine-sediment levels greater than 20 percent cover are known to have adverse effects on streambed life (Clapcott et al, 2011; Burdon et al, 2013).

For more detail see Environmental indicators Te taiao AotearoaStreambed sedimentation [Stats NZ].

Wetland habitats have been greatly reduced

Wetlands perform many functions. They filter nutrients and sediment from water, absorb floodwaters, and provide habitat for plants, fish, and other animals. The draining of our wetlands due to land-use change and farming practices has led to a loss of biodiversity and natural function (Clarkson et al, 2013).

Although we have no national information on the health of our wetlands, we have information on their extent. Before human habitation, wetlands covered approximately 2.5 million hectares of New Zealand’s land area. In 2008, the extent of wetlands reduced to approximately 250,000 hectares – 10 percent of their original extent (Ausseil et al, 2008).

Wetland losses occurred historically through drainage and conversion to farmland (McGlone, 2009). Although we are less clear on contemporary patterns of national wetland extent, we know that losses are still occurring today. For example, in Southland a loss of 1,235 hectares, which equates to 10 percent of wetlands outside the area’s public conservation land, occurred between 2007 and 2014–15 (Ewans, 2016).

The West Coast has the greatest extent of wetlands (84,000 hectares), followed by Southland (47,000 hectares), and Waikato (28,000 hectares) (Ausseil et al, 2008).

For more detail see Environmental indicators Te taiao AotearoaWetland extent [Stats NZ].

Structures in water bodies can obstruct fish migrations

Structures commonly found in our streams and rivers, such as dams, weirs, culverts, and tide gates, affect river flows and can obstruct fish migration (Franklin et al, 2014). A significant proportion of New Zealand native fish species need to migrate to and from the sea to complete their life cycles (McDowall, 2010). In-stream barriers prevent fish from reaching suitable habitats and food sources and completing their life cycles. Ultimately, this restricts the quantity and quality of habitat that is available for these species to colonise, as they can only use areas that retain access to the sea.

Barriers to fish can result in changes to the composition of fish community above and below the barrier. For example, research into dams in New Zealand found sites above dams had fewer fish species, lower proportions of diadromous fish species (fish that spend parts of the life cycle in fresh water and the ocean), and higher proportions of exotic fish species (Jellyman & Harding, 2012).

The combined physical characteristics of the barrier and surrounding environment, and the swimming or climbing ability of a fish species, determine the extent to which a barrier may negatively impact a fish community (Franklin et al, 2014). For example, īnanga (one of the five whitebait species) are weak swimmers and cannot climb, whereas juvenile eels/tuna can move through small spaces both in and out of water, and climb most wetted surfaces (Stevenson & Baker, 2009). However, not all barriers are detrimental to fish communities; in some circumstances, the barrier will prevent invasive species from reaching the fish community, thereby protecting the ecosystem. For example, our native galaxid fishes can persist and thrive in habitats above barriers that keep out exotic trout (Woodford & Mcintosh, 2010; Townsend & Crowl, 1991).

The extent and degree to which human-made structures in our water bodies obstruct fish passage are not known nationally. Several regional councils do collect this information, but in inconsistent ways. In one regional example, data from Hawke’s Bay shows that 80 of 240 (33 percent) culverts, weirs, and stormwater pump stations were identified as barriers to fish passage during some or all flow conditions between 2002 and 2010 (analysed by Ministry for the Environment and Stats NZ).

An example of a barrier to some fish.

An example of a barrier to some fish.

Source: NIWA

For more detail see Environmental indicators Te taiao AotearoaSelected barriers to freshwater fish in Hawke’s Bay [Stats NZ].

Pest species pose a major threat to fresh water biodiversity

Many algae, plant, and animal species introduced intentionally or accidentally into our environment (pests) have altered freshwater ecosystems and contributed to the decline in native freshwater species (Collier & Grainger, 2015). New Zealand is considered one of six global hotspots for non-native fish introductions, with 21 species of non-native fish in our freshwater ecosystems. The number of non-native freshwater fish species in New Zealand has increased from 12 in the 1930s to 21 in 2010 (Collier & Grainger, 2015).

Pest species can reduce native biodiversity by preying on native species, competing for food and habitat, and damaging freshwater habitats (see Koi carp in New Zealand). Freshwater plant pests can cause economic losses by blocking water intakes for hydroelectricity generation, impeding drainage and irrigation, and affecting cultural and recreational values and landscape aesthetics.

Koi carp in New Zealand

Of all our introduced freshwater fish, the Koi carp has the most adverse impact on the key components of freshwater ecosystems (Rowe & Wilding, 2012). Presumed to have arrived in New Zealand in the 1960s, the Koi carp feeds on material in sediment, stirring this material as it feeds. This resuspension of sediment and nutrients reduces water clarity and can lead to algal blooms. The Koi carp also feeds on invertebrates and the eggs of other fish, and competes with our native freshwater species. It is found in the North Island, and is widespread in Auckland and Waikato. Since its eradication from Nelson in 2001–03, it is no longer present in the South Island (Collier & Grainger, 2015).

Koi carp in New Zealand              Koi carp in New Zealand

Source: Bruno David

Introduced species, such as trout, can also have recreational and food-gathering benefits. The brown trout was brought to New Zealand in 1867. Its successful establishment has been a boon for recreational anglers, but has also had negative impacts on New Zealand streams (Townsend, 2003). In many streams, trout have replaced native galaxiid fish and altered how kōura (freshwater crayfish) and other large invertebrates are distributed (Mcintosh et al, 2010; Usio & Townsend, 2000). Because trout prey on invertebrate species that graze algae, algal biomass has been found to be six times higher in some streams with trout compared with neighbouring streams without trout (Townsend, 2003).

Once established, freshwater pest species can be difficult to control or eradicate. For example, didymo (Didymosphenia geminata), an introduced algae first discovered in New Zealand in 2004, is now found in many South Island rivers. A study of 20 South Island rivers found that fish biomass tends to decline to up to 90 percent when there is a heavy infestation of didymo (Jellyman & Harding, 2016). Controls of didymo have been trialled but finding an effective and environmentally acceptable control is challenging, especially considering didymo can regenerate from a single cell (Ministry for Primary Industries, nd).

In 2013, nine fish species, 11 invertebrate species, and 41 algae and plant species were identified as pests of greatest concern to our freshwater environments (see table 6; Champion et al, 2013). This means these species have the potential to form self-sustaining populations in other freshwater ecosystems beyond their present range, indicating that pressures from these pest species could get worse.

Table 6: Number of pest species of greatest concern by major taxonomic group, 2013

Species group


Species thought to no longer be present in New Zealand at time of publication
in 2013

‘Unwanted’ or ‘notifiable’ species listed in the Biosecurity Act 1993











Plants (algae)




Plants (freshwater weeds)




Source: Champion et al, 2013

For more detail see Environmental indicators Te taiao AotearoaFreshwater pests [Stats NZ].