You are here: Home» NZFFA Library» Why trees?» Trees for Erosion Control/Soil Conservation» Report: Trees for steep slopes» Tree Species

Report: Trees for steep slopes

Dean Satchell
Sustainable Forest Solutions

Reviewed by Mike Marden, July 2018.

PDF download of this report ».
Please note that the web report is regularly updated whereas the pdf download above is dated July 2018.

Tree Species

To control erosion on slopes "the need is for trees, not necessarily radiata pine." (Hocking, 2006a). Species preference can sometimes be from personal foresight, sometimes aesthetic considerations and sometimes economic drivers. Preference may also be driven by management choices and soil reinforcement performance that offers a reduced risk for erosion events to occur.

Species selection

The ideal tree species for erosion mitigation on hillslopes has the following traits (Clinton et al. 2009):

  • Tolerance to New Zealand's soils and climatic conditions;
  • Easy to establish;
  • An extensive root system that can tolerate wet conditions and even burial in aggraded gullies;
  • Capable of regenerating or coppicing freely.
  • Long lived.
  • No potential to become a pest.
  • Production of high-quality timber of sufficient value that could be partially harvested.

Canopy growth rates, degree of canopy closure, along with root strength and root growth rates all determine the individual tree species suitability for controlling mass movement erosion. If alternatives to radiata were available that had similar growth rates but were suited to longer rotation lengths and had the ability to coppice, these could be promoted for improved erosion control forestry plantations (Phillips et al. 2012).

Radiata pine has the advantage of rapid early growth rates that offer good root reinforcement of soils by the age of 8-10 years (O'Loughlin, 2005). However, the traditional regime of clearfelling at 25 to 30 years old is not optimal for long term stability of sensitive slopes (O'Loughlin, 2005). Longer rotation lengths result in less frequent erosion.

Being the most studied plantation forest species in New Zealand, radiata pine provides a benchmark with which to compare performance of other species in terms of controlling erosion (Phillips et al. 2015a). However, even for radiata pine there is a limited understanding of structural root system development under different soil types and climatic zones (Phillips et al. 2015a).

Radiata pine, being the dominant plantation forestry species with well developed log markets and a long history of research and development, could be seen as the clear choice for steepland afforestation. Breeding programmes have produced a 30% increase in recoverable volume over the last 50 years (Moore, 2017) and most production forest companies in New Zealand only have experience with radiata pine. Phillips et al. (2015) suggested that radiata at standard-practice clearfell rotation length and planting densities may not be the appropriate species for erosion-prone terrain where there is a high risk of storm-induced landslides and asked if we could do better in terms of the window of vulnerability. In some cases such as where it is difficult to extract trees or where storm damage history suggests that erosion risk is too high for standard practice re-establishment, decisions about what to do next might need to take into consideration the option of changing species or tactical withdrawal and reversion (Phillips et al. 2015a). Consideration of the 'window of vulnerability' is key to such decisions (Phillips et al. 2015a), and this varies with species.

Biological risk is an important factor to consider in the selection of plantation forest species. The high up-front costs of establishing plantation forests and the long rotation lengths give rise to considerable investment risk. The plantation forest industry, being one of the largest primary industry sectors, is currently considering an industry levy in order to be able to meet costs of responding to any incursion that threatens the industry. The level to which industry would fund incursions on "alternative" species to radiata pine is open to speculation, but such a levy would primarily be collected from radiata growers.


Radiata pine and Douglas fir are produced in bulk, therefore seedlings are likely to be lower cost than "alternative" species. In order to keep tree stocking up and costs down, growers may wish to interplant their selected species with radiata or Douglas fir, which would be culled later. Because of the vigour of radiata, it is not generally recommended as a nurse for other species because it tends to overshadow them before they are well enough established and old enough to be released. Douglas fir, on the other hand, makes a good nurse crop for most species listed in this report and could be left as a nurse until about ten years of age. However, it should be kept in mind that the selection ratio of crop trees is reduced by half if half of the trees are Douglas fir to be culled. Unimproved seedlings available for alternative species require high selection ratios, so generally these should be planted at high stocking rates to produce a well selected crop of good trees.

Alder may prove to be a particularly good nurse in eroded steeplands because it grows well in slip faces devoid of topsoil, fixes atmospheric nitrogen into the soil and provides leaf litter to rebuild topsoil. It also is vigorous as a young tree but "runs out of steam" allowing the crop trees to take over after being forced upward, thus minimising branch index.

Manuka has been suggested as a suitable nurse crop on erodible steeplands that also yields valuable honey and thus a short term income stream while the timber crop establishes. Manuka is naturally succeeded by climax species so such a regime would mimic natural reversion, rather than planting manuka as the primary crop and then having to maintain the species dominance through management.


Quality characteristics by species

Estimating the suitability for plantation species to control erosion on steeper slopes has not yet been refined and quantified. Subjective species ratings and quality ratings are presented here using professional judgement, with the goal of refining these as research improves our understanding of the importance of each quality and how the species perform comparatively. This method could be seen as a first attempt to develop a method for evaluating the suitability of different species for controlling erosion on ESC red zoned land identified in the NES-PF, on a long term basis.

The qualities or species characteristics that vary between species and that influence decisions on species choice are listed below.

The importance of species characteristics is presented here as preliminary ratings for each quality. Some qualities hold higher importance values than others and these importance values could be refined via evaluation studies.

Quality attributes or features

Quality Importance
Early growth rate 0.2 Early growth rate is very important in terms of the window of vulnerability. However, increased stocking rate can substitute for early growth rate so the importance of this may be negated. Radiata pine is given an early growth rate of 10 and other species comparative early growth rates are estimated.
Permanent canopy 0.2 The suitability for the species to be grown under continuous cover regimes is important for erosion control on steeper, more erodible slopes. This is only important where clearfell harvesting is not viable or where the level of landslide mitigation requires improvement.
Root decay 0.2 Root decay rate is important because it has an influence on the duration of the window of vulnerability. Species that coppice have slow root decay rates and it is assumed that for species that don't coppice, root decay is related to wood durability. That is, the durability performance of the timber is a useful proxy for root durability and thus is a measure of the time after clearfell that root reinforcement of the soil is lost.
Productivity 0.2 Productivity on erodible hill country. Productivity is based on annualised productivity rather than rotation length and therefore ignores the time value of money.
Timber value 0.1 An estimate of the timber value based on species appearance and physical properties. Timber value is assumed to be less important than other qualities, but still an important consideration when selecting a plantation forestry species for erosion control.
Coppicing 0.1 The ability to coppice from stumps that stay alive after harvesting. Coppicing is an important species attribute in terms of erosion control.
Total rating value 1 The importance of each quality characteristic, added together equal 1

Each species is rated out of ten for each quality. The total rating for the species is then calculated by multiplying the species rating for each quality by the quality importance rating for that quality, then adding these quality results together for the species:

Quality ratings for species

  Alder Beech, Southern Blackwood Cedar, Japanese Cypress Douglas-fir Eucalyptus Fir, silver Kauri Larch Manuka Pine, radiata Poplar Redwood, coast Sequoia, giant Totara
Early growth rate 8 1 7 5 9 5 8 4 1 6 n/a 10 9 6 4 4
Permanent canopy 7 8 8 10 8 6 10 10 8 6 n/a 3 8 10 9 9
Root decay rate 5 7 7 7 8 5 8 4 9 5 n/a 2 8 9 7 8
Productivity 4 2 3 8 7 8 9 8 2 6 n/a 10 9 8 7 5
Timber value 5 9 9 6 8 6 8 5 9 7 n/a 3 3 6 6 8
Coppicing 8 0 8 0 0 0 9 0 0 0 n/a 0 10 10 2 0
Total rating for species 6.1 4.5 6.7 6.6 7.2 5.4 8.7 5.7 4.3 5.3 n/a 5.75 8.1 8.2 6.2 6.0

The higher the overall rating for the species, the better it should be suited for production plantation forestry on erodible hill country. These ratings are subjective only, but serve to demonstrate that this method could be developed into a more objective tool. Rotation length was not taken into account because this is of less importance than annualised productivity in terms of species selection for erodible steepland country.

Early growth rate is an estimate for each species growth rate compared with radiata pine. A species with a score of 5 equates to half the growth rate of radiata which is 10, therefore twice the stocking is required to achieve the same level of root reinforcement over the same length of time. If the steeplands being harvested and reforested were in radiata pine and a replant stocking rate is determined for radiata that adequately mitigates erosion, stocking rate for an alternative species, to be equivalent in terms of erosion mitigation, would be represented by the equation: radiata early growth rate/alternative species early growth rate * radiata stocking rate.

Root decay rate and early growth rate together could determine the required stocking rate required to close the window of vulnerability quickest following harvesting and subsequent planting of the cutover.

Models of canopy closure and root occupancy have been developed for radiata pine at different planting densities (Marden et al. 2016). These could potentially be developed for other species to provide the same level of mitigation to that provided by radiata pine at a range of stocking rates. Other factors that might need to be taken into account include rotation length and coppicing ability because these also influence overall risk, along with harvest practice which might include continuous cover, which negates the risk of erosion following harvest.

Root grafting

Information on root grafting of individual species is mostly unavailable, but it can be assumed that trees of the same species will root graft and therefore this quality is equivalent between species. Natural root grafting is common between conifers of the same species (Cerezke, as cited in Harry and Smith, 1964) and grafts between roots of individuals of the same species is common in both conifers and hardwoods (Thomas, 2000). The frequency that root grafting occurs tends to be related to spacing between trees and is less prevalent as trees become further apart (Thomas, 2000).


Why alternative species to radiata pine?

Radiata pine is well known to be an adaptable species that can be successfully grown throughout most of New Zealand (Van Kraayenoord and Hathaway, 1986). In stark contrast, Van Kraayenoord and Hathaway (1986) contended that all other species "have much more specific site requirements", "are prone to various pests and diseases” and "are relatively untested for timber production". These issues suggest a higher level of risk with growing any alternative species to radiata pine.

Eyles (2014), on the other hand, held concerns that "the current emphasis on mass-producing a low-value timber as a monoculture will need to change", in reference to both the biological risk the plantation forest industry faces by depending on one species, and also in recognising the potential for producing higher value timber from other species. Among growers who recognise such opportunities there is a sense of frustration that research into management of alternative species over the last quarter of a century has been discontinuous, ad-hoc and insufficient, resulting in large information gaps that stifle any opportunity for profitably growing alternatives under longer-term rotations in steep unstable hill country (Eyles, 2014). It is well understood that forestry research needs to have a long-term focus and Eyles (2014) pointed the finger squarely at central government for abandoning any national co-ordination of rural sustainability during the 1990's, with the consequence being a lack of research effort into siting of alternative species. The current knowledge pool now resides mostly among farm foresters who have experimented with species and sites (Moore, 2017). This may well be considered insufficient by any serious investor considering planting large areas of these species. Despite these constraints, Moore (2017) suggested that by including carbon in the land-use equation, steep and remote land could be planted in species that produce high quality timber even if rotations were considerably extended.

Radiata forestry on steep slopes is usually considered marginal because of high harvesting costs and commodity log prices. In order to improve returns, either higher log volumes or higher log prices are required. High-value timber species offer the prospect of higher log prices that may produce annualised revenues that exceed those for radiata despite lower volumes, but the current reality is that lack of recognition of specialty timber quality and lack of consistent volumes in the market has resulted in:

  • a poorly functioning value chain; 
  • underdeveloped markets for specialty timber; and 
  • low log prices despite high retail prices for the timber (Gordon, 2014a).

Adding value to New Zealand-grown specialty timbers in order to generate demand for logs and interest in growing them may require some industry commitment in terms of product promotion, to overcome the "chicken and egg scenario" and drive interest in the species. Of course this conundrum might be viewed by some as a niche marketing opportunity. Moore (2017) suggested that "trees will grow and produce value in the future, irrespective of today’s market. Markets change, and a high quality species on a remote site has more potential than a low quality species".

Extreme environmental conditions such as high winds, poor soil structure, low available nutrients and low soil moisture content can all be present on steepland sites and for trees to thrive they must be adaptable and easy to establish (Van Kraayenoord and Hathaway, 1986).

Knowledge, of course, does continue to improve from anecdotes and experienced enthusiasts in the farm forestry community are willing to share what they have learned after "giving it a go" with alternative species. Formalising such evaluations into research trials that validate the hypotheses may or may not be necessary as knowledge evolves, but because of the inevitably long time frames involved in plantation forestry, knowledge tends to evolve slowly.


Genetic gain as recoverable volume per hectare has risen by around 30% in 50 years for radiata pine (Moore, 2017) as a result of a comprehensive and well-funded breeding programme. Meanwhile, alternative species to radiata pine and Douglas fir have suffered from an "almost complete lack of any selection for improved performance" (Gordon, 2014a). One method for overcoming lack of genetic improvement is to plant at a high stocking to provide a high selection ratio for growth and form. High tree stockings can also reduce branch index, but thinning in multiple progressive operations would be required to minimise windthrow in crop trees while ensuring adequate diameter increments. High initial tree stockings also reduce erosion risk in steeplands, so regimes could be deployed with unimproved seedlings that are nevertheless productive, but do require a higher initial capital outlay than genetically improved radiata.


Assessing risk is about identifying and evaluating those risks, which include wind, fire, snow, insect damage and diseases (Moore, J. 2014). That risk varies between species and sites and radiata pine is by no means immune to these; for example red needle cast outbreaks and periodic snow damage in some localities. In terms of biological risk, Van Kraayenoord and Hathaway (1986) questioned the advisability of planting monocultures and perhaps the opportunity steepland forestry offers to the sector is how to do things better.

Strong wind can break tops out of radiata pine and detrimentally affect both growth rates and form (Moore, J. 2014). Log grades can be lower from windy sites both because of malformation and also larger branches (Moore, J. 2014). Radiata grown in windy sites can also produce wood with less stiffness and more resin pockets compared with trees grown in more sheltered sites (Moore, J. 2014). Perhaps species better suited to windy sites could produce better returns than radiata in exposed steeplands. Unfortunately, evidence of the relative wind-firmness of alternative species to radiata pine is mostly anecdotal (Moore, J. 2014), albeit with some species clearly demonstrating high resistance to wind and others less so.

Management decisions also influence risk of wind damage to plantations. Thinning increases the risk for the residual stand to suffer from wind damage, especially where thinning is late and trees are tall and thin (Moore, J. 2014). Harvesting also increases the vulnerability of residual trees to wind damage that were left after harvesting trees adjacent to them (Moore, J. 2014). Increased rotation length also may increase the risk of storm damage (Moore, J. 2014). On steepland slopes, high initial stockings of trees may be required to shorten the window of vulnerability, especially with slower-growing species. These plantings may require multiple progressive thinning operations to reduce the risk of windthrow and produce sufficient crop tree diameters. Cost effective methods for progressive thinning to waste include ringbark thinning (Satchell, 2018).

Site requirements

Site requirements vary considerably between species. For example, pine species that are more suitable than radiata to colder, drier areas of the South Island include Pinus nigra subsp. LaricoP. muricata (blue strain) and P. ponderosa (Van Kraayenoord and Hathaway, 1986). Within species, determining suitability of provenances may be necessary to ensure climatic compatibility (Van Kraayenoord and Hathaway, 1986). Where a site has extreme variability in terms of soil structure, moisture, temperature and light intensity, an adaptable plant species is required, or alternatively a good understanding is required of the species adaptability to site (Van Kraayenoord and Hathaway, 1986).

Species with weed potential should be avoided for soil conservation plantings (Van Kraayenoord and Hathaway, 1986), which precludes planting of Douglas fir, larch and some pine species where these have potential to spread as wildings. Site also influences a species weed potential, which includes neighbouring land's vulnerability (Ledgard, 1999). Decision support systems are provided by MPI under the NES-PF, including the Wilding Tree Risk Calculator, which rates the spread risk using the following indicators:

  • Tree species type (spreading vigour)
  • Palatability of tree species to stock
  • Site location
  • Site characteristics
  • Existing vegetation on site

See the MPI Wilding Tree Risk Calculator.

Root habit may be important for preventing erosion. Species that are desirable for erosion control have a root system that develops rapidly, is extensive and binds soil together to resist tensile stresses (Van Kraayenoord and Hathaway, 1986).

Establishment of trees on seasonally dry sites can be difficult. This may require complete elimination of weeds and grass competition from around the tree during its establishment by using herbicides (Van Kraayenoord and Hathaway, 1986). 

Colonising plants are the most suitable for rapid revegetation of eroded soils where topsoil is depleted, especially species that produce new plants from suckers or root fragments, or plants that fix atmospheric nitrogen to enrich the soil ecosystem (Van Kraayenoord and Hathaway, 1986). Tolerance of stem burial and development of adventitious root systems and other means of regeneration after disturbance offer improved erosion-proofing (Van Kraayenoord and Hathaway, 1986).

Native timber trees generally have slower growth rates compared with exotic species, so reinforcement of soils would be delayed (Jones et al. 2008).

Timber production

Commercial plantation forestry investment demands a species that maximises production of logs that are of a quality and value that provide adequate returns to the investor. However, a survey of small forest owners found that primary drivers for forest investment also include best land use (West and Satchell, 2017), suggesting that some owners may be willing to trade off returns for environmental outcomes. Where erosion control is the primary objective, species selection criteria may be different from production-based plantation forestry.

Converting logs into value-added products that command high prices in the market and in sufficient volumes that fulfil market expectations may be a prerequisite to generating interest in planting high-value species in New Zealand (Gordon, 2014a; Satchell, 2015). Moore (2017) emphasised that the lack of a current market for locally grown specialty wood is only an education problem and that markets do change (Moore, 2017), so should deployment precede market development?

For some species market demand for logs has increased dramatically in the last few years as the export log market has matured and log exporters buy most available species, often at a premium (A. Laurie, pers. comm). Trost (2005) found the local market for logs to be much more sensitive to log quality and limited in capacity, with a preference for logs that produce lengths of timber that are clear or timber with tight knots for discerning appearance applications. This may be changing as local specialty mills compete with the export market for logs and market demand for knotty appearance timber grows (J. Fairweather, pers. comm). Although "the market for special purpose timbers is relatively small, and there is only a limited number of sawmillers who are experienced in handling these species" (Trost, 2005), this may well change within a rotation length.

Other impediments constrain market demand for high-value timber species in New Zealand. Our performance-based building code has a history of compliance paths offered for radiata pine and Douglas fir, but not for other species, despite their superior properties. By not being code-complaint, a range of products and applications for specialty timbers are effectively removed from the market, negatively impacting on demand for timber and prices for logs (Satchell et al. 2016). Developing markets for high-value specialty timbers could of course be viewed as a hurdle rather than a brick wall, with current industry efforts attempting to address this with the Specialty Wood Products programme led by the forest industry in partnership with MBIE.

Disclaimer: The opinions and information provided in this report have been provided in good faith and on the basis that every endeavour has been made to be accurate and not misleading and to exercise reasonable care, skill and judgement in providing such opinions and information. The Author and NZFFA will not be responsible if information is inaccurate or not up to date, nor will we be responsible if you use or rely on the information in any way.


Farm Forestry - Headlines

Article archive »