Ecosourcing in Aotearoa

MATT McGLONE & PETER HEENAN (Manaaki Whenua-Landcare Research, Lincoln)

WHY RESTORE? AND HOW?

Native plants have been shuffled around New Zealand since the first Polynesian settlers arrived. Māori cultivated karaka well south of its Northland range for its carbohydrate-rich kernels. European settlers established ornamental gardens and amenity plantings and sought out suitable native plants, spreading many of them well outside their range. However, only in recent decades has deliberate planting of natives to restore degraded or vanished vegetation become commonplace. As restoration has an important role in enhancing New Zealand biodiversity, critical thinking is needed about why and how to do it.

We first need to understand why restoring vegetation as close as possible to the original landcover is so important. There are four main reasons for this being an appropriate goal.

• First, the original species are likely to grow well at the restoration site.

• Second, such a restoration is more likely to lead to positive interactions with remaining original ecosystems and will not introduce species likely to turn into aggressive native weeds. It will be more open to colonisation of other organisms in the region – insects, spiders, fungi – which often will be well adapted to the regional vegetation. 

• Third, adopting a policy of restoring original vegetation is a powerful way of co-ordinating individual restoration effort.

• And finally, there is the pleasure of knowing that a restoration closely resembles the original, and that it is in some way authentic.  

The best approach is to let nature do the restoration. Where intact native vegetation is close by, many species will disperse naturally. Exclusion of mammalian herbivores, suppression of weeds and protection against fire will be essential, so this is not necessarily a much cheaper option. It will generally be a slower process, because seeding or planting permit skipping of successional stages. However, the advantages are great. The self-selection of arriving plants ensures that they will be very well adapted to the site. The resultant vegetation will look ‘right’. Try as we might, planted stock nearly always has an unnatural look as replicating the random patterns of naturally dispersed plants is difficult. Also, planted stock often is less genetically diverse.   Finally, the pleasure and intrinsic interest of watching a natural succession unfold is great. However, natural regeneration can be very slow and stalled successions common.  Therefore, hybrid efforts in which natural succession is enhanced by planting of preferred species – often podocarps – is an option.  

But what if natural regeneration is not feasible? Over large areas of New Zealand – particularly in the drier lowlands – a natural succession approach will not work on a time scale less than centuries. Often only small patches of original vegetation persist, and these are often atypical of what was once common, and lacking sensitive or rare species. Historical documents, palaeoecological findings and modelling of current vegetation can give a broad idea of what the vegetation composition was like in the past. However, a comprehensive and all-inclusive analysis usually eludes us. Too many species are gone without trace. In these instances, much thought and investigation must be put into deciding what was the most likely original vegetation cover. Consideration of soil, disturbance regime and climate and comparison with similar areas elsewhere will be needed. And this is where ecosourcing becomes important.

ECOLOGICAL GUIDELINES FOR ECOSOURCING

Early restoration attempts often paid little attention to the provenance of a given species: the main issue was whether it was a native and available.  Concern over inappropriate restoration efforts has grown and, especially, the movement of native species outside of their natural range, and procurement of plant seed from areas remote from the restoration. The Department of Conservation and local authorities now provide guidelines to assist potential restoration efforts. These guidelines stress the need for restoration stock to come from nearby sites. In a recent publication(1), we re-examined the basis for ecosourcing in New Zealand.  We argued that the advice commonly given was too restrictive.

Current ecosourcing guidelines tend to be focussed on distance, with further away being always worse than nearby. Some guidelines suggest a kilometre range (usually just a few km). Others suggest the Department of Conservation New Zealand Ecological District scheme(2) as a suitable template (Figure 1).

This scheme finely divides the landscape into 286 districts and 85 ecological regions. The districts are based on the concept that distinctive landscapes (according to geology, topography, soil and ecology) can be identified.  The scheme was introduced to provide a holistic basis for biodiversity protection at a landscape scale. It serves this purpose very well indeed. However, it is based on subjective criteria and has no underpinning quantitative data concerning the distribution and abundance of the species therein.

Species largely ignore the district boundaries and so it has limited value for ecosourcing purposes. If plant-relevant data is needed to assist with environment-matching, Land Environments of New Zealand is more useful(3). But as each species is unique, there are no short-cuts.  Decisions about where to find suitable plant material should be based on the distribution and ecology of the species under consideration.

Genetics also is an important consideration. Populations with limited gene exchange tend to become differentiated from neighbouring populations while those with strong gene flow may be little differentiated throughout the country.  Limited gene flow, especially in small, fragmented populations can lead to inbreeding and deleterious outcomes for plant health. On the other hand, indiscriminate movement of plants can lead to obliteration of the original landscape genetic pattern. Our view is that,  if sufficient attention is paid to the ecology and distinctiveness of the species involved as per our guidelines below, increasing the genetic diversity of a population at a restoration site should take precedence.

SELECTING YOUR SPECIES

For ecosourcing purposes, plant species fall into three broad groups.

• Abundant and widespread at a regional or national scale. Just a handful of species in a structural group (successional, overstory, understory, ground, climber) will make up the vast bulk of the plants present in any natural ecosystem.

• Species that are nowhere common or abundant but are widespread.

• Species with extremely limited ranges.  

One set of ecosourcing guidelines will not cover all three groups.

Widespread and abundant species are well connected throughout their range through seed dispersal and pollen flow. Interchange of genetic material has been frequent, at least until the disruptions caused by human settlements. We now have a number of comprehensive genetic analyses of populations of widespread and common species (for instance, mānuka, kānuka, kōwhai group, rāta, silver beech, the red beech group, lancewood etc) and can make some general conclusions.  Genetic differentiation is mostly low throughout their ranges and there is relative uniformity over wide areas. Genetic discontinuities in species tend to converge in the same general areas.  For instance, a long-established discontinuity is the Kauri Line at around  38°S, where numerous northern species find their limit.

Based on this genetic data, biogeographical discontinuities, and a consideration of the history of the last glacial-interglacial cycle, we have recently provided a map indicating the broad areas within which material can be freely sourced in most cases (Figure 2). Such a map is a guide only.  The zone borders should be regarded as broad transitions rather than strict demarcations. However, we would regard it as unwise to source material from deep within one zone for planting in another in the absence of some compelling evidence of compatibility.

Sometimes an otherwise widespread species will show major gaps within its range that have no apparent or obvious environmental cause. Beech is the classic example – the Westland Beech Gap extends over many kilometres on the west coast of the South Island - but many widespread species have range gaps of all sizes (Figure 3).

How should these be treated?

The purist approach would be to respect them. However, we should also be guided by the species itself. If it is a long-lived species that is known to be poorly dispersed – such as the beech species – preserving the original distribution pattern would seem to be the most authentic way of approaching the problem. Silver beech north and south of the Westland beech gap come from different lineages and the gap is clearly of long-standing, probably twelve thousand years or more.  On the other hand, in the case of short-lived, well distributed species (for instance swamp maire) gaps have most likely been caused by recent historical events. Filling in gaps in the range of such a plant species is unproblematic.

 Once a decision is made to use a widespread species, the question is where to get it from.  Despite gene flow, local site adaptation always occurs, and this is important. More attention then should be paid to site and environment in sourcing plant material than geographic proximity.

 Very distinctive local entities with marked environmental preferences called “ecotypes” may arise within in a freely interbreeding species. In New Zealand, the distinction between ‘ecotype’ and ‘species’ in some plant genera, such as the snow grasses, the red beech group, kōwhai, coprosmas etc, is somewhat blurred because of recent gene flow across populations. A prime example is the well-studied kānuka complex, where ten ecotypes (initially identified as varieties or species) have been named (Heenan et al. 2023).  Some of these ecotypes are restricted to specialised habitats such as frosty, dry sites, consolidated sandy soils along the coast, or thermally heated ground (Figure 4).  Ecotypes and site environment are therefore of great importance. The ecology and inferred history of the individual species should always be paramount over any rule of thumb.

The second category is the “widespread but nowhere very common” group, where current/recent interchange of genes is unlikely. Are such species naturally rare with a spotty distribution, or do they result from recent habitat disruption caused by human settlement? In the first case, they should be not used in a restoration; in the second, there is no reason why not. Genetic information can help distinguish between these options but, in its absence, the nature of the habitat can help. Sparsely distributed species often favour an uncommon environment, examples being limestone outcrops or thermal areas. Clearly these should only be used in restoration of these specific environments.  On the other hand, duneland, wetland and semi-arid vegetation has been severely reduced in extent by fire, drainage and various forms of development.  It is a safe assumption then that these now fragmented ecosystems once supported species with a more continuous distribution.

Our third group is species with highly restricted ranges. Tecomanthe speciosa – a vine with but one individual on Three Kings, Metrosideros bartlettii with only 31 trees at Cape Reinga are the most extreme examples, but there are many more.   Such species are mostly rare for a reason – usually stringent environmental requirements or competitive pressure from better adapted plants. They should be left out of restoration plans unless plant rescue is the point.

CONSIDERATIONS FOR RESTORATION

All restorations sit on a spectrum from gardening to plant-and-forget. Any restoration that needs constant and long-term intervention will eventually fail. A restoration that does not include sufficient space for turnover and regeneration will always be at risk. It is inevitable over time in small restorations that species will be lost through chance, and that genetic diversity will reduce. A restoration should therefore be as large as the budget can sustain and include varied topography, diverse soils, and waterways if at all possible.  It needs to be buffered against the possibility of fire, floods, and slips, or large enough to be resilient in the face of such events.

However, exceptionally large restorations absorb considerable resources and eventually will have to become a community asset if they are to persist. That is, they need financial as well as physical buffering. Community involvement and access is therefore essential if a project is to have lasting success. Good examples have been some wetland and riverine restoration efforts, and these have mostly had considerable local or regional council support.  

Regarding ecosourcing itself, where possible:

• Collect seed from 50 or more unrelated plants in the wild to maintain genetic diversity.

• Collect seed from larger rather than smaller populations and avoid small, fragmented populations where possible.

• Take account of ecotypic variation

• Gather soil as well to inoculate seedlings with appropriate mycorrhizae.

• Record provenance of seed collections, and number of source plants.

Access to plants for seed collection is a key issue. The larger commercial nurseries have their own sources, often on private land, and they can be cagey about giving details.

Information on the exact provenance or diversity of the stock therefore can be sketchy. If commercial material is used, there is always a possibility that siblings or close relatives are used in a planting – something that is genetically highly undesirable. Getting permission to source seeds from public land can be a rather tiresome process. This is not good. And it is unjustified. Plants produce seed well in excess of their needs and, as long as they are not damaged unduly while being harvested, little if any harm is done. A far more permissive attitude by both private and public landholders is called for.

Finally, climate change needs some consideration. The New Zealand climate will become warmer overall, and wetter in some districts, and drier in others.  Increased variability will lead to more droughts and more flooding. Many species in the north of the North Island currently constrained by cool temperatures may naturally move south. Some, like pōhutukawa, are doing so anyway as a response to plantings and naturalizing well south of their northern limit. Would it be wise to future-proof a restoration by including species naturally found only in warmer areas?

Our advice would be no. New Zealand plants are well adapted to a variable climate. Most lowland and montane plants are resilient and can grow under a wide range of conditions. The only justification for ‘future-proofing’ plantings would be if the resident vegetation is under or likely to come under damaging climate stress.  There is no definitive evidence yet that New Zealand vegetation is suffering from such stress. Although drought and frost events have caused tree deaths over wide areas, such tree-killing events seem no more frequent now than in the past. While climate change is inevitable, exactly how it will be delivered and when is still very uncertain and we are best to leave well alone.

Acknowledgements

We thank James McCarthy for reviewing this article and for Figures 1 & 3.

References

1. McEwen WM. ed. 1987. Ecological regions and districts of New Zealand. Wellington: Department of Conservation.

2. Heenan, P.B., Lee, W.G., McGlone, M.S., McCarthy, J.K., Mitchell, C.M., Larcombe, M.J. & Houliston, G.J. (2023) Ecosourcing for resilience in a changing environment. New Zealand Journal of Botany, 1-27.

3. Leathwick, J.R., Wilson, G., Rutledge, D., Wardle, P., Morgan, F., Johnston, K., McLeod, M. & Kirkpatrick, R. (2003) Land Environments of New Zealand. David Bateman, Auckland.

4. Heenan, P.B., McGlone, M.S., Mitchell, C.M., McCarthy, J.K. & Houliston, G.J. (2023) Genotypic variation, phylogeography, unified species concept, and the ‘grey zone’of taxonomic uncertainty in kānuka: recognition of Kunzea ericoides (A. Rich.) Joy Thomps. sens. lat.(Myrtaceae). New Zealand Journal of Botany, 1-30.