Report: Eucalypt timber species evaluation
By Angus Gordon, July 2007.
This project set out to test the survival and subsequent growth of a group of eucalypt species that have the potential for the production of naturally durable solid timber. The genesis of this project was the current and on going interest showed by farm forestry association members in these particular species, and also those people who see a potential niche market in the production of naturally durable posts for use in the viticulture industry.
Sustainable Farming Fund Project L03/007
Submitted: July 2007
Submitted by: The New Zealand Farm Forestry Association Eucalypt Action Group.
Project manager and report writer: Angus Gordon
Supporting Ensis staff:
Thanks must also go to the fifty-five farm foresters who gave so much time and money to provide the sites where these trials took place.
A special thank you must go to the Sustainable Farming Fund, as without their ongoing financial support, this project would never have happened.
The focus was primarily on the stringybark group of eucalypt species, and a few other non stringybark eucalypts species that have very similar wood properties. Two species, E. fastigata and E. maidenii were included as indicator species due to the large amount of trial data that already exists for them both in New Zealand and world wide. The stringybark group was chosen due to their inherently good sawing properties, good drying characteristics, general good health and natural durability.
Presently members this species group, which numbers in excess of 25 different identifiable species are grown over a wide number of sites in New Zealand, with some occurring on sites that were previously thought to be unsuitable for growing eucalypts, let alone the members of the stringybark group. This was contrary to the generally held consensus, which was that stringybarks were only suitable for warm, lower altitude North Island sites, and warm sites in the upper South Island. High altitude sites in the North Island and sites in the South Island that experience very cold conditions were considered to be unsuitable for these species, but it has been found isolated examples of stringybarks growing in unfavourable locations do exist .This presented the potential stringybark eucalypt grower with a problem, to plant only in areas that currently are regarded as suitable, which limits their opportunity, or to plant in areas that are currently not regarded as suitable based on a few isolated examples of species success, but with higher risk of failure.
To try and ease this decision for potential growers this trial was proposed, which set out to test a uniform range of species over a broad range of sites, and to try and find which ones are better suited to the many generalised climate areas of New Zealand.
It was hoped that this project would provide much clearer siting recommendations for the stringybarks, and with the ongoing growth measurement of the surviving trees, a set of recommendations will be able to be produced matching individual species, or groups of species to particular regions of New Zealand.
There is the potential to expand this trial series to include other eucalypt species that were not tested in this trial, where those species show potential for providing durable timber, or other properties, and where they had previously been unrecognised by the current tree planting fraternity. This could be an ongoing theme for the trial hosts that wish to commit to future use of their sites.
This project will continue beyond the presentation of this report to the Sustainable Farming Fund, and updates will be posted on the NZFFA website.
This project by virtue of its very nature, supplying plant material to 55 sites nationwide has been logistically challenging to say the least.
Firstly the trial sites had to be sourced. This was achieved by the strong network that exists throughout the farm forestry association. A series of six meetings were held during 2003 in Whangarei, Puhoi, Rotorua, Masterton, Blenheim and Balclutha. These were held in conjunction with field-days, a sustainable farming fund tech transfer project and other forestry related events and were supported by farm forestry association members, the staff of ENSIS, formerly the Forest Research Institute at Rotorua, and members of the general public. An information pack was sent out to over 100 people interested in the project, to outline the responsibilities and expected costs that would be incurred.
Plant material and sites:
A total of 55 sites were eventually offered for planting, with the majority of them planted in the spring of 2004. The plants were container grown by the FRI nursery at Rotorua, with seed obtained from Australian seed sources. There had to be enough plant stock available to supply each site with 150 trees, 15 for each of the 10 species that were to be planted at each site, which meant a requirement to grow in excess of 10,000 trees. These trees were purchased by each trial host.
Due to susceptibility of the seedlings of some of the species to fungal attack, there was a total failure of one of the original trial species and reduction in the numbers for some of the others. Trees were able to be sourced from another nursery to make up the numbers, from another project that was running concurrently in another part of the country that was using the same mix of species. Many of the trial blocks planted in the spring of 2005 had plant material sourced from both nurseries.
Trial hosts were sent a detailed instruction sheet describing pre planting site preparation, and a reply post return form to indicate when and where the plant material was to be delivered and the intended planting date. All trial plants were generally lifted from their containers in the week prior to planting, packaged and couriered to trial host. There was considerable communication between the trial manager and the nursery manager at FRI during the planting season to ensure that all the seedlings were delivered on time.
The Actual trial sites were ideally rectangular in shape and measuring 30 metre by 45 metre in dimension. These dimensions were to facilitate the planting of the ten species row-plots so that they aligned up and down the slope, with a 3 metre spacing between each row. Each species row-plot contained fifteen seedlings with a 3 metre spacing between each seedling. In reality the shape of the trial areas did not always match the ideal, due to the availability of appropriate parcels of land.
On site management:
The site preparation was left entirely to the trial hosts involved. In general they were required to fence the trial area, do a pre-planting spot spray of the site, plant the species row-plots in the designated order and carry out any weed control post planting as required, and prevent damage from wild and domesticated animals. On three sites deer and goats proved impossible to exclude, leading those sites to be abandoned. The planting followed a randomised planting plan that was unique to each site.This plan was produced by Ensis staff and was designed to avoid any successful species continually dominating the same adjacent neighbouring species. While good in theory, in terms of not having any particular species in the same position relative to its neighbours across all sites, it did in practice cause some problems with the interpretation of the instructions, which were re-written by the trial manager early in the 2004 planting season. Once the establishment of the trial sites was completed the only other requirement for the trial hosts was to assess the individual species plots for survival, which took place in the autumn of 2007 . That data has been recorded and interpreted for this report.
This has been achieved in two different ways. Firstly the trials were assessed and reported by the trial hosts using a survey form that was sent to them. Secondly some sites were inspected either by Ian Nicholas from Ensis, or the trial manager Angus Gordon to cross check the data or in some cases to interpret trial sites where unforeseen events lead to the trial host not being able to supply the data.
Other site variables such as mean annual rainfall, altitude, aspect of the trial site, soil type and any events that may have influenced the success or failure of the trial were recorded.
The trial manager has recorded the survey data and collated the results in a Microsoft Excel spread sheet.
Some trial hosts were contacted by telephone to query particular issues that arose on their sites, and to hurry along the late respondents. At the time of writing the data from ten sites is still outstanding, with that data expected in the near future. As this late data arrives the trial results that are posted on the farm forestry website will be updated.
The data which was initially arranged alphabetically using the trial host surname was re-arranged so that sites with similar survival performance could be grouped together for further analysis.
The trial sites have been grouped into three broad survival zones, which were generally of similar altitude. The eucalypt species that were trialled were divided into three species groups based on their relative survival across the three previously mentioned climate zones. These survival zones and species groups will form the basis of the discussion of the report outcomes.
The analysis of the results of the trial sites led to the sites effectively being divided into three broad survival zones, and the eucalypt species being divided into three broad survival groups. Those groups are displayed in table 2 and also described below.
A set of XY scatter plots showing the group survival performance is displayed in the appendices.
Table 2: Species group descriptions
Group 1 contains three species. C. maculata, E. microcorys and E. pilularis. They have proved to be the most tender with respect to survival, and have only performed well in warm and generally benign climate conditions. This certainly backs up the general consensus among some farm foresters with respect to C. maculata and E.microcorys, that is that neither of them cope at all well with frosts and in general like very mild sites, and that E. pilularis is only slightly more site tolerant. Only one of these three species is from the stringybark group, that being E. pilularis. C maculata is from the bloodwood group and E. microcorys is from the tallowwood group. It must also be noted that E. tereticornis has provisionally been included in this group, but due to the limited number of trial sites (4), few conclusions can be drawn about its performance.
Group 2 contains seven species. E baxteri, E. cameronii, E. globoidea, E. laevopinea, E. longifolia, E. macrorhyncha and E. muelleriana. All of these species are stringybarks except for E. longifolia (Woolybutt) which belongs to the eastern blue gum group of species. E. cameronii was planted on a limited range of sites (4) so few conclusions can be drawn about its performance.
This group had superior performance over a larger range of sites than group one, with its range pushing up over 500 m of altitude under some circumstances, but with the highest survivals being at the lowest altitudes and in what appears to be most benign climatic areas.
Group 3 contains five species, E blaxlandii, and E. youmanii, both stringybarks, E. fastigata and E. obliqua, both ash group eucalypts, and E. maidenii one of the southern blue gum group. The members of group three were the hardiest that were tested in this trial, and had good survival at some sights right up to 700m of altitude , with one site performing well at just under 900m in the central north island. It is worth mentioning that the most consistent survivals occurred at the lowest altitudes and again in what appears to be the most benign climatic areas. E. blaxlandii was planted in a limited range of sites so some caution is needed when interpreting the results for this species.
Table 3: Percentage of species row-plots with surviving plants, within groups and zones
Table 4: Percentage of surviving plants, within groups and zones
Table 5: Individual species survival percentages within groups and zones:
Note: ( * ) indicates that there are 3 sites or less in the sample set
(np) denotes that this species was not planted in that particular zone
Group one species, C. maculata, E. microcorys and E. pilularis all performed moderately well in survival zone A, with a range of survival percentages at the trial site level ranging from 40% up to 100%. The average survival across the zone being 61%.
As we move from zone A and into zone B, the average survival percentages drop and we start to see complete species mortality at some sites. The average survival percentage has lowered to 12% including nine trial plots that completely failed. The results for group 1 species vs zone C show the ultimate result of pushing a group of species too far out of their normal comfort zone, 100% mortality!!
In summary the three group one species have performed well in the most benign of the survival zones, but have struggled in both of the other two.
A copy of the XY scatter plots for the group one species results are in the appendices.
Group two species, E. cameronii, E. globoidea, E. laevopinea, E.baxteri, E muelleriana, E. macrorhyncha, and E. longifolia all performed well in survival zone A. The average survival for this group was 78%, and a relatively even performance across the range of altitudes present was observed, from altitudes near sea level through to 250 m above sea level. As you move through the results from zone A and into zones B and C a trend of decreasing performance begins to show up with average survival percentage dropping to 31 % in zone B where the first total species row-plot mortalities start to show. Once the results for zone C are observed it shows us that most of the species are struggling to survive with 13% of species row-plots having surviving plants, and an individual seedling survival of only 2%.
In summary, group 2 species have performed very well in zone A, but are struggling to perform in zone B with higher levels of mortality. They are well out of their preferred range in zone C as indicated by their almost total mortality.
A copy of the XY scatter plots are in the appendices.
Group three species, E. maidenii, E. fastigata, E. obliqua, E. blaxlandii and E. youmanii all performed very well in zone A, with average survival being around 83% and no obvious performance trends due to increasing altitudes being observed. As we move from zone A into to zone B we can see a steadily decreasing level of performance with average survival level dropping from 83% down to 61%. It is in zone B that the first species row-plot mortality occurs. This group of species only begins to show excessive mortality in zone C with the average seedling survival being 12%. There seems to be no direct link between altitude and mortality for this group of species in this zone suggesting that factors other than altitude are at play here, with the most likely candidate being the prevalence of hard frosts , especially out of season ones in this zone.
In summary the group 3 species have performed excellently in zone A, and well in zone B, and general poorly in zone C.
So far in this report we have described the three species groups quite adequately, but the actual definition of the survival zones has been a little harder to define.
Some of the range of survival observed will almost certainly result from the actual trial hosts themselves, and the way that they managed their sites. This should not be interpreted as a criticism, as they have on the whole done a fantastic job. It is more to do with the timing of jobs such as planting and weed control, and the actual site locality that they chose to put the trial into. A large part in defining what makes up the survival zones will be where in the landscape each trial site lies in respect to exposure to wind, sunlight, frost drainage, the length of the growing season, the maximum and minimum daily temperatures especially in summer and probably most importantly the number of frost free days per year, or in the case of the latter, the chance of getting an out of season frost when the trees at a specific site are actively growing. Out of season frost can be quite devastating to the survival of many eucalypt species, even when the trees reach adulthood if the frosting is severe enough.
The Ensis staff supporting this project have been able to source from NIWA a series of nationwide climate maps for median summer and winter air temperatures, and also the median number of frost free days per year. It is intended that we will overlay the position of our trial sites on this type of resource material so that we can better describe the common factors that exist within our survival zones. We hope that in the future we can accurately predict areas where groups of eucalyptus species with similar site requirements will not only survive, but also thrive, as ultimately the growers of these trees want live trees that are big, healthy, fast growing and profitable.
At this late stage of this species survival trial there is insufficient time or money to do this part of the analysis but it points to a way to get further information out of this data set, especially when the growth rates of the sites that have sufficient live trees are examined.
We have a total of fifty five trial sites nationwide of which approximately forty five have a good population of surviving trees. The next part of this project will be to develop site profiles that match the survival and growth that is being exhibited in the three survival zones that have been identified in this report.
The matching the exact trial site locations to the NIWA climate data set will be important for the extension of this work, thus giving a greater degree of predictability for future eucalypt planters. It is intended that the data set matching will be finished before the end of 2007 and that the height and diameter growth will be collected from the autumn of 2008 onwards. This will give us a fairly robust picture of what sort of survival and growth can be expected across a range of New Zealand sites for the species that we have tested and a range of species that have similar siting requirements.
The end use of such data will be to identify and plant the most successful species for any particular location, thereby developing a local plantation resource for future utilisation.
As far as the end of this project is concerned, well... it probably doesnt have one, as there is no reason why for as long as the trees are alive and the trial hosts are agreeable, that information cannot be collected from them well into the future.
This report will be available on the Sustainable Farming Fund website, and the New Zealand Farm Forestry Association website. It will also be incorporated into the electronic version of the Best practice for growing eucalypts handbook that is presently being prepared by Ensis staff and Farm Forestry Association members. This report will be presented to the NZFFA Eucalypt Action Group at their annual weekend field-day in September 2007.
A poster version of the highlights of this report will be available for presentation at the farm forestry association stand at the mystery creek agricultural field days, and at the "Plantation Eucalypts for High Value Timber" conference in Melbourne, Australia in October 2007.
The staff of Ensis in Rotorua will be making use if this report and its raw data whereever it is useful for their ongoing research projects.
Lastly but by no means least, all the trial hosts who provided sites for the trials will get a copy of this report.
Updated A.W. Gordon project manager 24/07/07