The NZ Farm Forestry Association eucalypt action group, through SFF support has established a nationwide series of eucalypt species trial plantings (research packs), covering the breadth of the country. With 56 sites included in these plantings farm foresters are in a unique position to contribute a powerful data set for GIS based siting recommendations. These research packs and other known species trial plantings will be visited, with site and growth variables measured. Evaluation of data using geographical information systems (GIS) will provide improved species and siting guidelines while also building baseline data necessary for more sophisticated decision support systems (DSS) to be developed in the future.
There is a lot of general information on siting eucalypt timber species across New Zealand, much of which confuses the average tree planter as it is based on broad generalisations, anecdotes and not specific site information. With nearly one third of all forest in NZ owned by small forest owners with less than 40ha, individual owners do not gain experience in matching appropriate species to their specific sites. Many of these small forest owners, often based on marginal land are interested in developing a diverse species mosaic across their land and explore the benefits that useful eucalypt species can provide. The benefits of eucalypts are high carbon sequestration rates, providing durable timbers for on farm use, and the provision of locally sourced, hard, strong, high value, naturally durable timbers to local industries that require these characteristics in their products. Matching species to site is critical to successful establishment.
The assessment for survival and productivity of the farm forestry research pack plantings, across the range of species and sites in the initial trials, will develop a matrix of site characteristics and species performances, that will help develop siting maps and to provide confidence for the diversification of planted forest species onto new sites across NZ.
Currently, NZ imports in excess of 20 million dollars worth of high value timbers primarily for the manufacture of house cladding, door and window frames, flooring, furniture, power pole cross arms, railway sleepers and other high value applications. Almost all of this imported wood could be substituted with NZ grown product with the same performance characteristics, thus providing business opportunity for local growers and industry.
The limitation in the NZ context for the growing of eucalypt hardwood species is that the specific site characteristics that are required for adequate survival and economic growth performance have only been quantified in a logical manner for a very narrow range of species. These are commonly non durable species that are used for pulp manufacture and export chip and are of little interest to small growers due to the low value associated with these commodity products
GIS mapping will allow the matching of measured trials against a well defined set of environmental factors that are limiting for the survival and growth of all eucalypt species
1. There were 56 eucalypt farm forestry research pack plantings established during 2004 and 2005. All of the surviving trials will be visited and the following data collected: site location; tree survival/growth, topex, aspect, slope, soil description.
2. Further site information; rainfall, mean daily temperature, frost patterns and potential evapo-transpiration and drainage will be provided by national data sets, LENZ and NIWA.
3. Data on failed sites is to be collated by survey.
4. Evaluation of the data using GIS maps and the national experts in the team will help identify key parameters for the development of siting guidelines for key farm forestry eucalypt species across New Zealand.
5. Data evaluation will have a regional focus to provide meaningful outputs for those regions that have distinct performance differences.The outputs developed from this project will also have potential to contribute to future Eucalypt DSS, particularlyasone of the species used as a benchmark in the research packs was E. fastigata, a species that has a large amount of professional research behind it.
6. Reporting of the findings will use the NZFFA website, “Tree Grower” magazine, “Eucalypt Action Group” newsletters, field days, in association with the “NZ Dryland Forest Initiative”, the “Trees on Farms” workshop series, and on the SFF website.
Over this past period, all sites that were to be evaluated were identified. These were primarily sites that still had live trees from the earlier Eucalypt survival trial SFF L03/007. All sites were visited and the following individual tree measurements taken: Tree height, tree stem diameter at breast height (DBH), stem form (straightness). The following site characteristics were also recorded at each trial area: GIS coordinates, TOPEX (a description of where the site sits in the landscape with respect to exposure or shelter from the surrounding topography or other vegetation), soil type, rainfall, slope, aspect, vegetation (competition), tree health, insect / animal browse.
A total of 56 trial sites have been assessed, of these 45 still have live healthy trees. Of the 11 sites that have no live trees 3 were lost due to animal browsing (goats) and at the other 8 sites lack of tree survival was due to the extreme climatic conditions that these sites have experienced in the 8 years since planting. Out of season frost and or extreme frost were the key climatic drivers in the failure of the trial plantings. The environmental characteristics of the failed sites are important in defining the climatic boundaries, when growing the selected species that were involved in this trial.
These failed site data sets rank equally alongside the best performing trials sites in terms of their relevance to predicting growth performance of the trialled species. Data for these failed sites has been obtained either by a telephone interview of the owner, or a site visit, using GIS tools and Google earth. A postal survey was not ultimately required for failed sites as there were only a limited number of sites that were not visited by the data collection team that required further contact. Due to the fact that the original site owners needed to be all contacted by telephone, the survey was conducted orally over the telephone whilst other site details were checked. This proved to be a more efficient mechanism as live interaction between the trial site owner and the project manager was very valuable, especially when Google earth was used to pin point the exact location of the trial site and other important physical features that could have impacted on the trial site performance
On Friday November 16 a full meeting of the project team was held. The meeting was attended by the A Gordon, I Nicholas, R May, P Millen, D Satchell, G Milligan, G Flemming, and D Hocking. In summary the project team was briefed on the data collection methods, the raw data manipulation and what raw data was then used for the GIS evaluation. The data entry and GIS evaluation has been carried out by Roger May. He addressed the other team members and took us through the process and how he has arrived at the results that he did. The results for all 17 species that were present at the trial sites were included. The species are as follows: C. maculata, E. baxteri, E. blaxlandii, E.bosistoana, E.fastigata, E.globoidea, E.laevopinea, E.longifolia, E.macrorhyncha, E.maidenii, E.microcorys, E.muelleriana, E.obliqua, E.pilularis, E.quadrangulata, E.tereticornis, E.youmanii.
The mean values for height growth of each species on each site, was plotted against each particular site descriptor. The site descriptors are as follows: latitude, altitude, mean annual rainfall, frost free days, February maximum temperature, July minimum temperature, land use capability class, soil PH, soil type, potential rooting depth, slope, exposure. An R squared regression line has been plotted through the results for each site descriptor. Some of the results sets show good relationships and the project team is confident that these truly reflect the sensitivities of those particular species. Some species were poorly represented in the results due to their limited plantings initially. These data sets should be treated with caution due to the lack of data and the way in which the curve fit function can fits the regression line.
Of the 17 species that were planted the project team felt that the data sets for five species were worthy of more work and should proceed to an analysis and mapping phase. This will yield a predictive map for those species to indicate potential success on a nationwide basis. The six species were E. globoidea, E.laevopinea, E. macrorhyncha, E.muelleriana, E.pilularis, E.youmanii.
These six species were chosen using a weighted scoring system based on the site occurrence and the experience of the project team. Species that were only capable of suitable performance in a very localized or restricted part of New Zealand were effectively removed from the process. Similarly, species that were only planted at a limited number of sites were excluded due to a lack of strong data. Members of the project team felt that an occurrence register for “low data” species may be an effective way of producing an anecdotal performance may for these species. An excel file with the voting has been included as an attachment. Note that E.fastigata was not included for further evaluation as this species was included in the original trial as a benchmark species. It has had a significant amount of research applied to it at a national level already by Scion and other research institutions.