The 100+ forests of the National Arboretum Canberra constitute a major biological resource based on a wide diversity of tree species that represent global biogeography, ethno-botany and species of international concern with ratings of endangered or highly endangered in the IUCN Red Book of endangered species.
Knowing how well all of these species are growing on the Arboretum site, and how their health is trending through time is important information to both managers of the collection and the wider public.
The Tree Growth Monitoring project aims to provide a quantitative, long term record of tree growth as measured by height and diameter measurements, done on about an annual cycle for the first five years, and possibly less often in later years.
The statistical design, provided by Dr Ross Cunningham from the Australian National University, ensures a rigorous underpinning to the data for determining height and diameter growth, and assessments of the impact of topography on growth, as well as the capacity to gauge genetic diversity of each species.
Dr Roger Hnatiuk, Member of the Friends of the National Arboretum Canberra. Dr Hnatuik is a retired, professionally trained and experienced plant ecologist with specific experience in forest ecology, primary productivity, micrometeorology and biogeography.
Citizen science volunteers who are members of the Friends of the National Arboretum Canberra.
The aims of the project are primarily to monitor the growth of the trees in the planted forests at the National Arboretum:
- to determine the heights and diameters of the forest trees as they develop and to track their rate of growth
- to determine if, and to what extent, tree growth is affected by local landscape position (slope, aspect)
- to determine the degree of genetic variability in the population of trees in each forest.
Each forest will be sampled separately:
- Measurements will be made of the diameter and height of a representative sample of trees from each Forest.
- Diameter is measured at 1.3m above ground. Callipers are used when diameters are too small for standard diameter tapes. Dimensions are recorded in centimetres. Where more than one trunk is present, either the dominant trunk will be measured, or in some circumstances, where no dominance is obvious, more than one trunk will be measured.
- Height is measured in metres to the highest point of the tree. Generally this will be the top of the highest green leaves of the dominant trunk, usually a single trunk but in some circumstances there may be more than one trunk, but the height will still be to the highest point as seen from an appropriate place on the ground. While trees are small, heights are measured using wooden poles with measuring tapes attached, or with 10cm bands marked for trees taller than about 1.5m. Height to the nearest cm will be recorded for trees up to about 2.0 m tall, above this height, to the nearest 10 cm. A special case concerns trees such as the palms, dragon trees, grass trees and cycads. In these instances, where the leaves make a significant contribution to the perceived height of the tree, two measures will be recorded. The first is the height at the top of the highest leaf in its natural position (ie. the leaves are not to be lifted for the measurement). The second is the height to the top of the estimated 'apex' of the stem. The latter will not be visible in most cases because of the cluster of leaf bases around that point. For practical purposes, the height will be taken to the base of the green leaves. In the case of grass trees, a third measurement will be taken to the top of mature inflorescences when they are present.
- Initially, forests were sampled using the following strategy to locate the trees to be measured. The forest was surveyed by walking throughout it to detect any major topographic variations as well as to see whether there were any sub- plantings that were clearly different in age. Blocks were then randomly located within each of the variations found. A block was a contiguous area, usually rectangular in outline, containing about 25-30 trees. All trees within the block were then measured. Each tree within the block was individually numbered (see below for details of tagging trees) so that subsequent re-measuring could be related to each tree separately for better estimates of growth rates.
- After trialling the first method and seeking advice from Dr Ross Cunningham, Centre for Research on Environment and Society, Australian National University, on sampling strategies, a new strategy was adopted that would allow a more rigorous system of analysis with respect to detecting within-forest variability. This new method will be used for all forests. Means, medians and variability of the samples are expected to be fairly similar between the methods, though the new methods will allow more rigorous analyses of within-site variations.
- The new method is a form of stratified random sampling. Using tree planting maps, each forest is divided into 6 or 8 rectangular blocks (usually with a line halfway between sides paralleling the long axis, and then three lines at regular intervals at right angles to the first. This plan is varied where the Forest contains significant irregularities in shape, in which case, blocks may not be rectangular in shape, may represent geographic outliers etc.
- Within each block, the trees are numbered. The total number of trees for the forest is calculated. A random sample of about 15 trees is apportioned across the forest according to the proportion of trees in each block compared to the total for the forest. These trees are marked on the maps and the nearest adjacent tree selected to constitute a 'pair'. The second member of the pair is chosen to be close to the first and on the same elevation contour so as to minimise environmental variability. The trees are marked on the digital maps. The 30 trees are then located in the field for measurement. Each tree is labelled in the field.
- This research will provide quantitative and statistically sound evidence of the growth rates of the full range of species making up the forests of the Arboretum. In some cases, these data may be the first time that their growth rate has been recorded, or for some, recorded for their early growth states. These data may be useful in assessing the potential and desirability of some species for urban horticulture (street trees for example).
They will also provide evidence for changes in health of the trees, as well as responses to weather and climate with respect to temperature and rainfall.
- The statistical design of the sampling strategy is such that the trees of the sample can not only be used for the growth study, but they can be used by other researchers as the basis for a potentially diverse range of other studies. The design allows the genetic diversity of each forest to be assessed. With the long run of data on growth of specific trees, from a random, block layout design, the use of these trees increases the value of other studies and the value of these specific trees to the Arboretum immediately and in the long run
This project is designed to last for a very long time, as by its very nature it falls into the long term ecological research mode. The longer it can be maintained, the more valuable the data become.
- The first public exposure of the project was in October 2013, at the 5th Global Botanic Gardens Congress held jointly with the Annual meeting of the Botanic Gardens of Australia and New Zealand, in Dunedin NZ.
- A short, general public presentation has been arranged with the Friends of the Australian National Botanic Gardens in April 2015.
- Detailed analysis of the data is anticipated but not planned.
- It has been the intention from the beginning that the raw data be made publically available. It is anticipated that the existing run of four + years of data will be ready for public release during 2015.
- A schedule of formal reports to the Arboretum will be negotiated.
- A short note reporting ongoing work is made annually to the Friends of the National Arboretum Canberra Annual General Meeting.