The AgMardt-funded project, ‘Matching elite high-value eucalypts to drylands’, is now complete.



The project aimed to establish hybrid physiological/mensurational yield models for durable dryland eucalypt species and to connect microsite variation with larger scale variation in site influences on tree species.

Key outcomes of the project

  • Users of NIWA’s Virtual Climate Station Network (VCSN) now know how precise VCSN estimates are likely to be for key variables. The VCSN is becoming widely used in forestry.
  • Those who lack LiDAR data now have a method for creating local digital elevation models (DEMs) from transects of GPS points. LiDAR coverage is sparse in New Zealand, and both managers and researchers make use of high resolution DEMs for decision-making.
  • Local variation in dryland eucalypt performance is now known to be influenced by exposure and topographical position. Growers may choose to not plant some areas of the properties where performance is likely to be poor.
  • Growers wishing to choose species for their sites now have access to estimates of site tolerance and utility of two dryland eucalypt species in on-line software at
  • A growth and yield model for E. globoidea is now available at This will enable growers to make estimates of future yield of that species following inventories of the stands.

Much of the work on this project was undertaken by Dr Serajis Selakin, from Bangladesh. Serajis’ PhD was based on this work. His thesis ‘Hybrid growth models for E. globoidea and E.bositoana: Explaining within and between site variability’ is available here.

Project methods: more details

Meteorological stations were established at seven NZDFI trial sites, and a total of 113 plots were measured. Soil moisture and temperature loggers were established on three sites where high resolution estimates of microsite influences were conducted. In addition, detailed soil surveys at two intensively monitored sites were conducted during the second half of 2015.

Digital elevation models (DEMs) were created at two sites using transects of points with a high resolution global positioning system, demonstrating the viability of this technique for sites with poor LiDAR coverage.

  • The research demonstrated that more understanding of local variation in tree performance could be gained from highly accurate DEMs than from intensive soil sampling on these sites. The predominant influences were wind exposure, represented by a wind exposure index from the DEMs, position index, distance from the top of a ridge line, and morphometric protection, also calculated from the DEMs.
  • Juvenile yield models were created for two eucalypt species, covering both local and global site variables. A growth and yield model was created for E. globoidea, for which rotation-length data were available.

Research updates and presentations

Check out our Resources, especially our six-monthly Project Updates, to read about how research has progressed over time.

Presentations on site x species work from two recent workshops are also available:

Key researchers

Dr Justin Morgenroth and  Professor Euan Mason are working together on the site-species matching research.

More information

Visit the School of Forestry website to learn more about matching species to site research.