Dr Trevor Moffiet

Biography

I have 40 years of cross-disciplinary work experience; Research, Teaching, Industry. Experiences include: Industrial Research (metallurgical processes, physical metallurgy), Statistical Consulting (Industry, research disciplines ranging from Engineering to Psychology to Food Science), Secondary Teaching (Maths, Science, Biology and Chemistry), University Teaching (Statistics), and Academic Research (Remote Sensing, Statistics, and Energy Efficiency of Housing)

My research higher degree (ARC linkage project Grant number LP0214188 with Queensland Department of Natural Resources) was in a cross-disciplinary area of statistics and remote sensing. With no prior experience in remote sensing, I am first author on one paper on Lidar Remote Sensing and one on Vegetation Indices and modelling of Foliage Projective Cover. In addition I hold a body of original work on vegetation indices and how to model vegetation cover that I have not published. I have not been in a position to complete that work to a level suitable for publication because I have not been employed in that field after graduating. However I retain an interest in that area. My interest is demonstrated by me assisting, in my own time, two PhD students from other univeristies that are doing research that involves remote sensing of vegetation.

In my Industrial Research role I attained the level of Principal Research Officer. (BHP and OneSteel). Additional to my industrial research my industrial experiences include: technical problem solving, process improvement and process improvement facilitation. In my production role I acted as the Technical Superintendent for production of galvanised steel wire, rural fencing products. I acted as a Process Improvement Facilitator for twelve project leaders in an intensive one year improvement exercise. I was an integral member of a research and engineering team that designed a novel galvanising ‘Jet Wiping’ process. The team took the project from fundamental research in a pilot plant through to final implementation and commissioning of that process as a large a scale commercial operation. Automatic control of the zinc coating on the wire during production required the establishment of robust calibration models between continuous electronic measurements of the coating during production and chemical measurements of samples in a laboratory. I developed the calibration model using conventional regression methods. Inadequacies with the conventional techniques provided the seed for a novel idea for improved calibration methods. I have carried that idea through to implementation withinmy Remote Sensing PhD research project and into statistical consulting where I have come across many diverse applications that would benefit from my new approach.

The new modelling method I have labelled in general as "bivariate relationship modelling on bounded spaces". More specifically I have called it Lineal Basis Modelling.

My current role is as a part time Research Associate in the priority research centre for energy at the University of Newcastle. My primary fuction within that role was to produce and maintain a database of thermal measurements taken from four experimental housing units. The data are collected at 5 minute intervals from 105 sensors in each of four modules, 24hrs per day since 2003. In that role I have made a significant research contribution to developing an index on the dynamic thermal performance of buildings. In essence the index is derived from elliptical shapes that are statistically fitted to hysteresis patterns of internal temperature plotted against external temperature. There is scope to write up this work as a new statistical modelling method.

Qualifications

• PhD, University of Newcastle, 2008
• Graduate Diploma in Mathematical Studies, University of Newcastle, 1995
• Graduate Diploma in Education, University of Newcastle, 1980
• Bachelor of Science (Metallurgy), University of Newcastle, 1977

Research

Research keywords

• Bivariate Relationships
• Bounded Spaces
• Foliage Cover
• Landsat
• Lidar
• Remote Sensing

Research expertise

Developed a new method for modelling of bivariate relationships on bounded spaces - the new method has wide ranging application in the bio sciences but also physical sciences. The method could be regarded as a new calibration method or errors in variables method and is suitable for allometry.

Modeling of bivariate relationships on bounded spaces, Application of statistical methods to industrial process control problems, Application of statistical models to satellite remotely sensed data.

Collaboration

I have been collaborating with or, more correctly, assisting in an unofficial and unsupported capacity two PhD students external to the university of Newcastle:

Sisira Ediriweera, School of Environment, Science and Engineering, Southern Cross University,

Lismore, NSW, Australia, "Effect of different topographic normalization methods on prediction accuracy of Foliage Projective Cover (FPC) in topographically complex landscapes"

Fatwa Ramdani, Tohoku University, Graduate School of Science, Earth Science Department, Miyagi, Japan, "Changes in local climate due to expansion of palm oil plantation in Indonesia".

Fields of Research

Centres and Groups

Centre

---

Teaching

Teaching keywords

• Statistics 1070

Teaching expertise

First year university statistics