Ihaka Lecture Series 2025

Patterns in space

The first law of geography is that everything is related to everything else, but near things are more related than distant things. This year’s Ihaka Lectures are about geographic data - facts on maps. Drawing maps and analysing them used to require specialised resources. Modern computing has made maps accessible as a form of data analysis, but they still require expertise. We will hear from experts in reasoning with geographic data and in presenting it for a wider audience.

Lectures commence at 6.30pm, MLT2/303-102, Building 303, 38 Princes Street

Refreshments will be available before each lecture, at 6pm.

Livestreams are available for each lecture - find the streaming links below.

Lecture 1: Thursday 4 September

Vis-ease - Using visualisation to move beyond the conventional

Professor Dan Exeter
School of Population Health, The University of Auckland

I often receive requests to “just create a heat map of these data”. What I’m really being asked is to produce a choropleth map of some results. Sometimes the mapped variable is a composite of many variables reduced to a single layer. However, today’s computational capacity, combined with a global network of open-source developers designing graphs, glyphs, maps and infographics, provides endless opportunities to communicate patterns hidden within tabular data.

In this presentation, I will show how I use a geospatial lens to identify and investigate differences in health and social outcomes. With examples at the nexus of geography, statistics, and public health, I will reflect on the evolving visualisation landscape that encourages and engages users with more effective narratives than just using heatmaps alone.

Dan Exeter is a quantitative health geographer whose research focuses on geographical variations in health outcomes. His research typically involves the use of Geographical Information Science, big-data and the secondary analysis of large administrative datasets such as those within Statistics New Zealand’s Integrated Data Infrastructure (IDI) to investigate the causes and consequences of inequities in health and social outcomes. He developed the Index of Multiple Deprivation (IMD) for NZ as well as measures of socioeconomic position and deprivation for the population aged 65+. Wherever possible, Dan maximises the potential for geovisualisation tools and techniques to help with the communication and dissemination of his research.

LIVESTREAM LINK:TBC

For more details, please view the Ihaka Lecture Series webpage

Lecture 2: Thursday 11 September

Connecting the dots with R

Professor Adrian Baddeley FAA
John Curtin Distinguished Emeritus Professor, Curtin University

A 'spatial point pattern' is a map covered with dots showing the spatial locations of events (accidents, crimes, disease cases) or things (gold deposits in a country, galaxies in the distant universe). Such maps played an important role in the discovery of the cause of cholera, the identity of the Yorkshire Ripper serial killer, the existence of new gold deposits in Western Australia, and the large-scale structure of the universe.

Analysing such maps is a problem which bedevilled statisticians for a century. Many leading experts in statistics once considered it difficult or impossible to apply standard statistical methods to spatial point pattern data, and declared that new specialised methods would be needed. Experts in probability theory developed a theory of random point patterns which was elegant and powerful, but almost completely incomprehensible to ordinary people. Instead, individual scientists in different fields were left to develop their own methods without much guidance from statisticians.

This unhappy situation was completely transformed by the arrival of R. Following the best traditions of statistical science, R made it possible to develop statistical ideas hand-in-hand with software which applied and tested the ideas. Previously insurmountable barriers were progressively overcome. Even abstract concepts of point process theory became accessible when represented as interactive graphical displays in R. Statistical methodology for spatial point patterns has reconnected with the `mainstream’ of statistical science, and is now a powerful scientific tool.

Adrian Baddeley is a statistician who develops methodology and software for spatial data. His contributions include the method of vertical sections in stereology (quantitative microscopy), metrics for measuring discrepancies between pixel images, and statistical methodology for spatial point patterns, including the ‘spatstat’ package. His current research investigates the effect of industrial pollution on indigenous rock art.

Adrian graduated from Australian National University and Cambridge University, and is a winner of the Pitman and Hannan medals. He is a John Curtin Distinguished Emeritus Professor at Curtin University, and a Fellow of the Australian Academy of Science.

LIVESTREAM LINK: TBC

For more details, please view the Ihaka Lecture Series webpage

Seminars, conferences and events may be photographed or filmed for use by the University and the media in hard copy and electronic publications, including in newsletters and brochures, on social media, and posted on the University or conference websites to promote subsequent events and the University. If you do not wish to be photographed or filmed, please inform the organiser or photographer. Please note that all University of Auckland campuses are entirely smoke-free and we ask that all alumni, staff, students and guests comply with this.