Academics and Research Institutes Unite to Discuss the Future of Environmental Data Science and Artificial Intelligence

On Monday, 19 August TAIAO hosted its first-ever Environmental Data Science and AI Summit at Victoria University of Wellington. Speakers from some of Aotearoa New Zealand’s top research institutes came together with academics to discuss how data science and AI can seek a deeper understanding of our environment.

Climate

The day began with Professor Karin Bryan from TAIAO and the University of Auckland providing an update on recent developments in earth observation and monitoring, showing how such information could be used in combination with recent AI applications developed for other disciplines. However, progress in moving from theory to practice has many impediments, ranging from effective communication to technical challenges.

To understand past climates, we use geological records – but key information such as temperature and rainfall cannot be observed. In his talk, Dr Mario Krapp from GNS Science explained a Bayesian approach to reconstructing the climate of the past. He demonstrated the use of Bayesian Inference in the paleoclimate context and how this approach could be used to address complex problems.

Dr Peter Gibson from NIWA – the National Institute of Water and Atmospheric Research – presented on opportunities and challenges for AI in climate downscaling. He showcased some of NIWA’s recent work that’s focused on using generative AI to emulate the traditional downscaling task of RCMs. He explained that the future is bright for AI-based RCM emulators, but comprehensive evaluation of the output and careful refinement of the AI training is needed to build trust in the climate projections produced.

Forestry and agriculture

Dr Dean Meason from Scion, the New Zealand Forest Research, provided an update on Forest Flows - an MBIE Endeavour Research Programme. Understanding how water flows through the land, including planted forests is essential to make the best use of water and land while maintaining environmental health. Forest Flows investigates this with the novel integration of various terrestrial and remote sensing technologies in Pinus radiata (D. Don) plantation forests. Dr Meason explained how Scion partnered with TAIAO to develop Green Artificial Intelligence (AI) – an environmental research framework that can analyse complex environmental data from the Forest Flows programme.

Professor Alan Brent, Chair of Sustainable Energy Systems at Victoria University of Wellington, presented on the Integration of AI and ML into the Operational and Planning Phases of Agrivoltaic Systems. Conventional solar farms have seen the conversion of productive agricultural land with solar arrays and sheep grazing for vegetation management. This is a concern as half of the country’s land area is utilised for agricultural purposes. Globally, agrivoltaic systems have emerged as an innovative solution to address this concern. Optimised systems allow for dual land-usage to both generate electricity and continue with efficient and effective agricultural production. Professor Brent explained how the placement, orientation, and spacing of solar arrays directly impact crop growth and yield, making it crucial to minimise shading while maximising energy production.

Professor Mengjie Zhang, Director of the Centre of Data Science and AI at the University of Waikato, presented two pieces of work using AIML for image segmentation. The first looked at segmentation of individual tree crowns images for Wellington City using Convolutional neural networks and variations, particularly different versions of Yolos. Recent extreme weather events have been leading to increased periods of low salinity in oyster farms, putting them at risk. Professor Zhang discussed three different methods to assess the viability of salinity prediction systems – the best being GP and CNN, both of which performed better than statistical models.

Health

Dr Alvaro Orsi joined us from The Institute of Environmental Science and Research (ESR) to discuss how AI and data science are revolutionising environmental health. He explained how the application of remote sensing and AI modelling can characterise on-site wastewater systems and soil characteristics in polluted areas to inform targeted remediation efforts. He also explored future directions for integrating data science into environmental health practice, emphasising the importance of interdisciplinary collaboration and the development of environmentally sustainable AI solutions.

The day was rounded off by a panel that discussed the future of data science and AI broadly – a key takeaway being that industry and government must work together to make data more available to progress environmental data science and AI.

A big congratulations to the TAIAO team for pulling off an exceptional conference!