Climate, weather, and avocados
Agriculture is highly dependent on climate and, as such, crop yield variability is affected by year-to-year climatic variability, with regards to both extreme events and changes in historical patterns of regional climate. Understanding the effect of weather variables on crop production is key to forecast and manage production. Currently, multiple systems, including remote and proximal sensing, collect data at high temporal frequency. However, it is challenging to identify the specific variables that can specifically have an impact on production.
The project we are currently working on uses NOAA GFS historical longitudinal data of the more than last ten years. We identify and extract multivariate weather factors that are known to have an impact on the avocado yield. New methodologies, based on machine learning and artificial neural networks, are investigated to identify the most important set of variables that have an effect on the avocado production. It is known that interactions between weather variables can have more impact than single variables effect, therefore it is important to build models that can directly consider these interactions. These weather factors are then used in the model along with the avocado yield that was recorded for the last ten growing seasons.
Jelena is a PhD Candidate at the University of Auckland. During her PhD, Jelena was working on developing a Bayesian network (BN) as a method of representing vineyard ecosystem. BN was used to model vineyard ecosystem incorporating chemical profiles, meteorological information, and other data at different time points in the life cycle of vineyards in order to discover the differences vineyard management techniques make with respect to resilience and profit. Last January Jelena joined PlantTech Research team as research scientist. Jelena's research focus on the development and application of machine learning methods to provide answers to some of the interesting horticultural problems.
Louis completed a PhD (Biology) from the University of Auckland in 2010. After his PhD, Louis worked as a postdoc at the Department of Statistics at the University of Auckland and also as bioinformatician for New Zealand Genomics Ltd. After a postdoc experience at the Australian National University, Louis returned to New Zealand to work as a machine learning engineer at Biomatters Ltd. As principal scientist for PlantTech Research Institute Ltd, Louis's research focus on the development and application of machine learning methods to the analysis of biological datasets, with a particular emphasis on the integration of multiscale and heterogenous data sources, from genomics to sensors data.
