Gustavo Olivares

Spatio-temporal PM 2.5 prediction by spatial data aided incremental support vector regression

Machine learning requires sufficient and reliable data to enhance the prediction performance. However, environmental data sometimes is short and/or contains missing data. Often existing prediction models built on machine learning fail to predict environmental problems accurately. We argue that spatial domain data can be used to facilitate the training of temporal prediction model. This paper formulates mathematically a spatial data aided incremental support vector regression (SalncSVR) for spatio-temporal PM2.5 prediction. We conduct spatio-temporal PM2.5 prediction over 13 monitoring stations in Auckland New Zealand, and compare the proposed SalncSVR with a pure temporal IncSVR prediction.

Vehicle emission prediction using remote sensing data and machine learning techniques

More and more researchers are using remote sensing technology to measure real-world, on-road automobile emissions of nitric oxide (NO), one of the most important and frequently studied pollutants. Partnered with the National Institute of Water and Atmospheric Research (NIWA) in New Zealand, we aim to establish a robust NO emission factor prediction model using remote sensing data to forecast future emissions. This model can be extremely useful to local transport authorities for monitoring urban pollution levels and validating the effectiveness of existing traffic control policies. We have conducted this research using real-world data that were collected over a 10-year span between 2005 and 2015. The experimental results have shown that the vehicle emission patterns are continuously changing and the relevance of remote sensing data for future predictions decays as they get older. We propose a 3-step machine learning approach to establish this model. Most notably we use quantile regression forest (a variation of random forest) as the base algorithm and use random forests variable importance measure to validate and interpret the features. The model yields error rates that compare favourably to linear model based recursive partitioning and the original random forest model.

Marine Aerosol Hygroscopicity and Volatility, Measured on the Chatham Rise (New Zealand)

The Surface Ocean Aerosol Production (SOAP) study was undertaken in February/March 2012 in the biologically active waters of the Chatham Rise, NZ. Aerosol hygroscopicity and volatility were examined with a volatility hygroscopicity tandem differential mobility analyser. These observations confirm results from other hygroscopicity-based studies that the dominant fraction of the observed remote marine particles were non-sea salt sulfates. Further observations are required to clarify the influences of seawater composition, meteorology and analysis techniques seasonally across different ocean basins.

What is sustainable air quality

For 40 years or more, air quality policy has been based on the paradigm of the air quality standard as a uniform criterion of acceptable environmental degradation, built on the foundations of the precautionary principle. However, developments in health science have undermined some of the underlying assumptions of this paradigm whilst technological emission controls have been offset by growing economic activity. Current trends are towards increasingly demanding notions of what constitutes acceptable air quality. Proposed future air quality standards could require a revolution in urban form and infrastructure. We need a new paradigm - sustainable air quality. In this paper, we discuss the need for a new paradigm of sustainable air quality management, its basis and what it might hope to achieve, including how further downward pressure can be exerted on emissions through urban form, urban design, transport policy and projects, energy strategy, etc.