Volcanic emissions and air pollution: Forecasts from time series models

The Kilauea volcano in Hawai'i has been erupting continuously since 1983. The eruptions are a major source of air pollution. Sulfur oxides released from the volcano react with sunlight, atmospheric gases and aerosols, and convert to fine particles. The resulting volcanic smog, known as vog, is a recognized health hazard. Data sets for volcanic emissions, sulfur dioxide gas and particulate matter are analyzed. Volcanic emissions are available at a daily resolution. Emissions show evidence of nonlinear variability, and intermittent trending over periods of several months. Forecasting tests at horizons of 1–7 days find that emissions can be predicted only with a substantial error. Air quality data is available hourly. Forecasting experiments for SO2 and particulate matter are run over horizons of 1–24 h and 1–7 days, using time series models. Volcanic emissions were tested as causal inputs in the models for the SO2 and particulate matter data, but were found to increase the forecast error. The daily and hourly SO2 series are predicted more accurately by ARIMAs. The results for particulate matter are more ambiguous. Over short horizons, they are often predicted better by regressions on levels or simple persistence models. At horizons associated with cycles (24 h, 7 days), ARIMAs and persistence forecasts achieve the most accurate results. Keywords: Volcanic emissions, Air pollution, Time series models

Abstract licence: CC BY-NC

A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.