Author : Sarah Alrefaei
Publisher :
ISBN 13 :
Total Pages : 0 pages
Book Rating : 4.:/5 (137 download)
Book Synopsis A Comparative Study of Box & Jenkins (ARIMA) Model and Kalman Filter Algorithm to Forecast the Impact of Ozone and PM10 Gases Emission in California by : Sarah Alrefaei
Download or read book A Comparative Study of Box & Jenkins (ARIMA) Model and Kalman Filter Algorithm to Forecast the Impact of Ozone and PM10 Gases Emission in California written by Sarah Alrefaei and published by . This book was released on 2022 with total page 0 pages. Available in PDF, EPUB and Kindle. Book excerpt: Air pollution is a significant public health issue in California. According to air monitoring, over 90% of Californians breathe harmful amounts of one or more air pollutants at some point during the year [1]. Although decades of progress, California continues to face some of the most serious issues in the United States for the most harmful and ubiquitous types of air pollution: particle and ozone pollutions [2]. We consider time series analysis of the pollutants to predict the future impact of ozone and particulate matter (PM10) emissions in California. This will facilitate implementation of proper mitigation policies to avoid worsening air pollution problems in California. A systematic study on two commonly prevalent pollutants - ozone and PM10 is carried out with time series analysis to predict the air quality in two cities in California, Los Angeles -Long Beach-Anaheim, and Bakersfield, by using the ARIMA model and the Kalman filter algorithm. The reason for choosing these two cities is that both were the most polluted cities in California in 2021 [3]. We apply the Autoregressive Integrated Moving Average (ARIMA) model to predict how ozone and PM10 affect Los Angeles -Long Beach-Anaheim, and Bakersfield. We also consider an adaptive filtering approach - the Kalman filter algorithm to predict the air quality index for ozone and PM10. Finally, we compare the results from both methods to determine the extent of variation in forecasted air quality indices using these two methods. The air quality indices for the last two and four weeks are forecasted using each method within each city to determine the impact of PM10 and Ozone pollutants. Our results demonstrate that the forecasted air quality values for two weeks are approximately similar to those in the last four weeks by both methods. In particular, the results from the ARIMA model and the Kalman Filter are similar for the forecasted air quality values in terms of PM10 emissions in Los Angeles -Long Beach-Anaheim, and Bakersfield. However, the forecasted air quality values in terms of Ozone emissions in both cities are different. Also, we found that the ARIMA model performed better than the Kalman Filter in most of the scenarios in predicting the air quality values in terms of PM10 and O3 emissions in Los Angeles -Long Beach- Anaheim, and Bakersfield, CA.