scholarly journals Estimation of PM2.5 Concentrations in New York State: Understanding the Influence of Vertical Mixing on Surface PM2.5 Using Machine Learning

Atmosphere ◽  
2020 ◽  
Vol 11 (12) ◽  
pp. 1303
Author(s):  
Wei-Ting Hung ◽  
Cheng-Hsuan (Sarah) Lu ◽  
Stefano Alessandrini ◽  
Rajesh Kumar ◽  
Chin-An Lin
Keyword(s):  
Machine Learning ◽  
New York ◽  
New York State ◽  
Vertical Mixing ◽  
Model Performance ◽  
Machine Learning Techniques ◽  
Joint Analysis ◽  
Ann Model ◽  
York State ◽  
Ann Models

In New York State (NYS), episodic high fine particulate matter (PM2.5) concentrations associated with aerosols originated from the Midwest, Mid-Atlantic, and Pacific Northwest states have been reported. In this study, machine learning techniques, including multiple linear regression (MLR) and artificial neural network (ANN), were used to estimate surface PM2.5 mass concentrations at air quality monitoring sites in NYS during the summers of 2016–2019. Various predictors were considered, including meteorological, aerosol, and geographic predictors. Vertical predictors, designed as the indicators of vertical mixing and aloft aerosols, were also applied. Overall, the ANN models performed better than the MLR models, and the application of vertical predictors generally improved the accuracy of PM2.5 estimation of the ANN models. The leave-one-out cross-validation results showed significant cross-site variations and were able to present the different predictor-PM2.5 correlations at the sites with different PM2.5 characteristics. In addition, a joint analysis of regression coefficients from the MLR model and variable importance from the ANN model provided insights into the contributions of selected predictors to PM2.5 concentrations. The improvements in model performance due to aloft aerosols were relatively minor, probably due to the limited cases of aloft aerosols in current datasets.

scholarly journals The impacts of transported wildfire smoke aerosols on surface air quality in New York State: A multi-year study using machine learning

2021 ◽  
pp. 118513
Author(s):  
Wei-Ting Hung ◽  
Cheng-Hsuan (Sarah) Lu ◽  
Stefano Alessandrini ◽  
Rajesh Kumar ◽  
Chin-An Lin
Keyword(s):  
Machine Learning ◽  
New York ◽  
Air Quality ◽  
New York State ◽  
York State ◽  
Wildfire Smoke

Daily Simulation of Ozone and Fine Particulates over New York State: Findings and Challenges

2007 ◽  
Vol 46 (7) ◽  
pp. 961-979 ◽  
Author(s):  
C. Hogrefe ◽  
W. Hao ◽  
K. Civerolo ◽  
J.-Y. Ku ◽  
G. Sistla ◽  
...  
Keyword(s):  
New York ◽  
New York City ◽  
Air Quality ◽  
Metropolitan Area ◽  
New York State ◽  
Model Performance ◽  
Concentration Gradients ◽  
Seasonal Differences ◽  
York State ◽  
Urban Rural

Abstract This study investigates the potential utility of the application of a photochemical modeling system in providing simultaneous forecasts of ozone (O3) and fine particulate matter (PM2.5) over New York State. To this end, daily simulations from the Community Multiscale Air Quality (CMAQ) model for three extended time periods during 2004 and 2005 have been performed, and predictions were compared with observations of ozone and total and speciated PM2.5. Model performance for 8-h daily maximum O3 was found to be similar to other forecasting systems and to be better than that for the 24-h-averaged total PM2.5. Both pollutants exhibited no seasonal differences in model performance. CMAQ simulations successfully captured the urban–rural and seasonal differences evident in observed total and speciated PM2.5 concentrations. However, total PM2.5 mass was strongly overestimated in the New York City metropolitan area, and further analysis of speciated observations and model predictions showed that most of this overprediction stems from organic aerosols and crustal material. An analysis of hourly speciated data measured in Bronx County, New York, suggests that a combination of uncertainties in vertical mixing, magnitude, and temporal allocation of emissions and deposition processes are all possible contributors to this overprediction in the complex urban area. Categorical evaluation of CMAQ simulations in terms of exceeding two different threshold levels of the air quality index (AQI) again indicates better performance for ozone than PM2.5 and better performance for lower exceedance thresholds. In most regions of New York State, the routine air quality forecasts based on observed concentrations and expert judgment show slightly better agreement with the observed distributions of AQI categories than do CMAQ simulations. However, CMAQ shows skill similar to these routine forecasts in terms of capturing the AQI tendency, that is, in predicting changes in air quality conditions. Overall, the results presented in this study reveal that additional research and development is needed to improve CMAQ simulations of PM2.5 concentrations over New York State, especially for the New York City metropolitan area. On the other hand, because CMAQ simulations capture urban–rural concentration gradients and day-to-day fluctuations in observed air quality despite systematic overpredictions in some areas, it would be useful to develop tools that combine CMAQ’s predictive capability in terms of spatial concentration gradients and AQI tendencies with real-time observations of ambient pollutant levels to generate forecasts with higher temporal and spatial resolutions (e.g., county level) than those of techniques based exclusively on monitoring data.

scholarly journals Epidemic Model Guided Machine Learning for COVID-19 Forecasts in the United States

2020 ◽  
Author(s):  
Difan Zou ◽  
Lingxiao Wang ◽  
Pan Xu ◽  
Jinghui Chen ◽  
Weitong Zhang ◽  
...  
Keyword(s):  
United States ◽  
Machine Learning ◽  
New York ◽  
Epidemic Model ◽  
New York State ◽  
The United States ◽  
Machine Learning Algorithms ◽  
York State ◽  
Link Type ◽  
The Us

AbstractWe propose a new epidemic model (SuEIR) for forecasting the spread of COVID-19, including numbers of confirmed and fatality cases at national and state levels in the United States. Specifically, the SuEIR model is a variant of the SEIR model by taking into account the untested/unreported cases of COVID-19, and trained by machine learning algorithms based on the reported historical data. Besides providing basic projections for confirmed and fatality cases, the proposed SuEIR model is also able to predict the peak date of active cases, and estimate the basic reproduction number (). In particular, the forecasts based on our model suggest that the peak date of the US, New York state, and California state are 06/01/2020, 05/10/2020, and 07/01/2020 respectively. In addition, the estimated of the US, New York state, and California state are 2.5, 3.6 and 2.2 respectively. The prediction results for all states in the US can be found on our project website: https://covid19.uclaml.org, which are updated on a weekly basis, and have been adopted by the Centers for Disease Control and Prevention (CDC) for COVID-19 death forecasts (https://www.cdc.gov/coronavirus/2019-ncov/covid-data/forecasting-us.html).

Sign in / Sign up

Export Citation Format

Share Document