Estimating Emission Reductions and Health Benefits from Episodic Controls on Woodburning

DAVID LIGHTHALL (1), Tim Tyner (2), David Nunes (1)

(1) San Joaquin Valley Air Pollution Control District, Fresno, (2) University of California, San Francisco-Fresno

     Abstract Number: 120
     Last modified: November 5, 2009

Abstract
The San Joaquin Valley of California experiences severe thermal inversions during winter months. Nighttime wood smoke has historically made a significant source contribution to the violation of the federal fine particulate standard from November through February. In 2003 the San Joaquin Valley Air Pollution Control District adopted episodic controls on domestic wood burning (Rule 4901). After three winters a research evaluation was funded to determine the effectiveness of the rule, focusing on (1) estimating reductions in PM 2.5 concentrations due to public compliance with daily bans, (2) using the EPA’s BenMAP program to estimate the corresponding reduction in PM 2.5-related disease (morbidity) and premature death (mortality), and (3) estimating the economic value of these health benefits. The Bakersfield and Fresno/Clovis metro areas were used as study areas.

As a threshold step in the analysis, comparisons of diurnal (24 hr.) emission concentrations from clusters of days with comparable meteorological conditions in the pre- and post-Rule 4901 period revealed clear differences in PM 2.5 concentrations. Days with wood burning bans were characterized by lower 24 hr. and, in particular, evening PM 2.5 levels when compared with days with comparable meteorological conditions but no controls. Second, a statistical analysis of the relationships between Fresno/Clovis PM 2.5 levels and meteorological parameters, including wind speed, temperature, and atmospheric stability, was conducted using data from the three winters prior to Rule 4901. This resulted in a meteorological model capable of accurately predicting evening and daily (24 hr.) PM 2.5 concentrations relative to observed levels. A second model derived from Bakersfield data also accurately predicted the observed PM 2.5 concentrations. Third, this model was applied to meteorological data from three years of the post-Rule 4901 period to see if model-predicted PM 2.5 levels were in excess of actually observed evening and daily concentrations. Over-prediction did consistently occur and it grew in successive years, providing evidence of progress in public compliance. The contribution of other mobile and stationary source controls to PM 2.5 reductions was negligible. The model over-estimated Fresno/Clovis PM 2.5 from 26% to 36% in the evening (9pm to 1am) and from 18% to 28% in overall daily PM 2.5. For Bakersfield, overestimations of evening PM 2.5 ranged from 22% to 35%. Converted to a 12 month metric, Rule 4901 contributed to an average annual daily PM 2.5 reduction of 13.63% in Bakersfield and 12.94% in Bakersfield. Finally, using the EPA’s BenMAP benefit estimation software, the reduced annual exposure to PM 2.5 contributed to mean annual avoided morbidity costs in Fresno/Clovis of $11 million to $26.6 million and $5.7 million to $14.1 million in Bakersfield. Avoided annual mortality costs were $367.5 million to $430.6 million in Fresno/Clovis and $189.1 million to $239.9 million in Bakersfield.

In tandem, the meteorological modeling and BenMAP benefit estimation provided clear evidence of reduced disease, pre-mature deaths, and associated cost savings attributable to wood burning controls. This scientific evidence played a key role in a further strengthening of Rule 4901 by the Valley Air District Board in 2008.