How do interannual climate variations influence air quality parameters in New England?

To address this research question, we have begun to quantify the linkages between surface O3 variability in New England, in the form of O3 mixing ratios, and associated interannual climate fluctuations, both surface- and circulation-based (Brown et al., 2005). The relatively long record of O3 observations in New England (1980-present), and previous studies that intimated a climate connection to O3 variations, suggest that this analysis of interannual climate controls on O3 concentrations will yield important insights into climate-air quality relationships. It will also provide a framework from which to subsequently examine other air quality parameters, specifically criteria pollutants, pollen, mold, and acidic aerosols, and the influence of interannual climate variations on those air quality measures. Furthermore, it will inform an analysis of interannual climate variability on longer time scales in New England, in particular an examination of multiple indicators of climate change (such as temperature, precipitation, snowfall, snow cover, ice-out dates, and growing season length) and their linkage to atmospheric circulation anomalies.

Our initial results indicate the presence of marked spatio-temporal variability of O3 conditions across New England on interannual time scales. A principal components-based regionalization analysis [Yarnal et al., 2001] of normalized (z-score) April-October daily 8-hr maximum mixing ratios for the period 1980-2004 reveals five spatially coherent regions (Regions 1 through 5 on Figure below). This spatio-temporal regionalization will be used as a framework to elucidate the large-scale influences on O3 variability in New England by identifying regional ÒsignaturesÓ where O3 - circulation relationships are strongest.

We are currently drawing linkages between interannual O3 mixing ratio anomalies and interannual surface climate variability using synoptic climatology methodologies including correlation and compositing (Yarnal et al, 2001). Measurements of daily maximum, minimum, mean, and dewpoint temperature, along with wind speed, wind direction, and precipitation, have been collected from first-order and cooperative NWS stations throughout New England for the same time period as the O3 data (i.e., Apr-Oct, 1980-2004). These data have been grouped into spatial regions matching those of the O3 data, and a regional average time series for each climate parameter, as well as O3, has been generated for each region. Pearson correlation coefficients are being calculated between the O3 and climate time series for each region; additionally, regional composites of each surface climate parameter are being calculated for high-event O3 days (defined as days when the normalized regional O3 value exceeds 2.0). In this fashion, we are illustrating spatio-temporal variations across New England of the linkages between O3 and various surface climate parameters.

We are also linking the occurrence of variations in seasonal O3 mixing ratios to atmospheric circulation anomalies and large-scale modes of interannual climate variability. The database for this analysis includes daily and monthly atmospheric circulation anomaly patterns derived from the NCEP/NCAR Reanalysis (Kalnay et al, 1996), daily New England synoptic weather types (e.g., Keim et al, 2005), and monthly/seasonal indices representing different modes (e.g., North Atlantic Oscillation [NAO], El-Nino Southern Oscillation [ENSO], Madden-Julian Oscillation and patterns (e.g., Bermuda High Index [BHI], Trough Axis Index/Trough Intensity Index [TAI/TII; Bradbury et al., 2002a]) of interannual climate variability. The methodological approach is again based on established synoptic climatology techniques. An indexing analysis is being employed to link circulation anomalies over New England, such as the occurrence of particular weather types as defined by Keim et al [2005], to inter-seasonal and interannual variations in large-scale modes of climate variability, as represented by teleconnection indices (e.g., BHI, Southern Oscillation Index [SOI]). A complementary compositing approach (Frakes and Yarnal, 1997) is providing a framework from which we are investigating the temporal variability of preferred patterns of atmospheric circulation over New England for specific time periods, such as during individual and multi-day high O3 events (normalized regional O3 values > 2.0). In both the indexing and compositing analyses, we are focused especially on the temporal linkages between the MJO and NAO modes of variability, and the occurrence of circulation anomalies over New England related to these modes that have been shown to be associated with O3 mixing ratio anomalies.

This analysis of O3 variability on interannual time scales, and the connection to large-scale modes and patterns of climate variability, represents the first step in a holistic exploratory investigation of interannual air quality-climate linkages in New England. We anticipate conducting follow-up studies involving several other measures of air quality (e.g., CO, SO2, PM10, PM2.5, acidic aerosols, pollen, and mold).