Impacts of the Bermuda High on Regional Climate and Ozone over the United States

Jinhong Zhu Earth System Science Interdisciplinary Center, University of Maryland, College Park, College Park, Maryland

Search for other papers by Jinhong Zhu in
Current site
Google Scholar
PubMed
Close
and
Xin-Zhong Liang Earth System Science Interdisciplinary Center, and Department of Atmospheric and Oceanic Science, University of Maryland, College Park, College Park, Maryland

Search for other papers by Xin-Zhong Liang in
Current site
Google Scholar
PubMed
Close
Full access

Abstract

Observations reveal that, in summer, westward extension of the Bermuda high enhances the Great Plains low-level jet (LLJ) that transports more moisture northward, causing precipitation increases in the Midwest and decreases in the Gulf States. Meanwhile, more warm air advection from the Gulf of Mexico to the southern Great Plains and stronger clear-sky radiative heating under high pressures over the Southeast result in warmer surface temperatures across the Gulf states. The enhanced LLJ transport of cleaner marine air from the Gulf reduces surface ozone across the southern Great Plains–Midwest. In contrast, larger transport of more polluted air from the Midwest to New England and more frequent air stagnation under high pressures in the Southeast increase ozone over most of the eastern coastal states. This Bermuda high–induced ozone change reversal between the southern Great Plains–Midwest and eastern coastal states, with a magnitude of 6 and 13.5 ppb, respectively, in summer-mean maximum daily 8-h average, exhibits strong decadal variations that should be considered in the U.S. air quality dynamic management.

The observed Bermuda high signatures over the Gulf states can be well captured by regional climate and air quality models. Notable model deficiencies exist over the northern Great Plains–Midwest that are more remote to the Bermuda high and LLJ control. The regional models largely reduce these deficiencies from general circulation models (GCMs). Only 7 out of 51 GCMs can represent all key regional signatures of the Bermuda high, while none can simulate its strong association with planetary sea surface temperature anomalies. The result indicates a great challenge for GCMs to predict Bermuda high variability and change.

Corresponding author address: Dr. Xin-Zhong Liang, Earth System Science Interdisciplinary Center, University of Maryland, College Park, 5825 University Research Court, College Park, MD 20740-3823. E-mail: xliang@umd.edu

Abstract

Observations reveal that, in summer, westward extension of the Bermuda high enhances the Great Plains low-level jet (LLJ) that transports more moisture northward, causing precipitation increases in the Midwest and decreases in the Gulf States. Meanwhile, more warm air advection from the Gulf of Mexico to the southern Great Plains and stronger clear-sky radiative heating under high pressures over the Southeast result in warmer surface temperatures across the Gulf states. The enhanced LLJ transport of cleaner marine air from the Gulf reduces surface ozone across the southern Great Plains–Midwest. In contrast, larger transport of more polluted air from the Midwest to New England and more frequent air stagnation under high pressures in the Southeast increase ozone over most of the eastern coastal states. This Bermuda high–induced ozone change reversal between the southern Great Plains–Midwest and eastern coastal states, with a magnitude of 6 and 13.5 ppb, respectively, in summer-mean maximum daily 8-h average, exhibits strong decadal variations that should be considered in the U.S. air quality dynamic management.

The observed Bermuda high signatures over the Gulf states can be well captured by regional climate and air quality models. Notable model deficiencies exist over the northern Great Plains–Midwest that are more remote to the Bermuda high and LLJ control. The regional models largely reduce these deficiencies from general circulation models (GCMs). Only 7 out of 51 GCMs can represent all key regional signatures of the Bermuda high, while none can simulate its strong association with planetary sea surface temperature anomalies. The result indicates a great challenge for GCMs to predict Bermuda high variability and change.

Corresponding author address: Dr. Xin-Zhong Liang, Earth System Science Interdisciplinary Center, University of Maryland, College Park, 5825 University Research Court, College Park, MD 20740-3823. E-mail: xliang@umd.edu

1. Introduction

The Bermuda high (also known as the Azores high) situated over the North Atlantic Ocean (Davis et al. 1997) is associated closely with the regional climate over the eastern United States (Stahle and Cleaveland 1992; Henderson and Vega 1996; Katz et al. 2003; Diem 2006; Ortegren et al. 2011; Li et al. 2011). In summer, the broad ridge of the high pressure system extending from the center around the Azores islands engulfs most of the eastern United States, producing clear sky and often stagnant air conditions therein. The west extension of the Bermuda high also defines the low-level jet (LLJ) of southerly flows over the southern Great Plains that transports warm, humid, and clean marine air from the Gulf of Mexico to the southern Great Plains–Midwest (Higgins et al. 1997). All these studies, however, have focused on climate responses only over the southeastern United States to Bermuda high variations.

Surface ozone concentrations are affected by regional circulation and climate conditions through complex dynamical, physical, and chemical processes (Vukovich and Sherwell 2003; Li et al. 2005; Camalier et al. 2007; Leibensperger et al. 2008; Jacob and Winner 2009). Episodes of high concentrations are usually related to local weather conditions that are conducive to photochemical formation of ozone, including high temperatures and clear skies, or that favor accumulation of precursors, such as low wind speeds and shallow mixing layers (Logan 1989; Jacob et al. 1993; Vukovich 1995; Bloomer et al. 2009). Meanwhile, atmospheric wind circulations strongly determine the geographic distribution of air pollutants through transport from remote sources (Auvray and Bey 2005; Kiley and Fuelberg 2006; Owen et al. 2006; Huang et al. 2008; Fang et al. 2009). Studies of episodes at a few monitoring stations have illustrated the importance of regional weather patterns on ozone distribution (Heidorn and Yap 1986; Comrie and Yarnal 1992; Hogrefe et al. 2004). In particular, ozone distribution is notably influenced by synoptic patterns that are associated with the enhancement and westward extension of the Bermuda high (Mao and Talbot 2004a,b; Chen et al. 2007; Darby et al. 2007; Hegarty et al. 2007). However, a concrete, quantitative relationship between U.S. ozone distribution and Bermuda high variation has yet to be established at a long-term climate scale. The present study offers just such a climatic effort to test this basic belief within the air quality community.

The Bermuda high, while critical for U.S. climate and air quality, was poorly simulated by the global general circulation models (GCMs) and regional climate models (RCMs) used in several climate change studies (Cooter et al. 2007; Gilliam et al. 2006; Gustafson and Leung 2007; Martin and Schumacher 2011). The strong influence of the driving GCM climate biases on the RCM-downscaled results was found to be inescapable (Liang et al. 2001, 2006; Gustafson and Leung 2007). Nolte et al. (2008) demonstrated that the biases in their RCM simulation over the eastern United States can be largely attributed to the unrealistic northeastward displacement of the Bermuda high by the driving GCM and consequently produce significant differences in the modeled ozone distribution from observations.

This study presents a systematic investigation of the Bermuda high impacts on U.S. climate and air quality at much broader temporal and spatial scales (than episodic cases or specific sites in the literature) using observations and simulations from both global and regional modeling systems. Section 2 describes the observational data and the model simulations used. Section 3 develops an index for the Bermuda high (BHI) to characterize its regional structure that is associated with LLJ. Section 4 quantifies the effects on the climate and ozone variations in the eastern United States. In particular, the observed relationships of BHI are identified with precipitation, surface air temperature (SAT), and surface ozone. Section 5 then compares the ability of the global and regional modeling system in reproducing these relationships. The skill of all available GCMs in simulating the Bermuda high is further examined in section 6 for the regional characteristics and in section 7 for remote responses to sea surface temperature (SST). Final conclusions are given in section 8.

2. Observational data and model simulations

Daily observational data for four fields during 1979–2010 are used to develop the BHI and its associated regional climate patterns as the base for model evaluation. Sea level pressure (SLP) and 850-hPa meridional wind (V850) are from the National Centers for Environmental Prediction (NCEP) reanalysis (Kalnay et al. 1996), available on the global 2.5° × 2.5° grid. For a quantitative comparison, they are mapped onto the 30 km × 30 km grid using bilinear spatial interpolation. Daily total precipitation and daily-mean (average of maximum and minimum) surface air temperature data are based on measurements from 7235 National Weather Service cooperative stations over the contiguous United States. They are mapped onto the 30 km × 30 km grid following the objective analysis of Liang et al. (2004b) with the topographic adjustment.

To investigate the Bermuda high impacts on air quality, surface ozone hourly measurements available for 1993–2009 at 1885 stations from the U.S. Environmental Protection Agency (EPA) Air Quality System (AQS; see online at http://www.epa.gov/ttn/airs/airsaqs/detaildata/downloadaqsdata.htm) are used to construct maximum daily 8-h average (MDA8). The sparse station distribution does not justify mapping onto the RCM grid. Instead, the observational analysis below is based on individual stations or averaging over broad regions with consistent variations. To explore the planetary mechanisms associated with BHI variations, the monthly-mean SST observational analysis data on global 1° × 1° grids since 1870 are adopted from Hurrell et al. (2008).

This study also analyzes climate and air quality simulations from both global and regional modeling systems. Over the globe, climate variations are simulated by the Parallel Climate Model (PCM; Washington et al. 2000), which in turn drive the Model for Ozone and Related Chemical Tracers (MOZART) to produce ozone distributions (Lin et al. 2008). These global simulations at the T42 resolution (~2.8° or 300 km) provide lateral boundary conditions that force the regional models to downscale regional climate and air quality onto 30 km × 30 km grids over the United States. The RCM is the climate extension of the fifth-generation Pennsylvania State University–National Center for Atmospheric Research (NCAR) Mesoscale Model (MM5; Dudhia et al. 2005), named CMM5 (Liang et al. 2004a). The CMM5 has considerable downscaling skill for the U.S. climate, including the diurnal cycle, annual cycle and interannual variation of precipitation (Liang et al. 2004a,b; Zhu and Liang 2007). This study adopts the CMM5 using the Grell (1993) cumulus scheme as driven by PCM, referred to as PGR. Given the MOZART chemical lateral forcing and CMM5 climate driving conditions, surface ozone variations on the 30 km × 30 km grids are predicted by the regional Air Quality Model (AQM; Chang et al. 1997; Huang and Chang 2001). The AQM has been shown capable of simulating the observed regional ozone distributions on diurnal to seasonal scales (Huang et al. 2007, 2008). The PCM and CMM5 climate simulations are available for 1991–2000, while the MOZART and AQM ozone simulations are limited to 1996–2000 mainly by the computational resources. For a consistent comparison, only the overlap period 1996–2000 is used in the analysis.

To evaluate the current capability of modeling the Bermuda high regional characteristics and underlying mechanisms, this study further compares all available global simulations by 16 separate GCMs worldwide [see Table 1 of Kunkel et al. (2006) for the basic model information and more details online at http://www-pcmdi.llnl.gov/ipcc/model_documentation/ipcc_model_ documentation.php]. Several GCMs have multiple runs differing only in initial conditions, resulting in a total of 51 realizations. Each realization is treated here as if it was an independent model simulation. Since the starting and ending dates vary among the realizations, the common period 1970–99 is used in the comparison.

3. Development of a regional Bermuda high index

There are several indices that depict Bermuda high variations. Stahle and Cleaveland (1992) first defined an index as the SLP difference between two specific locations, Bermuda (40°N, 60°W) minus New Orleans (30°N, 90°W). Subsequent studies computed similar indices using the same site at New Orleans but different locations near Bermuda, such as (32.5°N, 65°W) by Katz et al. (2003) and (35°N, 65°W) by Ortegren et al. (2011). These indices measure the pressure gradient across the southern United States and southwestern North Atlantic (Stahle and Cleaveland 1992). Positive values often indicate the western edge of the Bermuda high located farther east than its normal position (Katz et al. 2003) and may correspond with enhanced southerly moisture advection and reduced stability over the southeast United States (Henderson and Vega 1996). Recently, Li et al. (2011) defined two indices based on the 850-hPa geopotential height distribution to separately depict the center intensity and western ridge of the Bermuda high and showed that the center intensity variability explains 38% of the ridge westward extension.

All the previous indices were specified from the general features of the Bermuda high itself rather than from the perspective of its impacts on the U.S. regional climate. As such, observed relationships with these indices have been limited to precipitation variations over the U.S. Southeast. This study seeks to define the BHI that emphasizes the Bermuda high impacts on the U.S. regional climate. The Bermuda high contains many modes of spatiotemporal variations (Davis et al. 1997) and also many structural characteristics, including the center intensity and location, the influence area exceeding a certain pressure, the ridge direction and extension, etc. The 1979–2008 summer-mean (June–August) climatological SLP distribution (Fig. 1a) shows that the broad ridge of the Bermuda high engulfs most of the eastern United States. Along the west edge of the Bermuda high is the LLJ, with mean V850 stronger than 6 m s−1 located exclusively in the southern Great Plains.

Fig. 1.
Fig. 1.

(a) Summer-mean SLP distribution with V850 stronger than 6 m s−1 shaded by scarlet and the summer interannual correlation coefficients of SLP with (b) the LLJ and (c) the BHI during 1979–2008. Correlations with absolute values greater than 0.35 are statistically significant at 95% confidence, assuming yearly independence. The black boxes in (b) are the two centers for defining the BHI.

Citation: Journal of Climate 26, 3; 10.1175/JCLI-D-12-00168.1

The LLJ plays an essential role on the eastern U.S. climate variation, especially for precipitation distribution in the southern Great Plains and the Midwest (Higgins et al. 1997; Zhu and Liang 2005). When the LLJ is intensified, stronger southerly flows prevail along the Great Plains and thus more moisture is transported toward the Midwest, producing heavier rainfall there and less precipitation to the south. The opposite occurs when the LLJ is weakened. Following Liang et al. (2004a), an index for the LLJ intensity is defined as the regional mean of V850 over the southern Great Plains with maximum speeds (28.3°–34°N, 100.5°–97.4°W) (see Fig. 1a for the boundaries). A subsequent interannual correlation analysis shows that in summer the LLJ intensity varies closely with the SLP gradient from the high over the Gulf of Mexico to the low in the southern Great Plains, as well as an overall expansion of the Bermuda high toward the south, east, and west (Fig. 1b). Thus, the intensity of the LLJ and the extension of the Bermuda high are closely coupled. As the Bermuda high extends its ridge more westward, SLP rises over the Gulf of Mexico but falls in the southern Great Plains. This results in a larger SLP gradient that produces a stronger southerly flow or the LLJ.

The above analysis leads to our development of the BHI to quantify the impacts of the Bermuda high on regional climate and air quality over the eastern United States. The BHI is defined as the difference of regional-mean SLP between the Gulf of Mexico (25.3°–29.3°N, 95°–90°W) and the southern Great Plains (35°–39°N, 105.5°–100°W) (see the outlined boxes in Fig. 1b). The summer interannual SLP correlation with the BHI (Fig. 1c), mostly replicating the SLP pattern associated with the LLJ (Fig. 1b), confirms that the BHI depicts the regional SLP oscillation between the Gulf of Mexico and the southern Great Plains caused by the westward extension or contraction of the Bermuda high. This oscillation or western edge movement of the Bermuda high also has strong positive correlations with SLP along its south and east flanks but rather weak association in the center. Therefore, a larger BHI shows a systematic expansion of the Bermuda high toward its south, east, and west. In opposite, a smaller BHI indicates a shrinking of the system.

The BHI developed in this study differs from the previous indices. Our BHI is defined as the regional-mean (rather than site-specific) SLP gradient and has its positive center over the Gulf of Mexico, which contains New Orleans, and the negative center in the southern Great Plains, which is in the west, far away from Bermuda. As such, our BHI characterizes the regional structure of the Bermuda high that determines the LLJ intensity and more directly affect the eastern United States over a broader area than previously identified (see below). The BHI has certain advantages over the LLJ index. In particular, observations of wind are less accurate than those of SLP. Records of measured or reconstructed proxy data for SLP are available from the beginning of the nineteenth century. Model simulations of SLP, including NCEP–NCAR reanalysis products and GCM outputs used here, are also better and more readily available than those of smaller-scale wind feature that are much more difficult to detect. As such, the BHI is more favorable to use for study of the regional circulation patterns and interannual to decadal changes of the Bermuda high over the eastern United States.

4. Observed Bermuda high impacts on the U.S. climate and air quality

Figure 2a illustrates the observed summer precipitation interannual correlation coefficients with the BHI during 1979–2008. There are significant positive correlations in the Midwest and negative correlations over the southern states along the Gulf of Mexico. For a positive BHI anomaly (relative to its 1979–2008 average), the Bermuda high extends farther westward than its climatological-mean position. Consequently, the LLJ along the west flank of the high is enhanced and transports more moisture farther north. This results in larger moisture convergence into the region of greater convection activities (Easterling 1990) and hence more rainfall in the Midwest. The opposite condition occurs over the southern Great Plains, suffering from drought due to abnormal moisture loss by transport to the north. Meanwhile the southeast states are drier because of the prevailing clear-sky conditions under the control of the Bermuda high extended westward. This pattern of opposite precipitation anomalies between the Midwest and the Gulf States covers more than half of the continental United States, much broader than the southeastern United States identified in previous studies (Stahle and Cleaveland 1992; Henderson and Vega 1996; Katz et al. 2003; Diem 2006; Ortegren et al. 2011).

Fig. 2.
Fig. 2.

Geographic distributions of the summer interannual correlation coefficients of the BHI with (a) precipitation, (b) SAT, and (d) V850 during 1979–2008 and (c) MDA8 at all monitoring sites with data records of at least 10 summers in 1993–2008. The statistical significance is as in Fig. 1, but MDA8 assumes monthly independence. (e) Normalized summer interannual anomalies of the BHI (black) and MDA8 over the southern Great Plains–Midwest (blue) and the eastern United States (red). Their standard deviations used for the normalization are listed in the legend. The black boxes in (a),(b),(d) are where the regional-mean precipitation (PR), SAT, and V850 indices are defined, and the colored ones in (c) are for MDA8 in (e).

Citation: Journal of Climate 26, 3; 10.1175/JCLI-D-12-00168.1

Figure 2b depicts the corresponding SAT correlation pattern. When the BHI is above average, there is a broad band of significant warmer temperature anomalies across virtually the entire Gulf States. The enhanced LLJ advects more warm marine air from the Gulf of Mexico to the southern Great Plains, while the westward extension of the Bermuda high reduces cloud cover and increases radiation over the Southeast. Both effects result in higher temperatures than normal over the respective regions under influence. In addition, relatively weak cold anomalies appear over the Northwest. They are likely regional responses to the circulation anomalies in the North Pacific also associated with the Bermuda high (see Fig. 1c).

The concentration of surface ozone, which is widely recognized as a major cause for respiratory disease, depends on local meteorological conditions and chemical reactions. Since the record of surface ozone measurements is relatively short (<17 yr and with missing values), we use monthly data to calculate correlation coefficients after removing the climatological mean for each month. Using monthly-mean data rather than seasonal-mean data triples the number of samples such that a meaningful statistical significance of the correlation result may be reached. In addition, ozone geographic distributions evolve largely during summer, requiring a shorter time scale to better depict its variations. The summer correlations for ozone are thus calculated based on all monthly values of June, July, and August during 1993–2008 at each monitoring site that has a sufficient data record (>10 yr with more than 30 samples). For consistency, the corresponding BHI data used in this calculation are also monthly anomaly values.

Figure 2c shows summer BHI interannual correlations with MDA8 at all monitoring sites with sufficient data records. The correlation is negative across the southern Great Plains to the Midwest and positive along the eastern coastal states. This result may be explained by the pollutant transport and stagnation effect, which are induced by the Bermuda high. In fact, for a positive BHI anomaly, the Bermuda high extends more westward, causing meridional wind V850 to significantly increase over a broad region across the southern Great Plains to the Midwest (Fig. 2d). The enhanced LLJ brings more relatively clean marine air from the Gulf of Mexico into the southern Great Plains, with the influence extended farther north to the Midwest. As a result of the northeastward transport, ozone decreases along the broad path across the southern Great Plains to the Midwest. On the other hand, larger transport of more polluted air from the Midwest leads to ozone increase in New England. Meanwhile, under the control of the Bermuda high extended westward, more frequent air stagnation occurs over the eastern United States. Both effects generate higher ozone concentrations along the eastern coastal states.

Note that the relationships as revealed in Fig. 2 of rainfall, temperature, and ozone with BHI interannual variations greatly differ, mainly because of the relative roles of meridional advection, between the central United States and eastern coast. Clouds and rainfall enhance the scavenging of tropospheric ozone and its precursors such as CH2O and NOx (Lelieveld and Crutzen 1991). Ozone also has little production during rainy days because of the lack of photolysis. Thus, rainfall tends to reduce ambient ozone concentration (Muralidharan et al. 1989). On the other hand, warmer temperatures are associated with higher photolysis and favors ozone production (Camalier et al. 2007; Bloomer et al. 2009). Therefore, surface ozone is anticipated to correlate negatively with rainfall and positively with temperature on the local basis. Such relationships are clearly manifested in the Southeast, where meridional advection is weak. They are, however, reversed in sign over the southern Great Plains, where southerly advection of moist, warm, and clean air from the Gulf of Mexico dominates the local effects.

Figure 2e presents interannual variations of the BHI during 1979–2010 and the summer MDA8 regional means over the southern Great Plains–Midwest and the eastern coastal states (the blue and red boxes, respectively, in Fig. 2c) during 1993–2009 using all the monitoring sites with sufficient data records. The variations of the MDA8 and the BHI are clearly out of phase in the southern Great Plains–Midwest but in phase over the eastern coastal states, with a significant correlation coefficient of −0.50 and 0.67, respectively. On average of the summer using a linear regression, the magnitude of the MDA8 variations induced by the 95% range change (± two standard deviations) of the BHI is 6 ppb in the southern Great Plains–Midwest and 13.5 ppb over the eastern coastal states. These numbers are substantially large as compared with the U.S. EPA National Ambient Air Quality Standard (NAAQS). In particular, the summer of 1993 was the extreme of the two decades, when the Bermuda high had abnormally stronger pressure in the center and extends much more westward than average, causing significantly lower (higher) MDA8 in the southern Great Plains–Midwest (eastern coastal states). While it brought a low ozone “breeze” for the southern Great Plains–Midwest, it also caused ozone level in the eastern states to violate the NAAQS.

The above finding on the Bermuda high impacts on U.S. ozone is consistent with earlier studies. Hogrefe et al. (2004) and Chen et al. (2007) found that surface ozone episodes at a few stations in New England were related to weather patterns influenced by the Bermuda high. The correlation patterns shown in Fig. 2c, however, are at much broader temporal and spatial scales. More importantly, the Bermuda high has clear decadal variations (Fig. 2e), with the peaks (more westward extensions) in 1980 and 1993, while the lows are in 1989, 1995, and 2004. As such, regional ozone also exhibits decadal variations, which should be accounted for in the dynamic management of the U.S. air quality.

Li et al. (2011) showed that the Bermuda high in the last 30 yr (1978–2007) become more intense, mainly because of anthropogenic warming, and its western ridge is displaced westward. In contrast, our BHI returned to normal in 2008–09 from negative one deviation in 2007 and exhibited no obvious trend since 1979. As discussed earlier, our BHI is defined to emphasize the regional manifestation of the Bermuda high over the United States and thus may capture its characteristics of variability different from previous studies.

5. Simulations by the global and regional modeling systems

Figure 3 compares interannual correlations of the BHI with precipitation, SAT, MDA8, and V850 as simulated by the regional (CMM5 and AQM) and global (PCM and MOZART) modeling systems. To maximize the statistical measure on possible signal identification from relatively short data records, the correlation analysis of the model simulations is based on monthly means in each summer (June–August) of all 5 yr during 1996–2000. As such, the total number of data samples is 15 for both climate and ozone simulations.

Fig. 3.
Fig. 3.

Geographic distributions of the summer interannual correlation coefficients of the BHI with precipitation for (a) PGR and (b) PCM; SAT for (c) PGR and (d) PCM; MDA8 for (e) AQM/PGR and (f) MOZART/PCM; and V850 for (g) PGR and (h) PCM based on summer monthly data for 1996–2000.

Citation: Journal of Climate 26, 3; 10.1175/JCLI-D-12-00168.1

The CMM5 reproduces the observed out-of-phase correlation pattern of the BHI with precipitation between the Midwest and the Gulf States (Fig. 3a). Although the negative center is overexpanded into the Northwest and the positive area shrinks within the Midwest compared to observations, the CMM5-downscaled result (PGR) largely correct the biases in New England and the southern Great Plains from the driving PCM (Fig. 3b). For the correlation pattern of the BHI with SAT, both the CMM5 and PCM capture the main signal, where warmer than normal temperatures prevail over the Gulf States in response to a positive BHI anomaly (Figs. 3c,d). However, as compared to observations (Fig. 2b), much stronger positive correlations are simulated by the PCM over the Southwest, although the CMM5 reduces this tendency somewhat.

The AQM captures the main characteristics of the BHI-induced MDA8 pattern, where surface ozone reverses variation tendency between the southern Great Plains and the eastern coastal states as the Bermuda high extends westward or contracts eastward (Fig. 3e). Compared to observations (Fig. 2c), the AQM-simulated negative correlations are weaker in the northern Great Plains and the Midwest. Similarly, the MOZART captures the opposite correlations over the southern Great Plains and the eastern coastal states (Fig. 3f). However, the strong positive correlations in the northern Great Plains and the Midwest are not evident in observations. This MOZART deficiency is largely reduced by the AQM, mainly because the Bermuda high–induced meridional wind responses and hence pollutant transport over the Great Plains are more realistic (see Fig. 2d for observations) in the downscaling CMM5 (Fig. 3g) than the driving PCM (Fig. 3h).

6. GCM ability for Bermuda high impacts on U.S. climate

As the essential provider of the climate conditions for the RCM downscaling and together driving the global and regional air quality simulations, the most recent GCMs are evaluated for their abilities in representing the Bermuda high impacts on U.S. climate. Figure 4 compares the observed and simulated BHI relationships with summer-mean precipitation, SAT, and V850 interannual variations averaged over the key centers of teleconnection (boxes marked in Figs. 2a,b,d). These centers contain strong correlations with the BHI in observations. Among the 51 realizations from 16 GCMs, about half capture the observed correlations with precipitation over either the southern Great Plains or the Midwest, with the statistical significance better than the 95% confidence. About two-thirds and almost 80% of the realizations correctly depict the significant positive BHI correlations with SAT and V850, respectively, over the southern Great Plains. However, only seven realizations, each of a different GCM, manifest the ability in representing simultaneously all the significant BHI relationships with precipitation, SAT, and V850 (Table 1).

Fig. 4.
Fig. 4.

Correlation coefficients between summer BHI time series and (a) precipitation averaged over the southern Great Plains, (b) precipitation averaged over the Midwest, (c) SAT averaged over the southern Great Plains, and (d) V850 averaged over the southern Great Plains for twentieth-century climate simulation (20C3M) model experiments and observations during 1970–99. The number after the comma in the label indicates the run number for those models with multiple realizations. The dashed horizontal line indicates the value of apparent statistical significance at the 95% level of confidence.

Citation: Journal of Climate 26, 3; 10.1175/JCLI-D-12-00168.1

Table 1.

Summer interannual correlations of the BHI with PR, SAT, and V850 averaged over the respective key regions (see the boxes in Figs. 2a,b,d) as observed and simulated by the GCM runs that well capture the observed regional signatures of the Bermuda high. The left-hand column is from observations, CNRM, Geophysical Fluid Dynamics Laboratory (GFDL), L’Institut Pierre-Simon Laplace (IPSL), Meteorological Research Institute (MRI), MPI, PCM, and Met Office (UKMO).

Table 1.

The above result indicates the challenge in modeling the Bermuda high variations and teleconnection patterns as revealed in Fig. 2. To our knowledge, there is no study that specifically compares GCM performance of these Bermuda high features. Note that no Goddard Institute for Space Studies Model H (GISS-H) realization shows signature of the Bermuda high impacts on the U.S. climate. With a lower resolution, GISS-R unexpectedly captures the observed BHI correlations with precipitation in the Midwest but fails in all other aspects. Sensitivity experiments with that GCM would be helpful to understand the interplay between the planetary and regional circulation aspects of the Bermuda high variations.

On the other hand, the Centre National de Recherches Météorologiques model (CNRM) and the PCM.1 perform best among all 51 realizations. Figure 5 displays the geographic distributions of summer interannual BHI correlations with precipitation, SAT, and V850 simulated by these two GCMs. Both models capture the major characteristics of the observed BHI patterns: the positive Midwest and negative Gulf States precipitation oscillation, the positive SAT correlations over the Gulf States, and the positive V850 responses over the southern Great Plains. However, both GCMs contain notable discrepancies from observations. The precipitation negative correlations over the Gulf States by the CNRM (Fig. 5a) and positive correlations in the Midwest by the PCM.1 (Fig. 5b) are overestimated with much broader areas than observations. The PCM.1 misses the Southeast negative precipitation correlation center. Both GCMs overestimate positive correlations with SAT in the Gulf States (Fig. 5c,d) but realistically simulate those with V850 over the southern Great Plains. In addition, the accompanying negative correlations around Cuba are correctly depicted by the PCM.1 but overestimated and misplaced by the CNRM.

Fig. 5.
Fig. 5.

Geographic distributions of the summer interannual correlation coefficients of the BHI during 1970–99 with precipitation for (a) CNRM and (b) PCM1; with SAT for (c) CNRM and (d) PCM1; and with V850 for (e) CNRM and (f) PCM1, which contains missing values below high terrains because of no extrapolation from the lowest model layer of a coarse-resolution grid as in the raw data.

Citation: Journal of Climate 26, 3; 10.1175/JCLI-D-12-00168.1

Note that the BHI correlation patterns with precipitation, SAT, and V850 simulated by the PCM (Figs. 3b,d,h) resemble those by the PCM.1 (Figs. 5b,d,f). This similarity arises from the use of the same PCM model configuration. However, important regional differences exist as a result of changes in the initial condition, integration period, and experiment design. For example, the PCM.1 simulates stronger positive correlations with V850 along the southern Great Plains, which agrees with observations (Fig. 2d) better than the PCM.

7. Bermuda high’s response to SST forcing

Large-scale climate systems, including the Bermuda high, strongly interact with global SST variations. Figures 6a,b shows the observed summer BHI (single yearly value) interannual correlations with global SST distributions for a 30-yr period of 1970–99 in the concurrent summer and prior winter. [The observed SSTs are from Hurrell et al. (2008), and the period is chosen to match the GCM data availability.] For the summer, strong negative SST correlations exist over the eastern Atlantic Ocean (EA) along the west coast of North Africa. Since the Bermuda high center generally sits to the west of that region (Fig. 1a), the negative correlations likely result from larger cold SST advection induced by the stronger northerly flow along the eastern flank of the intensified Bermuda high. In the prior winter, strong positive SST correlations occur over the eastern Pacific (EP) near California. Thus, the SSTs in the EP may contain predictive signal for the Bermuda high strength, defined here as the SLP gradient between the Gulf of Mexico and the southern Great Plains. The higher the prior winter SSTs are over the EP, the stronger the Bermuda high is in the summer.

Fig. 6.
Fig. 6.

Geographic distributions of the summer interannual correlation coefficients of the BHI during 1970–99 with observed SST (a) in the previous winter and (b) in the concurrent summer; with CNRM-simulated SST (c) in the previous winter and (d) in the concurrent summer; and with PCM1-simulated SST (e) in the previous winter and (f) in the concurrent summer. The black boxes in (a),(b) are the EP and EA regions defining the SST index, respectively.

Citation: Journal of Climate 26, 3; 10.1175/JCLI-D-12-00168.1

The corresponding BHI–SST correlation patterns simulated by the CNRM and PCM.1 are shown in Figs. 6c–f. As discussed earlier, the two GCMs perform best among all 51 simulations in depicting observed BHI correlations with regional precipitation, SAT, and V850 in the United States. However, neither of them is able to capture the BHI relationships with SSTs over the two key regions (winter EP and summer EA) identified above. Figure 7 compares the summer BHI interannual correlations with the regional-mean SSTs over the EP in the prior winter and the EA in the concurrent summer from all 51 GCM realizations against observations. No GCM reproduces both the positive winter EP and negative summer EA SST correlations. Separately, the EP signal exists only in the Max Planck Institute model realization 1 (MPI.1), while the EA signal appears only in the Model for Interdisciplinary Research on Climate medium-resolution realization 3 (MIROCm.3). All other realizations produce correlations that are either not statistically significant or totally opposite to observations. None of the seven GCM runs that have the correct BHI signatures on the U.S. regional climate as listed in Table 1 can simulate these SST signals. The result indicates a great challenge for the GCMs to predict the Bermuda high variability and change as well as substantial uncertainties for using these GCMs’ simulations to study the Bermuda high impacts on U.S. regional climate and air quality.

Fig. 7.
Fig. 7.

Correlation coefficients between summer BHI time series and (a) SST in previous winter averaged over the key area boxed in Fig. 6a and (b) SST in the concurrent summer averaged over the key area boxed in Fig. 6b.

Citation: Journal of Climate 26, 3; 10.1175/JCLI-D-12-00168.1

8. Conclusions

This paper presents the observed evidence of the Bermuda high impacts on regional climate and air quality variations over the eastern United States and evaluates the global and regional models’ ability in simulating these relationships. The BHI, defined as the SLP difference between the Gulf of Mexico and the southern Great Plains, is used to characterize the regional structure of the Bermuda high, including its westward extension and close LLJ correspondence. For a positive BHI anomaly in summer, the Bermuda high extends more westward. This enhances the LLJ along the west flank of the high, which in turn transports more moisture farther north. As a result, rainfall increases in the Midwest but decreases to the south in the Gulf States. The enhanced LLJ advects more warm air from the Gulf of Mexico to the southern Great Plains, while the westward extension of the Bermuda high reduces cloud cover and increases radiation over the Southeast. Together the two effects produce significant warmer temperature anomalies across virtually the entire Gulf States. The opposite conditions apply for a negative BHI anomaly.

Since the marine air over the Gulf of Mexico is also cleaner than inland, the enhanced transport reduces ozone concentrations along the LLJ path across the southern Great Plains to the Midwest. On the other hand, larger transport of more polluted air from the Midwest to New England and more frequent air stagnation under the control of the Bermuda high extended westward over the Southeast lead to higher ozone concentrations along most of the eastern coastal states. During the last two decades with sufficient data records, summer MDA8 interannual variations are strongly correlated with the BHI, out of phase in the southern Great Plains–Midwest but in phase over the eastern coastal states, with the respective magnitude of 6 and 13.5 ppb in summer seasonal-mean MDA8. While it brings a low ozone “breeze” for the southern Great Plains–Midwest, the enhanced Bermuda high increases ozone hazards over the eastern states. In addition, the Bermuda high has clear decadal variations, with the peaks (stronger central pressures or more westward extensions) in 1980 and 1993 and the lows in 1989, 1995, and 2004. As such, regional circulation–induced ozone also exhibits decadal variations, which should be accounted for in the U.S. air quality regulation.

The major characteristics of the above-observed Bermuda high impacts on regional climate and air quality over the eastern United States can be well simulated by the regional CMM5–AQM as driven by the global PCM–MOZART. However, notable model deficiencies exist, especially over the northern Great Plains and the Midwest that are more remote to the BHI and LLJ control. In this regard, the downscaling regional models are more skillful than the driving global models in reproducing the U.S. climate and air quality responses to the Bermuda high variations, with significant reductions of biases over the remote areas.

As the necessary provider of the driving conditions for regional models, the GCMs with all available simulations are further examined on their abilities in capturing the observed Bermuda high variations and its impacts on the U.S. climate. Only 7 out of 51 realizations, each of a different GCM, can simultaneously represent all the significant BHI relationships with the key regional means of precipitation, SAT, and V850 over the eastern United States. Observational analysis also reveals that summer BHI interannual variations have a strong predictive signal of positive SST anomalies over the eastern Pacific near California in the prior winter and a significant response of negative SST anomalies over the eastern Atlantic Ocean along the west coast of North Africa in the concurrent summer. Unfortunately, none of the GCM simulations, including the seven realizations that have the correct BHI signatures on the U.S. regional climate, can simulate these SST patterns. The result indicates a great challenge for the GCMs to predict the Bermuda high variability and change as well as substantial uncertainties for using these GCMs’ simulations to study the Bermuda high impacts on U.S. regional climate and air quality.

In conclusion, the Bermuda high plays a critical role on regional climate and air quality variations over the eastern United States. Its swing between westward extension and eastward contraction can cause significant regional anomalies of precipitation, surface air temperature, and surface ozone. The strong decadal variations of such swing and likely future changes under global warming should be considered in the dynamic management of the U.S. air quality. To better understand and more credibly predict the U.S. climate and air quality, it is essential to further improve the ability of the global and regional models in representing the Bermuda high system structure. The general failure of the current GCMs in capturing the Bermuda high association with planetary SST anomalies casts a serious doubt on their direct use to provide credible climate change projections at regional scales. The downscaling by the regional models, due to resolution refinement and physics improvement (e.g., Liang et al. 2004b, 2006, 2012), is demonstrated to notably reduce the GCMs’ deficiency in depicting the regional signatures of the Bermuda high. We speculate that improving GCMs’ ability on the Bermuda high planetary features will enable accurate prediction of the regional climate and air quality variations and future changes over the eastern United States, especially through nesting with the regional models.

Acknowledgments

We thank Shenjian Su and Hang Lei for providing the AQM and MOZART simulations of ozone. We acknowledge NOAA/ESRL and NCSA/UIUC for the supercomputing support. The research was partially supported by the United States Environmental Protection Agency under Awards EPA RD-83337302 and RD-83418902. The views expressed are those of the authors and do not necessarily reflect those of the sponsoring agencies or the University of Maryland.

REFERENCES

  • Auvray, M., and I. Bey, 2005: Long-range transport to Europe: Seasonal variations and implications for the European ozone budget. J. Geophys. Res., 110, D11303, doi:10.1029/2004JD005503.

    • Search Google Scholar
    • Export Citation
  • Bloomer, B. J., J. W. Stehr, C. A. Piety, R. J. Salawitch, and R. R. Dickerson, 2009: Observed relationships of ozone air pollution with temperature and emissions. Geophys. Res. Lett., 36, L09803, doi:10.1029/2009GL037308.

    • Search Google Scholar
    • Export Citation
  • Camalier, L., W. Cox, and P. Dolwick, 2007: The effects of meteorology on ozone in urban areas and their use in assessing ozone trends. Atmos. Environ., 41, 71277137, doi:10.1016/j.atmosenv.2007.04.061.

    • Search Google Scholar
    • Export Citation
  • Chang, J. S., S. Jin, Y. Li, M. Beauharnois, C.-H. Lu, H.-C. Huang, S. Tanrikulu, and J. Damassa, 1997: The SARMAP Air Quality Model final report. California Environmental Protection Agency Air Resource Board Rep., 160 pp.

  • Chen, M., R. Talbot, H. Mao, B. Sive, J. Chen, and R. J. Griffin, 2007: Air mass classification in coastal New England and its relationship to meteorological conditions. J. Geophys. Res., 112, D10S05, doi:10.1029/2006JD007687.

    • Search Google Scholar
    • Export Citation
  • Comrie, A. C., and B. Yarnal, 1992: Relationships between synoptic-scale atmospheric circulation and ozone concentrations in metropolitan Pittsburgh, Pennsylvania. Atmos. Environ., 26B, 301312.

    • Search Google Scholar
    • Export Citation
  • Cooter, E. J., J. Swall, and R. Gilliam, 2007: Comparison of 700-hPa NCEP-R1 and AMIP-R2 wind patterns over the continental United States using cluster analysis. J. Appl. Meteor., 46, 17441758.

    • Search Google Scholar
    • Export Citation
  • Darby, L. S., and Coauthors, 2007: Ozone differences between near-coastal and offshore sites in New England: Role of meteorology. J. Geophys. Res., 112, D16S91, doi:10.1029/2007JD008446.

    • Search Google Scholar
    • Export Citation
  • Davis, R. E., B. P. Hayden, D. A. Gay, W. L. Phillips, and G. V. Jones, 1997: The North Atlantic subtropical anticyclone. J. Climate, 10, 728744.

    • Search Google Scholar
    • Export Citation
  • Diem, J. E., 2006: Synoptic-scale controls of summer precipitation in the southern United States. J. Climate, 19, 613621.

  • Dudhia, J., D. Gill, Y. R. Guo, K. Manning, W. Wang, and C. Bruyere, 2005: PSU/NCAR mesoscale modeling system tutorial class notes and user’s guide: MM5 modeling system version 3.7. [Available online at http://www.mmm.ucar.edu/mm5/documents/tutorial-v3-notes.html.]

  • Easterling, D. R., 1990: Persistent patterns of thunderstorm activity in the central United States. J. Climate, 3, 13801389.

  • Fang, Y., A. M. Fiore, L. W. Horowitz, A. Gnanadesikan, H. Levy II, Y. Hu, and A. G. Russell, 2009: Estimating the contribution of strong daily export events to total pollutant export from the United States in summer. J. Geophys. Res., 114, D23302, doi:10.1029/2008JD010946.

    • Search Google Scholar
    • Export Citation
  • Gilliam, R. C., C. Hogrefe, and S. T. Rao, 2006: New methods for evaluating meteorological models used in air quality applications. Atmos. Environ., 40, 50735086.

    • Search Google Scholar
    • Export Citation
  • Grell, G. A., 1993: Prognostic evaluation of assumptions used by cumulus parameterizations. Mon. Wea. Rev., 121, 764787.

  • Gustafson, W. I., Jr., and L. R. Leung, 2007: Regional downscaling for air quality assessment: A reasonable proposition? Bull. Amer. Meteor. Soc., 88, 12151227.

    • Search Google Scholar
    • Export Citation
  • Hegarty, J., H. Mao, and R. Talbot, 2007: Synoptic controls on summertime surface ozone in the northeastern United States. J. Geophys. Res., 112, D14306, doi:10.1029/2006JD008170.

    • Search Google Scholar
    • Export Citation
  • Heidorn, K. C., and D. Yap, 1986: A synoptic climatology for surface ozone concentrations in southern Ontario, 1976-1981. Atmos. Environ., 20, 695703.

    • Search Google Scholar
    • Export Citation
  • Henderson, K. G., and A. J. Vega, 1996: Regional precipitation variability in the southern United States. Phys. Geogr., 17, 93112.

  • Higgins, R. W., Y. Yao, E. S. Yarosh, J. E. Janowiak, and K. C. Mo, 1997: Influence of the Great Plains low-level jet on summertime precipitation and moisture transport over the central United States. J. Climate, 10, 481507.

    • Search Google Scholar
    • Export Citation
  • Hogrefe, C., J. Biswas, B. Lynn, K. Civerolo, J.-Y. Ku, J. Rosenthal, C. Rosenzweig, R. Goldberg, and P. L. Kinney, 2004: Simulating regional-scale ozone climatology over the eastern United States: Model evaluation results. Atmos. Environ., 38, 26272638.

    • Search Google Scholar
    • Export Citation
  • Huang, H.-C., and J. S. Chang, 2001: On the performance of numerical solvers for chemistry submodel in three-dimensional air quality models: Box-model simulations. J. Geophys. Res., 106 (D17), 20 17520 188.

    • Search Google Scholar
    • Export Citation
  • Huang, H.-C., X.-Z. Liang, K. E. Kunkel, M. Caughey, and A. Williams, 2007: Seasonal simulation of tropospheric ozone over the midwestern and northeastern United States: An application of a coupled regional climate and air quality modeling system. J. Appl. Meteor., 46, 945960.

    • Search Google Scholar
    • Export Citation
  • Huang, H.-C., and Coauthors, 2008: Impacts of long-range transport of global pollutants and precursor gases on U.S. air quality under future climatic conditions. J. Geophys. Res., 113, D19307, doi:10.1029/2007JD009469.

    • Search Google Scholar
    • Export Citation
  • Hurrell, J. W., J. J. Hack, D. Shea, J. M. Caron, and J. Rosinski, 2008: A new sea surface temperature and sea ice boundary dataset for the Community Atmosphere Model. J. Climate, 21, 51455153.

    • Search Google Scholar
    • Export Citation
  • Jacob, D. J., and D. A. Winner, 2009: Effect of climate change on air quality. Atmos. Environ., 43, 5163.

  • Jacob, D. J., J. A. Logan, G. M. Gardner, R. M. Yevich, C. M. Spivakovsky, S. C. Wofsy, S. Sillman, and M. J. Prather, 1993: Factors regulating ozone over the United States and its export to the global atmosphere. J. Geophys. Res., 98 (D8 14 81714 826.

    • Search Google Scholar
    • Export Citation
  • Kalnay, E., and Coauthors, 1996: The NCEP/NCAR 40-Year Reanalysis Project. Bull. Amer. Meteor. Soc., 77, 437471.

  • Katz, R. W., M. B. Parlange, and C. Tebaldi, 2003: Stochastic modeling of the effects of large-scale circulation on daily weather in the southeastern U.S. Climatic Change, 60, 189216.

    • Search Google Scholar
    • Export Citation
  • Kiley, C. M., and H. E. Fuelberg, 2006: An examination of summertime cyclone transport processes during Intercontinental Chemical Transport Experiment (INTEX-A). J. Geophys. Res., 111, D24S06, doi:10.1029/2006JD007115.

    • Search Google Scholar
    • Export Citation
  • Kunkel, K. E., X.-Z. Liang, J. Zhu, and Y. Lin, 2006: Can CGCMs simulate the twentieth-century “warming hole” in the central United States? J. Climate, 19, 41374153.

    • Search Google Scholar
    • Export Citation
  • Leibensperger, E. M., L. J. Mickley, and D. J. Jacob, 2008: Sensitivity of US air quality to mid-latitude cyclone frequency and implications of 1980-2006 climate change. Atmos. Chem. Phys., 8, 70757086.

    • Search Google Scholar
    • Export Citation
  • Lelieveld, J., and P. J. Crutzen, 1991: The role of clouds in tropospheric photochemistry. J. Atmos. Chem., 12, 229267.

  • Li, Q., and Coauthors, 2005: North American pollution outflow and the trapping of convectively lifted pollution by upper-level anticyclone. J. Geophys. Res., 110, D10301, doi:10.1029/2004JD005039.

    • Search Google Scholar
    • Export Citation
  • Li, W., L. Li, R. Fu, Y. Deng, and H. Wang, 2011: Changes to the North Atlantic subtropical high and its role in the intensification of summer rainfall variability in the southeastern United States. J. Climate, 24, 14991506.

    • Search Google Scholar
    • Export Citation
  • Liang, X.-Z., K. E. Kunkel, and A. N. Samel, 2001: Development of a regional climate model for U.S. Midwest applications. Part I: Sensitivity to buffer zone treatment. J. Climate, 14, 43634378.

    • Search Google Scholar
    • Export Citation
  • Liang, X.-Z., L. Li, A. Dai, and K. E. Kunkel, 2004a: Regional climate model simulation of summer precipitation diurnal cycle over the United States. Geophys. Res. Lett., 31, L24208, doi:10.1029/2004GL021054.

    • Search Google Scholar
    • Export Citation
  • Liang, X.-Z., L. Li, K. E. Kunkel, M. Ting, and J. X. L. Wang, 2004b: Regional climate model simulation of U.S. precipitation during 1982–2002. Part I: Annual cycle. J. Climate, 17, 35103528.

    • Search Google Scholar
    • Export Citation
  • Liang, X.-Z., J. Pan, J. Zhu, K. E. Kunkel, J. X. L. Wang, and A. Dai, 2006: Regional climate model downscaling of the U.S. summer climate and future change. J. Geophys. Res., 111, D10108, doi:10.1029/2005JD006685.

    • Search Google Scholar
    • Export Citation
  • Liang, X.-Z., and Coauthors, 2012: Regional Climate-Weather Research and Forecasting Model (CWRF). Bull. Amer. Meteor. Soc., 93, 13631387.

    • Search Google Scholar
    • Export Citation
  • Lin, J.-T., D. J. Wuebbles, and X.-Z. Liang, 2008: Effects of intercontinental transport on surface ozone over the United States: Present and future assessment with a global model. Geophys. Res. Lett., 35, L02805, doi:10.1029/2007GL031415.

    • Search Google Scholar
    • Export Citation
  • Logan, J. A., 1989: Ozone in rural areas of the United States. J. Geophys. Res., 94 (D6), 85118532.

  • Mao, H., and R. Talbot, 2004a: Relationship of surface O3 to large-scale circulation patterns during two recent winters. Geophys. Res. Lett., 31, L06108, doi:10.1029/2003GL018860.

    • Search Google Scholar
    • Export Citation
  • Mao, H., and R. Talbot, 2004b: Role of meteorological processes in two New England ozone episodes during summer 2001. J. Geophys. Res., 109, D20305, doi:10.1029/2004JD004850.

    • Search Google Scholar
    • Export Citation
  • Martin, E. R., and C. Schumacher, 2011: The Caribbean low-level jet and its relationship with precipitation in IPCC AR4 models. J. Climate, 24, 59355950.

    • Search Google Scholar
    • Export Citation
  • Muralidharan, V., G. M. Kumar, and S. Sampath, 1989: Surface ozone variation associated with rainfall. Pure Appl. Geophys., 130, 4755.

    • Search Google Scholar
    • Export Citation
  • Nolte, C. G., A. B. Gilliland, C. Hogrefe, and L. J. Mickley, 2008: Linking global to regional models to assess future climate impacts on surface ozone levels in the United States. J. Geol. Res., 113, D14307, doi:10.1029/2007JD008497.

    • Search Google Scholar
    • Export Citation
  • Ortegren, J. T., P. A. Knapp, J. T. Maxwell, W. P. Tyminski, and P. T. Soulé, 2011: Ocean–atmosphere influences on low-frequency warm-season drought variability in the Gulf Coast and southeastern United States. J. Appl. Meteor. Climatol., 50, 11771186.

    • Search Google Scholar
    • Export Citation
  • Owen, R. C., O. R. Cooper, A. Stohl, and R. E. Honrath, 2006: An analysis of the mechanisms of the North American pollutant transport to the central North Atlantic lower free troposphere. J. Geophys. Res., 111, D23S58, doi:10.1029/2006JD007062.

    • Search Google Scholar
    • Export Citation
  • Stahle, D. W., and M. K. Cleaveland, 1992: Reconstruction and analysis of spring rainfall over the southeastern U.S. for the past 1000 years. Bull. Amer. Meteor. Soc., 73, 19471961.

    • Search Google Scholar
    • Export Citation
  • Vukovich, F. M., 1995: Regional-scale boundary layer ozone variations in the eastern United States and their association with meteorological variations. Atmos. Environ., 29, 22592273.

    • Search Google Scholar
    • Export Citation
  • Vukovich, F. M., and J. Sherwell, 2003: An examination of the relationship between certain meteorological parameters and surface ozone variations in the Baltimore–Washington corridor. Atmos. Environ., 37, 971981.

    • Search Google Scholar
    • Export Citation
  • Washington, W. M., and Coauthors, 2000: Parallel Climate Model (PCM) control and transient simulations. Climate Dyn., 16, 755774.

  • Zhu, J., and X.-Z. Liang, 2005: Regional climate model simulation of U.S. soil temperature and moisture during 1982–2002. J. Geophys. Res., 110, D24110, doi:10.1029/2005JD006472.

    • Search Google Scholar
    • Export Citation
  • Zhu, J., and X.-Z. Liang, 2007: Regional climate model simulations of U.S. precipitation and surface air temperature during 1982–2002: Interannual variation. J. Climate, 20, 218232.

    • Search Google Scholar
    • Export Citation
Save
  • Auvray, M., and I. Bey, 2005: Long-range transport to Europe: Seasonal variations and implications for the European ozone budget. J. Geophys. Res., 110, D11303, doi:10.1029/2004JD005503.

    • Search Google Scholar
    • Export Citation
  • Bloomer, B. J., J. W. Stehr, C. A. Piety, R. J. Salawitch, and R. R. Dickerson, 2009: Observed relationships of ozone air pollution with temperature and emissions. Geophys. Res. Lett., 36, L09803, doi:10.1029/2009GL037308.

    • Search Google Scholar
    • Export Citation
  • Camalier, L., W. Cox, and P. Dolwick, 2007: The effects of meteorology on ozone in urban areas and their use in assessing ozone trends. Atmos. Environ., 41, 71277137, doi:10.1016/j.atmosenv.2007.04.061.

    • Search Google Scholar
    • Export Citation
  • Chang, J. S., S. Jin, Y. Li, M. Beauharnois, C.-H. Lu, H.-C. Huang, S. Tanrikulu, and J. Damassa, 1997: The SARMAP Air Quality Model final report. California Environmental Protection Agency Air Resource Board Rep., 160 pp.

  • Chen, M., R. Talbot, H. Mao, B. Sive, J. Chen, and R. J. Griffin, 2007: Air mass classification in coastal New England and its relationship to meteorological conditions. J. Geophys. Res., 112, D10S05, doi:10.1029/2006JD007687.

    • Search Google Scholar
    • Export Citation
  • Comrie, A. C., and B. Yarnal, 1992: Relationships between synoptic-scale atmospheric circulation and ozone concentrations in metropolitan Pittsburgh, Pennsylvania. Atmos. Environ., 26B, 301312.

    • Search Google Scholar
    • Export Citation
  • Cooter, E. J., J. Swall, and R. Gilliam, 2007: Comparison of 700-hPa NCEP-R1 and AMIP-R2 wind patterns over the continental United States using cluster analysis. J. Appl. Meteor., 46, 17441758.

    • Search Google Scholar
    • Export Citation
  • Darby, L. S., and Coauthors, 2007: Ozone differences between near-coastal and offshore sites in New England: Role of meteorology. J. Geophys. Res., 112, D16S91, doi:10.1029/2007JD008446.

    • Search Google Scholar
    • Export Citation
  • Davis, R. E., B. P. Hayden, D. A. Gay, W. L. Phillips, and G. V. Jones, 1997: The North Atlantic subtropical anticyclone. J. Climate, 10, 728744.

    • Search Google Scholar
    • Export Citation
  • Diem, J. E., 2006: Synoptic-scale controls of summer precipitation in the southern United States. J. Climate, 19, 613621.

  • Dudhia, J., D. Gill, Y. R. Guo, K. Manning, W. Wang, and C. Bruyere, 2005: PSU/NCAR mesoscale modeling system tutorial class notes and user’s guide: MM5 modeling system version 3.7. [Available online at http://www.mmm.ucar.edu/mm5/documents/tutorial-v3-notes.html.]

  • Easterling, D. R., 1990: Persistent patterns of thunderstorm activity in the central United States. J. Climate, 3, 13801389.

  • Fang, Y., A. M. Fiore, L. W. Horowitz, A. Gnanadesikan, H. Levy II, Y. Hu, and A. G. Russell, 2009: Estimating the contribution of strong daily export events to total pollutant export from the United States in summer. J. Geophys. Res., 114, D23302, doi:10.1029/2008JD010946.

    • Search Google Scholar
    • Export Citation
  • Gilliam, R. C., C. Hogrefe, and S. T. Rao, 2006: New methods for evaluating meteorological models used in air quality applications. Atmos. Environ., 40, 50735086.

    • Search Google Scholar
    • Export Citation
  • Grell, G. A., 1993: Prognostic evaluation of assumptions used by cumulus parameterizations. Mon. Wea. Rev., 121, 764787.

  • Gustafson, W. I., Jr., and L. R. Leung, 2007: Regional downscaling for air quality assessment: A reasonable proposition? Bull. Amer. Meteor. Soc., 88, 12151227.

    • Search Google Scholar
    • Export Citation
  • Hegarty, J., H. Mao, and R. Talbot, 2007: Synoptic controls on summertime surface ozone in the northeastern United States. J. Geophys. Res., 112, D14306, doi:10.1029/2006JD008170.

    • Search Google Scholar
    • Export Citation
  • Heidorn, K. C., and D. Yap, 1986: A synoptic climatology for surface ozone concentrations in southern Ontario, 1976-1981. Atmos. Environ., 20, 695703.

    • Search Google Scholar
    • Export Citation
  • Henderson, K. G., and A. J. Vega, 1996: Regional precipitation variability in the southern United States. Phys. Geogr., 17, 93112.

  • Higgins, R. W., Y. Yao, E. S. Yarosh, J. E. Janowiak, and K. C. Mo, 1997: Influence of the Great Plains low-level jet on summertime precipitation and moisture transport over the central United States. J. Climate, 10, 481507.

    • Search Google Scholar
    • Export Citation
  • Hogrefe, C., J. Biswas, B. Lynn, K. Civerolo, J.-Y. Ku, J. Rosenthal, C. Rosenzweig, R. Goldberg, and P. L. Kinney, 2004: Simulating regional-scale ozone climatology over the eastern United States: Model evaluation results. Atmos. Environ., 38, 26272638.

    • Search Google Scholar
    • Export Citation
  • Huang, H.-C., and J. S. Chang, 2001: On the performance of numerical solvers for chemistry submodel in three-dimensional air quality models: Box-model simulations. J. Geophys. Res., 106 (D17), 20 17520 188.

    • Search Google Scholar
    • Export Citation
  • Huang, H.-C., X.-Z. Liang, K. E. Kunkel, M. Caughey, and A. Williams, 2007: Seasonal simulation of tropospheric ozone over the midwestern and northeastern United States: An application of a coupled regional climate and air quality modeling system. J. Appl. Meteor., 46, 945960.

    • Search Google Scholar
    • Export Citation
  • Huang, H.-C., and Coauthors, 2008: Impacts of long-range transport of global pollutants and precursor gases on U.S. air quality under future climatic conditions. J. Geophys. Res., 113, D19307, doi:10.1029/2007JD009469.

    • Search Google Scholar
    • Export Citation
  • Hurrell, J. W., J. J. Hack, D. Shea, J. M. Caron, and J. Rosinski, 2008: A new sea surface temperature and sea ice boundary dataset for the Community Atmosphere Model. J. Climate, 21, 51455153.

    • Search Google Scholar
    • Export Citation
  • Jacob, D. J., and D. A. Winner, 2009: Effect of climate change on air quality. Atmos. Environ., 43, 5163.

  • Jacob, D. J., J. A. Logan, G. M. Gardner, R. M. Yevich, C. M. Spivakovsky, S. C. Wofsy, S. Sillman, and M. J. Prather, 1993: Factors regulating ozone over the United States and its export to the global atmosphere. J. Geophys. Res., 98 (D8 14 81714 826.

    • Search Google Scholar
    • Export Citation
  • Kalnay, E., and Coauthors, 1996: The NCEP/NCAR 40-Year Reanalysis Project. Bull. Amer. Meteor. Soc., 77, 437471.

  • Katz, R. W., M. B. Parlange, and C. Tebaldi, 2003: Stochastic modeling of the effects of large-scale circulation on daily weather in the southeastern U.S. Climatic Change, 60, 189216.

    • Search Google Scholar
    • Export Citation
  • Kiley, C. M., and H. E. Fuelberg, 2006: An examination of summertime cyclone transport processes during Intercontinental Chemical Transport Experiment (INTEX-A). J. Geophys. Res., 111, D24S06, doi:10.1029/2006JD007115.

    • Search Google Scholar
    • Export Citation
  • Kunkel, K. E., X.-Z. Liang, J. Zhu, and Y. Lin, 2006: Can CGCMs simulate the twentieth-century “warming hole” in the central United States? J. Climate, 19, 41374153.

    • Search Google Scholar
    • Export Citation
  • Leibensperger, E. M., L. J. Mickley, and D. J. Jacob, 2008: Sensitivity of US air quality to mid-latitude cyclone frequency and implications of 1980-2006 climate change. Atmos. Chem. Phys., 8, 70757086.

    • Search Google Scholar
    • Export Citation
  • Lelieveld, J., and P. J. Crutzen, 1991: The role of clouds in tropospheric photochemistry. J. Atmos. Chem., 12, 229267.

  • Li, Q., and Coauthors, 2005: North American pollution outflow and the trapping of convectively lifted pollution by upper-level anticyclone. J. Geophys. Res., 110, D10301, doi:10.1029/2004JD005039.

    • Search Google Scholar
    • Export Citation
  • Li, W., L. Li, R. Fu, Y. Deng, and H. Wang, 2011: Changes to the North Atlantic subtropical high and its role in the intensification of summer rainfall variability in the southeastern United States. J. Climate, 24, 14991506.

    • Search Google Scholar
    • Export Citation
  • Liang, X.-Z., K. E. Kunkel, and A. N. Samel, 2001: Development of a regional climate model for U.S. Midwest applications. Part I: Sensitivity to buffer zone treatment. J. Climate, 14, 43634378.

    • Search Google Scholar
    • Export Citation
  • Liang, X.-Z., L. Li, A. Dai, and K. E. Kunkel, 2004a: Regional climate model simulation of summer precipitation diurnal cycle over the United States. Geophys. Res. Lett., 31, L24208, doi:10.1029/2004GL021054.

    • Search Google Scholar
    • Export Citation
  • Liang, X.-Z., L. Li, K. E. Kunkel, M. Ting, and J. X. L. Wang, 2004b: Regional climate model simulation of U.S. precipitation during 1982–2002. Part I: Annual cycle. J. Climate, 17, 35103528.

    • Search Google Scholar
    • Export Citation
  • Liang, X.-Z., J. Pan, J. Zhu, K. E. Kunkel, J. X. L. Wang, and A. Dai, 2006: Regional climate model downscaling of the U.S. summer climate and future change. J. Geophys. Res., 111, D10108, doi:10.1029/2005JD006685.

    • Search Google Scholar
    • Export Citation
  • Liang, X.-Z., and Coauthors, 2012: Regional Climate-Weather Research and Forecasting Model (CWRF). Bull. Amer. Meteor. Soc., 93, 13631387.

    • Search Google Scholar
    • Export Citation
  • Lin, J.-T., D. J. Wuebbles, and X.-Z. Liang, 2008: Effects of intercontinental transport on surface ozone over the United States: Present and future assessment with a global model. Geophys. Res. Lett., 35, L02805, doi:10.1029/2007GL031415.

    • Search Google Scholar
    • Export Citation
  • Logan, J. A., 1989: Ozone in rural areas of the United States. J. Geophys. Res., 94 (D6), 85118532.

  • Mao, H., and R. Talbot, 2004a: Relationship of surface O3 to large-scale circulation patterns during two recent winters. Geophys. Res. Lett., 31, L06108, doi:10.1029/2003GL018860.

    • Search Google Scholar
    • Export Citation
  • Mao, H., and R. Talbot, 2004b: Role of meteorological processes in two New England ozone episodes during summer 2001. J. Geophys. Res., 109, D20305, doi:10.1029/2004JD004850.

    • Search Google Scholar
    • Export Citation
  • Martin, E. R., and C. Schumacher, 2011: The Caribbean low-level jet and its relationship with precipitation in IPCC AR4 models. J. Climate, 24, 59355950.

    • Search Google Scholar
    • Export Citation
  • Muralidharan, V., G. M. Kumar, and S. Sampath, 1989: Surface ozone variation associated with rainfall. Pure Appl. Geophys., 130, 4755.

    • Search Google Scholar
    • Export Citation
  • Nolte, C. G., A. B. Gilliland, C. Hogrefe, and L. J. Mickley, 2008: Linking global to regional models to assess future climate impacts on surface ozone levels in the United States. J. Geol. Res., 113, D14307, doi:10.1029/2007JD008497.

    • Search Google Scholar
    • Export Citation
  • Ortegren, J. T., P. A. Knapp, J. T. Maxwell, W. P. Tyminski, and P. T. Soulé, 2011: Ocean–atmosphere influences on low-frequency warm-season drought variability in the Gulf Coast and southeastern United States. J. Appl. Meteor. Climatol., 50, 11771186.

    • Search Google Scholar
    • Export Citation
  • Owen, R. C., O. R. Cooper, A. Stohl, and R. E. Honrath, 2006: An analysis of the mechanisms of the North American pollutant transport to the central North Atlantic lower free troposphere. J. Geophys. Res., 111, D23S58, doi:10.1029/2006JD007062.

    • Search Google Scholar
    • Export Citation
  • Stahle, D. W., and M. K. Cleaveland, 1992: Reconstruction and analysis of spring rainfall over the southeastern U.S. for the past 1000 years. Bull. Amer. Meteor. Soc., 73, 19471961.

    • Search Google Scholar
    • Export Citation
  • Vukovich, F. M., 1995: Regional-scale boundary layer ozone variations in the eastern United States and their association with meteorological variations. Atmos. Environ., 29, 22592273.

    • Search Google Scholar
    • Export Citation
  • Vukovich, F. M., and J. Sherwell, 2003: An examination of the relationship between certain meteorological parameters and surface ozone variations in the Baltimore–Washington corridor. Atmos. Environ., 37, 971981.

    • Search Google Scholar
    • Export Citation
  • Washington, W. M., and Coauthors, 2000: Parallel Climate Model (PCM) control and transient simulations. Climate Dyn., 16, 755774.

  • Zhu, J., and X.-Z. Liang, 2005: Regional climate model simulation of U.S. soil temperature and moisture during 1982–2002. J. Geophys. Res., 110, D24110, doi:10.1029/2005JD006472.

    • Search Google Scholar
    • Export Citation
  • Zhu, J., and X.-Z. Liang, 2007: Regional climate model simulations of U.S. precipitation and surface air temperature during 1982–2002: Interannual variation. J. Climate, 20, 218232.

    • Search Google Scholar
    • Export Citation
  • Fig. 1.

    (a) Summer-mean SLP distribution with V850 stronger than 6 m s−1 shaded by scarlet and the summer interannual correlation coefficients of SLP with (b) the LLJ and (c) the BHI during 1979–2008. Correlations with absolute values greater than 0.35 are statistically significant at 95% confidence, assuming yearly independence. The black boxes in (b) are the two centers for defining the BHI.

  • Fig. 2.

    Geographic distributions of the summer interannual correlation coefficients of the BHI with (a) precipitation, (b) SAT, and (d) V850 during 1979–2008 and (c) MDA8 at all monitoring sites with data records of at least 10 summers in 1993–2008. The statistical significance is as in Fig. 1, but MDA8 assumes monthly independence. (e) Normalized summer interannual anomalies of the BHI (black) and MDA8 over the southern Great Plains–Midwest (blue) and the eastern United States (red). Their standard deviations used for the normalization are listed in the legend. The black boxes in (a),(b),(d) are where the regional-mean precipitation (PR), SAT, and V850 indices are defined, and the colored ones in (c) are for MDA8 in (e).

  • Fig. 3.

    Geographic distributions of the summer interannual correlation coefficients of the BHI with precipitation for (a) PGR and (b) PCM; SAT for (c) PGR and (d) PCM; MDA8 for (e) AQM/PGR and (f) MOZART/PCM; and V850 for (g) PGR and (h) PCM based on summer monthly data for 1996–2000.

  • Fig. 4.

    Correlation coefficients between summer BHI time series and (a) precipitation averaged over the southern Great Plains, (b) precipitation averaged over the Midwest, (c) SAT averaged over the southern Great Plains, and (d) V850 averaged over the southern Great Plains for twentieth-century climate simulation (20C3M) model experiments and observations during 1970–99. The number after the comma in the label indicates the run number for those models with multiple realizations. The dashed horizontal line indicates the value of apparent statistical significance at the 95% level of confidence.

  • Fig. 5.

    Geographic distributions of the summer interannual correlation coefficients of the BHI during 1970–99 with precipitation for (a) CNRM and (b) PCM1; with SAT for (c) CNRM and (d) PCM1; and with V850 for (e) CNRM and (f) PCM1, which contains missing values below high terrains because of no extrapolation from the lowest model layer of a coarse-resolution grid as in the raw data.

  • Fig. 6.

    Geographic distributions of the summer interannual correlation coefficients of the BHI during 1970–99 with observed SST (a) in the previous winter and (b) in the concurrent summer; with CNRM-simulated SST (c) in the previous winter and (d) in the concurrent summer; and with PCM1-simulated SST (e) in the previous winter and (f) in the concurrent summer. The black boxes in (a),(b) are the EP and EA regions defining the SST index, respectively.

  • Fig. 7.

    Correlation coefficients between summer BHI time series and (a) SST in previous winter averaged over the key area boxed in Fig. 6a and (b) SST in the concurrent summer averaged over the key area boxed in Fig. 6b.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 1660 393 68
PDF Downloads 873 202 16