| 英文摘要: | Surface ozone, a major air pollutant toxic to humans and damaging to ecosystems1, 2, is produced by the oxidation of volatile organic compounds in the presence of nitrogen oxides (NOx = NO + NO2) and sunlight. Climate warming may affect future surface ozone levels3, 4, 5, 6 even in the absence of anthropogenic emission changes, but the direction of ozone change due to climate warming remains uncertain over the southeast US and other polluted forested areas3, 4, 5, 6, 7, 8, 9, 10. Here we use observations and simulations to diagnose the sensitivity of August surface ozone to large-scale temperature variations in the southeast US during 1988–2011. We show that the enhanced biogenic emissions and the accelerated photochemical reaction rates associated with warmer temperatures both act to increase surface ozone. However, the sensitivity of surface ozone to large-scale warming is highly variable on interannual and interdecadal timescales owing to variation in regional ozone advection. Our results have important implications for the prediction and management of future ozone air quality. Summertime surface ozone production over polluted forested areas is often dominated by the photochemical oxidation of biogenic isoprene (C5H8), whose emission from vegetation is nonlinearly dependent on temperature and sunlight intensity11, and thus highly sensitive to climate warming. Previous climate–chemistry model (CCM) studies predicted surface ozone changes due to climate warming (ΔCWO3) by simulating two climate scenarios—one for a future time slice and one for the present-day—in the absence of anthropogenic emission changes. These predictions showed no consensus on even the sign of ΔCWO3 over polluted forested areas, such as the southeast US (SEUS), western Europe, and parts of East and South Asia3, 4, 5, 6, 7, 8, 9. This lack of consensus has been attributed to the uncertain model assumptions regarding ozone precursor (particularly isoprene) chemistry4, 5, 7, 12, as well as the different predictions of regional climate changes (for example, temperature changes, ΔCWT) across CCMs and for different time horizons5, 7. In this study, we diagnosed the sensitivity of surface ozone to large-scale temperature variations (hereafter referred to as dO3/dTLS) over the SEUS as a means to understand the response of surface ozone to climate warming (ΔCWO3). Figure 1a shows the August surface temperature over the SEUS (TSEUS) during 1988–2011. There was no significant trend in TSEUS, but oscillated interannually over a range of approximately 3 K. We found that this interannual variation (IAV) of TSEUS was a manifestation of the first empirical orthogonal function (EOF) of the IAV of August surface temperature over the contiguous US during 1988–2011 (Fig. 1a, r2(TSEUS, first EOF) = 0.69). The first EOF of US August temperature was characterized by an oscillation where almost the entire contiguous US is in the same phase (Fig. 1c). Subsequent US August temperature EOFs were of finer spatial scales and not significantly correlated with TSEUS (Fig. 1b, d). Thus, during 1988–2011 and on the interannual timescale, SEUS August surface ozone was perturbed by large-scale temperature variations, which offers a unique opportunity to diagnose dO3/dTLS. In contrast, studies have analysed the sensitivity of SEUS surface ozone to daily temperature variations13, 14, but that more likely reflects the response of ozone to synoptic weather.
|
