Chemosphere - Global Change Science最新文献

C₂–C₆ non-methane hydrocarbons (NMHC), including isoprene, were measured in a forested area of Portugal from June 20 to July 12, 1994 as part of the FIELDVOC'94 project. The day-to-day variability of the measured NMHC was clearly linked to the air mass origin. The longest-lived ethane did not show any significant diurnal variation. The diurnal variability of the C₃–C₅ alkanes and of acetylene presented low amplitude attributed to the changeable wind direction between northwest and southeast. The short-lived alkenes showed a well-defined diurnal variation with maximum mixing ratios during night-time and minimum during daytime resulting from both dynamical and photochemical sinks. Isoprene presented a very different trend with minima of 0.1 ppbv at night and maxima up to 12 ppbv occurring between 16:00 and 18:00 (UT). In continental air masses, the contribution of isoprene to the daytime ozone production was 3- to 6-fold greater than that of the other NMHC. In air masses of marine origin, the effect of isoprene on ozone and hydroxyl radicals was significantly reduced relative to the anthropogenic NMHC, which were dominating at night. In all cases, isoprene was a more efficient consumer of nitrate radicals than these NMHC. The observed decrease of isoprene at night was consistent with the measured nitrate radical mixing ratios of a few pptv.

Measurements of NO,NO_x,NO_y,CO,O₃,J_NO2 (the photolysis frequency of NO₂) and of meteorological parameters made during the FIELDVOC'94 campaign in Tábua, Portugal, are described. The situation at the site was dominated by a land–sea breeze system coupled with thermal up-slope on the mountain ridge, which caused northwesterly flow from the coastal area during daytime and southeasterly flow at night. O₃ up to 100 ppbv was observed in photochemically aged air masses that reached the site in the afternoon from the coastal areas. At night, O₃ mixing ratios at the surface dropped below 10 ppbv. NO_x mixing ratios varied between 200 pptv during daytime and 3 ppbv at night with peaks up to 6 ppbv. NO_y (the sum of NO_x and its oxidation products) varied between 1 and 6 ppbv and CO between 100 and 280 ppbv. Simultaneous measurements of NO₂ by photolytic conversion/NO-chemiluminescence and by tunable diode laser spectroscopy showed a tight correlation, in particular at night. Taking the identified difference in calibration between the two methods (8%) into account, the disagreement in the atmosphere, was ⩽2%, on average. Photochemical ozone formation in air transported from the coast was identified by a positive correlation between O₃ and the oxidation products of NO_x referenced as NO_z. The slope of 16 represents an upper limit to the efficiency of NO_x in catalyzing ozone formation in air. The relatively high ozone formation efficiency is supported by the large O₃/CO ratio of about 0.4. The correlation between CO and NO_y indicates an upper limit for the CO/NO_x emission ratio of 28.

Gradients of CO and CO₂, taken between 12:00 and 15:00 local time, from the boundary layer over the tropical rainforest in Surinam were determined as 29 pmol/mol km⁻¹ and −8.9 nmol/mol km⁻¹, respectively, with a distance of south from the coast. For one CO₂ molecule fixed in tropical forests 0.33% CO was produced. From an extrapolation of the CO gradient to the global scale we deduce that approximately 19% of C emitted as isoprene from tropical forests is converted to C in CO. From an extrapolation of the CO₂ gradient we estimate that approximately 1.2% of the global atmospheric CO₂ is converted each year into tropical seasonal and rainforests.

CO production from isoprene was calculated using an explicit gas-phase photochemical model but was found to be insufficient to account for the gradients measured. Diurnal variation in CO was controlled by a complex interplay between advection, chemical formation from natural NMHCs, direct soil emissions, removal by HO, and possible night-time uptake by soil. Diurnal variations of CO₂ in the boundary layer were controlled by vegetation.

Over the 12.5 km altitude range of the aircraft, a high degree of variability was observed in CO, CO₂ and northerly wind components. Two biomass burning events were identified and the ratio of delta CO and delta CO₂ was determined in each case with a two-sided linear regression. A ratio of 12.1% was found in plumes from smouldering cooking or clearing fires at low-altitude. A ratio of 5.9% was found for a plume encountered between 10–12 km. The lower ratio of delta CO and delta CO₂ indicates hotter, more complete, flame burning than the cooking or clearing fires. Emissions from savanna fires in the Brazil, Colombia and Venezuela regions coupled with deep convection and long-range advection are proposed to explain the observations.

In Japan, five types of vehicles, light-duty gasoline [(LDG); 550 cc < engine displacement (ED)], super-light-duty gasoline [(SLD); ED < 550 cc], liquid petroleum-fueled [(LPG); 550 cc < ED < 3000 cc], light-duty diesel [(LDD); 550 cc < ED < 5000 cc] and heavy-duty diesel [(HDD); 5000 cc < ED] cars have been used widely to give the complex influences on the hydrocarbon components at roadsides and in city areas, which have not been explained sufficiently. In this study, we investigated the ambient 16 hydrocarbon components at two roadsides in commercial and industrial areas of Osaka City along with the traffic densities of five types of vehicles to propose the significant effects of emissions from SLD and LPG cars besides that of LDG car through a comparison of the ambient [ethylene]/[acetylene] and [i-, n-pentanes]/[i-, n-butanes] ratios with those of the source profiles published in Japan.

The dominant processes affecting the concentration of tropospheric methane on interannual timescales are the biospheric and anthropogenic sources and changes in the abundance of the hydroxyl radical caused by the changes in the UV flux which result from changes in stratospheric ozone abundance. We have carried out an empirical study of the sensitivity of the methane to fluctuations in ozone column abundance. This analysis was carried out using monthly mean surface methane concentrations measured by the National Oceanic and Atmospheric Administration – Climate Monitoring and Diagnostics Laboratory (NOAA-CMDL) Global Cooperative Air Sampling Network from 1983 to 1998 and ozone column abundances obtained by the Total Ozone Mapping Spectrometer (TOMS) and the EP TOMS instruments over the same time period. We focused on interannual variability with periods between 15 and 60 months, in which interval the dominant ozone fluctuation is the quasi-biennial oscillation (QBO), with a period of approximately 29 months. In order to isolate the response of methane to ozone from the effects of variability in the sources and transport of methane, we restricted our analysis to data at mid-latitudes in the southern hemisphere. A statistical study shows that the sensitivity factor α≡−d(ln[CH₄])/d(ln[O₃])=−0.038±0.009. The response of CH₄ lags approximately 6 months behind the forcing by O₃. A simple model was used to interpret the empirical results. Our results confirm that any mechanism that affects stratospheric ozone impacts the oxidizing potential of the troposphere. CH₄ fluctuations provide a quantitative measure of this important effect linking the upper and the lower atmosphere.

Vehicle exhaust has been a very important hydrocarbon source in a big city. In Japan, five types of vehicles, light-duty gasoline (LDG), super-light-duty gasoline (SLD), liquid petroleum gas-fueled (LPG), light-duty diesel (LDD) and heavy-duty diesel (HDD) cars have been used widely for a long time. These vehicles have different fuels and exhaust gas treatment systems, which have not changed so largely since the late 1980s. The hydrocarbons [total hydrocarbon (THC), methane and 16 speciated volatile hydrocarbon components] emitted from 12 cars (one LDG, two SLD, two LPG, three LDD and four HDD) were investigated. As a result, the remarkable peculiarities of hydrocarbon emissions from the five types of vehicles were observed. Especially, the emissions from SLD and LPG cars have the peculiarities of high concentrations of total hydrocarbon (THC) and C₂ (ethylene and acetylene), C₃ (propane or propylene) or C₄ (i-, n-butanes) hydrocarbons in 16 components, which may be proposed to give a significant effect on 16 component hydrocarbons of the ambient air in big cities of Japan in addition to those from LDG and diesel (LDD and HDD) cars.

Influences of initial and boundary conditions were quantified by analysis of a simplified governing equation followed with a tracer and O₃ simulations using a three-dimensional air quality transport model. The analytical solution derived from the governing equation indicated that the impacts of initial conditions on a given site decrease with simulation time and significantly affect the species concentrations before the arrival of the boundary conditions. Boundary influences, which decrease during the downwind transport, are significant to a selected site when the arrival time of boundary condition is short and the species lifetime is long.

Non-reactive tracers and O₃ were then added to SARMAP Air Quality Model (SAQM), a three-dimensional transport model, and the results showed that tracer lifetime and arrival time determine the boundary influences on a given site consistent with the analysis obtained from the governing equation. Moreover, boundary conditions followed initial conditions to affect the calculated ozone concentrations at a given site and the influences are proportional to the magnitude of the prescribed boundary conditions, arrival time, and are more obvious at local nighttime. When the arrival time is 15 h, an average of 64% of initial ozone boundary conditions will contribute to the calculated ozone concentrations. For arrival times of 23, 20, and 7 h, the ratios are 26%, 34%, and 79%, respectively. This study indicated that the derived analytical solution could be used by air quality modelers in estimating the boundary influences as long as the species lifetime are known.

The annual sink strengths of soils under rabi and kharif crops, forests, pastures and long fallow areas have been estimated to be 0.11, 0.34, 0.01 and 0.05 Tg yr⁻¹, respectively for the atmospheric methane. The total annual sink provided by Indian soils is 0.51 Tg yr⁻¹ and is about 4% of the total annual methane emission attributed to anthropogenic sources in India.

A global two-dimensional chemistry model is used to study the long-term trends of CH₄, CO, and OH from pre-industrial times to 2020 with given emission scenarios according to the increase of world population. The calculated global-averaged concentration of CO is 27 ppbv before 1840, the concentration of CO is 76 ppbv in 1991, and is estimated to be 105 ppbv in 2020. From 1840 to 1991, the concentration of OH changed from 7.17×10⁵ to 5.79×10⁵ molecules/cm³, i.e., decreased by 19%. The long-term trends of CH₄ derived from the model are in good agreement with observation results. The annual increase of CH₄ during 1983–1991 is 12.1–13.3 ppbv by this model and 11.1–11.6 ppbv by observation. The calculated growth rate of CO in 1980s is 1.03–1.06%/yr i.e., 6.9–7.9 ppbv/yr. The model is used to investigate why the CO concentration decreased at the beginning of 1990s. We find that the decrease of CO emissions and depletion of stratospheric ozone are the best explanation, which account for 70% and 30% of the decrease of CO concentration, respectively. The model results also show that possible reduction of CH₄ emission has little influence on the change of CO concentrations though the reduction of CO emission can counteract the growth of CH₄ significantly.