摘要:

A seasonal trend analysis of Dobson total ozone data that have been critically reevaluated and revised is performed for 29 northern hemisphere stations located between 19°N and 64°N latitude using data through 1986. The trend model considered for these data allows for a different linear trend for each month of the year, so that the seasonal as well as the latitudinal and regional nature of the total ozone trend behavior can be examined. The trend model also incorporates the 10.7‐cm solar flux series and the 50‐hPa equatorial zonal wind series as additional explanatory factors for solar and quasi‐biennial oscillation induced ozone variations. Regression random effects models are then used for the individual station seasonal trend estimates to obtain trend estimates as a function of latitude for different seasons of the year. The results of this seasonal trend analysis indicate significantly more negative trends during the winter months (December‐March) than during the summer months (May–August), notably at higher latitudes, with the trends in winter becoming more negative with increasing latitude. The trends in the winter are estimated to be of the order of −1.2%, −2.1%, and −3.0% per decade for latitudes 35°N, 45°N, and 55°N, respectively, while trends during the summer are of the order of −0.6% per decade with no distinct pattern as a function of latitude. The year‐round or annual trend over all latitudes is estimated to be about −0.84±0.82% per decade. The trends are found to display some regional variation, with trends in Japan being considerably less negative than those in North America and Europe. Sensitivity studies are also performed to investigate the effects on ozone trend estimates due to certain factors such as abnormal ozone behavior in 1983 and 1985, the use of ozone data prior to 1965, and nuclear weapons testing in the early 1960s. The seasonal trend analysis is also performed using published (unrevised) Dobson data. Trend results based on published data are on average less negative than trends from revised Dobson data for European stations, by about 1.0% per decade across all seasons, with only small average differences for stations

ISSN:0148-0227    

DOI:10.1029/JD095iD07p09785

年代:1990

数据来源: WILEY

摘要:

Measurements of ambient levels of PAN (peroxyacetyl nitrate or, more properly, peroxyacetic nitric anhydride), C2–C5alkyl nitrates, NOx(NO+NO2), HNO3, and NO3−particulates were made concurrently with a measurement of total reactive nitrogen species (NOy) at Scotia, Pennsylvania, from July 16 to August 31, 1988. Each of the organic nitrates exhibited a marked diurnal variation. The concentration of PAN varied from 0.05 to 3 ppbv (parts per billion by volume). The sum of the alkyl nitrate concentrations varied from less than 2 to 200 pptv (parts per trillion by volume). The NOylevels varied from 1 to 30 ppbv. During periods of high photochemical activity the contributions to NOywere 30–45% from HNO3, 20–25% from PAN, 15–25% from NOx, 5% from NO3−, and an additional 1.5% from the C2–C5alkyl nitrates, on average. During these periods, about 15% of the measured NOycould not be accounted for by the sum of the individually measured species. The unidentified species that constitute this 15% exhibit behavior similar to that of the measured org

ISSN:0148-0227    

DOI:10.1029/JD095iD07p09809

年代:1990

数据来源: WILEY

摘要:

Commercial mass flow controllers are widely used in atmospheric research where precise and constant gas flows are required. We have determined, however, that some commonly used controllers are far more sensitive to ambient pressure than is acknowledged in the literature of the manufacturers. Since a flow error can lead directly to a measurement error of the same magnitude, this is a matter of great concern. Indeed, in our particular application, were we not aware of this problem, our measurements would be subject to a systematic error that increased with altitude (i.e., a drift), up to a factor of 2 at the highest altitudes (∼37 km). In this note we present laboratory measurements of the errors of two brands of flow controllers when operated at pressures down to a few millibars. The errors are as large as a factor of 2 to 3 and depend not simply on the ambient pressure at a given time, but also on the pressure history. In addition there is a large dependence on flow setting. In light of these flow errors, some past measurements of chemical species in the stratosphere will need to be revise

ISSN:0148-0227    

DOI:10.1029/JD095iD07p09817

年代:1990

数据来源: WILEY

摘要:

We have analyzed atmospheric spectra obtained with the Fourier transform spectrometer on Kitt Peak, in Arizona, for absorption by CO2and CH4. During the period 1979–1985 we find that the CO2total column amount was increasing at a rate of 0.4%/yr and that CH4was increasing at a rate of 1.1%/yr. This is in substantial agreement with air sampling measurements from other sites. In addition, a weak seasonal variation of ∼2% is detected in CO2, but none is seen in

ISSN:0148-0227    

DOI:10.1029/JD095iD07p09823

年代:1990

数据来源: WILEY

摘要:

A one‐dimensional photochemical model has been used to calculate future changes in tropospheric O3and OH due to CO/NOx/CH4emissions and to possible changes in stratospheric O3and tropospheric H2O. Perturbations are simulated for various chemically coherent regions (e.g., urban and continental mid‐latitudes, and marine and continental low latitudes) from 1985 to 2035. Estimates of global changes in O3and OH are made by averaging over these regions. Two types of scenarios are simulated. “Global” scenarios assume that increases of CH4and CO continue at current rates in all regions. A second set of scenarios, based on an analysis Of CH4, CO, and NO budgets in each region, assumes that emissions will be controlled in some regions and not in others. Both global and region‐specific scenarios predict a global tropospheric O3increase of ∼10–15% from 1985 to 2035 with OH decreasing 10–15%. In the regionally varying scenarios, O3will increase in some regions and decrease in others; for examples in regions of rapid CH4and CO increase, growth in boundary layer O3may be as high as 40%. Calculations that assume stratospheric O3depletion and climate warming from 1985 to 2035 show near cancellation of the tropospheric O3enhancement and OH loss. All scenarios of CH4/CO/NO changes considered in this study imply a 1–1.5% increase in total ozone from 1985 to 2035, assuming that tropospheric O3is 10% of the total O3column. This may complicate detection of stratospheric O3change by monit

ISSN:0148-0227    

DOI:10.1029/JD095iD07p09829

年代:1990

数据来源: WILEY

摘要:

The Measurement of Air Pollution from Satellite (MAPS) experiment measured the distribution of middle tropospheric carbon monoxide (CO) from the space shuttle during October 1984. The data represent average mixing ratios in the middle troposphere between 57°N and 57°S. Approximately 75,000 individual CO measurements were obtained during the 9‐day mission. The data are presented in maps that show the CO mixing ratios averaged over 5° latitude by 5° longitude areas for 6 days of the mission. Comparisons with concurrent, direct measurements taken aboard aircraft show that the inferred concentrations are systematically low by 20–40% depending upon which direct measurement calibration standard is used. The data show that there are very large CO sources resulting from biomass burning over South America and southern Africa. Measured mixing ratios were high over northeast Asia and were highly variable over

ISSN:0148-0227    

DOI:10.1029/JD095iD07p09845

年代:1990

数据来源: WILEY

摘要:

Measurements of H2O2and related species were made by aircraft on 18 flights over 11 days during June 1987 over Ohio. Measurements typically encompassed an altitude range from the surface to 5.5 km above mean sea level and were conducted under a variety of meteorological conditions, but focussed on conditions conducive to the formation of convective storms. Concentrations of H2O2ranged from<0.2 to 7 ppb and exhibited as much as a factor of 10 variability even on a given day at fixed altitude, as well as between days. Hydrogen peroxide concentration typically was low near the surface, rose to a maximum near the top of the boundary layer, then slowly decreased with altitude. In contrast, SO2and NOywere largely confined to the surface layer. A simple photochemical calculation shows that where NOxconcentrations are low, [H2O2] is strongly influenced by the water vapor concentration. The [H2O2] tracked changes in dew point, and measured and calculated H2O2concentrations were in general agreement. The low [H2O2] in the boundary layer is ascribed to consumption of radicals by reactions other than combination reactions, and to loss of H2O2by dry deposition. Comparison of the profile shapes for H2O2and SO2suggests that mixing processes will have a large influence on the extent to which these two species will react in convective systems.

ISSN:0148-0227    

DOI:10.1029/JD095iD07p09857

年代:1990

数据来源: WILEY

摘要:

Dry deposition of mineral particles was investigated on the windward coast of Oahu (Hawaii) using a surrogate surface technique. The collected particles were analyzed microscopically for the number of particles in different size classes. These data were used to estimate deposition rates to the deposition plate. Relating these fluxes to the corresponding concentrations of mineral particles in air, mean dry deposition velocities were determined. For the size classes 2–4, 4–8, and 8–15 μm these were 0.3, 1.2, and 2.6 cm s−1, respectively. Additionally, the dry deposition velocity curve in the range between 2 and 15 μm was determined on the assumption that the mineral particle mass size distributions in the air and in dry deposition were lognormal. This curve was compared with theoretical data calculated by a model for the deposition of particles to the sea surface. The discrepancies, in particular for smaller particles, are probably due to the different deposition conditions at the sea surface and the deposition plate. Also, the aggregates that were found in some samples and which consist of sea salt and mineral particles will affect the experimentally derived dry deposit

ISSN:0148-0227    

DOI:10.1029/JD095iD07p09873

年代:1990

数据来源: WILEY

摘要:

This article presents the present‐day capability of three‐dimensional analysis of air pollution, by both differential absorption lidar and advanced numerical models. These two techniques exhibit numerous similarities, both at the level of spatial resolution and on a working scale. Two measurement campaigns have been performed to determine emission characteristics and concentrations of sulfur dioxide (SO2) in order to illustrate these similarities. The results of the intercomparisons are satisfactory, and differences are usually within

ISSN:0148-0227    

DOI:10.1029/JD095iD07p09879

年代:1990

数据来源: WILEY

摘要:

A method is presented for simultaneous determination of the aerosol optical thickness (τa), particle size (rm, geometric mean mass radius for a lognormal distribution) and the single scattering albedo (ω0, ratio between scattering and scattering +absorption) from satellite imagery. The method is based on satellite images of the surface (land and water) in the visible and near‐IR bands and is applied here to the first two channels of the Advanced Very High Resolution Radiometer (AVHRR) sensor. The aerosol characteristics are obtained from the difference in the upward radiances, detected by the satellite, between a clear and a hazy day. Therefore the method is mainly useful for remote sensing of large‐scale air pollution (e.g., smoke from a large fire or concentrated anthropogenic pollution), which introduces dense aerosol into the atmosphere (aerosol optical thickness ≥0.4) on top of an existing aerosol. The method is very sensitive to the stability of the surface reflectance between the clear day and the hazy day. It also requires accurate satellite calibration (preferably not more than 5% error) and stable calibration with good relative values between the two bands used in the analysis. With these requirements, the aerosol optical thickness can be derived with an error of Δτa= 0.08–0.15. For an assumed lognormal size distribution, the particle geometrical mean mass radiusrmcan be derived (if good calibration is available) with an error of Δrm= ±(0.10–0.20)μm, and ω0with Δω0= ±0.03 for ω0close to 1 and Δω0= ±(0.03–0.07) for ω0about 0.8. Uncertainties in the type of the size distribution further increase the uncertainty inrm. The method was applied to AVHRR images of a forest fire smoke. The aerosol optical thickness and particle size derived from the AVHRR images are compared to the analysis of sun photometer m

ISSN:0148-0227    

DOI:10.1029/JD095iD07p09895

年代:1990

数据来源: WILEY