In a recent paper we derived an analytical expression for the deposition velocity of molecular hydrogen on soil that includes the action of a dry top soil layer without H 2 removal (Ehhalt and Rohrer, 2013). With this short note we would like to expand that expresion to include a production of H 2 within the soil. (Published: 6 September 2013) Citation: Tellus B 2013, 65, 20620, http://dx.doi.org/10.3402/tellusb.v65i0.20620

In the past two types of laboratory experiments have been employed to determine the dependence of H-2 uptake by soils on temperature and moisture: Head space and flow experiments. The former actually measure the rate constant of the H-2 removal from the head space, k(H), the latter the uptake rate of H-2, UH2, both caused by a given volume of soil. From an analytical solution of the diffusion equation in the soil we derive a mathematical relation between k(H) and k(s), the desired uptake rate constant of H-2 in soil. Another equation relates UH2 with k(s). Both types of experiments actually determine the product of k(s) with Theta(a), the air-filled pore volume fraction. k(s)center dot Theta(a) for eolian sand and loess loam show zero uptake at very low and high moisture contents and a well defined maximum in between. Unlike soil moisture which also acts on the soil properties, the soil temperature, T, acts essentially on the enzyme activity only. Thus k(s)(T) is directly proportional to k(H)(T) or UH2(T) and the data of all experiments can be superimposed by scaling. The resulting average k(s)(T) shows a broad maximum around 30 degrees C with zero uptake below -20 degrees C and above 80 degrees C.

The literature on the distribution, budget and isotope content of molecular hydrogen (H-2) in the troposphere is critically reviewed. The global distribution of H-2 is reasonably well established and is relatively uniform. The surface measurements exhibit a weak latitudinal gradient with 3% higher concentrations in the Southern Hemisphere and seasonal variations that maximize in arctic latitudes and the interior of continents with peak-to-peak amplitudes up to 10%. There is no evidence for a continuous long-term trend, but older data suggest a reversal of the interhemispheric gradient in the late 1970s, and an increase in the deuterium content of H-2 in the Northern Hemisphere from 80 parts per thousand standard mean ocean water (SMOW) in the 1970s to 130 parts per thousand today. The current budget analyses can be divided in two classes: bottom up, in which the source and sink terms are estimated separately based on emission factors and turnovers of precursors and on global integration of regional loss rates, respectively. That category includes the analyses by 3-D models and furnishes tropospheric turnovers around 75 Tg H-2 yr(-1). The other approach, referred to as top down, relies on inverse modelling or analysis of the deuterium budget of tropospheric H-2. These provide a global turnover of about 105 Tg H-2 yr(-1). The difference is due to a much larger sink strength by soil uptake and a much larger H-2 production from the photochemical oxidation of volatile organic compounds (VOC) in the case of the top down approaches. The balance of evidence seems to favour the lower estimates-mainly due to the constraint placed by the global CO budget on the H-2 production from VOC. An update of the major source and sink terms yields: fossil fuel use 11 +/- 4 Tg H-2 yr(-1); biomass burning (including bio-fuel) 15 +/- 6 Tg H-2 yr(-1); nitrogen fixation (ocean) 6 +/- 3 Tg H-2 yr(-1); nitrogen fixation (land) 3 +/- 2 Tg H-2 yr(-1); photochemical production from CH4 23 +/- 8 Tg H-2 yr(-1) and photochemical production from other VOC 18 +/- 7 Tg H-2 yr(-1). The loss through reaction of H-2 with OH is 19 +/- 5 Tg H-2 yr(-1), and soil uptake 60(-20)(+30) Tg H-2 yr(-1). All these rates are well within the ranges of the corresponding bottom up estimates in the literature. The total loss of 79 Tg H-2 yr(-1) combined with a tropospheric burden of 155 Tg H-2 yields a tropospheric H-2 lifetime of 2 yr. Besides these major sources of H-2, there are a number of minor ones with source strengths < 1 Tg H-2 yr(-1). Rough estimates for these are also given.

The literature on the distribution, budget and isotope content of molecular hydrogen (H2) in the troposphere is critically reviewed. The global distribution of H2 is reasonably well established and is relatively uniform. The surface measurements exhibit a weak latitudinal gradient with 3% higher concentrations in the Southern Hemisphere and seasonal variations that maximize in arctic latitudes and the interior of continents with peak-to-peak amplitudes up to 10%. There is no evidence for a continuous long-term trend, but older data suggest a reversal of the interhemispheric gradient in the late 1970s, and an increase in the deuterium content of H2 in the Northern Hemisphere from 80 standard mean ocean water (SMOW) in the 1970s to 130 today. The current budget analyses can be divided in two classes: bottom up, in which the source and sink terms are estimated separately based on emission factors and turnovers of precursors and on global integration of regional loss rates, respectively. That category includes the analyses by 3-D models and furnishes tropospheric turnovers around 75 Tg H2 yr−1. The other approach, referred to as top down, relies on inverse modelling or analysis of the deuterium budget of tropospheric H2. These provide a global turnover of about 105 Tg H2 yr−1. The difference is due to a much larger sink strength by soil uptake and a much larger H2 production from the photochemical oxidation of volatile organic compounds (VOC) in the case of the top down approaches. The balance of evidence seems to favour the lower estimates—mainly due to the constraint placed by the global CO budget on the H2 production from VOC. An update of the major source and sink terms yields: fossil fuel use 11±4 TgH2 yr−1; biomass burning (including bio-fuel) 15 ± 6 Tg H2 yr−1; nitrogen fixation (ocean) 6 ± 3 Tg H2 yr−1; nitrogen fixation (land) 3 ± 2 Tg H2 yr−1; photochemical production from CH4 23 ± 8 Tg H2 yr−1 and photochemical production from other VOC 18 ± 7 Tg H2 yr−1. The loss through reaction of H2 with OH is 19 ± 5 Tg H2 yr−1, and soil uptake 60+30 −20 Tg H2 yr−1. All these rates are well within the ranges of the corresponding bottom up estimates in the literature. The total loss of 79 Tg H2 yr−1 combined with a tropospheric burden of 155 Tg H2 yields a tropospheric H2 lifetime of 2 yr. Besides these major sources of H2, there are a number of minor ones with source strengths > 1 Tg H2 yr−1. Rough estimates for these are also given.

[1] On 9 April 2001, while approaching the West Coast of the North American continent, flight 20 of the NASA TRACE P campaign penetrated deeply into a stratospheric intrusion. From measurements aboard that flight we derive vertical profiles of nonmethane hydrocarbons (NMHCs) and other short-lived trace gases in the lower stratosphere using Dichlorofluoromethane, CF2Cl2, as the altitude scale. All profiles show an exponential decrease, which permits the description of their vertical distribution by a single parameter, the scale height. These scale heights are shown to be related to the respective lifetimes in a unique fashion. Using the approximation of a 1-D diffusive model with a constant eddy diffusion coefficient, K, and assuming constant lifetimes, we establish an analytical solution for this relation. By fitting this theoretical expression to the experimental data we can estimate K and thus obtain an approximate age spectrum of the form given by Hall and Plumb (1994). A much better fit to the experimental scale heights is obtained, when we allow the lifetimes to be height-dependent and calculate the theoretical scale heights numerically from a 1-D model. An optimization also suggests a constant K, but with a value of 0.46 m2/s, larger than those obtained from the fit of the analytical solution. The obtained age spectra should be valid for transit times not longer than 300 days.

This paper reanalyzes the measurements of the D content in tropospheric water vapor by Ehhalt ( 1974) correcting for the isotopic contamination by wall water in the sampling tubes. The resulting corrections decrease the original D content. They are small for the data from the flights through 1967, which extended from the surface to 9 km altitude, but are large for the flights beginning 1971, which ranged from 6 to 13 km altitude and collected smaller amounts of water vapor. No correction therefore was attempted for the latter. The corrected data of the earlier flights are presented in the form of seasonally averaged profiles over Scotts Bluff, Nebraska; Death Valley, California; and the Pacific offshore of San Luis Obispo, California. As to be expected, the vertical profiles from the earlier flights show a decrease in the D content with altitude and a seasonal variation at all altitudes. However, when plotted against the water vapor mixing ratio, the D data from all seasons collapse on a line which closely follows that given by Rayleigh condensation with a fractionation factor alpha = 1.1, constant with altitude. These data can be explained by a simple one-dimensional convection model and the assumption that condensed water is lofted along with the water vapor.

On the basis of a one‐dimensional (1‐D) analysis the decay time of the lowest eigenmode, τ1, for the stratospheric distribution of a conserved tracer is derived from measured vertical profiles of the mean age of stratospheric air. Two case studies (a and b) give τ1,a = 3.8 ± 0.8 years and τ1,b = 5.3 ± 1.1 years. These semiobservational times are considerably longer than most of the τ1 derived from 2‐D and 3‐D models. At the same time they are shorter than the observational eigentime, τ1,HTO = 7.7 ± 2 years, determined from the decay of the tritium (T) content in stratospheric water vapor, following the thermonuclear test explosions in the early 1960s. Part of the differences among the observational eigentimes can be explained by the assumptions that had to be made to extract τ1,HTO from the trend in the T content of stratospheric water vapor (namely, the cosmogenic background of tritiated water vapor and the trend in stratospheric water vapor). This leads to a revised value τ1,HTO = 6.3 ± 0.9 years for the time period 1975–1983. Allowing for a possible temporal trend in Γ and hence τ1, the value for the current τ1 decreases to 5.3 ± 1 years.

Airborne measurements of formaldehyde (CH 2 O) were acquired employing tunable diode laser absorption spectroscopy (TDLAS) during the 2000 Tropospheric Ozone Production About the Spring Equinox (TOPSE) study. This study consisted of seven deployments spanning the time period from 4 February to 23 May 2000 and covered a wide latitudinal band from 40°N to 85°N. The median measured CH 2 O concentrations, with a few exceptions, did not show any clear temporal trends from February to May in each of five altitude and three latitude bins examined. Detailed measurement-model comparisons were carried out using a variety of approaches employing two different steady state models. Because recent emissions of CH 2 O and/or its precursors often result in model underpredictions, background conditions were identified using a number of chemical tracers. For background conditions at temperatures warmer than -45°C, the measurement-model agreement on average ranged between -13% and +5% (measurement-model/measurement), which corresponded to mean and median (measurement-model) differences of 3 ± 69 and -6 parts per trillion by volume (pptv), respectively. At very low temperatures starting at around -45°C, significant and persistent (measurement-model) differences were observed from February to early April from southern Canada to the Arctic Ocean in the 6-8 km altitude range. In such cases, measured CH 2 O was as much as 392 pptv higher than modeled, and the median difference was 132 pptv (83%). Low light conditions as well as cold temperatures may be important in this effect. A number of possible mechanisms involving the reaction of CH 3 O 2 with HO 2 to produce CH 2 O directly were investigated, but in each case the discrepancy was only minimally reduced. Other possibilities were also considered but in each case there was no compelling evidence to support any of the hypotheses. Whatever the cause, the elevated CH 2 O concentrations significantly impact upper tropospheric HO x levels at high latitudes (>57°N) in the February-April time frame.

The historic and partly unpublished measurements of the tritium content in stratospheric water vapor made at the National Center for Atmospheric Research between 1975 and 1983 are reanalyzed. The resulting vertical profiles of the T content, mainly at 32°N latitude, show little variation with altitude above 20 km but a strong decay with time. This decay is a consequence of the large T injections into the stratosphere by the atmospheric tests of high‐yield thermonuclear devices prior to 1963 and seems to proceed with a single e‐fold time of 5.12 years. Correcting for the radioactive decay of HTO within the stratosphere and for a temporal increase in stratospheric H2O, we obtain a decay time for stratospheric HTO of 7.7 ± 2.0 years. This decay time, which is solely due to the transport of HTO into the troposphere, is much longer than the age of stratospheric air at these altitudes or the accepted values for stratospheric residence times. The differences are discussed and resolved by interpreting the HTO decay time as the Eigentime of the longest‐lived mode of the stratospheric transport equations. This Eigentime should provide a useful constraint in modeling stratospheric transport.

The global distribution and budget of atmospheric molecular hydrogen (H2) is simulated with a global Chemistry‐Transport Model (CTM). Surface emissions include technological sources (industry, transportation and other fossil fuel combustion processes), biomass burning, nitrogen fixation in soils, and oceanic activity and totals 39 Tg/yr. The photochemical production (31 Tg/yr) from formaldehyde photolysis accounts for about 45% of the total source of H2. Soil uptake (55 Tg/yr) represents a major loss process for H2 and contributes for 80% to the total destruction. H2 oxidation by OH in the troposphere contributes the remainder. The global burden of H2 in the atmosphere is 136 Tg. Its overall lifetime in the atmosphere is 1.9 years. H2 is rather well‐mixed in the free troposphere. However, its distribution shows a significant seasonal variation in the lower troposphere where soil uptake dominates. This loss process shows a strong temporal variability and is maximum over the northern hemisphere landmass during summer. Strong vertical gradients result from this surface uptake. In these regions, H2 varies by more than 30% between the maximum mixing ratio in winter and the summer minimum. Our results stress the important role played by the tropics in the budget of H2. In these regions a strong seasonal cycle is also predicted due to the annual variation in biomass burning emissions, soil uptake, and rapid transport by convection of H2 depleted air masses from the boundary layer to the upper troposphere. A comparison with the observed H2 distribution allows to test some of the model predictions. Good agreement is found for the global burden and the annually averaged latitudinal gradient in the southern hemisphere and the tropics. A detailed comparison of the seasonal cycles of H2 in surface air indicates that the use of the net primary productivity to prescribe the seasonal and geographical pattern of soil uptake in the model leads to an underestimate of the deposition velocity during winter and spring over the continents in the northern hemisphere.

OH and the major parameters determining its concentration were measured during a field campaign in August 1994 at Mankmoos, a rural, relatively unpolluted site in northeastern Germany. The measured OH concentrations were previously shown to depend mainly on the intensity of solar UV and on the mixing ratio of NO 2 . In this paper we develop a simple parameterization of the dependence on solar UV and on NO 2 . The photolysis of O 3 to O 1 D , of NO 2 to NO, and of HCHO to HCO and H, all contribute significantly to the total dependence of OH on solar UV. We demonstrate that the photolysis frequency of O 3 , J O 1 D , is a suitable measure for that dependence which is slightly less than linear. The highly nonlinear dependence of OH on NO x is approximated by a Padé function. The parameterization provides a tool for a future quantitative intercomparison of the measured and modeled dependences of OH on UV and NO 2 . It also allows the removal of the variation in the measured OH induced by the dependences on the variables, UV and NO 2 , and thus enables a search for dependences on other, less influential parameters.

This paper explores a new approach to estimating atmospheric hydroxyl radical concentrations from regional measurements of a suite of hydrocarbons. The approach is guided by the study of a suite of synthetic tracers, i , with uniform continental sources and constant but different lifetimes of 1, 2, 5, 20, 50, and 100 days, whose global distributions are calculated from a three‐dimensional chemical tracer model. With the help of the model we show that in a grid box the standard deviation σ i divided by the average concentration is a unique function of the chemical lifetime τ i . In favorable cases, for instance, in surface air within a specific region sampled by the Pacific Exploratory Mission (PEM) West B campaign, that function takes a simple form: , with α = 0.48, very close to 1/2. An analogous relation is found for the alkanes, ethane through n‐hexane, measured during the PEM West B campaign in the same domain, with their reaction rate constant with OH, k OH. That relation has the form , with α′ = 0.49. Using the alkenes, for example propene, which also react with O 3 , the dependence on k OH,i can be related to a dependence on τ i . This allows us to estimate the OH concentration, 6×10 5 cm −3 , with an error of roughly a factor of 2 for this region (boundary layer, 30°N‐40°N latitude, and 135°E‐155°E longitude in March). This estimate is essentially based on empirical relations only and the assumption that the considered hydrocarbons have the same source distribution. As a by‐product, we show that the α defined above is related to the slope in the (logarithmic) correlation plot between the mixing ratios of two trace gases with different lifetimes. We also show that the global distribution of a appears to be a useful tool to diagnose fast regional transport.

In-situ OH measurements by laser-induced fluorescence (LIF) spectroscopy and folded long-path differential optical absorption spectroscopy (DOAS) were carried out in a rural environment in North-East Germany as part of the field experiment POPCORN in August 1994. The large set of OH data obtained allowed an intercomparison of both techniques based on relative diurnal profiles and simultaneously measured absolute concentrations. Most of the time the two OH instruments encountered the same air and agreed well in the measured relative diurnal variations. Only on a few occasions the measurements significantly disagreed due to a perturbation of the DOAS measurements by a local OH source in the north-western wind sector. Excluding data from this wind direction, the statistical analysis of 137 data pairs yields a correlation coefficient of r = 0.90 and a weighted linear fit with a slope of 1.09 ± 0.12. The correlations are carefully analyzed. The comparison of both instruments is discussed in the light of newly published effective absorption cross-sections for H2O and O2 that affect the calibration of LIF.

Tropospheric hydroxyl radical (OH) concentrations were measured by laser-induced fluorescence (LIF) during the POPCORN field campaign in August 1994 at a rural site in the North East of Germany. Ambient air spectra were recorded by tuning the laser wavelength over a spectral region covering the Q11(3), Q21(3), and P11(1) rotational transitions of the (0-0) band in the A-X system of OH around 308 nm. The observed spectra clearly identify the OH radical in the atmosphere. Besides the OH absorption lines there was no sign of any other narrow-band spectral structure nearby demonstrating the high specificity of the method. For OH measurements with a typical time resolution of 60–100 seconds per data point the laser wavelength was tuned repetitively over small spectral intervals covering the peak position of the P11(1) OH-line and background positions. A total of 2300 measurements were recorded including diurnal cycles of OH with more than 300 data points. The OH as well as the LIF background signal data will be presented. In a first analysis the background signal will be characterized and the correlation between OH and the ozone photolysis frequency will be derived.

Ambient mixing ratios of NO, NO2, and O3 were determined together with the photolysis frequency of NO2, JNO2, at a rural, agricultural site in Germany. The data were collected during the POPCORN-campaign from August 1 to August 24, 1994, in a maize field 6 m above ground. The medians of the NO, NO2, and O3 mixing ratios between 10:00 and 14:00 UT were 0.25, 1.09, and 45 ppbv, respectively. The corresponding median of JNO2 was 6.0 · 10−3 s−1. NOx = NO + NO2 showed a strong diurnal variation with maximum mixing ratios at night, suggestive of a strong local surface source of NO, probably by microbial activity in the soil. The estimated average emission rate was 40 ng(N) m−2 s−1 of NOx, the major part of it probably in the form of NO. The available measurements allowed the estimation of the local NOx budget. At night the budget is almost closed and the measured NOx mixing ratios can be explained by the local source, local dry deposition of NO2, formation of NO3 and N2O5, and vertical exchange of air across the nocturnal inversion. During day-time, the local surface source of NO is not sufficient to explain the measured mixing ratios, and horizontal advection of NOx to the site must be included. The NO2/NO ratio during the morning und late afternoon is lower than predicted from the photostationary state owing to the local NO surface source, but is regulary higher during the hours around noon. For noon, August 10, 1994, the NO2/NO ratio was used to derive the momentary lower limit for the concentration of the peroxy-radicals of 2.2 · 109 cm−3 (86 pptv).

A highly sensitive OH measurement instrument has been developed. It is based on laser‐induced fluorescence (LIF) detection of OH using the A²Σ+v′ = 0‐X²Π v″ = 0 transition at 308.15 nm at low pressure. The LIF instrument detects OH directly and with high specificity, a fact that was demonstrated by recording laser excitation spectra (Q1(3), Q21(3) and P1(1) lines) of ambient OH. For high time resolution (typ. 60–100 s), the laser wavelength was modulated on‐/off‐ resonance with the P1(1) line. Here, we report some of the OH measurements obtained by this technique during its first application in a tropospheric field campaign (“POPCORN”), which was conducted in August 1994 in a rural environment in the North‐East of Germany. These include diurnal OH concentration profiles with maximum OH concentrations up to 1.4×107cm−3at noon. Minimum OH concentrations were measured in the morning and evening down to the detection limit of (3–6) × 105cm−3(SNR=2, measurement time 1 min.). During the day, OH fluctuations were observed on a time scale of minutes and hours. These were highly correlated to the flux of the solar UV radiation which is responsible for the primary OH production by photolysis.