20 m

60 m

120 m

200 m

20-200 km

200-500 km

seacitiesgrassland

NL

country borders

Cabauw towerUtrecht

Rotterdam

the Hague

Amsterdam

H2 and δD(H2) observations at the Cabauw tall tower:very depleted δD source signature suggests microbial H2 production in Dutch pasture soil

A.M. Batenburg1,2,3, M.E. Popa1, A.T. Vermeulen4,5, W.C.M. van den Bulk4, P.A.C. Jongejan4, R.E. Fisher6, D. Lowry6, E.G. Nisbet6, and T. Röckmann1

1: H2 at the Cabauw towerWe measured the molecular hydrogen (H2) mixing ratio (χ(H2)) and its isotopic composition (δD(H2)) in samples from the Cabauw tall tower to better understand the atmospheric cycle of this potential alternative fuel.The Cabauw tower at the CESAR site is located in a relatively rural, mostly agricultural, central part of the Netherlands, within tens of km from the four major Dutch cities (Fig 1). Its tubing system has inlets at 20, 60, 120 and 200 m. The tower is immediately surrounded by pastures.

3: Low δD(H2) signatureWe estimated the isotopic signature of the Cabauw source mix from the y-intercept of a linear fit to a "Keeling" plot (δD(H2) plotted vs 1/χ(H2), Fig 3(a)). To obtain a realistic error estimate , and to account for possible arbitrariness in our data quality control, we applied a bootstrapping routine. The resulting distribution of intercepts is broad, but almost all are below -400 ‰ (Fig 3(b)).This is more D-depleted than any published source signatures for H2 from combustion sources, even those measured after modern engines with catalytic converters under "congested" simulated driving conditions. Microbial production of H2 has been observed in the clover-rich grassland soil around Cabauw (Chen et al., 2015) and produces very D-depleted H2, possibly explaining the low apparent δD(H2) source signature.

4: Height profilesCabauw is the first location where vertical profiles of δD(H2) in the boundary layer have been obtained (Fig 4). Lower χ(H2) values and higher δD(H2) values are expected close to the ground at locations with strong soil uptake. Fig 4 does not show this, probably because of the soil type (peat/clay) and high ground water table. In fact, χ(H2) is signicantly higher at 20 m than at 200 m, and δD(H2) is signicantly lower at the lower than at the higher levels.The δD(H2) difference may point to a difference in the fossil fuel combustion source signature between the footprint regions of the sampling levels, or, again, to local microbial H2 production in the soil.

2: Time seriesFig 2 shows the time series of χ(H2) and δD(H2) collected at Cabauw, together with similar data from Mace Head on the Irish west coast. The Mace Head data form the lower bound for the Cabauw χ(H2) and the higher bound for the Cabauw δD(H2) data. Especially in winter, excursions to high χ(H2) values occur regularly at Cabauw and are associated with very low δD(H2) values. These features indicate that Cabauw is heavily influenced by H2 from (anthropogenic) surface sources, which produce H2 that is depleted in deuterium.

‰

‰

5: ConclusionsOur χ(H2) and δD(H2) observations at this anthropogenically influenced site show features that were not seen in previous work with mostly background stations (2). The very depleted apparent source signature (3) and the observed differences between sampling levels (4) indicate that microbial H2 production may play a larger role in the local H2 isotope budget than expected.

This work was published as:A.M. Batenburg et al., AtmosEnv, 147, 98-108, 2016, doi: 10.1016/j.atmosenv.2016.09.058Other references:M.E. Popa et al., ACP, 11,6425-6443, 2011, doi:10.5194/acp-11-6425-2011A. M. Batenburg et al., ACP, 11, 6985-6999, 2011, doi:10.5194/acp-11-6985-2011Q. Chen et al., ACPD, 15, 23457-23506, 2015, doi:10.5194/acpd-15-23457-2015

Author affiliations:1: IMAU, Utrecht University, Utrecht, the Netherlands (NL)2: Department of Applied Physics, University of Eastern Finland, Kuopio, Finland3: Particle Chemistry Dept., Max Planck Institute for Chemistry / Johannes Gutenberg University of Mainz, Mainz, Germany 4: Energy research Centre of the Netherlands, Petten, NL5: ICOS-ERIC, Carbon Portal, Lund, Sweden6: Dept. of Earth Sciences, Royal Holloway, University of London, Egham, Surrey, UK

Fig 1: Cartoon representation of the geographic features within the different footprint regions of the tower. Insert: Land cover map of the region (Popa et al., 2011), showing mostly agricultural land around the site.

Fig 3(a): "Keeling" plot of δD(H2) vs 1/χ(H2) of the Cabauw flask data, with bivariate linear fit. Grey bar indicates y-axis intercept and error. (b) Distribution of intercepts obtained from a bootstrapping routine where the fit was applied to random samples of the data.

Fig 2: χ(H2) (a) and δ(H2) (b) time series of the flask samples from Cabauw. The Mace Head data with fits(Batenburg et al., 2011) are plotted for comparison. Open symbols indicate datapoints that did not pass quality control.

Fig 4: Box plots of χ(H2) (a) and δD(H2) (b) on days where more than two sampling heights were sampled. Red lines indicate medians, box edges indicate lower and upper quartiles and whiskers indicate lower and upper 95th percentiles.

Geophysical Research AbstractsVol. 19, EGU2017-8503, 2017EGU General Assembly 2017© Author(s) 2017. CC Attribution 3.0 License.

Observations of molecular hydrogen (H2) mixing ratio and stable isotopiccomposition at the Cabauw tall tower; very depleted source signaturesuggests microbial H2 production in Dutch pasture soil.Anneke Batenburg (1,2,3), Elena Popa (1), Alex Vermeulen (4,5), Pim van den Bulk (4), Piet Jongejan (4),Rebecca Fisher (6), Dave Lowry (6), Euan Nisbet (6), and Thomas Röckmann (1)(1) Institute for Marine and Atmospheric research Utrecht, Utrecht University, Utrecht, the Netherlands, (2) Department ofApplied Physics, University of Eastern Finland, Kuopio, Finland, (3) Particle Chemistry Department, Max Planck Institute forChemistry / Johannes Gutenberg University of Mainz, Mainz, Germany, (4) Energy research Centre of the Netherlands,Petten, the Netherlands, (5) Integrated Carbon Observation System European Research Infrastructure Consortium(ICOS-ERIC), Carbon Portal, Lund, Sweden, (6) Department of Earth Sciences, Royal Holloway, University of London,Egham, Surrey, United Kingdom

Molecular hydrogen (H2), though not toxic or a greenhouse gas itself, may influence air quality and climate indi-rectly by affecting the atmosphere’s oxidative capacity. So as increased use of hydrogen fuel is expected, a betterunderstanding of the global, regional and local atmospheric H2 cycles is needed. Studying the stable isotopic com-position of H2 (δD(H2)) is a promising way to achieve this. Since the start of this century, the isotope effects inH2 source and sink processes have been estimated, δD(H2) has been incorporated into chemical transport models,and larger sets of environmental observations of δD(H2) have appeared. The latter, however, were mostly obtainedfrom samples collected in remote regions of the atmosphere, which is not sufficient to fully characterize the H2

cycle or to assess the possible environmental effects of H2 leakage in urbanized regions. To address this gap, flasksamples were collected at the Cabauw tall tower at the CESAR site in the Netherlands. The air was sampled frominlets at 20, 60, 120, and 200 meter altitude for the analysis of H2 mixing ratio (χ(H2)) and δD(H2). More than250 samples were collected and analysed over a period of four years.

The H2 mixing ratios in the samples show frequent excursions to high values above the background. Previouslypublished continuous χ(H2) observations at Cabauw and other (sub)urban sites showed a similar pattern. Withthe isotope observations, we can now see that these high χ(H2) excursions are accompanied by very low δD(H2)values; probably at least partly a result of anthropogenic emissions of deuterium(D)-depleted H2.

However, with a simple “Keeling plot” analysis, we obtained an apparent source signature (-515 ± 26 h) that wasmuch below the range of published values for H2 emissions from the combustion of fossil fuels. Since the resultof the fit depended markedly on the quality selection of the samples that were included, we applied a bootstrapmethod to this fit to obtain a realistic picture of the uncertainty of the result. This showed a wide distributionwith more than 99 % of the values below -400 h, suggesting that the H2 cycle at Cabauw is under the influenceof a source mix that is much more D-depleted than currently accepted values for fossil fuel combustion. Sincemicrobial production of very D-depleted H2 has been observed previously at Cabauw, we consider it likely thatthis contributes to the low apparent source signature.

A comparison of the samples from different sampling heights shows that there is a significant shift to lower δD(H2)values at the lower sampling levels. This shows that the uptake of H2 by the soil, which preferentially removes“light” H2, is relatively weak at the site. It also points again to local to regional microbial production of H2, andpossibly to differences between national vehicle fleets.

Further information about this work

A more detailed description of the work on this poster has been published as an open access journal article in Atmospheric Environment:

A.M. Batenburg, M.E. Popa, A.T. Vermeulen, W.C.M. van den Bulk, P.A.C. Jongejan, R.E. Fisher, D. Lowry, E.G. Nisbet, and T. Röckmann, Observations of

molecular hydrogen mixing ratio and stable isotopic composition at the Cabauw tall tower in the Netherlands, Atmos. Env., 147, 98-108, 2016,

http://dx.doi.org/10.1016/j.atmosenv.2016.09.058

The Cabauw flask data are available from:

- Zenodo (record 155112)

- Pangaea (https://doi.pangaea.de/10.1594/PANGAEA.864969, or follow link from AtmosEnv page)

- www.projects.science.uu.nl/atmosphereclimate/Data.php.

The corresponding author (A.M. Batenburg) can be reached through annekebatenburg@gmail.com.

Related work from the IMAU-APCG group

H2 production and uptake in soil around Cabauw:

Q. Chen, M.E. Popa, A.M. Batenburg, and T. Röckmann, Isotopic signatures of production and uptake of H2 by soil, Atmos. Chem. Phys., 15, 13003-13021, 2015,

http://dx.doi.org/10.5194/acp-15-13003-2015

Semi-continuous H2 mixing ratio observations at Cabauw:

M.E. Popa, A.T. Vermeulen, W.C.M. van den Bulk, P.A.C. Jongejan, A.M. Batenburg, W. Zahorowski, and T. Röckmann, H2 vertical profiles in the continental

boundary layer: measurements at the Cabauw tall tower in the Netherlands, Atmos. Chem. Phys., 11, 6425- 6443, 2011, http://dx.doi.org/10.5194/acp-11-

6425-2011

H2 isotopic observations at background stations around the world:

A.M. Batenburg, S. Walter, G. Pieterse, I. Levin, M. Schmidt, A. Jordan, S. Hammer, C. Yver, and T. Röckmann, Temporal and spatial variability of the stable

isotopic composition of atmospheric molecular hydrogen: observations at six EUROHYDROS stations, Atmos. Chem. Phys., 11, 6985-6999, 2011,

http://dx.doi.org/10.5194/acp-11-6985-2011