Mesoscale modelling of atmospheric CO2 across Denmark

Research output: Book/anthology/dissertation/reportPh.D. thesisResearch

  • Anne Sofie Lansø
It is scientifically well-established that the increase of atmospheric CO2 affects the entire globe and will lead to higher surface temperatures. Although anthropogenic CO2is emitted straight into the atmosphere, it does not all contribute to the existing atmospheric CO2 reservoir. Approximately 29% is taken up by the global oceans, due to under-saturation of CO2 in the surface waters, while another 33 % is taken up by the terrestrial biosphere, via photosynthesis. In order to estimate the effects of increasing anthropogenic emissions of CO2 more accurately in the future, it is essential to understand the processes controlling the sources and sinks of atmospheric CO2.

This PhD dissertation attempts to increase our understanding of the importance of accounting for high spatiotemporal variability in estimates of COexchanges between the atmosphere and the surface. For this purpose, a mesoscale modelling system is constructed, centred around Denmark, based on an atmospheric transport model. In this study, the main areas of focus have been on improving the spatial surface representation, for both land and sea, and investigating the influence of the temporal resolution on the air–sea CO2 exchange.

Until the present study, no area-specific representation had been developed for the surface water pCOof the Baltic Sea and Danish inner waters. A surface water monthly climatology was implemented in the mesoscale modelling framework, and further improved with a near coastal climatology for the Danish inner waters. In the modelling framework, the heterogeneous land surfaces of Denmark were assessed by means of a detailed land surface classifications map and the inclusion of a high temporal and spatial resolution biosphere model. Available measurements of surface water pCO2, from stationary sites within the Baltic Sea, were used to construct a realistic set of monthly diurnal cycles of surface water pCO2 within the study region, in order to include short-term variability in surface water pCO2 in the model setup.

The air-sea CO2 exchange, for the coastal area of the Baltic Sea and the Danish inner waters, was consistently examined by means of the constructed mesoscale modelling framework. From a six year (2005-2010) simulation, the average annual flux for the study region was found to be a small sink of atmospheric CO2. The influence of short-term variability in atmospheric CO2 was found to have a significant impact on the annual air–sea CO2 exchange. A simulation with constant monthly fields of atmospheric CO2, reduced the winter release of CO2 for the six year period, resulting in an increase of 67% in the average annual uptake by the Baltic Sea and Danish inner waters.

The inclusion of short-term variations in surface water pCO2 increased the simulated release of COto the atmosphere. For the two different surface fields of pCO2tested during different simulations in the modelling framework, the annual air–sea CO2 exchange for 2011 showed an increase in one case, while in the other, when short-term variability in surface water pCOwas included, the annual uptake changed to an annual release of atmospheric CO2. Besides showing the impact of short-term variability in surface water pCO2, these simulations also showed that the choice of surface water pCO2 fields had a notable impact on the annual air–sea CO2exchange. Similarly, this was also evident for the near coastal climatology for the Danish inner waters: these areas resolve to release CO2 annually to the atmosphere when the near-coastal climatology is included, while act as annual sinks if not.

The Danish terrestrial biosphere was divided into seven land use classes, covering forest type, agricultural usage and grasslands. The land use class, found to be the most influential on the monthly surface exchanges was grassland, while winter crops were most influential on an annual basis. A westward gradient in numerical size of surface exchange was found across Denmark, and was reflected in the short-term variability in the atmospheric CO2 concentrations. However, the local land–sea signal was difficult to detect at the Risø tall tower site. A first estimate of the Danish CO2 budget was made for year 2011. The Danish biosphere and ocean were found to take up an amount of CO2 that nearly corresponds to all the CO2 emitted by fossil fuel use in Denmark. However, the biospheric uptake might have been overestimated.

The spatiotemporal resolution on land and sea has been greatly improved for the focus area during this study. The calculations of the annual air–sea CO2 exchange showed a sensitivity to short-term variability in the partial pressure of CO2. The air–sea exchange is also dependent on surface maps of pCO2, and with the improvements surface representation made in this PhD, the Baltic Sea is found to be a small sink, while near-coastal Danish inner waters are a net annual source of CO2 ot the atmosphere. It is evident that the developed modelling systems is capable of simulating the biosphere–atmosphere exchange of CO2 and, compared to the tall tower measurements of CO2, a good level of consistency was found. The variability of the simulated atmospheric CO2 across Denmark was, in particular, affected by the Danish terrestrial surface exchanges and its temporal variability.
This study urges all future modelling studies of air–sea CO2 to include short-term variability in pCO2. To capture the full heterogeneity of the surface exchanges including land-sea, fjord or lake signal, this study recommend high spatial resolution of both model and input parameters for studies investigating such complexities.
Original languageEnglish
Number of pages148
Publication statusPublished - 27 Jul 2016

See relations at Aarhus University Citationformats

ID: 101501851