Datensatz

Zusammenfassung

The global budget and trends of atmospheric methane (CH4) have been simulated with the EMAC atmospheric chemistry – general circulation model for the period 1997 through 2014. Observations from AGAGE and NOAA surface stations and intercontinental CARIBIC flights indicate a transient period of declining methane increase during 1997 through 1999, followed by seven years of stagnation and a sudden resumed increase after 2006. Starting the simulation with a global methane distribution, scaled to match the station measurements in January 1997 and using inter-annually constant CH4 sources from eleven categories together with photochemical and soil sinks, the model reproduces the observations during the transient and constant period from 1997 through 2006 in magnitude as well as seasonal and synoptic variability. The atmospheric CH4 calculations in our model setup are linearly dependent on the source strengths, allowing source segregated simulation of eleven biogenic and fossil emission categories (tagging), with the aim to analyze global observations and derive the source specific CH4 steady state lifetimes. Moreover, tagging enables a-posteriori rescaling of individual emissions with proportional effects on the corresponding inventories and offers a method to approximate the station measurements in terms of lowest RMS. Enhancing the a priori biogenic tropical wetland emissions by ~29Tg/y, compensated by a reduction of anthropogenic fossil CH4 emissions, the all-station mean dry air mole fraction of 1792 nmol/mol could be simulated within a RMS of 0.37%. The coefficient of determination R2=0.87 indicates good agreement with observed variability and the calculated 2000–2005 average interhemispheric methane difference between selected NH and SH stations of 119nmol/mol matches the observations. The CH4 samples from 95 intercontinental CARIBIC flights for the period 1997–2006 are also accurately simulated by the model, with a 2000–2006 average CH4 mixing ratio of 1786nmol/mol, and 65% of the measured variability being captured. This includes tropospheric and stratospheric data. To explain the growth of CH4 from 2007 through 2013 in term of sources, an emission increase of 28.3Tg/y CH4 is needed. We explore the contributions of two potential causes, one representing natural emissions from wetlands in the tropics and the other anthropogenic shale gas production emissions in North America. A 62.6% tropical wetland contribution and of 37.4% by shale gas emissions optimally fit the trend, and simulates CH4 from 2007–2013 with an RMS of 7.1nmol/mol (0.39%). The coefficient of determination of R2=0.91 indicates even higher significance than before 2006. The 4287 samples collected during 232 CARIBIC flights after 2007 are simulated with an RMS of 1.3% and R2=0.8, indicating that the model reproduces the seasonal and synoptic variability of CH4 in the upper troposphere and lower stratosphere.