Atmospheric Distribution of HCN from Satellite Observations and 3-D Model Simulations

Abstract. We use a tracer version of the TOMCAT global 3-D chemical transport model to investigate the physical and chemical processes driving the abundance of hydrogen cyanide (HCN) in the troposphere and stratosphere over the period 2004–2016. The modelled HCN distribution is compared with version 4.1 of the Atmospheric Chemistry Experiment - Fourier Transform Spectrometer (ACE-FTS) HCN satellite data, which provides profiles up to around 42 km, and with ground-based column measurements from the Network for the Detection of Atmospheric Composition Change (NDACC). ACE-FTS has so far provided over 17 years of data, from 2004, which allow us to monitor both the seasonal and interannual variations of HCN and its transport through the atmosphere. In particular, by analysing the long time series, we are able to detect the effects on atmospheric composition of large wildfire events like those observed in 2006 and 2015 in Indonesia. Our 3-D model simulations confirm previous lower altitude balloon comparisons that the currently recommended NASA Jet Propulsion Laboratory (JPL) reaction rate coefficient of HCN with OH greatly overestimates the HCN loss. The use of the rate coefficient proposed by Kleinböhl et al. (2006) in combination with the HCN oxidation by O(¹D) gives good agreement between ACE-FTS observations and the model. Furthermore, investigation of the individual photochemical loss terms shows that the reduction of the HCN mixing ratio with height in the middle stratosphere is mainly driven by the O(¹D) sink with only a small contribution from reaction with OH. From comparisons of the model tracers with ground-based HCN observations we test the magnitude of the ocean sink in two different published schemes (Li et al. 2000, 2003). We find that in our 3-D model the two schemes produce HCN abundances which are very different to the NDACC observations but in different directions. A model HCN tracer using the Li et al. (2000) scheme overestimates the HCN concentration by almost a factor two, while a HCN tracer using the Li et al. (2003) scheme underestimates the observations by about one-third. To obtain good agreement between the model and observations we need to scale the magnitudes of the global ocean sinks by the factors of 0.25 and 2 for the schemes of Li et al. (2000) and Li et al. (2003), respectively. This work shows that the atmospheric photochemical sinks of HCN now appear well constrained but improvements are needed in parameterising the major ocean uptake sink.