Global sensitivity and uncertainty analysis of an atmospheric chemistry transport model: the FRAME model (version 9.15.0) as a case study

Abstract

Abstract. Atmospheric chemistry transport models (ACTMs) are widely used to underpin policy decisions associated with the impact of potential changes in emissions on future pollutant concentrations and deposition. It is therefore essential to have a quantitative understanding of the uncertainty in model output arising from uncertainties in the input pollutant emissions. ACTMs incorporate complex and non-linear descriptions of chemical and physical processes which means that interactions and non-linearities in input–output relationships may not be revealed through the local one-at-a-time sensitivity analysis typically used. The aim of this work is to demonstrate a global sensitivity and uncertainty analysis approach for an ACTM, using as an example the FRAME model, which is extensively employed in the UK to generate source–receptor matrices for the UK Integrated Assessment Model and to estimate critical load exceedances. An optimised Latin hypercube sampling design was used to construct model runs within ±40 % variation range for the UK emissions of SO2, NOx, and NH3, from which regression coefficients for each input–output combination and each model grid ( > 10 000 across the UK) were calculated. Surface concentrations of SO2, NOx, and NH3 (and of deposition of S and N) were found to be predominantly sensitive to the emissions of the respective pollutant, while sensitivities of secondary species such as HNO3 and particulate SO42−, NO3−, and NH4+ to pollutant emissions were more complex and geographically variable. The uncertainties in model output variables were propagated from the uncertainty ranges reported by the UK National Atmospheric Emissions Inventory for the emissions of SO2, NOx, and NH3 (±4, ±10, and ±20 % respectively). The uncertainties in the surface concentrations of NH3 and NOx and the depositions of NHx and NOy were dominated by the uncertainties in emissions of NH3, and NOx respectively, whilst concentrations of SO2 and deposition of SOy were affected by the uncertainties in both SO2 and NH3 emissions. Likewise, the relative uncertainties in the modelled surface concentrations of each of the secondary pollutant variables (NH4+, NO3−, SO42−, and HNO3) were due to uncertainties in at least two input variables. In all cases the spatial distribution of relative uncertainty was found to be geographically heterogeneous. The global methods used here can be applied to conduct sensitivity and uncertainty analyses of other ACTMs.

Advanced methods for uncertainty assessment and global sensitivity analysis of an Eulerian atmospheric chemistry transport model

“…For computationally demanding models, such as ACTMs, a local one-at-a-time (OAT) sensitivity analysis is the most commonly used approach (Ferretti et al, 2015). However, unlike global sensitivity analysis, the local OAT approach does not take into account the non-linearities in the model response and the interactions between the input parameters (Saltelli and Annoni, 2010;Aleksankina et al, 2018).…”

“…Different meta-modelling approaches have been used for uncertainty and sensitivity analysis; these techniques include regression smoothers (Storlie and Helton, 2008;Storlie et al, 2009), Gaussian process emulator (Oakley and O'Hagan, 2004), high-dimensional model representation (Rabitz and Alış, 1999;Ziehn and Tomlin, 2009), and polynomial chaos expansion (Sudret, 2008). Meta-models have been applied for uncertainty and sensitivity analyses in earth science fields such as ecological modelling (Luo et al, 2013;Parry et al, 2013), hydrological modelling (Asher et al, 2015;Gladish et al, 2017), and atmospheric aerosol modelling (Lee et al, 2011;Carslaw et al, 2013;Chen et al, 2013;Christian et al, 2017).…”

A hybrid model approach for estimating health burden from NO2 in megacities in China: a case study in Guangzhou

“…emission inventories), meteorological parameters, and background concentrations, which are currently not fully available/accessible in most cities in China. Aleksankina et al (2018Aleksankina et al ( , 2019 have investigated model uncertainties in relation to emission input data in detail, highlighting that atmospheric chemistry transport models show relatively robust responses to changes in emission input data. LUR is an efficient modelling approach, but in areas with limited monitoring sites such as Guangzhou, with only 11 monitoring sites, the selected variables may overfit the model and hence cause bias in health-effect estimates (Basagaña et al 2012).…”

Author response to RC2

“…The FRAME model code is not available in the public domain as the model is the intellectual property of the Centre for Ecology & Hydrology and is only made available to students and researchers who are collaborating directly with CEH staff. However, all the following output data are available at https://doi.org/10.5281/zenodo.1145852 (Aleksankina, 2018). The data contain (i) all FRAME model outputs (raw data) for both actual input uncertainty and ±40 % input ranges, (ii) R scripts used to calculate RCs, SRCs, and uncertainty ranges, and (iii) RCs, SRCs, and uncertainty ranges calculated for every FRAME output variable and which are presented in all figures in this paper.…”