ammonia emissions range from 6.9 Tg N year
À1 (Xu et al. 2016) to 13.2 Tg N year
À1
(Dong et al. 2010) and even 15.6 Tg N year
À1 (Zhang et al. 2017b) in China. The
spatial distribution and seasonal variation also show significant discrepancies among
different ammonia emission inventories. In addition, the surface-atmosphere bidirectional NH 3 fluxes are not considered in current deposition model simulations,
which may lead to an overestimation of NH x deposition.
Uncertainties also exist in the model chemistry and deposition mechanisms.
Current models mainly simulate transformation and deposition of inorganic N
compounds, and only dry deposition of a few organic N species such as PANs and
isoprene nitrates is considered. In situ surface measurements have indicated that
soluble organic N species can contribute 20–25% of the bulk deposition in China
(Zhang et al. 2008; Du and Liu 2014). These soluble organic N species are not
included in the model simulations of wet deposition as our understanding of their
sources and sinks is rather limited. Uncertainties in dry deposition patterns often
result from lack, or oversimplified descriptions, of some important processes in
ecosystems. For example, vertical resolved canopy structure is necessary for accurately calculating dry deposition velocities, while detailed canopy morphological
information is difficult to obtain at regional scales for input to atmospheric chemistry
models. Future developments on the deposition mechanisms combined with more
accurate Nr emission estimates will improve model simulation of N deposition to
China.
Quantitative uncertainty analysis is an important means of diagnosis modelling.
Combined with sensitivity analysis, it is possible to identify the sources of key
uncertainties that influence model simulations among a large number of model
elements and provide guidance for model improvement. A number of uncertainty
analysis methods have been used such as Monte Carlo method (MCM), Latin
hypercube sampling (LHS), polynomial chaos expansion (PCE), stochastic response
surface method (SRSM), etc. However, the current methodology has limitations for
accurately and efficiently conducting uncertainty analysis for deposition modelling.
Future work is needed to establish a framework of quantitative uncertainty analysis
that combines sensitivity analysis, inputs uncertainty quantification, uncertainty
propagation, and assessment.
Acknowledgments The authors gratefully acknowledge the financial support by the National Key
Research and Development Program of China (2017YFC0210100).
References
Allen DJ, Pickering KE (2002) Evaluation of lightning flash rate parameterizations for use in a
global chemical transport model. J Geophys Res-Atmos 107:3851–3876
Baklanov A, Sorensen JH (2001) Parameterisation of radionuclide deposition in atmospheric longrange transport modelling. Phys Chem Earth Pt B 26:787–799
Barfield BJ, Gerber JF (1979) Modification of the aerial environment of plants. American Society of
Agricultural Engineers, St. Joseph, Mich
4 Modelling Atmospheric Nitrogen Deposition in China
81
À1 (Xu et al. 2016) to 13.2 Tg N year
À1
(Dong et al. 2010) and even 15.6 Tg N year
À1 (Zhang et al. 2017b) in China. The
spatial distribution and seasonal variation also show significant discrepancies among
different ammonia emission inventories. In addition, the surface-atmosphere bidirectional NH 3 fluxes are not considered in current deposition model simulations,
which may lead to an overestimation of NH x deposition.
Uncertainties also exist in the model chemistry and deposition mechanisms.
Current models mainly simulate transformation and deposition of inorganic N
compounds, and only dry deposition of a few organic N species such as PANs and
isoprene nitrates is considered. In situ surface measurements have indicated that
soluble organic N species can contribute 20–25% of the bulk deposition in China
(Zhang et al. 2008; Du and Liu 2014). These soluble organic N species are not
included in the model simulations of wet deposition as our understanding of their
sources and sinks is rather limited. Uncertainties in dry deposition patterns often
result from lack, or oversimplified descriptions, of some important processes in
ecosystems. For example, vertical resolved canopy structure is necessary for accurately calculating dry deposition velocities, while detailed canopy morphological
information is difficult to obtain at regional scales for input to atmospheric chemistry
models. Future developments on the deposition mechanisms combined with more
accurate Nr emission estimates will improve model simulation of N deposition to
China.
Quantitative uncertainty analysis is an important means of diagnosis modelling.
Combined with sensitivity analysis, it is possible to identify the sources of key
uncertainties that influence model simulations among a large number of model
elements and provide guidance for model improvement. A number of uncertainty
analysis methods have been used such as Monte Carlo method (MCM), Latin
hypercube sampling (LHS), polynomial chaos expansion (PCE), stochastic response
surface method (SRSM), etc. However, the current methodology has limitations for
accurately and efficiently conducting uncertainty analysis for deposition modelling.
Future work is needed to establish a framework of quantitative uncertainty analysis
that combines sensitivity analysis, inputs uncertainty quantification, uncertainty
propagation, and assessment.
Acknowledgments The authors gratefully acknowledge the financial support by the National Key
Research and Development Program of China (2017YFC0210100).
References
Allen DJ, Pickering KE (2002) Evaluation of lightning flash rate parameterizations for use in a
global chemical transport model. J Geophys Res-Atmos 107:3851–3876
Baklanov A, Sorensen JH (2001) Parameterisation of radionuclide deposition in atmospheric longrange transport modelling. Phys Chem Earth Pt B 26:787–799
Barfield BJ, Gerber JF (1979) Modification of the aerial environment of plants. American Society of
Agricultural Engineers, St. Joseph, Mich
4 Modelling Atmospheric Nitrogen Deposition in China
81