Sensitivity of simulated CO&amp;lt;sub&amp;gt;2&amp;lt;/sub&amp;gt; concentration to sub-annual variations in fossil fuel CO&amp;lt;sub&amp;gt;2&amp;lt;/sub&amp;gt; emissions

Xia Zhang,David Baker,Kevin R Gurney,Yu-Ping Liu,Peter Rayner

doi:10.5194/acp-16-1907-2016

Abstract

Abstract. Recent advances in fossil fuel CO2 (FFCO2) emission inventories enable sensitivity tests of simulated atmospheric CO2 concentrations to sub-annual variations in FFCO2 emissions and what this implies for the interpretation of observed CO2. Six experiments are conducted to investigate the potential impact of three cycles of FFCO2 emission variability (diurnal, weekly and monthly) using a global tracer transport model. Results show an annual FFCO2 rectification varying from −1.35 to +0.13 ppm from the combination of all three cycles. This rectification is driven by a large negative diurnal FFCO2 rectification due to the covariation of diurnal FFCO2 emissions and diurnal vertical mixing, as well as a smaller positive seasonal FFCO2 rectification driven by the covariation of monthly FFCO2 emissions and monthly atmospheric transport. The diurnal FFCO2 emissions are responsible for a diurnal FFCO2 concentration amplitude of up to 9.12 ppm at the grid cell scale. Similarly, the monthly FFCO2 emissions are responsible for a simulated seasonal CO2 amplitude of up to 6.11 ppm at the grid cell scale. The impact of the diurnal FFCO2 emissions, when only sampled in the local afternoon, is also important, causing an increase of +1.13 ppmv at the grid cell scale. The simulated CO2 concentration impacts from the diurnally and seasonally varying FFCO2 emissions are centered over large source regions in the Northern Hemisphere, extending to downwind regions. This study demonstrates the influence of sub-annual variations in FFCO2 emissions on simulated CO2 concentration and suggests that inversion studies must take account of these variations in the affected regions.

Highlights

Quantification of the spatial and temporal distribution of carbon sources and sinks is critical for projecting future atmospheric CO2 concentrations and climate change (Field et al, 2007)
In atmospheric CO2 inversions, fossil fuel CO2 (FFCO2) emissions are often treated as a known quantity in the system; uncertainty in FFCO2 emissions is not considered explicitly and errors in the distribution of simulated atmospheric FFCO2 are translated into errors in the terrestrial biospheric flux estimates
To understand the temporal variations in the input FFCO2 emission fields used in the simulations, we focus attention on areas of the planet with large FFCO2 emissions, what we refer to as the “large source regions” (LSRs)

Summary

Introduction

Quantification of the spatial and temporal distribution of carbon sources and sinks is critical for projecting future atmospheric CO2 concentrations and climate change (Field et al, 2007). In atmospheric CO2 inversions, fossil fuel CO2 (FFCO2) emissions are often treated as a known quantity in the system; uncertainty in FFCO2 emissions is not considered explicitly and errors in the distribution of simulated atmospheric FFCO2 are translated into errors in the terrestrial biospheric flux estimates. This problem has not been well studied, due mainly to limitations such as the coarse resolution of traditional FFCO2 inventories, the sparse monitoring of atmospheric CO2 concentrations, and sub-grid parameterization of atmospheric transport models. The network of atmospheric high-frequency CO2 concentration

Methods

Results

Conclusion