Uncertainties of biogenic VOC emissions caused by land cover data and implications on ozone mitigation strategies for the Yangtze river Delta region

Abstract

Biogenic volatile organic compounds (BVOC) play a crucial role in ground-level ozone formation, yet emission quantification faces considerable uncertainties stemming from input data. Land cover data, which determines the distribution of growth forms, is an essential driving input of BVOC emissions. This study explores the uncertainty in BVOC emission estimates arising from using two land cover datasets, specifically the MODIS and ESA land cover data, and the implications for ozone mitigation strategies in the Yangtze River Delta (YRD) region. By employing the MEGAN model for the year 2021, we estimated that ESA land cover data, with a finer 10 m resolution, yields a total BVOC emission estimate of 2299 Gg, which is over 52.9% higher than the MODIS dataset with a resolution of 500 m, attributed to the higher tree coverage (33.9% in ESA vs. 17.6% in MODIS) identified in the ESA dataset. We then simulated the ozone concentrations by both emission estimates with the WRF-CMAQ model. The elevated BVOC emissions based on the ESA data improved the underestimation of simulated isoprene concentrations and consequently resulted in more than 30% higher domain-averaged MDA8 ozone concentration compared to the MODIS-derived BVOC emission. Regarding ozone mitigation strategy, simulation with the ESA-based BVOC emissions results in 0.3–1.2 μg/m3 less ozone reductions compared to that with the MODIS-based emissions when reducing anthropogenic VOC emissions, but 0.7–2.9 μg/m3 higher ozone reductions when controlling anthropogenic NOx emissions. These disparities are more pronounced at the city level, highlighting the importance of accurate BVOC emission estimates for effective ozone control strategies.