1Research Institute for Global Change, Japan Agency for Marine-Earth Science and Technology, Yokohama, Japan
2Department of Geography, Tokyo Metropolitan University, Hachioji, Japan
3Atmospheric and Ocean Research Institute, The University of Tokyo, Kashiwa, Japan
4National Institute for Environmental Studies, Tsukuba, Japan
5National Institute of Polar Research, Tokyo, Japan
6Hydrospheric Atmospheric Research Center, Nagoya University, Nagoya, Japan
Abstract. To examine the influence of both crop cultivation and surface air temperatures (SATs) on annual global isoprene and monoterpene emissions, which can lead to the formation of secondary organic aerosols (SOAs), we simulated, on a monthly basis, the annual emissions of volatile organic compounds (VOCs) during the period 1854–2000. The model estimates were based on historical climate data such as SATs, and downward solar radiation (DSR) reproduced with an atmospheric-ocean circulation model, as well as a time series of the global distribution of cropland (to test the hypothesis that conversion of forests into croplands lowers emissions). The simulations demonstrated that global SAT, DSR, the combination of SAT and DSR, and the expansion of cropland all affected emissions. The effect of cropland expansion (i.e., forest conversion) on annual emissions during this period was larger for isoprene (~7% reduction on a global scale) than for monoterpenes (~2% reduction), mainly because of the reduction in broadleaf evergreen forests (BEFs) in Southeast Asia, which have the highest and most constant emissions of isoprene and where both temperature and radiation are high all year round. The reduction in the Amazon region and in parts of Africa, which are other primary sources of annual global isoprene emissions, but where the conversion of BEF to cropland has been much smaller than in Southeast Asia, was less remarkable, probably because the broadleaf deciduous forests and C4 grasslands in these areas have lower and seasonal emissions; hence, their conversion has less effect. On the other hand, the difference in the emission factors (ε) between cropland and the other vegetation types was much lower for monoterpenes than for isoprene, although the ε for cropland was generally the lowest for both compounds. Thus, the expansion of cropland also contributed to the reduction in monoterpene emissions to some degree, but had less effect. A ~5% increase in emissions due to rising SAT was more than offset by the decrease in isoprene emissions and a concurrent ~2% reduction caused by a decrease in DSR. Overall, annual global isoprene emissions in 2000 were lower than in 1854 by 13 TgC yr−1, whereas annual global monoterpene emissions were higher by 2.3 TgC yr−1.