Modelling isoprene emissions over Southern Africa based on climate change scenarios

Abstract

Biogenic volatile organic compounds (BVOCs), in the presence of nitrogen oxide gases (NOx), play a role in the production of tropospheric ozone (O3) which is an effective greenhouse gas and is hazardous to human health (Haagen-Smit, 1952, Chameides et al, 1988, Atkinson, 2000, Kanakidou et al, 2004). Isoprene is a single BVOC that accounts for over 50% of all emitted BVOCs. Isoprene emissions are species specific and vary according to temperature, light and leaf area index. Climate change studies predict that the geographic distribution of species, temperature ranges, light intensity and leaf area index will shift, thus altering future isoprene emissions. Several attempts to model BVOC emissions have been undertaken in an effort to quantify BVOC emission rates and the impact on ozone formation. The most widely used and empirically tested emission algorithms to date were developed by Guenther et al (1993) and are incorporated into the emission model Model of Emissions of Gases and Aerosols from Nature (MEGAN). MEGAN is used in this study to model isoprene emission rates over southern Africa under current and future climate conditions. Current and future climate conditions are taken from the regional climate model, Conformal-Cubic Atmospheric Model (C-CAM), which has been shown to simulate current climate well for the region. Emissions were modelled for January and July only, to represent summer and winter conditions. January isoprene emission rates for the current climate range from 0 to 1.41 gm-2month-1 and total 0.938 Tg of isoprene for the study domain. The highest emission rates are caused by combinations of driving variables which are: high temperature only; high temperature and high leaf area index; high emission factor and high leaf area index. Emission rates effectively shut down in July due to low temperatures and low leaf area index. July emission rates range from 0 to 0.61 gm-2month-1 and total 0.208 Tg of isoprene. Temperature is shown to cause the greatest variation in isoprene emission rates, and thus future scenarios represent an increase in temperature only. The spatial distribution of future emission rates does not shift when compared to current emission rates, but does show an increase in magnitude. Future emission totals for January increase iv by 34% to 1.259 Tg of isoprene and the July emission total increases by 38% to 0.289 Tg of isoprene. Future emission rates responded to temperature as expected, increasing in magnitude, rate of change and range of temperature over which the greatest rate of change occurs. Three areas demonstrating the highest increase in emission rates and highest future emission rates were identified. As temperature was the only variable altered in future scenarios, these areas can be deemed as areas most sensitive to changes in temperature. These areas are situated near the Angola-Namibia border, the Northern Interior of South Africa and the low-lying areas of Mozambique.