Emissions of isoprene, monoterpenes, and other VOC were calculated using the
GLOBEIS model (Guenther et al., 1999) which estimates biogenic VOC emissions
as a function of foliar density, an emission capacity (the emission at specified
environmental conditions), and an emission activity factor that accounts for
variations due to environmental conditions. The foliar densities, emission capacities,
and algorithms used to determine emissions activities for both 2000 and 2100
are the same as those described by Guenther et al. (1995). One difference between
this work and that of Guenther et al. (1995) is that here we used hourly temperatures
for each hour of a month to determine the monthly average emission rate whereas
Guenther et al. (1995) used monthly average temperatures to drive emission algorithms.
This results in about a 20% increase in isoprene emissions in 2000 and 10% increase
in emissions of other biogenic VOC.
Global annual emissions of monoterpenes for 2000 were 146 Tg (compared to
the estimate of 127 Tg in Guenther et al. (1995) which was used for the workshop
2000 emissions). These increase by 23% for 2100 relative to the 2000 scenario.
The changes are much higher at certain seasons and locations. The spatial distribution
of changes in total monoterpene emissions range from a 17% decrease to a 200%
The simple model used for this analysis does not consider a number of other factors that could significantly influence long-term trends in biogenic VOC emissions. For example, we did not consider changes in soil moisture which could significantly impact emission rates. In addition, we did not consider changes in future concentrations of OH, O3, and NO3 in determining the yield of aerosol products. Instead, a constant yield of 11% of the terpene emissions was assumed for all future emissions.
The meteorological variables used to compute the dust emission for 2000 corresponded
to those computed for daily meteorological data from the Data Assimilation Office
analysis for 1990 (the GEOS-1 DAS, see Schubert et al., 1993) using algorithms
outlined by P. Ginoux. For this computation, the total dust flux was scaled
to yield a total emission of about 2000 Tg/yr. Emissions in four size categories
were specified (diameter 0.2 to 12.0 mm).
In order to calculate dust flux in 2100, the monthly mean variables were computed
by averaging the wind speed at the lowest level (100 m) and the soil moisture
content over the ten years ending in 1999 and 2099. The fluxes for the 21st
century were calculated using the same meteorological variables, corresponding
to 1990, scaled by the monthly mean ratio of the equivalent variables computed
by the NCAR CSM model.
Overall, predicted dust emissions increase by approximately 10%. There are substantial increases in some seasons and locations (e.g. increases in Australia (85%) in winter and in Europe (86%) and East Asia (42%) in summer). Because dust may be especially important as an ice-nucleating agent in clouds, these possible changes add substantial uncertainty to the projected future indirect forcing.
Other reports in this collection