The basic question of Global Change is how humans have altered
atmospheric composition and climate. The industrial revolution is
usually taken as point of reference to determine changes with respect to
the 'pristine baseline'. But human societies have influenced the land
surface, terrestrial trace gas emissions and atmospheric chemical
composition ever since the taming of fire -although it was not until the
mid-Holocene that advances in agriculture and increasing use of coastal
margin productivity went hand in hand with the establishment of sizeable
civilizations. The associated deforestation was negligible for the
atmospheric CO_2 concentration but effects of land clearance on
emissions of non-CO_2 trace gases could well have been noteworthy since
atmospherically reactive chemical species are not well mixed (and thus
of importance to regional chemistry even when changes are not detectable
globally). Frequently they are (unlike, e.g., CO_2 or CH_4 ) not
preserved in ice cores. To assess their pre-anthropogenic baseline
therefore requires modelling that is process-based to be applicable over
environmental conditions encountered over centuries to millennia and
that accounts for human population growth atop of climate effects.
We apply a recently developed model that couples a process-based leaf
isoprene and monoterpene emission algorithm to the dynamic global
vegetation model framework LPJ GUESS. Our model analysis differs in
essential aspects from empirical algorithms used in previous work.
Importantly, the incomplete coupling of assimilation to isoprene
metabolism that underlies the observed CO_2 inhibition of its production
is accounted for. Deforestation associated with the spread and change of
agricultural practices is estimated for the last 6000 years based on the
work by Olofsson and Hickler [2007], and includes estimates of human
biomass burning.
The effects of early human settlements on simulated changes in regional
emission patterns are considerable with, for instance, isoprene
emissions in Europe declining visibly already 3000 years ago. By 1750,
emissions were approximately 30% below the potential natural vegetation
levels, and they continued to decline over the last 200 years as the
CO_2 inhibition of leaf isoprene production became visible. In some
regions increasing human activity resulted in an altered isoprene to
monoterpene ratio. The changes in the regional terrestrial BVOC signals
are also visible in the global totals. The associated fire related
emission patterns are complex and depend on the type of agriculture
(slash and burn /vs/. permanent). Our results point to a discernible
human influence on atmospheric chemistry that dates back much further
than the second half of the 18^th century and that must be considered in
atmospheric chemistry simulations.
21.11.2007
13:30 h