Tropical climates - The Global Climate System: Patterns, Processes, and Teleconnections - page 78

30
25
20

Figure 3.10 Annual
deforestation rates in the
Brazilian Amazon. (Data
from Brazilian Ministry of
Science and Technology)

15
10
5
0
1991

1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002

results are not, at this time, totally reliable. As Shukla et al.( 1990 ) note, climatic
fluctuations over northern middle latitudes are large, and forced model perturba-
tions make it difficult to draw any definitive conclusions.
At the more local scale it is clear that a number of climate impacts must occur
when a rainforest is replaced by a savanna grassland. There will be increased
surface albedos leading to a modification of day-night temperatures and chang-
ing surface properties that will decrease both roughness and infiltration rates. A
reduction in precipitation is anticipated and may well lengthen the dry seasons that
are experienced in some of the forest climates. The impacts have been shown to
occur in models as illustrated in the research of Dickinson and Kennedy ( 1992 ).
A number of research organizations have been responsible for obtaining and
modeling climate in Amazonia. The type of research completed as part of
ABRACOS (Anglo-Brazilian Amazonian Climate Observation Study) is illu-
strated in the assessment of Amazonian deforestation on climate. In their study,
Gash and Nobre ( 1997 ) review measurements and modeling completed by the
group and demonstrate that energy balance differences between forest and
clearings give higher temperatures in the clearings. Where substantial deforesta-
tion has occurred the higher sensible heat fluxes in the cleared areas produce
deeper convective boundary layers. This leads to variations in cloud cover and
meso-scale circulations.
The Large-Scale Biosphere-Atmosphere Experiment (LBA) in Amazonia is
an international research program led by Brazil. It was designed to generate new
information needed to understand the ''climatological, ecological, biogeochem-
ical, and hydrological functioning of Amazonia, the impact of land use changes
on these functions and the interactions between Amazonia and the Earth
system'' (http:daac.ornl.gov/lba).
The LBA has many components and is perhaps the most comprehensive
source for current Amazonia scientific research. Analyses of largely unknown
properties of the climate are providing some interesting results. For example, it
has been shown (Roberts et al. 2001 ) that cloud condensation nuclei are surpris-
ingly low. Thus the increase that will occur because of human activity in the
region will have a stronger impact on climate than it might in other continental
areas. In the study of ''smoking clouds'' over the Amazon (Andreae et al. 2004 )