Abstract. Understanding climate change effects on forests is
important considering the role forests play in mitigating climate change. We
studied the effects of changes in temperature, rainfall, atmospheric carbon
dioxide (CO2) concentration, solar radiation, and number of wet days (as a
measure of rainfall intensity) on net primary productivity (NPP) of the
Zambian Zambezi teak forests along a rainfall gradient. Using 1960–1989 as
a baseline, we projected changes in NPP for the end of the 21st century
(2070–2099). We adapted the parameters of the dynamic vegetation model,
LPJ-GUESS, to simulate the growth of Zambian forests at three sites along a
moisture gradient receiving annual rainfall of between 700 and more than
1000 mm. The adjusted plant functional type was tested against measured
data. We forced the model with contemporary climate data (1960–2005) and
with climatic forecasts of an ensemble of five general circulation models
(GCMs) following Representative Concentration Pathways (RCPs) RCP4.5 and
RCP8.5. We used local soil parameter values to characterize texture and
measured local tree parameter values for maximum crown area, wood density,
leaf longevity, and allometry. The results simulated with the LPJ-GUESS
model improved when we used these newly generated local parameters,
indicating that using local parameter values is essential to obtaining
reliable simulations at site level. The adapted model setup provided a
baseline for assessing the potential effects of climate change on NPP in the
studied Zambezi teak forests. Using this adapted model version, NPP was
projected to increase by 1.77 % and 0.69 % at the wetter Kabompo and by
0.44 % and 0.10 % at the intermediate Namwala sites under RCP8.5 and
RCP4.5 respectively, especially caused by the increased CO2 concentration by
the end of the 21st century. However, at the drier Sesheke site, NPP
would respectively decrease by 0.01 % and 0.04 % by the end of the
21st century under RCP8.5 and RCP4.5. The projected decreased NPP under
RCP8.5 at the Sesheke site results from the reduced rainfall coupled with
increasing temperature. We thus demonstrated that differences in the amount
of rainfall received in a site per year influence the way in which climate
change will affect forest resources. The projected increase in CO2
concentration would thus have more effects on NPP in high rainfall receiving
areas, while in arid regions, NPP would be affected more by the changes in
rainfall and temperature. CO2 concentrations would therefore be more
important in forests that are generally not temperature- or precipitation-limited; however, precipitation will continue to be the limiting factor in
the drier sites.
Historical climate controls soil respiration responses to current soil moisture
