Modeled gross primary productivity in an old-growth tropical rain forest: Effects of climate and canopy complexity

We used MAESTRA, a canopy processes model, to estimate gross primary productivity (GPP) in a Costa Rican tropical rain forest.  The model was parameterized with detailed canopy structure and leaf physiology data gathered from 55 tower transects throughout the old growth forest.  We validated MAESTRA with eddy covariance measurements of the same forest.  Response surfaces of GPP vs. temperature, vapor pressure deficit, and radiation were similar whether modeled with MAESTRA or measured with eddy covariance.  We used MAESTRA to address two hypotheses about the effects of stand development and climate on GPP:  (1) GPP of a younger forest with a thin, uniform canopy will be greater than GPP of an old growth forest with a deep, complex canopy, and (2) GPP will decrease when MAESTRA is run with weather data from El Niño Southern Oscillation year (ENSO, 1998) as opposed to a normal year (1999).  Hypothesis (1) was based on the assumption that net primary production declines with forest age, and we tested whether or not this decline could be the result of decreasing GPP.  This hypothesis was rejected, as we found absorbed radiation and therefore GPP was greater in more complex canopies than thin, uniform canopies, given the same leaf area.  The second hypothesis was supported, as GPP was greater during a normal year than when modeled with weather data from an ENSO year.  This change could account for measured differences in eddy covariance net ecosystem exchange during the ENSO year.