Testing a Mechanistic Model of Forest-Canopy Mass and Energy Exchange Using Eddy Correlation: Carbon Dioxide and Ozone Uptake by a Mixed Oak-Maple Stand

Js Amthor,Sc Wofsy,Ml Goulden,Jw Munger

doi:10.1071/pp9940623

Abstract

A big-leaf model of C3-canopy mass and energy exchange was used to predict hourly CO2 and O3 uptake by a mixed deciduous Quercus-Acer (oak-maple) stand in central Massachusetts, USA. The model is based on canopy-radiation interactions, leaf mesophyll metabolism (photosynthetic electron transport, carboxylation and oxygenation of ribulose 1,5-bisphosphate [RuP2] by RuP2 carboxylase/oxygenase [Rubisco], and respiration), physical transport conductances of mass and heat above and within the canopy, conductances of mass at the leaf surface and in the mesophyll, and mass and energy exchange at the soil surface (forest floor). Predictions of hourly CO2 and O3 uptake were compared to independent whole-forest CO2 and O3 exchange measurements made by the eddy correlation method during a 68 day period in the summer and early autumn of 1992. Predicted hourly CO2 exchange rate was strongly correlated (r ≈ +0.91) with measured hourly CO2 exchange, but mean day-time predicted whole-forest CO2 uptake was c. 13% (c. 1.13 μmol CO2 m-2 s-1) greater than CO2 uptake measured by eddy correlation. The model tended to overpredict CO2 uptake during late afternoon, but was accurate during the rest of the day. Predicted and measured O3 uptake rates also were positively correlated (r ≈ +0.76). The diurnal patterns of predicted and measured O3 uptake indicated that stomata1 conductance (gs) was accurately predicted during the morning, but in the afternoon the model overpredicted gs. This pattern was consistent with the overprediction of afternoon CO2 uptake, and suggested that a feedback inhibition of photosynthesis occurred in the afternoon. This might have been related to source-sink imbalance following several hours of photosynthesis. On the whole, and in spite of the simplifications inherent in the big-leaf representation of the canopy, the model is useful for predicting forest-environment interactions and for interpreting mass and energy exchange measurements.

Highlights

There are many reasons to quantify the exchange of mass and energy between a forest canopy and the atmosphere
We want the canopy model to be simple enough to be dependent on only a small number of input data and parameters, because our long-term goal is the application of the model to regional biospheric C02 exchange with the atmosphere in order to: (1) assess present C02 source and sink activity of the biosphere, and (2) predict effects of elevated atmospheric COz and other environmental changes on regional ecosystem processes
The eddy correlation measurements did not show a clear PPFD-saturation point, whereas the model predicted PPFD-saturation for the conditions of the study at about 1000 pmol photons m-2 s-', or just over half the maximum hourly PPFD measured during the test period. Both measured and predicted C 0 2 uptake were generally slower at a given PPFD in the afternoon than they were in the morning, as discussed below

Summary

Introduction

There are many reasons to quantify the exchange of mass and energy between a forest canopy and the atmosphere. Forest canopy mass and energy exchanges are key factors in forest growth and ecosystem function (Waring and Schlesinger 1985), forest photosynthesis and net primary production are thought to be large components of the global carbon cycle (Bolin et al 1979), and regional forest canopy physiology may influence weather and hydrology (Shukla et al 1990; Eltahir and Bras 1993). Simulation models must play a role in clarifying the significance of canopy physics and physiology to biospheric productivity, boundary layer meteorology, regional and global carbon cycling, and impacts of global environmental change on forest ecosystems. Models might be used to extrapolate site measurements that are made of forest mass and energy exchange to regional or global scales. We want the canopy model to be simple enough to be dependent on only a small number of input data and parameters, because our long-term goal is the application of the model to regional biospheric C02 exchange with the atmosphere in order to: (1) assess present C02 source and sink activity of the biosphere, and (2) predict effects of elevated atmospheric COz and other environmental changes on regional ecosystem processes

Objectives

Results

Conclusion