Modelling ozone effects on adult beech trees through simulation of defence, damage and repair costs: implementation of the CASIROZ ozone model in the ANAFORE forest model

Ozone affects adult trees significantly, but effects on stem growth are hard to prove and difficult to correlate with the primary sites of ozone damage at the leaf level. To simulate ozone effects in a mechanistic way, at a level relevant to forest stand growth, we developed a simple ozone damage and repair model (CASIROZ model) that can be implemented into mechanistic photosynthesis and growth models. The model needs to be parameterized with cuvette measurements on net photosynthesis and dark respiration. As the CASIROZ ozone sub-model calculates effects of the ozone flux, a reliable representation of stomatal conductance and therefore ozone uptake is necessary to allow implementation of the ozone sub-model. In this case study the ozone sub-model was used in the ANAFORE forest model to simulate gas exchange, growth, and allocation. A preliminary run for adult beech (Fagus sylvatica) under different ozone regimes at the Kranzberg forest site (Germany) was performed. The results indicate that the model is able to represent the measured effects of ozone adequately, and to distinguish between immediate and cumulative ozone effects. The results further help to understand ozone effects by distinguishing defence from damage and repair. Finally, the model can be used to extrapolate from the short-term results of the field study to long-term effects on tree growth. The preliminary simulations for the Kranzberg beech site show that, although ozone effects on yearly growth are variable and therefore insignificant when measured in the field, they could become significant at longer timescales (above 5 years, 5% reduction in growth). The model offers a possible explanation for the discrepancy between the significant effects on photosynthesis (10 to 30% reductions simulated), and the minor effects on growth. This appears to be the result of the strong competition and slow growth of the Kranzberg forest, and the importance of stored carbon for the adult beeches (by buffering effects on carbon gain). We finally conclude that inclusion of ozone effects into current forest growth and yield models can be an important improvement into their overall performance, especially when simulating younger and less dense forests