The Use of Leaf Functional Traits for Modeling the Timing and Rate of Canopy Development

Abstract:

Leaves vary in their habit, with some being short lived and possessing high intrinsic photosynthetic rates and others being long lived with lower photosynthetic capacity. Longer lived leaves will thus tend to cost more to produce and be able to assimilate carbon over a longer period of time. The timing and seasonality of forest canopies is a cost benefit strategy for the exploitation of favorable environmental conditions and avoidance of unfavorable conditions. Because of the selective pressure for plants to gather a return on leaf investment in relation to their leaf habit we propose that there is a relationship between plant functional traits and the timing and rate of canopy development. In a recent study it was shown that errors in predicted canopy dynamics could be reduced via a single parameter (τ) which modified the timing and rate of canopy development (Savoy & Mackay 2015). If τ is related to underlying mechanisms of plant physiology then it should vary predictably. To test this we will first examine the relationship between τ and observable biophysical variables which vary in ecologically meaningful ways. Then we will develop a model based on leaf traits which will regulate the timing and rate at which vegetation reaches peak rates of assimilation. The model will then be tested at eddy covariance sites which span a range environmental conditions. Preliminary results demonstrate a strong relationship (R² = 0.58) between estimated values of τ and leaf carbon to nitrogen ratio, which is important for representing the costs of leaf construction and nitrogen investment into photosynthetic machinery of leaves. By developing a canopy seasonality model based on plant functional traits and rooted in the framework of leaf economics it is possible to have a more flexible and generalized model. Such a model will be more adept at making predictions under novel environmental conditions than purely correlative empirical models.