How much do radiative transfer models influence red:far-red simulation and subsequent plant photomorphogenesis modelling ?

Abstract

Red:far-red ratio (R:FR) plays an important role in the architectural dynamics of vegetation. The integration of its effects into the crop model and/or into modelling work on plant dynamics over years requires new methods for describing R:FR spatial and temporal variability. This study assesses the sensitivity of simulating plant morphogenesis to the methods of R:FR modelling. The approach consisted in using a generic individual-based legume model coupled with radiative transfer models allowing the computation of R:FR values. Three methods of computation of R:FR were evaluated based on reference radiative transfer model CANESTRA and on turbid-medium model RIRI. The effects of R:FR simulated by the three methods on the simulation of plant morphogenesis were evaluated for the petiole and internodes of two contrasted plant architectures (alfalfa and white clover) at different stages of plant development and density. Plant morphogenesis expressed by the simulated petiole and internode length was sensitive to R:FR values provided by the models, especially at an early stage of development. Contrasted plant architectures exhibit different ranges of sensitivity to R:FR computed by the different models. However, this sensitivity follows a similar pattern between the two types of plant architecture, also for other conditions such as density or stage of development. This study highlights that the choice of radiative transfer model is of main importance for modelling plant morphogenetical responses, in particular at an early stage of plant development. The role of coupling of the FSP and radiative transfer models to address photomorphogenetic issues in order to consider plant-to-plant interactions is discussed.