DO3SE modelling of soil moisture to determine ozone flux to forest trees

The DO3SE (Deposition of O3 for Stomatal Exchange) model is an established tool for estimating ozone (O3) deposition, stomatal flux and impacts to a variety of vegetation types across Europe. It has been embedded within the EMEP (European Monitoring and Evaluation Programme) photochemical model to provide a policy tool capable of relating the flux-based risk of vegetation damage to O3 precursor emission scenarios for use in policy formulation. A key limitation of regional flux-based risk assessments has been the assumption that soil water deficits are not limiting O3 flux due to the unavailability of evaluated methods for modelling soil water deficits and their influence on stomatal conductance (gsto), and subsequent O3 flux. This paper describes the development and evaluation of a method to estimate soil moisture status and its influence on gsto for a variety of forest tree species. This DO3SE soil moisture module uses the Penman-Monteith energy balance method to drive water cycling through the soil-plantatmosphere system and empirical data describing gsto relationships with pre-dawn leaf water status to estimate the biological control of transpiration. We trial four different methods to estimate this biological control of the transpiration stream, which vary from simple methods that relate soil water content or potential directly to gsto, to more complex methods that incorporate hydraulic resistance and plant capacitance that control water flow through the plant system. These methods are evaluated against field data describing a variety of soil water variables, gsto and transpiration data for Norway spruce (Picea abies), Scots pine (Pinus sylvestris), birch (Betula pendula), aspen (Populus tremuloides), beech (Fagus sylvatica) and holm oak (Quercus ilex) collected from ten sites across Europe and North America. Modelled estimates of these variables show consistency with observed data when applying the simple empirical methods, with the timing and magnitude of soil drying events being captured well across all sites and reductions in transpiration with the onset of drought being predicted with reasonable accuracy. The more complex methods, which incorporate hydraulic resistance and plant capacitance, perform less well, with predicted drying cycles consistently underestimating the rate and magnitude of water loss from the soil. A sensitivity analysis showed that model performance was strongly dependent upon the local parameterisation of key model drivers such as the maximum gsto, soil texture, root depth and leaf area index. The results suggest that the simple modelling methods that relate gsto directly to soil water content and potential provide adequate estimates of soil moisture and influence on gsto such that they are suitable to be used to assess the potential risk posed by O3 to forest trees across Europe.