Evaluate dry deposition velocity of the nitrogen oxides using Noah-MP physics ensemble simulations for the Dinghushan Forest, Southern China

There has been a rapid growth of reactive nitrogen (Nr) deposition over the world in the past decades. The Pearl River Delta region is one of the areas with high loading of nitrogen deposition. But there are still large uncertainties in the study of dry deposition because of its complex processes of physical chemistry and vegetation physiology. At present, the forest canopy parameterization scheme used in WRF-Chem model is a single-layer big leaf model, and the simulation of radiation transmission and energy balance in forest canopy is not detailed and accurate. Noah-MP land surface model (Noah-MP) is based on the Noah land surface model (Noah LSM) and has multiple parametric options to simulate the energy, momentum, and material interactions of the vegetation-soil-atmosphere system. Therefore, to investigate the improvement of the simulation results of WRF-Chem on the nitrogen deposition in forest area after coupled with Noah-MP model and to reduce the influence of meteorological simulation biases on the dry deposition velocity simulation, a dry deposition single-point model coupled by Noah- MP and the WRF-Chem dry deposition module (WDDM) was used to simulate the deposition velocity (V-d). The model was driven by the micro-meteorological observation of the Dinghushan Forest Ecosystem Location Station. And a series of numerical experiments were carried out to identify the key processes influencing the calculation of dry deposition velocity, and the effects of various surface physical and plant physiological processes on dry deposition were discussed. The model captured the observed V-d well, but still underestimated the V-d. The self-defect of Wesely scheme applied by WDDM, and the inaccuracy of built-in parameters in WDDM and input data for Noah-MP (e.g. LAI) were the key factors that cause the underestimation of V-d. Therefore, future work is needed to improve model mechanisms and parameterization.