Updated soil map and soil moisture initialization in weather research and forecasting model

Abstract

Numerical weather prediction uses mathematical models of the atmosphere and oceans to predict the weather based on current weather conditions. In order to obtain realistic weather predictions, they require accuracy initial conditions (ICs) and boundary conditions (BCs) to drive the meteorological model. In this study, we focus on the surface heat exchange which depends strongly with soil composition and its moisture. Conventionally, the soil moisture is obtained from the NCEP FNL (Final) Operational Global Analysis data which are on 1-degree by 1-degree grids prepared operationally every six hours. The default soil map is generated from FAO-UNESCO soil classification system which was originally a soil legend correlating the variety of soil surveys throughout the world. The two inputs have coarse resolutions and appear the lack of accuracy, in particular the areas in China. We updated the China soil map with data from the Beijing Normal University with resolution down to 1km by 1km and proposed an alternative method for soil moisture initialization. The simulations of the WRF model showed that the spinning-up soil moisture improve the near-surface temperature prediction by using different soil moisture initialization. By performing process analysis, we are able to explain how the spinning-up soil moisture improve the near-surface temperature and the subsequent effect on air quality modelling by using the box model.

Date of Award	2014
Original language	English
Awarding Institution	The Hong Kong University of Science and Technology