Incorporating vegetation feedbacks in regional climate modeling over West Africa

Wednesday, 16 December 2015: 11:35
3006 (Moscone West)
Amir Erfanian1, Guiling Wang2, Miao Yu3, Kazi F Ahmed1 and Richard O Anyah4,5, (1)University of Connecticut, Department of Civil and Environmental Engineering, Storrs, CT, United States, (2)University of Connecticut, Department of Civil and Environmental Engineering, Groton, CT, United States, (3)NUIST Nanjing University of Information Science and Technology, Nanjing, China, (4)University of Connecticut, Groton, CT, United States, (5)University of Connecticut, Natural Resources and the Environment, Storrs Mansfield, CT, United States
Despite major advancements in modeling of the climate system, incorporating vegetation dynamics into climate models is still at the initial stages making it an ongoing research topic. Only few of GCMs participating in CMIP5 simulations included the vegetation dynamics component. Consideration for vegetation dynamics is even less common in RCMs. In this study, RegCM4.3.4-CLM4-CN-DV, a regional climate model synchronously coupled with a land surface component that includes both Carbon-Nitrogen (CN) and Dynamic-Vegetation (DV) processes is used to simulate and project regional climate over West Africa. Due to its unique regional features, West Africa climate is known for being susceptible to land-atmosphere interactions, enhancing the importance of including vegetation dynamics in modeling climate over this region. In this study the model is integrated for two scenarios (present-day and future) using outputs from four GCMs participating in CMIP5 (MIROC, CESM, GFDL and CCSM4) as lateral boundary conditions, which form the basis of a multi-model ensemble. Results of model validation indicates that ensemble of all models outperforms each of individual models in simulating present-day temperature and precipitation. Therefore, the ensemble set is used to analyze the impact of including vegetation dynamics in the RCM on future projection of West Africa‚Äôs climate. Results from the ensemble analysis will be presented, together with comparison among individual models.