HR: 1340h
AN: B43D-0303    [Abstracts]
TI: Evaluation of Different Parameterizations for Land Surface Processes Over Drying Terrain in West Africa
AU: * Schuettemeyer, D
EM: dirk.schuettemeyer@wur.nl
AF: Meteorology and Air Quality Group Wageningen University and Research Center, Duivendaal 2, Wageningen, 6701 AP Netherlands
AU: Moene, A F
EM: arnold.moene@wur.nl
AF: Meteorology and Air Quality Group Wageningen University and Research Center, Duivendaal 2, Wageningen, 6701 AP Netherlands
AU: Holtslag, B
EM: bert.holtslag@wur.nl
AF: Meteorology and Air Quality Group Wageningen University and Research Center, Duivendaal 2, Wageningen, 6701 AP Netherlands
AU: de Bruin, H
EM: henk.debruin@wur.nl
AF: Meteorology and Air Quality Group Wageningen University and Research Center, Duivendaal 2, Wageningen, 6701 AP Netherlands
AU: Kroon, L
EM: leo.kroon@wur.nl
AF: Meteorology and Air Quality Group Wageningen University and Research Center, Duivendaal 2, Wageningen, 6701 AP Netherlands
AB: In this study different parameterizations for land surface processes currently used in meteorological models at ECMWF and NCEP (TESSEL and NOAH, respectively) are evaluated for two test sites in the Volta Basin in Ghana, West Africa. For this evaluation a new data set is utilized including surface fluxes obtained by scintillometry. The parameterizations are selected because they differ in physical aspects and the degree of complexity with which land surface processes are parameterized. Both parameterizations use the big leaf approach in combination with a formulation of bare soil evaporation. The canopy conductance is parameterized following Jarvis and Stewart. TESSEL uses the "tile" approach for calculating surface fluxes to differentiate between high and low vegetation and bare soil. In NOAH the fractional vegetation cover varies on a monthly basis whereas in TESSEL it is fixed in time. Additionally an approach for NOAH is tested to calculate canopy conductance based on plant physiology, where the photosynthetic assimilation is coupled to the leaf stomatal conductance. This approach has the advantage that it is more physically based and less parameter are needed as compared to the Jarvis-Stewart method mentioned above. Furthermore carbon dioxide fluxes can be calculated based on commonly observed atmospheric variables. All parameterizations have been run in offline mode for a seasonal cycle in 2002/2003 using observations as forcings at two test sites. One site is located in the humid tropical southern region, and one in the dryer northern region. The measurements include the rapid wet-to-dry transition after the wet season at both sites. As a general trend it is found that during the wet part of the season net radiation is described well by all parameterizations. During the drying up the errors in modeled net radiation increases. For the sensible heat flux similar results are obtained, whereas for latent heat flux the results are better. Furthermore the differences in the simulations provide deeper insights into individual characteristics of the parameterizations independent of atmospheric conditions. It is found that TESSEL simulates surface fluxes physically more realistically compared to NOAH with varying fractional vegetation cover. The results for the plant physiology approach are comparable to those from TESSEL. It is concluded that it might be useful to utilize more physically based parameterizations with less parameters in such a climatically sensitive and ecologically unstable region.
DE: 0426 Biosphere/atmosphere interactions (0315)
DE: 0495 Water/energy interactions (1878)
SC: Biogeosciences [B]
MN: Fall Meeting 2005