Impact of improved land surface representation on modeling land surface atmosphere interactions under heterogeneous soil moisture conditions
Abstract
The purpose of this study was to focus on how soil moisture and vegetation heterogeneity can play an important role on surface processes such as atmospheric feedback under drought and non-drought conditions. The study also aims to improve land surface representation climatology in terms of single and coupled modes of land surface and weather forecasting models. The general methodology was applied to evaluate the performance of the offline high resolution Noah land surface model (version 3.1) versus the Noah land surface model with the photosynthesis-based Gas exchange Evapotranspiration Model (Noah GEM) using different land types such as forest and agricultural areas (i.e., the Niwot Ridge Ameriflux site, USA and the Avignon CarboEurope site, France) The coupled model Weather Research - Advanced Research Version (ARW ver. 3) was also employed to help understand the coupled processes between biochemical plant physiology, soil moisture, and atmosphere. Three cases were conducted: 1) an LLJ (low level jet) event observed on 3 June 2002 over the IHOP (International H20 Project) field experiment, 2) a severe drought from 11-19 June 2006 over the Southern Great Plains region (SGP); and 3) deep and shallow convection from 10-13 June 2007 over the CLASIC (Cloud and Land Surface Interaction Campaign) SGP region. Field experiment and aircraft data from the Ameriflux and CarboEurope site, IHOP, and the CLASIC campaign were used to calibrate and validate the models. All three hypotheses have been answered showing 1) The improved land surface initial conditions (soil moisture and temperature) using a high resolution land data assimilation system (HRLDAS) will lead to enhanced predictions of summer daytime and nighttime mesoscale forcing under both weak and intense synoptically-driven cumulus convection conditions. 2) Vegetation transpiration is more efficient than soil evaporation in transporting moisture from the land surface to the atmosphere during convection simulations. 3) Interactions between local and large-scale land surface heterogeneity can affect regional convection in the Southern Great Plains for both IHOP and the CLASIC field phase (June 2002 and June 2007). This study provides some of the first results highlighting land surface-vegetation-soil moisture-atmospheric feedback as an important factor not only for daytime processes but also for improved simulation of early morning and nighttime convection. Also the improved Noah land surface model predicted more accurate energy flux, cloud radiation, rainfall, soil moisture, and soil temperature during extreme drought conditions and shallow cumulus convection. Future works include the use of finer-scale data assimilation and long-term soil moisture climatology to improve model performance. Additional plant physiological biochemistry formulations need further evaluation. The impact of convection triggers and vegetation transpiration on deep convection also need further investigation.
Degree
Ph.D.
Advisors
Niyogi, Purdue University.
Subject Area
Geographic information science|Environmental science
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.