Luận án tiến sĩ mô hình vận chuyển độ ẩm đất 3D - Hyun Il Choi

3-D VOLUME AVERAGED SOIL-MOISTURE TRANSPORT MODEL: A SCALABLE SCHEME FOR REPRESENTING SUBGRID TOPOGRAPHIC CONTROL IN LAND-ATMOSPHERE INTERACTIONS BY HYUN IL CHOI B., Korea University, 1995 DISSERTATION Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Civil Engineering in the Graduate College of the University of Illinois at Urbana-Champaign, 2006 Urbana, IIinois UMI Number: 3242821 INFORMATION TO USERS The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleed-through, substandard margins, and improper alignment can adversely affect reproduction. In the unlikely event that the author did not send a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion.

® UMI UMI Microform 3242821 Copyright 2007 by ProQuest Information and Learning Company. All rights reserved. This microform edition is protected against unauthorized copying under Title 17, United States Code. ProQuest Information and Learning Company 300 North Zeeb Road P.

Box 1346 Ann Arbor, MI 48106-1346 © 2006 by Hyun Il Choi. All rights reserved. CERTIFICATE OF COMMITTEE APPROVAL University of Illinois at Urbana-Champaign Graduate College September 15, 2006 We hereby recommend that the thesis by: HYUN IL CHOI Entitled: 3-D VOLUME AVERAGED SOIL-MOISTURE TRANSPORT MODEL: A SCALABLE SCHEME FOR REPRESENTING SUBGRID TOPOGRAPHIC CONTROL IN LAND-ATMOSPHERE INTERACTIONS Be accepted in partial fulfillment of the requirements for the degree of: Doctor of Philosophy „CS —. Signatures: DirectoPaRResearch - Prof.

Praveen Kumar Head of Department - Committee on Final Examination* LimoconCo__, ChairperXgn- Prof, Praveen Kumar 10 I4. Corfimlitee Member - of. Valocchi Mt —Con Committee Member - Prof, Ximing Cai Yann bo Committee Mémber - Prof. Tracy Twine Committee Member - Dr.

Xin®&hbng Liang Committee Member - * Required for doctoral degree but not for master’s degree Abstract Climate models, both global and regional, have increased in sophistication and are being run at increasingly higher resolutions. The land surface models (LSMs) coupled to these climate models have also evolved from simple bucket models to the new generation models needed to support sophisticated linkages and process interactions at small scales to assess their cumulative impact at larger scales. This is possible by substantially improving the land-surface parameterization in these models to account for subgrid processes. Although topographic data is one of the most readily available high resolution products with continen- tal and global coverage, heterogeneity induced by topographic characteristics, such as slopes and curvatures, is generally not well represented in the models.

These data offer unprece- dented opportunity for representing the scaling issues of the processes, which are controlled by topographic attributes, such as subsurface moisture transport. In most current LSMs, however, soil-moisture transport equations limited to the vertical soil-moisture transport are unable to capture the spatial variability of soil water induced by topography, and models are thus limited in predictability for land surface fluxes as well. Some recent models use the Topmodel framework based on a basin or catchment scale to overcome these shortcomings, but the underpinning assumptions remain questionable at the scale of regional climate model (RCM) applications. Moreover, although the surface runoff is also one of the important com- ponents for the terrestrial hydrologic cycle, most LSMs simplistically estimate it using the soil water budget without any explicit simulation or routing schemes regarding runoff travel time over the basins.

Even this roughly generated runoff is not used as the boundary condi- tion for the subsurface flow calculation, which may result in a mass balance error in water ili cycle. The model errors, in the absence of appropriate parameterizations, often manifest as non-linear drifts in the dynamical response. For the significant improvement in these crude parameterizations in current terrestrial hydrologic schemes of LSMs, the following research tasks are explored in this study. The 3-dimensional (3-D) volume averaged soil-moisture transport (VAST) formulation is derived from the Richards equation to incorporate the lateral flow and subgrid hetero- geneity due to topographic characteristics.

Parameters characterizing subgrid variability are represented as scale dependent statistical functions, of soil-moisture variability dependence on subgrid topographic attributes, with second order approximations. These parameteriza- tions based on limited data sets serve to illustrate the role of subgrid variability, although they are not meant to serve as a universal model. I hope that this illustration will serve to catalyze a more consistent data collection. I find that the lateral and subgrid flux contri- bution plays a significant role in total soil-moisture dynamics, and the flux due to subgrid spatial variability is as much or larger than grid averaged flux, especially in drier condition.

One of the approximated forms of the Saint-Venant equations is the non-inertia diffusion wave (DW) model, which can account for the downstream backwater effect and is known as an efficient method in accuracy and computational time. I have developed a conjunctive surface-subsurface flow model at a large scale, a 1-D diffusion wave (DW) model for the surface flow interacting with the 3-D VAST model for the subsurface flow, for the compre- hensive terrestrial water and energy predictions in LSMs. This conjunctive flow model can explicitly simulate the surface runoff due to both rainfall excess and soil-surface saturation. A selection of numerical implementation schemes is employed for each flow component.

The 3-D VAST model is implemented using a time splitting scheme applying an explicit method for lateral flow after a fully implicit method for the vertical how. The 1-D DW model is then solved by the MacCormack finite difference scheme with the second-order accuracy in both space and time. iv For the implementation of the new developed model, this study also focuses on the de- velopment and construction of appropriate data sets, especially surface boundary conditions (SBCs) specifically designed for mesoscale RCM applications. The new SBCs development motivated by the limitations and inconsistencies of existing SBCs can be readily incorpo- rated into any RCM suitable for U.

climate and hydrology modeling studies. The primary SBCs are currently presented in a RCM domain for the U.S applications at 30-km spacing. The raw data sources and processing procedures, however, are elaborated in detail, by which the SBCs can be readily constructed for any specific RCM domain anywhere in the world. The new conjunctive surface-subsurface flow model is substituted for the existing hydro- logic scheme in the CLM model, the state-of-the-art LSM, to improve the model predictabil- ity and to understand the topographic impact on the terrestrial water and energy balance.

Model simulations are performed at the time step of 10 minutes for a study domain with a basin size of 450,000 km? around the Ohio Valley region using the North American regional reanalysis (NARR) forcing dataset from 1995 to 2000 in the off-line mode. All model simula- tions are performed using the published model parameter values without calibration, except for the hydraulic conductivity reference depth Z, and anisotropy ratio ¢ which are estimated through the sensitivity analyses. The predicted stream flow hydrographs are compared with the observations at the four United States geological survey (USGS) stream gauges selected within the study domain. The simulation results show that the lateral and subgrid fluxes play a significant role in total the soil-moisture dynamics and the spatial distribution of soil water that has a large impact on the surface energy balance as well.

It is also found from the new coupled model simulations that the interaction between surface and subsurface flows and the flow routing scheme improve the stream flow predictions significantly. Ignoring the role of surface flow depth on the infiltration rate causes errors in both surface and subsurface flow predictions. The new CLM model coupled to the improved terrestrial hydrologic scheme using realistic SBCs provides a full suite of modeling capability to characterize surface water and energy fluxes for regional, continental, and global hydrologic studies. To my parents, my sister, my brother, my wife Eunjin, and my daughter Angela.

vi Acknowledgements I wish to express my sincere appreciation to my advisor, Professor Praveen Kumar, for all of his patient guidance, encouragement, and advice throughout my Ph. He kindly invited me to his distinguished research group, when I was a sort of the ugly duckling in the Hydrosystems Laboratory. I again voice my appreciation to him for giving me this opportunity to make me as I am today. I also wish to express my profound gratitude to my co-advisor and supervisor, Dr.

Xin-Zhong Liang, for his strong guidance, advice, and support. I would not be writing this thesis without his full support and counsel. My advisors have always been my role model of the real accomplished scholars. I greatly appreciate my committee members, Professors Albert J.

Valocchi, Ximing Cai, and Tracy E. Twine, for extensive discussions and helpful comments. I am also very grateful for the support from my ex-supervisor, Dr, Yanqing Lian. I would like to thank my colleagues in the State Water Survey, and several former and current fellow graduate students in the Hydrosystems Laboratory.

Special thanks to Ji Chen for providing the motivation, Min Xu and Mingjian Zeng for providing meteorological data, Kingsley Allen, Amanda White, and Ben L. Ruddell for providing the GIS know-how, and Namrata Batra and Francina Dominguez for providing careful review and assistance, for the thesis. I deeply appreciate Professors Marcelo H. Garcia, Larry Di Girolamo, Praveen Kumar, Chris R.

Valocchi, and Robert Wilhelmson who provide fundamental and advanced knowledge to me through intensive and thorough course work. I also want to thank Mrs. Mary Pearson and Robin Ray for friendly assistance with all the vii paper work required for my course work and graduation. The research work for this thesis was supported by the National Oceanic and Atmospheric Administration Center for Atmospheric Sciences (NCAS) Grant 634554172523 awarded to Dr.

Xin-Zhong Liang of Illinois State Water Survey (ISWS). ISWS also provides several other resources for the completion of the work, including office facilities, GIS assistance and observational data. Any opinions, findings, and conclusions or recommendations expressed in this publication are those of the author and do not necessarily reflect the views of NCAS and ISWS. Model development and experiments were supported by the National Center for Supercomputing Applications (NCSA) supercomputing allocation.

Many thanks beyond description go to my parents and my parents in law in Korea for their care, concern, and trust for me. In their eyes, probably, I am still no more than a little kid on the road. Please be happy and healthy! I owe much to my wife, Eunjin. I will never forget her unconditional support, understanding, and devotion to me and our daughter, Angela (Yoonseo).

In closing allow me to dedicate all of my accomplishment to God. I am always looking for Him at the last moment, but He is always steadfast with unending and prodigal love. This is a new starting point in my life. I will live to pursue the real truth.

"Then you will know the truth, and the truth will set you free." (John 8:32) viii Table of Contents List of Tables. eee xii List of Figures 6. xiii Chapter1 INTRODUCTION .1 Land Surface Models. uc c c c ng c tt g g TA v v v vynt ĩ 1.

c c c Q c Q Q c c cv và Tà vài 8 Chapter 2 3-D Volume Averaged Soil-moisture Transport Model. odoir(icấiicicaaaỶŸÝỶÝỶÝeg 9 VAN: .(( ÁNÁÁSÁÀNằ—ããặắặẽềẽềaaặaaaaẦđẦđũÝđỶŸŸÝŸỶÝỶÝ es 13 2.3 Approximations and Closure Parameterizations. Relative Contribution of Subgrid Variabilty .1 The role of subgrid fluxes .2 VAST model application coupled to CLM .aaTA 38 Chapter 3 Surface Boundary Conditions .2 CWRF Representation of Surface Processes .3 General Considerations for Comprehensive SBCs.4 Construction of the CWRF SBCs .1 Surface Characteristic Identification (SCI) .2 Surface Elevation and Derivatives (HSFC, HSDV, HSLD, and HCVD) 55 3.3 Bedrock, Lakebed, or Seafloor Depth (DBED) .4 Soil Sand and Clay Fraction Profiles (SAND and CLAY) .4ð Bottom Soil Temperature(TBS) .6 Land Cover Category (LCC).7 Fractional Vegetation Cover (FVC) .8 Leaf and Stem Area Index (LAI and SAI) .9 Soil Albedo Localization Factor(SALE).