Publications

Detailed Information

A Comparative Study on the GLUE Likelihood Definitions for a Hydrologic Model : 수문 모델을 위한 GLUE 우도함수 정의에 관한 비교 연구

DC Field Value Language
dc.contributor.advisorVan Thinh Nguyen-
dc.contributor.author한기돈-
dc.date.accessioned2019-10-18T15:07:40Z-
dc.date.available2019-10-18T15:07:40Z-
dc.date.issued2019-08-
dc.identifier.other000000156922-
dc.identifier.urihttps://hdl.handle.net/10371/160965-
dc.identifier.urihttp://dcollection.snu.ac.kr/common/orgView/000000156922ko_KR
dc.description학위논문(석사)--서울대학교 대학원 :공과대학 건설환경공학부,2019. 8. Van Thinh Nguyen.-
dc.description.abstractA comparative study on the formal and informal likelihood definitions in GLUE methodology was performed. In the case of informal likelihood, uncertainty interval for high siginificance level could not cover extreme values, resulting low statistical reliability. Compared to this, the uncertainty interval from formal likelihood covered extreme values well, resulting better statistical reliability. The reason for this is because formal likelihood case considered stable distribution which can include heavy-tailed distribution as a descriptor of model error term.-
dc.description.abstract본 연구에서는 GLUE 방법론에 등장하는 비형식 우도함수와 형식 우도함수에 대한 비교연구가 수행되었다. 비형식 우도함수의 경우, 불확실성 구간의 크기는 작지만 높은 유의 수준에서 극값을 포함하지 못해 통계적 신뢰성이 떨어지는 것으로 나타났다. 이에 반해 형식 우도함수의 경우, 높은 유의수준에서 극값들을 잘 포함함으로서 통계적 신뢰성이 있는 구간이 산정되었다. 극값을 잘 포함할 수 있었던 이유는 모델 오차항을 고려할 때 꼬리가 두꺼운 분포인 안정분포를 사용하였기 때문인 것으로 판단되었다.-
dc.description.tableofcontentsAbstract i
Table of Contents iii
Lists of Figures vi
List of Tables ix
Chapter 1. Introduction 1
1.1 Background and Necessity of Study 1
1.1.1 Watershed model 1
1.1.2 Uncertainty analysis 3
1.1.3 Necessity of study 7
1.2 Objectives 7
Chapter 2. Theoretical background 10
2.1 Modified Generalized Watershed Loading Function (MGWLF) 10
2.1.1 Water Balance 11
2.1.1.1 Runoff calculation 12
2.1.1.2 Evapotranspiration 14
2.1.1.3 Percolation 15
2.1.1.4 Groundwater discharge and deep seepage 15
2.1.1.5 Streamflow 16
2.1.2 Dissolved nutrient load 17
2.1.2.1 Rural runoff load (DR) 17
2.1.2.2 Groundwater load (DG) 18
2.1.2.3 Septic system load (DS) 18
2.1.3 Solid-phase nutrient load 21
2.1.3.1 Rural runoff load (SR) 21
2.1.3.2 Urban runoff load (SU) 25
2.2 Generalized Likelihood Uncertainty Estimation (GLUE) 27
2.2.1 Parameter sampling 30
2.2.2 Identification of an epistemic error 31
2.2.3 Likelihood measure 32
2.2.3.1 Informal likelihood 32
2.2.3.2 Formal likelihood 36
2.2.4 Parameter posterior probability distribution 42
2.2.5 Uncertainty interval 43
2.2.6 Model validation and prediction 45
Chapter 3. Methodology 46
3.1 SNU-WS 46
3.1.1 Pre-processing 47
3.1.2 MGWLF component 49
3.2 Uncertainty Analysis tool 51
3.2.1 Development of uncertainty analysis input interface 51
3.2.2 Development of uncertainty analysis output interface 54
3.3 Application 57
3.3.1 Parameter setting 57
3.3.2 Reducing epistemic uncertainty 58
3.3.3 Informal likelihood definition 63
3.3.4 Formal likelihood definition 63
Chapter 4. Result 72
4.1 Monthly flow 72
4.1.1 Informal likelihood 72
4.1.2 Formal likelihood 76
4.1.3 Comparison 79
4.2 Daily flow 80
4.2.1 Informal likelihood 80
4.2.2 Formal likelihood 90
4.2.3 Comparison 98
Chapter 5. Summary and Conclusion 100
REFERENCES 103
초록 107
-
dc.language.isoeng-
dc.publisher서울대학교 대학원-
dc.subjectUncertainty analysis-
dc.subjectGLUE methodology-
dc.subjectformal likelihood-
dc.subjectinformal likelihood-
dc.subjecterror structure-
dc.subject.ddc624-
dc.titleA Comparative Study on the GLUE Likelihood Definitions for a Hydrologic Model-
dc.title.alternative수문 모델을 위한 GLUE 우도함수 정의에 관한 비교 연구-
dc.typeThesis-
dc.typeDissertation-
dc.contributor.AlternativeAuthorHan, Gi Don-
dc.contributor.department공과대학 건설환경공학부-
dc.description.degreeMaster-
dc.date.awarded2019-08-
dc.contributor.major수공학-
dc.identifier.uciI804:11032-000000156922-
dc.identifier.holdings000000000040▲000000000041▲000000156922▲-
Appears in Collections:
Files in This Item:

Altmetrics

Item View & Download Count

  • mendeley

Items in S-Space are protected by copyright, with all rights reserved, unless otherwise indicated.

Share