View : 1122 Download: 0

Full metadata record

DC Field Value Language
dc.contributor.advisor유창현-
dc.contributor.authorDISASO, DAGMAWIT AMAN-
dc.creatorDISASO, DAGMAWIT AMAN-
dc.date.accessioned2018-03-06T16:30:49Z-
dc.date.available2018-03-06T16:30:49Z-
dc.date.issued2018-
dc.identifier.otherOAK-000000147783-
dc.identifier.urihttp://dcollection.ewha.ac.kr/common/orgView/000000147783en_US
dc.identifier.urihttps://dspace.ewha.ac.kr/handle/2015.oak/240417-
dc.description.abstractEthiopia is located in tropical climate region where high seasonal rainfall variability is its main characteristics. Regardless of the fact that seasonal variability over the country is common, the majority of the population is dependent on natural rain for their livelihood, agriculture. Specifically, the Northern Highlands of Ethiopia (8o to 14o N latitude and 36o to 40o E longitude) is the major crop growing area. However, a few studies have been conducted concerning rainfall seasonal (June to September) predictability over the place, specially its association with lag time state of ocean as teleconnection. Historical monthly data from 1981 to 2015 of rainfall, SST and wind were used from different sources. predictors are selected based on correlation analysis between June to September Northern Highlands of Ethiopia (NHE) rainfall index and February to April global sea surface temperature, zonal and meridional wind anomalies. Empirical Orthogonal Function was used to extract and compute the data of the NHE as one homogenous region. By computing seasonal values and standardizing all variable data, cross correlation analysis was conducted with standardized seasonal value of rainfall over the region one by one. The region where the value passes the significance test at 95% confidence interval was selected as candidate predictor. Two models are developed using the different candidate variables of six independent box regions. The one is sea surface temperature north east pacific, zonal wind Northern Pacific and meridional wind North Africa with coefficient of determination r2 = 0.68, Pearson correlation r = 0.82. The other is sea surface temperature north pacific, Northern Indian and zonal wind north Pacific. The coefficient of determination is r2 = 0.58, Pearson correlation is r = 0.76 is found for the second multiple regression model.;에티오피아는 높은 계절적 강수의 변화가 주요 특징인 열대기후 지역이다. 이러한 계절적 변동이 일반적임에도 불구하고, 대다수의 인구는 그들의 생계와 농업을 위한 강수에 의존한다 특히, 에티오피아의 북부 고원지대 (북위8o-14o , 동경36o-40o ) 는 주요 농작물 재배 지역이다. 그러나 이 지역에 대한 강수의 계절적 (6월 - 9월) 예측 특히, 원격 상관과 같은 해양의 시간차와 관련된 연구는 적은 상황이다. 1981년부터 2015년까지의 강수, 해수면 온도, 그리고 바람의 과거 월별 자료를 이용하였다. 예측 변수는 6월부터 9월의 에티오피아 북부 고원(NHE)의 강수 지수와 2월부터 4월의 전 지구적 해수면 온도, 동서 바람과 자오선 바람 편차 간의 상관 분석을 통해 얻은 값을 이용하였다. 또한 NHE의 자료를 동일한 지역으로 계산하고 추출하기 위해 EOF를 사용하였다. 계절별 값을 계산하고 모든 자료를 표준화함으로써, 이 지역의 계절별 강수량을 하나씩 표준화하여 교차 상관 분석을 실시하였다. 95%의 신뢰구간에서 신뢰도 검증을 통과한 지역은 예측 변수가 될 수 있는 후보로 선택된다. 두 모형은 여섯 개의 독립적인 지역에서 각각 다른 예측 변수를 이용하여 설계 되었다. 첫 번째 모델은 동태평양 북부의 해수면 온도, 북태평양의 동서 바람, 북아프리카의 자오선 바람으로 r2 = 0.68 의 결정 계수와, r = 0.82의 Pearson 상관계수를 가진다. 또 다른 하나는 북태평양과 북인도양의 해수면 온도, 북태평양의 동서 바람이다. 두 번째 다중 회귀 모형의 결정 계수는 r2 = 0.58, Pearson 상관계수는 0.82임을 발견하였다.-
dc.description.tableofcontentsI. INTRODUCTION 1 A. Background 1 B. Problem statement 3 C. Objective 3 II. LITERATURE REVIEW 4 A. The rainfall brings systems of summer season over NHE 4 1. Intertropical Convergence Zone (ITCZ) 4 2. Formation of subtropical semi-permanent high-pressure systems 5 3. East African Low-Level Jet (EALLJ) 5 4. African Easterly Jet (AEJ) 6 5. Tropical Easterly Jet (TEJ) 6 B. Ethiopia rainfall and global teleconnection 9 C. The seasons and climate Ethiopia 10 D. Seasonal forecasting system 11 E. Study area 15 1. Geographical description 15 2. Summer rainfall over the Northern highland of Ethiopia 16 III. DATA AND METHODOLOGY 17 A. Data 17 B. Methodology 18 1. Preparing predictand dataset (rainfall) for cross correlation analysis 20 2. Preparing predictor dataset for cross correlation analysis 20 3. Multiple regression model development and leave one out cross validation 21 IV. RESULT AND DISCUSSION 22 A. EOF Analysis 22 B. Association of predictor and JJAS rain over NHE for first model 25 1. North east Pacific SST and JJAS rainfall of NHE 25 2. Zonal wind at 850 hPa and JJA 26 3. Meridional wind at 600 hPa and JJAS rain over NHE 27 4. Leave-one-out cross validation technique used to build the first model 29 C. Association of predictor and JJAS rain over NHE for second model 33 1. North Indian and North Pacific SST and JJAS rainfall of NHE 33 2. Zonal wind at 850 hPa and JJA 34 3. Leave-one-out cross validation technique used to build the first model 36 V. SUMMARY 42 ACKNOWLEDGMENTS 44 REFERENCE 45 APPENDIX 47 ABSTERACT (IN KOREA) 51-
dc.formatapplication/pdf-
dc.format.extent4020262 bytes-
dc.languageeng-
dc.publisher이화여자대학교 대학원-
dc.subject.ddc628-
dc.titleSummer Rainfall Predictability over Northern Highlands of Ethiopia-
dc.typeMaster's Thesis-
dc.format.pagevii, 52 p.-
dc.identifier.thesisdegreeMaster-
dc.identifier.major대학원 기후·에너지시스템공학과-
dc.date.awarded2018.2-
Appears in Collections:
일반대학원 > 기후·에너지시스템공학과 > Theses_Master
Files in This Item:
There are no files associated with this item.
Export
RIS (EndNote)
XLS (Excel)
XML


qrcode

BROWSE