서브메뉴
검색
Developing High Resolution Soil Moisture Maps Using In Situ, Satellite, and Model-Derived Data in the Contiguous United States
Developing High Resolution Soil Moisture Maps Using In Situ, Satellite, and Model-Derived Data in the Contiguous United States
상세정보
- 자료유형
- 학위논문 서양
- 최종처리일시
- 20260202105634
- ISBN
- 9798297962187
- DDC
- 910
- 저자명
- Eva, Eshita.
- 서명/저자
- Developing High Resolution Soil Moisture Maps Using In Situ, Satellite, and Model-Derived Data in the Contiguous United States
- 발행사항
- [Sl] : The Ohio State University, 2025
- 발행사항
- Ann Arbor : ProQuest Dissertations & Theses, 2025
- 형태사항
- 219 p
- 주기사항
- Source: Dissertations Abstracts International, Volume: 87-05, Section: B.
- 주기사항
- Advisor: Quiring, Steven M.
- 학위논문주기
- Thesis (Ph.D.)--The Ohio State University, 2025.
- 초록/해제
- 요약Soil moisture is a fundamental state variable in climatology, meteorology, and hydrology. Many of the available soil moisture products have a coarse spatial resolution that is not useful for agricultural applications. Therefore, this project will address the critical need to enhance the accuracy and utility of national soil moisture (SM) products by integrating the available data sources such as in-situ, satellite, and model-derived products and downscale them to the finer scale (2-km, 1-km, 700-m, and 400-m). The purpose of this research is to incorporate new sources of soil moisture data and to harness improved downscaling techniques to generate more accurate field-scale soil moisture estimates. The downscaling will apply machine learning and deep learning approaches to generate high resolution soil moisture datasets for the contiguous United States (CONUS). This will be achieved through the following three objectives: (1) determine which variables are most important for downscaling of soil moisture for two volumetric water content (VWC) and percentiles, (2) identify the most appropriate machine learning and deep learning techniques for downscaling soil moisture to field scale, (3) determine the optimal spatial resolution for the downscaled soil moisture. Two distinct sources of soil moisture data were utilized: satellite-derived soil moisture from NASA's Soil Moisture Active Passive (SMAP) mission and model-estimated soil moisture from the North American Land Data Assimilation System (NLDAS).Objective 1 demonstrated that dew point temperature is the most important variable for downscaling SMAP percentiles (0.18), NLDAS VWC (0.27), and NLDAS percentiles (0.17) over CONUS, indicating SMAP percentiles, NLDAS VWC, and NLDAS percentiles contribute 18%, 27%, and 17% of the total power to make it a dominant feature. While elevation is the most important variable for downscaling SMAP VWC (0.28). Dew point temperature is crucial for downscaling in most regions of the United States, except in the South and WestNorthCentral, where elevation is the most important feature. The accuracy of the downscaling varies by region. In the South, SMAP VWC and NLDAS VWC downscaling are relatively accurate. Both have mean absolute errors of ~0.07. The MAE in the South region was 0.196 for SMAP percentiles and 0.175 for NLDAS percentiles.Objective 2 demonstrated that RF was the best method for downscaling soil moisture. It was more accurate than Support Vector Machine (SVM) and Gradient Boosting (XGB) for downscaling VWC at the national level (RF: 0.0816 (SMAP) and 0.0828 (NLDAS), SVM: 0.0873 (SMAP) and 0.0914 (NLDAS), and XGB: 0.0818 (SMAP) and 0.0818 (NLDAS). After RF, XGB also showed good accuracy. XGB also has an advantage because it was faster to run. The SVM had larger errors than the other two methods, and it was slower to run.Objective 3 evaluated how the accuracy of the downscaled soil moisture varied as a function of spatial resolution (2-km, 1-km, 700-m, and 400-m). The results demonstrated that errors tended to decrease slightly at the finer resolutions. For example, the errors associated with SMAP VWC downscaled to 2-km, 1-km, 700-m, and 400-m were 0.0804, 0.0798, 0.0794, and 0.0787, respectively. These findings were consistent between NLDAS and SMAP. While there was some regional variability, the analysis demonstrated that it is possible to generate accurate field-scale (400 m) soil moisture estimates across CONUS.Overall, this doctoral research developed a framework for accurately downscaling satellite and model-derived soil moisture datasets to different spatial resolutions of CONUS. This study represents the national-scale downscaling of soil moisture that is essential for multiple purposes, including hydrology, agriculture, and climatology.
- 일반주제명
- Geography
- 일반주제명
- Soil sciences
- 키워드
- Soil moisture
- 기타저자
- The Ohio State University Geography
- 기본자료저록
- Dissertations Abstracts International. 87-05B.
- 전자적 위치 및 접속
- 로그인 후 원문을 볼 수 있습니다.
MARC
008260126s2025 us c eng d■001000017360891
■00520260202105634
■006m o d
■007cr#unu||||||||
■020 ▼a9798297962187
■035 ▼a(MiAaPQ)AAI32384087
■035 ▼a(MiAaPQ)OhioLINKosu1751561249131565
■040 ▼aMiAaPQ▼cMiAaPQ
■0820 ▼a910
■1001 ▼aEva, Eshita.
■24510▼aDeveloping High Resolution Soil Moisture Maps Using In Situ, Satellite, and Model-Derived Data in the Contiguous United States
■260 ▼a[Sl]▼bThe Ohio State University▼c2025
■260 1▼aAnn Arbor▼bProQuest Dissertations & Theses▼c2025
■300 ▼a219 p
■500 ▼aSource: Dissertations Abstracts International, Volume: 87-05, Section: B.
■500 ▼aAdvisor: Quiring, Steven M.
■5021 ▼aThesis (Ph.D.)--The Ohio State University, 2025.
■520 ▼aSoil moisture is a fundamental state variable in climatology, meteorology, and hydrology. Many of the available soil moisture products have a coarse spatial resolution that is not useful for agricultural applications. Therefore, this project will address the critical need to enhance the accuracy and utility of national soil moisture (SM) products by integrating the available data sources such as in-situ, satellite, and model-derived products and downscale them to the finer scale (2-km, 1-km, 700-m, and 400-m). The purpose of this research is to incorporate new sources of soil moisture data and to harness improved downscaling techniques to generate more accurate field-scale soil moisture estimates. The downscaling will apply machine learning and deep learning approaches to generate high resolution soil moisture datasets for the contiguous United States (CONUS). This will be achieved through the following three objectives: (1) determine which variables are most important for downscaling of soil moisture for two volumetric water content (VWC) and percentiles, (2) identify the most appropriate machine learning and deep learning techniques for downscaling soil moisture to field scale, (3) determine the optimal spatial resolution for the downscaled soil moisture. Two distinct sources of soil moisture data were utilized: satellite-derived soil moisture from NASA's Soil Moisture Active Passive (SMAP) mission and model-estimated soil moisture from the North American Land Data Assimilation System (NLDAS).Objective 1 demonstrated that dew point temperature is the most important variable for downscaling SMAP percentiles (0.18), NLDAS VWC (0.27), and NLDAS percentiles (0.17) over CONUS, indicating SMAP percentiles, NLDAS VWC, and NLDAS percentiles contribute 18%, 27%, and 17% of the total power to make it a dominant feature. While elevation is the most important variable for downscaling SMAP VWC (0.28). Dew point temperature is crucial for downscaling in most regions of the United States, except in the South and WestNorthCentral, where elevation is the most important feature. The accuracy of the downscaling varies by region. In the South, SMAP VWC and NLDAS VWC downscaling are relatively accurate. Both have mean absolute errors of ~0.07. The MAE in the South region was 0.196 for SMAP percentiles and 0.175 for NLDAS percentiles.Objective 2 demonstrated that RF was the best method for downscaling soil moisture. It was more accurate than Support Vector Machine (SVM) and Gradient Boosting (XGB) for downscaling VWC at the national level (RF: 0.0816 (SMAP) and 0.0828 (NLDAS), SVM: 0.0873 (SMAP) and 0.0914 (NLDAS), and XGB: 0.0818 (SMAP) and 0.0818 (NLDAS). After RF, XGB also showed good accuracy. XGB also has an advantage because it was faster to run. The SVM had larger errors than the other two methods, and it was slower to run.Objective 3 evaluated how the accuracy of the downscaled soil moisture varied as a function of spatial resolution (2-km, 1-km, 700-m, and 400-m). The results demonstrated that errors tended to decrease slightly at the finer resolutions. For example, the errors associated with SMAP VWC downscaled to 2-km, 1-km, 700-m, and 400-m were 0.0804, 0.0798, 0.0794, and 0.0787, respectively. These findings were consistent between NLDAS and SMAP. While there was some regional variability, the analysis demonstrated that it is possible to generate accurate field-scale (400 m) soil moisture estimates across CONUS.Overall, this doctoral research developed a framework for accurately downscaling satellite and model-derived soil moisture datasets to different spatial resolutions of CONUS. This study represents the national-scale downscaling of soil moisture that is essential for multiple purposes, including hydrology, agriculture, and climatology.
■590 ▼aSchool code: 0168.
■650 4▼aGeography
■650 4▼aSoil sciences
■653 ▼aSoil moisture
■653 ▼aDeep learning approaches
■653 ▼aGradient Boosting
■690 ▼a0366
■690 ▼a0800
■690 ▼a0481
■71020▼aThe Ohio State University▼bGeography.
■7730 ▼tDissertations Abstracts International▼g87-05B.
■790 ▼a0168
■791 ▼aPh.D.
■792 ▼a2025
■793 ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17360891▼nKERIS▼z이 자료의 원문은 한국교육학술정보원에서 제공합니다.


