Spatiotemporal analysis of wildfire severity in California, United States via satellite imagery and in-situ data, 2000–2023.
學年 114
學期 1
出版（發表）日期 2025-11-12
作品名稱 Spatiotemporal analysis of wildfire severity in California, United States via satellite imagery and in-situ data, 2000–2023.
作品名稱（其他語言）
著者 Christina W. Tsai; Jasmin Yu-Jen Tung; Raegeane Hastanjaya; Yu-Ju Hung; Shiv Mohan
單位
出版者
著錄名稱、卷期、頁數 International Journal of Applied Earth Observation and Geoinformation 145 ,p. 104891
摘要 California experiences hundreds of wildfires each year, posing increasing risks to ecosystems and communities. This study investigates how hydrometeorological drivers influence wildfire occurrence and severity across four ecological zones from 2000 to 2023, using satellite imagery and in situ data. Complementary Ensemble Empirical Mode Decomposition (CEEMD), Pseudo–Tri-Dimensional CEEMD, and Time-Dependent Intrinsic Correlation (TDIC) were applied to decompose short- and long-term variability, and Intensity–Duration–Frequency (IDF) curves were utilized to characterize event intensity and duration. Results show that wildfire severity has increased statewide even as occurrence rates diverged regionally. In the Sierra, the number of fires declined by approximately 25 %, yet the average daily burned area doubled, indicating fewer but more destructive events. In the North Coast, fire frequency decreased after 2010, but severity continued to rise, while the Central and South Coasts experienced more frequent ignitions that peaked around 2018. TDIC-based analysis revealed that wildfire occurrence decreases sharply with higher precipitation (≈ −0.7) and soil moisture (≈ −0.8), while hot and dry conditions strongly amplify fire activity, with temperature (≈ +0.8) and vapor pressure deficit (VPD; ≈ +0.6) emerging as dominant drivers. Vegetation greenness, measured by the Normalized Difference Vegetation Index (NDVI), exerted contrasting effects: in the Sierra, dense summer biomass fueled larger fires once dried (+0.7), whereas in the South Coast, senescent grasses supported more ignitions (−0.8). IDF analysis further demonstrated that two- to five-day events often reached daily intensities exceeding 500,000 ha, while longer events accumulated greater total burned areas despite lower daily rates. These findings provide quantitative evidence that wildfire–climate interactions in California are region-specific and increasingly shaped by hotter, drier conditions, underscoring the need for adaptive fire management tailored to both short-term extremes and long-term fuel dynamics.
關鍵字 Wildfire;California;Complementary ensemble empirical mode decomposition (CEEMD);Pseudo-Tri-Dimensional complementary ensemble empirical mode decomposition (Pseudo-TCEEMD);Time-dependent intrinsic correlation (TDIC);Satellite image time series (SITS)
語言 en
ISSN
期刊性質 國外
收錄於 SCI
產學合作
通訊作者
審稿制度 是
國別 USA
公開徵稿
出版型式 ,電子版
相關連結

機構典藏連結 ( http://tkuir.lib.tku.edu.tw:8080/dspace/handle/987654321/129231 )