日本地球惑星科学連合2022年大会

講演情報

[E] 口頭発表

セッション記号 A (大気水圏科学) » A-HW 水文・陸水・地下水学・水環境

[A-HW25] Near Surface Investigation and Modeling for Groundwater Resources Assessment and Conservation

2022年5月25日(水) 15:30 〜 17:00 301B (幕張メッセ国際会議場)

コンビーナ:Tsai Jui-Pin(National Taiwan University, Taiwan)、コンビーナ:谷口 真人(総合地球環境学研究所)、Yu Hwa-Lung(National Taiwan University)、Chairperson:Jui-Pin Tsai(National Taiwan University, Taiwan)、Hwa-Lung Yu(National Taiwan University)、Ping-Yu Chang(National Central University)

16:45 〜 17:00

[AHW25-06] Groundwater Withdrawal Estimation Using Space-time Pattern Analysis Method and Time Series Decomposition Technique – A Case of Zhoushui Alluvial

*Huating Tseng1Hwa-Lung Yu1 (1.National Taiwan University )

キーワード:Empirical orthogonal function, Timeseries decomposition, Groundwater withdrawal

Groundwater withdrawal estimation is one of the most difficult tasks of groundwater management in agriculture flourished and semi-arid area. This is due to numerous unregistered pumping wells. To efficiently managing water resources and avoiding geological hazard, effective groundwater withdrawal estimation is necessary. The key objective of this study is estimating groundwater pumping pattern just from groundwater level data. And calculate groundwater withdrawal by given storage coefficient. Groundwater level can be considered as composition of several components, e.g., natural recharge, pumping, lateral recharge, etc. That is, pumping pattern can be extracted by removing other components from waterlevel data. Different items have individual spatial distribution (spatial pattern) and frequency. Empirical orthogonal function (EOF) was used to reveal spatial pattern and its correspond temporal pattern from space-time data. The EOF components which have wide-range spatial pattern and its temporal pattern have similar path with rainfall will be considered as natural recharge component in this research. Water level also include low-frequency signal causing by water cycle and some over high frequency signal. This was removed by Ensemble Empirical Mode Decomposition (EEMD), which a time series decomposition method that can decompose signal into several frequencies. Here, we estimate groundwater withdrawal of Zhoushui alluvial for 2008-2019. Pumping pattern was extracted from water level observed data by removing recharge signal and extreme low or high frequency components. Moreover, the storage coefficient, which was estimated in other research by groundwater model i.e. MODFLOW and timeseries method, was used in here to quantize pumping pattern. The result shows the average groundwater withdrawal in Zhoushui Alluvial is 1.63 billion ton per year.