Premium
Multiple Wavelet Coherence to Evaluate Local Multivariate Relationships in a Groundwater System
Author(s) -
Gu Xiufen,
Sun HongGuang,
Zhang Yong,
Yu Zhongbo,
Zhu Jianting
Publication year - 2021
Publication title -
groundwater
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.84
H-Index - 94
eISSN - 1745-6584
pISSN - 0017-467X
DOI - 10.1111/gwat.13068
Subject(s) - groundwater , riparian zone , environmental science , hydrology (agriculture) , bivariate analysis , water level , statistics , geology , geography , mathematics , ecology , geotechnical engineering , cartography , habitat , biology
Abstract Groundwater level fluctuations are affected by surface properties due to complex correlations of groundwater‐surface water interaction and/or other surface processes, which are usually hard to be accurately quantified. Previous studies have assessed the relationship between groundwater level fluctuations and specific controlling factors. However, few studies have been conducted to explore the impact of the combination of multiple factors on the groundwater system. Hence, this paper tries to explore the localized and scale‐specific multivariate relationships between the groundwater level and controlling factors (such as hydrologic and meteorological factors) using bivariate wavelet coherence and multiple wavelet coherence. The groundwater level fluctuations of two wells in areas covered by different plant densities (i.e., the riparian zone of the Colorado River, USA) are analyzed. Main findings include three parts. First, barometric pressure and river stage are the best factors to interpret the groundwater level fluctuations at small scales (<1 day) and large scales (>1 day) at the well of low‐density plants stand, respectively. Second, at the well of high‐density plants stand, the best predictors to control the groundwater level fluctuations include barometric pressure (<1 day), the combination of barometric pressure and temperature (1‐7 days), temperature (7‐30 days), and the combination of barometric pressure, temperature, and river stage (>30 days). The best predictor of groundwater head fluctuations depends on the variance of the vegetation coverage and hydrological processes. Third, these results provide a suite of factors to explain the groundwater level variations, which is an important topic in water‐resource prediction and management.