Characterizing the spatiotemporal variability of groundwater levels of alluvial aquifers in different settings using drought indices

Abstract

Abstract. To improve the understanding of how aquifers in different alluvial settings respond to extreme events in a changing environment, we analyze standardized time series of groundwater levels (Standardized Groundwater level Index -SGI), precipitation (Standardized Precipitation Index -SPI), and river stages of three subregions within the catchment of the river Mur (Austria). Using correlation matrices, differences and similarities between the subregions, ranging from the Alpine upstream part of the catchment to its shallow foreland basin, are identified and visualized.Generally, river stages exhibit the highest correlations with groundwater levels, frequently affecting not only the wells closest to the river, but also more distant parts of the alluvial aquifer. As a result, human impacts on the river are transferred to the aquifer, thus affecting the behavior of groundwater levels. Hence, to avoid misinterpretation of groundwater levels in this type of setting, it is important to account for the river and human impacts on it.While the river is a controlling factor in all of the subregions, an influence of precipitation is evident too. Except for deep wells found in an upstream Alpine basin, groundwater levels show the highest correlation with a precipitation accumulation period of 6 months (SPI6). The correlation in the foreland is generally higher than that in the Alpine subregions, thus corresponding to a trend from deeper wells in the Alpine parts of the catchment towards more shallow wells in the foreland.Extreme events are found to affect the aquifer in different ways. As shown with the well-known European 2003 drought and the local 2009 floods, correlations are reduced under flood conditions, but increased under drought. Thus, precipitation, groundwater levels and river stages tend to exhibit uniform behavior under drought conditions, whereas they may show irregular behavior during floods. Similarly, correlations are found to be weaker in years with little snow as compared with those with much snow. This is in agreement with typical aquifer response times over 1 month, suggesting that short events such as floods will not affect much of the aquifer, whereas a long-term event such as a drought or snow-rich winter will.Splitting the time series into periods of 12 years reveals a tendency towards higher correlations in the most recent time period from 1999 to 2010. This time period also shows the highest number of events with SPI values below −2. The SGI values behave in a similar way only in the foreland aquifer, whereas the investigated Alpine aquifers exhibit a contrasting behavior with the highest number of low SGI events in the time before 1986. This is a result of overlying trends and suggests that the groundwater levels within these subregions are more strongly influenced by direct human impacts, e.g., on the river, than by changes in precipitation. Thus, direct human impacts must not be ignored when assessing climate change impacts on alluvial aquifers situated in populated valleys.

Analysis of hydrological data with correlation matrices: technical implementation and possible applications

“…A detailed description of this platform can be found in Haas and Birk (2017). As mentioned in the "Introduction" section, there are other possible sources for data and the tools provided herein can be adapted to other data sources or CSV files structured differently.…”

“…One aspect only demonstrated tangentially by Haas and Birk (2017) is the ability to use the correlation matrix as a tool for data quality control and for the identification of one or several time series showing a behavior deviating from the others. Figure 2 shows the correlation matrix for the Aichfeld region in the Styrian Mur catchment adapted from Haas and Birk (2017), which is used to illustrate this aspect in more detail.…”

“…In cases less severe than those, the module simply pads missing data with the previous water level or precipitation amount. We deem this an acceptable workaround for the low number of affected time series [less than 10% of the data used in Haas and Birk (2017) for example], but for larger issues it is upon the user to assess a more appropriate way to deal with missing data.…”

“…Haas and Birk (2017) demonstrate that this approach of standardizing data and visualizing it using correlation matrices can be useful for analyzing and comparing different types of hydrological data by conducting an analysis of the Austrian Mur valley, showing the effects of drought and flood conditions and the differences between different areas of the catchment. This paper illustrates the methodological details of this approach using python code and discusses potential applications and extensions as well as limitations using examples from further use cases and regions.…”

“…Various methods for standardizing the data such as the Standardized Precipitation Index-SPI (McKee et al 1993), Standardized Groundwater Index-SGI (Bloomfield and Marchant 2013), Standardized River Stages Index-SRSI (Haas and Birk 2017) or the Standardized Precipitation Evapotranspiration Index-SPEI (VicenteSerrano et al 2010) allow the user to compare different time series at different locations. Haas and Birk (2017) demonstrate that this approach of standardizing data and visualizing it using correlation matrices can be useful for analyzing and comparing different types of hydrological data by conducting an analysis of the Austrian Mur valley, showing the effects of drought and flood conditions and the differences between different areas of the catchment.…”