On the Optimal Spatial Design for Groundwater Level Monitoring Networks

Ohmer, Marc; Liesch, Tanja; Goldscheider, Nico

doi:10.1029/2019wr025728

Cited by 21 publications

(5 citation statements)

References 62 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Traditional statistical methods can only evaluate the overall variation of the variables in the region but cannot evaluate the spatial distribution of the variables. Geostatistics method can effectively characterize the spatial distribution of its variables by using variogram as the main tool (Júnez-Ferreira et al 2019;Ohmer et al 2019). Kriging interpolation is an interpolation method based on spatial variogram theory and structural analysis (Aryafar et al 2020).…”

Section: Study Area and Data Sourcementioning

confidence: 99%

Spatiotemporal variation and periodic evolution characteristics of groundwater in the Xining area of China, eastern Qinghai–Tibet Plateau

Zhan

et al. 2021

Environ Earth Sci

View full text Add to dashboard Cite

Groundwater, an important source of water, has profound effects on global human survival and production. The spatiotemporal characteristics of groundwater and the periodic variation law of different time scales must be analyzed to grasp the dynamic situation of groundwater and provide scientific guidance for the rational utilization and management of groundwater resources. In this study, the temporal and spatial variability and periodicity of groundwater level were studied in the Xining region of the eastern Qinghai-Tibet Plateau China using traditional statistical method, geographic information system and Morlet wavelet analysis.Results show that the groundwater dynamics in the study area are mainly controlled by three factors, namely, the amount of river water infiltration, discharge from the groundwater, and artificial exploitation. The aforementioned factors can be divided into three dynamic types according to their combination relationship: hydrological, hydrological exploitation, and runoff-discharge types. The groundwater depth showed a trend of first increasing, then stabilizing, and finally decreasing from 1980 to 2020. The analysis in spatial variability demonstrated that the groundwater depth in different periods has a great spatial difference, with a moderate spatial variation intensity. Moreover, the spatial correlation of groundwater in the abundant season is lower than that in the dry season, which is mainly caused by the strengthening of artificial exploitation. The groundwater depth in the Xining region presents a pattern of deep in the south and east and shallow in the north and west by Kriging interpolation of spherical model in geographic information system. Meanwhile, the inter-annual groundwater has continued to decline since the chronic overexploitation between the 1960s and 2000s, with a maximum cumulative depth of 15 m. Then, the amount extraction had been further reduced, and rainfall had significantly increased in recent years.Environmental Earth Sciences (2021) x:x Page 2 of 32 Springer Accordingly, the cone of depression has undergone an evolutionary process from an expansion period to a stable period to a shrinking period. Furthermore, the annual groundwater level of most monitoring wells in study area has the same multi-year scale time variation characteristics with an evident regular periodic variation on the 9-14a and 17-25a time scales by using Morlet wavelet transform analysis. The temporal sequence of groundwater from 1980 to 2020 has the first and second main periods of 12a and 21a. In accordance with the two time-scales, the groundwater level will continue to rise in the short term in the future, which provides a scientific theoretical basis for the long-term sustainable development of groundwater resources and government decision-making.

show abstract

Section: Study Area and Data Sourcementioning

confidence: 99%

Spatiotemporal variation and periodic evolution characteristics of groundwater in the Xining area of China, eastern Qinghai–Tibet Plateau

Zhan

et al. 2021

Environ Earth Sci

View full text Add to dashboard Cite

show abstract

“…Over the past few decades, extensive research have been conducted on designing, optimization methods, and applications of monitoring networks (Mishra & Coulibaly, 2009; Behmel et al., 2016; Chacon‐Hurtado et al., 2017; Jiang et al., 2020). The proposed methods include statistical (Stedinger & Tasker, 1985; Morrissey et al., 1995; H. Xu et al., 2013; Varekar et al., 2016), spatial interpolation (Bayat et al., 2021; Jeong et al., 2019; Ohmer et al., 2019), information‐based (T. Husain & Caselton, 1980; Kornelsen & Coulibaly, 2015; Sreeparvathy & Srinivas, 2020; Ursulak & Coulibaly, 2021), optimization‐based (Kornelsen & Coulibaly, 2015; Luo et al., 2016; Mooley & Ismail, 2009; Ursulak & Coulibaly, 2021; Yoo et al., 2003), sampling strategy (Bras et al., 1988; Tarboton et al., 1987; Tsintikidis et al., 2002), fractal‐based (Capecchi et al., 2011; Esfahani & Datta, 2018; Lovejoy & Mandelbrot, 1985), expert recommendation (Chapman et al., 2016; Laize, 2004; Moss & Karlinger, 1974; Samuel et al., 2013; Skok, 2006), and hybrid methods (Markus et al., 2003; Pardo‐Igúzquiza, 1998). Additionally, the data‐worth (Chadalavada & Datta, 2007; Dausman et al., 2010; Shlomo P.; Man et al., 2017; Neuman et al., 2012) and sensitivity‐based experiments, including D‐optimality for minimizing the determinant (Catania & Paladino, 2009; Knopman et al., 1991; Sciortino et al., 2002; Siade et al., 2017), A‐optimality for minimizing the trace of covariance matrix (Hsu & Yeh, 1989; Knopman et al., 1991; Nishikawa & Yeh, 1989; Vasco et al., 1997), and E‐optimality for minimizing the spectral radius (Sciortino et al., 2002; S. Liu et al.,...…”

Section: Introductionmentioning

confidence: 99%

“…The proposed methods include statistical (Stedinger & Tasker, 1985;Morrissey et al, 1995;H. Xu et al, 2013;Varekar et al, 2016), spatial interpolation (Bayat et al, 2021;Jeong et al, 2019;Ohmer et al, 2019), information-based (T. Husain & Caselton, 1980;Kornelsen & Coulibaly, 2015;Sreeparvathy & Srinivas, 2020;Ursulak & Coulibaly, 2021), optimization-based (Kornelsen & Coulibaly, 2015;Luo et al, 2016;Mooley & Ismail, 2009;Ursulak & Coulibaly, 2021;Yoo et al, 2003), sampling strategy (Bras et al, 1988;Tarboton et al, 1987;Tsintikidis et al, 2002), fractal-based (Capecchi et al, 2011;Esfahani & Datta, 2018;Lovejoy & Mandelbrot, 1985), expert recommendation (Chapman et al, 2016;Laize, 2004;Moss & Karlinger, 1974;Samuel et al, 2013;Skok, 2006), and hybrid methods (Markus et al, 2003;Pardo-Igúzquiza, 1998). Additionally, the data-worth (Chadalavada & Datta, 2007;Dausman et al, 2010;Shlomo P.;Man et al, 2017;Neuman et al, 2012) and sensitivity-based experiments, including D-optimality for minimizing the determinant (Catania & Paladino, 2009;Knopman et al, 1991;Sciortino et al, 2002;…”

mentioning

confidence: 99%

Integration of Deep Learning and Information Theory for Designing Monitoring Networks in Heterogeneous Aquifer Systems

Chen

Dai

Dong

et al. 2022

Water Resources Research

View full text Add to dashboard Cite

Groundwater monitoring networks are direct sources of information for revealing subsurface system dynamic processes. However, designing such networks is difficult due to uncertainties in the spatial heterogeneity of aquifer parameters such as permeability (k). This study combines deep learning and information theory with an optimization framework to address network design problems in heterogeneous aquifer systems. The framework first employs a generative adversarial network to parameterize heterogeneous k distribution using a low-dimensional latent representation. Then, surrogate models are developed based on the deep neural networks to perform uncertainty quantification of pressure heads and solute concentrations at locations of pre-designed candidate monitoring stations. The monitoring stations are then ranked using the greedy search algorithm based on the maximum information minimum redundancy (MIMR) criterion. In order to depict the importance of each candidate monitoring location, the hotspot maps of the selection probability (P s ) are derived from MIMR repetition results. Comprehensive monitoring networks derived from the hotspot maps are then conducted as the final monitoring stations to improve monitoring information compared to MIMR results. Additionally, nine entropy quantization strategies are compared to evaluate their effects on monitoring network optimization results. Results indicate that caution should be taken when selecting entropy quantization strategies to achieve the accuracy required for model calibration and to improve the efficiency of monitoring optimization. Considering high-dimensional uncertainties associated with aquifer parameters, the developed framework can provide important insights for monitoring network designs in various earth observational projects. Plain Language SummaryGroundwater is widely distributed throughout the crust and lithosphere playing a critical role in geological evolution processes and environmental problems. This study presents an integrated framework for designing groundwater monitoring networks in heterogeneous aquifer systems. The proposed model takes advantage of the strength of deep-learning algorithms and deals with high-dimensional and highly non-linear problems of entropy-based methods for designing the monitoring networks. The framework can efficiently provide monitor locations for complex aquifers. The monitoring data collected by the designed network can significantly reduce uncertainties associated with high-dimensional parameters and contain the least amount of redundant information. Further, the proposed method is significant in practice as credible groundwater forecasting requires model calibrations that use monitoring data.

show abstract

“…GMN optimization approaches are commonly divided into the following three categories based on the techniques applied: (a) those based on hydrogeological conceptual models and hydrogeological expert knowledge, (b) those based on numerical groundwater flow models (Kim and Lee, 2007;Singh and Datta, 2016;Thakur, 2017;Sreekanth et al, 2017), and (c) those based on data analysis with (geo-)statistical techniques. Many studies have focused on the geostatistical ability of kriging frameworks to determine new monitoring wells based on the reduction of estimation variance as the optimization criterion (Nunes et al, 2004;Li et al, 2011;Varouchakis and Hristopulos, 2013;Bhat et al, 2015;Thakur, 2015;Ohmer et al, 2019). With the steady increase in computational capacity in recent years, there are a growing number of studies that tackle these optimization problems using traditional datadriven heuristic optimization criteria such as genetic algorithms (GAs; Dhar and Patil, 2012;Reed and Kollat, 2013;Khader and McKee, 2014;Puri et al, 2017;Pourshahabi et al, 2018;Ayvaz and Elçi, 2018;Yudina et al, 2021;Komasi and Goudarzi, 2021), artificial neural networks (ANNs; Alizadeh et al, 2018), particle swarm optimizations (Gaur et al, 2013;Guneshwor et al, 2018;De Jesus et al, 2021), support vector machines (Asefa et al, 2004;Bashi-Azghadi and Kerachian, 2010, SVMs;) and relevance vector machines (RVMs; Khalil et al, 2005;Ammar et al, 2008), or a combination of these approaches.…”

Section: Introductionmentioning

confidence: 99%

Spatiotemporal optimization of groundwater monitoring networks using data-driven sparse sensing methods

Ohmer

Liesch

Wünsch

2022

Hydrol. Earth Syst. Sci.

Self Cite

View full text Add to dashboard Cite

Abstract. Groundwater monitoring and specific collection of data on the spatiotemporal dynamics of the aquifer are prerequisites for effective groundwater management and determine nearly all downstream management decisions. An optimally designed groundwater monitoring network (GMN) will provide the maximum information content at the minimum cost (Pareto optimum). In this study, PySensors, a Python package containing scalable, data-driven algorithms for sparse sensor selection and signal reconstruction with dimensionality reduction is applied to an existing GMN in 1D (hydrographs) and 2D (gridded groundwater contour maps). The algorithm first fits a basis object to the training data and then applies a computationally efficient QR algorithm that ranks existing monitoring wells (for 1D) or suitable sites for additional monitoring (for 2D) in order of importance, based on the state reconstruction of this tailored basis. This procedure enables a network to be reduced or extended along the Pareto front. Moreover, we investigate the effect of basis choice on reconstruction performance by comparing three types typically used for sparse sensor selection (i.e., identity, random projection, and SVD, respectively, PCA). We define a gridded cost function for the extension case that penalizes unsuitable locations. Our results show that the proposed approach performs better than the best randomly selected wells. The optimized reduction makes it possible to adequately reconstruct the removed hydrographs with a highly reduced subset with low loss. With a GMN reduced by 94 %, an average absolute reconstruction accuracy of 0.1 m is achieved, in addition to 0.05 m with a reduction by 69 % and 0.01 m with 18 %.

show abstract

On the Optimal Spatial Design for Groundwater Level Monitoring Networks

Cited by 21 publications

References 62 publications

Spatiotemporal variation and periodic evolution characteristics of groundwater in the Xining area of China, eastern Qinghai–Tibet Plateau

Spatiotemporal variation and periodic evolution characteristics of groundwater in the Xining area of China, eastern Qinghai–Tibet Plateau

Integration of Deep Learning and Information Theory for Designing Monitoring Networks in Heterogeneous Aquifer Systems

Spatiotemporal optimization of groundwater monitoring networks using data-driven sparse sensing methods

Contact Info

Product

Resources

About