Comparing two data driven interpolation methods for modeling nitrate distribution in aquifer

Tutmez, Bulent; Hatipoglu, Z.

doi:10.1016/j.ecoinf.2009.08.001

Cited by 28 publications

(4 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Previous research comparing IDW and OK found that in general, the two methods perform very similarly, and that IDW often produces a lower error, and this research agrees with that finding [29][30][31]. Furthermore, previous research has also shown that TS fuzzy modeling has the ability to outperform kriging [25,32,33] as it has in this study, albeit by a small amount. Although not clear in this research, the TS fuzzy model did produce a higher mean score than OK for data reduction increments C-F.…”

Section: Discussionsupporting

confidence: 89%

An Investigation of Takagi-Sugeno Fuzzy Modeling for Spatial Prediction with Sparsely Distributed Geospatial Data

Thomas

Khan²,

Valeo

et al. 2021

Environments

View full text Add to dashboard Cite

Fuzzy set theory has shown potential for reducing uncertainty as a result of data sparsity and also provides advantages for quantifying gradational changes like those of pollutant concentrations through fuzzy clustering based approaches. The ability to lower the sampling frequency and perform laboratory analyses on fewer samples, yet still produce an adequate pollutant distribution map, would reduce the initial cost of new remediation projects. To assess the ability of fuzzy modeling to make spatial predictions using fewer sample points, its predictive ability was compared with the ordinary kriging (OK) and inverse distance weighting (IDW) methods under increasingly sparse data conditions. This research used a Takagi–Sugeno (TS) fuzzy modelling approach with fuzzy c-means (FCM) clustering to make spatial predictions of the lead concentrations in soil. The performance of the TS model was very dependent on the number of outliers in the respective validation set. For modeling under sparse data conditions, the TS fuzzy modeling approach using FCM clustering and constant width Gaussian shaped membership functions did not show any advantages over IDW and OK for the type of data tested. Therefore, it was not possible to speculate on a possible reduction in sampling frequency for delineating the extent of contamination for new remediation projects.

show abstract

Section: Discussionsupporting

confidence: 89%

An Investigation of Takagi-Sugeno Fuzzy Modeling for Spatial Prediction with Sparsely Distributed Geospatial Data

Thomas

Khan²,

Valeo

et al. 2021

Environments

View full text Add to dashboard Cite

show abstract

“…These conditions can lead to coma and eventually death. Stomach cancer in adults is another health effect associated with high concentrations of nitrate [7,10]. Increased nitrogen levels also detrimentally affect coastal waters by expediting eutrophication, which impedes growth of submerged aquatic vegetation, necessary for nursery, spawning, and feeding for many species and by decreasing water's dissolved oxygen (DO) level [11].…”

Section: Introductionmentioning

confidence: 99%

“…Nitrate contamination of aquifers is a significant problem in many agricultural areas [6]. Nitrate is the primary form of nitrogen in natural waters [7]. Nitrate is a groundwater contaminant of particular concern because it is both soluble and mobile.…”

Section: Introductionmentioning

confidence: 99%

Predicting Nitrate Concentration and Its Spatial Distribution in Groundwater Resources Using Support Vector Machines (SVMs) Model

Arabgol

Sartaj

Asghari

2015

Environ Model Assess

View full text Add to dashboard Cite

In this paper, a support vector machine (SVM) model was developed to predict nitrate concentration in groundwater of Arak plain, Iran. The model provided a tool for prediction of nitrate concentration using a set of easily measurable groundwater quality variables including water temperature, electrical conductivity, groundwater depth, total dissolved solids, dissolved oxygen, pH, land use, and season of the year as input variables. The data set comprised of 160 water samples representing 40 different wells monitored for 1 year. The associated parameters for the optimum SVM model were obtained using a combination of 4-fold cross-validation and grid search technique. The optimum model was used to predict nitrate concentration in Arak plain aquifer. The SVM model predicted nitrate concentration in training and test stage data sets with reasonably high correlation (0.92 and 0.87, respectively) with the measured values and low root mean squared errors of 0.086 and 0.111, respectively. Finally, the map of nitrate concentration in groundwater was prepared for all four seasons using the trained SVM model and a geographic information system (GIS) interpolation scheme and compared with the results with a physics-based (flow and contaminant) model. Overall, the results showed that SVM model could be used as a fast, reliable, and cost-effective method for assessment and predicting groundwater quality.

show abstract

“…One involves plumes of high pollution concentration, usually associated with a known point source 6,11,12 and the other involves monitoring of major ions. [13][14][15] In both these cases the observed pollution concentration data follow well-behaved statistical distributions, and the geostatistical tools developed over the last decades provide good interpolation solutions. However, many anthropogenic pollutants are very rare in the natural environment and routine observations of their concentrations are expected to be, and in many cases are indeed, of zero concentration.…”

Section: Introductionmentioning

confidence: 99%

Interpolation of extensive routine water pollution monitoring datasets: methodology and discussion of implications for aquifer management

Yuval

Rimon

Graber

et al. 2014

Environ. Sci.: Processes Impacts

View full text Add to dashboard Cite

A large fraction of the fresh water available for human use is stored in groundwater aquifers. Since human activities such as mining, agriculture, industry and urbanisation often result in incursion of various pollutants to groundwater, routine monitoring of water quality is an indispensable component of judicious aquifer management. Unfortunately, groundwater pollution monitoring is expensive and usually cannot cover an aquifer with the spatial resolution necessary for making adequate management decisions. Interpolation of monitoring data is thus an important tool for supplementing monitoring observations. However, interpolating routine groundwater pollution data poses a special problem due to the nature of the observations. The data from a producing aquifer usually includes many zero pollution concentration values from the clean parts of the aquifer but may span a wide range of values (up to a few orders of magnitude) in the polluted areas. This manuscript presents a methodology that can cope with such datasets and use them to produce maps that present the pollution plumes but also delineates the clean areas that are fit for production. A method for assessing the quality of mapping in a way which is suitable to the data's dynamic range of values is also presented. A local variant of inverse distance weighting is employed to interpolate the data. Inclusion zones around the interpolation points ensure that only relevant observations contribute to each interpolated concentration. Using inclusion zones improves the accuracy of the mapping but results in interpolation grid points which are not assigned a value. The inherent trade-off between the interpolation accuracy and coverage is demonstrated using both circular and elliptical inclusion zones. A leave-one-out cross testing is used to assess and compare the performance of the interpolations. The methodology is demonstrated using groundwater pollution monitoring data from the coastal aquifer along the Israeli shoreline. The implications for aquifer management are discussed.

show abstract

Comparing two data driven interpolation methods for modeling nitrate distribution in aquifer

Cited by 28 publications

References 12 publications

An Investigation of Takagi-Sugeno Fuzzy Modeling for Spatial Prediction with Sparsely Distributed Geospatial Data

An Investigation of Takagi-Sugeno Fuzzy Modeling for Spatial Prediction with Sparsely Distributed Geospatial Data

Predicting Nitrate Concentration and Its Spatial Distribution in Groundwater Resources Using Support Vector Machines (SVMs) Model

Interpolation of extensive routine water pollution monitoring datasets: methodology and discussion of implications for aquifer management

Contact Info

Product

Resources

About