Skip to main content

Advertisement

Log in

Application of “panel-data” modeling to predict groundwater levels in the Neishaboor Plain, Iran

Application d’une modélisation avec “panneau de données” à la prédiction des niveaux de nappe dans la Plaine de Neishaboor, Iran

Aplicación de “datos de panel” para predecir niveles de agua subterránea en la planicie de Neishaboor, Irán

“面板数据”模型在预测伊朗Neishaboor平原地下水位上的应用

Aplicação da modelação de dados em painel para prever os níveis de água subterrânea na planície de Neishaboor, Irão

  • Paper
  • Published:
Hydrogeology Journal Aims and scope Submit manuscript

Abstract

The aim of this research was to predict groundwater levels in the Neishaboor plain, Iran, using a “panel-data” model. Panel-data analysis endows regression analysis with both spatial and temporal dimensions. The spatial dimension pertains to a set of cross-sectional units of observation. The temporal dimension pertains to periodic observations of a set of variables characterizing these cross-sectional units over a particular time span. Firstly, the available observation wells in the Neishaboor plain were clustered according to their fluctuation behavior using the “Ward” method, which resulted in six areal zones. Then, for each cluster, an observation well was selected as its representative, and for each zone, values of monthly precipitation and temperature, as independent variables, were estimated by the inverse-distance method. Finally, the performance of different panel-data regression models such as fixed-effects and random-effects models were investigated. The results showed that the two-way fixed-effects model was superior. The performance indicators for this model (R 2 = 0.97, RMSE = 0.05 m and ME = 0.81 m) reveal the effectiveness of the method. In addition, the results were compared with the results of an artificial-neural-network (ANN) model, which demonstrated the superiority of the panel-data model over the ANN model.

Résumé

Le but de cette recherche était de prédire les niveaux de la nappe dans la plaine de Neishaboor, Iran, en utilisant un modèle avec “panneau de données”. L’analyse avec “panneau de données” fonde une analyse de régression comportant à la fois des dimensions spatiale et temporelle. La dimension spatiale se rapporte à un ensemble d’unités d’observation transversales. La dimension temporelle se rapporte aux observations périodiques d’un ensemble de variables caractérisant ces unités transversales sur un laps de temps donné. Dans un premier temps, les puits d’observation disponibles dans la Plaine de Neishaboor ont été regroupés en fonction de leur mode de fluctuation, en utilisant la méthode “Ward” qui a abouti à la définition de 6 zones. Ensuite, dans chaque groupe, un puits d’observation a été choisi pour sa représentativité et dans chaque zone, les valeurs de précipitations et de températures mensuelles, considérées comme variables indépendantes, ont été estimées par la méthode “inverse de la distance”. Enfin, la performance des différents modèles de régression par “panneau de données”, tels que les modèles à paramètres fixes et les modèles à paramètres aléatoires, a été examinée. Les résultats montrent que le modèle à paramètres fixes binaire est meilleur. Les indicateurs de performance de ce modèle (R 2 = 0.97, RMSE = 0.05 m et ME = 0.81 m) révèlent l’efficacité de la méthode. De plus, les résultats ont été comparés à ceux d’un modèle de réseau de neurones artificiel (RNA), ce qui a démontré la supériorité du modèle avec “panneau de données” sur le modèle RNA.

Resumen

En esta investigación se predicen los niveles de agua subterránea en la planicie de Neishaboor, Irán, usando un modelo de “datos de panel”. El análisis de “datos de panel” está dotado de un análisis de regresión para las dimensiones espaciales y temporales. La dimensión espacial se refiere a un conjunto de unidades de secciones transversales de observación. La dimensión temporal se refiere a observaciones periódicas de un conjunto de variables que caracterizan estas unidades de secciones transversales en un lapso de tiempo determinado. En primer lugar, los pozos de observación disponibles en la planicie de Neishaboor se agruparon de acuerdo al comportamiento de las fluctuaciones usando el método “Ward”, lo cual se tradujo en seis zonas de áreas. Luego, para cada grupo se seleccionó un pozo de observación como representativo, y para cada zona se estimaron los valores de precipitación y temperatura mensual, como variables independientes, por el “método inverso de la distancia”. Finalmente, se investigaron las performances de diferentes modelos de regresión de “paneles de datos” tales como los modelos de efectos fijos y los modelos de efectos aleatorios. Los resultados mostraron que las dos formas del modelo de efectos fijos fueron superiores. Los indicadores de performance para este modelo (R 2 = 0.97, RMSE = 0.05 m and ME = 0.81 m) revelan la efectividad del método. Además, .los resultados se compararon con los resultados de un modelo artificial de redes neuronales (ANN), lo cual demostró la superioridad del modelo de “datos de panel” sobre el modelo ANN.

摘要

本次研究旨在利用“面板数据”模型预测伊朗Neishaboor平原的地下水位. “面板数据”可进行分析时间和空间尺度上的回归分析. 空间尺度上适用于一系列代表性单元的观测, 时间尺度上适用于一系列以典型时间跨度上的代表性单元为特征的变量的周期性观测. 首先, 根据使用“监视”方法得到的水位波动情况对Neishaboor平原的可利用的观测井进行分组, 分为六个区域. 然后, 对于每一个群组, 以观测井作为代表, 对每一个区域的月降水量和温度值作为独立变量, 通过“反距离方法”进行估计. 最后对不同的面板数据回归模型, 如固定效应和随机效应模型的效果进行研究. 结果表明, 双向固定效应模型更好. 模型的成绩衡量指标 (R 2 = 0.97, RMSE = 0.05 m and ME = 0.81 m) 显示了该法的有效性. 并将该结果与人工神经网络 (ANN) 模型的结果进行比较, 表明面板数据模型优于ANN模型.

Resumo

O objectivo desta investigação consistiu em prever os níveis de água subterrânea na planície de Neishaboor, Irão, usando um modelo de dados em painel (“panel-data”). A análise dados em painel contempla uma análise de regressão com dimensões espaciais e temporais. A dimensão espacial está relacionada com um conjunto de unidades de perfis de observação. A dimensão temporal está relacionada com observações periódicas de um conjunto de variáveis que caracterizam aquelas unidades de perfis durante um intervalo de tempo determinado. Primeiro, todos os furos de observação da planície de Neishaboor foram agrupados de acordo com o seu comportamento no que toca às oscilações, usando o método “Ward”, de que resultaram seis zonas superficiais. Depois, foi seleccionado um furo representativo de cada grupo e, para cada zona, foram estimados os valores de precipitação mensal e temperatura, como variáveis independentes, usando o método do inverso da distância. Por último, foi investigado o desempenho de diferentes modelos de regressão de dados em painel, tais como os modelos de efeitos-fixos (“fixed-effects”) e de efeitos-aleatórios (“random-effects”). Os resultados mostraram que o modelo de efeitos-fixos duplo sentido (“two-way”) foi superior. Os indicadores de desempenho deste modelo (R 2 = 0.97, RMSE = 0.05 m and ME = 0.81 m) evidenciam a eficácia do método. Adicionalmente, os resultados foram comparados com os resultados de um modelo de rede neuronal artificial (RNA), tendo-se demonstrado a superioridade do modelo de dados em painel em relação ao modelo de RNA.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8

Similar content being viewed by others

References

  • Allen RG, Pereira LS, Raes D, Smith M (1998) Crop evapotranspiration: guidelines for computing crop water requirements, irrigation and drain, Paper No. 56. FAO, Rome, 300 pp

  • Allen DM, Schuurman N, Zhang Q (2007) Using fuzzy logic for modeling aquifer architecture. J Geogr Sys 9:289–310

    Article  Google Scholar 

  • Alsaaran NA (2005) Experimental performance of spatial interpolators for groundwater salinity. Arabian J Sci Eng 30(1A):3–15

    Google Scholar 

  • Arbués F, Barberán R, Villanúa I (2004) Price impact on urban residential water demand: a dynamic panel data approach. Water Resour Res 40:W11402. doi:10.1029/2004WR003092

    Article  Google Scholar 

  • Arellano M (2003) Panel data econometrics. Oxford University Press, Oxford

    Book  Google Scholar 

  • Baltagi B (2005) Econometric analysis of panel data, 3rd edn. Wiley, New York, pp 1–77

    Google Scholar 

  • Bierkens MFP (1998) Modeling water table fluctuations by means of a stochastic differential equation. Water Resour Res 34(10):2485–2499

    Article  Google Scholar 

  • Blaney HF, Criddle WD (1950) Determining water requirements in irrigated area from climatologically irrigation data. US Department of Agriculture, Soil Conservation Service, Tech. Paper No. 96, USDA, Washington, DC

  • Breusch TS, Pagan AR (1980) The Lagrange multiplier test and its applications to model specification in econometrics. Rev Econ Studies 47:239–253

    Article  Google Scholar 

  • Browne MW (1974) Generalized least squares estimators in the analysis of covariance structures. South African Statist J 8:1–24

    Google Scholar 

  • Chow GC (1960) Tests of equality between sets of coefficients in two linear regressions. Econometrica 28:591–605

    Article  Google Scholar 

  • Coppola JM, Szidarovszky F, Poulton M, Charles E (2003) Artificial neural network approach for predicting transient water levels in a multi layered groundwater system under variable state, pumping, and climate conditions. J Hydrol Eng 8(6):348–360

    Article  Google Scholar 

  • Coulibaly P, Anctil F, Aravena R, Bernard B (2001) Artificial neural network modeling of water table depth fluctuations. J Hydrol 307(4):92–111

    Google Scholar 

  • Daliakopoulos IN, Coulibaly P, Tsanis IK (2005) Groundwater level forecasting using artificial neural networks. J Hydrol 309(4):229–240

    Article  Google Scholar 

  • David WS (1997) Cluster analysis: multivariate statistics—concepts, models, and applications. Available via DIALOG. http://www.psychstat.missouristate.edu/multibook/mlt04.htm. Cited 20 May 2008

  • De Cian E, Lanzi E, Roson R (2007) The impact of temperature change on energy demand: a dynamic panel analysis. NOTA DI LAVORO 46, FEEM, Milan, Italy

  • Ghahraman B (2006) Time trend in the mean annual temperature of Iran. Turkish J Agri For 30:439–448

    Google Scholar 

  • Ghahraman B, Taghvaeian S (2008) Investigation of annual rainfall trends in Iran. J Agri Sci Tech 10:93–97

    Google Scholar 

  • Gujarati D (2003) Basic econometrics, 4th edn. McGraw Hill, New York

    Google Scholar 

  • Hall BH (1987) The relationship between firm size and firm growth in the US manufacturing sector. J Indus Econ 35:583–606

    Article  Google Scholar 

  • Hargreaves GH, Samani ZA (1982) Estimating Potential Evapotranspiration. J Irri Drain Div 108(3):225–230

    Google Scholar 

  • Hausman JA (1978) Specification tests in econometrics. Econometrica 46:1251–1271

    Article  Google Scholar 

  • Hildreth C (1949) Preliminary considerations regarding time series and/or cross section studies. Cowles Commission Discussion Paper No. 333, Cowles Commision, Colorado Springs, CO

  • Hildreth C (1950) Combining cross section data and time series. Cowles Commission Discussion Paper No. 347, Cowles Commision, Colorado Springs, CO

  • Hoseini A, Farajzadeh M, Velayati S (2005) The water crisis analysis in Neishaboor plain with considering environmental planning (in Persian). Khorassan-Razavi Regional Water Company, Mashad, Iran

  • Hsiao C (2003) Analysis of panel data, 2nd edn. Cambridge University Press, Cambridge, 67 pp

  • Izady A, Soheili M, Davari K, Alizadeh A (2009) Groundwater level forecasting using multivariate-multiple regression model. Proceedings of the 1st International Conference on Water Resources, Shahrood, Iran, 16–18 August 2009

  • Kaastra I, Boyd MS (1995) Forecasting futures trading volume using neural networks. J Futures Markets 15(8):953–970

    Article  Google Scholar 

  • Kangasharju A (2000) Regional variations in firm formation: panel and cross-section data evidence from Finland. Pap Regional Sci 79:355–373

    Article  Google Scholar 

  • Karunanithi N, Grenney WJ, Whitley D, Bovee K (1994) Neural networks for river flow prediction. J Computing Civil Eng 8(2):201–220

    Article  Google Scholar 

  • Knotters M, Van Walsum PE (1997) Estimating fluctuation quantities from time series of water table depths using models with a stochastic component. J Hydrol 197:25–46

    Article  Google Scholar 

  • Leng L, Zhang T, Kleinman L, Zhu W (2007) Ordinary least square regression, orthogonal regression, geometric mean regression and their applications in aerosol science. J Phys Conf Ser 78:012084

    Google Scholar 

  • Liu S (1999) Matrix results on the Khatri-Rao and Tracy-Singh products. Linear Algebra Appl 289:267–277

    Article  Google Scholar 

  • Maier HR, Dandy GC (2000) Neural networks for the prediction and forecasting of water resources variables: a review of modeling issues and applications. Environ Modeling Softw 15:101–124

    Article  Google Scholar 

  • Masters T (1993) Practical neural network recipes in C++. Academic, San Diego, CA

    Google Scholar 

  • Moeltner K, Stoddard S (2004) A panel data analysis of commercial customers’ water price responsiveness under block rates. Water Resour Res 40:W01401. doi:10.1029/2003WR002192

    Article  Google Scholar 

  • Mundlak Y (1978) On the pooling of time series and cross section data. Econometrica 46:69–85

    Article  Google Scholar 

  • Nayak P, Satyaji Rao YR, Sudheer KP (2006) Groundwater level forecasting in a shallow aquifer using artificial neural network approach. Water Resour Manage 2(1):77–99

    Article  Google Scholar 

  • Nerlove M (2000) An essay on the history of panel data econometrics. Working paper, Department of Agricultural and Resource Economics, University of Maryland, College Park, MD

  • Sami SM, Nasr IM, Galal HG (2002) Evaluation of groundwater resources by using fuzzy Logic in Wadi Dara Area, Gulf of Suez, Egypt. 2nd International Conference on Earth Observation and Environmental Information, Cairo, Egypt, November 2000

  • Tabios GQ, Salas JD (1985) A comparative analysis of techniques for spatial interpolation of precipitation. J AM Water Resour Assoc 21(3):365–380

    Article  Google Scholar 

  • Tankersley CD, Graham WD, Haltfield K (1993) Comparison of uni-variate and transfer function models of groundwater fluctuations. Water Resour Res 29(10):2517–3533

    Article  Google Scholar 

  • Trenkler G (1995) A Kronecker matrix inequality with a statistical application. Econometric Theory 11:654–655

    Article  Google Scholar 

  • Van Geer FC, Zuur AF (1997) An extension of Box-Jenkins transfer/noise from time series of groundwater head series. J Hydrol 192:65–80

    Article  Google Scholar 

  • Velayati S, Tavassloi S (1991) Resources and problems of water in Khorasan province (in Persian). Razavi, Mashhad, Iran

    Google Scholar 

  • Wansbeek TJ, Kapteyn A (1982) A simple way to obtain the spectral decomposition of variance components models for balanced data. Commun Statist A11:2105–2112

    Google Scholar 

  • Wansbeek TJ, Kapteyn A (1983) A note on spectral decomposition and maximum likelihood estimation of ANOVA models with balanced data. Statist Probability Lett 1:213–215

    Article  Google Scholar 

  • Ward JH (1963) Hierarchical grouping to optimize an objective function. J AM Statist Assoc 58:236–244

    Google Scholar 

  • Wooldridge JM (2002) Econometric analysis of cross section and panel data. MIT Press, Cambridge, MA

  • Yaffee R (2003) A primer for panel data analysis. Available via DIALOG. http://www.nyu.edu/its/pubs/connect/fall03/yaffee_primer.html. Cited 15 April 2008

  • Zhang X, Fan Sh (2001) How productive is infrastructure? New approach and evidence from rural India. EPTD discussion paper No. 84, EPDT, Washington. DC

Download references

Acknowledgements

We thank Dr Hadi Jabbari Nooghabi and Dr Majid Sarmad from Department of Statistics of Ferdowsi University of Mashhad for their insightful suggestions and recommendations. The authors also wish to acknowledge the editor Dr. Vincent Post, the associate editor Dr Jerry Fairley and Dr Paul Seward, and the other two respectful reviewers whose comments greatly improved the quality of the manuscript. Finally, we would like to thank Richard Boak and the technical editorial advisor Sue Duncan for a thorough technical edit of the manuscript.

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to A. Izady.

Rights and permissions

Reprints and permissions

About this article

Cite this article

Izady, A., Davary, K., Alizadeh, A. et al. Application of “panel-data” modeling to predict groundwater levels in the Neishaboor Plain, Iran. Hydrogeol J 20, 435–447 (2012). https://doi.org/10.1007/s10040-011-0814-2

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10040-011-0814-2

Keywords

Navigation