Abstract
An artificial neural network model (ANN) and a geographic information system (GIS) are applied to the mapping of regional groundwater productivity potential (GPP) for the area around Pohang City, Republic of Korea. The model is based on the relationship between groundwater productivity data, including specific capacity (SPC) and its related hydrogeological factors. The related factors, including topography, lineaments, geology, and forest and soil data, are collected and input into a spatial database. In addition, SPC data are collected from 44 well locations. The SPC data are randomly divided into a training set, to analyse the GPP using the ANN, and a test set, to validate the predicted potential map. Each factor’s relative importance and weight are determined by the back-propagation training algorithms and applied to the input factor. The GPP value is then calculated using the weights, and GPP maps are created. The map is validated using area under the curve analysis with the SPC data that have not been used for training the model. The validation shows prediction accuracies between 73.54 and 80.09 %. Such information and the maps generated from it could serve as a scientific basis for groundwater management and exploration.
Résumé
Un modèle de réseau de neurones artificiels (RNA) et un système d’information géographique (SIG) sont utilisés pour cartographier le potentiel de productivité des aquifères au niveau régional (PPA) pour la région autour de Pohang City, République de Corée. Le modèle est basé sur la relation entre les données de productivité des aquifères comprenant le débit spécifique (DS) et les facteurs hydrogéologiques associés. Les facteurs associés tels que topographie, linéaments, géologie, couvert végétal de type forêt et données pédologiques sont collectés et introduits dans une base de données spatiales. De plus, les données de DS sont obtenues au niveau de 44 puits. Ces données sont réparties de manière aléatoire en un ensemble de points d’apprentissage pour analyser le PPA en utilisant le RNA, et en un ensemble de tests afin de valider la carte de potentiel prévisionnelle. L’importance relative et le poids de chaque facteur sont déterminés par des algorithmes rétro-propagation et appliqués au facteur d’entrée. La valeur de PPA est ainsi calculée utilisant les pondérations et des cartes de PPA sont créées. La carte est validée en utilisant la surface sous l’analyse de courbe avec les données de débit spécifique qui n’ont pas été utilisées dans la phase d’apprentissage du modèle. La validation montre des précisions de prévision entre 73.54 et 80.09 %. De telles informations et les cartes ainsi générées peuvent servir de base scientifique pour la gestion des eaux souterraines ainsi que pour la prospection des aquifères.
Resumen
Se aplicaron un modelo de redes neuronales artificiales (ANN) y un sistema de información geográfica (SIG) para el mapeo de la productividad potencial del agua subterránea regional (GPP) para el área alrededor de Pohang City, Republica de Corea. El modelo está basado en la relación entre los datos de productividad de agua subterránea, incluyendo la capacidad específica (SPC) y sus factores hidrogeológicos relacionados. Se recopilaron y reunieron los factores relacionados, incluyendo topografía, lineamientos, geología y datos de forestaciones y suelos y se lo introdujo en una base de datos espaciales. Además, se recolectaron los datos SPC a partir de de 44 ubicaciones de pozos. Los datos SPC son divididos aleatoriamente en un conjunto de entrenamiento, para analizar los GPP usando los ANN, y un conjunto de prueba para validar el mapa de potencial predicho. La importancia y el peso relativo de cada uno de los factores se determinaron por un algoritmo de entrenamiento de retropropagación y se aplicaron a los factores de entrada. Luego se calcula el valor de GPP usando los pesos y se crean los mapas GPP El mapa es validado usando el área bajo la curva de análisis con los datos SPC que no han sido usados para entender el modelo. La validación muestra exactitudes entre 73.54 y 80.09 %. Tal información y los mapas generados a partir de ello podrían servir como una base científica para el manejo y exploración de agua subterránea.
摘要
应用人工神经网络模拟(ANN)和地理信息系统(GIS)绘制韩国浦项市的区域地下水开采潜力(GPP)图。模型基于地下水开采数据之间的关系,包括单位出水量(SPC)及其相关的水文地质要素。收集包括地形、轮廓、地质以及森林土壤数据在内的相关要素,并输入到空间数据库。另外,SPC数据从44个井位中收集,并随机分为一个用ANN分析GPP的训练集和一个用于识别预测潜力图的测试集。 每个要素的相对重要性和权重由反向传递训练算法决定,并应用于输入要素。GPP值通过权重计算,从而产生GPP图。利用未用于训练模型的SPC数据进行曲线下面积分析,对地图进行验证。验证表明预测的准确度在73.54到80.09 %,这些信息和由此产生的地图可为地下水管理及开采提供科学依据。
요약
인공신경망 모델(ANN)과 지리정보시스템(GIS)이 대한민국의 포항시 지역에 대해 광역적 지하수 부존 가능성(GPP)도 작성에 적용되었다. 본 모델은 비양수량(SPC)를 포함하는 지하수 부존 자료와 이와 관련된 수리지질학적 요인들간의 상관관계에 기반을 두었다. 지형, 선구조, 지질, 임상, 토양자료를 포함하는 관련된 요인들은 수집되고 공간데이터베이스로 입력되었다. 또한 SPC자료는 44개의 우물로부터 수집되었다. 이 SPC 자료는 GPP분석을 위한 훈련자료 군과 예측된 GPP도 검증을 위한 검증자료 군으로 무작위적으로 분류되었다. 각 요인의 상대적인 중요성 즉 가중치는 역전파 훈련알고리즘에 의해 결정되고, 입력요인에 적용되었다. 그런 후 GPP 값이 가중치를 이용하여 계산되고 GPP도가 만들어졌다. 이 지도는 훈련에 사용되지 않은 SPC 자료를 이용한 선 아래 면적 방법에 의해 검증되었다. 그 검증 결과 예측정확도는 73.54와 80.09 % 사이로 나타났다. 이러한 정보와 이로부터 만들어진 도면은 지하수 관리와 탐사를 위한 과학적 기초자료로 이용될 수 있다.
Resumo
Utilizam-se um modelo de rede neural artificial (RNA) e um sistema de informação geográfica (SIG) para mapear o potencial de produtividade das águas subterrâneas (PPAS) a nível regional para a área envolvente da Cidade de Pohang, República da Coreia. O modelo baseia-se na relação entre dados de produtividade, nomeadamente a capacidade específica (CES), e os fatores hidrogeológicos associados. Nestes fatores incluem-se a topografia, os lineamentos, a geologia, dados florestais e dados de solos, cuja informação é recolhida e inserida numa base de dados espaciais. Adicionalmente, são recolhidos dados de CES de 44 captações, que são divididos de forma aleatória num conjunto experimental, para analisar o PPAS utilizando a RNA, e num conjunto de teste, para validar o mapa de potenciais estimados. Para cada fator, a importância e o peso relativos são determinados através de algoritmos experimentais por retro-propagação e subsequentemente aplicados. Os valores do PPAS são posteriormente calculados usando os pesos, construindo-se de seguida os mapas de PPAS. Estes são validados com base nos dados de CES não utilizados nas experiências do modelo, aplicando-se a análise de área sob a curva. A validação revela precisões de previsão entre 73.54 e 80.09 %. Esta informação, bem como os mapas gerados a partir dela, poderiam servir como uma base científica para a gestão e exploração das águas subterrâneas.
Similar content being viewed by others
References
Adinarayana J, Krishna NR (1996) Integration of multi-seasonal remotely-sensed images for improved landuse classification of a hilly watershed using geographical information systems. Int J Remote Sens 17:1679–1688
Atkinson PM, Tatnall ARL (1997) Neural networks in remote sensing. Int J Remote Sens 18:699–709
Chenini I, Mammou AB (2010) Groundwater recharge study in arid region: an approach using GIS techniques and numerical modeling. Comput Geosci 36:801–817
Corsini A, Cervi F, Ronchetti F (2009) Weight of evidence and artificial neural networks for potential groundwater spring mapping: an application to the Mt. Modino area (Northern Apennines, Italy). Geomorphology 111:79–87
Dar IA, Sankar K, Dar MA (2010) Remote sensing technology and geographic information system modeling: an integrated approach towards the mapping of groundwater potential zones in Hardrock terrain, Mamundiyar basin. J Hydrol 394:285–295
Engman ET, Gurney RJ (1991) Remote sensing in hydrology. Chapman and Hall, London
Garrett JH (1994) Where and why artificial neural networks are applicable in civil engineering. J Comput Civil Eng 8(2):129–130
Gleick PH (1993) Water in crisis: a guide to the world’s fresh water resources. Oxford University Press, New York
Hardcastle KC (1995) Photolineament factor: a new computer-aided method for remotely sensing the degree to which bedrock is fractured. Photogramm Eng Rem S 61:739–747
Hines JW (1997) Fuzzy and neural approaches in engineering. Wiley, New York
Hwang JH, Kim DH, Cho DL, Song KY (1996) Explanatory note of the Andong sheet. Korea Institute of Geoscience and Mineral Resources, Daejeon, Korea
Israil M, Al-hadithi M, Singhal DC (2006) Application of resistivity survey and geographical information system (GIS) analysis for hydrogeological zoning of a piedmont area, Himalayan foothill region, India. Hydrogeol J 14:753–759
Jaiswal RK, Mukherjee S, Krishnamurthy J, Saxena R (2003) Role of remote sensing and GIS techniques for generation of groundwater prospect zones towards rural development: an approach. Int J Remote Sens 24:993–1008
Jasmin I, Mallikarjuna P (2011) Review: satellite-based remote sensing and geographic information systems and their application in the assessment of groundwater potential, with particular reference to India. Hydrogeol J. doi:10.1007/s10040-011-0712-7
Jha MK, Peiffer S (2006) Applications of remote sensing and GIS technologies in groundwater hydrology: past, present and future. Bayreuth University Press, Bayreuth, Germany
Jha MK, Chowdhury A, Chowdary VM, Peiffer S (2007) Groundwater management and development by integrated remote sensing and geographic information systems: prospects and constraints. Water Resour Manag 21:427–467
Kane WF, Peters DC, Speirer RA (1996) Remote sensing in investigation of engineered underground structures. J Geotech Eng 122:674–681
Kenny JF, Barber NL, Hutson SS, Linsey KS, Lovelace JK, Maupin MA (2009) Estimated use of water in the United States in 2005. US Geol Surv Circ 1344, 52 pp
Kiesel J, Fohrer N, Schmalz B, White MJ (2010) Incorporating landscape depressions and tile drainages of a northern German lowland catchment into a semi-distributed model. Hydrol Process 24:1472–1486
Kim DH, Hwang JH, Park KH, Song KY (1998) Explanatory note of the Busan sheet. Korea Institute of Geoscience and Mineral Resources, Daejeon, Korea
Kim NW, Lee JW, Lee J, Lee JE (2010) SWAT application to estimate design runoff curve number for South Korean conditions. Hydrol Process 24:2156–2170
Koike K, Nagano S, Kawaba K (1998) Construction and analysis of interpreted fracture planes through combination of satellite-image derived lineaments and digital elevation model data. Comput Geosci 24:573–583
Korea Institute of Geoscience and Mineral Resource (1964) Geological map, 1:50,000. KIGMR, Daejeon, Korea
Korean Ministry for Food, Agriculture, Forestry and Fisheries (1985–2005) Rural groundwater survey report. KMFAFF, Seoul
Korean Ministry of Land, Transport and Maritime Affairs (1995) National groundwater monitoring network construction report. KMLMA, Seoul
Korean Ministry of Land, Transport and Maritime Affairs (1998) National groundwater monitoring network construction report. KMLMA, Seoul
Korean Ministry of Land, Transport and Maritime Affairs (2001) National groundwater monitoring network construction report. KMLMA, Seoul
Korean Ministry of Land, Transport and Maritime Affairs (2003) Report of groundwater in Pohang area. KMLMA, Seoul
Korean Ministry of Land, Transport and Maritime Affairs (2006) National groundwater monitoring network construction report. KMLMA, Seoul
Korean Ministry of Land, Transport and Maritime Affairs (2009) National groundwater monitoring network construction report. KMLMA, Seoul
Kumar MG, Bali R, Agarwal AK (2009) Integration of remote sensing and electrical sounding data for hydrogeological exploration: a case study of Bakhar watershed, India. Hydrolog Sci J 54:949–960
Lee S, Dan NT (2005) Probabilistic landslide susceptibility mapping in the Lai Chau province of Vietnam: focus on the relationship between tectonic fractures and landslides. Environ Geol 48:778–787
Lee S, Kim YS, Oh HJ (2012) Application of a weights-of-evidence method and GIS to regional groundwater productivity potential mapping. J Environ Manage 96:91–105
Livingstone DJ, Manallack DT, Tetko IV (1997) Data modelling with neural networks: advantages and limitations. J Comput Aid Mol Des 11:135–142
Madrucci V, Taioli F, de Araújo CC (2008) Groundwater favorability map using GIS multicriteria data analysis on crystalline terrain, São Paulo State, Brazil. J Hydrol 357:153–173
Murthy KSR (2000) Ground water potential in a semi-arid region of Andhra Pradesh: a geographical information system approach. Int J Remote Sens 21:1867–1884
Murthy KSR, Mamo AG (2009) Multi-criteria decision evaluation in groundwater zones identification in Moyale–Teltele subbasin, South Ethiopia. Int J Remote Sens 30:2729–2740
National Academy of Agricultural Science (1977) Detailed soil map, 1:25,000. NAAS, Seoul
National Geographic Information Institute (1997) Digital topographic map, 1: 5,000. NGII, Seoul
O’Leary DW, Friedman JD, Pohn HA (1976) Lineament, linear, lineation; some proposed new standards for old terms. Geol Soc Am Bull 87:1463–1469
Oh H-J, Kim Y-S, Choi J-K, Park E, Lee S (2011) GIS mapping of regional probabilistic groundwater potential in the area of Pohang City, Korea. J Hydrol 399:158–172
Paola JD, Schowengerdt RA (1995) A review and analysis of backpropagation neural networks for classification of remotely sensed multi-spectral imagery. Int J Remote Sens 16:3033–3058
Park YJ, Lee KK, Kim JM (2001) Effects of highly permeable geological discontinuities upon groundwater productivity and well yield. Math Geol 32:605–618
Pradhan B (2009) Groundwater potential zonation for basaltic watersheds using satellite remote sensing data and GIS techniques. Cent Eur J Geosci 1:120–129
Ranganai RT, Ebinger CJ (2008) Aeromagnetic and Landsat TM structural interpretation for identifying regional groundwater exploration targets, south-central Zimbabwe Craton. J Appl Geophys 65:73–83
Sander P, Chesley MM, Minor TB (1996) Groundwater assessment using remote sensing and GIS in a rural groundwater project in Ghana: lessons learned. Hydrogeol J 4:40–49
Saud MA (2010) Mapping potential areas for groundwater storage in Wadi Aurnah Basin, western Arabian Peninsula, using remote sensing and geographic information system techniques. Hydrogeol J 18:1481–1495
Singh CK, Shashtri S, Singh A, Mukherjee S (2011) Quantitative modeling of groundwater in Satluj River basin of Rupnagar district of Punjab using remote sensing and geographic information system. Environ Earth Sci 62:871–881
Solomon S, Quiel F (2006) Groundwater study using remote sensing and geographic information system (GIS) in the central highlands of Eritrea. Hydrogeol J 14:729–741
Srivastava P, Bhattacharya A (2006) Groundwater assessment through an integrated approach using remote sensing, GIS and resistivity techniques: a case study from a hard rock terrain. Int J Remote Sens 27:4599–4620
Todd DK (1980) Groundwater hydrology. Wiley, New York, pp 111–163
Todd DK, Mays LW (2005) Groundwater hydrology. Wiley, New York
Vijay R, Panchbhai N, Gupta A (2007) Spatio-temporal analysis of groundwater recharge and mound dynamics in an unconfined aquifer: a GIS-based approach. Hydrol Process 21:2760–2764
Yeh HF, Lee CH, Hsu KC, Chang PH (2009) GIS for the assessment of the groundwater recharge potential zone. Environ Geol 58:185–195
Zhou W (1999) Verification of the nonparametric characteristics of backpropagation neural networks for image classification. IEEE T Geosci Remote 37:771–779
Acknowledgements
This research was supported by the Basic Research Project of the Korea Institute of Geoscience and Mineral Resources (KIGAM) funded by the Ministry of Knowledge and Economy of Korea.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Lee, S., Song, KY., Kim, Y. et al. Regional groundwater productivity potential mapping using a geographic information system (GIS) based artificial neural network model. Hydrogeol J 20, 1511–1527 (2012). https://doi.org/10.1007/s10040-012-0894-7
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10040-012-0894-7