بهبود نتایج حاصل از مدل دراستیک با استفاده از هوش مصنوعی جهت ارزیابی آسیب پذیری آبخوان آبرفتی دشت رامهرمز

نوع مقاله: مقاله پژوهشی

نویسندگان

1 دانشجو/دانشگاه شهید چمران اهواز

2 دانشیار/دانشگاه شهید چمران اهواز

3 استادیار/دانشگاه شهید چمران اهواز

4 کارشناس ارشد/دانشگاه شهید چمران اهواز

چکیده

سابقه و هدف: آلودگی آب‌های زیرزمینی یک فرآیند پیچیده و پر از عدم قطعیت، در مقیاس منطقه‌ای می‌باشد. توسعه یک روش یکپارچه جهت ارزیابی آسیب‌پذیری آبخوان‌ها، می‌تواند به منظور مدیریت بهینه و حفاظت از آن‌ها کارامد باشد. دشت رامهرمز به دلیل داشتن خاک حاصلخیز و منابع آب کافی دارای زمین‌های مستعد کشاورزی است که به دلیل توسعه کشاورزی، استفاده از کودهای شیمیایی و مواد آفت‌کش‌ همواره در معرض خطر آلودگی قرار دارد. یکی از راه‌های مناسب برای جلوگیری از آلودگی آب‌های زیرزمینی، شناسایی مناطق دارای پتانسیل آلودگی می‌باشد. هدف از مطالعه حاضر، تهیه نقشه آسیب‌پذیری آبخوان آبرفتی دشت رامهرمز با استفاده از مدل دراستیک و سپس بکارگیری روش‌های هوش مصنوعی جهت بهبود نتایج حاصل از مدل دراستیک است. با توجه به اهمیت منابع آب زیرزمینی در منطقۀ مورد مطالعه که برای مقاصد مختلف از جمله کشاورزی مورد استفاده قرار می‌گیرد، مطالعۀ آسیب پذیری آبخوان و حفاظت این مناطق برای توسعه و مدیریت بهینه منابع آب ضروری به نظر می‌رسد.
مواد و روش‌ها: در این مطالعه، ارزیابی آسیب‌پذیری آبخوان آبرفتی دشت رامهرمز در ابتدا با استفاده از مدل دراستیک انجام شد و در ادامه از روش‌های هوش مصنوعی جهت بهینه‌سازی مدل استفاده گردید. مدل دراستیک شامل پارامترهای: عمق تا سطح ایستابی، تغذیه، جنس سفره، نوع خاک، توپوگرافی، مواد تشکیل دهنده منطقۀ غیراشباع و هدایت هیدرولیکی می‌باشد که در ارزیابی آسیب‌پذیری سفرۀ آب‌زیرزمینی موثر هستند. این روش بر اساس وزن‌های استاندارد پارامترهای مدل دراستیک و لایه‌های بدست آمده برای هر یک از هفت پارامتر میزان آسیب‌پذیری آبخوان را محاسبه می‌نماید. پس از آماده-سازی لایه‌ها، آسیب‌پذیری آبخوان آبرفتی دشت رامهرمز با استفاده از روش‌ دراستیک، تعیین گردید. هم‌چنین نقشۀ آسیب‌پذیری آبخوان و شاخص دراستیک برای کل منطقه محاسبه شد. به منظور ارزیابی دقت نتایج این مدل، از داده‌های غلظت نیترات موجود در آبخوان جهت صحت‌سنجی استفاده شده است. در ادامه به منظور بهبود نتایج، مدل دراستیک با روش‌های شبکه عصبی مصنوعی، منطق فازی( سوگنو و ممدانی) و سیستم استنتاج تطبیقی عصبی- فازی تلفیق شد و چهار نقشه آسیب‌پذیری با استفاده از مدل‌های مختلف هوش مصنوعی حاصل گردید.
یافته‌ها: نقشۀ آسیب‌پذیری آبخوان نسبت به آلودگی، با تقسیم‌بندی به سه محدودۀ آسیب‌پذیری کم، متوسط و زیاد تهیه و شاخص دراستیک برای کل منطقه بین 48 تا 156 محاسبه گردید. ضریب همبستگی 97/0 بین شاخص دراستیک و غلظت نیترات نشان دهنده دقت نسبتاً مناسب این روش است. نتایج نشان داد که مدل‌های هوش مصنوعی به کار گرفته شده، قابلیت بهبود نتایج مدل دراستیک اولیه را دارا می‌باشند. با مقایسه نتایج مدل‌ها می‌توان نتیجه گرفت که مدل سیستم استنتاج تطبیقی عصبی- فازی بهترین نتیجه را در بردارد.
نتیجه‌گیری: ضریب تعیین (R2) برای مدل‌های سیستم استنتاج تطبیقی عصبی- فازی، شبکه عصبی و مدل‌های فازی سوگنو و ممدانی به‌ترتیب 99/0، 94/0، 98/0 و 87/0 بدست آمد. طبق مدل نهایی، نواحی جنوب- جنوب شرقی منطقه دارای بیشترین میزان پتانسیل آلودگی هستند.

کلیدواژه‌ها

عنوان مقاله [English]

Improve the results of the DRASTIC model using artificial intelligence methods to assess groundwater vulnerability in Ramhormoz alluvial aquifer plain

نویسندگان [English]

  • nazanin ghanbari 1
  • Kazem Rangzan 2
  • Mostafa Kabolizade 3
  • Poria Moradi 4

چکیده [English]

Background and objectives: Groundwater pollution is a complex and full of uncertainty process, on a regional scale. Development of an integrated method for assessing groundwater vulnerability, can be efficient in order to optimized management and protection of them. Because of fertile soil and sufficient water resources, Ramhormoz plain is suitable area for agriculture that by development of agriculture, use of chemical fertilizers and pesticide, this plain always is at risk of contamination. One of the suitable approach to prevent groundwater contamination, identify areas of potential contamination. The aim of this study is to produce vulnerability map of Ramhormoz plain alluvial aquifer using DRASTIC model, and then use artificial intelligence techniques to improve the results of the DRASTIC model. Due to the importance of groundwater resources in the study area that are used for various purposes including agriculture, Aquifer vulnerability study and protect these areas for development and management of water resources is essential.
Materials and methods: In this study, first, vulnerability evaluation of Ramhormoz alluvial aquifer plain was performed using DRASTIC model and in the following, artificial intelligence methods was used to optimize the model. DRASTIC model includes the following parameters: depth to water table, net recharge, aquifer media, soil media, topography, impact of vadose zone and hydraulic conductivity that are effective in groundwater vulnerability assessment. This method, based on the standard weights of DRASTIC model and obtained layers for each of the seven parameters, calculates the amount of aquifer vulnerability. After preparation of the layers, vulnerability of Ramhormoz alluvial aquifer plain was determined using drastic model. Also the groundwater vulnerability map and DRASTIC index was calculated for the entire area. In order to evaluation of accuracy of the obtained results from the model, nitrate concentration data existing in groundwater have been used for verification. Following In order to improve results, DRASTIC model was integrated by artificial neural networks, fuzzy logic (Sugeno and Mamdani) and Adaptive Neuro-Fuzzy Inference System methods and four vulnerability maps was obtained using different models of artificial intelligence.
Results: the groundwater vulnerability map toward the contamination was prepared by the division into three vulnerability ranges including low, medium and high and DRASTIC index was calculated for the entire area between 48 and 156. Correlation coefficient 0.97 between DRASTIC index and nitrate concentration reflects the relatively good accuracy of this method. Also, the results of the implementation of these models showed that the used artificial intelligence models have the ability to improve the primary DRASTIC model results. By comparing the results of the models can be concluded that the Adaptive Neuro-Fuzzy Inference System model has the best result.
Conclusion: The determination coefficient, R2, for the Adaptive Neuro-Fuzzy Inference System, neural networks and Mamdani fuzzy and Sugeno fuzzy models, is 0.99, 0.94, 0.98 and 0.87 respectively. According to the final model, South- Southeast regions have the highest potential for contamination.

کلیدواژه‌ها [English]

  • groundwater vulnerability
  • DRASTIC model
  • geographic information systems
  • Artificial intelligence
1.Ahmadi, J., Akhondi, L., Abbasi, H., Khashei-Siuki, A., and Alimadadi, M. 2013.
Determination of aquifer vulnerability using DRASTIC model and a single parameter
sensitivity analysis and acts and omissions (Case study: Salafchegan-Neyzar Plain). J. Water
Soil Cons. 20: 3. 1-25. (In Persian)
2.Ahmadzadeh Gharah Gwiz, K., Mirlatifi, S., and Mohammadi, K. 2010. Comparison of
Artificial Intelligence Systems (ANN & ANFIS) for Reference Evapotranspiration
Estimation in the Extreme Arid Regions of Iran. J. Water Soil. 24: 4. 679-689. (In Persian)
3.Aller, L., Bennet, T., Lehr, J.H., Petty, R.J., and Hacket, G. 1987. DRASTIC: a standardized
system for evaluating groundwater pollution using hydrological settings. Ada, OK, USA:
Prepared by the National Water Well Association for the US EPA Office of Research and
Development.
4.Antonakos, A.K., and Lambrakis, N.J. 2007. Development and testing of three hybrid
methods for the assessment of aquifer vulnerability to nitrates, based on the drastic model, an
example from NE Korinthia, Greece. J. Hydrol. 333: 288-304.
5.ASCE Task Committee on Application of Artificial Neural Networks in Hydrology. 2000.
Artificial neural network in hydrology, part I and II. J. Hydrol. Engin. 5: 115-137.
6.Asghari Moghaddam, A., Fijani, A., and Nadiri, A. 2015. Optimization of DRASTIC model
by Artificial Intelligence for Groundwater Vulnerability Assessment in Maraghe-Bonab
Plain. Engineering and Environmental Geology. 24: 94. 169-176. (In Persian)
7.Asghari Moghaddam, A., Nadiri, A., and Fijani, A. 2010. Spatial Prediction of Fluoride
Concentration Using Artificial Neural Networks and Geostatic Models. Water and soil
science. 19: 2. 129-145. (In Persian)
8.Aslani, M., Alesheikh, A.A., and Shad, R. 2011. Landslide Susceptibility Mapping, Using
Fuzzy Inference System and GIS (Case study: Sections of Mazandaran Province). Iran. J.
Rem. Sens. GIS. 2: 2. 35-54. (In Persian)
9.Baghapour, M., Nasser, T., Sayed Hamidreza, T., and Amir, F. 2014. Assessment of
groundwater nitrate pollution and determination of groundwater protection zones using
DRASTIC and composite DRASTIC (CD) models: the case of Shiraz unconfined aquifer.
J. Health Sci. Surv. Sys. 2: 2. 54-65.
10.Demuth, H., Beale, M., and Hagan, M. 2010. Neural Network Toolbox™ 6 User's guide.
11.Dixon, B. 2005a. Applicability of neuro-fuzzy techniques in predicting groundwater
vulnerability: a GIS-based sensitivity analysis. J. Hydrol. 309: 1-4. 17-38.
12.Dixon, B. 2005b. Groundwater vulnerability mapping: a GIS and fuzzy rule based integrated
tool. J. Appl. Geograph. 25: 327-347.
13.Fijani, E., Nadiri, A.A., Asghari Moghaddam, A., Tsai, F., and Dixon, B. 2013. Optimization
of DRASTIC method by supervised committee machine artificial intelligence to assess
groundwater vulnerability for Maragheh–Bonab plain aquifer, Iran. J. Hydrol. 503: 89-100.
14.Guo, O., Wnag, Y., Gao, X., and Ma, T. 2007. A new model (DRARCH) for assessing
groundwater vulnerability to arsenic contamination at basin scale: a case study in Taiyuan
basin, northern China. Environmental Geology. 52: 5. 923-32.
15.Hooshangi, N., and Alesheikh, A.A. 2015. Evaluation of ANN, ANFIS and fuzzy systems in
estimation of solar radiation in Iran. J. Geomat. Sci. Technol. 4: 3. 187-200. (In Persian)
16.Hopfield, J.J. 1982. Neural network and physical systems with emergent collective
computational abilities. Proceeding of National Academy of scientists. 79: 2554-2558.
17.Karami shahmaleki, N., Behbahani, S.M., Masahbavani, A., and Khodai, K. 2010.
Optimization of DRASTIC model by statistical nonparametric methods. Iran. J. Geol.
4: 14. 73-82. (In Persian)
18.Khazaii, A., Al Sheikh, A., Karimi, M., and Hassan Vahidnia, M. 2012. Prediction and
modeling of carbon monoxide concentration with the combination of an adaptive
neuro-fuzzy network and GIS. J. Appl. RS & GIS Techniq. Natur. Resour. Sci. 3: 3. 21-35.
(In Persian)
19.Kia, M. 2011. Neural networks in MATLAB. Kian Rayan Sabz Press. (In Persian)
20.Lee, K.H. 2004. First Course on Fuzzy, Theory and Applications. Springer, Berlin, 335p.
21.Legrand, H.E. 1964. System for evaluating the contamination potential of some waste sites.
J. AWWA. 56: 959-974.
22.Li-Xin Wang, A. 1997. Course in Fuzzy Systems and Control, Prentice Hall PTR.
Pp: 192-205.
23.Merchant, J. 1994. GIS-based groundwater pollution hazard assessment: a critical review of
the DRASTIC model. Photogrammetric Engineering and Remote Sensing. 60: 9. 1117-1127.
24.Moghaddamnia, A., Ghafari Gousheh, M., Piri, J., Amin, S., and Dawei, H. 2009.
Evaporation estimation using artificial neural networks and adaptive neuro-fuzzy inference
system techniques. Advances in Water Resources. 32: 88-97.
25.Motkan, A.A., Naseri, H.R., and Ostad Hashemi, Z. 2008. Correction of DRASTIC method
based on GIS using statistical methods and Analytical Hierarchy Process: A Case Study of
Hamadan plain. Iran. J. Appl. Geol. 4: 3. 205-222. (In Persian)
26.Nakhaii, M., Amiri, V., and Rahimi Shahrebabaki, M. 2013. Evaluation of the contamination
potential and sensitivity analysis using DRASTIC model based on GIS. J. Adv. Appl. Geol.
3: 8. 1-10. (In Persian)
27.Nazifkar, M., and Asghari, K. 2011. Adaptive Neuro-fuzzy inference system using fuzzy
clustering in runoff predicting. The Sixth National Congress of Civil Engineering, Semnan
University, Iran. (In Persian)
28.Neshat, A., Biswajeet, P., and Mohsen, D. 2014. Groundwater vulnerability assessment using
an improved DRASTIC method in GIS. J. Resour. Cons. Recycl. 86: 74-86.
29.Newton, S.C., Pemmaraju, S., and Mitra, S. 1992. Adaptive fuzzy leader clustering of complex
data sets in pattern recognition. IEEE Transactions on Neural Networks. 5: 794-800.
30.Panagopoulos, G., Antonakos, A., and Lambrakis, N. 2006. Optimization of DRASTIC
model for groundwater vulnerability assessment, by the use of simple statistical methods and
GIS. Hydrogeol. J. 14: 894-911.
31.Report of Knowledge studies of available water resources in the Ramhormoz study area.
2006. Ministry of Power, Khuzestan Water and Electricity Company. April, Khuzestan.
(In Persian)
32.Report of weather, climate and water resources in Khuzestan province. 2009. General
Directorate of economic studies and surveys. Winter, Khuzestan. (In Persian)
33.Sabziparvar, A.A., and Bayat Varkeshi, M. 2011. Evaluation of artificial neural network
(ANN) and adaptive neuro-fuzzy inference system (ANFIS) methods in prediction of global
solar radiation. 10: 4. 347-357. (In Persian)
34.Sabziparvar, A.A., Zare Abyaneh, H., and Bayat Varkeshi, M. 2010. A model Comparison
between Predicted Soil Temperatures Using ANFIS Model and Regression Methods in Three
Different Climates. J. Water Soil. 24: 2. 274-285. (In Persian)
35.Sajadi, Z., Kalantari, N., Makvandi, M., Keshavarzi, M., Ghafari, H., Ahmadnejad, Z., and
Booslik, Z. 2011. Study of the aquifer vulnerability in Assaluee using DRASTIC model.
First national conference on water and wastewater science and technology. Islamic Azad
University of Ahwaz, April. (In Persian)
36.Samey, A.A., and Gang, C. 2008. A GIS Based DRASTIC Model for the Assessment of
Groundwater vulnerability to pollution in West Mitidja: Blida city, Algeria. Res. J. Appl.
Sci. 3: 7. 500-507.
37.Sener, E., and Sehnaz, S. 2015. Evaluation of groundwater vulnerability to pollution using
fuzzy analytic hierarchy process method. J. Environ. Earth Sci. 73: 12. 8405-8424.
38.Soper, R.C. 2006. Groundwater vulnerability to agrochemicals: A GIS-based DRASTIC
model analysis of Carrol, Chariton, and Saline Counties, Missouri, USA. M.Sc. Thesis,
University of Missouri-Columbia.
39.Sugeno, M., and Yasukawa, T. 1993. A Fuzzy-Logic-based Approach to Qualitative
Modeling. IEEE Transactions on Fuzzy Systems. 1: 1. 7-31.
40.Vrba, J., and Zaporozec, A. 1994. Guidebook on Mapping Groundwater Vulnerability.
International Association of Hydrogeologists–International Contributions to Hydrogeology 16
Water and Environ. J. 26: 3. 381-391.
41.Yarmohamadi, A., Chitsazan, M., and Rangzan, K. 2006. Calculation of the amount of
DRASTIC model Parameters impact on Aghili plain aquifer vulnerability. Twenty-fifth
Conference on Earth Sciences, Geological survey of Iran, Tehran. (In Persian)
42.Zounemat-Kermani, M., and Teshnehlab, M. 2008. Using adaptive neuro-fuzzy inference
system for hydrological time series prediction. Applied Soft Computing. 8: 2. 928-936.