Identification of Land Subsidence Susceptibility Areas in the Salmas Plain Using the GARDLIF Framework and Machine Learning

Document Type : Research paper

Authors

1 M.Sc. Graduate, Department of Geology, Faculty of Science, Urmia University, Urmia, Iran.

2 2- Assistant Professor, Department of Geology, Faculty of Sciences, Urmia University, Urmia, Iran.

3 Assistant Professor, Department of Geology, Faculty of Sciences, Urmia University, Urmia, Iran.

Abstract

Land subsidence, a geo-environmental hazard, is mainly associated with long-term decline in groundwater level and excessive groundwater extraction. The Salmas plain (NW Iran) experienced considerable subsidence (~12cm/yr). This study aimed to determine the areas susceptible to land subsidence in the Salmas plain. The GARDLIF framework was used to assess the spatial susceptibility to land subsidence based on seven hydrogeological and environmental parameters: groundwater-level decline (G), aquifer media (A), groundwater recharge (R), aquifer discharge or pumping (D), land use (L), aquifer thickness (I), and distance from faults (F). The thematic layers were ranked, weighted, and integrated using ArcGIS to evaluate the GARDLIF susceptibility index. In addition, the Grasshopper Optimization Algorithm along with an Artificial Neural Network (GOA-ANN) was applied to optimize the susceptibility assessment. According to the GARDLIF model, the Salmas Plain was classified into moderate, high, and very- high areas, accounting for 14.1%, 58.1 %, and 27.8% of the study area, respectively. Based on the GOA-ANN model, the Salmas Plain was classified as 8.2%, 45.3%, and 46.5% of the area into the moderate, high, and very-high susceptibility classes, respectively. A strong correlation (r= 95%) was observed between the two approaches. Based on sensitivity analysis, groundwater-level decline (G) and aquifer discharge (D) were the most influential parameters, with mean sensitivity values of 0.27 and 0.15, respectively. Whereas distance from faults (F) had the lowest influence (0.0025). The highest susceptibility areas occur mainly in western and southern parts of the plain, including areas around Qareh Qeshlaq, Tazeh Shahr, and Salmas. Field evidence, including sinkholes and surface cracks, is consistent with the predicted susceptibility patterns. Groundwater-level decline and excessive groundwater pumping are the principal factors controlling land-subsidence susceptibility in the Salmas Plain. The close agreement between the GARDLIF and GOA–ANN models, together with field evidence, supports the reliability of the susceptibility assessment. 

Keywords

Main Subjects

References

اسدی، م، گنجائیان، ح، جاودانی، م، و قادری حسب، م. 1400. ارزیابی ارتباط بین عوامل طبیعی و میزان فرونشست در دشت ایوانکی با استفاده از تصاویر رادار. هیدروژئولوژی، 6(1): 13-22.
جانباز فوتمی م، خلقی م، عبده­کلاهچی ع، روستایی م.آ. 1399. بررسی فرونشست زمین ناشی از تغییرات سطح ایستابی آب زیرزمینی با استفاده از روش تداخل سنجی راداری: مطالعه موردی استان قزوین. مجله تحقیقات منابع آب ایران، 16(3): 133-147.
صالحی متعهد ف، حافظی مقدس ن، لشکری­پور غ.ر.، دهقانی م. 1397. ارزیابی فرونشست زمین به‌کمک تلفیق روش تداخل‌سنجی راداری و اندازه‌گیری‌های میدانی و بررسی دلایل و اثرات آن بر شهر مشهد. زمین­شناسی مهندسی دانشگاه خوارزمی، 13(3): 435-462.
علیزاده، ا. 1374. اصول هیدرولوژی کاربردی. نهمین دوره چاپ، صفحه: 35.
فیجانی، ا. و زارعی، پ. 1403. ارزیابی پتانسیل فرونشست زمین در آبخوان هومند -آبسرد با استفاده از روشهای سنجش از دور و ارتباط آن با برداشت آبهای زیرزمینی. هیدروژئولوژی، 9 (1): 1- 15.
گزارش آب منطقه ای آذربایجان غربی. 1390. مطالعات بهنگام سازی بیلان منابع آب محدوده های مطالعاتی حوضه آبریز ارومیه.
نادری ک، ندیری ع.، اصغری مقدم ا. و کرد م. 1397. روشی جدید برای شناسایی و تعیین مناطق در معرض خطر فرونشست (مطالعة موردی: آبخوان دشت سلماس). مجله اکو هیدرولوژی، 5(1): 85-97.
هلالی، ل، باقری، ر، و مومنی، ع ا. 1403. بررسی فرونشست زمین با استفاده از روش تداخل سنجی راداری (مطالعه موردی: دشت سمنان). هیدروژئولوژی، 9(1: 86- 100.
Ambikesh G, Rao SS, Chandrasekaran K. 2024. A grasshopper optimization algorithm-based movie recommender system. Multimedia Tools and Applications. 83: 54189–54210.
Aalipour M, Malekmohammadi B, Ghorbani Z. 2022. Mapping land subsidence susceptibility due to groundwater decline using fuzzy pixel-based models. Arabian Journal of Geosciences. 15: 1014.
Arora S, Anand P. 2019. Chaotic grasshopper optimization algorithm for global optimization. Neural Computing and Applications. 31: 4385–4405.
Asadi M, Ganjaeian H, Javedani M, Ghaderi Hasab M. 2021. Evaluation of the relationship between natural factors and subsidence in Ivanaki Plain using radar imaging. Hydrogeology. 6(1): 13–22.
Babiker IS, Mohamed MA, Hiyama T, Kato K. 2005. A GIS-based DRASTIC model for assessing aquifer vulnerability in Kakamigahara Heights, Gifu Prefecture, central Japan. Science of the Total Environment. 345(1): 127–140.
Bagheri-Gavkosha M, Hosseini SM, Ataie-Ashtiani B, Sohani Y, Ebrahimian H, Morovat F, Ashrafi Sh. 2021. Land subsidence: A global challenge. Science of the Total Environment. 778: 146193.
Castellazzi P, Arroyo-Dominguez N, Martel R, Calderhead A, Normand J, Garfias J, Rivera A. 2016. Land subsidence in major cities of Central Mexico: Interpreting InSAR-derived land subsidence mapping with hydrogeological data. International Journal of Applied Earth Observation and Geoinformation. 47: 102–111.
Kuehn F, Albiol D, Cooksley G, Duro J, Granda J, Haas S, et al. 2010. Detection of land subsidence in Semarang, Indonesia, using stable points network (SPN) technique. Environmental Earth Sciences. 60(5): 909–921.
Lohani AK, Goel NK, Bhatia KKS. 2011. Comparative study of neural network, fuzzy logic and linear transfer function techniques in daily rainfall–runoff modelling under different input domains. Hydrological Processes. 25(2): 175–193.
Nourani V, Kisi Ö, Komasi M. 2011. Two hybrid artificial intelligence approaches for modeling rainfall–runoff process. Journal of Hydrology. 402(1–2): 41–59.
Poland JF, Davis GH. 1969. Land subsidence due to withdrawal of fluids. Engineering Geology. 2: 187–269.
Rosen L. 1994. A study of the DRASTIC methodology with emphasis on Swedish conditions. Ground Water. 32(2): 278.
Saremi S, Mirjalili S, Lewis A. 2017. Grasshopper optimisation algorithm: Theory and application. Advances in Engineering Software. 105: 30–47.
Scanlon B, Healy R, Cook P. 2002. Choosing appropriate techniques for quantifying groundwater recharge. Hydrogeology Journal. 10(1): 18–39.
Seifi A, Ehteram M, Singh VP, Mosavi A. 2020. Modeling and uncertainty analysis of groundwater level using six evolutionary optimization algorithms hybridized with ANFIS, SVM, and ANN. Sustainability. 12(10): 1–42.
Todd DK, Mays LW. 2005. Groundwater Hydrology. 3rd ed. John Wiley & Sons, Inc., Hoboken, New Jersey, 652 p.
Trinh MT, Fredlund DG. 2000. Modeling subsidence due to groundwater extraction in the Hanoi City area. Journal of Geology Technology. 37: 621–637.
Tsai FT-C, Li X. 2008. Multiple parameterizations for hydraulic conductivity identification. Groundwater. 46(6): 851–864.
Yonaba H, Anctil F, Fortin V. 2010. Comparing sigmoid transfer functions for neural network multistep ahead streamflow forecasting. Journal of Hydrologic Engineering. 15(4): 275–283.
Zadeh MR, Amin S, Khalili D, Singh VP. 2010. Daily outflow prediction by multilayer perceptron with logistic sigmoid and tangent sigmoid activation functions. Water Resources Management. 24(11): 2673–2688.
Hydrogeology
Volume 10, Issue 2
May 2026
Pages 95-114

Files

Share

How to cite

Statistics