Tận dụng sản phẩm khí quyển GNSS cho dự đoán lún đất dựa trên học máy

Springer Science and Business Media LLC - Tập 16 Số 4 - Trang 3039-3056 - 2023
Melika Tasan1, Zahrasadat Ghorbaninasab2, Saeid Haji-Aghajany3, Alireza Ghiasvand2
1Department of Civil Engineering, Faculty of Environmental Engineering and Geodesy, Wrocław University of Environmental and Life Sciences, Wroclaw, Poland
2Department of Surveying Engineering, South Tehran Branch, Islamic Azad University, Teheran, Iran
3Institute of Geodesy and Geoinformatics, Wrocław University of Environmental and Life Sciences, Wrocław, Poland

Tóm tắt

Lún đất là một hiện tượng nguy hiểm đòi hỏi phải có dự đoán chính xác để giảm thiểu thiệt hại và ngăn chặn thương vong. Nghiên cứu này khám phá việc sử dụng phương pháp Tăng giảm Dài-ngắn hạn (LSTM) để dự đoán chuỗi thời gian lún đất, xem xét các yếu tố ảnh hưởng như mức nước ngầm, loại đất và độ dốc, đặc tính tầng chứa nước, độ bao phủ thực vật, cách sử dụng đất, độ sâu đến mực nước ngầm, khoảng cách đến các giếng khai thác, khoảng cách từ sông, khoảng cách từ các đứt gãy, nhiệt độ, và các sản phẩm khí quyển ẩm. Do sự biến đổi không gian lớn của các tham số khí quyển ẩm, việc sử dụng các mô hình thời tiết số để trích xuất là không thực tiễn, đặc biệt là ở những khu vực có mạng lưới trạm synoptic thưa thớt. Điều này cản trở việc thu được kết quả dự đoán chính xác vì các sản phẩm khí quyển ẩm đóng vai trò quan trọng trong việc dự đoán lún đất và không thể bị bỏ qua trong quá trình dự đoán lún đất. Trong nghiên cứu này, các sản phẩm khí quyển GNSS, bao gồm Hơi nước tích hợp (IWV) và Bốc hơi - Truyền (ET), được sử dụng như một giải pháp thay thế. Hai kịch bản đã được xem xét: một kịch bản kết hợp sản phẩm GNSS với các tham số khác, và kịch bản còn lại chỉ dựa vào các tham số còn lại khi không có sản phẩm khí quyển GNSS. Dữ liệu thực địa từ các phép đo dịch chuyển Radar tổng hợp can thiệp (InSAR) đã được sử dụng để đánh giá và thử nghiệm. Kết quả cho thấy rằng việc bao gồm các sản phẩm khí quyển GNSS đã làm tăng đáng kể độ chính xác dự đoán, với giá trị Sai số bình phương gốc (RMSE) là 3.07 cm/năm trong kịch bản đầu tiên. Trong kịch bản thứ hai, việc thiếu thông tin khí quyển ẩm dẫn đến kết quả dự đoán kém hơn, làm nổi bật vai trò quan trọng của dữ liệu khí quyển ẩm trong phân phối không gian. Tuy nhiên, bằng cách sử dụng các sản phẩm khí quyển thu được từ quan sát GNSS, các dự đoán về sự thay đổi dịch chuyển đạt được độ chính xác hợp lý. Nghiên cứu này nhấn mạnh tầm quan trọng của các chỉ số khí quyển và thể hiện tiềm năng của phương pháp LSTM kết hợp với các quan sát GNSS cho việc dự đoán lún đất hiệu quả, giúp cải thiện các biện pháp phòng ngừa và chiến lược giảm thiểu ở những khu vực thiếu phủ sóng dữ liệu synoptic.

Từ khóa

#Lún đất #phương pháp LSTM #sản phẩm khí quyển GNSS #dự đoán dịch chuyển #ảnh hưởng của môi trường.

Tài liệu tham khảo

Agnihotri G, Chaurasia V, Kumar S (2021) A comparative study of different machine learning approaches for Precise Tropospheric Parameter Estimation using GNSS Data. IEEE J Sel Top Appl Earth Observations Remote Sens 14:3732–3744

Akbari V, Motagh M (2012) Improved ground subsidence monitoring using small baseline SAR interferograms and a weighted least squares inversion algorithm. IEEE Geosci Remote Sens Lett 9(3):437–444

Allen RG, Pereira LS, Raes D, Smith M (1998) Crop evapotranspiration. FAO Irrigation and Drainage, Pa-per No. 56. Food and Agriculture Organization of the United Nations, Rome

Anderssohn J, Wetzel HL, Walter TR, Motagh M, Djamour Y, Kaufmann H (2008) Land subsidence pattern controlled by old alpine basement faults in the Kashmar Valley, northeast Iran: results from InSAR and levelling. Geophys J Int 74:287–294. https://doi.org/10.1111/j.1365-246X.2008.03805.x

Andreas H, Abidin Z, Gumilar H, Sidiq I., P., Sarsito T., A., D., Pradipta D (2018) Insight into the correlation between Land Subsidence and the Floods in regions of Indonesia. IntechOpen. https://doi.org/10.5772/intechopen.80263

Azarakhsh Z, Azadbakht M, Matkan A (2022) Estimation, modeling, and prediction of land subsidence using Sentinel-1 time series in Tehran-Shahriar plain: a machine learning-based investigation. Remote Sens Appl Soc Environ 25:100691 [Google Scholar] [CrossRef]

Berardino P, Fornaro G, Lanari R, Sansosti E (2002) Small baseline subset (SBAS) interferometry: a Novel Method for Monitoring Elevation Changes Applied to volcanoes using ERS synthetic aperture Radar Data. IEEE Trans Geosci Remote Sens 40(11):2375–2383

Bevis M, Businger S, Herring TA, Rocken C, Anthes RA, Ware RH (1992) GNSS meteorology: Remote sensing of atmospheric water vapor using the global positioning system. J Geophys Research: Atmos 97(D14):15787–15801

Bock O, Bouin MN, Walpersdorf A, Doerflinger E (2003) Comparison of GNSS precipitable water vapor to Independent observations and numerical weather prediction models. J Geophys Research: Atmos 108:D12

Boehm J, Heinkelmann R, Schuh H (2006) Short note: a global model of pressure and temperature for geodetic applications. J Geodesy 80(12):771–781

Chen H, Luo Z, Zhou Z, Qi S (2019) Investigation on soil moisture and its influence on land subsidence based on remote sensing and field observations in Beijing plain, China. Environ Earth Sci 78(19):575. https://doi.org/10.1007/s12665-019-8552-y

Dach R, Lutz S, Walser P, Fridez P (2015) Bernese GNSS Software Version 5.2. User manual. Astronomical Institute, University of Bern, Bern Open Publishing. https://doi.org/10.7892/boris.72297. ISBN: 978-3-906813-05-9

Dach R, Arnold D, Grahsl A, Ge M (2020) Global tropospheric maps based on Global Navigation Satellite Systems: a review. Remote Sens 12(4):660

Davis JL, Yu Z, Wdowinski S, Zhang P, Lee H (2013) Surface deformations caused by the 2010 Maule Earthquake, Chile: Ground-based and GNSS measurements. J Geophys Research: Solid Earth 118(2):823–8303

Davoodijam M, Motagh M, Momeni M (2015) Land subsidence in Mahyar Plain, Central Iran, investigated using Envisat SAR Data. In: Proceedings the 1st international workshop on the quality of geodetic observation and monitoring systems (QuGOMS’11). Springer, pp 127–130. https://doi.org/10.1007/978-3-319-10828

Dehghani M, Valadan Zoej MJ, Entezam I, Mansourian A, Saatchi S (2009) InSAR monitoring of Progressive land subsidence in Neyshabour, northeast Iran. Geophys J Int 178:47–56 [CrossRef]

Ding P, Jia C, Di S, Wang L, Bian C, Yang X (2020) Analysis and prediction of land subsidence along signifcant linear engineering. Bull Eng Geol Environ 79(10):5125–5139

Elman JL (1990) Recurrent neural networks. Cogn Sci 14(2):179–211

Emardson TR, Simons M, Webb HH (2003) Neutral atmospheric delay in interferometric synthetic aperture radar applications: statistical description and mitigation. J Geophys Res 108(B5):2231. https://doi.org/10.1029/2002JB001781

European Space Agency (ESA). (n.d.). Sentinel-2. Retrieved from https://sentinel.esa.int/web/sentinel/missions/sentinel-2

Farshbaf A, Mousavi MN, Shahnazi S (2023) Vulnerability assessment of power transmission towers affected by land subsidence via interferometric synthetic aperture radar technique and finite element method analysis: a case study of Zanjan and Qazvin provinces. https://doi.org/10.1007/s10668-023-03127-x. Environ Dev Sustain

Faunt CC, Sneed M, Traum J, Brandt JT (2016) Impact of Climate Variability on Land Subsidence in the San Joaquin Valley, California, USA. Hydrogeol J 24(3):675–686

Ferretti A, Prati C, Rocca F (2007) InSAR Time Series Analysis: methods and applications for Earth Surface Deformation Monitoring. Rep Prog Phys 70(7):1–74

Gers FA, Schmidhuber J, Cummins F (2000) Learning to forget: continual prediction with LSTM. Neural Comput 12(10):2451–2471

Goodfellow I, Bengio Y, Courville A (2016) Deep learning. MIT Press

Graves A (2013) Generating sequences with recurrent neural networks. arXiv preprint arXiv:1308.0850

Gray AL, Mattar KE, Sofko G (2000) Influence of ionospheric electron density fluctuations on satellite radar interferometry. Geophys Res Lett 27:1451–1454

Greff K, Srivastava RK, Schmidhuber J (2017) LSTM: a search space odyssey. IEEE Trans Neural Networks Learn Syst 28(10):2222–2232

Haji-Aghajany S (2021) Function-Based Troposphere Water Vapor Tomography Using GNSS Observations. PhD Thesis, Faculty of Geodesy and Geomatics Engineering, K. N. Toosi University of Technology

Haji-Aghajany S, Amerian Y (2018) An investigation of three dimensional ray tracing method efficiency in precise point positioning by tropospheric delay correction. J Earth Space Phys 44:39–52. https://doi.org/10.22059/JESPHYS.2018.236885.1006913

Haji-Aghajany S, Amerian Y (2020) Atmospheric phase screen estimation for land subsidence evaluation by InSAR time series analysis in Kurdistan, Iran. J Atmos Solar Terr Phys 209:105314. https://doi.org/10.1016/j.jastp.2020.105314

Haji-Aghajany S, Amerian Y (2020a) Assessment of InSAR tropospheric signal correction methods. J Appl Remote Sens 14(4):044503. https://doi.org/10.1117/1.JRS.14.044503

Haji-Aghajany S, Voosoghi B, Yazdian A (2017) Estimation of North Tabriz Fault parameters using neural networks and 3D tropospherically corrected Surface Displacement Field. Geomatics Nat Hazards Risk 8(2):918–932. https://doi.org/10.1080/19475705.2017.1289248

Haji-Aghajany S, Voosoghi B, Amerian Y (2019) Estimating the slip rate on the north Tabriz fault (Iran) from InSAR measurements with tropospheric correction using 3D ray tracing technique. Adv Space Res 64(11):2199–2208. https://doi.org/10.1016/j.asr.2019.08.021

Haji-Aghajany S, Pirooznia M, Raoofian Naeeni M, Amerian Y (2020) Combination of Artificial neural network and genetic algorithm to Inverse Source parameters of Sefid-Sang Earthquake using InSAR technique and Analytical Model Conjunction. J Earth Space Phys 45(4):121–131. https://doi.org/10.22059/JESPHYS.2019.269596.1007065

Haji-Aghajany S, Amerian Y, Amiri-Simkooei A (2022) Function-based Troposphere tomography technique for optimal downscaling of precipitation. Remote Sens 14(11):2548. https://doi.org/10.3390/rs14112548

Haji-Aghajany S, Amerian Y, Amiri-Simkooei A (2023) Impact of Climate Change parameters on Groundwater Level: implications for two subsidence regions in Iran using Geodetic observations and Artificial neural networks (ANN). Remote Sens 15:1555. https://doi.org/10.3390/rs15061555

Hanssen RF (2001) Radar interferometry: data interpretation and error analysis. Springer Science & Business Media

Hersbach H, Dee D ERA5 reanalysis is in production; ECMWF Newsl 147; ECMWF: Reading, UK, 2016.

Hochreiter S, Schmidhuber J (1997) Long short-term memory. Neural Comput 9(8):1735–1780. https://doi.org/10.1162/neco.1997.9.8.1735

Hu J, Motagh M, Guo J, Haghighi MH, Li T, Qin F, Wu W (2022) Inferring subsidence characteristics in Wuhan (China) through multitemporal InSAR and hydrogeological analysis. Eng Geol 297:106530

Jolivet R, Agram PS, Lin NY, Simons M, Doin MP, Peltzer G, Li Z (2014) Improving InSAR geodesy using global atmospheric models. J Geophys Res Solid Earth 119:2324–2341.https://doi.org/10.1002/2013JB010588

Khalili MA, Voosoghi B, Guerriero L, Haji-Aghajany S, Calcaterra D, Di Martire D (2023) Mapping of mean deformation rates based on APS-corrected InSAR data using unsupervised clustering algorithms. Remote Sens 15(2):529. https://doi.org/10.3390/rs15020529

Khanlari G, Heidari M, Momeni AA, Ahmadi M, Beydokhti AT (2012) The effect of groundwater overexploitation on land subsidence and sinkhole occurrences, western Iran. Q J Eng GeolHydrogeol 45(4):447–456

Kingma DP, Ba J (2014) Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980.

Kleijer F (2004) Troposphere delay modelling and filtering for precise GNSS levelling, Ph.D. thesis, Mathematical Geodesy and Positioning, Delft University of Technology

Li GD, Yamaguchi D, Nagai M (2007) A GM(1,1)–Markov chain combined model with an application to predict the number of Chinese international airlines. Technological Forecast Social Change Technol Forecast Soc Change 74:1465–1481

Lipton ZC, Berkowitz J, Elkan C (2015) A critical review of recurrent neural networks for sequence learning. arXiv preprint arXiv:1506.00019.

Mahmoudpour M, Khamehchiyan M, Nikudel MR, Ghassemi MR (2016) Numerical simulation and prediction of regional land subsidence caused by groundwater exploitation in the southwest plain of Tehran. Iran Eng Geol 201:6–28. https://doi.org/10.1016/J.ENGGEO.2015.12.004

Mohseni N, Sepehr A, Hosseinzadeh SR, Golzarian MR, Shabani F (2017) Variations in spatial patterns of soil–vegetation properties over subsidence-related ground fissures at an arid ecotone in northeastern Iran. Environ Earth Sci 76(6):234. https://doi.org/10.1007/s12665-017-6559-z

Moriasi DN, Arnold JG, van Liew MW, Bingner RL, Harmel RD, Veith TL (2007) Model evaluation guidelines for sys-tematic quantification of accuracy in watershed simulations. Trans ASABE 50:885–900

Motagh M et al (2008) Land subsidence in Iran caused by widespread water reservoir overexploitation. J Geophys Res Lett L0338:14. https://doi.org/10.1029/2008G

Motagh M, Shamshiri R, Haghshenas Haghighi M, Wetzel H-U, Akbari B, Nahavandchi H, Roessner S, Arabi S (2017) Quantifying groundwater exploitation induced subsidence in the Rafsanjan plain, southeastern Iran, using InSAR time-series and in situ measurements. Eng Geol 1–18. https://doi.org/10.1016/j.enggeo.2017.01.011

Paparrizos J, Gravano L (2015) k-shape: Efficient and accurate clustering of time series. In Proceedings of the ACM SIGMOD International Conference on Management of Data, Malbourne, VIC, Australia, 31 May–4 June; pp. 1855–1870

Peltzer G, Crampé F, Hensley S, Rosen P (2001) Transient strain accumulation and fault interaction in the Eastern California shear zone. Geology 29(11):975–978

Qi Y, Li Q, Karimian H, Liu D (2019) A hybrid model for spatiotemporal forecasting of PM2.5 based on graph convolutional neural network and long short-term memory. Sci Total Environ 664:1–10. https://doi.org/10.1016/j.scitotenv.2019.01.333

Remy D, Bonvalot S, Briole P, Murakami M (2003) Accurate measurement of tropospheric effects in volcanic areas from SAR interferometry data: application to Sakurajima volcano (Japan). Earth Planet Sci Lett 213:299–310

Rosen PA, Hensley S, Joughin IR (2000) Synthetic aperture radar interferometry. Proc IEEE 88(3):333–382

Saastamoinen J (1972) Atmospheric correction for the Troposphere and stratosphere in radio ranging of satellites. The Use of Artificial Satellites for Geodesy 15:247–251

Sak H, Senior A, Beaufays F (2014) Long short-term memory recurrent neural network architectures for large scale acoustic modeling. In Fifteenth Annual Conference of the International Speech Communication Association

Sorkhabi OM, Kurdpour I, Sarteshnizi RE (2022) Land subsidence and groundwater storage investigation with multi sensor and extended Kalman filter. Groundw Sustain Dev 1(19):100859

Statistical Center of Iran (2018) Available online: http://www.amar.org.ir (accessed on 11

Thornthwaite CW (1948) An approach toward a rational classification of climate. Geogr Rev 38:55–94 [CrossRef]

Tregoning P, Ramillien G (2014) Water in the balance. Nat Geosci 7(9):613–614

Ty TV, Minh HVT, Avtar R, Kumar P, Hiep HV, Kurasaki M (2021) Spatiotemporal variations in groundwater levels and the impact on land subsidence in CanTho, Vietnam. Groundw Sustain Dev 15:100680. https://doi.org/10.1016/j.gsd.2021.100680

Wang H, Jia C, Ding P et al (2023) Analysis and prediction of Regional Land Subsidence with InSAR Technology and Machine Learning Algorithm. KSCE J Civ Eng 27:782–793. https://doi.org/10.1007/s12205-022-1067-4

Webster R, Oliver MA (2007) Geostatistics for environmental scientists. John Wiley & Sons

Ye S, Zhou G, Peng Y, Liu X, Sun C (2020) Impacts of soil moisture and temperature variations on land subsidence in Shanghai, China. J Coastal Res 95(sp1):57–63. https://doi.org/10.2112/SI95-011.1

Zebker HA, Villasenor J (1992) Decorrelation in interferometric radar echoes. IEEE Trans Geosci Remote Sens 30(5):950–959

Zhang K, Lu C, Chen W, Hu C, Zhang J, Jiao W (2016) Zenith wet delay estimation using ground-based GNSS and surface pressure observations: a comparative study. J Geodesy 90(10):951–964

Zhao Q, Sun T, Zhang T, He L, Zhang Z, Shen Z, Xiong S (2021) High-Precision potential evapotranspiration model using GNSS Observation. Remote Sens 13:4848. https://doi.org/10.3390/rs13234848

Zhou D, Zuo X, Zhao Z (2022) Constructing a large-scale Urban Land Subsidence Prediction Method based on neural Network Algorithm from the perspective of multiple factors. Remote Sens 14:1803. https://doi.org/10.3390/rs14081803