چکیده:
رودشکنها از لندفرمهای پرشیب رودخانهای هستند که در تحول سیستمهای رودخانهای اهمیت دارند. این پژوهش با هدف شناسایی عوامل مؤثر بر ایجاد رودشکن و تعیین مناطق مستعد ایجاد رودشکن در حوضۀ قلعه شاهرخ با استفاده از روش رگرسیون لجستیک باینری انجام شده است. بدین منظور عوامل مؤثر بر ایجاد رودشکن انتخاب شدند و سپس ارتباط آنها با پراکنش رودشکنها بررسی شد؛ در ادامه متغیرهای تأثیرگذار و میزان تأثیر آنها بر رودشکن تعیین و مدل پیشبینی با رگرسیون لجستیک روی این متغیرها انجام شد. نتایج آزمون درستنمایی در مدل ارائهشده نشان میدهد زمینشناسی و فاصله از مرزهای سازندهای زمینشناسی در مدل معنادارند. تحلیل نتایج نشان میدهد با کاهش مقدار در شاخصهای بریدگی و سطوح همپایه، فاصله از مرزهای سازندهای زمینشناسی و با افزایش برش عمودی احتمال وجود رودشکن افزایش مییابد. دربارة سایر عوامل استفادهشده، رابطهای دیده نشد. دقت 87درصدی دادههای آزمون در نمودار راک بیانکنندة دقت زیاد مدل در تشخیص درست نقاط رودشکن در حوضة قلعه شاهرخ است. نتیجۀ شاخص یودن برای دادههای آزمون 66/0 است که ارائة اطلاعات درست از وضعیت احتمال نقاط رودشکن بهویژه برای دادههای آزمون مدل را نشان میدهد. نتیجة ضریب توافق کاپا برای دادههای آزمون 60/0 است که تطابق و توافق هر دو روش را با مقادیر مشاهداتی نقاط رودشکن نشان میدهد. براساس نتایج این پژوهش، سازندهای زمینشناسی و توپوگرافی در رخداد رودشکنها در منطقة مطالعهشده نقش مهمی دارند و رگرسیون لجستیک نیز، مدل مناسبی برای پیشبینی وقوع رودشکن است.
Extended Abstract: Introduction: Rivers react to subsidence at their baseline by cutting and digging topographic features. The development of an upstream incision is often accompanied by a steep fracture called a river break (Loget & Van Den Driessche, 2009). The presence of river breaks in a geographical landscape is an indication of a steady-state in river systems. Therefore, the presence of knickpoints shows the system instability. The study of knickpoints can be used in the field of studies related to the evolution of valleys, identification of tectonic active areas and rock outcrops, river surface changes, erosion and sedimentation, and geomorphological changes in river systems. The basin studied in this study is located in the Qaleh Shahrokh-Chelgard area in the northeastern part of Chaharmahal and Bakhtiari province, Iran. The reason for selecting this basin is the extensive activities of the Zagros fault along the northwest-southeast and the existence of a hydrographic network affected by the trend of faults and the potential for knickpoints. Methodology: In this study, the locations of knickpoints were detected from the Radiometrically Terrain-Corrected (RTC) model which is extracted from the active microwave sensor ALOS PALSAR with a spatial resolution of 12 meters (Logan et al., 2014) as input data to the MATLAB executive toolbox called Tec DEM. Tec DEM is an executable toolbox in MATLAB software and uses a Digital Elevation Model (DEM) as input for morphotectonics in the basin. Tec DEM tool can be used in a variety of fields in the analysis of surface anomalies, drainage network and surface dynamics of basins, production of base maps, incisions (local roughness), vertical dissection and drainage density of basins and sub-basins, determination of turning points or knickpoints, hypsometric analysis and slope and concavity index of canal profiles (Shahzad & Gloaguen, 2011). The determination of knickpoints according to the shape of the longitudinal profile of the river is done semi-automatically. In this study, these points in the study areas were investigated according to field observations. In this study, geological variables and geomorphic variables related to knickpoints were used to identify the knickpoints. Information layers including geology, distance from the fault, distance from the boundary of geological formations, surface roughness index, fractal dimension, base surfaces, local roughness, and the vertical dissection as predictor variables and the layers of knickpoints as the prediction variables were used for modeling. For geological and tectonic studies of the region, geological maps of 100,000 sheets of Chadegan and Fereydunshahr and 250,000 sheets of Shahrekord were used. A total of 8 raster layers were used to analyze and predict the possibility of the presence of a knickpoint in the study area. Since 8 layers have different units and are not suitable as direct input for logistic regression, the input parameters were normalized in the range of 0 to 1. Nominal layers, such as geological data, became sequential variables between 0 and 1. All of these layers were then re-sampled as a network format with a cell size of 195*195 m using the nearest neighbor method, to allow all layers to be combined. Then, a matrix of square cell structure was prepared for the study area. It consisted of a matrix of 273 rows and 273 columns representing a total of 39,650 cells. Of these, 74 cells were identified as knickpoint points. These areas were identified with code 1 (presence of knickpoint) and the rest of the cells that did not have knickpoints were recorded with code 0 (absence of knickpoint). Discussion: The probabilistic relationship of the presence of a knickpoint as one of the important results of the research was obtained by the logistic regression method. This relationship predicted the probability of the presence of knickpoints based on geological and geomorphic variables. The probability map of the knickpoints in the study area was obtained based on the statistical relationship. According to the results, there is a possibility of river knickpoints in the southwestern regions and parts of the northeastern Basin. The results of the probability ratio test to determine the statistical significance of each of the independent variables in the proposed model showed that the geology and the distance from the boundaries of the geological formations in the model were significant. The results of the Yuden index for the training dataset, validation, and test data were equal to 0.72, 0.76, and 0.66, respectively, which indicated the accurate information on the probability status of knickpoint points, especially for the test data of the model. The results of the Kappa agreement coefficient for training, validation, and test data were also equal to 0.62, 0.73, and 0.60, respectively, which indicated the agreement of both methods with the observed values of knickpoints. Conclusion: The results of this study showed that at the boundary of lithology, because of the presence of joints and cracks due to differences in the type of rocks, the probability of the presence of river break was more than other parts of the region. Although the presence of some relatively high slope knickpoints indicated active tectonics in that area, in the present study, the effect of the fault system or active tectonics in the formation of knickpoints was not statistically significant. Particularly, the reduction of local roughness index and baselines was associated with less tectonic activity, but in this study, the appearance of knickpoints has been associated with a decrease in these two factors. Keywords: Ghaleh Shahrokh Basin, Logistic Regression, Knickpoint, Probabilistic Modeling. References: - Alexander, J., & Leeder, M. R. (1990). Geomorphology and Surface Tilting in an Active Extensional Basin, SW Montana, USA. Journal of the Geological Society, 147(3), 461-467. - Aman, M. A., Yunus, A. P., & Javed, A. (2020). Fluvial Knickpoint Identification and Their Characterizations in the Drainage Basins of Western Ghats, India. Spatial Information Research, 1-10. - Ayalew, L., & Yamagishi, H. (2005). The Application of GIS-Based Logistic Regression for Landslide Susceptibility Mapping in the Kakuda-Yahiko Mountains, Central Japan. Geomorphology, 65(1-2), 15-31. - Berlin, M. M., & Anderson, R. S. (2007). Modeling of Knickpoint Retreat on the Roan Plateau, Western Colorado. Journal of Geophysical Research: Earth Surface, 112(3),1-16. - Bishop, P. (2007). Long‐Term Landscape Evolution: Linking Tectonics and Surface Processes. Earth Surface Processes and Landforms. The Journal of the British Geomorphological Research Group , 32(3), 329-365. - Boulton, S. J. (2020). Geomorphic Response to Differential Uplift: River Long Profiles and Knickpoints from Guadalcanal and Makira (Solomon Islands). Frontiers in Earth Science, 10(8), 1-23 - Bowman, D., Shachnovich-Firtel., Y., & Devora, S. (2007). Stream Channel Convexity Induced by Continuous Base Level Lowering, The Dead Sea, Israel. Geomorphology, 92(1-2), 60-75. - Gardner, T. W. (1983). Experimental Study of Knickpoint and Longitudinal Profile Evolution in Cohesive, Homogeneous Material. Geological Society of America Bulletin, 94(5), 664-672. - Gilbert, G. K. (1896). Niagara Falls and Their History. USA: American Book Company. - Gonga-Saholiariliva, N., Gunnell, Y., Harbor, D., & Mering, C. (2011). An Automated Method for Producing Synoptic Regional Maps of River Gradient Variation: Procedure, Accuracy Tests, and Comparison with Other Knickpoint Mapping Methods. Geomorphology, 134(3-4), 394-407 - Hayakawa, Y. S., & Oguchi, T. (2006). DEM-Based Identification of Fluvial Knickzones and Its Application to Japanese Mountain Rivers. Geomorphology, 78(1-2), 90-106. - Heine, R. A., & Lant, C. L. (2009). Spatial and Temporal Patterns of Stream Channel Incision in the Loess Region of the Missouri River. Annals of the Association of American Geographers, 99(2), 231-253. - Kamintzis, J. E., Irvine-Fynn, T. D. L., Holt, T. O., Jones, J. P. P., Tooth, S., Griffiths, H. M., & Hubbard, B. (2019). Knickpoint Evolution in a Supraglacial Stream. Geografiska Annaler: Series A, Physical Geography, 101(2), 118-135. - Lee, S. I., Lee, H., Abbeel, P., & Ng, A. Y. (2006). Efficient L1 Regularized Logistic Regression. In Aaai, 6, 401-408. - Loget, N., & Van Den Driessche, J. (2009). Wave Train Model for Knickpoint Migration. Geomorphology, 106(3-4), 376-382. - Mackey, B. H., Scheingross, J. S., Lamb, M. P., & Farley, K. A. (2014). Knickpoint Formation, Rapid Propagation, and Landscape Response Following Coastal Cliff Retreat at the Last Interglacial Sea-Level Highstand: Kaua ‘i, Hawai ‘i. Geological Society of America Bulletin, 126(7-8), 925-942. - Marrucci, M., Zeilinger, G., Ribolini, A., & Schwanghart, W. (2018). Origin of Knickpoints in an Alpine Context Subject to Different Perturbing Factors, Stura Valley, Maritime Alps (North-Western Italy). Geosciences, 8(12), 443. - Miller, J. R. (1991). The Influence of Bedrock Geology on Knickpoint Development and Channel-Bed Degradation Along Downcutting Streams in South-Central Indiana. The Journal of Geology, 99(4), 591-605. - Muehlbauer, J. D., & Doyle, M. W. (2012). Knickpoint Effects on Macroinvertebrates, Sediment, and Discharge in Urban and Forested Streams: Urbanization Outweighs Microscale Habitat Heterogeneity. Freshwater Science, 31(2), 282-295. - Nakas, C. T. (2014). Developments in Roc Surface Analysis and Assessment of Diagnostic Markers in Three-Class Classification Problems. REVSTAT – Statistical Journal, 12(1), 43-65. - Ouimet, W. B., Whipple, K. X., Royden, L. H., Sun, Z., & Chen, Z. (2007). The Influence of Large Landslides on River Incision in a Transient Landscape: Eastern Margin of the Tibetan Plateau (Sichuan, China). Geological Society of America Bulletin. 119(11-12), 1462-1476. - Pavano, F., Pazzaglia, F. J., & Catalano, S. (2016). Knickpoints as Geomorphic Markers of Active Tectonics: A Case Study from Northeastern Sicily (Southern Italy). Lithosphere, 8(6), 633-648. - Peifer Bezerra, D., & Persano, C. (2017). Identifying Knickpoints Using Elevation Breaks and Offsets in Slope-Area Scaling. Geophysical Research Abstracts of EGU General Assembly Conference. - Phillips, J. D., McCormack, S., Duan, J., Russo, J. P., Schumacher, A. M., Tripathi, G. N., ... & Pulugurtha, S. (2010). Origin and Interpretation of Knickpoints in the Big South Fork River Basin, Kentucky–Tennessee. Geomorphology, 114(3), 188-198. - Saadat, M., Khandelwal, M., & Monjezi, M. (2014). An ANN-Based Approach to Predict Blast-Induced Ground Vibration of Gol-E-Gohar Iron Ore Mine, Iran. Journal of Rock Mechanics and Geotechnical Engineering, 6(1), 67-76. - Schumm, S. A., Dumont, J. F., & Holbrook, J. M. (2002). Active Tectonics and Alluvial Rivers. Cambridge: Cambridge University Press. - Shahzad, F., & Gloaguen, R. (2011a). TecDEM: A MATLAB Based Toolbox for Tectonic Geomorphology, Part 1: Drainage Network Preprocessing and Stream Profile Analysis. Computers & Geosciences, 37(2), 250-260. - Shahzad, F., & Gloaguen, R. (2011b). TecDEM: A MATLAB Based Toolbox for Tectonic Geomorphology, Part 2: Surface Dynamics and Basin Analysis. Computers and Geosciences, 37(2), 261-271. - Verdel, C., Wernicke, B. P., Ramezani, J., Hassanzadeh, J., Renne, P. R., & Spell, T. L. (2007). Geology and Thermochronology of Tertiary Cordilleran-Style Metamorphic Core Complexes in the Saghand Region of Central Iran. Geological Society of America Bulletin, 119(7-8), 961-977. - Viera, A. J., & Garrett, J. M. (2005). Understanding Interobserver Agreement: The Kappa Statistic. Family Medicine Research Series, 37(5), 360-363. - Wohl, E. E. (1992). Gradient Irregularity in the Herbert Gorge of Northeastern Australia. Earth Surface Processes and Landforms, 17(1), 69-84. - Zhu, W., Zeng, N., & Wang, N. (2010). Sensitivity, Specificity, Accuracy, Associated Confidence Interval and ROC Analysis with Practical SAS Implementations. NESUG Proceedings: Health Care and Life Sciences, Baltimore, Maryland, 19, 67.
خلاصه ماشینی:
براساس نتايج اين پژوهش ، سازندهاي زمين شناسي و توپوگرافي در رخداد رودشکن ها در منطقۀ مطالعه شده نقش مهمي دارند و رگرسيون لجستيک نيز، مدل مناسبي براي پيش بيني وقوع رودشکن است .
Run–riffle–pool sequence مدل سازي احتمالاتي در بررسي عوامل مؤثر بر ايجاد رودشکن ها حامد ادب و همکاران 63 کوچک جغرافيايي شکل مي دهد؛ از طرفي براي اکولوژيست ها نيز جالب توجه است ؛ زيرا در مقياس ميکرو، اين نقاط محـل تنوع زيستي گونه هاي متفاوتي هستند (٢٨٣ :٢٠١٢ ,Muehlbauer and Doyle).
Independent variables used to investigate the statistical relationship with river knickpoint (به تصویر صفحه رجوع شود)مدل سازي احتمالاتي در بررسي عوامل مؤثر بر ايجاد رودشکن ها حامد ادب و همکاران 69 روش رگرسيون لجستيک باينري الگوريتم هايي تحليلي يا آماري وجود دارند که پراکنش نقاط را با ارتباط دادن با لايه هاي متغيرهـاي جغرافيـايي تحليـل و پيش بيني ميکنند.
Comparison of the predictive power of binary logistic regression in identifying the knickpoints (Authors, 1399) 74 ١٤٠٠ رودشکن ها وقفۀ ناگهاني توپوگرافي را در پروفيل کانال طولي نشان ميدهند و عموما به مثابۀ شکل هاي غيرتعـادلي در نظر گرفته ميشوند؛ براي نمونه مطالعات ميلر (١٩٩١) و هاياکاوا و همکاران ١ (٢٠٠٦) نشان ميدهند رودشـکن هـا درنتيجۀ عملکرد نيروهاي مختلفي مانند تأثير سنگ شناسي، فرايند هيدرولوژيکي و همچنـين زمـين سـاخت و سيسـتم گسلي ايجاد ميشوند.