نوع مقاله : مقاله پژوهشی

نویسندگان

1 استادیار دانشکده علوم زمین، دانشگاه شهید بهشتی

2 دانشیار مرکز مطالعات سنجش از دور، دانشگاه شهید بهشتی

3 دانشجوی دکتری دانشگاه شهید بهشتی

چکیده

مدل سازی تغییرات کاربری اراضی، ابزاری ضروری برای تجزیه و تحلیلهای محیط زیستی، برنامهریزی  و مدیریت محسوب میگردد. در حال حاضر آشکار سازی و مدلسازی تغییرات کاربری اراضی با استفاده از تصویر ماهوارهای ابزاری سودمند برای درک تغییرات زیست محیطی در رابطه با فعالیتهای انسانی به حساب میآیند. ناحیه مورد مطالعه یکی ازمناطق ایران است که هدف تجاوزساخت و سازهای بی رویه و بدون برنامه قرار گرفته است.  توسعه شهری و رشد جمعیت منجر به تغییرات الگوی فضایی شده و کاربری بخش زیادی از منابع طبیعی را  تحت تأثیر قرار داده است. در این تحقیق از تصاویر ماهواره لندست در سالهای ۱۹۸۶، ۲۰۰۲، ۲۰۱۸ برای طبقهبندی و آشکارسازی تغییرات کاربری اراضی استفاده شده است.پس از رفع خطاهای تصاویر ماهوارهای چهار کلاس عارضه، ساخت و ساز مسکونی و غیر مسکونی، پوشش گیاهی، کوه و مرتع و راه، جهت بررسی تغییرات در نظر گرفته شد. عملیات میدانی و برداشت عوارض نمونه، با گیرندههای GPS دو فرکانسه در محدوده مورد مطالعه انجام شد. سپس این عوارض به نرم افزار معرفی و با روش ماشینهای بردار پشتیبان[1] طبقهبندی روی تصاویر سه دوره انجام و میانگین دقت کلی و میانگین ضریب کاپا [2]  در این روش به ترتیب  ۶۲ /۹۶% ، ۳۳/۸۵% محاسبه گردید. بیشترین تغییرات مربوط به کلاس کاربریهای مسکونی و غیر مسکونی و راه میباشد. بیش ترین تغییرات مربوط به ساخت و ساز مسکونی ۰۶/۹ درصد و راه ۱ درصد میباشد، که این روند رو به افزایش سبب کاهش دو کلاس عارضه کوه و مرتع و پوشش گیاهی به ترتیب به میزان ۰۷/۹ و۱/۰ درصد شده است. در ناحیه مورد مطالعه اکثر پوششهای گیاهی و زمینهای کشاورزی تبدیل به شهرکهای صنعتی و ویلاهای تفریحی شده است.  در راستای چنین تغییراتی زنجیره مارکوف توانایی خوبی برای پیشبینی احتمال تغییرات را دارد و بر پیش بینیهای تغییرات کاربری اراضی متمرکز  است در حالی که اتوماتای ​​سلولی به عنوان یک روش قدرتمند در تشخیص تغییرات مؤلفه مکانی فضایی است. به این منظور جهت پیشبینی تغییرات در کمیت و فضا  از مدل ترکیبی زنجیره مارکوف و سلولهای خودکار استفاده گردید و نقشه کاربری اراضی برای سال ۲۰۵۰  شبیه سازی شد. نتایج نشان داد که مدلهای مارکوف اطلاعات مفیدی در اختیار ما قرار میدهد که میتواند برای برنامه ریزی کاربری اراضی در آینده  مفید واقع شود.



[1]- SVM


[2]- Overall Accuracy

کلیدواژه‌ها

عنوان مقاله [English]

Detection and prediction of land use changes using CA-Markov model Case study: Tehran - Damavand

نویسندگان [English]

  • Naser Shafiei Sabet 1
  • Alireza Shakiba 2
  • Ashkan Mohammadi 3

1 Assistant Professor,University Shahid Beheshti

2 Associate Professor,University Shahid Beheshti

3 ۤPh.D Student, , Faculty of science S.B.U

چکیده [English]

Extended Abstract
Introduction
Nowadays,satellite imagery is used as a suitable toolforproduction of land use maps. It is also considered to be an important resource used for urban and rural land use planning. Due to the general coverage of different phenomena and natural resources, satellite imageriesplay a major role in spatial and temporal analysis. Using these images in various fields can show us their capabilities and limitations. The important point is to consider increasing advances in their spectral and spatial capabilities. Systematicexploitation of natural resources requires patterns and models of the region, so that related regulations are observedand sustainable utilization is also considered.Obviously,exact, accurate, fast and economic estimate of these changes is impossible without modern technologiesused for regional and environmental studies.Land use change modelingis an indispensable tool for environmental analysis, planning and management. Eastern parts of Tehran metropolis are among regions facing unstructuredand unscheduled constructions in Iran. Urban development and population growth have led to rapid changes in spatial patterns and have severely affected land use and natural resources.
 
Materials and methods
In order to investigate land use changes, the present study takes advantage of satellite imageries, remote sensing techniques and spatial information systems.The trend of land use changeswas separately extracted from satellite imageries received in1986, 2002, and 2018.After visual interpretation and error correction,four categories were selected (residential and non-residential construction, vegetation, mountain and grassland) based on which changes were investigated. After data collection (including imageries received from Landsat satellite and TM, ETM and OLI sensors) classification and detection commenced.Then, suitable band was selected for classification, spectral reflectance curves of each land use class were evaluated and bands correlation histograms were compared.since changing bandsgives a comprehensive understanding of the classes, their relations and resolution, two-band diagram of pixels’ distribution in two different bands was used.Properties of the texture were extracted using GLCM matrix and principal component analysis was performed. Support Vector Machine was selected as an optimal classification method. Feature vectors and the training rangeweregiven to this algorithm as its input.Markov chain works well in predicting probability of change, and especiallyland use changes. Cellular automaton is also a powerful method used for detecting changes in spatial component. Thus,Markov chain and automated cells model were both used in order to predict changes in quantity and space, and land use map was predicted and simulated for 2050.Results indicate that Markov models provide useful information which can be beneficial for future land use planning.
 
Results and discussion
Calculations indicate thatdue to creeping discrete growth and in some areas continuous growth, most changes in Damavand (in Tehran)have happened in the category of residential construction (9.06%) and road (1%).This increasing trend has reduced two classes of mountain/grassland and vegetation cover by 9.07% and 0.1%, respectively. After field operations and sampling with dual-frequency GPS receivers, data was introduced to software and classification was performed using support vector machines with an average overall accuracy of 96.62% and a mean kappa coefficient of 85.33%. Change detection studiesindicate that in time period of 1986 to 2002,most changes have occurred in residential and non-residential construction category. In fact, ​​residential and non-residential construction has reached from 3.1% in 1986 to 6.1% in 2002 year, while mountain and grassland category has faced 2.96% decrease. Also, vegetation cover has decreased by 0.76%.Likewise, we also saw a 6.15% increase in residential and non-residential construction, a 6.11% decrease in mountain and grassland and a 0.22% decrease in vegetation cover of the study area in the time period of 2002 to 2018.Road category had an 81% increase in the first time period and an 18% increase in the second time period. Overall, residential/non-residential construction and roads have increased, while mountains/grassland and vegetation cover have decreasedin the time period of 1986 to 2018. Due to population overflow in recent decades, and unplanned construction, land uses like vegetation cover and grassland have changed into residential construction, and especially industrial land use in the area under study (Jajrood, Kamard, KhorramDasht, Shamsabad, Mehrabad, Pardis and Siasang).
 
Conclusion
While investigating spatial evolution and agricultural land use changes, it is important to distinguish betweenrapidly changing phenomenon, and slowly changing one.Results of the present study indicate that compared to other land uses,vegetation cove has changed more severely. Therefore, without necessary policies and actions to prevent this process,pressure on naturalresources, land use changes, and consequently destruction of valuable resourceswill result in harmful environmental impacts. This will also change the economic performance of the villages, and have many negative spatial, socio-economic consequences.

کلیدواژه‌ها [English]

  • Detection
  • Markov chain model
  • Cellular automata model
  • Land use change
1. رضازاده، میراحمدی؛ راضیه، مهرداد (1388). مدل اتوماسیون سلولی، روش نوین در شبیه سازی رشد شهری، نشریه علمی پژوهشی فناوری آموزش، دوره چهارم، شماره یکم، تهران
2. شفیعی ثابت، ناصر (1393)، خزش کلان شهر تهران و ناپایداری کشاورزی روستاهای پیرامونی، نشریه علمی پژوهشی آمایش محیط، دوره هفتم، شماره بیست و چهار.
3. شفیعی ثابت، خاکسار؛ ناصر، سوگند (1396)، پیامدهای محیطی ـ اکولوژیک خزش شهری در سکونتگاه‌های روستایی پیرامون شهر همدان , علوم محیطی,  15(3) ،55-74.
4. Arsanjani, Jokar , Wolfgang , Mousivand ,Jamal, Kainz , Ali, Jafar, (2011), Tracking dynamic land use change using spatially explicit Markov Chain based on cellular automata، the case of Tehran”, International, Journal of Image and Data Fusion, 2، 4, United Kingdom
5. Clark Labs, (2006), IDRISI Geographic Information Systems and Remote Sensing Software; Clark Labs، Worcester, MA, USA.
6. Deep, Saklani, (2014). Urban sprawl modeling using cellular automata. The Egyptian Journal of Remote Sensing and Space Science. 17(2), 179-187.
7. Eastman J.R, (2006), IDRISI Andes Tutorial. Clark-Labs, Clark University, Worcester, 284 pp.
8. Gross, JE, Goetz, SJ, Cihlar, J, (2009), Application of remote sensing to parks and protected area monitoring Introduction to the special issue, Remote Sensing of Environment, 113, 7, 1343-1345
9. Hartter, J. South worth, J. Dwindling ,(2009),  Resources and fragmentation of landscapes around parks، Wetlands and forest patches around Kibale National Park, Uganda. Landsc. Ecol, 24, 643–656.
10. Hua, A., (2017), Application of Ca-Markov model and land use/land cover changes in Malacca RiverWatershed,Malaysia. Appl. Ecol. Environ. Res., 15, 605–622.
11. Hu, Zhiyong, and Lo, C. P.  , (2007), Modeling urban growth in Atlanta using logistic regression. Computers, Environment and Urban Systems, 31، 6. United Kingdom.
12. Iqbal Sarwar Md. Billa M. Paul, Alak.,(2016), Urban land use change analysis using RS and GIS in Sulakbahar ward in Chittagong city, Bangladesh. Internatinal Journal Of Geomatics and geosciences. 1، 7, Pp 1-10.
13.Jamal Jokar Arsanjani, Marco Helbich, Wolfgang Kainz, Ali Darvishi Boloorani, (2013), “Integration of logistic regression, Markov chain and cellular automata models to simulate urban expansion”, International Journal of Applied Earth Observation and Geoinformation,    21, pp 265-275, doi،10.1016/j.jag.12.014.
14. Jensen, J. R. ,(2015),  Introductory digital image processing 4 rd edition, In Upper Saddle River، Prentice hall.
15. Kamusoko, Masamu, Bongo, Courage, Aniya, Adi, (2009), Rural sustainability under threat in Zimbabwe-simulation of future land use/cover changes in the Bindura district based on the Markov-cellular automata model, Applied Geography 29 .435–447.
16. Krieger, D. J. ,(1999) , Saving open spaces: Public support for farmland protection. American Farmland Trust, Center for Agriculture in the Environment
17. Kuldeep, Tiwari. And Kamlesh, Khanduri ,(2011), Land Use / Land cover change detectionin Doon valley (Dehradun Tehsil), Uttarakhand، using GIS& Remote Sensing Technique, International Journal of Geomatics and Geosciences, 2 (1)، Pp 34-41
18. Li, HaiFeng. Inohae Takuro and Su Weici and Nagaie Tadashi and Hokao Kazunori, (2011), Modeling urban land use change by the integration of cellular automaton and Markov model. Ecological Modelling, 222، 20, Netherlands.
19. Mas, Jean-François, Melanie, Kolb, Martin, Paegelow, María Teresa, Camacho lmedo, and Thoma, Houet,(2014), Inductive pattern-based land use/cover change models, A comparison of four software packages, Environmental Modelling & Software, 51, 94-111.
20. Messina, J. P. Walsh, S. J., 2. 5D Morphogenesis،(2001), Modeling landuse and landcover dynamics in the Ecuadorian Amazon. Plant Ecol, 156, 75–88.
21. Mitsova, D, Shuster, W, Wang, X, (2011), A cellularautomata model of land cover change to integrate urbangrowth with open space conservation, Landscape andUrban Planning, 99, 2, 141-153.
22. Pijanowski, Brown, Manik, (2002), Using neural nets and GIS to forecast land use changes، a land transformation model, Computers, Environment and Urban Systems 26 (6) 553–575.
23.Rahel Hamad, heiko Balzter,(2018), Predicting Land Use/Land Cover Changes Using a CA-Markov Model under Two Different Scenarios
24. Taubenböck, Hannes. Thomas Esch and Andreas Felbier and Michael Wiesner
And Achim Roth, and Stefan Dech,(2012), Monitoring urbanization in mega cities from space. Remote sensing of Environment, 117, Netherland
25. Vaz, Eric. Nijkamp Peter and Painho Marco, and Caetano Mario,(2012), A multi-scenario forecast of urban change، a study on urban growth in the Algarve, Landscape and Urban Planning, 104، 20, Netherland
26. Verburg PH, Soepboer W, Veldkamp A,Limpiada R, Espaldon V, Mastura SS, (2002),.Modeling the spatial dynamics of regional land use: the CLUE-S model. Environmental Management, 30(3): 391-405.
27.Verburg, P. H. ; de Nijs, T. C. M. ; Ritsema van Eck, J. ; Visser, H. ; de Jong, K., A. (2004), Method to analyse neighbourhood characteristics of land use patterns. Comput. Environ. Urban Syst,28, 667–690.
28. Verda Kocabas, Suzana Dragicevic, ,(2006), Assessing cellular automata model behaviour using a sensitivity analysis approach, Computers, Environment andUrban Systems 30 , 921–953.
29. Wang, Shi Qing, Zheng, Xizinqi, and Zang, X.B., (2012), Accuracy assessments of land use changesimulation based on Markov-cellular automata model,Procedia Environmental Sciences, 13, 1238-1245
30.Weng, Q,(2002), Land use change analysis in theZhujiang Delta of China using satellite remote sensing,GIS and stochastic modelling, Journal of Environmental Management, 64, 3, 273-284.
31. Yang X, Zheng X-Q, Lv L-N, (2012), Aspatiotemporal model of land use change based on ant colony optimization, Markovchain and cellular automata. Ecological Modelling, 233: 11-19.
32.Zeng, C.H, Liu, Y, Stein, A, Jiao, L., (2015). Characterization and spatial modeling of urban sprawl in the Wuhan Metropolitan Area, China, International Journal of Applied Earth Observation and Geoformation. 35, 10-24. ussc.html, The Christian Science Monitor- and taxes septic systems.