Analyzing and Forecasting the Expansion of Mashhad City from 2000 to 2025 through Multi-temporal Satellite Imagery and Markov Chain

Document Type : Research Article

Authors

• Ebrahim Akbari ¹
• Rahman Zandi ²
• Rogue kalatah Meymari ³

¹ University of Tabriz

² University of Kahim Sabzevari

³ Ferdowsi University of Mashhad

10.22067/geo.v0i0.78082

Abstract

1 Introduction
The physical expansion of cities is a dynamic and continuous process through which the city boundaries and the physical spaces in vertical and horizontal directions are increased both quantitatively and qualitatively. If urban growth is uncontrollable and unplanned, it may negatively affect spatial benefits and ultimately lead to urban expansion. The present study aims to investigate the expansion of Mashhad City in Iran from 2001 to 2017 and then forecast the changes until 2026. A Markov chain is a mathematical and probabilistic method which acts as a random process via which the future state of a pixel depends only on its predecessor and is predicted on it. The direct result of this model is the transmission probability matrix, but in this model no geographic perceptions are obtained. In addition, modeling a single map and representing the spatial distribution of classes is not generated. To solve this problem, the CA-Markov model was designed by John Von Neumann in the 1950s in order to add a spatial feature to the Markov model.
2 Materials and Methods
In the present research, the ETM images of the 2000 and 2009, and OLI images of 2016 from the Landsat Satellite were employed. The images were taken in 28.06.2000, 6. 6. 2009, and 25. 5.2016, respectively. Initially, Landsat Satellite images were geometric and radiometric corrections to reduce the satellite imagery errors. As such, the study area is separated from the images and it is attempted to classify the satellite data. The method used to classify the information is the monitoring method via which educational samples are used to classify the pixels. This means that by defining the specific pixels of each image for each of the classes, classification is performed in the form of the considered classes. Moreover, the maximum similarity algorithm is used for the classification of monitoring. In this method, the reflective value method and each pixel are unknown. In terms of the variance and covariance of a particular spectral reaction class, it is assumed that the distribution of the data of each class is based on the normal distribution around the pixel average of the given class. Practically, the variance and covariance, and the mean of the various classes of each satellite image are calculated for the classification of phenomena; thus, each pixel belongs to a class whose presence in that class is more likely to occur. In order to find out the changes in land use in the study area (Mashhad City), including gardens and agricultural land uses, built-up areas, grassland, and rangelands, the Markov chain module was employed. In the Markov chains, the land-cover classes are used as the chain states. According to this analysis, we always use two raster maps called case models. In addition, the two maps illustrate the time interval between two the images and the predicted interval in the 1401 horizon in the CA-Markov model. The output of the Markov model also includes the possibility of turning the status and matrix of the converted areas in each class, and finally the images are probably conditional for different land use conversions. In this study, Cohen's kappa coefficient (κ) was used for confirming the classification.
3 Results and Discussion
In this research, changes in land cover in gardens and agricultural land uses, and built-up areas, meadows, and rangelands using satellite imagery in a period from 2000 to 2016, and also maximum similarity algorithm, monitoring method and Markov chain model were employed. In the Markov chain model, the cover classes are used as the chain states. In fact, the transmission area matrix represents the number of pixels which are converted from a class to another and has changed from any land-use to another within the same time since 2000-2016.
Based on the CA-Markov model, four floors of the land cover in the land uses mentioned in the strategic perspective document for 1404 (2025) were forecasted. In this regard, it was determined that during the years of 2016, 2009, and 2000, the areas of land ​​uses of the built-up areas were significantly increased, and also the rangelands were expanded relatively; however, the areas of ​​ gardens and agricultural land uses have severely decreased. In addition, the areas of ​​meadows have decreased. The area of ​​land uses in the strategic perspective document for 1404 (2025) is somehow similar to those in 2016; therefore, the gardens and agricultural land uses will change as 90.74%, built-up land uses as 121.57%, meadows as 26.21%, and rangelands as 100%. 
In fact, it was determined that the largest area of ​​transmission in different land uses was in the period between 1999 to 2009 for grasslands, gardens and agricultural land uses, built-up areas, and meadows. In addition, the highest probability of transmission area is in the period from 2000 to 2009 is related to meadows, rangelands, gardens and agricultural land uses, and built-up areas, respectively. The largest transmission area of the mentioned land uses is from 2009 to 2016, which is related to meadows, rangelands, built-up areas, and gardens and agricultural land uses, respectively. The highest transmission area of land uses is within the time interval of 2009-2016 related to meadows, rangelands, built-up areas, gardens and agricultural lands, respectively. It is also revealed that the area of ​​gardens and agricultural land uses was expanded in 2009 as compared to that of 2000, whereas it was significantly reduced in 2015.
The area of the built-up areas during the mentioned years was expanded, and in 2016 (as 359973900 m2) it was significantly increased. Meadows’ area was decreased in 2009 as compared to 2000, but it increased in 2016 as compared to 2009. In 2009, the area of rangelands was decreased as compared to 2000; however, it was again increased significantly in 2016 as compared to the years of 2009 and 2000.
4 Conclusion
According to the results obtained from the maps, in the three years of 2000, 2009, 2016, the most changes were related to built-up areas. Therefore, during this period the construction and physical growth of the city were mostly in the northwest direction. Moreover, because constructions usually are done on gardens and agricultural land uses, the decrease in the area of gardens and agricultural land uses, and consequently the increase of the built-up areas can be observed in this part of the city. According to the map of 2016, the gardens and agricultural land uses still remains in the south-east part of the city. One reasons for this issue may be the lack of growth of the city in this direction. One of the important issues in the field of urban planning is how urban spatial development and its resulting pattern are. The pattern derived from the spatial distribution of urban human activities called urban form is changing due to the dynamic and changing nature of cities. Urban growth is horizontal and vertical, and Mashhad has a horizontal urban growth. This form of urban growth over time has led to the disappearance of gardens and agricultural land uses and their transformation into built-up areas, resulting in economic, social and environmental crises. Therefore, land use management planning is one of the most important issues in Mashhad. If land use management planning is done correctly, optimally and accurately, many urban issues and problems may be solved.
 

Keywords

• Landsat
• Land use change
• Markov Chains
• Urban Physical Development
• Mashhad

---

References

اطهاری، کمال؛ 1379. به سوی کارآمدی دخالت دولت در بازار زمین شهری. فصلنامه اقتصاد مسکن. شماره 18. سازمان ملی زمین و مسکن شماره30. صص 245-237.

رمضانی، نفیسه، جعفری، رضا؛ ۱۳۹۳. آشکارسازی تغییرات کاربری و پوشش اراضی در افق ۱۴۰۴ با استفاده از مدل زنجیره‌ای CA مارکوف (مطالعه موردی: اسفراین). فصلنامه تحقیقات جغرافیایی. شماره 4. صص96-83.

رهنما، محمدرحیم، عباس زاده، غلامرضا؛ 1387. اصول مبانی و مدل‌های سنجش فرم کالبدی شهر. انتشارات جهاد دانشگاهی مشهد.

علوی پناه، کاظم؛ 1384. کاربرد سنجش‌ازدور در علوم زمین. انتشارات دانشگاه تهران.

علوی پناه، کاظم؛ متین فر، ح. ر؛ 1383. ارزیابی کاربری داده‌های ماهواره‌ای از نظر صرفه جویی وقت. کنفرانس ملی بهره وری. فرهنگستان علوم ایران. ص 478.

علیزاده ربیعی، حسن؛ 1392. سنجش‌ازدور (اصول و کاربرد). انتشارات سازمان مطالعه و تدوین کتب علوم انسانی دانشگاه‌ها (سمت). صفحه218.

فیضی زاده، بختیار؛ حاجی میری، محمود؛ 1387. آشکار سازی تغییرات کاربری اراضی با استفاده از روش طبقه‌بندی شیءگرا (مطالعه موردی: شهرک اندیشه). همایش ژئوماتیک 87. تهران: سازمان نقشه برداری کشور.

ماجدی، حمید؛1378. زمین مساله اصلی توسعه شهری. مجله آبادی. شماره33. مرکز مطالعات و تحقیقات شهرسازی و معماری ایران.

محمد اسماعیل، زهرا؛ 1389. پایش تغییرات کاربری اراضی کرج با استفاده از تکنیک سنجش‌ازدور. مجله پژوهش-های خاک (علوم خاک و آب). شماره1. صص88-81.

مرکز آمار ایران. سال‌های؛ 1335-1390.

مهندسین مشاور بافت شهر؛1381. گزارش طرح راهبردی و گردشگری منطقه طرقبه و شاندیز- مشهد.

مهندسین مشاور فرنهاد؛ 1389. طرح راهبردی توسعه و عمران شهر شاندیز. وزارت مسکن و شهر سازی خراسان رضوی.

واحدیان بیکی، لیلا، پوراحمد، احمد، سیف الدینی، فرانک؛ ۱۳۹۰. فصلنامه علمی- پژوهشی نگرشی نو در جغرافیای انسانی- سال چهارم. شماره 1. صص46-30.

Acevedo, W., Forestman, T. W., Buchanan,J. T., 1996. Origins and philosophy of building a temporal database to examine human transformation processes.Proceedings, ASPRS/ ACSM Annual Convention and Exhibition, Baltimore,, 24:148–161.

Ewing, R., 1997. Is Los Angelez-style sprawl desirable?.J. Am. Plan. Assoc. 63.107-27.

Fan, Fenglei, Wang, Yunpeng, and Wang, Zhishi., 2008, Temporal and spatial change,detecting (1998–2003) and predicting of land use and land cover in Core corridor of Pearl pp: 127-147.

Goward, S. N., G. D. Cruickshanks, and A. S.Hope., 1985. Observed relation between thermal emission and reflected spectral radiance of a complex vegetated landscape, Remote Sensing of Environment, 18: 137-146.

Guan, DongJie, Li, HaiFeng, Inohae, Takuro, Su Weici, Nagaie, Tadashi, and Hokao, Kazunori., 2011, Modeling urban land use change by the integration of cellular automaton and Markov model, Ecological Modelling, 222 (20–22), 3761-3772.

Gupta, R. P., 2003, Remote Sensing Geology. Second Edition, Springer-Verlag, Berlin,p: 655.

Kamusoko, Courage, Aniya, Masamu, Adi, Bongo and Manjoro, Munyaradzi, 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.

Knorn, J. A. Rabe, C.V. Radeloff, T. Kuemmerle, J. Kozak and Hostert, P., 2009. Land cover mapping of largeareas using chain classification of neighboring Landsat satellite images. Remote Sens, Environ. 113: 957– 964.

Mas, Jean-François, Melanie, Kolb, Martin, Paegelow, Maria Teresa, Camacho Olmedo, and Thoma, Houet., 2014, Inductive pattern-based land use/cover change models: Acomparison of four software packages, Environmental Modelling & Software, 51: 94-111.

Matsushita, B. Xu, M. and Fukushima, T., 2006. Characterizing Changes in Landscape Structure in the Lake Kasumigaura Basin, Japan Using a High-Quality GIS Dataset, Journal of Landscape and Urban Planning, 78: 241-250.

modeling as a practical tool for decision-makers, Journal of Environmental Management,129: 235-243.

Pontius, Robert. Gilmore. and Chen, Hao., 2006. Geomod Modeling, USA: Clark

River Delta (China) by using TM and ETM+ images, Environmental Monitoring and Assessment, 137: 127-147.

Sohl, Terry L. and Claggett, Peter R., 2013. Clarity versus complexity: Land-use

Torrens, P.M., Sullivan, D. O. 2000. Cellular Automata and Urban Simulation: Where do we go from here, Environment and Planning, 28: 163-168.

Tso. B. and P.M. Mather., 2009. Classification Methods for Remotely Sensed Data. Chapter 2-3. 2nd ed., Taylor and Francis Pub., America Upadhyay, Thakur, Solberg, Birger, and Sankhayan, Prem Lall., 2006, Use of models to analyse land-use changes, forest/soil degradation and carbon sequestration with special reference to Himalayan region: A review and analysis, Forest Policy and Economics, 9: 349-371.

Wang, Shi Qing, Zheng, Xizinqi, and Zang, X. B., 2012. Accuracy assessments of land use change simulation based on Markov-cellular automata model, Procedia Environmental Sciences, 13: 1238-1245.

Wassmer, R.W., 2002. Influences of the fiscalization of land Use and urban-Growth Boundaies,www.csus. edu/indiv/w/wassmerr/sprawl.html.

Whitford, Walter G., Translated by, Azarnivand, Hossein, and Malekian, Arash. 2008. Ecology of desert systems, Tehran: University of Tehran. P. 340.

Yuan,F.,K.E,Sawaya.B.C,Loeffelholz& M.E,Bauer., 2005. "Land cover lassification andchange analysis of the twin Cities (Minnesota).

Zhang, Zh., V. Lieven, C. Eva De, X. Ou and De.W. Robert., 2008. Vegetation change detection using artificial neural networks with ancillary data in Xishuangbanna, Yunnan Province, China. Chin. Sci. Bull. 52: 232-243.