Monitoring land use changes in Lake Urmia and its surroundings using various methods of statistical training theory

Document Type : Original Research

Authors
H. Nazmfar 1
M. Shirzad 2
1 university of Mohaghegh Ardabili, Ardabil, Iran.
2 University of Mohaghegh Ardabili, Ardabil, Iran
Abstract
Extended abstract

Introduction

Land use is one of the most important biophysical and socio-economic characteristics in any watershed. The science of land change has recently been introduced as one of the fundamental components of global environmental change and sustainable development research. Monitoring land changes is important in future planning and natural resource management. Therefore, the need to detect such changes in an ecosystem is very important. Therefore, the need to detect such changes in an ecosystem is very important to take appropriate action if necessary. . Due to the fact that Lake Urmia is an important ecotourism center in Azerbaijan, with the drying up of the lake, Greater Azerbaijan and all the areas affected by this phenomenon will face a recession of domestic tourists. These factors, in turn, will lead to the migration of residents of the villages of this region to the surrounding cities and social problems in these cities. Its catchment area has been one of the water resources of this area[r11] . But the extent to which these changes, and especially the change in land use, have taken place, requires special study. In general, it is possible to study land use changes in both terrestrial and remote sensing methods. However, in recent decades, with the development of hardware and software facilities for processing satellite images, as well as the ease of access to multi-spectral and ultraviolet images, the use of remote sensing techniques to produce land use maps has become more common. The use of remote sensing technology has a special place in natural resource studies. Multi-time comparison, information updates, digital processing, data diversity, and data transfer speeds have made remote sensing the most important technology in detecting changes.

Methodology

The approach of the present study is developmental-applied and its descriptive-analytical method. According to the subject of the research and in line with the objectives defined in this research, satellite image with the specifications listed in Table (1) and the softwares of Google Earth, ENVI4.8, ArcGIS10.2 have been used. To use satellite imagery to perform techniques, all images must have the same coordinates. Remote sensing techniques, especially those used to classify land use and detect changes, are usually monitored and analyzed based on similar pixels in multi-time images; Corrections, images are not properly geometrically and radiometrically corrected, research accuracy is reduced. Thus, the satellite images of 1989, 2000, 2016, and 2019 were returned to the image with an RMS error of 0.42 pixels, capturing 20 control points from the image surface to the image method. In geometric correction, the ground control points were tried to have a good distribution at the image level so that the mathematical model used to calculate the unknown coefficients in the equation would have less error. To convert the corrected image coordinates to the non-corrected image, a second-order function was used. . In this study, the numerical value reduction method of dark pixels for radiometric correction of images has been used. In this method, a constant value of the total value of the pixels in a given band is reduced to apply radiometric corrections to each satellite image. In the next step, the images were mosaic due to the location of the study area in two women (1368-348)[r12] . Then, using field visits and the global location apparatus, instructional samples for each use (lake, agriculture, salt marsh, other lands) were identified in the study area.

Results and discussion

In this study, three supervised classification methods (neural network, backup vector machine and maximum probability) have been used to extract land use maps. By comparing the accuracy of the classification obtained from the methods mentioned in Table (2), it was found that the classification method of the backing vector machine with a cap rate of 99.75% is more accurate than other methods. According to the results of both classification methods of machine vector support and neural network, precise methods for extracting land uses and in separating the phenomena that have close spectral behavior are very successful, especially support vector machine , which . Which was a bit successful.[r13]

Conclusion

In this study, first, images of measuring satellites (MSS-TM-OLI) were used and the map of Urmia Lake, lake landscaping and its surroundings were was extracted by applying supervised classification (support vector machine, neural network and maximum probability). . Comparison of image stratification methods showed that the support vector machine method has more classification accuracy than the other two methods due to its general accuracy and higher capability coefficient. The results also show that satellite imagery has a significant ability to extract land uses. Also, in order to investigate the trend of land use change, maps extracted from satellite imagery in 1989, 2000, 2016 and 2019 were compared. Examination of land use maps in the three mentioned periods showed significant changes in land cover. These changes include: Agricultural land use area has increased significantly from 1989 to 2019 due to the favorable area for agriculture and drilling wells. Numerous and the use of aquifers has been underground . Analysis of Landsat satellite images showed that significant fluctuations in the lake's water level have occurred over the years. So so that the water level changes of Urmia Lake from 1989 to 2016 have increased from 5348 to about 2705 square kilometers. However, from 2016 to 2019, due to heavy cross-sectional rains, it had an increase in water area of ​​1644 square kilometers. The images also show that the coastline, especially in the east and southeast of the study area, has a significant number of boys. From 1989 to 2000, the area of ​​this land use increased by 378 square kilometers. Also, between 2000 and 2016, its area continued to rise and increased to 786 square kilometers. However, due to the increase in cross-sectional rainfall during 2016 to 2019, the water level of the lake has increased and some of the salt marshes have been submerged and the land use area of ​​the salt marshes has decreased by 838 square kilometers.








Keywords

Subjects

-- آرخی، صالح. ادیب نژاد، (1390). ارزیابی کارایی الگوریتم های ماشین بردار جهت طبقه بندی کاربری اراضی با استفاده از داد ههای ماهوار های ETM لندست )مطالعه موردی: حوزه سد ایلام(، فصلنامه علمی پژوهشی تحقیقات مرتع و بیابان ایران.
- اکبری، علی .شکاری، الهه (1392). پردازش و استخراج اطلاعات از داده‌های ماهوارهای با استفاده از نرم افزار ENVI با نمونه های کاربردی در علوم زمین،نقشه برداری، جغرافیا و محیط زیست، انتشارات ماهواره،جلد اول.
- دهستانی، غلامرضا(1377). طبقه‌بندی تصاویر چند طیفی سنجش از دور با استفاده از شبکه عصبی، پایان نامه کارشناسی ارشد، دانشگاه تربیت مدرس.
- رضائی‌مقدم، محمد حسین. ولی‌زاده کامران، خلیل. رستم‌زاده، هاشم.رضائی، علی(1391).ارزیابی کارایی داده‌های سنجنده MODIS در برآورد خشکسالی(مطالعه موردی: حوضه‌ی آبریز دریاچه‌ی ارومیه)، جغرافیا و توسعه پایدار محیط، شماره 5،صص 52-37.
- ربیعی، حمیدرضا. ضیائیان،پرویز. محمدی، عباسعلی(1384).کشف و بازیابی تغییرات کاربری و پوشش اراضی شهر اصفهان به کمک سنجش از دور وسیستم اطلاعات جغرافیایی، فصلنامه مدرس علوم انسانی،شماره4، صص32-19.
- سنجری، صالح. برومند، ناصر(1392). پایش تغییرات کاربری/پوشش اراضی در سه دهۀ گذشته با استفاده از تکنیک سنجش از دور)مطالعۀ موردی:منطقۀ زرند استان کرمان(، مجلۀ کاربرد سنجش از دور و GIS در علوم منابع طبیعی، سال چهارم، شمارۀ 1، صص 67-57.
- شنایی، هویزه. زارعی، حیدر (1395). بررسی تغییرات کاربری اراضی طی دو دهۀ دوره زمانی (مطالعۀ موردی: حوزۀ آبخیز ابوالعباس)، پژوهشنامۀ مدیریت حوزۀ آبخیز، شماره14،صص244-237.
- علی محمدی، عباس؛ موسیوند، علی جعفر و شایان سیاوش(1389)." پیش بینی تغییرات کاربری اراضی و پوشش زمین با استفاده از تصاویر ماهواره ای و مدل زنجیزه مارکوف" نشریه برنامه ریزی و آمایش فضا.د 14. ش 3. صص 130-117.
- عبدالهی، علی اصغر؛ خبازی، مصطفی و درانی زهرا(1399). " مدلسازی تغییرات کاربری اراضی با استفاده از شبکه عصبی پرسپترون(مطالعه موردی: شهر لاهیجان)" نشریه برنامه ریزی و آمایش فضا. د 24. ش 1،صص 79-49.
- علوی‌پناه، سیدکاظم(1384). کاربرد سنجش از دور در علوم زمین،انتشارات دانشگاه تهران، تهران.
- فاطمی، سید باقر. رضایی، یوسف (1389). مبانی سنجش از دور، چاپ دوم، انتشارات آزاده، تهران.
- فتحیان، فرشاد.مرید، سعید(1391). بررسی متغییرهای هواشناسی و هیدرولوژیکی حوضه دریاچه ارومیه با استفاده از روش-های غیر پارامتری، مجله تحقیقات آب و خاک ایران، صص269-259.
- قربانی، مهدی. نظری سامانی، علی اکبر . کوهبنانی، حمیدرضا. اکبری ؛ فاطمه ؛ جلیلی پروانه، زهرا(1389). ارزیابی روند تغییرات کاربری اراضی حوزۀ آبخیز طالقان، مجموعه مقالات چهارمین کنگره بین المللی جغرافیدانان جهان اسلام، زاهدان.
- کشاورز، احمد و قاسمیان یزدی، محمد حسین(۱۳۸۴). یک الگوریتم سریع مبتنی بر ماشین بردار پشتیبان برای طبقه بندی تصاویر ابر طیفی با استفاده از همبستگی مکانی،نشریه مهندسی برق و مهندسی کامپیوتر ایران. صص37-44.
- مالیان، عباس. محمدی، علی. محمدی، عباس علی. ولی اللهی، جلال(1395).آشکارسازی و پیش‌بینیروند تغییرات دریاچه-ی ارومیه و محیط پیرامونی آن طی نیم قرن گذشته برپایه‌ی تحلیل‌های مکان مبنای دورکاوی، هیدروژئومورفولوژی. شماره9،صص62-43.
- مسیبی، مرضیه.ملکی، محسن(1393). پایش تغییرات کاربری اراضی با استفاده از داده‌های سنجش از دور و سیستم اطلاعات جغرافیایی(مطالعه موردی شهرستان اردبیل)، فصلنامه سنجش از دور و سامانه اطلاعات جغرافیایی در منابع طبیعی. سال پنجم، شماره1.
- محمدیاری، فاطمه. پورخباز، حمیدرضا. توکلی، مرتضی. اقدر، حسین(1393). تهیۀ نقشۀ پوشش گیاهی و پایش تغییرات آن با استفاده از تکنیک‌های سنجش از دور و سیستم اطلاعات جغرافیایی )مطالعۀ موردی: شهرستان بهبهان(، فصلنامۀ علمی-پژوهشی اطلاعات جغرافیایی سپهر، دوره 23، شماره9،صص 23-34.
- محمدی، مجید؛ امیری، مجتبی و جعفر دستورانی(1394)." مدلسازی و بررسی تغییرات کاربری اراضی شهرستان رامیان در استان گلستان". نشریه برنامه ریزی و آمایش فضا.ج 19.د 19.ش 4. صص 158-141.
- نادر صفت، محمد حسین(1390). ویژگی‌های ژئومورفولوژی دریاچه ارومیه وتاثیرآن در اکوسیستم این منطقه، مجله دانشنامه جغرافیا، دانشگاه آزاد اسلامی واحد علوم و تحقیقات.
- یوسفی، صالح. مرادی، سید حسینی. میرزایی، سمیه(1390). پایش تغییرات کاربری اراضی ،مریوان مجلۀ کاربرد سنجش از دور و GIS در علوم منابع طبیعی، سال چهارم، شمارۀ 1،صص.105-97.
- Arekhi, S., and Adibnejad, M. (2011). Efficiency assessment algorithms, support vector machines to classify land use and desertification of pasture research, the eighteenth year, No. 3(44). (In Persian).
- Akbari, Ali. Shakari, Elahe (2013). Processing and extracting information from satellite data using ENVI software with applied examples in Earth Sciences, Mapping, Geography and Environment, Satellite Publications, Volume One. (In Persian).
- -Ali Mohammadi, Abbas; Musavand, Ali Jafar and Shayan Siavash (2010). "Predicting land use change and land cover using satellite images and Markov chain model" Journal of Spatial Planning and Planning. (In Persian).
- - Abdollahi, Ali Asghar; Khabazi, Mostafa and Durrani Zahra (1399). "Modeling Land Use Change Using Perspective Neural Network (Case Study: Lahijan City)" Journal of Spatial Planning and Planning. D 24. Sh1, pp. 79-49. (In Persian).
- Bhagawat, R. (2013). Application of remote sensing and gis, land use/land cover change in kathmando metropolitan city, Nepal Journal of Theoretical and Applied Information Tech. 3: 2-7.
- Dehestani, Gholamreza (2000). Classification of multi-spectral images of remote sensing using neural network, Master's thesis, Tarbiat Modares University. (In Persian).
- Esam, I., F. Abdalla and N. Erich (2012). Land Use and Land Cover Changes of West Tahta Region, Sohag Governorate, Upper Egypt, Journal of Geographic Information System, 4: PP. 483-493.
- Feyzizadeh, B., Azizi, H. & Valizadeh, K.KH., 2007, Extraction Land Uses Malekan City Using Satellite Images EThr (In Persian), Amayesh, Vol. 2, No.3, PP. 1-10.
- Fatemi, Seyed Baqer. Rezaei, Yousef (2010). Fundamentals of Remote Sensing, Second Edition, Azadeh Publications, Tehran.
- Fathian, Farshad. Marid, Saeed (2012). Investigation of meteorological and hydrological variables of Urmia Lake basin using non-parametric methods, Iranian Journal of Soil and Water Research, pp. 269-259. (In Persian).
- Chander, G., Markham, B. L., & Helder, D. L. (2009).Summary of current radiometric calibration coefficients for Landsat MSS, TM, ETM+, and EO-1 ALI sensors. Remote sensing of environment, 113(5), 893-903.
- Chavez jr، P.S.( 1988).An Improved Dark-Object Subtraction Technique for AtmosPheric Scttering Correction of Multispectral Data. Remote Sensing of Environment، 24:459-479.
- Ghorbani, Mehdi. Nazari Samani, Ali Akbar. Kuhbanani, Hamidreza. Akbari; Fatima ; Jalili Parvaneh, Zahra (2010). Assessing the trend of land use change in Taleghan watershed, Proceedings of the 4th International Congress of Geographers of the Islamic World, Zahedan. (In Persian).
- Keshavarz, Ahmad and Qasemian Yazdi, Mohammad Hossein (2005). A fast algorithm based on support vector machine for classifying spectral cloud images using spatial correlation, Iran Journal of Electrical Engineering and Computer Engineering pp. 37-44. (In Persian).
- Liu, T., and X. Yang (2015). Monitoring land changes in an urban area using satellite imagery, GIS and landscape metrics. Applied Geography, 56: 42-54.
- Malian, Abbas. Mohammadi, Ali. Mohammadi, Abbas Ali. Vali Elahi, Jalal (2016). Revealing and predicting the changes in Lake Urmia and its surroundings over the past half century based on the analysis of the location of Durkhevi, hydrogeomorphology. No. 9, pp. 62-43. (In Persian).
- Mosayebi, Marzieh. Maleki, Mohsen (2014). Monitoring land use changes using remote sensing data and geographic information system (a case study of Ardabil city), remote sensing quarterly and geographic information system in natural resources. Fifth Year, No. 1. (In Persian).
- -Mohammadi, Majid; Amiri, Mojtaba and Jafar Farmanani (2015). "Modeling and reviewing land use changes in Ramian city in Golestan province". Journal of Spatial Planning and Planning, Vol. 19., 19. 19. 4. pp. 158-141. (In Persian).
- Mohammadyari, Fatemeh. Pourkhabaz, Hamidreza. Tavakoli, Morteza. Iqdar, Hussein (2013). Preparation of vegetation map and monitoring of its changes using remote sensing techniques and geographical information system) Case study: Behbahan city (, Sepehr Geographical Information Scientific-Research Quarterly, Volume 23, Number 9, pp. 23-34. (In Persian).
- Nader Sefat, Mohammad Hossein (2010). Characteristics of Geomorphology of Urmia Lake and its effect on the ecosystem of this region, Journal of Geography, Islamic Azad University, Science and Research Branch. (In Persian).
- Rawat, J.D., Kumar, M. (2015). Monitoring land use/cover change using remote sensing and GIS techniques: A case study of Hawalbagh block, district Almora, Uttarakhand, India, The Egeptyan Journal of Resmoet Senaing and space scinces, 18: PP. 77-84.
- Richards J., A. (2013). Remote sensing digital image analysis, fifth edition, springer, 494 p.
- Rezaei Moghaddam, Mohammad Hussein. Valizadeh Kamran, Khalil. Rostamzadeh, Hashem. Rezaei, Ali (2012). Evaluating the Efficiency of MODIS Measuring Data in Drought Assessment (Case Study: Urmia Lake Basin), Geography and Sustainable Environmental Development, No. 5, pp. 52-37 . (In Persian).
- Rabiee, Hamidreza. Ziaian, Parviz. Mohammadi, Abbas Ali (2005). Discovery and recovery of land use changes and land cover in Isfahan with the help of remote sensing and geographic information system, Quarterly Journal of Humanities Teacher, No. 4, pp. 32-19. (In Persian).
- Şatır, O., & Berberoğlu, S. (2012). Land Use/Cover Classification Techniques Using Optical Remotely Sensed Data in Landscape Planning. Landscape Planning, Dr. Murat Ozyavuz (Ed.), ISBN, 978-953.
- Sanjari, Saleh. Boroumand, Nasser (2013). Monitoring land use change / land cover in the last three decades using remote sensing technique) Case study: Zarand area of Kerman province (Journal of Remote Sensing Application and GIS in Natural Resources Sciences, Fourth Year, No. 1, pp. 67-57). (In Persian).
- Swimming, Hoveyzeh. Zarei, Haidar (2016). Investigation of land use changes during two decades of time period) Case study: Abol Abbas watershed (Research Journal of Watershed Management, No. 14, pp. 244-237). (In Persian).
- Swain, P.H. and S.M. Davis. (1987). Remote Sensing: the Quantitative Approach. McGraw-Hill. USA.
- Shalaby, A., and R. Tateishi. (2007). Remote sensing and for mapping and monitoring land cover and land use changes in the Northwestern coastal zone of Egypt. Applied Geography. 27: 28-41.
- Tayebi, A., Delavar, M.R., Saeedi, S., Amini, J. and Alinia, H ( 2008). Monitoring Land Use Change By MIulti-Temporal Landsat Remote Sensing Imagery the International Archives the Photogrammetry. Remote Sensing Information Science. XXXVII Part B7 Beijing.
- Yousefi, Saleh. Moradi, Seyed Hosseini. Mirzaei, Somayeh (2011). Monitoring land use changes in Marivan land in GIS sciences, Applied Journal of Remote Sensing and .Landsat Satellite ETM + and TM Using Natural Resource Meters(In Persian).
The Journal of Spatial Planning and Geomatics
Volume 24, Issue 3
October 2020
Pages 201-229