Spatio-temporal Modeling of urban development,with an emphasis on landuse change by using of Markov-CA Model Case study:parsabad moghan city

Introduction
With the increasing development of human societies, urban societies have also undergone many changes and complexities. This transformation and complexity encompasses all aspects of human life in cities. One of the most important categories in urban development is having a dynamic economy. Parsabad city is one of the a few new cities in Iran that has been created under the influence of this theory and the category of comparative advantage of regions in the second quarter of the present century. The city was established in line with the development plan of the Moghan Plain Civil Enterprise in 1953 and in order to exploit the fertile lands of the plain as a new and fledgling city. Because this development was done rapidly in fertile lands .It is considered the best agricultural lands in the Moghan plain, and with the expansion of the city, the agricultural lands around the city were being reconstruct day by day, causing changes in the landuses The study of physical, economic, social and biological conditions indicates the development of the city and changes in its land use. Since the use of traditional methods to determine urban development and land use changes is very time consuming and costly, so using multi-time satellite images and digital maps and a variety of models can model urban development landuse changes and simulated and examined the relevant implications. The purpose of this research is:

Assess using quantity models in Predicting urban development trends and land use changes.
Assess the future pattern of urban development and land use in Parsabad city.

 Methodology
This research with respect to type of purpose and nature were classified  applied and descriptive-analytical research. The instruments used in this research are Landsat satellite images of TM, ETM +, OLI sensors included in 1989, 2002, June 2015 period. For analysis of images , ENVI 5, IDRISI SELVA, ARC GIS software were used . All classification options in ENVI 5 software are situated in the Classification menu. The unsupervised and supervised classification methods were placed in unsupervised and supervised option. In this study, we used the supervised classification method for classification.
 Results and discussion
In general, the process of simulating urban development and land use changes are categorized in three general stages:
Create a transition matrix using Markov chain analysis
• Create competency maps for each land use class using Multi-Criteria evaluations (MCE) Weighted Linear Combination (WLC)
• Accomplishing the transition matrix from Markov chain analysis and spatial fit maps of land use classes and land use / land cover map for simulation automated cells-based.
To obtain the transition probability matrix, we use the user map which obtained from satellite images from 2002 to 2015. The weighted linear composition and multi-criteria evaluation method were  used to create spatial fit maps for each land use. These criteria was considered in the form of factors and constraints. Factors: distance from water levels, distance from city center, distance from road, distance from constructed lands, landuses, slope to percentage. Limitations of these criteria were e constructed lands, water levels and its 50-meter area.
For simulating the land use map of 2028, the land use map of 2015 has been used. All the simulation steps have been done in IDRISI SELVA software environment using markov-ca model. In 2028, the amount of constructed lands increased by 400 hectares, instead, the use of farm lands decreased by 410 hectares. As presented in figures the greatest decrease and change of landuse in this period related to farm lands, and the barren lands decreased approximately 20 hectares by  Maid, Exploiting from forest and  water land use show a relative increase. By comparing the map of 2015 and the simulated map for 2028, it can be concluded that the development of constructed lands have been unsuitable  to barren and farm land use., Also with the development of constructed lands, the use of ​​farm and barren lands area, especially farms, has been reduced.
 Conclusion
With respect to urban development and land use changes in Parsabad city, it can be concluded that: In response to the first question, the study of land use maps of previous periods showed that during the last 26 years, the area of barren lands and surrounding farms of city has been reduced. This amount of reduction has been more to the benefit of urban land use so that the land use area of urban land has increased to278,567 and 820 hectares in 1989 , 2002 and 2015 periods, respectively. In response to the second question of the research, the main level of land use change in the coming years will take place on the outskirts of the southwestern part of the city, including the residential towns of Fajr, Farhangian; And the future development model of Parsabad city is somewhat out of the scattered state and will be done more on the surrounding agricultural and barren lands and without use inside the city, especially in the southwestern part of the city. It was clear that during the last 26 years, the most changes and urban development has taken place in this part of the city. Agricultural lands in this part of the city gradually become to barren land use and eventually changed to urban landuse; the fact that from 1989 to 2015, the area of ​​the city has been increased by 70 hectares only in this part of the city.