A STUDY ON VARIATIONS OF SHORELINE CHANGES AND TEMPORAL-SPATIAL POTENTIALITY FOR CLOUD SEEDING AT URUMIA LAKE
Keywords: Shoreline Changes, Neural Network, Urmia Lake, Landsat Satellite, Temporal- Spatial Potentiality Evaluation
Abstract. Protecting the living environment has become one of the greatest human concerns; sudden increase of population, industry and technology development, unrestrained over consumption of the citizens and climate changes, have all caused many problems for mankind. Shores are special zones that are in contact with three Atmosphere, Hydrosphere and Lithosphere environments of earth. Shore lines are of the most important linear features of the earth’s surface which have an animated and alive nature. In this regard, optimized management of the shores and environmental protection for stable development require observing the shorelines and their variations. Protection of shorelines within appropriate time periods is of high importance for the purpose of optimized management of the shores. The twenty first century has been called the era of information explosion. A time that, through benefits of new technologies, information experts attempt to generate more and up to date information in various fields and to provide them for managers and decision makers in order to be considered for future planning and also to assist the planners to arrange and set a comprehensive plan.
Aerial images and remote sensing technology are economic methods to acquire the required data. Such methods are free from common time and place limitations in survey based methods. Among remote sensing data, the ones acquired from optical images have several benefits which include low cost, interpretation simplicity and ease of access. That is why most of the researches concerning extraction of shorelines are practiced using optical images. On the other hand, wide range coverage of satellite images along with rapid access to them caused these images to be used extensively for extracting the shorelines.
The attempt in this research is to use satellite images and their application in order to study variations of the shorelines. Thus, for this purpose, Landsat satellite images from TM and ETM+ sensors in the 35 time period has been used. In order to reach better results, images from MODIS satellite has been used as auxiliary data for the images that are with an error margin. Initial classification on the images was conducted to distinguish water and non water applications. Neural network classification was applied with specific scales on the images and the two major applications were thereby extracted. Then, in order to authenticate the proceedings, Error matrix and Kappa coefficient has been applied on the classified images. Base pixel method of neural network was used for the purpose of information extraction while authenticity of that was evaluated too. The outcomes display the trend of Urmia shoreline has been approximately constant between the years of 1976 to 1995 and has experienced very low variations. In 1998 the lake experienced increase of water and therefore advancement of the shoreline of the lake due to increase of precipitation and the volume of inflowing water to the basin. During 2000 to 20125, however, the lake’s shoreline has experienced a downward trend, which was intensified in 2007 and reached to its most critical level ever since, that is decreasing to about one third.
Further, temporal and spatial potentiality evaluation of clouds seeding in Urmia lake zone has been studied as a solution for improvement and recovery of the current status of the lake, and an algorithm was proposed for optimized temporal- spatial study on could seeding. Ecological, meteorological and synoptic data were used for timing study of the cloud’s seeding plan, which upon study; it is easy to evaluate precipitation potential and quality of the system. At the next step, the rate of humidity and also stability of the precipitating system can be analyzed using radar acquired data. Whereas extracted date from MODIS images are expressing the spatial position, therefore in order to study the location of the cloud’s seeding, MODIS images of the selected time intervals along with applying MCM algorithm were used to conclude thick clouds. Also, with interpolation of the TRMM data, it is possible to deduce maximum precipitation in the form of spatial arena. One of the data categories that is used both for temporal and spatial analysis is radar images which in addition to time, displays the existing humidity range, movement direction, and positions of accumulated precipitation cores. Therefore, using this algorithm, it is possible to conclude the most optimized spatial position in order to execute the seeding plan.