Statistical Analysis Of Lineaments Using Landsat 8 Data: A Case Study Of The Fergana Valley (East Uzbekistan)

This paper describes a statistical analysis of lineaments based on multi-temporal Landsat 8 imageries in the Fergana Valley (East Uzbekistan). The results of the statistical analysis showed that determined that the count of lineament structures changes by months. It was also noted that the Namangan region is more prone to the manifestation of lineament structures. The maximum count of lineament structures was in July. And in November, we observe a sharp decrease in lineament structures. According to the results of the rose diagrams, various orientations are observed for these months. There is a coincidence of directions.


INTRODUCTION
In the developed countries of the world with the intensive development of remote sensing, to pay attention to the detection and analysis of lineament structures using modern satellite data.Usually, visual interpretation of satellite data to identify linear structures on the Earth's surface is a more complex process, which often requires a significant investment of time and effort.While an automated process avoids such difficulties, it has high efficiency, costeffectiveness, and especially objectivity.

OCLC -
Currently, the leading scientists of the world have carried out large significances of research on the study of modern movements of the earth's crust.In most of the works, the authors used modern satellite data, along with software for automatic extraction of lineament structures.Shevyrev S.L. (2017) in his work describes an algorithm for structural-tectonic analysis of remote sensing data.He divides the analysis of satellite data into several stages of processing: generalization, taking into accounts the scale of the analyzed objects with the creation of a series of images of different spatial resolution -"pyramids"; analysis of the "pyramids" of the image; application of graphic filters [1].Such stages of data processing using Landsat and SRTM DEM made it possible to confirm the localization of ore objects within the fractured permeable areas of the territory and to clarify the position and severity of regional fracture structures.In many scientific works, a comparative analysis of optical satellite data and radar satellite data is carried out for automatic lineament extraction.The basis of the automatic approach involves determining the optimal parameters for automatic lineament extraction with a combination of edge detection and line-linking algorithms and determining the appropriate bands based on optical data suitable for lineament mapping.To verify the results obtained, the extracted lineament structures are compared with the lineaments obtained by hand digitizing from geological maps [2].The work presents a processing method using multi-temporal Landsat 8 imageries were determined on base LINE algorithm in software PCI Geomatica in the territory of the water reservoir [3].The result showed that water level fluctuations have a greater influence on the appearance of the lineaments structure than periods of water filling and downstream in the reservoir.
This paper describes a statistical analysis of lineaments based on multi-temporal Landsat 8 imageries in the Fergana Valley (East Uzbekistan).

STUDY AREA
The Fergana Depression is the largest intermountain depression in Central Asia and is located in the eastern territory of the Republic of Uzbekistan and borders on the Republic of Tajikistan and Kyrgyzstan.
Several large geological faults pass in the Fergana Valley: Tamdi-Karachatir, the approximate length of the fault is 1089.72 km, Besapano-South Fergana, the approximate length of the fault is 1082.19km, the North Fergana, the approximate length of the fault is 393.52 km, Kumbel the approximate length of the fault is 352.31 km, Andijan the approximate length of the fault is 255.49km and Kenkol-Arashan, the approximate length of the fault is 148.92 km.
In figure 1

MATERIALS AND METHODS
The first step in the automated process of extracting lineament structures is to select the original data.The main criterion for choosing a suitable image was cloud-free.Data processing was carried out for the period May -November 2019.The Landsat 8 has a spatial resolution of 30 m; to improve the quality of lineament structures extraction, it is necessary to improve the spatial resolution of the images.
To do this, it is necessary to carry out pan sharpening for each satellite image.Automatic interpretation of lineament structures was carried out by Principal Component Analysis (PCA).PCA was performed using six Landsat 8 spectral channels (channels 1, 2, 3, 4, 5, 7).Bands 6 (thermal) were not used in the study.Analysis of these components showed that the most informative components from the point of view of displaying geological structures are the sixth, fifth and third components with a standard deviation of 1.5.The automated process of extracting lineament structures was performed in the LEFA (Lineament Extraction and Fracture Analysis) program in the Matlab environment.

RESULTS AND DISCUSSION
In figure 2

OCLC -
The results of statistical analysis showed that the maximum count of lineament structures was in July; their count was 654 with a maximum length of 3.5 km and a minimum length of 1.05 km.Then there is a gradual decrease in the count of lineament structures.
In August, the count of lineaments was 537 with a maximum length of 3.09 km and a minimum length of 1.05 km.In September, the count of lineaments was 435 with a maximum length of 4.6 km and a minimum length of 1.05 km.In October, the count of lineaments was 209 with a maximum length of 3.3 km and a minimum length of 1.05 km.And in November, we observe a sharp decrease in lineament structures; their count was 97 with a maximum length of 3.8 km and a minimum length of 1.05 km.But in May and June, the opposite dynamics was observed, if in the second half of the year there was a gradual decrease in lineament structures, then the opposite is true.In May, the count of lineaments was 504 with a maximum length of 4.03 km and a minimum length of 0.15 km.In June, the count of lineaments was 386 with a maximum length of 3.4 km and a minimum length of 1.05 km.
The analysis of the orientation of lineament structures is carried out.As a result, it was revealed that in May, June, and September the main direction of the lineament structures was the north and south.Also in May, the western and eastern directions are observed, which disappear in June, but their appearance is observed in July, August, and September, and then these directions decrease in October, and in November they practically disappear.In May, the direction is north-west and southeast, which in the following months is the main direction -July, August, and October.In June, there is an almost mirror direction northwest, southeast, east-northeast, and westsouthwest.

Figure 1 .
Figure 1.Fergana Valley according to the Landsat 8 satellite with geological faults