CC BY
15
Section 7. Agricultural sciences
Djalilova Gulnora Tulkunovna, Tashkent institute of irrigation and agricultural mechanization engineers Associate Professor, Department of Ecology and water resources management E-mail: gulnora_jalilova@rambler.ru. Gafurova Lazizakhon Akramovna, National University of Uzbekistan, Professor, the Faculty of Biology E-mail: glazizakhon@yandex.ru Sherimbetov Vafabay Khalilullaevich, Art. Lecturer, Department of Soil science and agronomy fundamentals E-mail: vafo_uz@mail.ru
FORMATION OF DATABASES ON PLANT COVER FOR MAP-MAKING OF DEGRADATION PROCESSES USING REMOTE SENSING MATERIAL PROCESSING
Abstract: This manuscript provides material on the advantages ofusing the data of remote sensing in soil degradation studies. The authors have stated that the application of space methods in soil cover mapping avoids considerable labor content of work and the subjectivity of information inherent in traditional methods of research.
Keywords: three-dimensional model of the terrain, remote sensing, land degradation, vegetation calculation index, decoding signs.
At present, soil degradation presents a serious en- achievements of space technology and surveying equip-
vironmental problem; it includes: erosion, salinization, ment made it possible to analyze, map, study and evalu-
dehumification, desertification, pollution, etc. For the ate the territories of various localities. Currentstage of
monitoring of soil condition and creation of cartograph- the development of space and ground facilities makes it
ic database of geo-information systems (GIS), the use possible to acquire images of a particular region several
of materials of remote sensing (MRS) of the Earth is times a day, and modern computing capabilities ensure
considered to be the most economically profitable and high rate of data processing. Also, the main advantage
acceptable solutionThe progress of soil mapping, the of the application of MRS in degradation study is ob-
process ofsoil cover studying, correction and making of taining information in any weather, at any time of the
soil maps cannot be imagined without the use of remote day and of a certain surface layer under investigation.But
sensing materials. Remote sensing is very important and the fact should be noted that the effective use of remote
often indispensable in the Earth exploration. Modern sensing (RS), among other things, is possible if two basic
conditions are met: the first is an essential knowledge of the subject or objects under study and the ability to exactly set the research task. From this follows the second condition - the need to know the properties and features of RS materials, the ability to choose certain types of them for solving specific problems.
Modern state-of-the-art of problem. Remote-sensing has received its modern development thanks to the improvement of aerospace survey methods, the installation of personal stations for receiving space information, and the emergence of geographic information systems. This was preceded by a whole era of remote sensing, which deserves attention from historical and scientific point of view.
The first aerial surveys, for the purpose of soil mapping, were carried out in 1929. The result was a soil map at a scale of 1: 50000. The first aerial survey of the Earth's surface was carried out by cosmonaut G. S. Titov on 6th ofAugust, 1961 from Vostok-2 spaceship [7]. Soyuzpro-gramdeveloped later, as one of the main tasks in the field of space exploration outlined a program of the principle of using manned spaceship for exploring the earth in the interests of various sectors of national economyVi-sual observations, as well as photographing the Earth, various forms and types of terrain relief were continued during the j oint flight of the Soyuz-4 and Soyuz-5 spaceships. Crews of manned spaceships Soyuz-6, Soyuz-7, Soyuz-8 have carried out studies on the reflectivity of forest areas and desert areas. Visual observations and aerial surveys were carried out in different areas of the spectrum. The obtained materials of space survey have found application in various branches of national economy in solution of the whole complex of problems.
The study of the materials of space surveyand their application in thematic mapping was carried out by V. L. Andronnikov [1; 3], A. M. Berlyant [4; 5], E. V. Volkova [6], E. N. Molchanov [10], E. A. Panko-va [11], D. I. Rukhovich [12], U. T. Tadjiev [15; 16], I.Yu. Savin [13], G. T. Jalilova [7; 8] and others. In these studies the fundamentals of remote sensing are discussed in detail, technical parameters and classification of surveying cameras are given according to their purpose. Varioustypes of space surveyhave been comprehensively analyzed, characteristics of space survey images are given as well as the assessment of their informative properties.
According to V. L. Andronnikov [1; 2], the term "remote sensing" was introduced in 1960 by geographer E. SelinPrunt (USA) and is now used in all countries of the world. M. S. Simakova [14] mentioned in her publications that topography and remote sensing materials are of particular importance in mapping soil cover. They give information on the various forms, the dimensions of terrain relief, allow researchers to mark out certain natural boundaries. Types, sizes, forms of relief are usually associated with geological structure of the terrain, the granulo-metric composition and the type ofsoil-forming rocks.The topographic map also contains a characteristic of vegetation in a generalized form, and remote sensing materials contain complex information on the relief, vegetation, and soils. According to E. K. Kurbanbekov and G. T. Sido-renko [9], when mapping soil cover, the advantages inherent in space images determined: the efficiency of information obtaining; high accuracy of contouring of soil differences; the possibility of mapping the components of natural environment, which conditions the scientific effectiveness of space methods of soil research. U. Tadjiev in [15; 16], reveals the advantages of space methods for studying natural environment, including soil cover, in terms ofthe globality, regularity, periodicity and complexity of observations. He notes that an important feature of space image is its efficiency, which makes it possible to obtain data on the state ofvegetation, the top-soil, the development of erosion processes, the saline areas and the state of agricultural crops at the moment. In addition, the application of space methods in top-soil mapping allows to avoid considerable labor content and subjectivity inherent in traditional methods of research [7; 8].
Objects and methods of research. A san object of research, there is an image of the top-soil of vertical zo-nation: dark sierozem soils, mountain brown carbonate, mountain brown typical and mountain brown leached soils located in the western branches of the Chatkal ridge, and also an image of the top-soil of horizontal zonation: irrigated sierozem-meadow, irrigated light si-erozem, irrigated marsh-meadow, irrigated meadow, irrigated meadow-sierozem soils, and also typical virgin sierozem, sierozem-meadow, mead ow soils and mead ow solonchaks (saline lands) prevalent in the territory of the Jizzakh steppe.
Studies on selection of remote sensing materials was carried out according to the generally accepted
methodology of aerospace-geological research "Methodological recommendations for the conduct of space-geological research in Uzbekistan" (1982), Direction on the organization of aerospace-geological research (AKGI) of the Republic of Uzbekistan (2002). Theprocessing of remote sensing images was carried out by software and hardware: Global mapper 17, ENVI, ERDAS, Ar-cGis10 and others.
Research results. To perform this project, it was necessary to select the space images of various types and modesof the territories under investigation. The images of remote sensing from the Earth satellite systems Landsat 7 TM, Terra (Aster), SRTM, SPOT-4 and QUICKBIRD (GOOGLE MAPS) were selected. Each space image of Landsat (Fig. 1. 2), Aster (Fig. 3) and QUICKBIRD (Fig. 4) types is a specific set of archive files provided to the user for further processing. The number of provided files depends on the type of sen-
sor and may reach 17 elements. For example, for Aster -it is14 archived channel images plus a graphic image of survey subject in the form of a graphic file, a metadata file and a snapshot caption file for automated processing. For Landsat 7 ETM + it consists of 12 files.
The Landsat program is the most long-term project for obtaining satellite images ofthe Earth. The first of the satellites under the program was launched in 1972, the last one, to date, Landsat 8 - on 11th of February, 2013. The equipment installed on the Landsat satellites has made billions of images. Pictures taken in the United States and in satellite receiving stations around the world are a unique resource for carrying out a wmultitude ofscientific research in the field of agriculture, cartography, soil science, geology, forestry, etc. As mentioned above, Landsat 7 takes images in 8 spectral bands with a spatial resolution from 15 to 60 meters per pixel. The frequency of data collection for the entire planet was originally 16-18 days.
Landsat 4 (4 chan- Landsat 7 (7chan- Landsat 7 (7chan- Landsat 8 (llchan-
nel-57 meters per pixel) - nel-28 meters per pixel) - nel-28 meters per pixel) - nel-28 meters per pixel) -1977 y. 1989y. 2002y. 2013y.
Figure 1. Space image of LANDSAT of the western branches of the Chatkal ridge
Landsat 4 (4 chan- Landsat 4 (4 chan- Landsat7 (7chan- Landsat8 (llchan-
nel-57 meters per pixel) - nel-57 meters per pixel) - nel-28 meters per pixel) - nel-28 meters per pixel) -1977 y. 1980y. 1989y. 2013y.
Figure 2. Space image of LANDSAT of the Djizzak steppe
Terra (Aster) program - satellite system Terra with optical-electronic sensor ASTER -was launched on December 18, 1999. Data obtained with ASTER sensor are used for a wide range of applied and research tasks due to the unique characteristics of this device: it allows one
to obtain images of earth's surface with a resolution from 15 meters in 14 different spectral channels.The possibility of these images in the infrared (thermal) range makes it possible to analyze the Earth's surface at any time of the day; they are used for geological and topographic
mapping, creating three-dimensional terrain models, studying global changes, analyzing elements of earth's
surface, for hydrological, climatological and soil studies.
Aster (14 channel - 15 meters per pixel) - 2005y. Aster (14 channel - 15 meters per pixel) - 2005y. Chatkal ridge Djizzak steppe
Figure 3.Space image of TERRA satellite system
(Pan-sharpened, high-resolution color image). Images have good decoding properties. If there is a qualitative Digital Model of Relief, they can be considered as an alternative to aerial survey to update the maps scale up to 1: 2000.
Space images QUICKBIRD have the best resolution ofall satellites, presented in remote sensing market. Images are transmitted as panchromatic (0.61m per pixel) and mul-tispectral (2.4m per pixel) ones, and in synthesized form
QuickBird (RGB, GOOGLE-1 meters per pixel) - 2012 y. QuickBird (RGB, GOOGLE-1 meters per pixel) - 2012 y.
Figure 3.Space image of QUICKBIRD satellite system
The processing of remote sensing images of the area under investigationto study land degradation (one should remember that the images were taken in the interval of several years) was started with a linking of the image, i. e., with definition of geographical location of the area reflected on it using a topographic map. Having made the linking, we proceed to our own interpretation of the image. The process of interpretation includes the detection and identification of decoded objects and is carried out according to decoded characteristics.
When decoding satellite images, first of all, the fact that top-soil can be depicted on satellite images - directly - in case of plowing up the territory and - indirectly - through the image of forest, natural grassy and cultivated vegetation, was taken into account. Various
plant associations were distinguished in the images by tone and pattern (texture) of the image. The tone of the vegetation image was significantly influenced by spectral brightness, texture of vegetation and projective cover of soil surface.
Further, a detailed analysis of decoding index of plant cover is given. NDVI (normalized difference vegetation index) is the most famous index, it is easy to calculate, it has the widest dynamic range amongfre-quently used vegetation indices, and the best sensitivity to changes in plant cover in GIS. Calculationformula for the vegetation index is a ratio of the difference between red and infrared channels to the sum of these channels. It is moderately sensitive to changes in soil and atmospheric conditions.
NDVI = (NIR-Red)/(NIR+Red) where NIR, Red - are the reflections in the near infrared and red regions of the spectrum, respectively. NDVI is an excellent indicator for assessing the state of vegetation and is one of the most commonly used indices for solving problems in the quantification ofplant cover. The calculation of NDVI is based on two most stable (not dependent on other factors) regions of spectral curve of the reflection of vascular plants.In the red region of the spectrum (0.6-0.7 ^m) is the maximum absorption of solar radiation by chlorophyll of higher vascular plants, and in the infrared region (0.7-1.0 ^m) is the region of maximum reflection of the cellular structures of the leaf. That is, high photosynthetic activity (associated, as a rule, with dense vegetation) leads to less reflection in the red region of the spectrum and to greater reflection -in the infrared. The use (instead of a simple ratio) of a normalized difference between the minimum and the maximum of reflections increases the accuracy of the measurement, reduces the influence of such phenomena as differences in image illumination, cloudiness, haze, radiation absorption by the atmosphere, etc. [7].
According to the data of QUICKBIRD satellite of the western branches of the Chatkal ridge, a preliminary classification was conducted by six main classes. The classification was made in order to find data that corresponded to any criterion of bare plots and properties. It is known, that classification is a process of sorting out (distribution by classes) of image elements (pixels) into finite number of classes based on their attribute values (DN -digital numbers).If a pixel satisfies a certain classification condition, it refers to a specific class that corresponds to this condition. As a result of using the classification, a set of brightness values corresponding to the input data is obtained, and an image is obtained, an example of which is shown in Fig. 5, the classification table (the so-called signature) is also presented. The signature is a collection of data that determines the brightness values of the pixels of a particular class group.Values in table columns RED, GREEN, BLUE characterize the brightness of pixels in this class on the corresponding channel (red, green, blue). In the course of the study, the main soil-protective features of plant associations are given, as well as their approximate indicators. It was stated that for herbaceous phytocenosis, the main features of soil protection capacity are: the ratio of the projective cover
of the grass and the true turf, the presence in the herbage of plant species belonging to various erosion-resistant groups by the type of root systems, and the current state of phytocenosis.
For woody and shrubby vegetation, the main signs of assessing the soil protection capacity are: the closeness of tree crowns and shrub layers and the state of the forest bedding and soil cover.Brown leached soils located in the most moistened, shady areas with lush vegetation prevail in this area.Vegetation is of meadow-steppe type, with a lot of herbage and shrubs, in some places - juniper, spruce, apple, and walnut trees standing separately. In this area an almost intact vegetative community is observed with good soil protection ability. The highest percentages of the projective cover are characteristic for high turf. High-resistantand medium-resistant species are found in the grass. Communities of this category do not need special measures for restoration (except in minor insolated eroded slopes).The only thingrequired - is a rational use. From the above we can say that the soils distributed on the shadow slopes, with the exception of eroded differences, always have a powerful fine-grained cover, which does not contribute to the formation of surface runoff. Therefore, these plots are the best lands in mountainous areas. In the area where brown typical soils at an altitude from 1300 to 1600 m above sea level are widespread, large-cereal-shrubby vegetation with shrub thickets, archa, deciduous mesophilous vegetation - walnut, apple, etc., is growing.In this category of land, plant communities with insignificant signs of damage are widespread: a reduction in turf percentage (with sufficiently high projective cover) and the introduction of low-erosion-resistant species. There is also a weak evidence of erosion processes (with the exception of eroded insolated slopes). These lands require partial surface improvement with usage regulation. Vegetation of this category requires partial reforestation measures aimed at increasing the closeness of crowns and the restoration of wood bedding, as well as favorable conditions for natural renewal of woody vegetation.
In the area where brown carbonate soils prevail in the lower part of the belt of mountain brown soils and at an altitude of 900-1300 meters above sea level, a decrease in the value of all the indicators is characteristic: further lowering in the percentage of projective cover and turf, mainly medium and weakly erosion-resistant
species in the grass stand. This calls for continuous surface improvement of plant cover with usage regulation.
Plant communities need protection, continuous reforestation measures and promoting natural regeneration.
Figure 5. Map of calculating the vegetation index NDVI of the territory of the Western branches of the Chatkal ridge (by zones)
Clustering of data on the territory of the Jizzakh steppe began with arbitrarily given values (average) or mean values taken from existing signatures. After assigning all possible pixels to one of the classes, the class centers shifted, and the process wascompletely repeated (the next interpretation). The process continues until the maximum number of interpretations is reached or the maximum percentage of pixels that have not changed their class is reached (convergence threshold). Landsat images obtained served as the main source of high-resolution images for detailed studies ofplant cover change. After decoding the image, it could be definitely said that, for example, the vegetation of the territory under study was undeveloped in 1980, as the NDVI index of territory vegetation was 25-30%. In 1989, with the development of this territory, this value rose to 87%, which is due to soil melioration during this period.In 2000, the NDVI index of vegetation dropped to 78%, and in 2009 to 73% due to soil salinity; so, the rate of veg-
etation degradation grows with deterioration of agro-physical, agrochemical properties and melioration state of soils in area under investigation. High resolution images show many features, such as different conditions of vegetation, barren areas formed as a result of cattle overgrazing.
Besides natural conditions that determine the nature of the plant groupings of the described area, economic activity of a man has a great impact; it changes the natural relationships of plant groupings due to centuries-old use of the territory as pastures, felling perennial plants for fuel, and plowing for growing grain crops. Thus, the degradation rates of plant cover of the area under study are determined by the following criteria: degradation of steppe lands (increase in degraded zones), reduction of fodder reserves in the area under study, deforestation of tree and shrub plants.
Figure 6. Map of calculatingthe vegetation index NDVI of the territory of the Jizzakh steppe
Proceeding from the above, it can be said that the creation of databases on plantcover with the use of processing remote sensingmaterials on the basis of special software ofgeoinformation systems provides a detailed spatial assessment of plant cover in various scales, including an automatic mode, and this, in its turn, will allow more correctly identify the main damagefocusesin degradation processes.
The obtained materials on areas under study after the processing of remote sensing images based on GIS
technologies can provide information on the actual state of the region's lands, top-soil cover, land degradation risks; this information makes possible to developthe algorithms for analyzing the suitability of lands for agricultural crops, the algorithms for assessing degradation risks, and to develop the technology of results optimization and assessments in the form of a series of optimal ecologically and economically sound scenarios for the allocation oflands and agricultural crops, forconducting soil-protective measures.
References:
1. Andronnikov V. L. Use of multi-zonal space images for studying the soil cover: Journal of Soil Science, Nauka, No. 1. - Moscow, - 1979. - P. 41-49.
2. Andronnikov V. L. Aerospace methods of soils studying: Kolos, - Moscow, - 1979. - 280 p.
3. Andronnikov V. L. Methods of decoding the soil cover of forest-steppe territories by aerial photographs: Thesis. cand. of geol. - min.sci. - Moscow, - 1958. - 17 p.
4. Berlyant A. M. Informational mapping: Science, - Moscow, - 2007. - 62 p.
5. Berlyant A. M. Geographical forecast based on the interpretation of cartographic and aerospace methods: Journal of the Earth study fromSpace, Nauka, Russian Academy of Sciences, - Moscow, - No. 2. - 1986. - P. 30-35.
6. Volkova E. V. Use of remote sensing data in agriculture: "Agrarian Science for Agriculture" IX International Scientific and Practical Conference, Barnaul, - 2014. - P. 413-415.
7. Gafurova L. A., Djalilova G. T. Modern approach to the study of erosion-hazardous lands in the Sukoksai basin using GIS technology: Fan va technology. - Tashkent, - 2017. - 144 p.
8. Gafurova L.A., Djalilova G. T. Creation of a digital model of terrain using GIS technologies (on the example of erosion-prone lands of the Sukoksai basin): IV International Scientific and Practical Conf. Agrarian Science forAgriculture, Barnaul, - 2007. - P. 103-105.
9. Kurbanbakov E. K., Sidorenko G. T. Use of space data in agricultural mapping: Interdepartmental transactions. Exploring the Earth's natural resources using space data, - Moscow, - 1985.
10. Molchanov E. N. Use of remote sensing methods is the main way to improve the mapping of soils in mountain areas: Experience in land management problem and the implementation offood supply program in the USSR. Proc. of Scientific and technical conference, - Kishenev, - 1985.
11. Pankova E. I., Rukhovich D. I. Remote sensing monitoring of salinization of irrigated soils of arid territories: Journal of Soil Science, - No. 2. - 1999. - P. 253-263.
12. Pankova E. I., Soloviev D. A. Salinization monitoring of irrigated soils of the Hungry Steppe according to remote sensing data: In: Modern Natural and Anthropogenic Processes in Soils and Geosystems. V. V. Dokuchaev Soil Institute, - Moscow, - 2006. - 369 p.
13. Savin I.Yu., Kiryanova E.Yu. On the possibilities of assessing the contrast of soil cover of the Saratov Volga region withLandsat satellite images: Digital soil cartography: theoretical and experimental studies, - Moscow, - 2012. -P. 189-209.
14. Simakova M. S., SavinI.Yu. Use of materials of aerial and space imagery in soil mapping (ways of development, state-of-the-art, problems): Journal of Soil Science, - No. 11. - 1998. - P. 1339-1347.
15. Tadjiev U., Namozov Kh. K. Mapping of soils based on the use of space imagery materials: FAN Publishing House, - Tashkent, - 2003.
16. Tajiev U. T., Namazov Kh. K., Akhmedov A. U., Zhuraev G. A. Mapping of soil salinization using space images: Problems of land management in modern conditions, Gorki, - 2004. - 285 p.
