mohammad hedayatfard; Hamid Gholami; Seyed Masoud Soleimanpour; Arashk Holisaz
Abstract
Gully erosion is one of the most important phenomenons in land degradation and desertification, which upsets the balance of environmental resources and its sustainability. The aim of this study was to determine the effective factors on creation of gully erosion using data mining methods in Bayan Watershed ...
Read More
Gully erosion is one of the most important phenomenons in land degradation and desertification, which upsets the balance of environmental resources and its sustainability. The aim of this study was to determine the effective factors on creation of gully erosion using data mining methods in Bayan Watershed in the north of Fars Province. K-Means clustering algorithms and CART decision tree were used to model gully erosion. The estimated accuracy based on the parameters affecting the creation area in the final model is 100% for the educational data set and 71.43% for the experimental data set. Results from measuring different variables from 40 selected gullies, including a target variable (creation area) and twelve predictive variables, like creation slope, vegetation, bare soil, surface gravel, surface litter, EC, pH, SAR, OM, clay, silt and sand in field and laboratory operations and the use of data mining techniques, showed that the creation of gullies in this watershed is a function of silt, acidity, bare soil, gravel and organic matter. These results showed the effect of vegetation and consequently, the presence of adequate organic matter in the soil as the most appropriate method for controlling gullies and the destructive and intensifying effect of bare lands without vegetation in intensifying gully erosion. Therefore, it is necessary to establish and establish permanent and compatible vegetation, to stabilize and improve soil structure and increase organic matter, and control and management programs, in lands with more silt, to be operated more carefully.
Alireza Arabameri; Mohammad hossein Ramshet; Khalil Rezaei; Masoud Sohrabi
Abstract
Nowadays, desertification as a great problem affects most of the countries in the world, especially developing countries. Desertification phenomenon that occurs in arid, semi-arid and dry semi-wet regions, will reduce the land potential. In this study MEDALUS model was applied for quantitative assessment ...
Read More
Nowadays, desertification as a great problem affects most of the countries in the world, especially developing countries. Desertification phenomenon that occurs in arid, semi-arid and dry semi-wet regions, will reduce the land potential. In this study MEDALUS model was applied for quantitative assessment of desertification risk with emphasis on five criterion in Shahroud-Bastam Basin, because of easy style and data accessibility and also compatible with the effective indices on land degradation. For this purpose, first, the study area was divided into five homogeneous units. Then, five criteria including soil quality, climate, vegetation, management, erosion and 13 indicators were used for quantitative evaluation of desertification in homogeneous units. Index layers for each criterion were prepared using GIS. These indices were ranked in accordance with Modified MEDALUS model. The geometric mean was then calculated and map was produced for each criterion. Land degradation map of each homogeneous unit was prepared using the geometric mean criteria. Finally, from their combination, desertification susceptibility map produced. According to the results, climatic and erosion criteria with an average weight of 1.607 and 1.467, had the most effectiveness on desertification and vegetation with average weight of 1.377 had the least effectiveness. As a results, from the total of study area, with area of 7737.71 km2, 40% (3096.311 km) is located in the C3 class or extreme critical erosion, that requiring more attention to implement desertification control programs. Result of this research is a managing tool available for decision-making regarding the selection of priority areas in the fight against desertification.
Abstract
Homogeneous sub watershed Are useful for comparison and better understanding of the complex structure of watershed and thereby facilitating the development and management approaches specific to each class to deal with the destructive phenomena such as soil erosion, drought and flood. In this study, the ...
Read More
Homogeneous sub watershed Are useful for comparison and better understanding of the complex structure of watershed and thereby facilitating the development and management approaches specific to each class to deal with the destructive phenomena such as soil erosion, drought and flood. In this study, the database contains 6 groups of information such as climatic and socio-economic, topography, soil, water, remote sensing were studied.These Factors converted to raster and then the pixel data converted to text data (ASCII). So matrix was produced with dimensions of 27 * 147 506. Eigenvalues and the variance associated with each factor showed that 6 first component includes a total of 83.27% of the changes. In order to select the optimal number of clusters and clustering evaluation and validation of the method, hierarchical clustering and discriminant analysis and clustering using k- mean cluster of the data was performed. Karoon Sub watersheds were classified into 16 groups.
