Font Size: a A A

Study On The Space-time Changes And Eco-economic Effects Of Cropland In The Yellow River Irrigation System Of Zhongwei, Ningxia Autonomous Region

Posted on:2013-02-02Degree:MasterType:Thesis
Country:ChinaCandidate:S F ZhangFull Text:PDF
GTID:2249330371986642Subject:Cartography and Geographic Information System
Abstract/Summary:PDF Full Text Request
Farmland is a kind of important resources and material basis to support human’s survival and development and guaranteed maintain production and access food by human. Farmland change is a core study domain of land use and land cover changes. The change and evolution of farmland reflect basic trends of the social and economic development.The Yellow River Irrigation System of Zhongwei originates at Shapotou hydro-junction port. It consists of the two parts:Zhongwei oasis in Ningxia Autonomous Region and Luanjingtan in Inner Mongolia Autonomous Region. In the study, the author extracted and detected space-time changes of cropland during1992and2011using Landsat TM/ETM+images and object-based image analysis method. This article consists of three parts:(1) image segmentation and segmentation quality evaluation;(2) the establishment of the classification rule organization and information extraction;(3) space-time changes detecyion and eco-economic effects.Image segmentation is a process of spliting the image into image objects of internal homogeneity and neighborhood heterogeneity Combine the spectrum, texture and shape feature. It is the first step in object-based image analysis, the image object is the basis of object-based image analysis, and the image segmentation quality directly determines the next image classification and information extraction accuracy. In the study, the author using vector-raster overlay analysis method to evaluate the quality of the segmentation results based on the Python scripting language call Arcpy module, and selects the best sub-parameters on image segmentation. In the paper, using six scale factor (for instance5,10,15,20,25,30), the range of shape factor and compactness factor is set to [0.1,0.9] and step0.1to Spliting image by eCognition software. And then use the Python scripting language call Arcpy module to generate a script to batch processing of partitioning the data set, the segmentation quality evaluation using PSE(Potential Segmentation Error)、NSR (Number-of-Segments Ratio) as well as ED2(Euclidean Distance2) proposed by Liu(2012). Ultimately, choice the optimal segmentation parameters set by ED2.In the study, the author extracted cropland using classification rule organization based on the membership function. First select the samples, and statistical analysis the spectrum, shape, texture feature, the relationship between classes, as well as the custom features such as vegetation indices and water index of the samples to determine the remote sensing image information extraction mechanism of the surface features of the study area. Using the water index extract the planting of the paddy fields and use vegetation indices to distinguish it with water; Vegetation index can be use to extracted cropland and the lush green of rice fields, but can not distinction with the artificial forest, requires a combination of spectral, shape and texture feature to distinguish cropland from artificial forest.Finally, take the overlay analysis of the classification results and analysis it’s eco-economic effects, take the following conclusions:That the cropland areas rapid growth of Luanjingtan during1992-1999, an increase of nearly6000hectares, mainly due to the Luanjingtan Pumping Irrigation Project is completed in1993reclaimed a lot of cropland, after1999growth leveled off, but continued to increase in the study period. The Luanjingtan cropland increase parts of all converted from the desert shrub, due to rational planning before construction, residential occupation of cultivated land less. The cropland area of Zhongwei continued to increase and the growth rate is basically the same between1992and2006, but in2006-2010fell by nearly3000hectares; the most important reason is a lot of arable land into facilities of agricultural land as well as the urban expansion occupying too much cropland. The most of increase parts of Zhongwei cropland converted from the desert shrub; small parts due to the standardization embankment project of the Yellow River converted from the old river channels and beaches. The occupation of cropland mainly converted to residential areas and facilities of agricultural land, a small part is converted to artificial forest and fish ponds.
Keywords/Search Tags:object-based image analysis (OBIA), Image Segmentation, Rulesset, Farmland, Yellow River Irrigation System, Zhongwei, Luanjingtan
PDF Full Text Request
Related items