EN中文

Archive

  • 2009 Volume 11 Issue 1
    Published: 25 February 2009
      
    ARTICLES
  • Select all
    |
    ARTICLES
  • ARTICLES
    Research on Earth System Scientific Data Sharing Platform Based on SOA
    ZHU Yunqiang, FENG Min, SONG Jia, LIU Runda
    2009, 11(1): 1-9.
    Download PDF ( )   Knowledge map   Save
    In order to integrate distributed and heterogeneous earth system scientific data resources and provide one-stop data sharing services for different users,a distributed data sharing platform is needed urgently in China. In this study,we analyzed interoperability requirements of the distributed earth system scientific data sharing plat- form(ESSDSP)in the first place.After compared with several architectures and technologies of distributed inter- operability,we chose SOA(service-oriented architecture)as the architecture of ESSDSP(hereafter called SOA4GS,SOA for Geo-data sharing).The basic principle of SOA4GS was to abstract a series of web services from the whole data sharing activities which included information synchronization service of data centers,distributed metadata synchronization management service and multi-resource data access service.Through deploying these web services in different data sharing nodes,the software can easily implement the interoperability among them.Finally we introduced the design,implementation and deployment of these web services in detail.SOA4GS has the following characteristics:(1)Distributed nodes with uniform logic ESSDSP are comprised of many centers which are deployed in different organizations at different locations.They have the uniform business logic to ensure they can communicate each other by web services.(2)With open and extensible architecture based on SOA,ESSDSP is an incompact architecture.So other systems based on web can be easily integrated into the platform after little function model revision for compliance with the interface of web services,while,on the other hand,systems such as e-government and e-commerce can use data sharing functions by calling web services to share the data provided by the platform.(3)Easy to use Web service platform is a web software module.Its client code can be created automatically from the web service' address by using some development tools such as Eclipse,JBuilder.So users can easily call the platform's web services or revise their system's functions.In practice,ESSDSP has been used in Earth System Scientific Data Sharing Network,one of National Infrastructure and Facility Development Environment Building for Science and Technology Industries Program of China.It comprises one main center,one authentication center and sub-centers.So far,it has integrated 10.23 T Byte data resources and Byte data resources have been downloaded by 34,597 users.
  • ARTICLES
    Review of Spatial Data Mining Methods Based on Concept Analysis
    QIN Kun, LI Zhenyu, DU Yi
    2009, 11(1): 10-17.
    Download PDF ( )   Knowledge map   Save
    The process of spatial data mining is closely related to the process of spatial concept formation and analysis. From the view of recognition,centered in the thought to extract the concepts in different hierarchies and granularities, based on the formal theories of concept formation and analysis,i.e.concept lattice and cloud model, this study analyzes and summarizes the methods and their application of spatial data mining based on concept analysis, including the methods and their application of spatial association rule mining based on concept lattice,those of conceptual clustering based on concept lattice,those of spatial association rules mining based on cloud model, those of spatial clustering/classification based on cloud model,etc.Concept lattice provides a kind of theory which can express and analyze spatial concepts and their concept structural relationships by mathematical and formal methods,which is good for the understanding of human,and is convenient to process by computers;Cloud model is a kind of uncertainty theory which can deal with fuzziness,randomness and their association,it is a more powerful and effective uncertainty theory than fuzzy sets and rough sets.The further research directions of spatial data mining based on concept analysis are as follows:(1)the formal expression of spatial concepts;(2)the formal analysis of the process of spatial concepts formation;(3)the formal expression and analysis of the structural relationships of spatial concepts;(4)the construction of the spatial concepts hierarchy and their formal expression and analysis;(5)the expression and analysis of spatial concepts and the concept structural relationships with uncertainty .
  • ARTICLES
    Real-time Copyright Protection for Spatial Data Files
    LI Anbo, LV Guonian, ZHOU Liangchen, LIN Bingxian, GU Zhu
    2009, 11(1): 18-23.
    Download PDF ( )   Knowledge map   Save
    Geospatial data is a kind of strategic information resource and is widely used in economic,social and environmental applications.GIS vector data is closely related to the overall situation of socio-economic development and national security due to its high production cost and high precisions.In recent years,researches about copyright protection of GIS vector data mainly focus on copyright marking techniques.To meet the requirements of copyright protection for spatial data files,a real-time copyright protection scheme for spatial data files is proposed in this study,based on the Copyright Marking technology,especially on the Digital Watermarking technology which was only used for the authentication,file system filter driver technology and encryption technology,which can be used for dynamic encryption and decryption of data files and real-time detection of copyright marking.This scheme achieves the purposes to keep confidential,detect copyright marking in a real-time way and control access.In addition, it lowers the limitations on the capacity of copyright marking information and the robustness of embedding algorithms.
  • ARTICLES
    A Study on Meteorological Metadata Catalogue Service System
    WANG Guofu, SUN Chao, HE Xiaoming, ZHU Jiang, CHEN Xuejun
    2009, 11(1): 24-29.
    Download PDF ( )   Knowledge map   Save
    Scientific data sharing system shall provide the functions of uniform data access in addition to metadata query and navigation.Metadata Catalogue Service(MCS)is the key component of meteorological data sharing service system for the purpose of scientific data discovery,access and management.Based on the analysis of user' s abundant requirements,we develop meteorological metadata Catalogue service system,which consists of metadata gateway,metadata server,Catalogue service,and metadata database.A meteorological Catalogue service system is built in this study based on the characteristics of meteorological data sharing service and implements the functions of distributed data management,consolidated Catalogue service and data access,unified security authentication and distributed authorization.MCS enhances the means of meteorological data sharing and improves the quality of on-line data service.MCS technology can also be applied in meteorological information management system and climate change database system in the future.
  • ARTICLES
    An XML-based Solution of Customizing and Extending of APDM
    WANG Shuang, WAN Qing
    2009, 11(1): 30-35.
    Download PDF ( )   Knowledge map   Save
    APDM is the pipeline data model based on Geodatabase technology issued by ESRI.The latest version is 4.0.In the entire pipeline integrity management system,APDM can effectively store and manage pipeline data. To successfully implement the digital pipeline's integrity management,the work of model maintenance and management is most important.There are two key problems in the management of APDM model,constructing and extending of the model.Before building pipeline Geodatabase,pipeline companies must construct APDM model according to their actual needs,which is model constructing work.When needs change,a mechanism is provided to ensure model modifying and extending.Considering the features of APDM model,this study proposes an XML-based method to realize the constructing and extending.First,the design principles,characteristics,history and conceptual model of APDM model are introduced,then detail descriptions on how the XML-based method works are provided, and its advantages are discussed.At last the APDM model management system is designed and implemented by visual studio.NET 2005 and ArcGIS Engine interface.By APDM model management system,people can quickly build and modify APDM model in visual user-interface provided by the system.The solution is proved feasible with a case study.
  • ARTICLES
    The Impact of Elevation Interpolation on the Accuracy of Gradient and Aspect from DEMs
    JIA Yini, TANG Guoan, LIU Xuejun
    2009, 11(1): 36-42.
    Download PDF ( )   Knowledge map   Save
    DEMs and the terrain features derivation accuracy are usually affected by many factors such as sample points density and distribution,DEM form structure and interpolation methods.Since only the elevation of the sample point can be obtained,interpolation will be used to estimate the simulative value of the other points.In this study,four interpolation methods,IDW interpolation,Spline interpolation,Kriging interpolation and TIN interpolation, are evaluated on a Gauss synthesized surface with peak and valley just similar as a real surface,by comparison with the true values of gradient and aspect.Interpolation result is more sensitive to the errors when gradient is less.The uncertainty of gradient,in the four methods,is the least for SPLINE interpolation,and increases in turn of KRIGING,IDW,and TIN interpolations.And the uncertainty of aspect in the four methods,as that of gradient, is the least for SPLINE interpolation,and increases in turn of KRIGING,TIN,and IDW interpolations.The results provide information for the selection of interpolation algorithms.
  • ARTICLES
    Slope Accuracy Associated with DEM Data Error and Terrain Representation Error
    JIA Dunxin, TANG Guoan, WANG Chun
    2009, 11(1): 43-49.
    Download PDF ( )   Knowledge map   Save
    DEM data error is different from DEM's terrain representation error.This paper focuses on the relationship among DEM data error,terrain representation error and slope accuracy based on high resolution and high accuracy DEM.The result shows that(1)data error reduces with the increase of the cell size,while terrain representation error rises rapidly;(2)slope error firstly reduces and then increases.The lowest point of slope error's trend is an important point.At this point,the corresponding cell size is the most suitable cell size of DEM.The cell size is not only related to the DEM data error,but also closely related to the terrain complexity.This paper makes them clear about the source and characteristics of slope error derived from DEM.
  • ARTICLES
    A New Algorithm to Process Depressions Based on DEM Model
    YU Miao, REN Liliang
    2009, 11(1): 50-55.
    Download PDF ( )   Knowledge map   Save
    A new method to process depressions in DEM data is introduced.It is based on a novel concept of spill elevation and integrating the priority queue data structure into the least-cost search algorithm progressively to determine the spill elevation of every grid.The final DEM data after such a processing procedure are the elevation data without depressions.The algorithm is finished with the support of Standard Template Library.It is not only simple to understand but also faster than the other algorithm in processing the different sizes and resolutions of DEMs and has a good drainage extracted from DEM based on the algorithm.
  • ARTICLES
    Design and Implementation of Antarctic Grove Mountains Nature Reserve Information System
    CHEN Nengcheng, CHEN Genhan, E Dongcheng, AI Songtao, YU Jie
    2009, 11(1): 56-61.
    Download PDF ( )   Knowledge map   Save
    In recent years,the Antarctic scientific expedition and research has become the focus of the world. Grove Mountains which locate in South Antarctic could help to resolve many important scientific problems because of their natural and unique geological landforms.Therefore,in order to protect particular scientific,aesthetic and wildness value in this area,our country has proposed to establish Antarctic Grove Mountains Nature Reserve.How to manage the information resource in this area,however,has become a serious problem.First,this paper analyzes the current situation of Grove Mountains Nature Reserve,such as the physical environment,research value and management activities and goals.Then,based on this situation,requirements analysis has been done from data, function and performance aspect,and related GIS application technologies have been discussed.Furthermore,the overall design of this system is introduced.Dual mode(C/S mode and B/S mode)and three-tier architecture(data tier,application tier and presentation tier)are utilized to construct the system.The functions was divided into five parts:data input,operation,edit,output and distribution.Vector data,image data and thematic data have been imported and stored in GeoDatabase using ArcGIS toolbox.The vector data consists of rock,ice,water bodies,water courses,sub-glacial lakes,mountains,glaciers,ice shelfs,coast line,contour line,station,place name,geodetic control point,and the vector data.The large scale vector data can be downloaded from UK Antarctica DigitalDatabase(ADD).The image data includes different observation phase TM and SPOT image.The thematic data includes coastline type,ice shelf,watercourse,human,continental Antarctica and management zones. The human layer includes building,tanks,wharves,landing area,refuges,and masts.The management zones include station limits,recreational limits,no go zones,and magnetic zones.The system shows that it is an effective approach for diverse Antarctica Grove Mountain's spatial related data and is useful for manager decision and data information sharing.
  • ARTICLES
    The Research of City Emergency Management Model Based on GIS Technology——Guangzhou Higher Education Mega Center as a Case Study
    CHEN Yingbiao, CHEN Jianfei, SU Qixin
    2009, 11(1): 62-69.
    Download PDF ( )   Knowledge map   Save
    With rapid development of city construction,emergency events that endanger national security and people's safety occurred on occasion.It is necessary to establish an emergency management system and comprehensive emergency information system.Pipe analysis is a main application of GIS in the City Emergency Management Model.We can manage urban underground pipeline network and store information into computer orderly for data updating and resource sharing.The pipeline data in Guangzhou higher education mega center are used to demon- strate the application.A city emergency management model is built in the study by including three analysis modules (section analysis,vertical distance analysis,and analysis of pipe burst),and Guangzhou higher education mega center is selected as a study area for the three analyses by the visualization platform,By studying the Guangzhou higher education mega center pipeline network the authors analyze the three modules on the scope of application at the same time.The administrators of the pipe management can also make analysis of pipe burst,which is the most important spatial analysis function in pipe network analysis.The three modules of the emergency management model can be applied in pipe design,construction and service.
  • ARTICLES
    Evaluated Result Management Information System of Standard Land-Value in Cities and Towns
    LIU Yanhua, CHEN Weiqing, WU Xiangbin
    2009, 11(1): 70-76.
    Download PDF ( )   Knowledge map   Save
    The evaluated results of standard land-value in cities and towns are managed presently with AutoCAD and Excel individually,maps with AutoCAD and tables with Excel.This kind of management lowers data renewal efficiency,brings inconsistency between maps and tables,and makes data analysis and bidirectional inquiry between maps and tables difficult.On the basis of management present situation and open questions,the study introduces the GIS technology by selecting MapInfo Geographic Information System software platform,and establishes standard land-value evaluated results management information system by the way of component development which integrates the evaluated results maps,tables and documents.The system,based on map and attribute database, includes four sub-systems such as document management,map management,table management and GIS topic analysis. The output results are given by five modules such as map operation,attribute management,spatial analysis, information inquiry and output mapping in its function.The system applications demonstrate that it is convenient for map locating,attribute analyzing,historical data reproducing and thematic mapping by using GIS to man- age standard land-value evaluated result.At the same time,it can guarantee the displaying and renewal consistency between maps and tables,and make data maintenance and analysis easier.So the system ean improve the management of standard land-value evaluated results.
  • ARTICLES
    Surface Parameters Retrieval from Alluvial Fan in Ejina Area of Inner Mongolia Using Multi-polarization SAR Data
    LIAO Jingjuan, PANG Zizhen
    2009, 11(1): 77-83.
    Download PDF ( )   Knowledge map   Save
    The surface parameters of alluvial fan consist of roughnesss and moisture.It is very important to show the characteristics and formation of alluvial fan using these parameters.Synthetic aperture radar(SAR)is sensitive to surface texture and roughness,so it is a power tool to retrieve the surface parameters of alluvial fan.This paper presents the method of retrieving surface parameters of alluvial fan using multi-polarization SAR data based on Genetic Algorithm combined with backscattering mode.The comparison of simulated results and field measurements shows that the method is efficient for surface parameters retrieval from alluvial fan.This method presented that the cross function of surface parameters inversion could be variable with the amount of data acquired.The data used for surface parameters inversion must be more than two scenes.The more data could generate the more accurate results.Then the surface parameters of the alluvial fan in Ejina Area of Inner Mongolia were estimated using ENVISAT ASAR and ALOS PALSAR data.The estimation results show the ground surface of Ejina alluvial fan is very flat,so the range of its roughness is small,and the root mean squared heights in most party of the alluvial fan are no more than 1.0 cm.The roughness in the area along the Heihe River is big,and it is very small in the other areas far from the Heihe River.The estimation result of soil moisture shows that this area is very arid,and the soil moisture volume in most part is no more than 10%.
  • ARTICLES
    Agrotype Recognition Using Dual-Frequency Multi-polarization SAR Data and Multi-spectral Data
    ZHAO Tianjie, LI Xinwu, ZHANG Lixin, WANG Fang
    2009, 11(1): 84-90.
    Download PDF ( )   Knowledge map   Save
    This paper,taking Changping District of Beijing as a study area,conducts a detailed analysis on crop characteristics combining ASAR image,new PALSAR dual-frequency images and TM multi-spectral image.Firstly, MIMICS is used to simulate the backscattering characteristics for spring maize and orchard based on extracted backscattering coefficient from images,and the modeling result is compared with the radar measurement result. Then,a classifier based on BP neural network and normal fuzzy distribution function was established utilizing redundancies and complementary of multisource remote sensing data sufficiently.This classifier used training samples generated by Gaussian fuzzy distribution function,taking the advantage of the study ability of fuzzy neural network to adjust membership functions and fuzzy rules,completed fuzzy inference,consequently made the system have adaptive characteristics.Results indicated that:the dual-frequency and multi-polarization radar data can afford more crucial information for crop classification,this kind of classifier is able to be used in crop recognition after training, and the classification precision increases by about 10%averagely,it is an effective crop recognition method.
  • ARTICLES
    Study on Dynamic Changes of Wetland Landscape Pattern in Yellow River Delta
    ZONG Xiuying, LIU Gaohuan, QIAO Yuliang, LIN Song
    2009, 11(1): 91-97.
    Download PDF ( )   Knowledge map   Save
    In this paper,the information about wetland landscape features of the Yellow River Delta in 1986, 1996,and 2006 were extracted respectively based on TM images and by support of Remote Sensing(RS)and Geographic Information System(GIS)techniques.By analyzing the information,we investigate the spatial and temporal changes of wetland landscape and explore the driving forces for wetland landscape pattern change in the Yellow River Delta within the past 20 years.Our results show that:(1)the area of man-constructed wetland demonstrated a sharp increase,whereas the proportion of natural wetland decreased accordingly.As an important part of the natural wetland,the tidal flat and Chinese tamarisk shrub wetlands in the Yellow River Delta has significantly shrank in coverage and degraded during the past two decades.(2)Since 1990s,some of the important landscape indices such as the number of patches,Shannon's diversity index,and Shannon's evenness index have increased.This indicates that the fragmentation degree of wetland landscape in the Yellow River Delta intensified, the type of landscape was much diversified,and each type is distributed uniformly in the wetland landscape and no dominant wetland landscape types can be recognized.(3)Landscape style transfer occurred between wetland and wetland,wetland and non-wetland.During the interval from 1986 to 2006 A.D,about 18.1%Reed meadow had been transferred to farmland,26.6%Seablite meadow transferred to salt pan and 11.9%Reed swamp transferred to Reed meadow.(4)The driving forces for wetland landscape pattern change analyses were conducted, and the result implies that both natural factors and anthropogenic factors have played important role in altering the pattern.The former includes decrease of discharge and sediment,self succession,while the latter mainly consists farmland reclamation,exploitation of tidal flat and oilfield development and construction.
  • ARTICLES
    Study of Water Body Information Extracting Methods Based on Spectrum Characteristics
    XU Jing, CHEN Xingwei, CHEN Wenhui
    2009, 11(1): 98-103.
    Download PDF ( )   Knowledge map   Save
    Landsat ETM+data are used to extract the water bodies in build-up and mountainous areas based on the spectrum characteristics of water and background objects,by the methods of threshold,spectro-photometric, index models and criteria function,respectively.The results of the test indicate that MNDWI index model has the best performance in the water body extracting,while the other four methods mix the other objects information with water body.Studies are also carried out on other types of water body in different areas in order to validate the above conclusion.The MNDWI index model is still the most effective method to identify water body information. However,restricted by the spatial resolution of remote sensing image,some trickles which occupy less than one pixel or distribute over several pixels simultaneously are difficult to be extracted.If the spectrum and spatial resolution are,appropriate,this method will be feasible in applications.Tests showed that MNDWI index model was effective in restraining shadow,residential area and vegetation information.It is convenient,fast and accurate in water extraction and can be used in many situations.
  • ARTICLES
    Quantitative Retrieval of Water Quality by Remote Sensing in Taihu Lake Based on MODIS data
    LU Heng, HUANG Jiazhu, JIANG Nan
    2009, 11(1): 104-110.
    Download PDF ( )   Knowledge map   Save
    MODIS has become a potential remote sensing data for monitoring inland lakes' water quality because of it's high spectral resolution,time resolution and radiance resolution.The correlation between water quality parameters and visible and near infra-red bands was analyzed and the liner model and BP neural net model were used to retrieve water quality concentration.It was demonstrated that SS in Taihu Lake has the highest correlation coefficient with MODIS band 13,chlorophyll-a in Taihu Lake has the highest correlation coefficient with MODIS band 2,in the hundreds of band combinations,SS in Taihu Lake has the highest correlation with M12/M17,chlorophyll-a has the highest correlation with M2/M8,so the chlorophyll-a concentration can be retrieved by band combination of M2/M8 and M2/M9,while solid suspended concentration can be estimated using band combination of M12/M17,M13/M17 and band 4,but the estimation accuracy is very low that can't be used in practical applications. So a two-layer BP neural net with 16 input nodes of visible near-infrared bands of MODIS and 1 output nodes of water quality,was constructed to infer the water quality concentration in Taihu Lake,the accuracy of this model was highly improved compared to the linear model.Water quality in Lake Taihu can be effectively retrieved via BP neural net and MODIS's visible and near-infrared bands.
  • ARTICLES
    The Dynamic Monitoring of Land Use Change in Guangzhou Based on RS and GIS
    ZHONG Kaiwen, SUN Caige, XIE Liang
    2009, 11(1): 111-116.
    Download PDF ( )   Knowledge map   Save
    Change detection was a technique which was used to determine the change information between two or more time periods of a particular object of the study area.Because of the advantages of macroscopic information, timeliness,periodicity and comprehensive nature,remote sensing has been popular in land use study.The land use change can be monitored quickly,objectively and precisely.In this paper,detailed introduction about the classification process of land use in Guangzhou is gaven a supported by RS and GIS software using the data of Landsat TM of 1997 and 2003.After being geometric corrected,subseted and enhanced,the images are classified into 9 classes as farmland,orchard,forest,disk-pond,urban and rural land,developing area,river,lake and unused land by using maximum likelihood classifier.The total classification accuracy was 85.4% for 2003 and 87.2% for 1997 respectively.And then the mathematic method is used to analyze the change of land use by the conversion matrix of land use between 1997 and 2003.Through the above analysis,the article draws the following conclusions:1)In 2003,the areas of orchard,forest land and rural land have higher percentage than the others in Guangzhou.And the areas of 9 types of land use are distributed unevenly in all the regions with obvious difference. 2)The characteristic of the change on the structure of land use in Guangzhou from 1997 to 2003 is that the change of farmland and forest land is not obvious while area of disk-pond and unused land decreases,orchard and that of urban and rural land and newly development zone increases.3)The change rate of newly development zone and urban and rural land is the highest followed by orchard and pond.4)Regional difference in land use change is obvious,and displays a dissimilar transformation situation for each type of the land use in space.
  • ARTICLES
    Remote Sensing Monitoring and Analyzing of the Urban Built-up Land Expansion in Kunming in 1974～2004
    TAN Wenbin, LIU bin, ZHANG Zengxiang, ZHAO Xiaoli, YI Ling, WANG Xiao
    2009, 11(1): 117-124.
    Download PDF ( )   Knowledge map   Save
    Urban expansion influences the regional environment remarkably because there are many human activities concentrating on it.This paper took Kunming which is a major city locating on the Yunnan-Guizhou Plateau as a study area.The expansion of Kunming urban area in different periods during 1974～2004 was obtained by using one Landsat MSS image and five Landsat TM images.The spatial-temporal features of the urban expansion and the relationship between population growth and urban expansion during the 30-year period were analyzed by using expansion speed,expansion elasticity and fractal dimension.The research shows that the built-up area in 2004 is 3.07 times larger than that of in 1974.The expansion rate is distinct from each other and Kunming has experienced two fast expansion stages in 1992-1996 and in 2000～2004.Considering the relationship between the construction land area and population,the urban sprawl rate of Kunming did not match urban population growth during the study years.The increase of the fractal dimension value before 1992 shows that the built-up land sprawled out from inner city in a radial way while the decreasing value after 1992 shows the city sprawls by the way of low-density fill between main roads.The cultivated land is the primary land use type converted to urban land use which accounts for one half of the urban area increment in each stage during 1974～2004.The natural environment is the limiting factor to the urban expansion.The economic development especially the increase of GDP and the growth of average salary is the fundamental driving force.The policy and the urban planning are the guidance of extension of the urban built-up land.
  • ARTICLES
    Detection of land use change in the Laohahe River Basin of West Liaohe River based on remote-sensing images
    FANG Xiuqin, REN Liliang
    2009, 11(1): 125-131.
    Download PDF ( )   Knowledge map   Save
    Land use/cover change is one of the most important areas of global change.On the basis of the Landsat TM and Landsat ETM+images and other auxiliary data,the rules of decision trees were developed in this study. Land use/cover classifications in 1989 and in 1990 were implemented based on the decision trees with satisfied accuracy. The land use changes from 1989 to 1999 and their spatial distribution were analyzed quantitatively with GIS techniques in semiarid Laohahe River Basin with a catchment area about 19,000km2 in North China.The results show that in this period,distinct land use changes happened in the study area because of the impacts of climate change and human activities.The changed area accounted for about 33.64%of the basin.The most important change was the conversion among agricultural lands,forest lands and grasslands.Water areas,bottomlands, forest lands,grasslands and fallow lands are decreased.And the fallow lands decreased especially seriously because large numbers of sandy lands and bare lands were exploited.Rural and urban areas,paddy fields and drycrop lands increased.The paddy field areas increased more than 10 times because of the change of agricultural cultivation. And rural and urban areas increased because of the expansion of Chifeng city.This study provides an important precondition for the study on the effects of land use change on evapotranspiration and hydrological processes in Laohahe River Basin.
  • ARTICLES
    A Study on Spacetime Evolution of Soil Erosion Based on ANN-CA Model
    DONG Tingting, ZUO Lijun, ZHANG Zengxiang
    2009, 11(1): 132-138.
    Download PDF ( )   Knowledge map   Save
    Soil erosion is one of the primary environmental problems in China even in the world.Until now those models applied in soil erosion research can only do some tardy forecasts.They can't achieve forehead forecast of soil erosion,let alone reveal space-time evolution law of soil erosion.This research utilizes the strong performance of cellular automata in spatial dynamic simulation.It constructs cellular automata(CA)model in soil erosion and realizes spatial prediction of soil erosion in the study area in 2010.In CA model,transform rule is essential.In this research it selects artificial neural network(ANN)as transform rule which has strong abilities in the respect of self-organization,self-adaptation and self-learning.Through ANN it makes sure the relationship of soil erosion between 1990 and 2000,and then it uses this relationship to predict soil erosion in 2010.Comparing computed soil erosion intensity type with predicted type in 2000,it evaluates the predicted precision.The precision of all the soil erosion intensity types is over 70%except moderate soil erosion intensity type.It also analyses the reasons for moderate soil erosion intensity type with relative low precision.The precision will be better by increasing samples of each soil erosion intensity type and improving classification standard.ANN-CA model reveals macroscopical variety law of soil erosion in the study area through microcosmic state change of cell and comes true soil erosion prediction in the future.All the above models this research constructed can also be applied into other regions.