The vegetation changes in Xiamen of SE China from 1989 to 2000.have been detected using remotely sensed data.Three Landsat TM/ETM+ images of 1989, 1995 and 2000 have been used in this study.The change detection and change information extraction techniques were employed to reveal the vegetation changes.However, the resultant different images of NDVI contained false change information that was not produced by the vegetation change.Therefore, a series of thresholds had to be determined to distinguish the false change information from the true one and mask out the false information from the different images.The final results indicate that the vegetated area in Xiamen has been significantly reduced during the study years.This was due largely to the rapid urban expansion.Nevertheless, the speed of the reduction has been slowed down in the last five years from 1995 to 2000.Obviously, both governors and people have paid more attention to the ecological environmental protection while doing various constructions.

As the realization of Geographical Information System (GIS) is a quite complicated course, the method of modeling and hierarchical structure is used widely. Furthermore, the applied design pattern makes it easier and more convenient to use successful designs and software architecture repeatedly than ever before, which benefits the development and reconstruction of GIS. The thesis, through the analysis of geometry model architecture of Geographical Information System feature model, fully uses and combines various kinds of design patterns to probe into the combination of composite pattern and visitor pattern together for feature organization, and makes it possible to access inhomogeneous geometry in geometry collects. Generally speaking, composite pattern and strategy pattern on rendering geometry for feature visualization enable us to separate rendering from feature naturally and sustain various kinds of rendering during the development and implementation of visualization of geographic feature. Advantages of this kind of structure lie in not only satisfying easily Open GIS abstract specification and implementing specification of simple geometry, but also keeping structure clearly and module democratized, and making the structure more flexible and natural. It is beneficial to the development, maintenance and expansion of the Geographical Information System.

In the process of analysing polarimetric imaging radar data, one is often faced with the situation,namely some geophysical parameters must be inferred from an area that exhibits significant natural variability in the scattering properties.Using Cloud's decompositon technique, an average stokes or covariance matrix can be decomposited into a sum of matrices representing single scatterers.Then the targets can be classified as single, double, and multiple scattering targets according to their scattering machenism.Akasayqin lake is a saline lake located to the north of the Qinghai-Tibetan Plateau, previous research revealed that the evolution of the lake is closely related with the environmental change caused by the uplift of the plateau.With the climate change in this area, a series of shore lines formed in the process of lake retreat during tens of thousands of years.To identify these shore lines accurately is very important for further research of the climate change.In this study, the Cloud's decompositon method was used to polarimetric SAR, Based on the target decomposition technique, some useful surface parameters include single,double, and multiple target backscattering coefficients can be extracted from the SIR-C du-frequency polarimetric radar data, by analysing these surface parameters, some useful information about the lake area can be got: 1.The L band Radar can penetrate the surface and get more information about the subsurface than C band Radar in lake area.2.Shore line features are easily discriminated from alluvium in co-polarization images than in cross-polarization images.3.As co-polarization images mainly reflect the result of simple scattering,and cross-polarization reflect multiple scattering, the poor exhibition of HV images in showing the variation of shorelines indicates that less depolarization was induced, due to small surface roughness of dry sand beds in L band.While for co-polarization images the radar's brightness and darkness features are enhanced due to its illumination to undulated shore lines.Results show that by using target decompositon method to polarimetric Radar, the multi-period bank lines can be identified, and the process of the climate change in this area can be known.

Web-shared digital map-symbol-base is an important research direction associated with sharing spatial information on the internet and interactive analyzing service. In this paper, we illustrate how the shared digital map-symbol-base was integrated into the Fujian province e-government information sharing platform service agent system, and analyze the present technique about designing and developing of shared digital map-symbol-base based on GIS component.A design principle on digital map-symbol-base faced network service application has been presented. By using GIS component ArcObject (ArcObject is built using Microsoft′s Component Object Model (COM) technology), the prototype of shared digital map-symbol-base has been established and its functions of symbol editing, storing and managing as well as its symbolizing process have been explored in detail. By integrating into the Fujian province e-government information sharing platform service agent system, the shared digital map-symbol-base can primarily meet the application demand on sharing spatial information based on the Internet and interactive analysis service in different departments and different realms of profession.

“模型”一词,在《说文》中早有记述:“模,法也”,清代人解释:“以木曰模……以土曰型……,皆法也”,都是指仿制或缩小仿制的意思。其有多种解释。当一个数学结构作为某种形式语言(即包括常用符号、函数符号、谓词符号的符号集)的解释时,称为模型。任何数学结构都由一个非空集合组成,这个集合称为模型的论域。形式语言中每一类符号都分别被确定地解释为论域中的元素、数学结构中具体的函数或关系,数学中出现的各种结构等都可以称为模型。如果一个数学结构使得形式理论中的每个公式在这个结构内解释为真,那么这个数学结构就称为这个理论的一个模型。对于一个理论模型不一定是唯一的(引自辞海)。

“数字中国”目前不再只是“热门话题”,而是真枪实弹地在启动、运作了。诸如“数字福建”建设规划的制订 ,建设全省公用信息平台 ,省政务信息网络工程和政务信息资源整合改造工程、组建福建省空间信息工程研究中心、制定省政务信息共享政策 ,同时启动“数字鼓楼”示范区。福建省是真抓实干 ,棋先一着 ,为“数字中国”领跑。另外 ,在全国国土资源信息化会议上 ,田凤山部长提出 ,以信息化带动国土资源工作的科学化和现代化。

“数字地球”如今不仅是信息社会的热门话题 ,而且已成为只争朝夕的行动。目前 ,各省 (区 )都陆续开展“数字省 (区 )”的研制 ,下面是“数字中国”的若干省区的数字化工程建设态势。( 3)数字山西充分利用已有的基础和优势 ,实施数字山西战略 ,合理利用资源 ,加强环境保护 ,以信息化带动工业化 ,提升传统产业的现代化水平 ,促进经济结构调整 ,全面推动山西经济的跨越式发展。数字山西建设将由省政府来抓 ,构建数字山西的基本方针是“政府主导、统筹规划、阶段发展、统一标准、联合建设、资源共享”。

原始创新的主题 ,党和政府都十分重视 ,企业界也非常关注。学习“三个代表”的指导思想之后 ,期望就更加殷切。大家都希望迅速提高全民的素质教育和科技原始创新的能力 ,让世纪英才 ,脱颖而出。有的发起向“诺贝尔”冲刺 ,有的致力引进“海归”人才 ;有的强调 SCI作为管理指标 ,有的采取猎头战术 ,应该说是煞费苦心 ,求贤若渴。这样千年一遇的机会 ,对于饱经风雨的“臭老九”来说 ,无疑是天上地下 ,换了人间 ;但是对于早已与世界接轨的科学精英们来说 ,只不过像是足球场上听到裁判员的几声哨音 ,未必给他们带来多少惊讶和兴奋。

The West China will be developing under a new era with network,knowledge based and globalized economy. The target of the national strategy is focused on decreasing the regional differenciation,spreeding the adjustment of economic structure, exchancing the friendship between all of the nationalities in the whole mother country.After the efforts of scientific surveying and exploration in the past 50 years, the geo information source in West China may be very useful for the regional planning or engineering design after the updating by the new satellite remote sensing data, GPS and GIS. Under the support by 'informatization', the 'modernization' of the West China will read the success step by step more quickly.

广东省深圳市启动一项计划 :1 0年半城披绿装工程。目标是 :全市城市建成区绿地率达 40 % ,绿化覆盖率达 5 0 %以上 ,强化绿化生态功能 ,建设具有岭南特色的亚热带海滨园林城市。云南省的昆明、安宁、开远等 5个城市已达园林城市标准。辽宁省 2 0 0 1年 9月召开河道工作会议 ,要求大力推广“绿堤工程”。种树种草 ,变大堤为绿色风景线。

城市是人类聚落的高级形式。城市是人类文明进步的重要标志。中国城市的发展,历史悠久。据考古发现,早期的城市可追溯到五、六千年之前。以往人们主要是依据考古发现实物,古籍文字记载,研究城市的萌芽及其发生发展历史。这无疑是研究城市发展重要的方法和手段。但是,还有一个不争的事实亦不应忘却,即古代城镇在其发展进程中的某一历史阶段,它的布局结构,如城廓范围、城门城墙、街道桥梁、寺观庙坛、宫学衙署、河流湖泊等自然、人文景观,或早或晚或详或略地被先民用图形、线条、文字,或示意性地或按一定比例,表示在绢、纸、木板、石碑等不同载体上。古代绘制的那些城市地图,有的流传下来了。

The paper probes the application of artificial neural network in basic land price evaluation after analyzing current methods of land evaluation and properties of artificial neural network. Two artificial neural network models for land evaluation are developed, which are called as the network model I and the network model Ⅱ. The network model I is based on the relation between general value of all factors which are taken into account in evaluation work and basic land price.The network model Ⅱ is based on the relation between the value of single factor and basic land price. The improved algorithm for back propagation of two kinds of models is presented. Taking Wuhan city as an example, the models developed in this paper are tested.After comparing different models, we can make conclusions as follows: 1.The training error of neural network model I is almost the same as the testing error of exponent model, which can reflect that the relation of general acting value of factors and basic land price is the exponent relation in the test area. 2.The training error of neural network model Ⅱ is relative smaller than that of exponent model, which shows the model can represent correctly the relation of acting factor and basic land price. The model do not need to decide the weight for each factor by the people and reduce the influence to the evaluation work by the people. 3.The result shows that the models of artificial neural network applied in land evaluation are objective and accurate.

The development and advancement of China national atlases since 1950's was discusses in the first part of the paper. And author also systematically summarized the theories, methods and technologies of the atlas compilation.

The meaning and characters of digital earth were briefly generalized and the transformation from Geo Information System to Geo Information Science was analyzed. The main content of Geo Information Science was generalized. Some key problems for founding Geo Information Science were put forward, which were spatial database and the description of geographical phenomenon.

At the early development stage of GISs, the planar graph model has been used for spatial data modeling. The multi dimensional and dynamic spatial entities were mapped into two dimensional points, lines and polygon objects with associated attributes. This resulted 2 or 2.5 dimensional geo spatial data framework, which has been widely used in various fields and has been found many limitations. There is now a demand from many applications to develop a new generation of geo spatial data framework for three dimensional and dynamic environment. This leads to the evolution of multi dimensional and dynamic geo spatial data framework. After a discussion of the concepts and history of geo spatial data framework, this paper tries to define the context of multi dimensional and dynamic geo spatial data framework, and identify a list of challenging issues for futureresearch.

The paper discusses how to establish metadata management system of National Resource and Environment Spatial Database. It gives the designing method of Unified Modeling Language (UML )static diagram for metadata framework designing and data dictionary for detail element definition. Using Extensible Markup Language (EML) to describe metadata standard, it carries out metadata management and metadata expansiong. It can help user to query effectively and find resource quickly in the database.

In the mid-1990s, we established the national operative dynamic information systems on natural resources and environment. In the process of establishment of the land-use database, 520 scenes of remotely sensed images of Landsat Thematic Mapper (TM) were interpreted at scale of 1∶100 000 under overall digital software environment after being geo-referenced and ortho-rectified. The vector map of land-use in China at the scale of 1∶100 000 was recently converted into a 1-km raster database that captures all of the high-resolution land-use information by calculating area percentage for each kind of land use category within every cell. Being designed as the operative dynamic information serving systems, monitoring the change in land-use/land-cover at national level was executed. We have completed the updating of LUCC database by comparing the TM data in the mid-1990s with new data sources received during 1999～2000 and 1989～1990. The LUCC database has supported greatly the national LUCC research program in China and some relative studies are incompletely reviewed in this paper.

The paper puts forward data frame of the remote sensing knowledge base. Oriented-object method is adopted to organize data of the knowledge base. The typical land surface objects, which are classified into vegetation, soil, water, snow, minerals, man-made object, are the center to be used to link measured data at field, prior spatial data, expert system data, images and models.The knowledge base consists of five aspects: 1)The database for measured spectrum and relevant environmental factors.The database consists of 6 kinds of data.The first is the measured data, which are made up of canopy spectrum, soil physical-chemical data (N, P, K, humidity etc), plant bio-chemical data (chlorophyll, nitrogen etc), vegetation structure (LAI, LAD etc) and biomass data, component (leaf, stalk etc) spectrum data, and data about microclimate and weather. The second is the instrument information. The third is data of campaign site condition. The fourth is the documents about observation methods and measured data processing methods. The fifth is the documents about surveyors. And the sixth is the metadata.2)The spatial database for landform,land-use and land-cover etc.3)The expert system for vegetation seasonal changes and spatial patterns.4)The remote sensing image base, which are acquired by field, aerial and space approaches such as multi-angle, multi-temporal, multi-spatial resolution, multi-spectral resolution, multi-platform etc. And 5) the remote sensing physical model base. The knowledge base can not only retrieve spectrum and environmental factors of typical land surface objects, but also simulate the spectrum of remote sensing pixel when the measured spectrum is unavailable. The model system consists of the analytical models and computer simulations for remote sensing mechanism. The spatial scale of the models is from the material (such as leaf, soil) to end-member (such as canopy) to remote sensing pixel (such as a pixel consisted of wheat, road and lake).The user enters object/place, time, observing direction, bands and spatial resolution,and the knowledge base provides the reference spectrum and relevant description. The spectrum of vegetation depends on the seasons, and it is impossible to measure the spectrum of vegetation at all times. So the vegetation parameters have to be extended temporally when the measured data is unavailable.Based on the knowledge of the vegetation expert system, the vegetation structure parameters and spatial pattern can be computed by the vegetation shaping models. The land surface parameters have to be scaled up/down when the pixel size is different to the size of view field of the measured data. The describing parameters of DTM and land-cover and soil should be based on the pixel size. Selecting remote sensing physical models for simulation depends on three factors:1)spectral band: visible light & near infrared, thermal infrared, and microwave;2)spatial scale on which the models are built;and 3)land surface type, grass or forest. In other words, radiation transfer model, geometric optics model or compositive scene model is more suitable to this type. The agriculture remote sensing applications based on the spectral knowledge base also is discussed. The approach is a feasible quantitative remote sensing way in agriculture.

This paper emphasized the economic development problems of West China during the process of economic globalization. The author systemically analyzed the relationship of economic globalization and western exploitation, strategic measures of western exploitation joining in economic globalization and how to take advantage of Eurasian Continental Bridge to promote West China joining in the process of economic Globalization.

我国著名地理学家薛风璇教授,执教香港大学地质地理学系并担任系主任;兼任香港城市大学校长。以其深厚的科学造诣,又长期参与香港社会经济与市政建设的策划和实践,对香港城市的发展历史、现状与未来,了如指掌。1985年,侯仁之院士主编《中国地理丛书》,我们曾特邀薛风璇教授篡写《香港与澳门》,出版之后,畅销海内外,不胫而走!最近读到他的新著《香港发展地图集》,三联书店(香港)有限公司出版,中华商业彩色印刷公司印刷8开版本(257×355cm),192页,2001年2月闻世。

We will discuss in this paper about how denominating to geographical names has been originated and evolved in all ages in China, what principles and ways can we find in the course of denominating to geographical names, and what the denominating rules are.

In this article there is a systematic analysis the difficulties on 'Big Development'.The paper put forward some countermeasures and suggestions.

Tu" in Chinese words, means maps representing spatial information in graphics , also including images, illustrates and other graphic forms of representing spa tial informatic;"Pu"is a system built according to thing's characteristics or time series. Geo Informatic Tupu has two characteristics with graphics and pedigree. Formed from a great deal of geo information in digit through graphic thinking and abstract generalization, Geo Informatic Tupu is a means and method of using computer multi dimensional and dynamic visualization technology to display and reveal the spatial configuration and spatial temporal change rule of earth system and its elements and phenomena. Five parts are discussed in this paper, they are: 1. the basic concept of Tupu; 2. the discussion of Geo Informatic Tu pu; 3. the basic process and steps of building Geo Informatic Tupu; 4.the example of Geo Informatic Tupu complex Informatic Tupu of natural landscape in China. 5. the meanings of and prospects for Geo Informatic Tupu.

The metadata standard of data in resources and environment information system(REIS) has been created, and plays an important role in data's management and organization, but for the model and method metadata, which is very important to the usage of data, have not been described in details, thus there is very difficult to query and index for the existing mathematics model and methods, and limits its use. With the research to the REIS, put forward to the idea that the metadata should included metadata for data, metadata for model and metadata for method; it also put forward a set of metadata database architecture that is suit to the REIS in order to realize the efficient management of the model and method. At last, bring forward the method of integrating the model and method metadata with the REIS, and the application and development in this system. Finally it draw a conclusion that it is feasible to realize the integrating management and sharing based on the metadata.

The conservation planning results of this project led to the establishment of the guidelines for zonal planning of the Central Mountain Greenway based on IUCN guideline of conservation planning. These guidelines could be used as the basis of the Central Mountain Conservation project and the references of developing process of Digital Taiwan Model. Digital Taiwan is an article, multi-directional and metadata technology convoluted from traditional spatial information and political management to overcome time-consuming planning, management and decision support. As a result, this study tries to establish the Digital Taiwan Model by using GIS and RS as the Land Planning tool to correlate the existing digital data of land resources, and apply the concepts of "Environment resources sustainable development", "Environmental costs", and "Bio-diversity conservation". This study will serve as pilot reference for executing the land planning and land planning decision-making support system.

The management of Remote Sensing images is based on files at present, but this management method has many disadvantages for data sharing, we need new methods to manage images data. ArcSDE, a key component of ArcGIS developed by Environment System Research Institute of America, is an open architecture and GIS gateway to connect GIS to RDBMS, such as Oracle, Microsoft SQL Server, DB2, Informix and so on. Based on the understanding of architecture and principles of ArcSDE, this paper found that using ArcSDE to build a RS image database is an effective way except for some limitations, and developed a method of building image database using ArcSDE tools, and took building remote sensing image database of Chinese coastal zone as a case study. During the case study, some problems were met and corresponding resolutions were identified:The storage of overlap parts between images is not very effective when mosaicing is done to them, especially for those Nodata area. During the storage when an overlap image data comes ArcSDE simply rewrites the data of these overlapped blocks. Therefore, the blocks with values have to be rewritten by other blocks with Nodata from another image.The authors provided a resolution to this problem, that is to customize the storing procedure when an overlap block comes. The system checks each pixel value of blocks that will rewrite and blocks that will be rewritten. If the rewriting blocks contain Nodata pixels while the rewritten blocks have values in corresponding locations, the rewriting action will be cancelled in these locations or pixels, otherwise, the rewriting action will be executed. This method can ensure that after mosaicing each pixel of overlap parts will have value as long as one of those images that take part in mosaic has value on that location. Searching and retrieving functions of raster data are not as flexible vector data, users can improve searching and retrieving functions of raster datasets if some vector feature classes were merged into these raster datasets, and we can even search and retrieve raster data through toponyms or placenames if a powerful gazetteer was constructed.

THe author proposed three suggestions on the development of West China. First, editing and analyzing the rich geo information about the natural resources and environment in the West China provides basic scientific data and facts for the West China development. Second, the development strategy and measures should be suitable to the local reality of vast area and outstanding regional differences. Third, it is important to make full use of second Eurasian continental bridge.

Nowadays both internal and external geographical history circles doesn't deeply understand the achievements of Before Qin geography in China.This paper think that the achievements of Before Qin geography is fairly splendid.It is the age of origin of Chinese geographical works.And it is also the period of Chinese ancient geography becoming a sort of knowledge.Before Qin geography had a good tradition of inspecting the nature and research the nature.From Qin-Han dynisty on,Chinese ancient geography have lost this good tradition,and paid more attention on textual research of yange geography and recording the social history and the social condition.These little by little formed Chinese tradition geography on main part of the study of classics geography and the study of history geography and the study of yange geography.This is a big fault in the development geographical history of China,between Before Qin and Qin-Han.This fault directly affected the achievements of the Chines ancient geography,and resulted the Chinese ancient geography in not developing to modern geography.

This paper maily introduces the design of multi-language placename index software, which forms a part of SINO-Japan GIS Coorperation The software can implement the following operation: Multi-object placename indexing, placename index data file Creation, and so on It may acts as a basic software for map data indexing in the future implementation of Clearing Houses (namely Spatial Data Exchange Web Site on the Internet)

The concept of stochastic resonance (SR) has been applied into the analysis of satellite thermal infrared images. Six kinds of anomalous phenomena related to crustal movement were recognized in satellite thermal infrared images. Six diagnostic indicators for the prediction of magnitude above 6.0 global earthquakes and their quantitative evaluation standards have been established. The microscopic behavior of global crustal movement is successfully controlled by using satellite thermal infrared imagery, and the occurrence time and magnitude of over 80% of global strong earthquakes occurred since the foundation of the observation station have been successfully predicted. It is believed that the combination of satellite thermal infrared information with macroscopic anomalous phenomena will play an important role in earthquake hazard reduction.