GIS should be able to support management, integration, sharing and analysis of different kinds of massive, dynamic, unstructured spatio-temporal data. The geometry, space, time, semantics, relationships, and attributes of geographical data should be integrated and expressed comprehensively. And the support for the description, analysis, problem solving and expression of real geographical environment and process should be maintained in a geographical information system. However,existing GIS technologies are still difficult to support the above-mentioned requirements from both theoretical and application perspectives. The spatio-temporal expression, data models, human-computer interaction and operational efficiency are still insufficient, which also lead to the lack of large demonstration projects in social applications. Therefore, it is necessary for GIS to essentially breakthrough such techniques as a unified spatio-temporal data model, effective data management, spatial index, GIS accelerator and various forms of expression and interactive capabilities. The GIS data model and organization should be spatio-temporal unified and need to represent the geographical entity and the spatial distribution of geographical phenomena. The GIS accelerator can support the progressive transmission of data according to the screen resolution constraints. And for the visualization and expression, it needs to develop various GIS expression and interaction methods based on the computer vision, true three-dimensional displays, virtual reality, actual situation fusion, etc. These technologies will form the key technology system supporting the formation of the social applications of GIS. Meanwhile, the correct comprehensive view at the world of the "GIS" concept should be established. In order to develop social applications of GIS, we must firstly establish the perspective of "comprehensive observation of the world" with GIS. Secondly, GIS develops from "geographical environment oriented" towards the "people-oriented", especially VideoGIS for the requirements of the mass media. In the construction of big demonstration projects of GIS social applications, genealogy GIS is a typical socialized GIS application, which plays important roles in the heritage of Chinese history and civilization. It should also promote the genealogy GIS which represented a major social engineering, in order to increase public awareness of participation, and promote the development of social applications GIS.
The quality of map symbolization and the display speed of the electronic map are always the hot issues for the cartographers. The availability of the electronic map mainly includes the effectiveness, efficiency and user subjective satisfaction. Currently, the map symbolization quality can meet to the basic needs of various user groups, but electronic map rendering speed has affected the feelings of map users. Since the electronic map involving large volumes of data, it needs a relatively long period of time to respond to user's operation, causing the user to generate anxiety and dissatisfaction because of waiting, so it is difficult to give map users more smooth operating experience. For the problem of improving the rendering speed, current studies mainly focused on data block processing, hierarchical processing and the establishment of efficient indexing model to improve the efficiency of electronic map display by reducing the amount of map data. Based on analysis and summary on the improved multi-sheet electronic map rendering speed, this research compares the characteristics between GDI and OpenGL. When we use GDI and OpenGL to draw 2D entities at the same time, if the number of primitive is small, the difference is not large. But if the number of primitive is large and primitives are static, the rendering speed using OpenGL is far faster than the GDI, because OpenGL can use display list technology, texture mapping and vertex data technology. In this paper we presented and implemented a symbolic method in three-dimensional interfaces for the different types of features. It proposes the way of expression and organization of multi-scale data in the three-dimensional space through multithreading technology and memory pool. The experimental results show that it is able to achieve the symbolization of point feature, line feature and area feature. And roaming, zooming in multi-sheet electronic map data, screen rendering speed remain at more than 17 frames per second. So, using methods and techniques mentioned in this paper can improve rendering speed and ensure the good interaction effect between user and map.
A road network with topology is the basis for optimal path finding. According to the definition of connectivity topology by ISOGDF4.0 (Geographic Data File) model and the requirements of road network connectivity topology during the optimal path finding procedure, this paper presents an efficient algorithm to construct connectivity topology for road network using R-tree and B-tree indexes. The efficient algorithm contains two parts: the first part breaks the roads at intersections and generates anew road network; the second part constructs connectivity topology based on the new network. The procedure of building new road network breaks the roads into line segments and gets intersections at first, then reconstructs new roads with the line segments set and the intersections. During the procedure of getting intersections of any two line segments, the algorithm builds R-tree spatial index on the line segments to improve query operation efficiency significantly. What's more, our algorithm also builds B-tree index on serialized data of the roads such as the identity codes while constructing connectivity topology to improve the query operation efficiency on serialized data. This paper also tests the time complexity of the algorithm using road networks with different scales. Experiment results show that our algorithm builds connectivity topology of large-scale road network in very short time. (For a road network that contains about 30 thousands line segments and 25 thousands points, the quick algorithm completes connectivity topology construction in no more than 3 seconds). The time cost of our algorithm increases slightly while the traditional algorithm increases tremendously, to be exact, our algorithm's time cost grows at the rate of [O(vlog(v))]while the traditional one grows with the speed of [O(v2)], so we concluded that the quick algorithm has high efficiency and is valuable to practicable application.
Spatial scan statistic method is a widely adopted spatial cluster detection method in the field of public health surveillance. It can detect a sub-zone where the number of disease cases rises abnormally, based on infectious disease surveillance data, and thus is able to make early warning on possible outbreak of infectious disease. Chinese Center for Disease Control and Prevention (China CDC) launched China Infectious Disease Automated-alert and Response System (CIDARS) in 2004, which handles the infectious disease surveillance data of all of the counties of China to detect possible case clusters. The making of candidate clusters is a key step to this method, which to some extent determines the accuracy and time efficiency of the spatial scan statistic method. There are two deficiencies if the existing candidate clusters making method is applied to a very big research area with a lot of sub-regions. The first is that, the inappropriate separation distance of grid points might miss a lot of possible candidate clusters, which affects the accuracy of detected result. The second is that, the existing method might duplicate a great number of candidate clusters, which could prolong the computing time of subsequent spatial scan operation. In this paper a new efficient method is proposed according to the former existing candidate clusters making method. Based on the correct setting to the separation distance of grid points, the new method could greatly reduce the possibility of missing of some possible candidate clusters. At the same time, applying multiple-sort arithmetic, the proposed new method could find and delete a great number of duplicate clusters in the original-making candidate clusters in a shorter time. Finally, the paper applies and tests the proposed method for the making of candidate clusters in 608 counties in southwest Shandong Province and proves that the method works satisfactorily in both two aims, that is, it reduced the computing time and reduced the missing of candidate clusters.
In this paper we aim to develop a location-allocation model for the school emergency shelter selection by using GIS. Through the studies of emergency shelter in recently years, a location-allocation model for school emergency shelter selection is proposed. Based on location-allocation model theory, the optimal evacuation model for school emergency shelter selection is made with constrains of the largest site service capacity and the shortest network distance. In the process of the model construction, we take school emergency shelter's safety and suitability into full consideration. According to the characteristics of this model, the environmental constraints model is constructed to make spatial overlay analysis of six factors by using ArcGIS 9.3. The six factors are terrain slope, rivers and lakes, earthquake fault zone, flammable and explosive site, collapsed range of High-rise buildings, and protection range of cultural relics. The environmental constraints model respectively discusses the five factors influence range when the earthquake happens. And the actual capacity of the evacuation demand and site supply is calculated in this way. In order to reduce the error, the straight-line distance as the evacuation path is not chosen in this paper. Through the network analysis of the unaffected road, the shortest evacuation road is found. Considering the largest site service capacity and the shortest network distance, is the key to solve the problem. By introducing the location-allocation model theory, the solution of the optimal evacuation model is obtained to select the school emergency shelter. In this paper, school emergency shelter in Xicheng District of Beijing City is selected by using this technical method which can effectively solve the problem of school emergency shelter selection. This optimization method is an effective measure to prevent the school from disaster and provide safe site for the student. This academic technology is introduced as the references to relevant researchers on the school emergency shelter selection.
Spatial data of forest resource is that related data which has characteristic of space geometry. It mainly comes from the forest resources survey, including the national forest resource continuous inventory and the local annual survey. It often lost information of time among the surveys. It causes difficult to achieve the characteristics of historical inquiry. Therefore, temporal-GIS was introduced in this paper. The sub-compartment is the basic unit of forest resource investigation and management. Taking it as the main research object, as the same time, introducing the time factor, we have established a data update model based on temporal-GIS. The object of this model is to improve the update rate realize the historical query of forest resources spatial data. In this paper we have analyzed the changes of forest resources spatial data and has understood those changes are due to reasons such as natural growth of trees, disasters, legal and illegal human activities and the survey error. According to the variation of space and attribute, the variation type of sub-compartment is divided into eight basic ones. Through described temporal and spatial information of sub-compartment updating we have improved the traditional base-state with amendments by the way of dynamic classification indexes. At the same time, we have extended model through object-oriented and version management. On this basis, we structured a new spatio-temporal data model named version-based multi-base state extended spatio-temporal data model with amendments which is short for VBMBSEA-STDM. The article has a systematic description of model mainly about the three aspects: data structure of sub-compartment object, basic spatial operation of sub-compartment object and data storage of sub-compartment. According to the study, by analyzing the data of Liangshui experimental forest in 1999 and 2009, it reaches the specific changes form of sub-compartment data. It has constructed a spatio-temporal data update model of forest resource to realize the historical comparison and backtracking.
Sudden gas leakage accidents which happened frequently threaten the security of the public heavily and brought great panic to the society. In order to provide great guidance for emergency departments to implement emergency rescue operations theoretically, it is necessary to discuss the model which is used to analyze the pollution of common gas leakage accident. Firstly, to adapt to simulating gas diffusion of the 3D terrain, based on the research and analysis of the typical gas leakage diffusion model, the terrain factors are introduced to refine the model in this thesis. Then the model is integrated with GIS, which has an advanced space displaying and spatial analysis ability, for analysis and comparison of gas concentration distribution acquired from the previous model and the improved one respectively. And on that basis, we made a further research on gas concentration distribution situation under the conditions of different DEM spatial resolutions, found that the improvement of DEM spatial resolution has some impact on the gas concentration distribution in 3D terrain conditions, and we analyzed the reasons of its effects. The results of the study reveal that the improved model is more accurate and closer to the real urban 3D terrain conditions. At the same time, the improvement of DEM resolution has a big impact on the simulation of the improved model. So, choosing an appropriate DEM resolution in the right time can help emergency services simulate th gas concentration distribution area more reasonable, reliable and timely, which has a certain practical reference value and guiding significance on gas leakage accident emergency.
Clipping function is one of the fundamental functions in Geographic Information System. The efficiency of clipping function can greatly affect the overall performance of data processing in GIS application, especially when clipping large vector data. This paper proposes a new clipping computing model to effectively fulfill the clipping function for different types of clipped layers according to their different attribute and geometry characteristic, such as point, line, polygon, multi-point, multi-line, multi-polygon and so on. This clipping computing model consists of the following three steps: first of all, using clip-layer's Minimum Bounding Rectangle to select features from clipped-layer, where the features must intersect with or within the clip-layer's Minimum Bounding Rectangle; secondly, using Feature ID to build revised quad-tree index for the clip-layer features; lastly, using thread-pool to fulfill clip computing in parallel. A performance test was carried out using four different vector data layers. The result shows that the performance of clipping function based on this clipping computing model is as efficient as the one fulfilled in ESRI's ArcGIS 10.0. CPU-intensive is one drawback of the implementation of this model at its current form.
A task oriented method for mobile collection and direct report of disaster information is proposed to meet emergency work requirements around the disaster site, such as disaster information report, check and assessment, disaster emergency response, remote sensing image interpretation and so on. It will be an operational application and the total solution of natural disaster mobile information platform is introduced in the paper at length. The core of the method is regarded as the task planning, which depends on the layer for disaster information collection around the disaster site. It solves several problems occurred to the disaster information collection in the field, including multi-type disaster information collection, rapid integration, instant report, real time monitoring, team collaboration and so on. The combination of satellite positioning, base map plotting and mobile communication is widely used for disaster information collection. Compared with the universal method, four advantages are involved in the new method. First, it is a flexible method and suitable to collect any disaster information around the disaster site. Second, post-data-processing is avoided, which is resulted from poor accuracy of satellite positioning system. Third, multi-type information can be collected in an integrative way and reported rapidly. Finally, several teams are serviced by the platform to collect information synergistically around the disaster site. Natural disaster mobile information platform is developed based on the method, which is composed of the terminal, the background server and the database. The terminal plays an important role in the platform, by means of which the operator can collect the multi-type disaster information synthetically around the disaster site, report the disaster data rapidly, do the online data integration in real time, share relative information quickly and so on. It was applied in collapsed house assessment around Jiegu Town, Qinghai Province after the Yushu Earthquake. The whole town was divided into 685 plots according to collapsed houses. Four mobile terminals are employed by the working group in the field to investigate and collect data of damaged and collapsed houses, such as the level of damage, the number of damaged houses, the unit-price of one house, the number of families, the number of affected persons and so on. The analysis of statistics could help to assess the loss from the Yushu Earthquake.
Due to intensified global warming, more and more global and regional issues become the focus of research. In the context of global scale change, the mechanism of how regional climate makes feedback to global environment change and how regional climate change influences large-scale environment in turn is a difficult point for climate research. However, the transformation between scales provides ideas to solve these difficulties. With the development of climate models, nesting regional climate model into global climate models to simulate is an effective downscaling way, however the resolution limitations of the regional climate model cause simulation accuracy of mountainous areas and other complex terrain much lower than expected. WRF model, a mesoscale climate model with resolution from several meters to thousands kilometers and with two-way nested features, provides opportunities for transforming between scales dynamically. In this paper, we firstly use the WRF model to simulate the temperature at different scales and compare with 15 climate-stations' in-site value. Through the comparison, we can conclude that the simulated values are more similar with the measured ones with the resolution becomes better. Then we used statistical downscaling method to downscale the temperature from 27km to 3km and compared with the WRF downscaling results and ANUSPLIN interpolation. The results showed that the downscaling from statistical method had the same tendency with the WRF downscaling and ANUSPLIN interpolation results, but compared with the latter, the former downscaling had obvious Mosaic effect. The statistical based downscaling method only considered the influence of elevation change on temperature not other factors, while the temperature is influenced by various factors such as slope, slope aspect and land cover types. In the further research we will consider more factors and analyze the relationship between temperature and related factors more comprehensively, in this way we can make the statistical downscaling more realistic.
It was an important precondition for wetland ecosystem health recovery, management improvements, and the sustainable use of resources to carry out an accurate and reasonable urban wetland function zoning. Taking Beijing wetlands as an example, according to the result of the wetlands ecological service functional value evaluation, the method of principal component analysis was used in comprehensive evaluation of urban wetland ecological service functions. Then the space partitioning method of weighted Voronoi diagram was applied, through the main steps of the attribute weights being determined by using Newton-Breaking point model, to build urban wetlands function zoning model based on weighted Voronoi diagram. Based on the method, the typical wetlands patches were selected in the research of a "bottom-up" quantitative Beijing wetland function zoning. The results showed that: firstly, the wetland ecosystem service function of north wetland groups was very important, especially the Miyun Reservoir wetlands comprehensive value indicating the most important role of the wetland ecosystem services in Beijing. Secondly, Beijing wetland function zoning showed a three hierarchical framework, which included 4 wetland functional zones for the first level, 15 wetland functional zones for the second level, and 43 wetland functional important units for the third level in the homogeneous plane space. At last, the function magnitude of value on wetland resources was combined with the spatial relationship in wetland patches, in order to implement wetland function zoning by a double distance constraint on both attribute and space. This research not only promoted the in-depth application on value assessment of wetland ecosystem services in the decision-making management, but also provided technical supports to the urban wetland function zoning, and met the needs of urban wetlands management as well.
Internet and other modern information space provide a new data source and interactive platform for the study of intangible cultural landscape. In this paper we took the culture of the Chinese paper-cuts as an example, basing on urban spatial statistics unit and using the regularized webpage acquisition tool, with the help of the Taobao API, extracted the basic information of the E-shops and the trade information of the products. Then we used the visualization and analysis methods of GIS to carry out the expression of cultural trading landscapes under a new C2C business models. The results showed that: the cultural trading landscapes under C2C mode has a significant coupling with the areas of traditional culture, and the E-shops and consumers distributed in Beijing, Tianjin, the Yangtze River Delta and Pearl River Delta region which formed the core areas. At the same time, the development of different paper-cuts genre is significantly different: the northern faction paper-cuts like northern Shaanxi and Yuxian paper-cuts are in the dominated position in the trading. The trading relationship is concentrated in China's eastern and central regions. It is an important extension of the research perspectives and research methods of cultural geography in the information age, and has practical significance for the comprehensive understanding of the transformation and development trend of China's non-material culture industry.
Traditional classification accuracy assessments in terms of overall accuracy or Kappa coefficient based on isolating pixels statistic, cannot capture the geometrical properties of the segmented image objects, and therefore do not provide an accurate evaluation in object based segmentation or classification. Considering the importance of bordering pixels and geometric shape in object segmentation, it is thus of our interesting to design some measuring indexes introduced the border pixels and geometric structure information to evaluate the object-based resultant image segmentation or classification. This paper improved five indexes originally proposed by Persello and Bruzzone to evaluate the resultant segmentation of high spatial resolution image. Realizing that the original indices cannot well capture the gap among different categories, resulting in difficulties in discriminating the assessment results of different categories, we use normalized techniques based on geometric shape to overcome it. The indices depend on the geometry features of each object of the thematic map including over segmentation, under segmentation, edge location, fragmentation error and shape error. Moreover, we realized a prototype system contained the aforementioned evaluated indexes based on IDL platform to support the object-based image processing and analysis. To validate these indexes, a subset QuickBird image located on Fuzhou was implemented and the results of the Meanshift segmented algorithm demonstrate that the proposed indexes can provide better accuracy evaluation of each land cover class, and can make users more effectively choose the best classification map. Moreover, our experiments also demonstrate that, compared with OA and Kappa coefficient, the proposed indexes have advantages on characterizing the detailed ground materials and are helpful in aiding the optimal parameters selection for the Meanshift segmented algorithm.
Based on fusion images of Hyperion and ALOS, using linear spectral mixing model, quantitative analysis on fusion effect and its factors was explored, which included atmospheric correction via FLAASH in ENVI, spatial resolution of image, threshold value for classification of rule images. The results showed that much information would be lost with spatial resolutions becoming coarser, as well as a threshold value existed. Still, correlative coefficients between raw images and coarser resolution images were all larger than 0.90 which indicated images kept their raw spectrum and could be used to identify surface of land. Meanwhile, total accuracy and Kappa coefficient presented downward trend. Total accuracy and Kappa values for classification corrected by FLAASH were larger than that for uncorrected images, while classification map was turned into pieces which were not in accord with the actual condition. Segment threshold of fraction was one of key factors when fraction images were separated into patches to create classification images. While threshold value was from 10% to 60%, total accuracy of classification maps displayed the opposite trend to decrease. The spectrum in a pixel within the area ratio may tend to be balanced, and there was not absolutely dominant spectrum in the pixels. This revealed the value of 50%, commonly adopted, should be used on condition. And Here the different thresholds should be given for different objects.
Owing to the tidal fluctuation, interstitial water content in intertidal flat changes frequently and spatially. It gives important influence to the reflectance of surface spectrum in this area. Precisely prediction of tidal flat moisture content can make the extraction of surface feature information by remote sensing easier and more accurate. Based on detailed analysis of the characteristics of 449 sediment spectral curves with different moisture contents, three kinds of water content index namely difference water index (DWI), ratio water index (RWI) and normalized differential water index (NDWI) were proposed. And then they were used for the construction of water content reversion models and applied for the accurate prediction of moisture content in tidal flat for different sediment types. Results showed that: (1) Moisture content for sediment samples had a good segment linear correlation with its reflectance in short-wave infrared band area. The corresponding moisture contents for segmentation points were 42% and 62% respectively; (2) Reflectance of characteristic bands with central wavelengths at 1165nm, 1336nm, 1568nm and 1780nm has very good responses to the changes of water contents for different sediment types. DWI, RWI and NDWI calculated by the combinations of reflectance at characteristic bands had a significant linear correlation with moisture contents of different soil types; (3) Among the three moisture content reversion indexes, DWI had a better prediction accuracy than RWI and NDWI and a good adaptability for water contents distributed in wide ranges. In comparison, RWI and NDWI were more suitable for a moderate moisture content level. Much smaller or much larger of the water content level would lead to a relatively larger deviation for its prediction; and (4) Using the model constructed by DWI1336,1780 (λ1336, λ1780), water contents were reserved and correlation coefficients between the predicted and observed ones were 0.891, 0.915, 0.920 and 0.905 for four different sediment types names silt, sandy silt, silty sand and sand while their root mean square errors were 9.87%, 3.56%, 4.24% and 2.98% respectively. So DWI is suitable for the prediction of spatial and temporal variation of the moisture contents in tidal flat areas.
Remote sensing technology is a kind of effective technical means to obtain information of the tidal muddy flat wetlands. It has important significance to explore the tidal flat wetland's remote sensing classification method. This study's aim is to get high-precision information of features of tidal flat wetlands according to image feature and muddy tidal flat wetlands characteristics of Yancheng National Nature Reserve in Jiangsu Province, China. The TM image of 2010 is taken as the data source. The practical problems are tidal flat wetland vegetation types' complexity and the difficulty in correctly identifying the ecotone information. Because tidal flat wetland surface features in remote sensing images have a complex spectrum, it is very difficult to effectively extract information through one single method. So, we comprehensively use methods of the NDVI, band reflectance spectral characteristics, environmental characteristics and ecological conditions, one level after another to simplify the complex tidal flat wetland classification. Firstly, we chose the best combination of the bands. Secondly, we established vegetation and non-vegetation's ROI to analyze the band reflectance spectral characteristics. The ROI include 1316 Spartina alterniflora Loisel, 633 mudflats, 1253 Suaeda salsa, 1002 Phragmites cammunis, 525 water body, and 4803 road samples. At last, we established decision tree to reality the practices. After selecting 3216 samples to accuracy inspection and compared with the same area of supervised classification, the results show that classification accuracy is up to 95.87% by ROI training area. It has advantages in identifying the ecotones' vegetation types and vegetation scattered patches. The method makes up for the lack of a single classification and has an important reference value and practical significance in effectively extracting tidal flat surface features.
The boreal forest biome is one of the largest on earth, covering 17% of total land area. The major ecosystems within the boreal zone contain more than 30% of terrestrial carbon stores, thus representing a major component of the global carbon budget. Wildfires are a main disturbance factor in the Canadian boreal forests, which play an important role in boreal ecosystems. In this paper, we mapped the forest burned area in Saskatchewan and Alberta of Canada during 2004-2011, using two data sets of Moderate Resolution Imaging Spectroradiometer (MODIS) Land Surface Temperature (LST) and Enhanced Vegetation Index (EVI) imagery at 1 km resolution. The MODIS global disturbance index algorithm (MGDI) was used to detect the forest burned area. It is based on the consistent radiometric relationship between annual maximum LST and the maximum EVI value that occurs after the LST during the same year at every pixel. The performance of the algorithm was evaluated by comparing the results with the MODIS MCD45A1 products and the inventory data from Forestry Bureau of Canada. The results show that the majority of fire events is captured by the MGDI techniques, and the burned area estimated by the fires pixels is better in good agreement with the inventory data than MODIS products. Compared with the Canadian Forestry Bureau data, the average relative error of our algorithm results is only 8.03%, while the MODIS product results is up to 70.53%. The results estimated that the average burned forest area is about 763 600 ha annually, which is about 3.36% of the total forest in this region. A mean annual burned forest is 468 300 ha for Saskatchewan and 295 300ha for Alberta respectively. The Canadian boreal forest fires show a significant interannual variation and severity, the burned scar mainly appeared in the years of 2006, 2008 and 2010-2011. Forest burned area mainly concentrated in the ecotone of the Boreal Shield, Taiga Shield, Taiga plains and the Wood Buffalo National Park located at the northeast of Boreal Plains.
The Tarim River Basin, especially the downstream area of it is one of the ecologically fragile areasin China. Based on technologies of multi-type, multi-temporal remote sensing imagery, key ecological elements, that is, cultivated land changes in whole Tarim River Basin, the vegetation, sandy and salinized desertification in the green corridor zone of Tarim River mainstream were monitored from 1999 to 2008, after the implementation of comprehensive harnessing engineering for the whole Tarim River Basin. The investigation results show that the ecological environment in the green corridor zone of Tarim mainstream was conspicuously improved, mainly appearing as the rise of vegetation cover rate, the reduction of sandy desertification land area and the weakening of intensity of sandy and salinized land. On the other hand, the cultivated area in the Tarim River Basin significantly increased from 1999 to 2008, which reached 40.34×104 hm2, for 24.13%. Among the whole basin, the cultivated area of the mainstream area increased most quickly. The growth rate of cultivated area of whole basin from 2004 to 2008 is significantly higher than those of the two periods from 1999 to 2002 and from 2002 to 2004. Moreover, the characteristics of sandy and salinized land in upstream of the main channel showed an increase in area of 2.69×104 hm2 and a reduction in increase rate from 2002 to 2004. And sandy desertification land area decreased 1.20×104 hm2 in the same period in lower reach of the basin. Vegetation cover rate of the whole basin apparently increased, especially the surrounding area of the Taitema Lake.Although the ecological environment presented an somewhat improvement after the implementation of comprehensive harnessing in the whole basin, the ecologically fragile environment need long-term supply of water resource and continuous concern from the whole society. The increase of irrigation water resulting from the substantial increase of cultivated area could undertake a long-term potential restraining effect on the harnessing of ecological functions of the Tarim River.
Grassland in Sichuan Province is one of the five most important pastoral areas in China. Available natural grassland accounts for 85% of the total grassland area of Sichuan Province. Accurate grass yield is significant for management of grassland and local economics. In this study, data of MODIS-NDVI/EVI with different resolution (250m, 500m, 1km) in July of 2011 was utilized to model the relationship between grass yield of the four main types (i.e., alpine meadow type, alpine shrub grassland, alpine swamp grassland and mountain drain grassland) and RS data, respectively. Vegetation coverage combined with NDVI/EVI was used to reduce the effect of mixed pixel. The main conclusions are as follows: NDVI was better than EVI to establish the relationship. Data of 250m was ideal among data of different resolution tested. Model for each main type of grassland was better than model for all the four grassland types. Exponential model was better to simulate the relationship compared with other models for the four main types of grassland except for alpine swamp grassland. And for alpine swamp grassland, power function model was better. Optimization model for each of the four main type grassland was obtained. Accuracy of each optimization model was more than 70% while R2 was larger than 0.75. Then these optimization models were utilized to estimate grass yield of Sichuan Province in 2011 and the accuracy of estimation was about 90%. This suggests that these optimization models are feasible and the accuracy is satisfied in practice.
Fractional vegetation cover is an important variable in ecological environment and a key parameter in remote sensing estimation, which is needed in the modeling of the land-atmosphere exchanges of momentum, energy, water, and trace gases. Determination of fractional vegetation cover exactly is necessary for studies on plant transpiration, ground surface evaporation, soil degradation and salinization. Excess green, highlighting vegetation and inhibiting the interference of soil and shadow, was used as a contrast enhancement for identifying plants from soil regions. This study uses modified excess green index to extract fractional vegetation cover by analyzing RGB color features of corn, sorghum, mung beans and weeds growing in saline land in western Jilin Province. The digital images are geometrically corrected in order to eliminate distortion. The automatic extraction approach using modified excess green indexes which is about 40 for vegetation growing on the saline land of western Jilin Province accurately distinguishes vegetation from soil, derives plant and soil binary images, then calculates fractional covers of the four vegetation types. This paper chooses maximum likelihood method to contrast the results of MExG automatic classification. The covers of corn, sorghum, mung beans and weeds calculated by these two methods were compared by visual interpretation and t-test. The visual interpretation shows a very high probability. The t-tests indicate that the means of the four vegetation types extracted by maximum likelihood and MExG automatic classification show a high consistency. In addition, the true values of corn, sorghum, mung bean are obtained by digitizing these images using ArcGIS software to validate MExG approach. The accuracy of MExG method can reach 99%. The results show that the MExG automatic approach which achieves good classification results and is less labor and time intensive than maximum likelihood, can be a viable ground-based method to validate fractional cover products generated by remote sensing.
