(Institute of Hydrogeology and Engineering Geology, Ministry of Geology and Mineral Resources, Zhengding, Hebei 050803)
The distribution, occurrence and development of geological disasters have spatial characteristics, and its influencing factors have their own characteristics and complexity. The ghm IAS- geological hazard graphic image analysis system developed by the author is an applied geographic information system with typical geological hazards as the main object and spatial information management and analysis as the main function. The system has a unique spatial analysis model extension, integrates vector, raster and Windows graphical user objects, supports the conversion of multiple data formats, has rich mapping functions and high-quality mapping output, and can quickly generate disaster thematic maps. Using GHMIAS system, the distribution law, occurrence mechanism, influencing factors and development trend of regional geological disasters in Beijing, Tianjin and Tangshan are widely analyzed, and beneficial results are obtained in the study of land subsidence process and future development trend in Tianjin urban area, which shows that GHMIAS system can play an important auxiliary role in solving the spatial distribution characteristics, occurrence and development law and evolution trend prediction of geological disasters.
Geographic information system; Spatial analysis; geological disaster prediction
1 Introduction
Geological disasters are serious natural disasters that endanger human survival. Their distribution, occurrence and development process have spatial characteristics, and their influencing factors have their own characteristics and complexity. However, at present, there are few graphic software systems suitable for the management and analysis of thematic spatial information of geological disasters. "Research on graphic and image aided analysis system of geological disasters" is a special subject set up in the national "Eighth Five-Year Plan" scientific and technological key project. It aims to develop a software system (GHMIAS) with the functions of graphic and image input, storage, processing, display, analysis and output by absorbing and digesting the characteristics of GIS and other graphic analysis systems at home and abroad, and apply it to the evaluation of geological disasters in Beijing, Tianjin and Tangshan as a demonstration.
Geological disasters are always associated with a certain spatial area. The geological disaster space we see is the result of the interaction of various complex factors, including the very complex relationship between natural factors and human factors. The method of geographic information system provides us with the possibility to systematically analyze the spatial relationship of various factors and their action results. The development of this system is based on the analysis method of geographic information system, and establishes a practical system that can manage and analyze the spatial information of geological disasters and serve the prediction, prevention and decision-making of geological disasters.
According to the characteristics of geological disasters, GIS should not only have the general functions of spatial information and attribute data collection, storage, spatial analysis and output, but also provide spatial evaluation and prediction capabilities for specific geological disasters, computer graphic expression technology of geological disasters, legend schema system and corresponding symbol library, thematic map generation and processing and other functions. At the same time, the design of graphic data structure and database system should reflect the distribution characteristics of professional information and make use of professional information as much as possible.
In the demonstration application analysis, through the collection, storage and spatial analysis of corresponding spatial information, this paper tries to provide decision-making basis for the evaluation and prediction of specialized geological disasters from the perspectives of spatial statistics, spatial compounding and spatial model prediction, and provide basic parameter information for other accurate mathematical model analysis.
Development of 2 GHMIAS system
2. 1 supporting environment
Hardware: PC386 or above (Pentium 586 model is recommended); At least 4MB of memory (more than 8MB is recommended); The hard disk has at least 40MB of remaining space; I/O devices supported by Windows system.
Software: Dos5.0 or above; Chinese Windows3. 1 or above, or Chinese Win95.
2.2 GHMIAS system composition
GHMIAS system consists of the following main functional modules: graphic input and editing, graphic decoration and output, gallery space query, graphic space analysis (including graphic statistics, graphic composition, spatial model and other sub-modules), image processing (figure 1), and auxiliary modules such as system help and system function demonstration.
Figure 1 GHMIAS System Structure Schematic Diagram
2.3 main features of ghmias system
As an analytical GIS system focusing on geological disasters, GHMIAS has the following characteristics:
(1)GHMIAS has the basic functions of general GIS, such as graphic input, editing management, query display, analysis and processing, graphic output, etc.
(2)GHMIAS has many spatial analysis models, such as graphic coverage analysis, graphic sequence spatial gray modeling and prediction, etc. The first application of spatial grey prediction model to graphic image analysis system is an extension of one-dimensional nonlinear grey prediction in space, which is of great significance to broaden the types of spatial analysis models and enrich the analysis ability of graphic image system, and is suitable for the demand of spatial analysis and modeling of geological disasters.
(3) The data structure is advanced and reasonable, integrating vector, grid and Windows graphical user objects, adopting the technology of structure contraction and data compression, having the mechanism of * * * storage and topology/direct vector format complementary conversion, supporting the mutual conversion of various data formats, and being able to share data with major GIS systems at home and abroad such as ARC/INFO, IDRISI and SPACEMAN.
(4)GHMIAS adopts microcomputer+Chinese Windows platform mode, with intuitive interface and simple operation, which meets the development trend of software and the needs of popularization and application.
2.4 Hierarchical data model
Data model and data structure are the core of geographic information system and the key to realize its functions completely and flexibly. Geological hazard information system deals with information bodies with certain spatial characteristics and complex attributes. The purpose of its data model and data structure design is to abstract the data structure of professional information and establish the data structure oriented to professional problems, thus laying the foundation for realizing the goal of highly specialized geographic information system.
2.4. 1 project
A project is the highest object of information management established for a specific goal in a specific information field, and it is a collection of information bases related to a specific field and goal. In other words, a project manages and controls the operation of more than one database.
An application task establishes a project, and at the same time, the project environment and database scheme will be initially established. In the next operation, you can gradually modify and expand the following levels of the project through hierarchical management.
Database (database)
A library is a collection of different types of information files recorded in different storage modes under the control of a project. A database consists of multiple information files.
A project can be composed of one or several libraries, and the number of libraries depends on the reasonable library division scheme determined after analyzing the information attributes of the object system.
2.4.3 Documents (documents)
It is the basic unit that the operating system manages the user information body in the computer. In the data structure of this system, the data file is an information body composed of multiple layers.
A graphics library can be composed of multiple graphics files, the number of which is limited by the maximum number of files in the directory supported by the operating system.
2.4.4 layer (layer)
Graphic information files are composed of information with different attributes. In order to distinguish the attributes of information and operate independently, different attribute types need to be managed hierarchically in a graphic file, which is the concept of layer.
A graphic file can contain as many as 256 layers. When operating a graphic file, you can determine the display range of a layer by setting the display properties of the layer.
2.4.5 elements
A graphic element is the smallest unit of graphic information. Developing GIS system in Windows environment can better realize the information recording system combining vector, raster and Windows standard primitives. Among them, user information is generally expressed in the form of points, lines and polygons. , with vector characteristics; Spatial images, photos and other scanned graphics are expressed in grid form; Regular graphic bodies such as rectangles and ellipses provided by Windows system can be used for symbol marking, drawing arrangement and so on. Realizing the storage of three types of pixels can greatly enhance the operability, output simplicity and graphic expression effect of GIS system.
The representation of primitives adopts the way of "primitive header"+"primitive body". The primitive header records the information of the primitive, such as its identity, display attribute, filling attribute, layer it belongs to, user attribute contact person, etc. Through the connection with the related attribute table, the primitive body records the spatial position relationship of the primitive.
In this way, we have established an information structure chain of "project → library → file → layer → primitive". For users, as long as the project is established, they can gradually build a complex information framework and structure under the guidance of the system to form a complete information system network. For the system, once this framework is implemented, this hierarchical structure can meet the requirements of users to update and edit information at any level.
2.5 System modules and functions
2.5. 1 vector graphics input editing module
Processing graphics with multi-document interface mainly includes the following functions.
Files: create, open, close, save, transfer in, transfer out, digitize, connect data, print and exit;
Edit: select, copy, delete, move, rotate, change shape, modify attributes, reduce graphic elements, and modify graphic file headers.
View: full window, zoom in, zoom out, redraw and display control;
Drawing: words, dots, lines, polygons, rectangles, squares, circles, ellipses, etc.
Settings: page size, layer, ruler, text features, line features, filling features, point types, changing color palette, etc.
Output: It supports surface finishing and combined output of vector graphics and raster graphics. It can be used on various output devices (from ordinary needle printers and laser printers to large pen plotters and large color inkjet plotters), and the output quality has reached a high level.
Help: index, theme setting help, terminology, information about this module.
2.5.2 raster graphic analysis module
On the basis of absorbing the advantages of related software at home and abroad, this module has the following main functions.
Documents: new drawings, opening, closing, scanning, etc.
Edit: modify, copy, delete, etc.
Drawing: basically the same as vector subsystem, but stored in grid format;
Operations: vector raster conversion, graphic attribute query, image file header modification, graphic assignment, attribute extraction, graphic reclassification, enlargement, reduction, transposition, splicing, window opening, filtering, etc.
Graphic statistics: histogram analysis, cross list, regression analysis, autocorrelation analysis, trend analysis, random image generation, etc.
Graphic algebra: graphic coverage, constant operation, area calculation, perimeter calculation, etc.
Spatial model: grey prediction model, distance model, cost surface, optimal path, excavation and filling analysis, classification, surface analysis, viewpoint analysis, watershed analysis, etc.
Image processing module
It mainly absorbs the functions of other systems and supports image recognition, classification, standardization, false color synthesis, striping, filtering, principal component analysis, fuzzy matrix analysis and other operations.
2.5.4 Spatial retrieval module of graphic image library
This module is a retrieval and query system of Beijing-Tianjin-Tangshan geological disaster graphic image database. After entering this module, use the mouse to query the graphic image library information of any point on the map of Beijing, Tianjin, Tang, Qin and the whole region.
2.5.5 System Guide and Help Module
Help module is a convenient tool for users to learn to use this system. This module and the help functions in each module constitute the help support system of GH-MIAS, which enables users to get corresponding help in the process of initial contact, operation and use, and after reaching proficiency, solve difficult problems that need to be understood in the use process, and obtain information such as data structure and file structure.
Application of 3 GHMIAS system in spatial analysis of land subsidence in Tianjin
3. 1 Temporal and spatial statistics and evaluation of water level change
The existing research shows that the main reason for the acceleration of land subsidence in Tianjin is that "groundwater overexploitation makes the groundwater level continue to decline". With the efforts of Tianjin geologists and various parties, the land subsidence was effectively controlled after the groundwater exploitation was reduced in the 1980s.
As the direct inducing factor of land subsidence, a lot of analysis and research have been done on the change of groundwater level in the past, but the method of direct statistical analysis of observation data, that is, the analysis of discrete data, is usually adopted. But the characteristics of real-world information are constantly changing in space and time. Using the principle of GHMIAS spatial analysis, we can achieve accurate spatial statistics of this kind of information (although our collection of real world information is discontinuous and discrete, with the support of GHMIAS system, we can use discrete interpolation or isoline interpolation to generate a simulated continuous spatial surface from discrete information and reflect the spatial characteristics of information in the form of "quasi-reality"). This is superior to the previous analysis form in analytical ability and accuracy, so it is easy to obtain more valuable supporting basis for the decision-making process.
The spatial analysis of water level change in Tianjin is based on the water level observation data of the second and third water-bearing formations 1980, 1985 and 1988 in Tianjin. The analysis process is shown in Figure 2.
Fig. 2 Basic process of spatio-temporal graphic analysis of land subsidence in Tianjin (taking water level analysis as an example)
The water level observation information extracted from the attribute database is processed by the discrete data interpolation surface of GHMIAS system to generate the groundwater level map (sketch) of each period, and then the graph coverage-constant operation is carried out by the grid space analysis tool of GHMIAS system to obtain the spatial distribution characteristic map of water level change at different time intervals, and then 1980, 1985,10.
From the statistical results of spatial analysis (table 1), it can be seen that the change of groundwater level in Tianjin urban area in the 1980 s, whether in the second water-bearing group or the third water-bearing group, the downward trend of groundwater level is weakening. As far as the second water-bearing formation is concerned, whether it is the spatial absolute eigenvalue (maximum value, minimum value and average value) of water level change, the ratio of falling area to rising area, or the volume change of water level change reflected in water-bearing space, etc. Both of them are developing in the direction of rising water level. Compared with the early 1980s, the total amount of water level rising has exceeded the total amount of water level falling, which led to 80 years.
Table1Characteristic value of spatial fluctuation of groundwater level in Tianjin urban and suburban areas in 1980s
* Negative value is the relative decline of water level, while positive value is the relative appreciation of water level. The rise of water level at the end of the generation is the dominant process. The situation of the third water-bearing group is similar to that of the second water-bearing group, but the change amplitude is not as obvious as that of the second water-bearing group, so that the overall change of water level in the early and late 1980s is still dominated by water level decline (the spatial statistical average is the decline value), and the total amount of water level decline still exceeds the total amount of water level rise (the equilibrium value of water-bearing spatial volume change is the volume decrease). This shows that the control measures of the third water-bearing group are not as powerful as those of the second water-bearing group.
3.2 Spatial analysis and evaluation of land subsidence change
Fig. 3 the spatial volume change of land subsidence in Tianjin urban area 1985 ~ 1992.
Table 2 1985 to 1992 Numerical Statistics of Settlement Distribution
* Negative value refers to rebound.
Based on the same principle as the spatial analysis of water level (see Figure 2), the spatial characteristics of land subsidence observation information in Tianjin 1985, 1988, 1990 and 1992 are analyzed. Statistics of spatial analysis results (Figure 3, Figure 4, Table 2) show that Tianjin's land subsidence control action has achieved remarkable results around 1988, and the soil volume compression caused by land subsidence has obviously decreased. After 1988, it entered a relatively stable period, and the settlement growth rate reached the minimum near 1990, but reached 65438+. The amount of land subsidence has increased slightly, which may mean that the effect of controlling subsidence has been almost played, but the settlement acceleration caused by new settlement factors has not been well controlled.
Fig. 4 The negative sign of the spatial statistical value of land subsidence classification distribution area 1985 ~ 1992 indicates rebound.
From the change of spatial distribution of settlement rebound, the same result is obtained, that is, the ground rebound driven by settlement reaches the maximum near 1988, and then begins to fall back, and the ground rebound of 1992 is obviously reduced compared with 1990. This also shows that after entering the 1990s, the land subsidence situation in Tianjin is still not optimistic. If no further measures are taken to control the land subsidence, the land subsidence will be aggravated again.
Since 1988, the area of land subsidence deceleration zone in Tianjin has been shrinking, from 1985 ~ 65438+549.34km2 in 0988 to 4 18.00km2 in1988 ~/990. However, the area of land subsidence acceleration zone is increasing, from 268.98km2 in 1985 ~ 1988 to134.09km2 and 1990+.
The absolute statistics of the distribution of annual settlement variation values in the time interval (Table 3) also show that the rebound is decreasing and the settlement is increasing after 1988. Compared with 1990, the distribution of land subsidence in the whole region (including urban and suburban areas) increased by 2.46mm on average.
Table 3 Statistics of Annual Settlement Change Range
* Negative value is settlement deceleration, while positive value is settlement acceleration.
Fig. 5 The negative sign of the spatial statistical distribution of annual settlement changes in different periods indicates rebound.
From the perspective of spatial distribution, the distribution of land subsidence centers and the variation value of land subsidence in different periods have also changed greatly. Compared with 1985 in 1988, the land subsidence speed in the study area is weakened in an all-round way, and the central area with the greatest weakening degree is distributed in the workers' new village, Fang Xinzhuang, perfume factory and other areas on the east side of the urban area. Compared with 1988 in 1990, the settlement in most areas has little change. Among them, the settlement in most central urban areas and the area of Limingzhuang-Huantuo in the northeast of the study area increased slightly, while the settlement in other areas continued to weaken. Compared with 1990 in 1992, the settlement rate in most areas continues to be basically stable, with the settlement in the northeast and northwest directions of the central city and the study area slightly weakened, while the settlement in the southwest corner of the study area obviously increased.
3.3 Grey Spatial Model Prediction of Land Subsidence
It is difficult to establish an accurate spatial analysis model of land subsidence prediction because the spatial distribution and time evolution information of land subsidence related factors we have is not systematic enough. The land subsidence observation sequence information managed by GH-MIAS system includes four periods: 1985, 1988, 1990 and 1992. Basically, we can use the spatial distribution characteristics of this information.
3.3. 1 Grey Prediction of Basic Characteristics of Land Subsidence 1994
According to the results of spatial statistical analysis, some characteristic values of future settlement changes can be predicted by grey method. Please refer to the description of the algorithm principle of the spatial analysis module for the theoretical method and data utilization technology of the prediction process. The prediction results are shown in Table 4.
Table 4: Estimated subsidence characteristic values of 1994 and 1996, and characteristic values of land subsidence change between 1992 and1994.
According to the prediction results, the land subsidence in Tianjin urban area in 1994 is about 5409538m3, and the predicted land rebound will be reduced to 6757.4m3 The minimum spatial distribution of land subsidence in 1994 is-1.8mm (the ground rebound is 1.8mm), and the maximum is. Because the data time series is short and the model's correction ability is limited, the deviation range of the prediction results is large. However, the comparison between the actual value and the calculated value of the prediction model shows that the fitting situation is ideal, which shows that it is reliable in the development trend, consistent with the previous statistical analysis results and has certain reference value.
3. 3. 2 1994 and 1996 Spatial prediction and distribution analysis of land subsidence.
The analysis process is as follows:
(1) Extract the ground subsidence observation data of 1985, 1988, 1990 and 1992 from the attribute database;
(2) Using the surface interpolation function of discrete point data of GHMIAS system, the settlement characteristic surfaces (sketches) of four periods are established;
(3) Run the grey spatial pattern prediction analysis option in the grid spatial analysis tool of GHMIAS system, select the four time periods of the generated surface to participate in the analysis, and generate the spatial characteristic surfaces for predicting land subsidence 1994 and 1996 (omitted);
(4) Analyze the spatial interval of residuals, determine the confidence interval of model prediction, and evaluate the accuracy and reliability of model prediction;
(5) Using the grid graphics analysis function of GHMIAS system, the forecast result graphics are statistically analyzed and extracted by layers, and the area and volume are calculated, and the statistical calculation results are analyzed.
The overall evaluation of the prediction and analysis results of land subsidence in Tianjin urban area 1994 and 1996 with grey space graphics is as follows:
(1) Compared with 1992, the land subsidence situation in Tianjin urban area will be basically stable, and the suburban subsidence will increase, resulting in a slight increase in the total subsidence level in the whole region. The average distribution of land subsidence in 1994 is 19.77 mm, and the average distribution of land subsidence in 1996 is 24.09mm, so it is necessary to strengthen measures to control land subsidence.
(2) The settlement acceleration center will further move to Hua Zhuangzi area in the southwest corner of the study area, and the settlement in the southeast of the study area may also increase during 1996.
(3) Compared with the previous direct gray prediction of the characteristic values of land subsidence distribution, it can basically be considered that the spatial prediction results of SGM model are enhanced to the extreme due to the limited length of time series, that is, the predicted maximum and minimum values of spatial distribution may be quite different from the actual values, but the average subsidence level is basically consistent with the results of characteristic analysis. The average value of settlement distribution of 1994 is 19.77 mm, and that of 1996 is 24.09mm, which is basically consistent with the results of direct gray prediction of eigenvalues, (1994 19.38mm,1999).
refer to
[1] Shi Jiansheng. Spatial analysis of the relationship between active faults and karst ground collapse in Tangshan city. China Journal of Geological Hazards and Prevention,1996,7 (3): 87 ~ 91.
[2] Wu Lun et al .. Guide to Geographic Information System. Peking University Publishing House, 1994, 1~5.
[3] J.T. Coppock et al ... Spatial analysis methods in GIS. International Journal of Geographic Information System,1990,4 (3).
[4] Michael F, Good Boy, Karen K Kemp, Introduction to Geographic Information System. NCGIA core courses, 199 1.
[5] Hu Huimin et al .. The development of land subsidence in major cities in northern China. China Journal of Geological Hazards and Prevention, 199 1, 2(4): 1~9.
[6] Cui Xiaodong et al. Mathematical model and numerical calculation of land subsidence in Tianjin. Geological hazards, Proceedings of Beijing International Symposium, 199 1, 3 10~3 12.
[7] Jin Dongxi, Niu Xiujun. Land subsidence in tianjin and its countermeasures. Geological hazards, Proceedings of Beijing International Symposium, 199 1, 3 13~3 18.
[8] Li minglang Main geological environment problems in Tianjin. China Journal of Geological Hazards and Prevention,1994,5 (3).
[9] Deng Julong. Grey control system. Huazhong University of Science and Technology Press, 1985.