Geographical Information System Processing - Essay Example

? GIS GIS Introduction The quality of data in relation to GIS processing has in the recent past raised a lot of concerns especially among GIS specialists. This is as a result of the GIS software influx in the market and the increased use of GIS technology in varied decision making and problem solving roles resulting to a lot of scrutiny with regards to its quality and reliability. The main area of concern about GIS is its relative error in its methodologies. This has led to the coming up with several recommendations that are practical which assist in the location of sources of possible errors and define the data quality. Sources of errors in GIS can be classified into operational and inherent, with the two responsible for reducing the products quality generated using geographical information systems. Inherent errors are normally witnessed in source data and documents. Operational errors on the other hand result from GIS manipulation functions and data capture. Possible operational errors sources include; digitizing human error, thematic maps areas mis-labelling, horizontal boundaries misplacement, classified error, human bias and inaccuracies in GIS algorithms. The main of GIS processing is to identify the possible error in data sources as well as minimizing the error amount resulting from the processing. Cost constraints make it easier to avoid errors rather than eliminating them afterwards. A GIS (geographical information system) is a tool that is computerised and is used for both mapping and analysis of existing geographical phenomenon and occurring events on the Earth surface. The technology behind GIS entails the integration of operations of common database like statistical and query analysis with exceptional geographical analysis and visualization benefits provided by maps. It is this ability that makes GIS stand out from the rest of information systems and makes it valuable to most of the private and public enterprises in predicting outcomes and explaining events strategy planning. Data accuracy mostly depends on the original input data quality as well as its precision when the data is being processed. This follows the many states in which inaccuracies and errors may occur within the GIS database. The most common sources of errors include; field measurement inaccuracies, use of equipments that are inaccurate, or in recoding procedures that are not correct. This implies that, higher accuracy calls for higher quality in the initial data and processing that is more precise. However, the two will increase the systems costs. Theoretically, the quality of GIS data is a comprise between the cost involved and the needs. Practically, the choice goes down to what is available at that time or what is acquirable in a reasonable cost or amount of time. The two will determine the quality of working level of a given dataset. DIGITIZING АND SCАNNING OF MАPS Maps are often scanned so as to make use digital data as their bases for other information on vector map as well as to have the scanned data converted to vector data which can be used in vector GIS. For scanning to be successful, the map to be scanned has to have clear defined text, lines and symbols; be of good quality, be free from extraneous strains and have their width lines being greater than 0.1 mm (Bolstad, 2005). The scanning process consists of binary encoding and scanning. In normal cases, scanning results in an 8-bit gray scale regular pixel.Pixels appear in columns and rows and every pixel coordinates are normally identified by the row and column number after which the number is converted to normal coordinates using transformation process. Now that maps that have been scanned do not have any information on the area structures or inner linear (Bolstad, 2005), it is impossible to associate attributes to structures in an effort to define whether the landmark in question is a river or roads. Reclassification involving cells grouping can be conducted on the maps that have already been scanned. Maps that have been scanned can also be vectorized so as to make use of GIS systems that are based on vector. Vectorization is vulnerable to defects and errors such as interruptions or deformations of lines which intersect at nodes, extraneous particles or stains on the map before it is scanned, alphanumeric text and information, line breaks that are not intended leading to vectors that are divided, dotted-line symbols leading to several small vectors, and jagged smooth curves. Features labelling in layers that are vectorized is important in order to distinguish between both the classes and subclasses. Taking an example of a line feature representing a river or road, such a line has to be properly labelled (Borrough, 2000). Automatic Scanning Several scanning devices have been introduced and are used in the automatic capturing of spatial data. The advantage of such techniques is that they are in a position to capture spatial features in any map even at a rapid high speed. Despite this advantage, scanning has not been recognized as a suitable alternative of most implementations of GIS. Scanners have proved to be expensive in their acquisition and operation. In addition, most devices used in scanning have limitations in relation to the capturing of selected features like symbol and text recognition. Most of the scanned data need a lot of manual editing before coming up with a data layer that is clean (Mailing, 2002). Manual Digitising Manual digitising has been identified as the most common approach of having map data transferred onto a GIS. In addition, the approach has also been identified as the most common source when it comes to positional error. Several factors are put together in manual digitizing approach to make it almost impossible for the user to follow the exact line shape being digitized. Such factors include; strain, hand wobble, overshoot or undershoot tendency and coordination of the eye and hand. Human digitising can be categorized into two types; physiological and psychological. Psychological errors entail hardships in perceiving the real line centre that is being digitized as well as difficulties experienced to accurately move the cursor-cross hair along it (Chang, 2006). The two will result in an offset consistent lateral arising between the digitized and true lines and overshoot or undershoot. Physiological errors are caused by involuntary muscles spasms resulting in displacements that are more random as well as polygonal knots. Some of the errors experienced in the output of the digitizer can be spotted easily like knots and spikes while others like consistent displacement to either direction of the real line prove to be problematic. It however goes down to the individual digitizer to determine how consistent their errors will turn out to be as they are in a position to come up with an error signature for every operator and go ahead to correct the error accordingly. Positional error differences can also be visible between the two digitising modes; point and stream mode. Point mode digitizing has it that a sample points set that has been sparsely but carefully picked can be based on in the creation of a line representation that is more faithful. The more the point’s numbers registered, the more details the operator achieves. This implies that the line shape that is recorded more overall and faithful will turn out to be more accurate but still the positional accuracy errors of single points remain the same (Lo and Yeung, 2002). The error resulting from the map document registration on the digitising table has also to be taken into consideration. In cases where such digitization is carelessly or incorrectly done then the entire co-ordinate set that is being digitized from the map will have systematic error. It is therefore essential for any digitiser to take great care during map document setting. Automatic Digitizing The error resulting from automatic digitising are different compared to those in manual digitizing although the sampling and registration errors remain the same. There are two automatic digitizing methods and they include; raster scanning and digital line following with the two resulting in raster and vector output respectively. In digital line following, the positional error is changed to machine tolerance making it more accurate with reference to manual digitizing while the source document quality seriously affects the produced data. Breaks are normally witnessed in the followed lines like counter lines used to permit the insertion of height labels, can result in problems. Error problems related to devices used in raster scanning are mainly related to the size of pixel used in the raster. The smaller the size of the raster the less the rasterising error will be witnessed, however, the amount of storage required will be greater now that vast data volumes may be produced (Maguire, 2005). The main problem with the two automatic digitising methods is that although less positional errors are experienced compared to the case in manual digitising, it is more difficult to clean up the output data and have it integrated into the database of GIS. Although it is not common for persistent errors to be of a greater magnitude, manual digitizing is more preferred as a more favourable data capture technique for paper maps. Digitizing on a Mountain Environment Uncertainties and errors are common cartographic features thus such aspects are also applicable in digitizing analogue maps. GIS has been identified as a powerful tool that can be applied in spatial analysis and it has the ability to increases both the importance and magnitude of errors dramatically in spatial databases. There are three main groups that have been identified as to govern errors associated with the processing of spatial data. They include; human error, errors caused by ordinal measurement and natural variations and processing errors (Goovaerts, 2000). The GIS application errors specifically to regions that are mountainous are significant following the terrain in such regions. However, no matter the terrain of the region under study, no map can be completely without errors. Errors such as aerial Interpolation error, measurement error and positional error increase with increase in slope thus are common in mountain environment. Calculations of parameters such as population density are normally exaggerated as the areas under study are normally underestimated following the sloping terrain. The sloping terrain actual surface area is normally greater that what is depicted by the geographical area on the map representing the earth surface as a flat surface. The actual area is reached at by getting the product of the average slope angle Cosine and the geographical area of the area under study. Linear calculation error in mountainous region: the over sloping land distances happen to be larger in reality that what is implied on the maps. The actual distances in such a case can be reached at through getting the product of the traversed average slope angle Cosine and the distance mapped. This means that the distanced given by GIS is wrong like lower drainage density and road distances that are shorter (Agresti, 2000). Following this, the costs deduced for road construction or transmission line will end up being a value that has been underestimated even in cases where the difference between the values is not that large. Mountainous regions also experience position errors following the shorter distance from GIS considering the mountainous terrain making point placement in accordance to linear measurement to be erroneous when used on sloping terrain. For example, points used as samples located 100 miles from a chosen point and their position reached using GIS in a two dimensional surface, the position that will be calculated will be far way than 100 miles as a result of the slope involved. Buffer zone errors are also prone in such region as the underestimation of linear distances result in a buffer zone from GIS which ignores the slope of land that would enclose a wider region than what was actually intended. The proportion of the added distance introduced into the buffer zone is directly proportional to the land’s slope angle. Mountainous regions are also prone to straight line accessibility derivation ineffectiveness as in the case of a plain land, the derivation is reached at simply by the generation of the multiple buffer from the line which represents the road for an area and from a point representing a settlement centre used in the evaluation of the village accessibility to various service centres. This approach is not applicable in mountain region where physical impediments like slope have to be taken into account while using larger scales than 1:50, 000 (Aronoff, 2000). Some of the accessibility maps have been developed by taking into consideration the counters but the presence of slopes that are un-traversable is not considered. LAND SURVEY Land survey method of data collection has been in existence for a very long time contrary to the GIS technology which is considered as a relatively new method. The digital data used by GIS has in the recent past become a main analysis and mapping tool for a number of applications in many fields. The data in GIS mapping systems are applicable in myriad uses like land survey research, land assessment, land planning and development and economic development (Cressie, 2000). The irony of this application is that as the data used in GIS is driven by its own geographic component in an ideal manner, little emphasis was placed on the accuracy of data in the early GIS development which is a key factor in land surveying. In the early days of its introduction, GIS did not take into consideration high standard data accuracy (Blachut, 2000). However, this trend has changed over the years with the development of geographical systems following the increased applications among local government as well as advances in software and hardware capabilities. In modern times, many GIS technology users expect their data to have a higher accuracy degree. Improved GPS accuracy capabilities, greater clarity in digital aerial photography, and higher data quality expectations are all responsible for this current situation. Many GIS users are also finding their original data less than satisfactory, and many early GIS programs are beginning to integrate land surveying protocols and methodologies into their mapping system to improve the overall accuracy of GIS data (Brimicombe, 2003). Different companies have come out strongly striving to integrate the protocols on accurate land survey within GIS applications. The result of such initiatives is a dramatic increase in the number of expertise and land surveyor who rely heavily on GIS application in their project development. GIS has turned out to be one of the most accurate tool applied in survey research. Land Information System Application GIS technology can be applied in different environments land information system such as infrastructure. Some of the areas of application include water, transportation and wastewater infrastructure. In such cases, GIS can be used to provide solution to multitiered project in cities. Such a project will bring together professional land surveyors from different and multiple disciplines, civil engineers and a firm in charge of aerial photography. A project like this will require mapping systems that are highly accurate (Buongiorno and Gilless, 2003). In such a project, the initial stage will be coming up with a GIS base map to be used in coordinating position for every part of the project and provide coordinate data that are highly accurate through GPS survey grade and coded attributed data to represent each point (Buongiorno and Gilless, 2003). The base map will then be developed into base map layers using highway plans, corner documents, and subdivision plats. The other data layers that will be applicable in this case include accurate railroad and road centerlines, index grids, blocks, lots, water boundaries as well as land ownership boundaries in the chosen city (Borrough and Frank, 2003). A city limit boundary that has to be comprehensive will also have to be developed basing on the annexation ordinances of the city. All the above information and data will then be prepared in a GIS geodatabase format. Conclusion In the past, most developers and users of GIS paid very little attention to the issues of inaccuracy, error and imprecision in the collection and application of geographical information. This resulted to the data captured and applied at that time to be full of imprecision and inaccuracy yet these GIS problems were not considered to be a great deal. Over the years, the situation has been changing and it has now been recognized that errors, imprecision and inaccuracy can result to more harm than good in any GIS project. This means if such errors are not checked, the GIS analysis results will be rendered worthless. The situation is complicated following the fact that the error devolves from the main GIS strength. GIS has the advantage of being able to collate as well as cross-reference different data types by location. The sad thing is that, for every new imported dataset, the GIS inherit the related error. Such errors may mix up with already existing errors in the database in ways that cannot be predicted. The idea here is that despite the fact that errors come in the way of accurate data analyses, ways have been brought forth to minimize the error through careful methods and planning for estimating GIS solutions effects. Awareness of the error problem is one of the useful ways of having GIS practitioners to be more sensitive to possible GIS limitations to achieve impossibly precise and accurate solutions. References Agresti, A. (2000). Categorical Data Analysis. New York: Wiley Aronoff, S. (2000). Geographic Information System: A Management Perspective. New York: Ottawa Blachut, T et al. (2000). Urban Surveying and Mapping. New York: Springer Bolstad, P. (2005). GIS Fundamentals: A First Text on Geographic Information Systems. Manhattan: Eider Press Borrough, P & Frank, A (2003). Geographic Objects With Indeterminate Boundaries. London: Taylor & Francis Borrough, P & McDonnell, A. (2000). Principles of Geographical Information Systems. Oxford: Oxford University Press Borrough, P. (2000). Principles of Geographical Information Systems for Land Resource Assessment. London: Clarendon Press Brimicombe, A. (2003). GIS Environmental Modeling and Engineering. London: Taylor & Francis Buongiorno, J & Gilless, J. (2003) Decision Methods for Forest resources Managers. Amsterdam: Academic Press Chang, K. (2006). Introduction to Geographic Information Systems. New York: McGraw Hill Cressie, N . (2000). Statistics For Spatial Data. New York: Wiley Goovaerts, P. (2000). Geostastistics for Natural Resources Evaluation. Oxford: Oxford University Press Lo, P & Yeung, K. (2002). Concepts and Techniques of Geographic Information Systems. New York: Prentice Hall Maguire, J. (2005). Geographic Information Systems and Science. New Jersey: Hoboken Mailing, D (2002). Measurement From Maps. New York: Pergamon Press Read More

 

Read More