Local government authorities (LGA) in the UK
undertake a range of core statutory tasks which include decision
making processes involving allocations of finite resources. Such
a core function is the determination, or “deciding”,
of development proposals relating to land resources. Land development
decisions are undertaken within a democratic decision process
which involves the views of expert and non-expert stakeholders.
The study examines, using a rural LPA as a case study, the use
of spatial information in the decision process and how environmental
Assets are represented and included within the process; and how
straightforward GIS customisations can create simple SDSS support
for spatial decision making under uncertainty using expert information.
Problem areas that are identified, and that would provide fundamental
difficulties for the implementation of more sophisticated spatial
decision support “tools”, are discussed and an approach
to providing a simple but extensible SDSS for use by decision
makers and expert stakeholders is described.
This dissertation will consider the enterprise
GIS model as a method of delivering consistent spatial data in
a geographically distributed work environment. The overall framework
of this study will be based on the geosciences in general and
mineral exploration in particular. Exploration and mining is,
by its nature, a geographically distributed industry making it
a good candidate for an enterprise GIS. Enterprise GIS will be
considered from the point of view that it has the potential of
becoming a strategic decision-making tool deployed within the
structure of traditional DBMS.
The paper has been laid out with chapters 2 to 5 presenting the
background components important to an enterprise GIS implementation
and chapters 6 to ten focusing on implementation and a series
of three case studies.
Geographic information has a significant role
to play in the modernising government agenda. Two projects of
a spatial nature are the National Land and Property Gazetteer
and the National Land Information Service. Both are time consuming
to develop and both will have an impact on business processes
within local authorities. The main issue considered by this study
is that the development of GIS within local authorities, for example
creating new datasets and implementing browser versions of GIS
software, is being affected because GIS staff are having to compile
Local Land and Property Gazetteers and capture the datasets required
for automatic completion of land charge searches. The affect of
the type of authority and the size of authority, in terms of both
population and area, is also considered. In addition to a questionnaire
survey sent to all street naming and numbering authorities in
England and Wales, case studies of two local authorities and their
approaches to NLPG and GIS are discussed.
In 1998 the government of British Columbia,
Canada formalized the Muskwa-Kechika Management Area act. This
act set aside 6.4 million acres in north-eastern British Columbia
as one of the world’s largest integrated resources management
areas in the world. The objective is to preserve ecosystems including
habitats and all values and still allow economically viable extraction
of resources. Imperative to the success of such a venture is a
current inventory of all resources.
Among these resources are the rangelands, which
need to be managed to ensure that livestock and wildlife have
the forage they need to thrive, without causing degradation of
the vegetation. Managers currently have questions regarding the
types of changes in vegetation which have occurred over time,
particularly in relation to fire suppression and large numbers
of non-native bison. Because this study area is extremely remote
it is accessible by only helicopter in many areas. Remote sensing
offers the only viable solution to studying vegetation in this
area.
This aim of this study is to provide a current
vegetation inventory using Landsat images as well as to investigate
the methods of post classification comparison and principal components
analysis as vegetation change detection techniques. Landsat images
for the years 1986, 1994 and 1999 were obtained. The 1999 Landsat
image was classified using the PCI ISODATA unsupervised classification
routine. Overall classification accuracy was 92% and the overall
Kappa statistic was 0.892%. The 1999 image acted as the reference
data for the 1994 and 1986 images, which were also classified
using ISODATA and classification maps were produced. The classified
images were placed into ArcGrid and subtracted by pairs 1986-1994,
1994-1999, 1986-1999, to create change detection maps. Standardized
principal components analysis was run using IDRISI software in
the same three date sets to produce 6 principal components each.
The three components from each set (1986-1994, 1994-1999, 1986-1999)
exhibiting the most change information were used to create change
images. These changes images correlated well with the change detection
maps derived from the post classification comparisons.
The resulting information was useful and helped
to confirm suspicions that grassland and alpine type vegetation
was decreasing, whilst forest type areas were on the increase.
More reference data should be collected and there are other methods
of change detection which might provide better results. Undoubtedly
there is great potential for remote sensing to be used as a tool
for inventory data collection, change detection and trend modelling.
The GI User Community (GUC) is starting to exploit
Extensible Markup Language (XML) for data transfer, data configuration
and data encoding purposes. Research revealed that XML is currently
used for augmentation to traditional business conducted with enterprise
spatial databases, or as a data transfer format.
In the enterprise setting, XML is used to retrieve
data from a spatial database and present the data to a device
or web page, in a format compatible with that device or web page.
An example of XML use within a data transfer format is the deployment
of Open GIS Consortium (OGC) Geography Markup Language (GML) within
the Ordnance Survey (OS) Mastermap product.
The aim of this thesis was to investigate the
possibility of storing and managing spatial data in XML format,
as opposed to only using XML as a means to configure or transfer
data. The use of a native XML database system (XDBMS) to store,
manage and distribute spatial data in XML format was proposed,
the hypothesis being that such a method of storage would preserve
the XML data in its original format and achieve complete interoperability
of spatial data.
Two experimental spatial databases were built
to store sample Mastermap data, in order to test the hypothesis.
A prototype system was built with Software AG Tamino XDBMS. A
benchmark system was built using an established method for the
storage and management of spatial data, namely Environmental Systems
Research Institute Spatial Database Engine (ESRI ArcSDE). The
prototype system was assessed for performance, management and
distribution of spatial data against the benchmark system, using
identified requirements for building spatial databases.
The experiment revealed potential for the use
of an XDBMS in enterprise spatial data management, particularly
with regard to interoperability and organisational autonomy. Further
research and development work required to optimise and exploit
such a system was also identified.
The thesis assesses biodiversity in Nunavut,
Canada using ArcView 3.2a GIS software and the Image Analysis,
Patch Analyst and Spatial Analyst extensions. Biodiversity is
measured at a variety of spatial scales to provide a baseline
biodiversity assessment that may then be used for environmental
planning purposes in Nunaut, Canada. The biodiversity models created
for bird species and land cover diversity indices have not been
verified by field observations and therefore caution should be
used when making management decisions based on the results without
first conducting proper field sampling programs aimed at verifying
these models. A landscape-level analysis, conducted for Nunavut’s
49 ecoregions, provides insight into landscape, AVHRR land cover
class and patch metrics that affect biodiversity.
Coastal habitats, wetlands and terrestrial riparian
habitats near fresh water sources appear to have high diversity
based on the results for land cover and bird species diversity
assessments. Another general trend that emerges, and is expected
based upon biogeographical theory, is a decrease in diversity
from the more vertically complex forest and shrub covered south
and south-west ecosystems in Nunavut to barren, ice-covered and
sparsely vegetated areas in the north and north-east. Diversity
is also higher in regions that have a high degree of topographic
relief due to quickly changing plant cover over short distances.
Trees are an important part of our urban environment.
Many National bodies (The National Urban Forestry Unit NUFU,1998)
now widely recognize the many benefits that trees bring to the
urban environment. These benefits include pollution reduction,
soil water management, energy savings as well as aesthetic improvements.
Land owners have a ‘duty of care’
to manage their tree stock in a responsible manner. Remote Sensing
in the form of colour aerial photography provides us with a means
of collecting urban tree data. This data collection has traditionally
been a slow manual process.
Traditional aerial photography techniques (including
the use and description of colour, shape, size and shadow) are
proved difficult to obtain and fully characterize in the urban
environment. Automated methods of interpretation analysis namely
texture analysis and fuzzy classification are also hampered by
the complex mosaic of agglomerated urban features. More specific
tree interpretation techniques are proved ineffective at differentiating
between species, however the differences in deciduous tree species
are often subtle. Tree measurement is possible to a certain extent
but only for unrestricted canopy views.
An evaluation of risk analysis modelling is
made, taking into consideration the complex urban environment.
An integrated data source approach is recommended.
The thesis details a methodology developed to
produce monthly 1km groundwater recharge maps covering the major
aquifers in England and Wales, using a study area covering most
of southern England.
The main spatial datasets required for analysis
are monthly rainfall and monthly evaporation, and percentage runoff
from the soil. The method also requires a model that distributes
the effective rainfall through time, as it travels downward through
the unsaturated zone to the groundwater below. These variables
chosen have the biggest effect on the amount of recharge to an
aquifer.
The water available for groundwater recharge,
calculated from 1km rainfall, evaporation and percentage runoff
grids, was distributed through time using th distribution functions
generated from the groundwater model. The groundwater model uses
point borehole water level data to calculate the recharge distribution
parameters for distributing the monthly water available for recharge
through time. Many methods of spatially interpolating the distribution
functions are tested; however thiessen polygons are used for the
final analysis. Although the thiessen polygon method is not ideal
in terms of spatial accuracy, the highly viariable nature of recharge
across a short distance made the production of other interpolation
methods difficult. Further work on spatially interpolating the
recharge distribution parameters is required, however initial
broadscale groundwater recharge results are sensible in water
balance terms and in relation to previous studies on groundwater
recharge.
Vehicle telematics combines the knowledge of
a diverse group of industries with modern technologies to produce
a range of services that offer numerous enhancements to vehicle
users in such areas as navigation, entertainment and safety. With
people spending ever-increasing amounts of time in vehicles either
as a driver or passenger, there is a captive market to which services
can be offered.
To understand the constraints in the design
and deployment of vehicle telematic applications this paper takes
into consideration all of the components that have influences
upon its development. Each sector of the economy, be they governments,
users or service providers have inputs into how the technology
is developed and utilised. The choice of technologies available
and which specific technologies are chosen for implementation
in the final design will also influence what services are available
to the system. It is by analysing these two market segments, the
industries and the technologies, along with implied constraints
that conclusions can be drawn about what services will be offered
and where the technology will be headed.
It is through the development of open standards
and the reduction of costs involved with the creation of telematic
products and services that will move the technology from a luxury
product to a mainstream product in wide use. Then with further
development and deployment of advanced technologies such as 3G
wireless technologies, the industry will alter again, enabling
the provision of more services from sources external to the vehicle.
This will not be an overnight alteration to the automotive industry,
it takes time to advance from an initial idea to market saturation.
The purpose of this dissertation is to define
the optimal business model and organizational structure for a
shared local government regional GIS.
The author creates a framework to help structure a literature
review. The framework identifies motivations and benefits for
creating a regional GIS, factors that aid or inhibit GIS implementation,
and classical organizational structures. Findings for these three
categories result in a preliminary organizational model, which
provides a background or context to formulate relevant questions
for subsequent case studies. Secondly, the model derived from
the literature research became a milepost or standard to compare
with the surveyed organizations. Eight local government shared
GIS’s are surveyed and profiled as case studies. Based on
the lessons learned from these organizations a final organizational
model is developed.
The study concludes that motivations and benefits
of a shared GIS result in a mandate that subsequently determines
the tasks to be carried out. Once the tasks are identified, it
is possible to derive an organizational structure. Factors that
aid or inhibit GIS implementation for such entities are similar
to stand alone GIS’s, with no unique factors discovered
for a shared local government GIS. An adhocracy type of organizational;
structure is identified in the literature research and confirmed
through case studies as being a viable model for a shared local
government GIS.
The author discovers that organizations are
very adaptable and that if the initial startup configuration is
not optimal the organization will identify this and change (optimize)
accordingly. While the author is able to offer a framework or
model for use by organizations embarking on a shared local government
GIS, he cautions that the internal organizational model (the staff
and their tasks) will vary and change according to the mandate.
Traditional methods of designing the alignment
of a new railway involve many days or weeks of pouring over topographic
and thematic mapping to find a route that costs the least in terms
of money, time, damage to the environment etc. Least Cost Pathway
(LCP) techniques seem to provide a fast, measurable and rigorous
alternative, however, the use of these functions without first
understanding how the algorithm interacts with the data may lead
to erroneous results and thus runs the risk of rejection by the
already sceptical engineering industry.
After an investigation into the factors that
need to be taken into account when designing the route of a new
railway, it can be seen that the LCP technique falls short in
two areas. The first involves the avoidance of undulating land
to reduce the requirement for the cut & fill of large quantities
of material. This requirement may be satisfied by a manual pre-processing
step. However, the promotion of LCP techniques is so that such
operations are removed from the design process.
The second factor is the need to impose a minimum
radius of curvature condition so as not to impact on the design
speed of the track. It is suggested that a LCP function may be
employed in an iterative process whereby the LCP is derived in
the usual manner and then analysed to determine if the minimum
radius condition has been broken. If it has then a high penalty
value is added to the input friction surface at the point of failure
and the LCP routine is run again. The process continues until
a route is found that satisfies the minimum radius condition.
At this stage the results of the procedure are
encouraging; however, the benefit of the technique can not be
fully determined until it is employed in parallel with traditional
methods on a real, railway alignment design project.
In several European countries, the national
mapping agencies have to finance at least a part of their expenses
by selling spatial data to the community. This makes it more important
than ever to have efficient and reliable systems for the retrieval
and delivery of spatial data to the users. These retrievals and
deliveries are often based on the operational system in the production
lines. This is the situation in the Danish National Survey and
Cadastre . This situation is not optimal – overlaying different
maps as the cadastre and topographic base map is hampered by the
circumstance that the data are to be retrieved from different
heterogeneous data sources and the production systems are affected
of the burden from the queries.
The purpose of this dissertation is to evaluate
if it is within the bounds of possibility to establish a GeoData
warehouse, where the different base maps regardless of there origin
are brought together in a common database, based on a standard
relational database management system (RDBMS) with spatial extensions.
The reason for such a warehouse is to make the retrieval and delivery
of spatial data more efficient, and to offload the operational
system with the burden from data delivery.
The queries in a database that is the basis
for a retrieval and delivery system will be very unpredictable
much like the ones used for Online Analytical Processing (OLAP)
in the business world. Because of this, it was decided to evaluate
the use of data warehousing as it is known from the world of OLAP
by doing a research in data warehousing, including the similarities
and differences in data warehouses used in the world of business
and data warehouses intended to be used for spatial data.
Because the data model has a great influence on the performance,
and on how the database can be used, there has been done research
in the different options for the data base scheme. The data model
can be based on either a normal Entity-Relation schema (E-R)
Groundwater recharge, defined as the entry of
water in to the saturated zone, is one of the most important and
complex hydrologic parameters to quantify. This process is dependent
on a large scale of controlling parameters and may vary significantly
in temporal and spatial scales. A sound knowledge of the spatially
variable distribution of groundwater recharge is a prerequisite
for sustainable groundwater resource management.
Due to the complexity of the process of groundwater
recharge and the lack of appropriate techniques for direct measurements
of recharge various methods have been developed to estimate recharge
rates. The most commonly used methods and models as well as current
GIS-based approaches for estimating regional groundwater recharge
are reviewed in this thesis. Cho9osing an appropriate modelling
approach for recharge estimation is a crucial part of regional
recharge studies. A thorough understanding of the attributes of
the different approaches is a prerequisite for their proper application.
In order to determine the spatial distribution
of mean annual groundwater recharge in the state of Hesse (Germany)
am empirically-based soil water balance model is integrated with
the GIS Arc/Info. In this raster-based modelling approach analysis
is carried out on a cell-by-cell basis by making use of Arc/Info’s
programming language as well as integrated map algebra-based functions.
The establishment of a common and suitable input database is required
for the application of the recharge model. The numerous steps
of data preparation and pre-processing that are necessary prior
to model application are outlined, giving an idea how time-consuming
this part of a recharge study may be.
The modelling results of this project reveal
that in the area of Hesse the process of direct groundwater recharge
by precipitation occurs predominantly during the winter months.
The spatial distribution of groundwater recharge is primarily
controlled by the distribution of precipitation and the topographic
characteristics of the study area. High recharge rates can generally
be observed in all mountainous areas, whereas low recharge rates
predominantly occur in the lowlands and river valleys.
For the validation and evaluation of modelling
results on groundwater recharge all simplifications and limitations
of the applied model have to be taken into consideration. This
thesis shows that the comparison of different modelling results
as well as the comparison of modelling results with lysimeter
data may be not suitable for validating modelling recharge estimates.
Because of the uncertainties associated with modelled recharge
estimates the use of many different and complementing modelling
approaches is recommended.
Classical cartographers work hard to obtain
perfection in presenting spatial information using graphics. They
differentiate street crossings with fine cut. Multilined-styles
are used and digitized without breadth in a fine shape. Text is
placed with art and honour. Their art tends to be done using raster.
By contrast, GIS analysts put this in the "background"
and instead analyse their thematics over and over. However. They
both utilise the same fundamental spatial data. This dissertation
examines the creation of a common system, termed here a "mycelium"
("SINGLE" spatial database), which can be used by both
thus enabling cartographers and GIS analysts to live in a symbiotic
manner with benificial synergy effects. A cartographic map has
endless information. Each map has its "analyzeable-scale"
(Kelnhofer 1996). This dissertation utilises this information
and opens further steps towards "MultiMedia Cartography"
(examples see Cartwright, Peterson, Gartner 1999).
The Austrian 1:50.000 Map series (KM50) is
produced using CAD systems and then additional raster software-modules.
The initial result is a MicroStation design file, which consists
of lines, areas, cells and text elements with linestyle libraries,
hatches, cell libraries for signatures and text-fonts. The final
product , however, is layered in raster format (wood, contours,
drainage, rocks, situation and additional thematics) for handing
over on CD's or printed to classic paper maps. There are 213 map
sheets in Gauß-Krüger Projection with three meridians.
Only some of the maps are fully vectorized. This dissertation
considers the using GeoInformationSystems (GIS) at the provisional
CAD map file stage preserving the same map styling but introducing
additional GIS functionality. This is termed "KMGIS50".
Initially different international map catalogues
are compared and this results in a first specified draft for an
Austrian 1:50.000 "KMGIS" mapping catalogue. The aim
is an international compatible data structure with flexible exchanges
to users' needs. Next an example of a GIS driven vector raw data
interface to the final rastered map 1:50.000 is examined. This
is discussed in terms of a "mycelic systematism between a
cartographic view and the geoinformation view" by which is
meant a common underlying system for both the cartographer and
the GIS analyst permitting a synergy in map production and use.
This study aims to investigate whether visual
impact assessment (VIA) in a GIS environment can help visual simulation
of the urban environment. It is mainly divided into four parts:
(a) literature review on the theory and practice of VIA and GIS
application in VIA; (b) four detailed case studies; (c) a GIS
implementation and (d) analysis and conclusion.
Literature review reveals that there are three
main domains in VIA, namely analysis, visualisation and decision
making domains. The limitation of GIS application in VIA stems
back from its 2D boundary mapping paradigm.
The four detailed case studies reveal that the
nature of business and professional background have major bearing
in the choice of tools and information in VIA. In addition, the
choice of software is a response to the general problem of lack
of tools and data in VIA. There is no major difference between
firms and organizations in the public and private sectors in the
VIA practice.
In the GIS implementation, it has been found
that 3D Analyst, a popular GIS software, cannot handle complex
3D shapes and the VIA analysis is restricted to TIN and GRID formats.
These weaknesses have significant bearing to VIA application in
a high-rise high-density urban environment in Hong Kong.
Due to the current limitations of GIS software,
VIA in a GIS environment can only partially meet the requirements
for the visual simulation of urban environment.
Transportation investments can have different
impacts on land value, most of them being positive. But the whole
project process, from construction to operation, can show different
impacts on land value. Studies have concluded that environmental
externalities such as noise, ground vibration, airborne pollution
and visual obstruction have a negative impact on residential values.
On the other hand accessibility generated by the location of a
rail transit station can have a positive effect on residential
value. Investigations performed on this subject take as the main
analytical tool statistical software to test their hypothesis.
A very reduced group have included the geospatial vision of problems
that GIS can provide as an investigative tool.
Living the experience of the construction of
the first metropolitan rail transit system in Puerto Rico, the
present research highlight the use of GIS to model Tren Urbano
nearest residential values and see if to this point the speculation
raised had have an influence on the value of land, residences
and structures along its corridor. Is expected to consider this
research as a base to compare value changes before and after Tren
Urbano is operational.
The methodology of investigation establishes
a 400 metres buffer area as the primary market area of study.
The 400 metres distance is based on Law 207 from year 2000. This
law gives the power to the Puerto Rico’s Highway and Transportation
Authority to work in conjunction with private and public entities
on development plans that help restructure land use pattern along
Tren Urbano alignment in order to intensify land use in the region.
The law establishes a 500 metres buffer as the intervention zone.
This research reduces this zone to 400 metres because transportation
planners have concluded that this is the optimum distance a person
is willing to walk to a transit station. A secondary 800 metres
buffer is establish to have more data for the statistical operations
to be executed. Residential sales data for 1995 and 2001 were
used to model residential values on both periods of times.
Mapinfo with SPSS (Statistical Package for the
Social Sciences) were used. Statistical functions were used to
test the reliability of the 1995 and 2001 data. Statistical functions
were also used to evaluate the correlation between residential
values and distance to the nearest station for both periods of
time under consideration.
At the end a residential value change maps was
calculated using both 1995 and 2001 residential sales databases.
For this purpose Arcview GIS with Spatial Analyst module was selected.
This map along with the statistical analysis and data from the
Planning Board gave the results discussed in the final chapter.
Organisations involved in the management
of the environment need access to each other's data. However,
access to spatial environmental data are often constrained by
issues of cost, manpower, data quality, data classification, data
standards, copyright, organisational problems and data knowledge.
To overcome these constraints, many countries are in the process
of integrating a Spatial Data Infrastructure (SDI), improving
the management of environmental information by making the best
use of digital map data.This dissertation examines how the establishment
of an SDI would help to overcome these constraints in the country
of Zimbabwe, focussing in particular on habitat data.
The Southern African Development Community (SADC)
is providing support to Zimbabwe's spatial environmental data
community, in particular by establishing an SDI in the whole of
the SADC region in Southern Africa.
A national SDI effectively integrated in
Zimbabwe will be part of the wider regional SDI in the SADC region,
and these in turn will contribute to the proposed Global SDI. The
Global SDI will affect and support, through considering needs, the
regional and national SDI's.
Through programmes of work to support environmental
information systems in African countries and taking an overview
on these countries' needs, EIS-AFRICA is also helping to promote
moves towards SDI's.
 Top of page
|
