D.R. Green and S.T. Ray
114
undertaking similar studies, by suggesting to what extent studies, such as the artificial
reef siting, can safely make use of existing datasets without risking the problems
associated with judgements based on inadequate information, and generated or inherent
error.
Introduction
Until quite recently most applications of Geographical Information Systems (GIS) have
focused on land-based studies rather than marine and coastal environments. A
monograph by Bartlett (1994) covers some of the wide range of applications undertaken
to date, whilst other papers by e.g. Fairfield, 1987; Davis and Davis, 1988; Riddell,
1992; Deakin and Diment, 1994; and Green, 1994a,b,c; 1995) illustrate some of the
specific ways in which GIS has been used to date. With increasing interest now being
shown in our marine and coastal environments, more GIS-based applications,
particularly those developing Spatial Decision Support Systems (SDSS) (see, for
example, Raal and Davids, 1995; Canessa and Keller, 1997) are being developed.
Demand from end-users has subsequently placed greater pressure on commercial
software developers to provide more user-friendly front-ends to GIS systems, e.g.
ArcView for ArcInfo, with the aim of assisting, e.g. coastal zone managers, to use this
technology in their work environment. Interest in the benefits of communications and
networking technology e.g. the Internet and the World Wide Web (WWW), has
recently been examined by a number of studies (e.g. Green, 1996; 1997).
As GIS becomes a more 'user-friendly' tool, and more people, not necessarily
GIS specialists, recognise the benefits of the technology in their work environment,
greater use will undoubtedly be made of GIS. Whilst this is good in itself, the ease with
which a GIS can now be used tends to overshadow the complexity of the technology
and the potential difficulties that can (and do) arise when the technology is not fully
understood. It is, for example, relatively easy to acquire digital data, to input, store,
manipulate and display this data, and the results of any analysis in a practical sense. The
problem is that there is seldom any mention of data quality associated with GIS
analyses.
In practice there are two routes open to the application of GIS technology in
environmental studies. The first is to collect entirely new data for the research. The
second is to make use of existing available datasets, both analog and digital. In an ideal
world ‘starting from scratch’ is probably the best approach to any research, except
where historical trends are an important component. Acquiring and subsequently using
the 'right data' for the 'right job' is then almost a certainty. The appropriate scale, spatial
sampling, and areal boundaries are then collected for the specific application in mind.
More often than not, however, there are many datasets already available in both
analogue and digital format which could potentially be used for other applications, but
as an aid to the problem, rather than for quantitative analysis. Unfortunately whilst some
datasets may be of use for an environmental application, many may not in fact be
appropriate, and may have limited use in practice.
A GIS is relatively simple to develop in theory and the data is easy to acquire,
whether it is archival or new, for a wide variety of applications. But, in practice, there
are many fundamental considerations to be taken into account when undertaking a GIS
114
undertaking similar studies, by suggesting to what extent studies, such as the artificial
reef siting, can safely make use of existing datasets without risking the problems
associated with judgements based on inadequate information, and generated or inherent
error.
Introduction
Until quite recently most applications of Geographical Information Systems (GIS) have
focused on land-based studies rather than marine and coastal environments. A
monograph by Bartlett (1994) covers some of the wide range of applications undertaken
to date, whilst other papers by e.g. Fairfield, 1987; Davis and Davis, 1988; Riddell,
1992; Deakin and Diment, 1994; and Green, 1994a,b,c; 1995) illustrate some of the
specific ways in which GIS has been used to date. With increasing interest now being
shown in our marine and coastal environments, more GIS-based applications,
particularly those developing Spatial Decision Support Systems (SDSS) (see, for
example, Raal and Davids, 1995; Canessa and Keller, 1997) are being developed.
Demand from end-users has subsequently placed greater pressure on commercial
software developers to provide more user-friendly front-ends to GIS systems, e.g.
ArcView for ArcInfo, with the aim of assisting, e.g. coastal zone managers, to use this
technology in their work environment. Interest in the benefits of communications and
networking technology e.g. the Internet and the World Wide Web (WWW), has
recently been examined by a number of studies (e.g. Green, 1996; 1997).
As GIS becomes a more 'user-friendly' tool, and more people, not necessarily
GIS specialists, recognise the benefits of the technology in their work environment,
greater use will undoubtedly be made of GIS. Whilst this is good in itself, the ease with
which a GIS can now be used tends to overshadow the complexity of the technology
and the potential difficulties that can (and do) arise when the technology is not fully
understood. It is, for example, relatively easy to acquire digital data, to input, store,
manipulate and display this data, and the results of any analysis in a practical sense. The
problem is that there is seldom any mention of data quality associated with GIS
analyses.
In practice there are two routes open to the application of GIS technology in
environmental studies. The first is to collect entirely new data for the research. The
second is to make use of existing available datasets, both analog and digital. In an ideal
world ‘starting from scratch’ is probably the best approach to any research, except
where historical trends are an important component. Acquiring and subsequently using
the 'right data' for the 'right job' is then almost a certainty. The appropriate scale, spatial
sampling, and areal boundaries are then collected for the specific application in mind.
More often than not, however, there are many datasets already available in both
analogue and digital format which could potentially be used for other applications, but
as an aid to the problem, rather than for quantitative analysis. Unfortunately whilst some
datasets may be of use for an environmental application, many may not in fact be
appropriate, and may have limited use in practice.
A GIS is relatively simple to develop in theory and the data is easy to acquire,
whether it is archival or new, for a wide variety of applications. But, in practice, there
are many fundamental considerations to be taken into account when undertaking a GIS