As Extensible Markup Language (XML) standards are developed for GIS applications
(led by the OpenGIS Consortium, http://www.opengis.org ) it will be easier for users to
analyze and display multiple geospatial data types from multiple sources as maps
(Schutzberg, 2000).
Although desktop GIS software is generally able to convert data among formats
and projections, IMS technologies are less flexible in combining data in different
projections, and work best with geographic projections. For integrating data, it can
require intensive effort to convert data in a common projection, and to deal with
artifacts from re-projection before different datasets can be displayed together
seamlessly.
Good documentation of metadata, both in the source products and in derived
products, is important for insuring that different data are used appropriately. Without
documentation of methods of production, it is difficult to evaluate how the incorporated
assumptions affect the quality of the product and its application to a particular problem
that may not have been envisioned by the original developer. In particular, issues of
spatial scale and assumptions underlying class definitions are important in making wise
use of shared data.
A key element for being able to combine multiple datasets is the accessibility of
data and data products from a variety of sources. To protect data ownership, many
products are now available for viewing on the Web using (IMS) Internet map server
technology; however, many fewer products are available for download and
incorporation into new analyses. Collaborative agreements may need to be negotiated
with owners of data products before the data can be used independently, and depending
on the needs of the original producer may be focused on scientific credit, control of data
products, or commercial interests. Both commercial and noncommercial data sources
may be seeking to recoup the costs of archive maintenance and data distribution, and
this can limit the accessibility of good global datasets. For example, commercial
products such as Geocover-LC are affordable for use in local studies, but purchasing
hundreds of scenes would be cost-prohibitive for regional or global studies.
5.1
DATA SYNTHESIS AND DISTRIBUTION EXAMPLE: REEFS AT RISK
AND REEFBASE
Led by the World Resources Institute, the Reefs at Risk in Southeast Asia project
had three primary goals: 1) to raise awareness about threats to coral reefs in Southeast
Asia and the linkages between human activity and coral reef condition; 2) to develop a
standardized indicator of likely threats to coral reefs from human activities; and 3) to
promote sharing and improvement of information about coral reefs through data
integration and distribution. Reefs at Risk in Southeast Asia (Burke et al., 2002) was
the first regional study to follow the global Reefs at Risk Assessment (Bryant et al.,
1998), and will be followed by other regional studies using the same approach.
In the Reefs at Risk in Southeast Asia project, information about coral reef status
was quite limited for most locations in the region. After assembling the best available
information on coral reef status and threats to coral reefs for the region, a major
component of the project involved modeling threats to coral reefs to produce a mapbased indicator that identified areas where coral reef degradation might be expected to
occur (or had already occurred) given current levels of human activity. The indicator
also considered mitigating factors such as effective management and the natural
vulnerability of a location to pollution and sediment.
298
Robinson, Andréfouët and Burke
(led by the OpenGIS Consortium, http://www.opengis.org ) it will be easier for users to
analyze and display multiple geospatial data types from multiple sources as maps
(Schutzberg, 2000).
Although desktop GIS software is generally able to convert data among formats
and projections, IMS technologies are less flexible in combining data in different
projections, and work best with geographic projections. For integrating data, it can
require intensive effort to convert data in a common projection, and to deal with
artifacts from re-projection before different datasets can be displayed together
seamlessly.
Good documentation of metadata, both in the source products and in derived
products, is important for insuring that different data are used appropriately. Without
documentation of methods of production, it is difficult to evaluate how the incorporated
assumptions affect the quality of the product and its application to a particular problem
that may not have been envisioned by the original developer. In particular, issues of
spatial scale and assumptions underlying class definitions are important in making wise
use of shared data.
A key element for being able to combine multiple datasets is the accessibility of
data and data products from a variety of sources. To protect data ownership, many
products are now available for viewing on the Web using (IMS) Internet map server
technology; however, many fewer products are available for download and
incorporation into new analyses. Collaborative agreements may need to be negotiated
with owners of data products before the data can be used independently, and depending
on the needs of the original producer may be focused on scientific credit, control of data
products, or commercial interests. Both commercial and noncommercial data sources
may be seeking to recoup the costs of archive maintenance and data distribution, and
this can limit the accessibility of good global datasets. For example, commercial
products such as Geocover-LC are affordable for use in local studies, but purchasing
hundreds of scenes would be cost-prohibitive for regional or global studies.
5.1
DATA SYNTHESIS AND DISTRIBUTION EXAMPLE: REEFS AT RISK
AND REEFBASE
Led by the World Resources Institute, the Reefs at Risk in Southeast Asia project
had three primary goals: 1) to raise awareness about threats to coral reefs in Southeast
Asia and the linkages between human activity and coral reef condition; 2) to develop a
standardized indicator of likely threats to coral reefs from human activities; and 3) to
promote sharing and improvement of information about coral reefs through data
integration and distribution. Reefs at Risk in Southeast Asia (Burke et al., 2002) was
the first regional study to follow the global Reefs at Risk Assessment (Bryant et al.,
1998), and will be followed by other regional studies using the same approach.
In the Reefs at Risk in Southeast Asia project, information about coral reef status
was quite limited for most locations in the region. After assembling the best available
information on coral reef status and threats to coral reefs for the region, a major
component of the project involved modeling threats to coral reefs to produce a mapbased indicator that identified areas where coral reef degradation might be expected to
occur (or had already occurred) given current levels of human activity. The indicator
also considered mitigating factors such as effective management and the natural
vulnerability of a location to pollution and sediment.
298
Robinson, Andréfouët and Burke