CoastGIS'99: geomatics and coastal environment
- different methods have been proposed to calculate coastal lengths
(IHO, 1990) which tend to become a reference unit for instance in the
calculation of equalization funds. Standardisation in terms of reference and procedure choice should be defined more accurately;
- some regulations are defined according to physiographic entities which
are variable in time, such as the water boundary at time “t”. All this
requires the implementation of hydrodynamic models or boundary
representation or extreme zones of extension.
Structuring geometric data
The vector form appears to be the most appropriate as regards the original management of the data concerned. The geometric data structuring must be the result of a comparison between firstly ease of access
to the required functional entities corresponding to the most common
uses, secondly the technical skills of the users in meeting their own requirements and thirdly, the generation and coherent management difficulties
in the updating process.
At the present time, the update and follow up of data propagation
from the operational point of view are proving to be quite tricky (Badard,
1998). Moreover, until now, the arcs available in most marine databases
are not codified, numerous arcs are common to several zones which in
turn overlap, the intersections being rarely broken down into significant
units and zone construction protocols are rarely explicit. For example,
the substitution of a new high watermark, without changing the other
boundaries of a zone, may have serious repercussions if the structuring
of the data has not been pre-programmed.
Ideally, it would be interesting to keep the physical links:
nodes/arc/arc_groupings/polygons/polygon_ groupings. In this case, a
functional “boundary” may be formed from one or several arcs and a
regulatory “zone” may constitute one or several polygons (originating
from the assembly of several functional boundaries).
The possibility of keeping construction links between the different
entities is not a common feature of GIS tools available for administration
use. Only limited management facilities are available in most cases, and
thus point, linear or surface objects are managed in a totally disjointed
manner (for instance ArcView software). In this case only manual
corrective operations are possible in terms of the polygons; unless, of
course, these polygons are reconstructed logically from the boundaries,
but this involves losing the attributes attached to them.
The Arc/Info software offers a solution which is close to the desired model,
in the form of “routes”, thus enabling arc associations to be defined
{i.e. legal boundaries) as well as “regions” formed by a set of arcs demarcating the zone in question or by a set of polygons. However, “a region”
cannot be defined by “routes”, there is no direct construction link
between the two objects.
In order to avoid duplication of common arcs, all the arcs can be assembled together in one single layer (fig. 1, 2).
276
- different methods have been proposed to calculate coastal lengths
(IHO, 1990) which tend to become a reference unit for instance in the
calculation of equalization funds. Standardisation in terms of reference and procedure choice should be defined more accurately;
- some regulations are defined according to physiographic entities which
are variable in time, such as the water boundary at time “t”. All this
requires the implementation of hydrodynamic models or boundary
representation or extreme zones of extension.
Structuring geometric data
The vector form appears to be the most appropriate as regards the original management of the data concerned. The geometric data structuring must be the result of a comparison between firstly ease of access
to the required functional entities corresponding to the most common
uses, secondly the technical skills of the users in meeting their own requirements and thirdly, the generation and coherent management difficulties
in the updating process.
At the present time, the update and follow up of data propagation
from the operational point of view are proving to be quite tricky (Badard,
1998). Moreover, until now, the arcs available in most marine databases
are not codified, numerous arcs are common to several zones which in
turn overlap, the intersections being rarely broken down into significant
units and zone construction protocols are rarely explicit. For example,
the substitution of a new high watermark, without changing the other
boundaries of a zone, may have serious repercussions if the structuring
of the data has not been pre-programmed.
Ideally, it would be interesting to keep the physical links:
nodes/arc/arc_groupings/polygons/polygon_ groupings. In this case, a
functional “boundary” may be formed from one or several arcs and a
regulatory “zone” may constitute one or several polygons (originating
from the assembly of several functional boundaries).
The possibility of keeping construction links between the different
entities is not a common feature of GIS tools available for administration
use. Only limited management facilities are available in most cases, and
thus point, linear or surface objects are managed in a totally disjointed
manner (for instance ArcView software). In this case only manual
corrective operations are possible in terms of the polygons; unless, of
course, these polygons are reconstructed logically from the boundaries,
but this involves losing the attributes attached to them.
The Arc/Info software offers a solution which is close to the desired model,
in the form of “routes”, thus enabling arc associations to be defined
{i.e. legal boundaries) as well as “regions” formed by a set of arcs demarcating the zone in question or by a set of polygons. However, “a region”
cannot be defined by “routes”, there is no direct construction link
between the two objects.
In order to avoid duplication of common arcs, all the arcs can be assembled together in one single layer (fig. 1, 2).
276