You are on page 1of 6

CHAPTER 21

Geomatics for the Management of Oyster Culture Leases and


Production
J. Populus, L. Loubersac, J. Prou, M. Kerdreux, and O. Lemoine
ABSTRACT: The coastal zone of Charente-Maritime in France has a leading position
in oyster culture production. Yet it is being jeopardized by a number of problems such
as overstocking, inadequate distribution of culture types, wild stock in excess, and
strong siltation processes. Some of these problems result from a lack of communication
and realization within the professional community, others due to obsolete management
tools. The introduction of geomatics is believed to help provide some answers, besides
alleviating the classical task of lease management. This chapter shows how the system
was built, the spatial data geo-referenced and linked to administrative data. Examples
of queries on the spatial data base reveal some otherwise invisible facts and allow
prioritization of management decisions. The system also acts as a communication tool
between the actors, i.e. the professionals, the local authorities and the research sector.
Perpectives on further uses are given.
Introduction
Heral (1986) described the relationship between annual productions and oyster biomass
over one century. He concluded that the observed oyster growth decrease was due to
overstocking that limited the carrying capacity of the bay. Moreover, large tidal flats do
offer big opportunities for shellfish cultivation but suffer from high biodeposition rates
(Sornin, 1983). On this basis, spatial management of oyster culture seems to be a key
factor for the future sustainable development of the oyster industry . In relation to
the administration in charge of this management (Affaires Maritimes), IFREMER
developed a GIS in order to provide a useful tool for integration of different spatial
features such as bathymetry, leasing ground location, sediment types, regulations,
261
D.R. Green and S.D. King (eds.), Coastal and Marine Geo-Information Systems, 261274.
2003 Kluwer Academic Publishers. Printed in the Netherlands.

262 J. Populus et al.


phytoplankton and hydrodynamic model simulations, etc... Cross analysis of these
layers should provide adequate advice for the better use of both environment and space.
Material and Methods
THE GIS PLATFORM
ArcView was selected as the GIS platform for the project, whereas Autocad was
selected for the digitization of cadastre maps. During data implementation, the formats
.dwg and .dxf were successively used, then transformed into Arcinfo coverages and
ArcView The shapefile is an ESRI proprietary format that, unlike a
coverage, does not store the topology, which makes it a lot easier to manipulate. When
the topology is needed, it is computed locally. Tabular data remain unchanged from the
coverage format. Raster data may be imported into ArcView and stored either as a
or as an image. Grids, which keep pixel values inside a table, can be
manipulated using the ArcView extension through various
functions : slopes, isolines, enhancement, area summarizing etc...
SCRIPT DEVELOPMENT
When clean data sets are ready for use in an ArcView project, it is desirable for the
end-user to perform his spatial analyses using only resident functions. The software
offers a number of standard facilities, plus a wealth of scripts that only need to be
copied from the library and run for specific purposes. For instance, to make the union
of several identical shape files, a specific script has to be invoked, are not available in
the interface toolkit. When the script has been loaded, if recurrent use is needed, it may
easily be either iconised in the ArcView interface or made resident in the project.
However, as little code as possible was written in the course of the project.
THE REFERENTIAL
In coastal zone mapping the Mercator projection system is the current one used for the

purposes of navigation. Inland, the French IGN (Institut Gographique National) uses
the Lambert conformal conical projection. Considering this, and also its links with
other data sources and its potential users, the latter land-based reference was more
appropriate than a maritime one. The Lambert projection was also adopted by the local
services of the Ministry of Transport in charge of dredging and harbour maintenance,
in particular for depth surveys.
THE INTERPOLATIONS
Interpolation is required when values of a given variable are needed at locations
without measurements. This is the case here for the output of depth soundings surveys,
where point samples have to be made into either a grid file or isolines. It may also

The Management of Oyster Culture Leases and Production 263


concern biological data such as samples of oyster biomass. Geostatistics help us to
understand the spatial structure of a phenomena (a variable) by building its
experimental variogram and applying the model of the structure to make estimates in
unknown places. After building the experimental variogram and interpreting it, a
model of the structure and a neighbourhood are chosen. The calculation of the value at
the unknown grid location is the weighted sum of all samples points contained in the
neighbourhood, the weights being provided by the model. Kriging was performed here
with Isatis, software developed and distributed by Govariances.
The Lease Cadastral Maps
DESCRIPTION
One hundred and seventy cadastral paper maps cover the tidal area of MarennesOlron (Fig. 1), containing 22000 leases for a total surface area of 2900 hectares.
264 J. Populus et al.
Their attribution and management is run by Affaires Maritimes, a body within
the Ministry of Transport. Until recently, this job was still performed by hand, i.e.
graphical corrections on paper maps. The whole system was traditionally georeferenced
to a local coordinate system centered on the spire of the church of the town
of Marennes. Although most sheets overlap correctly, there has never been any control
of the consistency of the whole structure. The system was viable. With
the advent of its digitization and the prospects thereby raised, the question of an
accurate geo-referencing could no more be avoided.
The digitization of the 22000 parcels was contracted to a chartered surveyor,
who delivered the work under the dwg (Autocad) format containing the following
layers: parcels, channels, (banks), coastline and miscellaneous. The sheet and
parcel numbers were attached to each parcel label point. Later, an ArcInfo routine had
to be written in order to transform the polylines and polygons into a topological
structure and to add to each parcel its unique number (a concatenation of the above
two). A comprehensive work of polygon cleaning, dangle correction and other trouble
shooting had to be carried out to end up with a reliable cadastre structure under
ArcView.
GEO-REFERENCING
In order to build a transform polynomial, some conspicuous points known in both
coordinates systems had to be found. The first operation consisted of assessing the
positional accuracy of various church spires, and conspicuous marks surrounding the
basin, which were likely to have been used for positioning. Discrepancies of up to 150
metres were found to exist between distances computed in local coordinates and in the
Lambert projection. As more points were necessary, a field survey was organized in
order to dGPS the few additional marks appearing on some sheets. An accuracy of 2
metres RMS was obtained, typical of a base station not farther than 20 km. The basin
was split in 4 zones where different polynomials were applied using surrounding tiepoints.
Table 1 shows a set of five points in the NE zone which exhibits an average
shift of 18.5 metres. The dispersion of 5.4 metres shows that the positions within the
local referential are not consistent. This will obviously generate a distortion on the

final positions computed by the Helmert polynomial.


However, this is probably the best achievable result using global polynomials
based on only a few points. The alternative would be to work on a sheet basis, in which
case 3 points per sheet would be necessary, a great deal of work. The lack of
conspicuous points (the corner of a lease on the map having no physical existence in
the field) would in all cases limit the accuracy.
BANK MAPPING
Parcels in the field are grouped into entities referred to as which are both
physical units and management units, i.e. under the supervision of a bank committee.
It is therefore quite a relevant geographical entity to be stored in the database. The way
The Management of Oyster Culture Leases and Production 265
to extract banks is as follows. First of all, a query is made of the field
within the administrative data, after a logical join has been
performed with the parcel graphical file. All parcels of a given bank are highlighted.
Spatial merging of all parcels may then be performed which results in one (or several)
global polygon (s). Completion is achieved using the interactive graphical editing tools
of ArcView in order to end up with a polygon, which is labelled after the bank name.
All separate banks are then unioned to form a new shapefile.
More attributes may be appended to the bank entity. Apart from bank
committee data for instance, some biological data are available per bank : this is the
case of the biomass of wild stock for instance.
THE ADMINISTRATIVE DATA BASE
The contains 32 items of information relevant to
each lease, i.e. its unique number, the owner number, the bank it belongs to, its
specific rearing activity, various dates (last movement, begin and end-dates) and so
forth. Another file is the with various civil status data, which may
be linked to the previous one through the owner number.
These data are mostly for administrative use, although some of it may be of
interest for management purposes. Previous experience of a similar data base
construction and management had been gained by the authors with the development of
a GIS for pearl oyster culture in French Polynesia (Chenon, 1990).
This whole semantic data base is centrally maintained by
for the whole of France and it undergoes frequent updates. Therefore, in
order to maintain the data integrity, it is never modified in the project and is exploited
through logical joints with the lease graphical file.
266 J. Populus et al.
The Depth Chart
THE INTERPOLATION METHOD
The whole area covered by oyster and mussel culture inside the Bassin is currently
surveyed by DDE (Harbour Maritime Service), in charge of coastal civil engineering.
Their main concern is a siltation trend that has been occurring for many years and that
is suspected to be the result of extensive oyster breeding on i.e. metal
structures that stick out two feet off the ground. Depth surveys were conducted in 1970,
1985 and recently over the period 1994-96. Soundings were performed every ten
metres on transects 100 metres apart and corrected for tidal effects, to provide a
mapping on scale of 1/10 000. Tidal constraints limited transects ashore to an altitude
of approximately 4 metres above chart datum.
The data are delivered as (X,Y,Z) ascii files, X and Y being Lambert coordinates in
metres and Z in centimetres positive for depths above the chart datum. The variograms
generated by Isatis (de Chambure, 1991; Constantin, 1996) in two orthogonal
directions, i. e. NNW and ENE, both local (range of 500 metres) and egional (range of
1500 metres) are shown in Fig. 2.
The fitting using a Gaussian function is made good up to 400 metres. A slight
of a few centimetres is adopted, in order to avoid matrix inversion
problems. The NNW regional fitting (up to 900 metres) is made to be better than the

ENE one, as sounding profiles are run East-West. The neighbourhood is an oval one,
with 100 and 250 metres as respective horizontal and vertical semi-axis values. The
interpolation is performed on a 20 metre mesh size grid encompassing the whole basin.
A section of the the depth digital terrain model (DTM) is shown on Fig. 3, with depth
varying between -12 and +4 metres.
MERGING DEPTH TO LEASES
The aim of this operation is to provide the average depth of each lease. It is a classical
problem of raster/vector conversion (Durand, 1994), the depth DTM and the leases
being respectively under these forms. The choice of the method depends on the relative
size of the objects concerned. If the parcel size and the DTM pixel are of comparable

The Management of Oyster Culture Leases and Production 267

size, it is recommended to convert pixels into polygons, perform an intersection and


compute the weighted sum of intersected depth polygons contained in each cadastre
polygon. In this present case, with a parcel size of around 40 metres and pixel size of
20 metres, this method is inappropriate, as the intersection of both polygons coverages
would yield over 100000 polygons. It is more reasonable (and much less time
consuming!) to use depth pixel as representative of the pixel value and
compute the average of all contained in a given parcel. This operation has
the advantage of being readily available in ArcView Spatial Analyst
It ends up with the attribute appended to the lease file as shown in table
2 (Elevation in metres is positive above chart datum). Fig. 6 illustrates the case of Banc
de Ronce, with parcels colour-coded according to their elevation.
268 J. Populus et al.
The Tidal Model
Shellfish leasing grounds are located on large tidal mudflats and therefore are
submitted to a variable immersion duration from sea to shore. Due to their filtering
activity, oyster growth depends directly on this parameter. A tidal model is written
under the programming language. The script running on ArcView provides
water level estimates at given times. The calculation is made by using formulae given
by the EPSHOM (Hydrographic Survey) manual. The results were compared to tables
given by EPSHOM and show very little discrepancy. However, this calculation is not
available when the distance with the reference harbour is too large.
This model is used to calculate the immersion duration at each depth. Water
level calculations are recorded in a ASCII file with a six minute time lag. By reading
this file between two given dates another script permits to generate a DBF
file where the frequencies of water level values at a 0,1 meter accuracy are
recorded. The immersion durations for each depth Z are calculated by summing
the frequencies over the values deeper than Z.
The DBF file is joined to every depth table to provide maps of immersion duration,
of oysters is a isometric function of Dtz where
X is a complex function for determination of growth per time unit. SFGz represents the
gain in flesh dry weight for an oyster. In this way, Fig. 4 illustrates the effect of a
theoretical siltation of 0,5 metres on the spatial distribution of immersion duration, in
per cent of loss of immersion time. This calculation provides direct coefficient
for calculation of at two different depths.

The Management of Oyster Culture Leases and Production 269


270 J. Populus et al.
Moreover, a ratio (PDTz) could be calculated by dividing for each depth the
immersion duration by the total duration.
Application to Dredging Plans
THE SEDIMENTOLOGICAL BALANCE OF THE BASIN
Obviously, the sedimentological balance is positive in the southern par of the basin.
Siltation is particularly strong in culture areas where two feet high
metallic structures produce heavy sediment deposition. However, the quantification of
this phenomenon remains very approximate and tools are needed to prioritize dredging

and cleaning operations.


The 1985 depth survey was processed with geostatistics exactly the same way
as described above for the 1994-96 survey. By subtracting the two situations, a
difference map was obtained and transformed into a grid file in ArcView. Using the
operation applied to the bank file, it is possible to display the average
sediment balance per bank. Fig. 5 denotes an almost homogeneous accumulation of
The Management of Oyster Culture Leases and Production 271
over 30 cm on the banc de Bourgeois, whereas deposition is more variable across banc
de Ronce with values between zero and 30 cm.
THE DREDGING OPERATION OF PERQUIS/RONCE
In this area prone to siltation, the idea is to improve the flush effect of the ebb tide
through channels surrounding some of the banks. The Conseil Gnral of Charente
Maritime is currently funding dredging operations carried out by contractors under
supervision of the local Harbour Maritime Service. Two operations are programmed
for impending execution in several phases, the Etier de Perquis in the East (2 phases)
and the Coursire des Lzards around the Ronce bank (Fig. 6).
The site width is 50 metres, which means all parcels intersecting this corridor
have to be expropriated and their owners granted equivalent leases in the vicinity. The
dredging corridor width itself is only 20 metres. The objective is to compute the
sediment volume between the present elevation and the projected one
The volume computation makes use of the depth grid file. The elevation to be
reached is 1.0 metre above chart datum. After keeping only the pixels above 1 metre in
the grid file (by a thresholding query on the grid table). An averaging of all pixels
values contained in the polygon is then performed (using the above-mentioned
tool) which yields its average depth The volume is computed

as:
In Etier de Perquis, the first phase concerns a surface to be dredged of
on an average sediment height of 0.7 metres, that is a volume of In
Coursire de Ronce, are concerned on a height of 1.6 metres, i.e. a volume of
Future Prospects
A number of further uses of the cadastre data base are foreseen.
- In biology, stock assessment is a major issue in this basin. In order to assess the
global biomass, the delineation of homogeneous zones of culture type is needed. As
regulations may not be respected, the only way to approach reality is to interpret
airphotos previously registered to the cadastre file. Every couple of years, B&W
airphoto campaigns are undertaken with simultaneous oyster biomass sampling
- Productivity, mortality, epizooties are also of highest concern. Ways are sought to
couple results of trophic and sedimentological models with field observations in test
parcels, together with the incidence of rainfall sewage outfall
- In socio-economy, there are prospects of a better understanding of the lease exchange
system by computing a lease value based on major bio 272
J. Populus et al.
The Management of Oyster Culture Leases and Production 273
physical parametres (essentially emerging time, position in terms of productivity,
etc...). As market values are known, the comparison of the two might allow to refine
the notion of anticipated parcel value
Conclusion
The shell culture cadastre clearly is an essential element of coastal management in
France, a) for the economic value of shell production, b) for its impact on coastal
environment. Its digitization is underway within Affaires Maritimes, the
Administration in charge, with technical assistance by Ifremer. For lease management
itself, the transition from paper maps to graphical files associated to a semantic data
base will not only alleviate the traditional task of Affaires Maritimes, but also open
new possibilities, namely the spatial illustration of queries made on the data base,
which was otherwise unfeasible.

For wider purposes, several bodies are interested in possessing the digitized
cadastre. Among them, Ifremer is in charge of dealing with biological and
environmental aspects and providing advice to all possible actors in the coastal zone.
The cadastre is a fundamental layer of anticipated GISs to be set up in its various
coastal stations. If only a couple of applications of data mixing were shown in this
274 J. Populus et al.
paper, indeed a host of them are expected in the near future, with a view to improving
decision-making.
References
Chenon, F., Varet, H., Loubersac, L., Grand, S., Hauti, A. (1990).
a Geographic Information System of the Fisheries and Aquaculture
Territorial Department. A tool for a better monitoring of public marine
ownerships and pearl oyster culture. Actes du colloque international
90, Tldtection et milieux insulaires du Pacifique, approche
IFREMER, Nouma-Tahiti, novembre 1990, pages 561-572.
Constantin, V. (1996). Cartographie du Golfe du Lion. Modle numrique de terrain
bathymtrique par les logiciels Isatis et Trismus, rapport IFREMER DRO/GM/96.34, 26 p.
de Chambure, L. (1991) Traitement par numrisation d'une carte bathymtrique et
calcul d'un MNT, Rapport Govariances, 38 av. Franklin Roosevelt, BP91,
77212 AVON, France. FRA/033/149/91, 20p.
Durand, H., Guillaumont, B., Loarer, R., Loubersac, L., Prou, J., Heral, M. (1994). "
An example of GIS potentiality for coastal zone management: pre-selection of
submerged oyster culture areas near Marennes Olron (France)". EARSEL
Workshop on Remote Sensing and GIS for Coastal Zone Management. Delft,
The Netherlands, 24 - 26 oct. 1994, 10 p.
Hral, M., Deslous-Paoli, J.M., Prou, J. (1986). Dynamiques des productions et des
biomasses des hutres creuses cultives (crassostrea angulata et crassostrea
gigas) dans le Bassin de Marennes-Oleron depuis un siecle. CIEM C.M.
1986/F:41.
Sornin, J.M., Feuillet, M., Hral, M., Deslous-Paoli, J.M. (1983). Effets des biodpts
de l'hutre Crassostrea gigas (Thunberg) sur l'accumulation de matires
organiques dans les pares du Bassin de Marennes-Olron. J.Moll.Stud.,Suppt.
12A, 185-197.

You might also like