CoastGIS’99: geomatics and coastal environment
75 t.km - 2 .yr -1 (Maneux et al., 1999b). Nevertheless, as the regional
model shows that there is no significant correlation between sediment
yields and other parameters, such empirical models must be used with
precaution.
However, at the regional level, the empirical model using the Soil Erosion Index, I
ero
, can estimate specific sediment yields and mean interannual TSM fluxes to the CZ quite reliably in basin areas exceeding
2 500 km 2 . For now, just as the TSM flux databases are not precise
enough, the spatial databases available to develop such empirical models
are not representative enough. Thus, in future, we will have to improve
the temporal representativeness of climatic ancl I.UC databases (integrating
seasonal variations) and to improve the thematic descriptions of LUC
and soil classes.
In spite of these facts, the study allowed us to assess the contribution
of river sediment discharge to the Bay of Biscay. TSM river flux measurements and extrapolation from environmental databases, according
to our empirical model, make it possible to estimate TSM river fluxes
to CZ, even when no TSM data were available (fig. 4).
Small mountainous rivers (19000 km 2 ) contribute to 1.9 Mt.yr - 1 , i.e.
40% of the total sediment input (4.7 Mt.yr - 1 ) to the CZ in the southern
part of the Bay of Biscay. Moreover, while the Garonne river has a larger
contributive area (53 000 knr’), its suspendecl secliment input into the
ocean is only 47% (2.2 Mt.yr - 1 ) of the total flux. Therefore, we emphasize
the importance of small mountainous rivers for sedimentological and
biogeochemical budget assessment in CZ.
Conclusions
On two different basin scales, the GIS approach and spatial analysis of
the soil erosion risk make it possible to develop empirical models for
river sediment yields. On a global scale, with a wide range of observed
sediment yield values, this method allows us to determine the most significant controlling parameters. On a regional scale, the range of values,
both for sediment yields and basin features, are too small to determine the
significant parameters.
The representativeness of available databases had to be improved, but
the main problem is that the sediment yield reliability decreases when
the homogeneity of watershed features (such as LUC) increases. Thus,
we propose that m future studies, if the main objective remains the
study of the main watersheds, a reliable sediment yield assessment
should also be made for small representative watersheds with drainage
areas around 1 000 square kilometres.
Finally, this study allowed us to extrapolate an average value of
100 t.km - 2 .yr _1 for the Cantabrian (Spain) and Basque coasts
(19000 km
2 ). This extrapolation highlights the fact that small coastal
rivers are the first source (52%) of TSM inputs to these coastal zones.
Thus, although the Garonne river is a major source of TSM to these
coastal zones (42%), the Pyrenean and Cantabrian small mountainous
252
75 t.km - 2 .yr -1 (Maneux et al., 1999b). Nevertheless, as the regional
model shows that there is no significant correlation between sediment
yields and other parameters, such empirical models must be used with
precaution.
However, at the regional level, the empirical model using the Soil Erosion Index, I
ero
, can estimate specific sediment yields and mean interannual TSM fluxes to the CZ quite reliably in basin areas exceeding
2 500 km 2 . For now, just as the TSM flux databases are not precise
enough, the spatial databases available to develop such empirical models
are not representative enough. Thus, in future, we will have to improve
the temporal representativeness of climatic ancl I.UC databases (integrating
seasonal variations) and to improve the thematic descriptions of LUC
and soil classes.
In spite of these facts, the study allowed us to assess the contribution
of river sediment discharge to the Bay of Biscay. TSM river flux measurements and extrapolation from environmental databases, according
to our empirical model, make it possible to estimate TSM river fluxes
to CZ, even when no TSM data were available (fig. 4).
Small mountainous rivers (19000 km 2 ) contribute to 1.9 Mt.yr - 1 , i.e.
40% of the total sediment input (4.7 Mt.yr - 1 ) to the CZ in the southern
part of the Bay of Biscay. Moreover, while the Garonne river has a larger
contributive area (53 000 knr’), its suspendecl secliment input into the
ocean is only 47% (2.2 Mt.yr - 1 ) of the total flux. Therefore, we emphasize
the importance of small mountainous rivers for sedimentological and
biogeochemical budget assessment in CZ.
Conclusions
On two different basin scales, the GIS approach and spatial analysis of
the soil erosion risk make it possible to develop empirical models for
river sediment yields. On a global scale, with a wide range of observed
sediment yield values, this method allows us to determine the most significant controlling parameters. On a regional scale, the range of values,
both for sediment yields and basin features, are too small to determine the
significant parameters.
The representativeness of available databases had to be improved, but
the main problem is that the sediment yield reliability decreases when
the homogeneity of watershed features (such as LUC) increases. Thus,
we propose that m future studies, if the main objective remains the
study of the main watersheds, a reliable sediment yield assessment
should also be made for small representative watersheds with drainage
areas around 1 000 square kilometres.
Finally, this study allowed us to extrapolate an average value of
100 t.km - 2 .yr _1 for the Cantabrian (Spain) and Basque coasts
(19000 km
2 ). This extrapolation highlights the fact that small coastal
rivers are the first source (52%) of TSM inputs to these coastal zones.
Thus, although the Garonne river is a major source of TSM to these
coastal zones (42%), the Pyrenean and Cantabrian small mountainous
252