French Guiana landscape map 2012

This landscape dataset was built using Shuttle Radar Topography Mission (SRTM) data with full resolution (1 arc sec ~ 30m). This map aims at evaluating the geodiversity of natural areas. Twelve landscape types into 5 main categories have been identied with respect to Migon's typology (2006): the coastal plains landscapes (AA. AB. AC); the typical multi-convex landscapes (B. I. J); the multi-concave and joint-valley landscapes (C.D); the more or less dissected plateau (E.F.G) and the all-slopes topography (H). The map was produced at the request of the French Guianan public forest manager of the Office National des Forêts (ONF).

Date (Creation): 2012-01-01

Identifier: CEBA_landscape2012

Status: Completed

Point of contact

Office national des forêts (ONF) - Guitet, Stéphane

Maintenance and update frequency: Not planned

GEMET - INSPIRE themes, version 1.0 INSPIRE themes

Geology

GEMET - Concepts, version 2.4

relief (land)
geomorphology

Geounits

South America
Americas
France
French Guiana

Keywords

national coverage
labex CEBA

Place

Use limitation: Please get in touch with a point of contact.

Access constraints: Intellectual property rights

Classification: Unclassified

Spatial representation type: Grid

Denominator: 1000000

Metadata language: eng

Character set: UTF8

Topic category

Geoscientific information

Begin date: 2012-01-01

End date: 2012-01-01

Supplemental Information: Bibliography: Guitet S.. Cornu J.-F.. Brunaux O.. Betbeder J.. Carozza J.-M. Richard-Hansen C.(2013) : Landform & landscape mapping. French Guiana (South America). Journal of Map. DOI:10.1080/17445647.2013.785371

Reference system identifier: EPSG / WGS 84 / UTM zone 22N (EPSG:32622) / 7.9

Distribution format

ESRI Shapefile ( 1.0 )

OnLine resource: Guitet S.. Cornu J.-F.. Brunaux O.. Betbeder J.. Carozza J.-M. Richard-Hansen C.(2013) : Landform & landscape mapping. French Guiana (South America). Journal of Map. DOI:10.1080/17445647.2013.785371 ( WWW:LINK-1.0-http--related )

Hierarchy level: Dataset

Statement: The landscape map was computed by analyzing the spatial distribution of the different types of landforms. This analysis was conducted in five steps. At first. a manual expert segmentation was performed based on landforms spatial distribution, searching to draw homogenous region based on repeated pattern or dominant type. Secondly, 3 local indices were computed from the landforms map on 1.5 x 1.5 km sliding windows: the Shannon-weaver diversity index, the contagion index, and the majority index (O'Neill et al.. 1988, Mc Garigal et al.. 2002). Thirdly, each boundary of the manual segmentation was compared with maps of the 3 indices for validation. In case of inconsistency, the landscape segmentation was modified in the light of the spatial variability of these indices. Then another Principal Components Analysis was performed with 14 variables computed for each region (average Shannon index. average Contagion index and relative proportions of the 12 different types). A hierarchical cluster analysis was performed to assign each unit to a natural landscape type. At last, permutations tests were used to highlight relations between landscape and landforms type and to interpret them. The different maps and models were compared to topographic field data that have been collected on 92 transects totalizing 260 km length. These data including natural habitat descriptions, topographic profiles, and soil collections were collected on 24 sites distributed through French Guiana in order to represent the maximum of geographic and ecological variability. Habitat descriptions are used to validate the hydromorphy model. Slope, elevation levels and topographic position occurrence measured on the transect help to interpret landforms and landscape classes. A Garmin GPS (60CSX) was used to geolocalize the data. Slope angles and distances were measured with Vertex laser rangefinder. Landscape analysis was conducted with Fragstat freeware v2 (Mc Garigal et al.. 2002) and statistical analyses with the R1.9 package (Team. 2009). The landscape map. based on landforms analysis. has to be used at the scale 1:1 000 000 to 1:2 000 000.