North Chin - Myanmar - 2020-2021 - Land cover map
The land cover maps published here were produced for four regions of the northern part of Chin state in Myanmar: Hakha, Falam, Tedim and Thantlang. This work was carried out as part of the ALIVE FNS project to monitor land use (forests, cultivated land, built-up areas, etc.). We used the Moringa processing chain, which is based on satellite imagery (Sentinel 2 free of charge time series and SPOT6-7 very high spatial resolution) and a supervised classification algorithm (Random Forest) trained on a reference database made of polygons associated with a land cover class. Generally, this database comes from ground GPS surveys, but it can be replaced by photo-interpretation of very high spatial resolution images if field collection is unavailable or impossible, as it is the case here in the State of Chin. The database was therefore obtained by photo-interpretation of Spot6/7 images acquired as part of the Dinamis programme. <br> The nomenclature includes 4 crop classes (irrigated crops - mainly rice, shifting cultivation, new shifting cultivation, old shifting cultivation) and 6 non-crop classes (open spaces with little or no vegetation, herbaceous vegetation, shrubland, wooded vegetation, water, built-up areas). The maps are available, for the years 2020 and 2021, at a spatial resolution of 1.5 m over the parts covered by SPOT6/7 imagery (approximately half of the study area) and at a spatial resolution of 10m using only Sentinel-2 imagery over the whole area comprising the 4 regions: Hakha, Falam, Tedim, Thantlang. <br> The overall and class accuracies (f-score) of the maps are available in a text file included in the archive containing the maps.
| Main Authors: | , , , |
|---|---|
| Format: | Dataset biblioteca |
| Language: | English |
| Published: |
CIRAD Dataverse
2022
|
| Subjects: | Agricultural Sciences, Computer and Information Science, Earth and Environmental Sciences, télédétection, Sentinel-2, Environnement, environment, spatial database, base de données spatiale, remote sensing, |
| Online Access: | https://doi.org/10.18167/DVN1/S8Q4LV |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | The land cover maps published here were produced for four regions of the northern part of Chin state in Myanmar: Hakha, Falam, Tedim and Thantlang. This work was carried out as part of the ALIVE FNS project to monitor land use (forests, cultivated land, built-up areas, etc.). We used the Moringa processing chain, which is based on satellite imagery (Sentinel 2 free of charge time series and SPOT6-7 very high spatial resolution) and a supervised classification algorithm (Random Forest) trained on a reference database made of polygons associated with a land cover class. Generally, this database comes from ground GPS surveys, but it can be replaced by photo-interpretation of very high spatial resolution images if field collection is unavailable or impossible, as it is the case here in the State of Chin. The database was therefore obtained by photo-interpretation of Spot6/7 images acquired as part of the Dinamis programme.
<br>
The nomenclature includes 4 crop classes (irrigated crops - mainly rice, shifting cultivation, new shifting cultivation, old shifting cultivation) and 6 non-crop classes (open spaces with little or no vegetation, herbaceous vegetation, shrubland, wooded vegetation, water, built-up areas). The maps are available, for the years 2020 and 2021, at a spatial resolution of 1.5 m over the parts covered by SPOT6/7 imagery (approximately half of the study area) and at a spatial resolution of 10m using only Sentinel-2 imagery over the whole area comprising the 4 regions: Hakha, Falam, Tedim, Thantlang.
<br>
The overall and class accuracies (f-score) of the maps are available in a text file included in the archive containing the maps. |
|---|