| Titre : |
Towards a system combining SAR and optical Sentinel data to monitor gold mining in the Guiana shield |
| Type de document : |
Article/Communication |
| Auteurs : |
Mathieu Rahm, Auteur ; Cédric Lardeux, Auteur ; Pierre-Louis Frison  , Auteur ; et al., Auteur |
| Editeur : |
Paris : Agence Spatiale Européenne ASE / European Space Agency ESA |
| Année de publication : |
2016 |
| Conférence : | LPS 2016, ESA Living Planet Symposium 09/05/2016 13/05/2016 Prague République tchèque programme sans actes | | Importance : |
60 p. |
| Format : |
21 x 30 cm |
| Note générale : |
bibliographie |
| Langues : |
Anglais (eng) |
| Descripteur : |
[Vedettes matières IGN] Applications de télédétection
[Termes IGN] analyse diachronique
[Termes IGN] Brésil
[Termes IGN] déboisement
[Termes IGN] fusion d'images
[Termes IGN] Guyana
[Termes IGN] Guyane (département français)
[Termes IGN] image Sentinel-MSI
[Termes IGN] image Sentinel-SAR
[Termes IGN] mine d'or
[Termes IGN] pollution des eaux
[Termes IGN] Suriname
|
| Résumé : |
(auteur) Included in the larger Guiana Shield ecosystem, Suriname, Guyana, French Guiana and the Brazilian state of Amapá possess one of the largest continuous tracts of pristine forest in the world. Under little threat until fifteen years ago, deforestation and forest degradation are of increasing concern in the region. Gold mining activities driven by the sustained increase of gold price has experienced a significant boom and represents nowadays one of the main driver. The pollution of rivers and streams by mercury used in small-scale gold mining is also expanding, which increases risks to local population health and freshwater biodiversity. In 2010, the French National Forest Office (ONF) showed by using optical satellite images at medium and high resolution (HR) that gold mining activities’ impacts on forest cover and freshwater increased approximately by a factor three in the region between 2001 and 2008. More recently, Alvarez-Berríos et al. (2015) pointed out a sustained acceleration of deforestation caused by gold mining in the Guiana shield between 2007 and 2013. However, this study which was performed using low resolution data at the scale of South America has limited capacity to detect gold mining, especially in the high forest cover of Guiana Shield where small- and medium-scale operations account for most of the deforestation. To overcome this limitation, the REDD+ for the Guiana Shield project conducted a study co-funded by WWF Guianas to update for 2014 the ONF 2001-2008 results, using optical multi-sensors data at medium and high resolution. The study was carried out following a unique collaborative and participatory approach involving a team of experts from the forestry and environmental services of each territory, namely SEMA (Amapá-Brazil), ONF (French Guiana-France), GFC (Guyana), and SBB (Suriname). The results confirmed the rapid expansion of the activity in the region where more than 92,000 ha were newly deforested between 2008 and 2014, compared to approximately 46,000 ha during the period 2001-2008. In 2014, more than 9,000 km of waterways were in direct contact with mining sites, which is approximately 6.5 times more than in 2001. Although a reliable, accurate and robust regional methodology has been developed and operationally implemented, the frequent and widespread cloud cover of the Guianan moist forest region represents a challenge for the use of optical HR data. The need to process time series of satellite images in most areas to reduce cloud cover is time-consuming. Despite processing more than two hundreds images, 3.6% of the study area remained masked by clouds. The recent free access to SAR HR Sentinel-1 data offers great opportunities to improve the process. SAR sensors can peer through clouds and their sensitivity to soil moisture can help to better detect small-scale mining sites. Therefore, the REDD+ for the Guiana Shield project has started to build capacities in the region on SAR image interpretation and processing using the Sentinel Application Platform (SNAP). A first mosaic of Sentinel-1 data covering Suriname, Guyana, French Guiana and the Brazilian state of Amapá has been created and automated pre-processing steps have been developed. The integration of Sentinel-1 data in the regional gold mining monitoring system has been successfully tested in four study sites, one in each country. The coming free access to optical HR Sentinel-2 data opens even more perspectives towards the development of cost-effective monitoring systems in the region, especially valuable in the context of REDD+. This paper first presents the results of the impact of gold mining activities on the forest cover and freshwater for 2014 and shows the evolution since 2001. Secondly, it provides the first outcomes towards the development of time- and cost-efficient forest monitoring systems in the region. |
| Numéro de notice : |
C2016-058 |
| Affiliation des auteurs : |
LASTIG MATIS+Ext (2012-2019) |
| Thématique : |
IMAGERIE |
| Nature : |
Communication |
| nature-HAL : |
ComAvecCL&ActesPubliésIntl |
| DOI : |
sans |
| En ligne : |
http://lps16.esa.int/page_session187.php#1306p |
| Format de la ressource électronique : |
URL abstract |
| Permalink : |
https://documentation.ensg.eu/index.php?lvl=notice_display&id=91912 |
|  |
Towards a system combining SAR and optical Sentinel data to monitor gold mining in the Guiana shield (2016)
