Cloud computing solution for monitoring dryland forests dynamics in Morocco
No hay miniatura disponible
Fecha
Título de la revista
ISSN de la revista
Título del volumen
Editor
FAO ;
Resumen
Descripción
The wide availability of free satellite imagery and the recent development of cloud-based geospatial analysis platforms dedicated to spatial Big Data that integrates both image archives from different providers, processing algorithms, distributed processing capabilities as well as an application programming interface that facilitate scripting and automation process, opened new perspectives for the use of vegetation observation time series over long time series and over large spatial scales.
This work aims at harnessing these technologies and building up an automated solution to monitor forests rehabilitation dynamics in arid lands and to assess the effectiveness of stakeholder’s management or restoration strategies. Such solution is based on graphical user interface that facilitate the process and on the use of analysis functions relaying on analyzing temporal trajectories (time series) of different spectral indices derived from satellite images (Landsat, MODIS or Sentinel) at the required spatial analysis scale.
The solution is implemented using java script language by the functions offered by Google Earth Engine (GEE) API. The graphical user interface of the first prototype is exploitable by the means of a standard web browser. It is accessible even to people without any background in regard to programming languages or to remote sensing skills and to overcome managers lack of technical skills or computing infrastructure capacities. The process was tested for two arid regions on Morocco mainly on sites recently rehabilitated: acacia ecosystems on the southern part of Morocco and the argan ecosystem which is an emblematic agrosilvopastoral system. The output has been qualified as promising solution and the prototype represent decision-support tool which contribute to managing and communicating forest information and management at different levels and facilitating the assessment of ecosystem trends and then to plan restoration interventions. Furthermore, field data collected by involving local communities can greatly facilitate results validation, which makes possible to compensate errors due to data sources.
Key words: Google earth engine; remote sensing; monitoring; Dryland forests, Morocco
ID: 3623147