<scp>Remap</scp> : An online remote sensing application for land cover classification and monitoring

Abstract Recent assessments of progress towards global conservation targets have revealed a paucity of indicators suitable for assessing the changing state of ecosystems. Moreover, land managers and planners are often unable to gain timely access to the maps they need to support their routine decision‐making. This deficiency is partly due to a lack of suitable data on ecosystem change, driven mostly by the considerable technical expertise needed to develop ecosystem maps from remote sensing data. We have developed a free and open‐access online remote sensing and environmental modelling application, the Remote Ecosystem Monitoring and Assessment Pipeline ( Remap ; https://remap-app.org ), that enables volunteers, managers and scientists with little or no experience in remote sensing to generate classifications (maps) of land cover and land use change over time. Remap utilizes the geospatial data storage and analysis capacity of Google Earth Engine and requires only spatially resolved training data that define map classes of interest (e.g. ecosystem types). The training data, which can be uploaded or annotated interactively within Remap , are used in a random forest classification of up to 13 publicly available predictor datasets to assign all pixels in a focal region to map classes. Predictor datasets available in Remap represent topographic (e.g. slope, elevation), spectral (archival Landsat image composites) and climatic variables (precipitation, temperature) that are relevant to the distribution of ecosystems and land cover classes. The ability of Remap to develop and export high‐quality classified maps in a very short (&lt;10 min) time frame represents a considerable advance towards globally accessible and free application of remote sensing technology. By enabling access to data and simplifying remote sensing classifications, Remap can catalyse the monitoring of land use and change to support environmental conservation, including developing inventories of biodiversity, identifying hotspots of ecosystem diversity, ecosystem‐based spatial conservation planning, mapping ecosystem loss at local scales and supporting environmental education initiatives.