Downstream Service/Application Development for Monitoring of Environmental Indicators (Africa)
The objective of the action is to identify remote sensing-based indicators to be used by researchers, public authorities, decision makers and other stakeholders for evaluating and monitoring the dynamics, locations and types of (informal) settlement development in fast growing African cities, as well as to attract new Copernicus users in the African context.
By using the Copernicus Sentinel data, the opportunity is given to have large-area, cost-free and consistent coverage of high resolution satellite data. In the context of this action, the objective is to identify and define useful parameters and indicators for localizing and characterizing the living environments of the urban poor in African cities.
Since consulting and integrating the interests of local stakeholders in the application development is one of the main objectives of this action, a user requirement workshop “Urbanization and Migration: Capabilities of remote sensing for monitoring in Africa” was organized on 24 March 2021. In total, 12 participants from 12 institutions attended the workshop. The programme included both presentations and interactive discussions. Three presentations were made by DLR: (1) an introduction to the main objectives of the Copernicus mission, the products, and future satellite launches, (2) an overview of DLR activities conducted in Africa, and (3) applications of Earth Observation for urbanization and migration, which was used as a basis to discuss at a later stage with the participants potential applications for Africa in the framework of this action, based on their needs and requirements.
After the presentations and a brief question-and-answer session, the floor was given to the stakeholders to present their views. The discussion on requirements and needs was initiated and stimulated by two polls (1) on the usage of Earth observation in day to day work and (2) on the availability of geo-information and products/indicators of interest. First, the majority agreed that the currently available geo-information is still not enough for their work in Africa (71%), while 14% were content with the availability or did not know for sure. As a consequence of the second question, four different indicators were proposed by DLR based on the aim of the action. The result showed that in general, participants were interested in land use/land cover (LULC) change, urban growth and slum mapping, which confirmed that these products and indicators will be a valuable and relevant asset in the African context. At the end of the workshop, DLR staff was exchanging with the workshop participants on the future possible collaboration and way ahead. During this session, a discussion on the selection of areas of interest for developing such indicators was held. All participants agreed that big cities and coast line cities should be the main focus.
Thereby, based on the suggestions from participants of the workshop, currently, different indicators are being developed. In order to identify potential test regions in Africa, available geoinformation were downloaded and pre-processed. By using these multisource datasets, a spatio-temporal analysis was performed to identify high dynamic areas in Africa that align the local stakeholder’s suggestions. As a result, a subset of test areas was selected based on the preliminary analyses performed.
In the following, the application concept was developed. First, a feasibility study testing the local climate zones (LCZ) framework in African cities was conducted and after the (re-)conceptualization of some classes to adapt the African landscapes, training data is being created with the aim to create multi-temporal maps for the test sites. The application uses Sentinel-2 imagery to categorize highly dynamic African cities into urban structural types (through the LCZs) in two different dates (2016-2021). Therefore, Sentinel-2 cloud free mosaics are being built for the time period of 2015-2016 and 2020-2021, since some cities are in the tropical zone and the presence of clouds is persistent.
As part of the application, the multi-temporal analysis using EO data will be complemented with data from social media produce by citizens (using Twitter) to test whether its inclusion unravels reasons and drivers of growth (e.g. migration, conflicts, climate change, etc.). Therefore, social media data from Twitter is being downloaded every day (in streaming) for the upcoming analyses in combination with EO data. An on-going exploration analysis of the Twitter data in Africa is being performed.
Outputs and Results
- User requirement workshop
- Downstream application concept development to support the monitoring of the slum growth of cities
- Demonstration of feasibility and benefit analysis
- User uptake workshop