Annual report iMMAP
In 2020, iMMAP established operational partnerships with key technology organizations, including satellite imagery providers, high-tech startups, and other entities specialized in different remote data collection capacities.
These alliances created synergies that allowed iMMAP to develop cutting-edge, tailored solutions for our partners’ data collection needs, facilitating prompt and detailed information that supported the delivery of data-driven humanitarian response services.
Examples of these capacities include mass internet surveys, mobile application data collection, and satellite imagery analysis. All these data collection processes are complemented by modeling, allowing for data samples that are turned into high-quality indicators to support data-led decisions.
Through iMMAP’s global technology department with a base in Berlin, Germany, and our dedicated units in multiple countries, iMMAP strives to leverage the latest innovations, technologies, partnerships, and digital pathways to support our partners by providing accurate and high-quality information that facilitates efficient humanitarian response operations.
Due to the vast number of Venezuelan migrant and refugee populations entering Colombia, the spread of informal settlements located in rural areas and the outskirts of cities, combined with access issues and the absence of migration control, is exacerbating the humanitarian crisis, making it almost impossible to know the exact details of the population profile and their basic needs. To address this issue, large-scale data collection systems are required.
Satellite imagery allows for rapid detection of these settlements that multiply along the Colombian-Venezuelan border and cannot be located by field teams using traditional methods of detection.
iMMAP, working closely with partner Thinking Machines, adopted an artificial intelligence model to detect and profile these settlements, producing methods that accelerate the detection and profiling process of new settlements, making high-quality information available for humanitarian responders that allows them to quickly respond to the changing needs of the refugee and migrant population from Venezuela.
Through this model, we develop quarterly maps of possible informal settlements across urban centers and areas along the border, to support organizations consolidate a common baseline between responders that reduces duplication and improves the quality of humanitarian assistance.
During 2020, 433 new settlements from 23 departments were detected. Of these, 341 settlements were validated by means of the Premise application. The detection model was executed in 96 municipalities, with settlements detected in 74 of those. A total of 36,277 housing units were detected and mapped one by one. A registry of 12,500 photos was cataloged with partner Premise, averaging 34 per settlement. La Guajira, Antioquia, and Arauca are the departments where a higher number of settlements were detected.
How does the process work?
The detection process begins with the generation of a model from machine learning, developed by the iMMAP partner, Thinking Machines, which uses Sentinel-2 low-resolution satellite imagery to generate a probability map of new settlement locations. With this map, a verification is carried out on Google Earth Pro to ascertain the emergence of settlements over the 2015-2020 time period, during which Colombia received the greatest number of people arriving from Venezuela.
To complement the remote verification, a ground-based validation is carried out to identify the presence of settlements and whether they are currently inhabited. This verification is conducted as the model may detect a settlement that no longer exists or pick up satellite images on Google Earth that are not up-to-date in some municipalities. The ground-based validations are used to provide confirmation of the existence of a settlement as well as profiling the current condition of these sites through the Premise application, a mobile crowdsourcing platform.
In order to meet the needs of refugee and migrant populations fleeing Venezuela, it is essential to identify the chosen routes and the temporary or permanent settlements of these communities. Surveys are usually the most common tool but, in most cases, this information lacks current accuracy and does not allow for the identification of a specific person’s movement over time.
To address this shortfall, iMMAP tracks the movement of Venezuelan refugees and migrants around Latin America on a biweekly basis by analyzing the Facebook advertising platform, an underexploited source of freely accessible data, which allows advertisers and researchers to consult information on socio-demographic characteristics of Facebook users. The data are estimates of the potential reach of a Facebook ad based on the segmentation of the audience on its platform. In doing so, iMMAP collects information that enables humanitarian organizations to better assist migrant and refugee populations.
To present these findings, iMMAP created an interactive dashboard that informs humanitarian actors on the location and movement of refugee and migrant populations, enabling a targeted response. In 2020, 78 organizations consulted iMMAP’s dashboard from the Health, Food Security, WASH, Education, and Socioeconomic and Cultural Integration sectors, as well as the NGO Forum and multiple government bodies.
iMMAP’s Analysis Ready Data Cube (ARDC) was designed to assist the humanitarian and development sectors in addressing crucial economic, environmental, and social challenges by harnessing earth observation data from satellites as well as climate models.
Employing the technology infrastructure behind the Open Data Cube initiative, iMMAP ARDC is the first Open Data Cube with a focus on the needs of the humanitarian sector.
iMMAP’s ARDC increases the impact of Earth Observation data aimed at supporting different clusters with On-Demand Analysis Ready Data and Information. With the advances in machine learning, data mining, and computing infrastructures, the iMMAP ARDC can manage big data queries and rapidly yield time-series analysis of large satellite data archives such as Sentinel 2 and Landsat spanning as far back as 30 years.
iMMAP and Data Friendly Space (DFS) develop tailored data analysis processes to support the humanitarian community by using DFS’ Data Entry and Exploration Platform (DEEP). DEEP generates repositories of pre-organized secondary data for each country, addressing the challenges in data and information comprehensiveness, consistency, and value. Together, iMMAP and DFS are generating extensive secondary data reviews of the reports, assessments, and research papers available at the field level that provide humanitarian actors with a thorough analysis of the impact of the COVID-19 pandemic and the condition of the people in each of the countries of intervention.
In 2020 iMMAP remained actively engaged on the Humanitarian Data Exchange (HDX) platform. Managed by OCHA’s Centre for Humanitarian Data, HDX is an open platform for sharing data across crises and organizations, making humanitarian data easy to find and use for analysis.
total views of iMMAP’s
datasets on HDX
individual users downloaded at least
one of our datasets
individual users visited iMMAP’s
datasets on HDX
total downloads of
iMMAP’s datasets from HDX