Search for: All records

Measuring socioeconomic indices at the scale of regions or countries is required in various contexts, in particular to inform public policies. The use of Deep Learning (DL) and Earth Observation (EO) data is becoming increasingly common to estimate specific variables like societal wealth. This paper presents an end- to-end framework ‘DeepWealth’ that calculates such a wealth index using open-source EO data and DL. We use a multidisciplinary approach incorporating satellite imagery, socio-economic data, and DL models. We demonstrate the effectiveness and generalizability of DeepWealth by training it on 24 African countries and deploying it in Madagascar, Brazil and Japan. Our results show that DeepWealth provides accurate and stable wealth index estimates with an 𝑅2 of 0.69. It empowers computer-literate users skilled in Python and R to estimate and visualize well-being-related data. This open-source framework follows FAIR (Findable, Accessible, Interoperable, Reusable) principles, providing data, source code, metadata, and training checkpoints with its source code made available on Zenodo and GitHub. In this manner, we provide a DL framework that is reproducible and replicable. 

Data Management Plans (DMP) are now a routine part of research proposals but are generally not referred to after funding is granted. The Belmont Forum requires an extensive document, a ‘Data and Digital Object Management Plan’ (D(DO)MP) for its awarded projects that is expected to be kept current over the life of the project. The D(DO)MP is intended to record team decisions about major tools and practices to be used over the life of the project for data and software stewardship, and for preservation of data and software products, aligned with the desired Open Science outcomes relevant to the project. Here we present one of the first instances of the use of Belmont’s D(DO)MP through a case study of the PARSEC project, a multinational and multidisciplinary investigation of the socioeconomic impacts of protected areas. We describe the development and revision of our interpretation of the D(DO)MP and discuss its adoption and acceptance by our research group. We periodically assessed the data management sophistication of team members and their use of the various nominated tools and practices. As a result, for example, we included summaries to enable the key components of the D(DO)MP to be readily viewed by the researcher. To meet the Open Science outcomes in a complex project like PARSEC, a comprehensive and appropriately structured D(DO)MP helps project leaders (a) ensure that team members are committed to the collaboration goals of the project, (b) that there is regular and effective feedback within the team, (c) training in new tools is provided as and when needed, and (d) there is easy access to a short reference to the tools and descriptions of the nominated practices.