The large-scale expansion of built infrastructure is profoundly reshaping the geographies of Africa, generating lock-in patterns of development for future generations. Understanding the impact of these massive investments can allow development opportunities to be maximised and therefore be critical for attaining the United Nations Sustainable Development Goals and Africa Union Agenda 2063. However, until now information on the types, scope, and timing of investments, how they have evolved, and their spatial-temporal impact was dispersed amongst various agencies. We developed the first comprehensive database of 79 ongoing and planned investment corridors across Africa, synthesizing data from multiple sources covering 184 projects on railways, wet and dry ports, pipelines, airports, techno-cities, and industrial parks. The georeferenced interlinked tabular and spatial database includes 22 attributes with sources provided for each observation. We expect this database will improve coordination, efficiency, monitoring, oversight, strategic planning, transparency, vulnerability risk, and impact assessments, among other uses for inter alia investment banks, governments, impact assessment practitioners, communities, conservationists, economists, and regional economic bodies.

The methodology was set in three main stages: first, a global review involving the searching and screening for data, second, the manual digitisation of spatial features, spatial-temporal distribution, and volume calculation procedures, and third, technical validation.

Search strategy: Standard methods were used to access, appraise, and synthesise peer-reviewed and grey literature information about infrastructure projects within development corridors in Africa. We developed a search protocol to ensure rigor, objectivity, verifiable procedures, and clarity of the study design (CEE 2013). The search strategy was established through discussions between researchers and stakeholders during eleven months of fieldwork in Tanzania and Kenya between 2018 and 2019, and consultations with the executive board of scientific experts in the Development Corridors Partnership (https://developmentcorridors.org) - a collaborative capacity-building research programme in Kenya, Tanzania, China, and the UK involving universities, development corridors agencies and NGOs. Three reviewers conducted searches and screening over 18 months between the 28^th of January 2020 and the 19^th of May 2021.

Sources of publications and key search terms: We combined a multitude of data sources (total n=191), and ran searches on websites of corridor authorities, regional economic bodies, national government bodies, private companies, INGOs, international and national media houses, research groups, public-private partnerships, regional associations, research groups, and multilateral agencies (see Supplementary Table 1 for a summary of these data sources). In addition, we searched multilateral development banks across the world (e.g., African Development Bank, Asian Development Bank, European Central Bank World Bank, Islamic Development Bank), which have approved funding for infrastructure development projects with wider socio-economic developmental benefits. We also searched infrastructure databases, including e-platforms for information and knowledge sharing on infrastructure development in Africa. Four key e-journals whose topic areas closely aligned with the research focus were hand searched, including Geographical Review, Economic Geography, African Studies Quarterly, and International Journal of Applied Engineering Research. We further contacted researchers involved in past and ongoing projects to source data, such as African Ecological Futures and the Global Infrastructure Mapping and Modelling Project

After testing the key search terms on 20 development corridors to confirm that the search strings captured relevant literature, and balanced specificity and sensitivity (Pullin et al., 2013), the final string consisted of: "transport corridor *", "economic corridor *", "development corridor *", "infrastructure corridor*", "resource corridor*", and "growth corridor*". These terms were applied to all databases and sources for all African countries, subject to their individual search requirements (Thorn et al., 2020). We used Boolean operator terms and wildcards, i.e., a character that can be used to substitute for other character(s) in a string, connected search terms, which were disaggregated using truncation (“*” in most databases). 

Inclusion criteria: We only included infrastructure projects for which we could see a clear connection to a development corridor programme as defined above. We divided infrastructure within development corridors into the following types: airports; dry ports; wet ports; electricity transmission lines; freight railways; industrial parks; passenger and freight railways; pipelines (oil); pipelines (water); resort cities; and roads. We included electricity transmission lines only when multiple countries had partnered to develop the infrastructure to enhance trade. Waterways included all forms of transport of goods and people traversing water, including catamarans and large ships. We included wet ports that cater for large shipping vessels, as well as dryland inland ports. We included international airports which are written into development strategic planning or policy documents, are international hubs that allow for the inflow and outflow of traded goods and services in development corridors, or are physically connected or near development corridors included in our database. We excluded dams, domestic electricity transmission lines, small ferry links, freshwater inland ports, local airports, agricultural corridors, bridges (unless part of roads in a development corridor), and other forms of infrastructure not connected to development corridors.

Searches were completed for 53 countries in continental Africa. The islands of Comoros, Mauritius, Reunion, Mayotte, and São Tomé and Principe were excluded because we did not find development corridors present in these areas. English, Swahili, Portuguese, and French sources were reviewed, but we did not include other languages due to the skillset of the review team. We expect that these languages captured most key documents regarding development corridors in Africa, and suggest future iterations include languages such as Arabic, Yoruba, and Hausa. There was no time limitation on the searches. Research articles, reviews, and reports were included, while book chapters, conference proceedings, and graduate theses were excluded.

Spatial digitisation of corridors: Mapping of the spatial features of the African Development Corridors Database was made using ArcGIS Desktop 10.8.1 and ArcGIS Pro 2.5.2. Each development corridor project was identified with a unique project code, to which all tabular and spatial parameters were associated, and recorded as point or line features. We sourced available base layers using the results of Laurance et al. (2015), Open Street Maps (https://download.geofabrik.de/), the World Food Programme GeoNode Global Airports dataset (https://geonode.wfp.org/layers/geonode%3Awld_trs_airports_wf) including international airports or airports that were within a buffer of 40km to nearby development corridors, and data from partners linked to the Development Corridors Partnership. We also looked for maps in reports and websites and digitized them. All data were digitized by the project team and sources are acknowledged in the accompanying reference table. Approximately 79.1% (n=140 projects) were fully mapped, while the remaining 44 (23.9%) were unmapped.

It is important to note that this study does not include all infrastructure projects in Africa. Instead, the focus was on mapping those infrastructure projects that were under the umbrella of an existing, in progress, or planned development corridor project (or similar term). See the Nature Scientific Data manuscript and methods for further details (Manuscript is In Press).

The files available in the compressed zip files are the same version of the database but in different formats: ESRI ArcGIS geodatabase (.gdb) named as "DCDB20210817.gdb", open access Geopackage format (.gpkg) named as "DCDB20210817.gpkg", and non-spatially explicit version of the database in .csv format named as "DCDB_Tabular_CSVs_20210817". Each includes the spatial data with its 22 attribute table, a references table, and a list of non-mapped corridors.

The African Development Corridors database can be explored at https://arcg.is/TynL90. It is also publicly available and can be accessed and downloaded through the Development Corridors data portal: https://dcp-unep-wcmc.opendata.arcgis.com/datasets/. All the data contained in the database and the database can be used under a CCO 1.0 license. They were digitized by the team using the sources cited in the methods. All sources are acknowledged in the reference table, which is linked to the spatial data by a unique identifier for each project that links the spatial data and the references table. There are several ways to access and download the information contained in the database in the Development Corridors data portal– with options to download the database in CSV with 22 attributes, KML, Shapefiles, or GEOJSON formats. Detailed information about the variables in the database can be found in Supplementary Table 2 of the published paper, which corresponds to the .csv files in the data portal.