Reference Dataset for Land Use Change Mapping in Ghana's Cocoa Landscape (2024–2025)
Bibliographic record
Abstract
This dataset was produced by the Centre for Remote Sensing and Geographic Information Services (CERSGIS) as part of the project Reference Data Collection for Improving Land Use Change Mapping in Ghana. The primary objective was to develop high-quality reference data to enhance the accuracy of remote sensing-based land use and land cover (LULC) change mapping using machine learning methods in Ghana’s cocoa production landscapes.The dataset comprises: cocoa_farms: 21,031 geocoded cocoa farm polygons, including agroforestry and shadeless cocoa plots - collected using OpenForis Ground homogeneous_cocoa_farm: 14,192 homogeneous cocoa polygons (shadeless) digitised from total cocoa plots other_land_uses: 20,035 additional geocoded points and polygons representing informal gold mining, degraded forest, oil palm (commercial and subsistence), and rubber (commercial and subsistence) - collected with Collect Earth Online gha_cocoa_hh_public: 485 anonymised cluster records derived from 4,444 individual household survey that complement the geospatial data and provide socioeconomic context - collected with KoboToolbox This dataset provides a critical foundation for automated land cover classification and change detection models in tropical forested regions, where land use is heterogeneous and dynamic. It was developed to support researchers, policymakers, and practitioners across sub-Saharan Africa engaged in monitoring commodity-driven deforestation, landscape restoration, and sustainable land management.This dataset was originally created with support from Lacuna Fund, the world’s first collaborative effort to provide data scientists, researchers, and social entrepreneurs in low- and middle-income contexts globally with the resources they need to produce labelled datasets that address urgent problems in their communities. Lacuna Fund is a funder collaborative that includes The Rockefeller Foundation, Google.org, Canada’s International Development Research Centre, the German Federal Ministry for Economic Cooperation and Development (BMZ) with GIZ as implementing agency, Wellcome Trust, Gordon and Betty Moore Foundation, Patrick J. McGovern Foundation, and The Robert Wood Johnson Foundation. See https://lacunafund.org/about/ for more information. Please contact fmensah@ug.edu.gh with any questions or report an issue on Github here. Let us know how you plan to use the dataset. We are very interested in potential collaborations. NOTE: The cocoa farm geospatial data does not represent property or farm boundaries and should not be used for compliance / legal purposes. This data was collected for the purposes of training remote sensing models for improved mapping of cocoa and other land covers, and not for geolocating specific farms for the purposes of compliance with any regulation. Field data collectors did not trace property boundaries in the field, and field data was checked for quality and potentially edited in GIS. Therefore, these polygons represent only portions of cocoa farms. The sizes of cocoa polygons in this dataset do not necessarily relate to the size of an entire farm for a given location Project Team: CERSGIS - Foster Mensah, Bashara Abubakari SERVIR/UAH - Jacob Abramowitz WRI - James Warburton, Ashleigh Zosel-Harper, Emma Hodoka Data Collection Team: CERSGIS, University of Ghana (Centre for Climate Change and Sustainability Studies, Department of Geography and Resource Development), YouthMappers (University of Ghana Chapter, University of Cape Coast Chapter).
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.003 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".