Accurate land use/land cover (LULC) classification remains a persistent challenge in rapidly urbanising regions especially, in the Global South, where cloud cover, seasonal variability, and limited ground truth data often undermine the reliability of traditional remote sensing approaches. While recent advances in satellite embedding models, such as AlphaEarth Foundations (AEF) offering, task-ready representations of Earth’s surface, there is limited empirical evidence on how different machine learning algorithms perform when applied to these embeddings, particularly in African urban contexts. This study addresses that gap by evaluating the performance of four leading boosting algorithms (Light Gradient Boosting Machine (LightGBM), Categorical Boost (CatBoost), AdaBoost, and XGBoost) used in conducting supervised classification for four dominant land cover types (Urban, Water, Bare Land, and Vegetation) in the Greater Accra Area, Ghana, using AEF embeddings. The findings indicate that LightGBM achieved the highest classification performance, with a superior F-score across all classes, an overall accuracy (OA) of 98.35%, and a Kappa coefficient of 0.978, reflecting excellent agreement with ground truth labels. XGBoost followed closely, with an OA of 98% and a Kappa value of 0.973. CatBoost ranked third, with an OA value of 97.5% and a Kappa value of 0.966. AdaBoost performed least effectively with an OA value of 96.7% and a Kappa value of 0.955. These results provide further clarity and insights into the comparative performance of boosting algorithms when applied to AEF embeddings for supervised land use/land cover (LULC) classification. Additionally, the results highlight the potential of embedding-based approaches for scalable and interpretable land cover mapping in complex, data-constrained urban environments.
References
[1]
Colkesen, I. and Ozturk, M.Y. (2022) A Comparative Evaluation of State-of-the-Art Ensemble Learning Algorithms for Land Cover Classification Using Worldview-2, Sentinel-2 and ROSIS Imagery. ArabianJournalofGeosciences, 15, Article No. 942. https://doi.org/10.1007/s12517-022-10243-x
[2]
Brown, C.F., Kazmierski, M.R., Pasquarella, V.J., et al. (2025) AlphaEarth Foundations: An Embedding Field Model for Accurate and Efficient Global Mapping from Sparse Label Data. Google DeepMind. https://arxiv.org/abs/2507.22291
[3]
Mahmoud, A.S., Mezaal, M.R., Hameed, M.R. and Naje, A.S. (2022) A Framework for Improving Urban Land Cover Using Object and Pixel-Based Techniques via Remotely Sensed Data. NatureEnvironmentandPollutionTechnology, 21, 2189-2200. https://doi.org/10.46488/nept.2022.v21i05.013
[4]
Addae, B. and Oppelt, N. (2019) Land-Use/Land-Cover Change Analysis and Urban Growth Modelling in the Greater Accra Metropolitan Area (GAMA), Ghana. UrbanScience, 3, Article No. 26. https://doi.org/10.3390/urbansci3010026
[5]
Akubia, J.E.K. and Bruns, A. (2019) Unravelling the Frontiers of Urban Growth: Spatio-Temporal Dynamics of Land-Use Change and Urban Expansion in Greater Accra Metropolitan Area, Ghana. Land, 8, Article No. 131. https://doi.org/10.3390/land8090131
[6]
Abdelmajeed, A.Y.A. and Juszczak, R. (2024) Challenges and Limitations of Remote Sensing Applications in Northern Peatlands: Present and Future Prospects. RemoteSensing, 16, Article No. 591. https://doi.org/10.3390/rs16030591
[7]
Faridatul, M.I. and Wu, B. (2018) Automatic Classification of Major Urban Land Covers Based on Novel Spectral Indices. ISPRS International Journal of Geo-Information, 7, Article No. 453. https://doi.org/10.3390/ijgi7120453
[8]
Houriez, L., Pilarski, S., Vahedi, B., Ahmadalipour, A., Honda Scully, T., Aflitto, N. and Andre, D. (2025) Scalable Geospatial Data Generation Using AlphaEarth Foundations Model. https://arxiv.org/abs/2508.11739
[9]
Alzubaidi, L., Bai, J., Al-Sabaawi, A., Santamaría, J., Albahri, A.S., Al-dabbagh, B.S.N., et al. (2023) A Survey on Deep Learning Tools Dealing with Data Scarcity: Definitions, Challenges, Solutions, Tips, and Applications. JournalofBigData, 10, Article No. 46. https://doi.org/10.1186/s40537-023-00727-2
[10]
Alvarez, C.I., Ulloa Vaca, C.A. and Echeverria Llumipanta, N.A. (2025) Machine Learning for Urban Air Quality Prediction Using Google Alphaearth Foundations Satellite Embeddings: A Case Study of Quito, Ecuador. RemoteSensing, 17, Article No. 3472. https://doi.org/10.3390/rs17203472
[11]
Choudhury, S., Aharoni, E., Suvarna, C., Tsogsuren, I., Kreidieh, A.R., Lu, C.-T. and Arora, N. (2025) S2Vec: Self-Supervised Geospatial Embeddings. ACM Conference Proceedings, Minneapolis, 3-5 November 2025, 1-22. https://arxiv.org/abs/2504.16942
[12]
Wu, M., Huang, Q. and Gao, S. (2025) Advancing Mixed Land Use Detection by Embedding Spatial Intelligence into Vision-Language Models. ISPRSAnnalsofthePhotogrammetry, RemoteSensingandSpatialInformationSciences, 4, 121-128. https://doi.org/10.5194/isprs-annals-x-4-w7-2025-121-2025
[13]
Feng, et al. (2025) TESSERA: Temporal Embeddings of Surface Spectra for Earth Representation and Analysis. https://arxiv.org/abs/2506.20380
[14]
Owusu, M., Engstrom, R., Thomson, D., Kuffer, M. and Mann, M.L. (2023) Mapping Deprived Urban Areas Using Open Geospatial Data and Machine Learning in Africa. UrbanScience, 7, Article No. 116. https://doi.org/10.3390/urbansci7040116
[15]
GSS (2021) 2021 Population and Housing Census: Greater Accra Region Analytical Report. Ghana Statistical Service. https://census2021.statsghana.gov.gh
[16]
Ekumah, B., Armah, F.A., Afrifa, E.K.A., Aheto, D.W., Odoi, J.O. and Afitiri, A. (2020) Assessing Land Use and Land Cover Change in Coastal Urban Wetlands of International Importance in Ghana Using Intensity Analysis. WetlandsEcologyandManagement, 28, 271-284. https://doi.org/10.1007/s11273-020-09712-5
[17]
Dogbey, F., Asilevi, P.J., Dzrobi, J.F., Koffi, H.A. and Klutse, N.A.B. (2022) Modelling Cloud Cover Climatology over Tropical Climates in Ghana. Atmosphere, 13, Article No. 1265. https://doi.org/10.3390/atmos13081265
[18]
Stow, D.A., Chen, D., Coulter, L.L. and Sandborn, A. (2013) Monitoring Land Cover Change in Urban and Peri-Urban Areas Using Multi-Temporal High Spatial Resolution Satellite Imagery. Remote Sensing of Environment, 129, 25-38.
[19]
Bramhe, V.S., Ghosh, S.K. and Garg, P.K. (2018) Extraction of Built-Up Area by Combining Textural Features and Spectral Indices from Landsat-8 Multispectral Image. TheInternationalArchivesofthePhotogrammetry, RemoteSensingandSpatialInformationSciences, 5, 727-733. https://doi.org/10.5194/isprs-archives-xlii-5-727-2018
[20]
Maloney, M.C., Becker, S.J., Griffin, A.W.H., Lyon, S.L. and Lasko, K. (2024) Automated Built-Up Infrastructure Land Cover Extraction Using Index Ensembles with Machine Learning, Automated Training Data, and Red Band Texture Layers. Remote Sensing, 16, 868.
[21]
Gómez, C., White, J.C. and Wulder, M.A. (2016) Optical Remotely Sensed Time Series Data for Land Cover Classification: A Review. ISPRSJournalofPhotogrammetryandRemoteSensing, 116, 55-72. https://doi.org/10.1016/j.isprsjprs.2016.03.008
[22]
Moharrami, M., Attarchi, S., Gloaguen, R. and Alavipanah, S.K. (2024) Integration of Sentinel-1 and Sentinel-2 Data for Ground Truth Sample Migration for Multi-Temporal Land Cover Mapping. RemoteSensing, 16, Article No. 1566. https://doi.org/10.3390/rs16091566
[23]
Tsiligaridis, J. (2023) Tree-Based Ensemble Models, Algorithms and Performance Measures for Classification. AdvancesinScience, TechnologyandEngineeringSystemsJournal, 8, 19-25. https://doi.org/10.25046/aj080603
[24]
Rizkallah, L.W. (2025) Enhancing the Performance of Gradient Boosting Trees on Regression Problems. JournalofBigData, 12, Article No. 35. https://doi.org/10.1186/s40537-025-01071-3
[25]
Natekin, A. and Knoll, A. (2013) Gradient Boosting Machines, a Tutorial. FrontiersinNeurorobotics, 7, Article No. 21. https://doi.org/10.3389/fnbot.2013.00021
[26]
Boldini, D., Grisoni, F., Kuhn, D., Friedrich, L. and Sieber, S.A. (2023) Practical Guidelines for the Use of Gradient Boosting for Molecular Property Prediction. JournalofCheminformatics, 15, Article No. 73. https://doi.org/10.1186/s13321-023-00743-7
[27]
Dietterich, T.G. (2000) Ensemble Methods in Machine Learning. First International Workshop, MCS 2000, Cagliari, 21-23 June 2000, 1-15. https://doi.org/10.1007/3-540-45014-9_1
[28]
Chen, T. and Guestrin, C. (2016) XGBoost: A Scalable Tree Boosting System. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, San Francisco, 13-17 August 2016, 785-794. https://doi.org/10.1145/2939672.2939785
[29]
Wiens, M., Verone‐Boyle, A., Henscheid, N., Podichetty, J.T. and Burton, J. (2025) A Tutorial and Use Case Example of the Extreme Gradient Boosting (XGBoost) Artificial Intelligence Algorithm for Drug Development Applications. ClinicalandTranslationalScience, 18, e70172. https://doi.org/10.1111/cts.70172
[30]
Zhang, A., Lipton, Z.C., Li, M. and Smola, A.J. (2020) Dive into Deep Learning. Cambridge University Press.
[31]
Dorogush, A.V., Ershov, V. and Gulin, A. (2018) CatBoost: Gradient Boosting with Categorical Features Support. https://arxiv.org/pdf/1810.11363
[32]
Prokhorenkova, L., Gusev, G., Vorobev, A., Dorogush, A.V. and Gulin, A. (2018) CatBoost: Unbiased Boosting with Categorical Features. Proceedings of the 32nd International Conference on Neural Information Processing Systems, Montréal, 3-8 December 2018, 6638-6648. https://proceedings.neurips.cc/paper/2018/hash/14491b756b3a51daac41c24863285549-Abstract.html
[33]
Fonte, C.C., See, L., Laso-Bayas, J.C., Lesiv, M. and Fritz, S. (2020) Assessing the Accuracy of Land Use Land Cover (LULC) Maps Using Class Proportions in the Reference Data. ISPRSAnnalsofthePhotogrammetry, RemoteSensingandSpatialInformationSciences, 3, 669-674. https://doi.org/10.5194/isprs-annals-v-3-2020-669-2020
[34]
Ursu, C., Benedek, J. and Temerdek-Ivan, K. (2025) Accuracy Assessment of Four Land Cover Datasets at Urban, Rural and Metropolitan Area Level. RemoteSensing, 17, Article No. 756. https://doi.org/10.3390/rs17050756
[35]
Becker, S.J., Maloney, M.C., Griffin, A.W.H., Lasko, K. and Sussman, H.S. (2025) Bare Ground Classification Using a Spectral Index Ensemble and Machine Learning Models Optimized across 12 International Study Sites. GeocartoInternational, 40, Article ID: 2465452. https://doi.org/10.1080/10106049.2025.2465452
[36]
Friedman, J.H. (2001) Greedy Function Approximation: A Gradient Boosting Machine. TheAnnalsofStatistics, 29, 1189-1232. https://doi.org/10.1214/aos/1013203451
[37]
Amarsaikhan, E., Enkhjargal, D., Jargaldalai, E. and Amarsaikhan, D., (2024) Comparison of Machine Learning and Parametric Methods for the Discrimination of Urban Land Cover Types. Geocarto International, 39, Article ID: 2380372. https://doi.org/10.1080/10106049.2024.2380372
