Published

2026-05-27

Machine Learning Approaches for Predicting Geoid Undulations to Improve Orthometric Height Determination in Data-Scarce Regions

Enfoques de Aprendizaje Automático para Predecir las Ondulaciones del Geoide y Mejorar la Determinación de Alturas Ortométricas en Regiones con Escasez de Datos

DOI:

https://doi.org/10.15446/esrj.v30n1.122297

Keywords:

Geoid undulations, GNSS, gradient boosted regression (GBR), machine learning, orthometric height., support vector regression (SVR) (en)
ondulación del geoide, GNSS, regresión con impulso gradiente (GBR), aprendizaje automático, altura ortométrica, regresión de vectores de soporte (SVR) (es)

Downloads

Authors

This study investigates the use of machine learning algorithms for geoid undulation modelling in data-sparse environments, using Ibadan, Nigeria as a case study. A total of 207 control points were utilized, with 70% allocated for training and 30% for testing. Six algorithms were assessed: Multiple Linear Regression (MLR), Random Forest (RF), Gradient Boosted Regression (GBR), Decision Tree (DT), Support Vector Regression (SVR), and K-Nearest Neighbors (KNN). Model evaluation was conducted using root mean square error (RMSE), mean absolute error (MAE), coefficient of determination (R²), and 5-fold cross-validation to ensure robustness. Among the models tested, GBR and SVR yielded superior performance. The GBR model achieved a test RMSE of 0.047 m with an R² of 0.9785, whereas the SVR model demonstrated a lower test RMSE of 0.026 m and a higher R² of 0.9934. Cross-validation results were consistent, with GBR yielding an RMSE of 0.043 m and R² of 0.9828, compared to SVR’s RMSE of 0.056 m and R² of 0.9719. These results highlight the strong generalization ability and practical applicability of both models. Additionally, the GBR model was applied to derive orthometric heights from GNSS-based ellipsoidal heights, and the output was validated against GNSS-derived orthometric heights, yielding an RMSE of 0.047 m. The study concludes that machine learning, particularly GBR and SVR, provides an effective complementary approach for geoid prediction and vertical height transformation in regions with limited access to gravimetric data, with important implications for geodetic infrastructure development, surveying, and vertical referencing improvement in developing regions.

Este estudio investiga el uso de algoritmos de aprendizaje automático para la modelización de las ondulaciones del geoide en entornos con escasez de datos, utilizando Ibadan, Nigeria, como caso de estudio. Se emplearon un total de 207 puntos de control, asignando el 70% para entrenamiento y el 30% para pruebas. Se evaluaron seis algoritmos: Regresión Lineal Múltiple (MLR), Bosques Aleatorios (RF), Regresión con Impulso Gradiente (GBR), Árbol de Decisión (DT), Regresión de Vectores de Soporte (SVR) y K-Vecinos Más Cercanos (KNN). La evaluación de los modelos se realizó utilizando el error cuadrático medio (RMSE), error absoluto medio (MAE), coeficiente de determinación (R²) y validación cruzada de 5 pliegues para garantizar robustez. Entre los modelos probados, GBR y SVR obtuvieron el mejor rendimiento. El modelo GBR alcanzó un RMSE de prueba de 0.047 m con un R² de 0.9785, mientras que el modelo SVR mostró un RMSE de prueba más bajo de 0.026 m y un R² más alto de 0.9934. Los resultados de la validación cruzada fueron consistentes, con GBR obteniendo un RMSE de 0.043 m y un R² de 0.9828, en comparación con el RMSE de SVR de 0.056 m y R² de 0.9719. Estos resultados destacan la gran capacidad de generalización y aplicabilidad práctica de ambos modelos. Además, el modelo GBR se aplicó para derivar alturas ortométricas a partir de alturas elipsoidales basadas en GNSS, y la salida se validó frente a alturas ortométricas derivadas de GNSS, obteniendo un RMSE de 0.047 m. El estudio concluye que el aprendizaje automático, en particular GBR y SVR, ofrece un enfoque complementario efectivo para la predicción del geoide y la transformación de alturas verticales en regiones con acceso limitado a datos gravimétricos, con importantes implicaciones para el desarrollo de infraestructura geodésica, la topografía y la mejora de la referencia vertical en regiones en desarrollo.

References

Albayrak, M., Özlüdemir, M. T., Aref, M. M., & Halicioglu, K. (2020). Determination of Istanbul geoid using GNSS/levelling and valley cross levelling data. Geodesy and Geodynamics, 11(3), 163-173. https://doi.org/10.1016/j.geog.2020.01.003

Alsadany, S. S., Fawaz, E. M., Elshewy, M. A., & Hamdy, A. M. (2024). Developing a local geoid model for Egypt using machine learning algorithms. Journal of Al-Azhar University Engineering Sector, 19, 102-118.

Cunderlík, R., Medľa, M., & Mikula, K. (2020). Local quasigeoid modelling in Slovakia using the finite volume method on the discretized Earth’s topography. Contributions to Geophysics and Geodesy, 50(3), 287-302. https://doi.org/10.31577/congeo.2020.50.3.1

Dawod, G. M., & Abdel-Aziz, T. M. (2020). Utilization of geographically weighted regression for geoid modelling in Egypt. Journal of Applied Geodesy, 14(1), 1-12. https://doi.org/10.1515/jag-2019-0009

Doganalp, S., & Selvi, H. Z. (2015). Local geoid determination in strip area projects by using polynomials, least-squares collocation and radial basis functions. Measurement, 73, 429-438. https://doi.org/10.1016/j.measurement.2015.05.030

Elshambaky, H. (2018). Application of neural network technique to determine a corrector surface for global geopotential model using GPS/levelling measurements in Egypt. Journal of Applied Geodesy, 12(1), 29-43. https://doi.org/10.1515/jag-2017-0017

Elshewy, M. A., Thanh, P. T., Elsheshtawy, A. M., Refaat, M., & Freeshah, M. (2024). A novel approach for optimizing regional geoid modeling over rugged terrains based on global geopotential models and artificial intelligence algorithms. The Egyptian Journal of Remote Sensing and Space Sciences, 27(4), 656-668. https://doi.org/10.1016/j.ejrs.2024.09.002

Erol, B., & Celik, R. N. (2006). Modelling local GPS/levelling geoid: Assessment of inverse distance weighting and geostatistical kriging methods. Geoinformation Science Journal, 6(1), 78-83.

Erol, B., & Erol, S. (2013). Learning-based computing techniques in geoid modeling for precise height transformation. Computers & Geosciences, 52, 95-107. https://doi.org/10.1016/j.cageo.2012.09.010

Erol, S., & Erol, B. (2021). A comparative assessment of different interpolation algorithms for prediction of GNSS/levelling geoid surface using scattered control data. Measurement, 173, 108623. https://doi.org/10.1016/j.measurement.2020.108623

Featherstone, W. E., & Kuhn, M. (2006). Height systems and vertical datums: A review in the Australian context. Journal of Spatial Science, 51(1), 21-41. https://doi.org/10.1080/14498596.2006.9635062

Fotopoulos, G. (2005). Calibration of geoid error models via a combined adjustment of ellipsoidal, orthometric and gravimetric geoid height data. Journal of Geodesy, 79(1–3), 111-123. https://doi.org/10.1007/s00190-005-0449-y

Heiskanen, W. A., & Moritz, H. (1967). Physical geodesy. W.H. Freeman.

Jayawardana, S. S. U., Jayasinghe, U. D. T. M., & Kekulawala, K. V. S. P. (2024). Improving accuracy of geoid undulation model using machine learning approaches for Sri Lanka. Proceedings of the Asian Conference on Remote Sensing (ACRS 2024).

Kaloop, M. R., Rabah, M., Hu, J. W., & Zaki, A. (2018). Using advanced soft computing techniques for regional shoreline geoid model estimation and evaluation. Marine Georesources and Geotechnology, 36(6), 688-697. https://doi.org/10.1080/1064119X.2017.1370622

Kaloop, M. R., Samui, P., Rabah, M., Al-Ajami, H., Hu, J. W., & Zaki, A. (2021). Improving accuracy of local geoid model using machine learning approaches and residuals of GPS/levelling geoid height. Survey Review, 53(383), 505-518. https://doi.org/10.1080/00396265.2021.1970918

Kao, S. P., Chen, C. N., Huang, H. C., & Shen, Y. T. (2014). Using a least squares support vector machine to estimate a local geometric geoid model. Boletim de Ciências Geodésicas, 20(2), 427-443. https://doi.org/10.1590/s1982-21702014000200025

Kohavi, R. (1995). A study of cross‐validation and bootstrap for accuracy estimation and model selection. 14th International Joint Conference on Artificial Intelligence, Montreal, Quebec, Canada, 1137-1143.

Konakoglu, B., & Akar, A. (2021). Geoid undulation prediction using ANNs (RBFNN and GRNN), multiple linear regression (MLR), and interpolation methods: A comparative study. Earth Sciences Research Journal, 25(4), 371-382. https://doi.org/10.15446/esrj.v25n4.91195

Pavlis, N., Holmes, S., Kenyon, S., & Factor, J. (2012). The development and evaluation of the Earth Gravitational Model 2008 (EGM08). Journal of Geophysical Research: Solid Earth, 117(B4), 1-38. https://doi.org/10.1029/2011JB008916

Rabah, M., & Kaloop, M. (2013). The use of minimum curvature surface technique in geoid computation processing of Egypt. Arabian Journal of Geosciences, 6(4), 1263-1272. https://doi.org/10.1007/s12517-011-0418-0

Rangelova, E., Fotopoulos, G., & Sideris, M. G. (2010). Implementing a dynamic geoid as a vertical datum for orthometric heights in Canada. In S. Mertikas (Ed.), Gravity, Geoid and Earth Observation, 135, 299-306. https://doi.org/10.1007/978-3-642-10634-7_38

Raufu, I. O. (2024). Evaluation of different digital elevation models (DEMs) for geospatial applications: A case study of Ibadan, Nigeria. African Journal on Land Policy and Geospatial Sciences, 7(4), 1137-1154. https://doi.org/10.48346/IMIST.PRSM/ajlp-gs.v7i4.49875

Raufu, I. O., & Tata, H. (2021). Accuracy assessment of different polynomial geoid models in orthometric height determination for Akure, Nigeria. Geodetski Glasnik, 52, 61-73.

Raufu, I. O., & Tata, H. (2022). Comparison of two corrector surface models of orthometric heights from GPS/levelling observations and global gravity model. Journal of Geospatial Information Science and Engineering, 5(1), 15-20. https://doi.org/10.22146/jgise.72531

Rodriguez, J. D., Perez, A., & Lozano, J. A. (2009). Sensitivity analysis of k-fold cross-validation in prediction error estimation. IEEE Transactions on Pattern Analysis and Machine Intelligence, 32(3), 569-575. https://doi.org/10.1109/TPAMI.2009.187

Şahin, M., Kaya, Y., & Uyar, M. (2013). Comparison of ANN and MLR models for estimating solar radiation in Turkey using NOAA/AVHRR data. Advances in Space Research, 51(5), 891-904. https://doi.org/10.1016/j.asr.2012.10.010

Sorkhabi, O. M. (2015). Geoid determination based on log sigmoid function of artificial neural networks: A case study: Iran. Journal of Artificial Intelligence in Electrical Engineering, 3(12), 18-24.

Stone, M. (1974). Cross-validatory choice and assessment of statistical predictions. Journal of the Royal Statistical Society: Series B (Methodological), 36(2), 111-133. https://doi.org/10.1111/j.2517-6161.1974.tb00994.x

Trojanowicz, M., Osada, E., & Karsznia, K. (2020). Precise local quasigeoid modelling using GNSS/levelling height anomalies and gravity data. Survey Review, 52(367), 76-83. https://doi.org/10.1080/00396265.2018.1525981

Zaletnyik, P., Völgyesi, L., & Paláncz, B. (2008). Modelling local GPS/levelling geoid undulations using support vector machines. Periodica Polytechnica Civil Engineering, 52(1), 39-43. https://doi.org/10.3311/pp.ci.2008-1.06

Ziggah, Y. Y., Youjian, H., Tierra, A. R., & Laari, P. B. (2019). Coordinate transformation between global and local data based on artificial neural network with k-fold cross-validation in Ghana. Earth Sciences Research Journal, 23(1), 67-77. https://doi.org/10.15446/esrj.v23n1.63860

How to Cite

APA

Raufu, I. O., Azeez, A. & Ojo, A. (2026). Machine Learning Approaches for Predicting Geoid Undulations to Improve Orthometric Height Determination in Data-Scarce Regions. Earth Sciences Research Journal, 30(1), 41–51. https://doi.org/10.15446/esrj.v30n1.122297

ACM

[1]
Raufu, I.O., Azeez, A. and Ojo, A. 2026. Machine Learning Approaches for Predicting Geoid Undulations to Improve Orthometric Height Determination in Data-Scarce Regions. Earth Sciences Research Journal. 30, 1 (May 2026), 41–51. DOI:https://doi.org/10.15446/esrj.v30n1.122297.

ACS

(1)
Raufu, I. O.; Azeez, A.; Ojo, A. Machine Learning Approaches for Predicting Geoid Undulations to Improve Orthometric Height Determination in Data-Scarce Regions. Earth sci. res. j. 2026, 30, 41-51.

ABNT

RAUFU, I. O.; AZEEZ, A.; OJO, A. Machine Learning Approaches for Predicting Geoid Undulations to Improve Orthometric Height Determination in Data-Scarce Regions. Earth Sciences Research Journal, [S. l.], v. 30, n. 1, p. 41–51, 2026. DOI: 10.15446/esrj.v30n1.122297. Disponível em: https://revistas.unal.edu.co/index.php/esrj/article/view/122297. Acesso em: 20 jul. 2026.

Chicago

Raufu, Ibrahim Olatunji, Abubakri Azeez, and Aderemi Ojo. 2026. “Machine Learning Approaches for Predicting Geoid Undulations to Improve Orthometric Height Determination in Data-Scarce Regions”. Earth Sciences Research Journal 30 (1):41-51. https://doi.org/10.15446/esrj.v30n1.122297.

Harvard

Raufu, I. O., Azeez, A. and Ojo, A. (2026) “Machine Learning Approaches for Predicting Geoid Undulations to Improve Orthometric Height Determination in Data-Scarce Regions”, Earth Sciences Research Journal, 30(1), pp. 41–51. doi: 10.15446/esrj.v30n1.122297.

IEEE

[1]
I. O. Raufu, A. Azeez, and A. Ojo, “Machine Learning Approaches for Predicting Geoid Undulations to Improve Orthometric Height Determination in Data-Scarce Regions”, Earth sci. res. j., vol. 30, no. 1, pp. 41–51, May 2026.

MLA

Raufu, I. O., A. Azeez, and A. Ojo. “Machine Learning Approaches for Predicting Geoid Undulations to Improve Orthometric Height Determination in Data-Scarce Regions”. Earth Sciences Research Journal, vol. 30, no. 1, May 2026, pp. 41-51, doi:10.15446/esrj.v30n1.122297.

Turabian

Raufu, Ibrahim Olatunji, Abubakri Azeez, and Aderemi Ojo. “Machine Learning Approaches for Predicting Geoid Undulations to Improve Orthometric Height Determination in Data-Scarce Regions”. Earth Sciences Research Journal 30, no. 1 (May 27, 2026): 41–51. Accessed July 20, 2026. https://revistas.unal.edu.co/index.php/esrj/article/view/122297.

Vancouver

1.
Raufu IO, Azeez A, Ojo A. Machine Learning Approaches for Predicting Geoid Undulations to Improve Orthometric Height Determination in Data-Scarce Regions. Earth sci. res. j. [Internet]. 2026 May 27 [cited 2026 Jul. 20];30(1):41-5. Available from: https://revistas.unal.edu.co/index.php/esrj/article/view/122297

Download Citation

CrossRef Cited-by

CrossRef citations0

Dimensions

PlumX

Article abstract page views

130

Downloads

Download data is not yet available.