Evaluating Multivariate Adaptive Regression Splines (MARS) for Modelling Observed Heights Above Mean Sea Level: A Case Study of the Tarkwa Local Geodetic Reference Network in Ghana
DOI:
https://doi.org/10.58825/jog.2025.19.2.246Keywords:
Data Mining, Reference System, Multivariate Adaptive Regression Splines, Regression Model, Model Evaluation and ValidationAbstract
This study examines the potential of Multivariate Adaptive Regression Splines (MARS) in predicting recorded heights above mean sea level within the Tarkwa Local Geodetic Reference Network in Ghana. Logistical and computational constraints of conventional techniques, such as spirit levelling and geostatistical interpolation, drive the assessment of MARS as a strong soft computing substitute. The MARS model was trained and verified using field-measured data gathered using a Total Station DTM 122A, and its performance was compared against the Polynomial Regression model (PRM) and Kriging models. Each model technique was assessed based on statistical models such as arithmetic mean absolute error (AMAE), arithmetic mean squared error (AMSE), arithmetic root mean squared error (ARMSE), arithmetic standard deviation (ASD), correlation coefficient (R), and coefficient of determination (R2). Statistical measures showed MARS's better accuracy utilizing near-perfect correlation (AMAE: 1.7963E-06 m; AMSE: 8.6775E-12 m) and low error margins. The results show MARS to be a possible, high-precision solution for orthometric height calculation, hence improving Ghana's geodetic network uses in environmental management, building, and surveying. This work not only confirms the effectiveness of MARS but also provides a basis for improving height measurement methods in local geodetic systems.
References
Al-Krargy, E. M., H. F. Mohamed, M. M. Hosney and G. M. Dawod (2017). A High-Precision Geoid for Water Resources Management: A Case Study in Menofia Governorate, Egypt. National Water Research Center (NWRC) Conference on Research and Technology Development for Sustainable Water Resources Management, Cairo, Egypt, pp. 1–13.
Boye, C. B., M. S. Peprah and N. K. Kodie (2018). Geographic Assessment of Telecommunication Signals in a Mining Community: A Case Study of Tarkwa and Its Environs. Ghana Journal of Technology, 2(2), 41–49.
Doganalp, S. and H. Z. Selvi (2015). Local Geoid Determination in Strip Area Projects by using Polynomials, Least Squares Collocation and Radial Basis Functions. Measurement, 73, 429–438.
Erol, B. (2011). An automated height transformation using precise geoid models. Journal of Scientific Research and Essays, 6(6), 1351–1363.
Erol, B. and R. N. Celik (2005). Modelling local GPS/levelling geoid with the assessment of inverse distance weighting and geostatistical kriging methods. Athens, Greece, 1–5.
Falchi, U., C. Parente and G. Prezioso (2018). Global geoid adjustment on local area for GIS applications using GNSS permanent station coordinate. Geodesy and Cartography, 44(3), 80–88.
Friedman, J. H. (1991). Multivariate adaptive regression splines. Annals of Statistics, 19, 1–67.
Herbert, T. and M. N. Ono (2018). A gravimetric approach for the determination of orthometric heights in Akure environs, Ondo State, Nigeria. Journal of Environment and Earth Science, 8(8), 75–80.
Joe-Asare, T., M. S. Peprah and M. M. Opoku (2018). Assessment of the potability of underground water from a small-scale underground mine: A case study. Ghana Mining Journal, 18(2), 61–67.
Kaloop, M. R., A. Zaki, H. Al-Ajami and M. Rabah (2019). Optimizing local geoid undulation model using GPS/levelling measurements and heuristic regression approaches. Survey Review, 1–12.
Konakoglu, B. and L. Cakir (2018). Generalized regression neural network for coordinate transformation. International Symposium on Advancements in Information Sciences and Technologies (AIST), Montenegro, 5–8 September 2018, pp. 66–78.
Kumi-Boateng, B. and M. S. Peprah (2020). Modeling local geometric geoid using soft computing and classical techniques: A case study of the University of Mines and Technology (UMaT) local geodetic reference network. International Journal of Earth Science Knowledge and Applications, 2(3), 166–177.
Najeeb, Z. A. (2017). Predicting digital elevation model height by multivariate adaptive regression splines (MARS) interpolation approach. Global Journal of Researches in Engineering: E Civil and Structural Engineering, 17(3), 1–8.
Ophaug, V. and C. Gerlach (2017). On the equivalence of spherical splines with least-squares collocation and Stokes’s formula for regional geoid computation. Journal of Geodesy, 91(11), 1367–1382
Peprah, S. M., Y. Y. Yevenyo and I. Issaka (2017). Performance evaluation of the Earth Gravitational Model (EGM2008) – A case study. South African Journal of Geomatics, 6(1), 47–72.
Poku-Gyamfi, Y. (2009). Establishment of GPS reference network in Ghana. MPhil Dissertation. Universität der Bundeswehr München, Werner Heisenberg-Weg 39, 85577, Germany, pp. 1–218.
Poku-Gyamfi, Y. and T. Schuder (2008). Renewal of Ghana’s geodetic reference network. 13th FIG Symposium on Deformation Measurement and Analysis, 4th IAG Symposium on Geodesy for Geotechnical and Structural Engineering, LNEC, Lisbon, 12–15 May 2008, pp. 1–10.
Soycan, M. (2014). Improving EGM2008 by GPS and levelling data at locale scale. Boletim de Ciências Geodésicas, 20(1), 3–18.
Yakubu, I., Y. Y. Ziggah and M. S. Peprah (2018). Adjustment of DGPS data using artificial intelligence and classical least square techniques. Journal of Geomatics, 12(1), 13–20.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2025 Journal of Geomatics
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
