In mountainous regions, a gravimetric geoid determination is a difficult task which needs special attention to obtain reliable results satisfying the demands, e.g. of the engineering infrastructure. The present study compares the KTH (Royal Institute of Technology) method with the remove-compute-restore (RCR) technique to attain a precise geoid model for the Konya Closed Basin (KCB) and thus aims to clarify the theoretical and numerical differences between these techniques. The KCB located in the central part of Turkey is a mountainous area with a very limited amount of the ground gravity data. Various data sources, such as the Turkish Digital Elevation Model with 3 '' x 3 '' resolution (TRDEM3), a recently published satellite-only global geopotential model and terrestrial gravity observations are combined in the gravimetric geoid models of the KCB by applying the KTH and RCR techniques, separately. The both gravimetric geoid models (KG-KTH11 and KG-RCR11) are compared with GNSS/levelling data at the control points in the target area, resulting the RMS (Root Mean Square Error) differences of +/- 6.7 cm and +/- 9.8 cm in the absolute sense, respectively. These results show that the KTH method gives more reasonable accuracy than the RCR technique in a mountainous area with scarce terrestrial gravity data. (C) 2012 Elsevier Ltd. All rights reserved.