This research aimed to determine the optimal interpolation methods for predicting unsampled groundwater levels in Abu Dhabi Emirate, UAE. Geostatistical and deterministic interpolation techniques including Inverse Distance Weighting, Ordinary Kriging, Radial Basis Function, Global Polynomial Interpolation (GPI) and Local Polynomial Interpolation (LPI) with different parameters were applied using ArcGIS software. A dataset of 257 wells spanning the period from 2002 to 2022 was utilized with two data selection approaches (random and non-random). The LPI and the Exponential Ordinary Kriging models demonstrated the highest efficiencies and lowest errors across both selected methods. In contrast, despite high cross-validation performance, GPI was found to be unsuitable due to model limitations. Comparisons between random and non-random data selection approaches revealed the superiority of models with random data selection, enabling comprehensive analysis and visualization of groundwater levels. These findings highlight the importance of selecting appropriate interpolation techniques and data strategies to support effective groundwater management, thereby guiding sustainable monitoring and management practices in arid environments such as Abu Dhabi Emirate.