Orbit selection for a space mission is a process rather than a set of explicit calculations. The general orbit selection process has a few steps. First, establishing orbit type, orbit-related mission requirements and orbit performance, evaluating orbit performance for many candidates, and finally, performing orbit trades and selecting several optimal design alternatives. The last selecting process is a bit fuzzy, as many design candidates exit, and usually, there is no precise answer or correct solution. The output of orbit selection is several different credible orbits rather than a single answer. However, finding such orbits is a complex task due to the fact that most candidates achieve comparable performance, making it difficult to pick up the recommended orbits by inspection. The reason behind this is the orbit candidates are represented by a multi-dimensional table containing numerous orbit parameters and many performance-related figures of merit. Therefore, an effective orbit selection method is required to aid the designers in discovering the recommended design alternatives. To address the issue, this study investigates the utility of tradespace exploration in orbit selection. As proof of concept, a pilot study was conducted on the orbit data from the Satellite Cross-Calibration Radiometer (SCR) series Pre-Phase A Study Report. The results demonstrate that the tradespace exploration provides a data-driven decision-making process for orbit selection. The principal component analysis (PCA) provides a clear visualisation of all design alternatives, and Pareto-optimization can be used to determine the optimal orbits. Moreover, using clustering analysis on the design parameters enables the discovery of hidden patterns that would otherwise be missed by human inspection analysis. The pilot study demonstrates that tradespace exploration is a promising method for supporting faster, more accurate, and more reliable orbit selection.