In geology context is everything, and repeat observations essential. The best way to store such data is in a Wiki knowledge-base
Many geological observations are through proxies and these proxies have developed over time. Well analyses need to be considered in light of sampling quality (core vs drill cuttings). In an ideal world all data should be checked and cross-checked, but in reality it is not, and anomalous data needs validating. What was found in the well nearby? or in outcrops onshore - but who did that work, using what method? These important points are naturally accommodated in a Wiki, but not in a relational database