Large vegetation-plot databases have opened unprecedented opportunities for investigating vegetation patterns and processes across large spatial and temporal scales. However, such databases are typically created by combining pre-existing smaller databases or datasets that often used different standards and protocols, leading to data inconsistencies (e.g., different taxonomy and nomenclature) and issues (e.g., wrongly georeferenced plots) that must be addressed during the pre-processing of data for analyses. This pre-processing phase is usually time-consuming and may repeat workflows already developed by other researchers. To facilitate the pre-processing of large vegetation-plot databases and avoid duplication of effort, we have compiled a list of what we consider to be the 15 most common data inconsistencies and issues encountered during this stage. Alongside descriptions of their origins and potential impacts on data analysis, we provide solutions for addressing each inconsistency and issue. Furthermore, we present a standardized workflow – an ordered sequence of steps to identify and fix data inconsistencies and issues – that can be adapted to different types of analyses. Besides supporting researchers in the pre-processing stage, this standardized workflow aims to improve transparency and reproducibility in the analysis of large vegetation-plot databases.
原文链接
分享
引用
