Data rescue
A major challenge in climate-related or impact-related studies is the lack of historical data against which to assess biological response to changes in the physical environment. As such, historical data are both invaluable and irreplaceable, yet the raw data needed for their modern re-analysis is seldom formally published. Raw data was frequently lost as researchers retired, and the samples themselves were often discarded after processing as facilities do not exist for their long-term curation.
"Data rescue" refers to a variety of activities directed at moving information at risk of being lost into electonic format so they can be used for future analyses. Such actvities include:
- Keying in raw data in tables of hard-copy reports
- Extracting raw data tables from notebooks of retired, or retiring, scientists
- Locating and processing biological samples that have never been analysed
- Re-processing samples originally analysed at low taxononomic resolution (i.e. not to species)
- Filling in importent missing details in existing datasets (or the files describing them)
ArcOD has been involved in all of these forms of data rescue, the products of which are now posted on this website and served by international data systems such as OBIS. Rescue of the WEBSEC program datasets provide examples of several of these different approaches.
Page Author: Russ Hopcroft
Revised: Sept 30, 2011