DataWiki

This is an old revision of the document!

Key ideas of the scheme design:

1. GGMs include data from multiple sources, while the definition of Interoperable includes Data or metadata must include qualified references to other datasets. Hence, the input data of each model needs to be described, categorized and ideally provided with a PID. So, I extended the RelatedItem field in the DataCite scheme to contain additional info specific to GGM
2. ICGEM proides GNSS Levelling for the hosted models. This is essentially a validation technique with independent measurements from different continents. The levelling shows that model quality might drastically vary from one country to another. For instance, EIGEN-6C4 model has the best quality for Canada region, but is only number 9 in quality for Mexico. Therefore, GNSS levelling information is essential for election of the proper model. I decided to defined a new child element of the metadata sceme to include GNSS Levelling information.
3. Adding conditional information is important, because models and input data sources differ by nature. Temporal and simulated models have a time range, where they are valid. So the description of temporal models will be enriched by the time range. This is inapplicable to static models. Satellite data changed the world of GGMs. If an input data source is a satellite observation, it should be included aalongside with [[level of data processing]https://climatedataguide.ucar.edu/climate-data/nasa-satellite-product-levels ]