Dataset upload format

• Datasets are submitted as a table with one event per row. Use .csv format.
• Columns are organised by interval and event, with an extra header row showing where each interval set/block begins.
• Multiple interval sets can be defined - e.g. zones and subzones - they just need to be defined in separate blocks of columns.
• The interval or event term (intervalx, eventx) must be given in the first col of the block otherwise the column order within each block is not important.
• If a column header is not a field in the database the column will be ignored.
• For valid column headers see the table in the simple format section
• Event definitions are given in the last block of columns.
• Interval definitions are given on the same row as the event marking the base of the interval.
• Gaps - add 'gap' or 'TOP' at the base of all gaps including base of table if bottom row is not an interval.
• added_abvs - use options table to automatically add these to intervals. Events need the addded abvs to be given BEFORE upload
• alternative_terms - for zonal names based on taxa use the auto_abbreviate option in the table below to add an set of alternative terms with the generic name abbreviated. (e.g. D. pentaradiatus zone from Discoaster pentaradiatus zone)
• A sort column should be added before the first data block - this is used to record the row of data in the original source workbook.
• Empty rows - rows without either eventx or intervalx data will be ignored so could be used to leave comments in the table.
• The eventx column needs to be filled for all rows with interval definitions, but including a pre-defined event (e.g base Albian, top Holocene) is ok. These will be used for the interval base but the stored event data will not be altered..
• The table below gives an example of the spreadsheet layout.
• Debugging - check results table for incomplete rows.