A new database for time-series monitoring data: the NitroEurope approach

The NitroEurope Integrated Project (2006-2011) aimed to gain a better understanding of the nitrogen cycle and its impacts on the European greenhouse gas balance. Over sixty project partners collected large volumes of data from sites all over Europe. There are currently more than 15 forest sites in 11 countries, including beech, oak, spruce and mixed forests. The collected data are at different temporal resolutions, from one-off measurements to 30-minute time-series data. Around 500 variables (plus attributes) are measured, including soil and plant data, details of cultivation management for natural and manipulated sites, long-term datasets of flux and concentrations of greenhouse gas and pollutants and their precursors, and micrometeorology data. The NitroEurope research community therefore needed a database to provide easy upload and extraction of data for analysis, interpretation and modelling. A bespoke database with a web-based graphical user interface (GUI) in a user-friendly and attractive online environment was produced by the NitroEurope (NitroEurope 2011) database development team. The software provides a flexible interface between Microsoft Excel workbooks and an Oracle database. The database “Form Manager” enables construction of database forms which reflect the project’s Excel data workbook templates. The software locates data according to Excel worksheet and cell references. The database “Uploader” extracts the specified data from completed Excel templates and uploads data directly to the Oracle relational database tables through the web front-end. Data are run through a series of automatic checks on upload which can be verified using an online graphing tool. Data are then validated and made available to the NitroEurope community for downloading. The database “Reporting” tools enable data from different sites and activities to be brought together and datasets can be previewed and graphed before they are downloaded. The database has almost 300 users and currently contains over eight million rows of data. The data will be made available to the wider scientific community two years after the project ends (Spring 2013). The database structure is extremely flexible and has been used for the storage and reporting of other time-series project data at our institute.