Filtering of semantically enriched environmental time series

In current time series processing there is a growing urge to add meaning to the information being processed in order to make it more readable for users as well as understandable/interpretable by machines. Especially in the environmental field of computer science, Semantic Web technologies have become more and more popular. Where we had simple time series data, consisting of time stamps an according values in the past, we are now dealing with complex time series slots consisting of not only time and information, but also a lot of additional meta-information. Therefore, we developed a method for semantic filtering of time series meta-information and annotations, as well as methods to visualize the results to the user and show how this meta-information is composed. Our result is a Visualization and Filtering Component, which is able to process annotations similar to simple time series, but filters tags based on users, timing and geo-location. The benefits of this component are the ability to use semantic for filtering of annotations, the visualization/representation of annotations in a time series manner and furthermore the possibility to apply time series processing calculations on semantic annotations. Within the TaToo project and as one of several show cases, we have implemented this method into an application called Climate Twins. Climate Twins demonstrates possible future changes of climate conditions at a regional level. This is done by indicating all regions, which have climate conditions today that are similar to future projected conditions in a certain region of interest. This similarity is based on the climate parameters 'temperature' and 'precipitation'. To be able to filter and visualize annotations on the Climate Twins application as time series, we integrated the Visualization and Filtering Component in the Climate Twins application.