As of this month we have introduced the flagging entries (manual quality checks) to our database. This does not sound astounding but this means that both datashop and API now can (and by default) do flag data.

Since these are usually manual (automatic checks happen before writing to the database) there is some delay but they provide a better end user experience and trust in our data

Plus they are quite useful for us too and our internal checks and dashboards. So far we have flagged almost 160k periods (over all timeseries, the flagged time period varies of course with the reason for the flagging and often more than one timeseries is flagged per stations). The maximum number of entries is almost 500 for one timeseries.

IT infrastructure changes everywhere, also at the University of Basel and the atmospheric sciences. Namely, we moved our database to a new server and data storage. Combined with this, we merged our meta information and slightly restructured our tables

For the time being, both old and new servers will operate side by side but this page (dolueg2) uses the new server. In the background our data pipelines needed to be adjusted but for the users, no changes should be evident. The next step is to allow better quality flagging - stay tuned...!

As of this July, the EC system at the tower top in BKLI has been redesigned and modernised to replace the more than 20 year old Gill HS and a at the time very progressive sampling solution with labview.

With changing IT needs and instrumentation, this has been modified to run similar to other stations directly on the logger. This facilitates the whole data pipeline. We hope for great measurements!

In the background of the Namib Fog Life Cycle Analysis project the need to see fog events together with station data from FogNet becamse apparent.

We thus tailored a GDAL pipeline using existing MSG data to create band composites, add stations and their data (wind and fog precipitation) to them and animate them. This has led to many videos that are crucial for interpretation of fog events and their often quite unique dynamics.

Fitting the overarching idea to consolidate the website of the MCR all the old project pages have been integrated. This allows interested visitors to learn easily more about the research by the MCR

This integration would also allow us to finetune future projects for display and information, an advantage that we could reap for NaFoLiCA

The old dolueg faced several issues, most severly the issue of plot management and adjustment. The design just relied too much on hardcoded plots and thus was also limited for the MCR. Additionally, the backend for plot production reliad on IDL, namely the PNG pipeline which has resolution issues in the version 7.5.X. A first development was to instead create EPS/PS plots which are vector type graphics and thus far more suitable for many timeseries related displays. At the time and with the given version however, we needed to use latex to convert the PS figures to SVG that is more suitable for webdisplay.

To match the new faciliate making of graphics the frontend of dolueg needed an urgent facelift, more modern redesign and core functionality to reduce manual work and hardcoding in the back. This was implemented with a dynamic search function and attritbution to station/variables. Another 10 years later, the workflow has remained the same despite many changes in stations, databases and new programming languages