We have recreated for several points in the area you had specified but if you look at our figure, the time series look fine. Please make sure you have the files in the correct order. If the problem continues please share your script with us
The metadata point is a good question – comprehensive metadata is included with the data provided on the web presence, and so I would assume that it should be enough to copy this into the QUACK where you have used this data ‘as is’. Any new data could refer back to this metadata, and perhaps include some further lineage so it is clear and transparent what has been done to the original dataset. Does that sound reasonable?