Artifacts in rqatrend due to gaps in the time series (example northern Normandie) #100
Description
I had a look at the strange looking tiles in northern Normandie. This tile had a lot higher rqatrend values compared to the other surrounding tiles.
This is tile EU-E039N018 and the increase in the RQATrend values might come from the fact that we have regular gaps in the time series due to the structure of the data.
The orbit A132 has a gap in the time series for the data in France and the northern half of tile E039N018 gets a bit of data while all other tiles get no data at all.
This gap appears from 2018-04-20 to 2019-12-11. I checked the CDSE Browser and the gap in the data seems to be real.
This gap might lead to an appearance of NaNs in the southern part of the tile due to our data loading.
Apparently the rqatrend code is not handling these regular data gaps well.