Huge amount of missing values for GSP: '139', '239', '87', '92'

[hzloc]

Hi, I've been preparing the dataset, and what have caught my attention was a few GSP station had unbelievable amount of missing data.
I was wondering if this is an issue of retrieving the data, or they are not there between 2022-01-01T12:00:00Z - 2023-01-02T12:00:00Z

[JamieTaylor-TUOS]

@hzloc Thanks for reporting this.

These GSPs are all in Northern Scotland and are either very isolated or serve very small geographical regions. As such, we have insufficient PV generator sample data to model the yield in these GSPs.

For 3 of the 4 you've identified (87, 92, 239), this isn't a problem, since our PV deployment dataset shows no PV systems installed in those regions. These regions are therefore reported via the API as having a generation_mw of 0.0:

• https://api.solar.shef.ac.uk/pvlive/api/v4/gsp/87?extra_fields=installedcapacity_mwp
• https://api.solar.shef.ac.uk/pvlive/api/v4/gsp/92?extra_fields=installedcapacity_mwp
• https://api.solar.shef.ac.uk/pvlive/api/v4/gsp/239?extra_fields=installedcapacity_mwp

The GSP with gsp_id 139 is G_EXTRA_12 which serves the Shetland Isles. Unfortunately, due to its isolated nature, we do not have sufficient sample data to model the PV yield. This isn't usually an issue though for most uses cases, since the estimated installed capacity of PV on the Shetland Isles is just 0.9 MWp.