Actually, I've also just decided to purge any SSI=12 reading from the historic data as well - so the query will work as it stands.

Thanks for this solution, it worked perfectly! I tried it against the real-world version of this data, which has over 5000 samples for each one of over 3500 waysides. It...

perfect. It was the outer join that was eluding me.

thank you!

I admit I'm not sure how this query works, but it returns one record for each address, with the latest time from either table:

wea ...

Thanks for the reply. When I substituted actual table/column names I got syntax errors:

the first table (3000 individual records) is rt_group_status. Second table (over 80 million records) is rt_group_coverage....