I see what you mean. Without going through locations one-by-one, I would like to determine which locations in the world most closely match this graph. Is there any way to do this quickly?

Hi, and thanks for posting. Short answer: history+ won't give you what you're after, because it contains historical weather simulations (a reanalysis-style archive), not an archive of the forecasts as they were originally issued. The Forecast API can return past forecasts via the history_days parameter, but only for the previous 4 days, so it isn't suitable for a 3-year comparison either. A few useful things to know: The history+ archive is built from weather simulations (NEMS, ERA5 and local high-resolution models), not from station measurements. It's a very good reference for "what the weather actually did" at a location, but it isn't raw observational data, and it isn't a record of what was forecast at the time. For a true forecast-versus-observation study over multiple years, you'd need (a) an archive of the forecasts as issued on each day, and (b) a reference dataset to compare against (either station observations or a reanalysis like ERA5 from history+ or the Dataset API). An archive of past forecasts going back years isn't part of the standard self-service offering. If this is for a specific project, the best route is to write to support@meteoblue.com with the locations, time range and variables you need. The team can advise on whether a custom dataset or a tailored verification extract is feasible. If it would help, let us know a bit more about what you're trying to achieve (research, due diligence on forecast quality for a site, model selection, etc.) and we can point you at the most useful next step. Best regards, The meteoblue team

Hello, and thank you for your question. The precipitation anomaly in meteoblue climate change data represents a deviation from the weekly or yearly mean but is not always calculated simply as the difference between the year mean and the historical mean. The anomaly calculation may include additional statistical methods or reference periods beyond direct subtraction. Furthermore, ERA5 precipitation data incorporates various precipitation types and is influenced by spatial and temporal resolution limitations, which can explain consistent differences such as the 10mm offset you noticed. Updates to datasets, model changes, and reanalysis methods can also contribute to these discrepancies. We hope we were able to answer your questions and wish you pleasant weather with meteoblue :)