Test name Check_user_percentile

[lisavalexander]

Tests user-specified percentiles (supplying user-generated percentiles)

[heroldn]

Not sure how this is to be tested. In any case, testing gridded datasets of percentiles would probably be processor and time consuming. For single time series data though (via the GUI) comparisons of the user-provided percentiles would be doable.

[lisavalexander]

Nick this specifically related to the GUI calculations. I think because we are pushed for time we can add this to the 'future issues' pile since it's the gridded output that we need to check in the short-term.

[cicatrixx]

Hi @heroldn! Is there a way to provide user-defined threshold for percentile based indices when using station data in batch processing? The user manual has a workflow to use for netcdf data but not so for station data. I want to use a historical base period for assessing percentile based indices for future time period. Currently I have to put in both historical and future timeseries as input, but doing so also evaluates the trend over the entire historical+future timeseries. I only want to look at trend for the future?

[heroldn]

Hi Sanita, there isn't a feature like that unfortunately, though you should be able to calculate the index for the whole time series (present plus future) then use the .csv output to calculate a trend just over the years you're interested in. Excel would give you a linear trend or use R for ones like sen's slope. Nick.

…

On 27/7/20 10:42 pm, Sanita Dhaubanjar wrote: Hi @heroldn <https://github.com/heroldn>! Is there a way to provide user-defined threshold for percentile based indices when using station data in batch processing? The user manual has a workflow to use for netcdf data but not so for station data. I want to use a historical base period for assessing percentile based indices for future time period. Currently I have to put in both historical and future timeseries as input, but doing so also evaluates the trend over the entire historical+future timeseries. I only want to look at trend for the future? — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#29 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ACW3FAK45KIXEFAJJRJFOKTR5VYZVANCNFSM4BAQDYZA>.