See Forecast Data fields for additional information for the v2/forecast_timeseries endpoint. The different output formats are shown below.
Plot
The time series shows the selected values for the deterministic hourly 5-day forecast.
The time series for daily timescale shows the mean and range of forecasted values by percentile.
The time series for the weekly, monthly, and quarterly forecasts is a series of box plots. The box plots show the (25th, 50th, and 75th) percentiles. 25th Percentile - Also known as the first, or lower, quartile. The 25th percentile is the value at which 25% of the values lie below that value, and 75% of the values lie above that value. 75th Percentile - Also known as the third, or upper, quartile. The 75th percentile is the value at which 25% of the values lie above that value and 75% of the values lie below that value.
CSV table format
- The first four (4) columns are the same across all timescales: Latitude, Longitude, Timescale, Model.
- For hourly, the next column is “Valid Time” (yyyy/mm/dd HH). The final column is the requested variable with units in the header name.
- For daily, the next column is “Valid Day” (yyyy/mm/dd). The final column is the requested variable with units in the header name.
- For sub-seasonal, seasonal, and long-range, the next two columns are “Forecast Start” and “Forecast End” (yyyy/mm/dd) denoting the beginning and end of the forecast window. Each column thereafter is a different forecast quantile with a header named “x% Quantile”. The variable and units are given as a title above the table on the page. In statistics and probability, quantiles are cut points dividing the range of a probability distribution into continuous intervals with equal probabilities.
netCDF format
- To ensure clarity for the forecast periods at each time scale above daily, the label describing the dimensions of the data is “forecast_date_<scale>”, where <scale> is one of weekly, monthly, quarterly. Similarly, the name of the variable in the dataset is “<key>_<scale>” where <key> is one of “vals” or “anom”.
FAQs
- For point requests, is there interpolation between grid cells? i.e. if a location on the western edge of a grid is selected, does it default to the grid the point is contained in OR does it do a weighted average of nearby cells?