Using the Hydrologic Ensemble Forecast Service (HEFS)

4.0 The Hydrologic Ensemble Forecast Service (HEFS) ยป 4.2 Meteorological Forcing Uncertainty

Flowchart for the HEFS

The Meteorological Ensemble Forcing Processor (MEFP) is designed to construct calibrated ensemble meteorological forecast forcings. This includes correcting for biases, merging time periods (i.e. forecasts that operate on different time horizons), and downscaling (i.e. to the hydrologic basin scale).

Meteorological inputs can be a major source of error in the hydrologic forecasts as we see in these QPF ensembles from the Global Ensemble Forecast System (GEFS).

Ensemble traces of QPF from GEFS at a recording station on the upper Ohio River near Pittsburgh in Feb 2017

Here is the 7-day quantitative precipitation forecast from the Global Ensemble Forecast System (GEFS) for the 21 ensemble members.

Question

What can you say about the Day 1 QPF compared to the QPF in days 4-7? (Choose all that apply.)

The correct answers are a, d.

SInce QPF becomes more uncertain as lead time increases, the hydrologic forecasts that use these inputs become more uncertain at longer lead times. The Days 4-7 forecasts contain uncertainty in both the magnitude (vertical spread) and timing (horizontal spread).

Please make a selection.
Cumulative QPF from GEFS at a recording station on the upper Ohio River near Pittsburgh in Feb 2017 with individual ensemble members, the ensemble mean, and the middle 40%

We can also view the GEFS QPF data as a cumulative value over the 7-day period. For the 21 ensemble members, the middle 40% are highlighted in orange, and the black line indicates the median for each time step. The spread in the cumulative outcomes, and thus the forecast uncertainty, clearly increases with lead time.