E1S2: Compute ensemble statistics on forecast data structures

To analyse the forecast uncertainty, ensemble statistics need to be computed on the forecast data structures.

hazard = HazardForecast(...)
haz1 = hazard.select(lead_time=t).max()  # Max over all ensemble members at lead time t
haz2 = hazard.quantile(0.75, dim="member")  # 75th percentile for each lead time
haz3 = hazard.median(dim="lead_time")  # Median over all lead times for each ensemble member

Statistics:

min, max
mean
quantile
median (quantile=0.5)
standard deviation (?)

API:

Public methods in derived classes.

Concat does not work. Workaround:

class HazardForecast:
    @classmethod
    def concat(cls, *obj):
        hazard = Hazard.concat(*obj)
        member, lead_time = np.concatenate(...)
        return cls(hazard, member, lead_time)

Navigation Menu

Repository navigation

E1S2: Compute ensemble statistics on forecast data structures #1120

Metadata

Assignees

Labels

Type

Projects

Milestone

Relationships

Development

E1S2: Compute ensemble statistics on forecast data structures #1120

Description

Activity

peanutfun commented on Dec 8, 2025

Metadata

Metadata

Assignees

Labels

Type

Projects

Milestone

Relationships

Development

Issue actions