precip_Niamey_2016: Precipitation forecasts and observations at Niamey, Niger in...

In reliabilitydiag: Reliability Diagrams Using Isotonic Regression

Precipitation forecasts and observations at Niamey, Niger in July to September 2016

Description

A data set containing 24-hour ahead daily probability of precipitation forecasts of four forecasting methods and corresponding observations of precipitation occurrence.

For a detailed description of the four prediction methods, see Vogel et al (2021).

Usage

precip_Niamey_2016

Format

A data frame with 92 rows and 6 variables:

date

a date from "2016-07-01" to "2016-09-30" in Date format.

Logistic

prediction based on logistic regression, as a probability.

EMOS

prediction based on EMOS method, as a probability.

ENS

prediction based on ECMWF raw ensemble, as a probability.

EPC

prediction based on EPC method, as a probability.

obs

observation, indicator variable where 1 represents the occurrence of precipitation.

Source

Vogel P, Knippertz P, Gneiting T, Fink AH, Klar M, Schlueter A (2021). "Statistical forecasts for the occurrence of precipitation outperform global models over northern tropical Africa." Geophysical Research Letters, 48, e2020GL091022. doi: 10.1029/2020GL091022.

This data set contains modified historic products from the European Center for Medium-Range Weather Forecasts (ECMWF, https://www.ecmwf.int/), specifically: ensemble forecasts of precipitation that have been summarized to a probability of precipitation (column ENS), and historical observations for the occurence of precipitation (column obs). The ECMWF licenses the use of expired real-time data products under the Creative Commons Attribution 4.0 International (CC BY 4.0, https://creativecommons.org/licenses/by/4.0/).

reliabilitydiag documentation built on June 29, 2022, 1:05 a.m.