1 million-year Synthetic Daily Weather Dataset

# 1 million-year Synthetic Daily Weather Dataset

## Overview

This dataset contains **synthetic daily time series of precipitation and mean temperature** generated using a **stochastic, multi-site, multivariate weather generator** as described in Nguyen et al. (2021, 2024). The weather generator was trained on observed weather data from the **Bamberg station** (Station ID 00282, DWD 2022) for the historical period **1961–2010**.

## Data Description

- **Variables**:  - Daily **precipitation** (mm)  - Daily **mean temperature** (°C)

- **Length**: 1,000,000 years of daily data (synthetic simulation)

- **File structure**:  - The dataset is provided as two compressed ZIP archives:    - `syn_prec_daily_1000000_years.zip`: Contains synthetic **daily precipitation** data.    - `syn_tavg_daily_1000000_years.zip`: Contains synthetic **daily mean temperature** data.  - Each ZIP file contains **10 subfolders**, each representing a **100,000-year chunk**.  - Each chunk includes **2,000 realizations** of **50-year periods**, all spanning the time frame:- **1 January 1961 to 31 December 2010**.  

- **Spatial coverage**: Single location: Bamberg;  ID= 00282;  lon/lat = 10.9206/49.8743

- **Temporal structure**: Daily values with stationary statistical properties

## Methodology

The dataset was generated using a **stationary version** of the weather generator by Nguyen et al., which captures both marginal distributions and temporal dependencies.

- **Precipitation**:  - Modeled using a **monthly-based extended Generalized Pareto Distribution (GPD)**.- **Temperature**:  - Modeled using a **normal distribution**, conditioned on precipitation occurrence.

The generator accounts for monthly seasonality but assumes stationarity across the entire simulated period.

## Applications

This synthetic dataset can be used for:- Hydrological and climate risk modeling- Extreme value analysis (e.g., block maxima, return periods)- Ensemble simulation for stress-testing models- Probabilistic impact assessments

## References

DWD: Climate Data Centre: Station ID 00282, DWD [dataset], https://opendata.dwd.de/climate_environment/CDC/observations_germany/climate/ (last access: 15 December 2022), 2022

Macdonald, E., Merz, B., Guse, B., Nguyen, V. D., Guan, X., and Vorogushyn, S.: What controls the tail behaviour of flood series: rainfall or runoff generation?, Hydrol. Earth Syst. Sci., 28, 833–850, https://doi.org/10.5194/hess-28-833-2024, 2024.

Nguyen, V. D., Merz, B., Hundecha, Y., Haberlandt, U., and Vorogushyn, S.: Comprehensive evaluation of an improved large-scale multi-site weather generator for Germany, Int. J. Climatol., 41, 4933–4956, https://doi.org/10.1002/joc.7107, 2021.

Nguyen, V. D., Vorogushyn, S., Nissen, K., Brunner, L., and Merz, B.: A nonstationary climate-informed weather generator for assessing future flood risks. Advances in Statistical Climatology, Meteorology and Oceanography, 10(2), 195–216, https://doi.org/10.5194/ascmo-10-195-2024, 2024.

## License

This dataset is published under the **Creative Commons Attribution 4.0 International (CC BY 4.0)** license. You are free to use, share, and adapt the data with appropriate credit.

## Contact

For questions or feedback, please contact:

**Viet Dung Nguyen (PhD)**  Section Hydrology  Email: [e-mail]  [GFZ Profile](https://www.gfz.de/en/staff/viet.dung.nguyen)  [ORCID](https://orcid.org/0000-0002-2649-2520)