Malcheva et al. (2026) Analysis of Precipitation and Regionalization of Torrential Rainfall in Bulgaria

Identification

• Journal: Climate
• Year: 2026
• Date: 2026-02-01
• Authors: Krastina Malcheva, Neyko Neykov, Lilia Bocheva, Anastasiya Stoycheva, Nadya Neykova
• DOI: 10.3390/cli14020039

Research Groups

National Institute of Meteorology and Hydrology, Sofia, Bulgaria

Short Summary

This study comprehensively analyzes precipitation regimes and their links to atmospheric circulation in Bulgaria (1991–2020) and proposes a regionalization of torrential rainfall, finding it primarily associated with low-pressure systems, easterly/northeasterly flows, and weak-gradient pressure fields.

Objective

• To provide a thorough overview of precipitation regimes in Bulgaria under contemporary climate conditions.
• To investigate the connections between precipitation regimes, hazardous rainfall, and atmospheric circulation over the country.
• To propose a regionalization specific to torrential rainfall in Bulgaria.

Study Configuration

• Spatial Scale: Bulgaria (Southeastern Europe, eastern Balkan Peninsula). Daily precipitation data from 355 meteorological stations (average area of approximately 320 km² per station). Mean sea level pressure (MSLP) data from NCEP-NCAR reanalysis covering the European sector (5° E–45° E, 25° N–60° N) on a 2.5° × 2.5° grid.
• Temporal Scale: Daily precipitation data and NCEP-NCAR reanalysis for 1991–2020. MeteoAlarm warning data for 2011–2020. Monthly precipitation normals calculated for 1991–2020.

Methodology and Data

• Models used:
  • Automated Jenkinson–Collison (JC) classification for Circulation Weather Types (CWTs).
  • Principal Component Analysis (PCA) in T-mode with varimax rotation.
  • k-means clustering of component scores.
  • R statistical software (version 4.4.2), RStudio (2024.09.1), QGIS Desktop (3.28.5).
• Data sources:
  • Daily precipitation data from the meteorological archive of the National Institute of Meteorology and Hydrology (NIMH) from 355 stations in Bulgaria.
  • Daily mean sea level pressure (MSLP) data from the NCEP-NCAR reanalysis.
  • MeteoAlarm (MA) warning data on heavy precipitation (yellow, orange, red codes) issued by NIMH.
  • Digital terrain model AW3D30 from the Japan Aerospace Exploration Agency (JAXA).

Main Results

• Annual precipitation normals in Bulgaria range from 450 mm to 550 mm in lowlands to 900 mm to 1200 mm in mountainous areas, with a country average of 663 mm.
• Seven distinct precipitation regimes were identified, including continental (summer maximum, winter minimum), Mediterranean (winter maximum, summer minimum), and transitional types.
• Hazardous Rainfall Events (HREs) exhibit clear seasonality, with 43.1% occurring in summer (May-July peak). Their contribution to annual precipitation ranges from 4% to 21.4%, with a median of approximately 10% for stations below the country average and 18.4% for stations with the highest annual normals.
• HREs are primarily associated with cyclonic circulation (53% of HREs), easterly and northeasterly flows (30% of HREs), and low-flow circulation (weak-gradient pressure fields, 16% of HREs).
• A new regionalization of torrential rainfall was proposed, consisting of nine clusters, each characterized by a unique combination of contributing CWTs.
• There is a good agreement (Spearman correlation coefficient of 0.670) between the weighted contribution of clusters to torrential rainfall by district and the calculated torrential rainfall from station data.

Contributions

• This study provides the first comprehensive analysis in Bulgaria linking precipitation regimes, hazardous rainfall, and atmospheric circulation using objective methods (Jenkinson–Collison classification, T-mode PCA, k-means clustering) and three independent data sources.
• It is the first time that MeteoAlarm warning data for hazardous rainfall have been used as a reference in climatological studies and for the regionalization of hazardous rainfall in Bulgaria.
• The proposed regionalization offers a physically meaningful basis for evaluating region-specific precipitation forecasting, particularly for extreme events, and for delineating regions with distinct circulation conditions that trigger torrential rainfall.
• The identified torrential rainfall patterns can support future assessments of extreme event shifts using CMIP6 projections and aid in flood hazard zoning by improving precipitation quantile estimations.

Funding

This research received no external funding.

Citation

@article{Malcheva2026Analysis,
  author = {Malcheva, Krastina and Neykov, Neyko and Bocheva, Lilia and Stoycheva, Anastasiya and Neykova, Nadya},
  title = {Analysis of Precipitation and Regionalization of Torrential Rainfall in Bulgaria},
  journal = {Climate},
  year = {2026},
  doi = {10.3390/cli14020039},
  url = {https://doi.org/10.3390/cli14020039}
}