Stošić et al. (2025) Multi-scale precipitation trends across Serbia: insights from ITA and MMK methods

Identification

• Journal: Theoretical and Applied Climatology
• Year: 2025
• Date: 2025-12-12
• Authors: Tatijana Stošić, Antônio Samuel Alves da Silva, Ivana Tošić, Borko Stošić, Vladimir Djurdjević
• DOI: 10.1007/s00704-025-05896-z

Research Groups

• Departamento de Estatística e Informática, Universidade Federal Rural de Pernambuco, Recife, PE, Brazil
• Faculty of Physics, Institute for Meteorology, University of Belgrade, Belgrade, Serbia

Short Summary

This study analyzed multi-scale precipitation trends in Serbia from 1961-2020 using Innovative Trend Analysis (ITA) and Modified Mann-Kendall (MMK) methods, finding ITA to be significantly more sensitive in detecting diverse temporal and spatial trends, particularly revealing more negative trends than MMK.

Objective

• To investigate multi-scale (annual, seasonal, vegetation period) precipitation trends across Serbia using the Innovative Trend Analysis (ITA) and Modified Mann-Kendall (MMK) methods, and to compare their sensitivity in identifying significant trends.

Study Configuration

• Spatial Scale: Serbia, covering the entire country with daily precipitation data from 14 meteorological stations.
• Temporal Scale: 60-year period from 1961 to 2020, analyzed on annual, seasonal (spring, summer, autumn, winter), and vegetation-period (April–October) scales.

Methodology and Data

• Models used:
  • Innovative Trend Analysis (ITA)
  • Modified Mann-Kendall (MMK) test
  • Sen's slope method (for trend magnitude)
  • Alexandersson test (for data homogeneity)
• Data sources: Daily precipitation data from 14 meteorological stations, obtained from the Serbian Hydrometeorological Service. Data underwent technical and critical quality control.

Main Results

• On an annual scale, both ITA and MMK identified significant positive precipitation trends, with magnitudes generally increasing from east to west across Serbia. ITA found significant trends at 9 stations, while MMK found them at 8.
• On a seasonal scale, both methods consistently showed positive trends in autumn across most stations (all 14 by ITA, 8 by MMK). However, ITA detected both positive and negative trends in spring (11 significant trends: 5 positive, 6 negative), summer (12 significant trends: 8 positive, 4 negative), and winter (13 significant trends: 9 positive, 4 negative), whereas MMK predominantly identified positive trends or fewer significant trends overall (e.g., 5 in spring, 3 in summer, 2 in winter).
• For the vegetation period (April–October), ITA identified significant positive trends at 6 stations, while MMK found positive trends at 4 stations and a negative trend at 1 station.
• Overall, ITA demonstrated higher sensitivity, identifying 65 statistically significant trends out of 84 total evaluations (14 stations x 6 temporal scales), compared to 31 significant trends identified by MMK.
• The strongest positive trends were consistently found in western Serbia, while the strongest negative trends were observed in the eastern region, particularly in spring, summer, and winter as revealed by ITA.
• Trend magnitudes were generally weak, typically below 2 mm/year for annual trends and less than 1 mm/year for seasonal trends.
• The correlation between ITA and MMK trend results was positive and statistically significant across all seasons, but its strength varied, being strongest in spring and weakest in summer, highlighting ITA's ability to capture more nuanced precipitation variability.

Contributions

• This is the first study to apply the Innovative Trend Analysis (ITA) method to seasonal and vegetation-period precipitation trends in Serbia, extending previous research that only used ITA for annual precipitation.
• Provides a comprehensive comparison of ITA and Modified Mann-Kendall (MMK) methods across multiple temporal scales, demonstrating ITA's superior sensitivity in detecting a wider range of precipitation trends (both positive and negative).
• Utilizes an extended and refined dataset of daily precipitation records up to 2020, offering updated insights into precipitation dynamics in Serbia.
• Offers crucial information for water resource management, agricultural planning, and disaster risk reduction strategies in Serbia by detailing multi-scale precipitation trends.

Funding

• Brazilian agency CNPq (grants No 309499/2022-4, 306590/2024-7, and 308782/2022-4)
• Ministry of Science, Technological Development and Innovation of the Republic of Serbia (grant No. 451-03-136/2025-03/200162)

Citation

@article{Stošić2025Multiscale,
  author = {Stošić, Tatijana and Silva, Antônio Samuel Alves da and Tošić, Ivana and Stošić, Borko and Djurdjević, Vladimir},
  title = {Multi-scale precipitation trends across Serbia: insights from ITA and MMK methods},
  journal = {Theoretical and Applied Climatology},
  year = {2025},
  doi = {10.1007/s00704-025-05896-z},
  url = {https://doi.org/10.1007/s00704-025-05896-z}
}