Gorsevski (2026) Predicted Streamflow Sensitivity to Climate Change Using TOPMODEL with CLIGEN Weather Generator in a Data-Sparse Medium-Sized Mediterranean Watershed

Identification

• Journal: Water Resources Management
• Year: 2026
• Date: 2026-02-21
• Authors: Pece V. Gorsevski
• DOI: 10.1007/s11269-026-04540-3

Research Groups

School of Earth, Environment & Society, Bowling Green State University, Bowling Green, OH 43403, USA

Short Summary

This study develops a transferable framework to assess future streamflow sensitivity to climate change in data-sparse Mediterranean regions, confirming historical declines and predicting significant future reductions of 32% to 50% under 1.5 °C and 3.0 °C temperature increases, respectively.

Objective

• To assess historical trends (1961–1996) in observed streamflow changes in a medium-sized Mediterranean watershed.
• To calibrate and validate a hydrological model (TOPMODEL) using observed data (1990–1997).
• To simulate and assess a 30-year period of future streamflow projections under climate change scenarios (1.5 °C and 3.0 °C temperature increases).

Study Configuration

• Spatial Scale: Strumica River Watershed (SRW), southeastern North Macedonia, covering 1,649 km² with elevations ranging from 186 m to 1540 m.
• Temporal Scale: Historical analysis from 1961 to 1996; model calibration from January 1994 to January 1997; model validation from September 1990 to July 1993; future streamflow projections for a 30-year period.

Methodology and Data

• Models used:
  • TOPography-based hydrological model (TOPMODEL) for streamflow simulation.
  • CLImate GENerator (CLIGEN) for generating daily climate data.
  • Generalized likelihood uncertainty estimation (GLUE) for uncertainty assessment.
  • Mann-Kendall statistic and Theil-Sen median trend estimator for historical trend analysis.
  • Shuffled Complex Evolution of the University of Arizona (SCE-UA) for parameter optimization.
• Data sources:
  • Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) Global Digital Elevation Model (GDEM) v3 with 30 m spatial resolution.
  • Observed daily rainfall and streamflow data (1961–1997) from the Hydrometeorological Service (HMS) at Novo Selo gauging station (ST104) and a meteorological station near Strumica.
  • Potential evapotranspiration (ET) calculated using the Hargreaves-Samani method.
  • Global Surface Temperature Change data (GISTEMP) from NASA Goddard Institute for Space Studies (NASA-GISS) for temperature trend adjustments.

Main Results

• Historical streamflow (1961–1996) showed a significant declining trend, corroborated by Mann-Kendall (tau = -0.267, p-value = 0.023) and Theil-Sen median trend estimator.
• TOPMODEL calibration achieved a Nash-Sutcliffe efficiency (NSE) of 0.73 and underestimated observed streamflow by 16%, with uncertainty bounds from -52.7% to 35.4%.
• TOPMODEL validation achieved an NSE of 0.61 and overestimated observed streamflow by 12%, with uncertainty bounds from -46.5% to 18.4%.
• Future streamflow projections indicated a decline of 32% with a 1.5 °C temperature increase scenario and a 50% decline with a 3.0 °C temperature increase scenario.
• Seasonal analysis showed streamflow decline in spring, fall, and winter, but a projected increase in summer streamflow, likely due to altered precipitation patterns and increased evapotranspiration.
• Temperature-induced processes were found to have a greater influence on streamflow decrease than precipitation variations.

Contributions

• Developed a transferable framework for climate impact assessment in data-sparse regions, integrating TOPMODEL and CLIGEN with quantified uncertainty analysis.
• Provided a rigorous separation of temperature and precipitation effects on streamflow sensitivity.
• Employed a scenario-agnostic approach using uniform temperature and precipitation changes to investigate hydrological responses to warming signals, reducing reliance on specific climate model realizations.
• Contributed to multimodel ensemble studies by shedding light on hydrological responses in data-sparse areas.

Funding

Not applicable.

Citation

@article{Gorsevski2026Predicted,
  author = {Gorsevski, Pece V.},
  title = {Predicted Streamflow Sensitivity to Climate Change Using TOPMODEL with CLIGEN Weather Generator in a Data-Sparse Medium-Sized Mediterranean Watershed},
  journal = {Water Resources Management},
  year = {2026},
  doi = {10.1007/s11269-026-04540-3},
  url = {https://doi.org/10.1007/s11269-026-04540-3}
}