Hydrologic response to changes in land use and variations of physical catchment characteristics
- 1Adigrat University
- 2Dire Dawa University
Abstract
This study evaluates land-use change and its modeled hydrological response in the study area, which covers 731.41 km2 of the Genfel watershed. The watershed ranges in altitude from 1764 to 3065 m a.m.l and its mean precipitation and temperature are approximately 597.9 mm and 18.8 °C, respectively. The Soil and Water Assessment Tool (SWAT) model was used to simulate watershed hydrological processes, while multi-temporal Landsat imagery from 1987, 2002, and 2017 was classified using supervised classification to assess land-use/land-cover changes. The SWAT model showed limited performance during calibration and validation, particularly at the monthly time step. For daily calibration, the coefficient of determination (R2) and Nash–Sutcliffe (ENS) were 0.5111 and 0.16, respectively, whereas monthly calibration produced R2 and NSE values of 0.6428 and −0.23, respectively. Daily validation yielded R2 and NSE values of 0.5147 and 0.16, respectively, while monthly validation yielded R2 and NSE values of 0.6133 and −0.05, respectively. These results indicate substantial uncertainty in the simulated hydrological responses and therefore require cautious interpretation. Land-use change showed a substantial increase in cultivation from 36.525% in 1987 to 48.383% in 2002 and 54.01% in 2017, while forest cover decreased from 7.624% in 1987 to 1.665% in 2002 and 0.237% in 2017. Given the SWAT model’s limited performance, the simulated results are interpreted as model-based scenario responses rather than definitive evidence of historical causation. The study’s novelty lies in integrating multi-temporal remote sensing, GIS-based watershed characterization, SWAT water balance simulation, and sub-watershed morphometric analysis in a data-scarce watershed. The results provide a basis for targeted watershed monitoring, soil and water conservation, and further field-based investigation.
Related
- Catchment hydrologic function is widely time-variant
- Variations in Arctic aerosol iron solubility in relation to leaching methodology, air mass characteristics, and seasonality
- Characteristics of seasonal snow cover variations and their effects on soil moisture and vegetation dynamics in Xinjiang
- Non-dendritic hydrologic networks: foundational patterns, algorithms, and a global empirical demonstration
- AI Agent for Hydrologic Modeling: Definition, Development, and Application
BibTeX
@article{zewde2026hydrologic,
title = {{Hydrologic response to changes in land use and variations of physical catchment characteristics}},
author = {Iyasu Zewde and Hayelom Belay and Awet Arefe and Kollege Girmay},
journal = {Water Science},
year = {2026},
month = oct,
doi = {10.1007/s44533-026-00082-0},
url = {https://doi.org/10.1007/s44533-026-00082-0},
note = {CC BY},
}
RIS
TY - JOUR
TI - Hydrologic response to changes in land use and variations of physical catchment characteristics
AU - Iyasu Zewde
AU - Hayelom Belay
AU - Awet Arefe
AU - Kollege Girmay
PY - 2026
DA - 2026/10/25
JO - Water Science
DO - 10.1007/s44533-026-00082-0
UR - https://doi.org/10.1007/s44533-026-00082-0
AB - This study evaluates land-use change and its modeled hydrological response in the study area, which covers 731.41 km2 of the Genfel watershed. The watershed ranges in altitude from 1764 to 3065 m a.m.l and its mean precipitation and temperature are approximately 597.9 mm and 18.8 °C, respectively. The Soil and Water Assessment Tool (SWAT) model was used to simulate watershed hydrological processes, while multi-temporal Landsat imagery from 1987, 2002, and 2017 was classified using supervised classification to assess land-use/land-cover changes. The SWAT model showed limited performance during calibration and validation, particularly at the monthly time step. For daily calibration, the coefficient of determination (R2) and Nash–Sutcliffe (ENS) were 0.5111 and 0.16, respectively, whereas monthly calibration produced R2 and NSE values of 0.6428 and −0.23, respectively. Daily validation yielded R2 and NSE values of 0.5147 and 0.16, respectively, while monthly validation yielded R2 and NSE values of 0.6133 and −0.05, respectively. These results indicate substantial uncertainty in the simulated hydrological responses and therefore require cautious interpretation. Land-use change showed a substantial increase in cultivation from 36.525% in 1987 to 48.383% in 2002 and 54.01% in 2017, while forest cover decreased from 7.624% in 1987 to 1.665% in 2002 and 0.237% in 2017. Given the SWAT model’s limited performance, the simulated results are interpreted as model-based scenario responses rather than definitive evidence of historical causation. The study’s novelty lies in integrating multi-temporal remote sensing, GIS-based watershed characterization, SWAT water balance simulation, and sub-watershed morphometric analysis in a data-scarce watershed. The results provide a basis for targeted watershed monitoring, soil and water conservation, and further field-based investigation.
SN - 2357-0008
SN - 1110-4929
ER -