Salau et al. (2025) A Novel Index for Integrative Drought Assessment in Agricultural Reservoirs

Identification

• Journal: Water Resources Management
• Year: 2025
• Date: 2025-12-29
• Authors: Rahmon Abiodun Salau, Bashir Adelodun, Mirza Junaid Ahmad, Qudus Adeyi, Adisa Hammed Akinsoji, Golden Odey, Kyung Sook CHOI
• DOI: 10.1007/s11269-025-04437-7

Research Groups

• Department of Agricultural Civil Engineering, Kyungpook National University, Daegu, Korea
• Arusha Climate and Environmental Research Centre, Aga Khan University, Arusha, Tanzania
• School of Resource and Environmental Management, Simon Fraser University, Burnaby, Canada
• Department of Agricultural and Biosystems Engineering, University of Ilorin, Ilorin, Nigeria
• Interdisciplinary Research Center for Membranes and Water Security, King Fahd University of Petroleum and Minerals, Dhahran, Saudi Arabia
• Institute of Agricultural Science & Technology, National University, Kyungpook, Daegu, Korea

Short Summary

This study developed a Multivariate Drought Index (MDI) using Principal Component Analysis (PCA) to integrate precipitation, inflow, and reservoir storage for comprehensive hydrometeorological drought assessment in 404 agricultural reservoirs across five major watersheds in South Korea, demonstrating its superior ability to detect earlier and longer droughts compared to existing indices like SPEI.

Objective

• To develop a Multivariate Drought Index (MDI) by blending hydrometeorological drought indices (Standardized Precipitation Index, Standardized Watershed Inflow Index, and Standardized Reservoir Storage Index) to enable comprehensive drought monitoring and prediction in agricultural reservoirs across South Korea.
• To evaluate the spatial and temporal variability of drought in agricultural reservoirs and identify reservoirs impacted by drought during the paddy rice season (April–September) from 1973 to 2017.

Study Configuration

• Spatial Scale: 404 agricultural reservoirs across five major watersheds (Hangang, Geumgang, Nakdonggang, Yeongsangang, and Seomjingang) in South Korea. Climate conditions were represented by data from 50 weather stations. Reservoir storage capacities ranged from 10,000 to 75,000 × 10^3 cubic meters, catchment areas from 800 to 36,550 × 10^3 square meters, and irrigated areas from 4,370 to 17,340 × 10^3 square meters.
• Temporal Scale: 1973–2017 (45 years), focusing on the paddy rice season (April–September) with drought indices calculated at a 3-month time scale using monthly data.

Methodology and Data

• Models used:
  • Principal Component Analysis (PCA) for developing the Multivariate Drought Index (MDI) and Standardized Drought Indices (SDI) in a two-stage aggregation process.
  • Hydrological tank model (used in prior work by Ahmad et al., 2022) for simulating watershed inflow and reservoir storage.
  • Gamma probability distribution for fitting precipitation, runoff, and water level data to calculate Standardized Precipitation Index (SPI), Standardized Watershed Inflow Index (SWII), and Standardized Reservoir Storage Index (SRSI).
• Data sources:
  • Daily rainfall data from the Korean Meteorological Administration (KMA).
  • Watershed inflow and reservoir storage data (simulated).
  • Standardized Precipitation Evapotranspiration Index (SPEI) downloaded from CSIC (spei.csic.es).
  • Historical drought records for validation from Hong et al. (2016).
  • Reference period for standardization of indices: 1973–2017.

Main Results

• The Multivariate Drought Index (MDI) successfully integrated precipitation, watershed inflow, and reservoir storage, with the Standardized Precipitation Index (SPI) being the most dominant contributor to the Standardized Drought Index (SDI) at the reservoir level (explaining 58% to 71% of variance).
• MDI detected earlier and longer drought durations compared to the Standardized Precipitation Evapotranspiration Index (SPEI); for example, in the Nakdonggang watershed, MDI indicated droughts lasting 81 months compared to 22 months for SPEI.
• MDI results showed strong alignment with historical drought events in South Korea (1973–2015), confirming its reliability.
• Frequency analysis revealed that approximately 69% of years experienced near-normal conditions based on SPI, 67% on SWII, 85% on SRSI, and 72% on SDI. Dry years had a higher incidence than wet years.
• June was consistently identified as the driest month during the paddy rice season (April–September), exhibiting severe to extreme drought conditions.
• In 2017 (identified as the driest year), 58% (236 out of 404) of reservoirs were affected by meteorological drought (SPI), while 42% (170 out of 404) and 43% (173 out of 404) were affected by hydrometeorological drought (SWII and SRSI), respectively.
• Hydrological droughts (SWII) exhibited longer cumulative durations (54,495 months across all reservoirs) compared to meteorological droughts (SPI, 52,850 months).

Contributions

• Development of a novel, comprehensive Multivariate Drought Index (MDI) specifically tailored for agricultural reservoirs by integrating key hydrological and operational variables (precipitation, watershed inflow, reservoir storage) using Principal Component Analysis (PCA).
• Addresses the limitations of single-variable drought indices by providing a more accurate and nuanced representation of drought duration, onset, and recovery, which is crucial for real-time reservoir operation and decision-making.
• Demonstrates superior performance of MDI over SPEI in detecting earlier and longer drought events, enhancing the effectiveness of drought management and mitigation strategies.
• Offers a scalable and adaptable methodology that can be applied to different geographical regions, temporal scales, and other water resource systems beyond agricultural reservoirs, by incorporating region-specific variables.

Funding

No funding was received.

Citation

@article{Salau2025Novel,
  author = {Salau, Rahmon Abiodun and Adelodun, Bashir and Ahmad, Mirza Junaid and Adeyi, Qudus and Akinsoji, Adisa Hammed and Odey, Golden and CHOI, Kyung Sook},
  title = {A Novel Index for Integrative Drought Assessment in Agricultural Reservoirs},
  journal = {Water Resources Management},
  year = {2025},
  doi = {10.1007/s11269-025-04437-7},
  url = {https://doi.org/10.1007/s11269-025-04437-7}
}