A Place-Based Approach to Drought Forecasting in South-Central Oklahoma

Area of Science:

• Environmental Science
• Climate Science
• Hydrology

Background:

• Water managers in south-central Oklahoma require reliable drought forecasts for future planning.
• Drought preparedness is crucial for regional water resource management.

Purpose of the Study:

• To produce reliable, place-based drought forecasts for south-central Oklahoma.
• To provide water managers with an early warning system for drought conditions.

Main Methods:

• Utilized multivariate regression with climate indices (Multivariate ENSO Index, Pacific Decadal Oscillation, Atlantic Multidecadal Oscillation) and Palmer Drought Severity Index (PDSI) as predictors.
• Developed a second-order model using a training dataset from 1901-2021, employing stepwise regression and cross-validation for predictor selection.
• Generated 1000 bootstrapped restricted models, averaging monthly predictions to forecast PDSI 18 months in advance.

Main Results:

• The developed model demonstrates sufficient skill for early drought warning.
• The model captures the timing of severe to exceptional drought periods, despite slight overestimation of maxima and minima.
• The adjusted R-squared value indicates an adequately accurate predictive model with good predictive ability.

Conclusions:

• The forecasting technique provides a valuable tool for water managers in Oklahoma.
• The model offers reliable, long-term drought intensity predictions, enhancing drought preparedness strategies.
• Place-based drought forecasting is essential for effective water resource management in drought-prone regions.