ABSTRACT

Shannon’s entropy theory measures the average uncertainty in the outcomes of an event. Since drought monitoring is an important issue, it is imperative to develop more user-friendly, modern methods for it. The present study investigates the uncertainty of the difference between monthly mean precipitation and potential evapotranspiration utilizing Shannon’s entropy over short-, mid-, and long-term dynamic time scales in the Karkheh Basin of Iran. An entropy-based precipitation–evapotranspiration index (EPEI) is defined for drought classification. EPEI is compared with the standardized precipitation–evapotranspiration index (SPEI) in terms of spatiotemporal patterns. The results indicate a higher similarity between EPEI and SPEI under existing conditions with low average precipitation. Over short scales, both indices yield similar results in duration, frequency, and intensity. Over long scales, the consistency of the drought duration, frequency, and intensity results declines. The main difference between the two is the EPEI’s ability to determine the early onset of the drought event.

摘要

香农的熵理论测量事件结果的平均不确定性。由于干旱监测是一个重要问题，因此必须为其开发更加用户友好的现代方法。本研究利用香农熵在伊朗 Karkheh 盆地的短期、中期和长期动态时间尺度上调查了月平均降水量和潜在蒸散量之间差异的不确定性。为干旱分类定义了基于熵的降水-蒸散指数（EPEI）。EPEI 在时空模式方面与标准化降水-蒸散指数 (SPEI) 进行了比较。结果表明，在平均降水量较低的现有条件下，EPEI 和 SPEI 之间的相似性更高。在短尺度上，这两个指数在持续时间、频率和强度方面产生了相似的结果。在长期范围内，干旱持续时间、频率和强度结果的一致性下降。两者之间的主要区别在于 EPEI 确定干旱事件早期发生的能力。