A review on baseflow separation methods

Abstract

Baseflow separation is essential for effective water management, drought assessment, and groundwater resources protection. Despite its importance, baseflow observations are often limited to small scale studies. To address this limitation, researchers have developed various baseflow separation methods. This paper reviews and analyzes existing studies which have developed or used the baseflow separation methods. A total of 43 studies are described, with a detailed review of 26 of them, focused on baseflow separation methods. Even if existing methods have already focused on baseflow separation, however, various methods produce divergent outcomes, primarily due to the inherent challenges indirectly observing the flow process associated with each technique. A minority of methods are anchored in physical science, particularly noticeable during waning stream flow periods. Notably, certain methods dynamically adjust baseflow estimates in response to precipitation intensity, an approach that, while intuitive, lacks a physical rationale and introduces subjectivity, especially when precipitation events conflate. Filter methods, despite their apparent rigor compared to graphical techniques, they suffer from a lack of physical underpinning regarding their operational frequency and orientation and are often constrained by arbitrary limits to avert baseflow estimates from surpassing total stream flow or descending into negative values. While the process based methodology enhances accuracy by employing physical principles to gauge baseflow across both arid intervals and rainy spells, the veracity of hydrological models is intimately tied to the data's availability and integrity. The main recommendations resulting from this review are that combining the strengths of different baseflow separation methods can lead to more robust results. For example, starting with a digital filter method for initial separation and refining it with physical based approaches. Leveraging advancements in computational power and algorithms can help in handling complex calculations and iterative processes more efficiently, leading to more accurate baseflow estimations.