Spatial and Temporal Variation in the Three Main Hydrological States of Temporary Streams in a Swiss Pre-Alpine Catchment

Abstract

There are three main hydrological states for temporary (i.e., non-perennial) streams: dry streambed, standing water (pools), and flowing water. These states and the changes between them uniquely influence the physical, chemical and biological conditions and processes in temporary streams. Therefore, it is important to characterise temporary stream dynamics based on the three states. However, there is a lack of high spatiotemporal resolution data of the three states across stream networks. In this study, a network of 30 multi-sensor monitoring systems was used to acquire 5-min data of the three states across a small pre-Alpine headwater catchment during a 3-month monitoring period. The standing water state was most common in the upper part of the catchment, while the flowing water state occurred more frequently downstream. The dry streambed state was dominant in a fault zone between two types of Flysch bedrock. The spatial variation of the hydrological state permanence was correlated to topography, specifically the local Topographic Wetness Index, channel slope and upslope contributing area, except for monitoring locations in the fault zone. The wetting pattern during precipitation events was a bottom-up pattern outside the fault zone and a top-down pattern in the fault zone. The spatial variation in the amount of precipitation prior to a state change and the soil moisture storage at the time of a state change were related to topography as well. The temporal variation in these wetness thresholds for state changes was influenced by the antecedent soil moisture conditions and precipitation intensity. Our findings highlight the influence of topography, geology, channel morphology and event characteristics on the variation of the three main hydrological states of temporary streams. Moreover, this work highlights the value of monitoring all three states and high temporal resolution state data. Monitoring only wet and dry states or at a lower temporal resolution (e.g., weekly) would not have captured any state changes for many of the monitoring locations, and, therefore, would have severely underestimated the temporary stream dynamics.