Resistance and resilience to droughts: hydropedological controls on catchment storage and run‐off response

Abstract

Hydropedological units are of critical importance in modulating catchment response in terms of storage and flux dynamics under changing hydrological conditions. We examined the short-term impacts of an extreme drought on the storage dynamics and runoff response in hydropedological units in a headwater catchment in the Scottish Highlands. These included poorly drained histosols in riparian zones and freely draining podzols on steeper hillslopes. To characterise the storage and runoff dynamics prior to, during, and after the drought period, precipitation, soil moisture, shallow ground water levels, and consequent runoff were monitored and stable water isotopes samples collected. Storage changes in the histosols were remarkably small (<40 mm), compared to those in moorland (~100 mm) and forest (~200mm) covered podzols. Although storage in all soils recovered soon after the drought, this took longest (3-4 months) for the forested podzols. During events, there was consistent threshold behaviour in most hydropedological units and the integrated response at the catchment scale, which was not affected by drying or wetting. The results suggest that during dry periods, large parts of the catchment were disconnected from the river network and runoff was generated mainly from the wet histosols. However, during events, there was an intermittent connection of the hillslopes that recharged the wetland and stream. This contributed to strong recovery and resilience of the catchment in its runoff response. Nevertheless, as future climate projections for northern environments suggest that prolonged dry periods are likely to become more frequent, further work is needed on the potential cumulative or carry over effects of consecutive drier periods.

Stream water age distributions controlled by storage dynamics and nonlinear hydrologic connectivity: Modeling with high-resolution isotope data

“…This is evident in terms of the disconnection of S up and its storage deficit (∼150 mm) that subsequently accumulates and is not replenished until the end of the year. It is also apparent in the prolonged, gradual drawdown of groundwater storage, although this probably partly reflects oversimplification in the model structure as some groundwater recharge in summer and early autumn was detected in some groundwater wells [ Geris et al ., ]. This simplification also likely helps explain the poor simulations of rewetting on flows and associated errors in the modeled stream isotope response.…”

“…Recent use of LiDAR in the catchment has shown restricted, more disconnected areas of saturation that are not captured by the simple lumped saturation algorithm used in the model and these are likely to contribute flows in the early rewetting phase as they connect and contribute to streamflow generation. In addition, there is evidence that aspect and land cover has an influence on evaporative losses and soil moisture deficits on the steeper catchment hillslopes with associated implications for groundwater recharge and slope‐wetland connectivity following drier periods [ Geris et al ., ]. Incorporating such processes would likely lead to improvements in flux estimates for the runoff model that will have indirect benefits for tracer modeling.…”

“…However, precipitation events following dry antecedent periods mainly generate runoff from the saturated area. At such times, precipitation on the drier hillslopes simply replenishes soil moisture deficits [ Tetzlaff et al ., ; Geris et al ., ].…”

“…Discharge measurements commenced in October 2007 and were derived from 15 min water level records (using an Odyssey capacitance probe) in a stable, rated section. Groundwater levels (Odyssey capacitance probes) and soil moisture (Campbell time domain reflectometers) measurements (15 min intervals) were also collected at the sites along the hillslope transect from June 2011 and collection continues to the present [ Geris et al ., ].…”

Using high resolution tracer data to constrain water storage, flux and age estimates in a spatially distributed rainfall‐runoff model

“…The spring in 2013 (January to April) was the coldest for over 50 years, whereas the summer 2013 (June to September) was the warmest and driest period for over 10 years. This was then followed by the wettest winter period on record (December 2013–January 2014) (Geris et al, ).…”