Premium
Time‐lapse animation of hillslope groundwater dynamics details event‐based and seasonal bidirectional stream–groundwater gradients
Author(s) -
Zimmer Margaret A.,
McGlynn Brian L.
Publication year - 2017
Publication title -
hydrological processes
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.222
H-Index - 161
eISSN - 1099-1085
pISSN - 0885-6087
DOI - 10.1002/hyp.11124
Subject(s) - groundwater , hydrology (agriculture) , event (particle physics) , animation , environmental science , geology , computer science , computer graphics (images) , geotechnical engineering , physics , quantum mechanics
Stream water–groundwater (SW‐GW) interactions and hydrological connectivity are two important processes that drive much of the variability seen in streamflow characteristics. Measurement “snapshots” of hillslope flowpath connections (Bracken et al., 2013; Nippgen, McGlynn, & Emanuel, 2015) and SW‐GW gradients and associated biogeochemical patterns (Lautz & Fanelli, 2008; Vidon & Hill, 2004) are commonly used to categorize different characteristics or states of catchment and stream response. Studies that have fine resolution or long‐term datasets still often distill this information by displaying representative periods or events (e.g., Nippgen, McGlynn, & Emanual, 2015). Although these snapshot approaches provide useful information, they can oversimplify complex interactions or conceal the varying time scales (e.g., event responses nested within seasonal dynamics) of hillslope connectivity and SW‐GW interactions, shifts, and flow reversals. Because of these issues, catchment hydrologists have been encouraged to move beyond the snapshot approach (e.g., Bracken et al., 2013). Perennial streams in headwaters of humid climates are common foci for catchment hydrology. In these perennial reaches, the degree of hillslope connectivity and SW‐GW gradient magnitudes and directions can temporally shift on seasonal or event bases dependent on hydrological characteristics such as catchment storage and hydromorphology (Jencso et al., 2009; Nippgen et al., 2015; Rodhe & Seibert, 2011; Todd, 1955). Although these perennial streams are often viewed within a gaining stream system framework (Winter, Harvey, Franke, & Alley, 1999), little work has characterized temporal SW‐GW interactions and hillslope connectivity within the ephemeral and intermittent sections of these headwater catchments. This study sought to address a lack of understanding regarding the temporal dynamics of SW‐GW interactions and process‐based stream–hillslope connectivity within ephemeral and intermittent drainages. Our study site is located in the Piedmont region of North Carolina, USA, which is characterized as low relief with highly weathered, deep soils. Although this landscape type is underrepresented in hillslope connectivity studies (Bracken et al., 2013), it is a globally ubiquitous landscape type that is undergoing increasing development and land use pressure (Terando et al., 2014). This study uses the power of animated data visualization to highlight a novel application of groundwater levels in uncovering event‐ and seasonally‐driven (dis)connections between temporary streamflow and the complex groundwater system common in highly weathered landscapes. We monitored fine temporal resolution (5 min) streamflow and groundwater levels across an adjacent hillslope for 1 year within a 3.3‐ha ephemeral‐to‐intermittent drainage network in the Duke Forest, North Carolina, USA (36°2′3.4728′′N, 79°4′52.752′′W; see Zimmer & McGlynn, in review for more site details). Groundwater wells were hand augered to refusal depths at lower, mid‐, and upper hillslope positions along a representative convergent hillslope. At these sites, we also installed shallow wells to the depth of the most prominent confining soil horizon (A/Bt horizon or Bt/C horizon interfaces), confirmed by field‐saturated hydraulic conductivity measurements using a constant head permeameter (Amoozegar, 1989). Water levels were monitored using pressure transducers (+/ −0.1 mm resolution, Solinst) and capacitance recorders (+/−1 mm resolution, TruTrack, New Zealand). Because of the minimal range in observed stream level (0–0.18 m) relative to hillslope length (76 m), the streambed elevation was used as the stream elevation, regardless of discharge magnitude. Received: 20 September 2016 Accepted: 2 January 2017