Untangling harvest‐streamflow responses in foothills conifer forests: Nexus of teleconnections, summer‐dominated precipitation, and storage

Abstract

This study re‐evaluated data from the historical Tri‐Creeks Experimental Watershed (1967–1988) in Alberta, Canada to address the initial question of forest harvest effects on streamflow and investigate the potential influence of teleconnections, summer‐dominated precipitation, and watershed storage on runoff generation. Tri‐Creeks has deep (up to 21 m) glacial deposits underlain by folded and faulted sedimentary bedrock with considerable potential for subsurface water storage. Timing of the conifer forest harvest experiment in two sub‐watersheds (>50% harvested) and one reference occurred near the 1976–77 Pacific Decadal Oscillation (PDO) phase change that led to less snowfall, but little difference in annual precipitation or runoff between phases after harvest. Established statistical and hydrological modelling methods that used regression techniques of observed and simulated streamflow to separately analyse sub‐watersheds did not detect change in average daily or annual runoff due to harvest. The interannual hydroclimatic variability influenced by the climate shift, attenuation of summer precipitation by the drier antecedent conditions in the warm period following harvest, and large potential for subsurface water storage contributed to shifts in runoff and uncertain detection of streamflow response. However, a hydrological modelling approach using calibrated parameters separately in the pre‐ and post‐harvest periods indicate significant change in rainfall‐generated peak runoff events and summer runoff following harvest, which was not detected in the reference watershed. Model calibration required less soil storage capacity in the treated watersheds in the post‐harvest period compared to the reference likely due to reduced transpiration that increased the likelihood of storm runoff during larger summer rainfall events. Within the context of streamflow responses to harvest in conifer dominated forest landscapes with seasonal snow cover, this study illustrates how complexity of climate variability and interaction with watershed storage and continental summer‐dominated precipitation may confound and mask the interpretation of harvest effects in paired‐watershed studies.

Runoff Threshold Responses in Continental Boreal Catchments: Nexus of Subhumid Climate, Low‐Relief, Surficial Geology, and Land Cover

“…Although runoff efficiencies are generally greater following snowmelt, in contrast to snow‐dominated regions, the overall contribution of snowmelt to annual R is low in the study catchments (Redding & Devito, 2010). Accumulation of snow water equivalent (CMw median = 120 mm) is low relative to potential soil and depression S (Redding & Devito, 2011), and there may also be considerable loss to groundwater recharge (Smerdon et al., 2008) resulting in minor snowmelt peaks in annual hydrographs in most years for BP catchments, as observed in summer P ‐dominated montane catchments (Goodbrand et al., 2022). Cyclonic summer patterns across the BP result in considerable range in magnitude, timing, and spacing of rain events, leading to dynamic antecedent moisture conditions that are synchronized with maximum rates of catchment ET (Devito et al., 2005; Mwale et al., 2011; Wells et al., 2017).…”

Flooding and the interannual variability of hydrologic quantities in the Missouri River basin

“…Because of the modification of the wind as it rises above the mountains, the descending wind on the lee side of the Rockies have high velocities, high temperatures, and high humidity deficits, which enables the wind to rapidly modify local , , and and consequently modify local (Equation ). A well‐known example of this orographic effect on wind is the Chinook wind that modifies water and heat energy fluxes in the valleys and plains on the lee side of the Rockies (Beran, 1967; Burrows, 1903; Goodbrand et al, 2022; MacDonald et al, 2018; Oard, 1993).…”

“…Another local source of horizontal temperature gradient that drive modification of T air (Equation 3) in the MRB is the local changes in T s and advected H owing to heterogenous land cover, such as snow cover, water surfaces, and irrigated surfaces (Harder et al, 2019;Neumann & Marsh, 1998;Philip, 1959;Schlögl et al, 2018). These land cover types account for a large proportion of the MRB land surface area during the year (Gleick, 1989;Mehta et al, 2013;Wise et al, 2018) and they undergo large thermal changes during the year, for example, snow accumulation/thaw, growing/fallow season, and frozen/unfrozen water surfaces.…”

Impact of land cover changes on Long‐Term Regional‐Scale groundwater recharge simulation in cold and humid climates

“…Others have found that land use and LC had little impact on GWR simulation, or that land use and LC impacts were smaller than those of climate change (Guerrero‐Morales et al, 2020; Mohan et al, 2018; Morgan et al, 2021; Shuler et al, 2021). These differences are most likely due to the fact that it can be challenging to distinguish the impacts of climate change from the impacts of land use and LC changes on GWR, and on all components of the water budget more generally (Assani et al, 2021; Goodbrand et al, 2022; Young et al, 2019). The study scale could also determine the importance of the impacts of land use and LC changes on GWR simulations, as local impacts of changes are not necessarily visible at larger scales (Shuler et al, 2021; Zomlot et al, 2017).…”

“…Similarly,Assani et al (2021) andGoodbrand et al (2022) identified average slope and soil water storage as countering the impact of agricultural development on forested land and wood harvesting on the river flow of six watersheds of different sizes in southern Quebec (Canada) and three headwater watersheds in Alberta (Canada). While an increase in fall and spring rainfall was the second reason identified byAssani et al (2021) for the increase in minimum flow during these seasons in southern Quebec,Goodbrand et al (2022) found that climate variability also reduced impacts of land-use changes on river flow. Similarly, moister future conditions expected in southern Quebec counter balanced the decrease in GWR rates simulated using HB LC due to afforestation taking place on lowpermeability surficial deposits, thus reducing the impact of LC changes on GWR (Figure6).…”