On the connection between terrestrial and riparian vegetation: the role of storage partitioning in water‐limited catchments

Abstract

Understanding how ecosystem functioning affects hydrological partitioning at the catchment scale is critically important to predict the annual water balance under climate-related land use change. Terrestrial ecosystems rely on rainfall infiltration while riparian ecosystems rely on the accumulation of surface and subsurface runoff in the riparian zone and the channel network. Some of the rainfall that infiltrates into the soils will be available for plants on the catchment's hillslopes. Some of that water may percolate down the soil profile where it can recharge a perched shallow aquifer or move even deeper to recharge bedrock aquifers. Lateral shallow subsurface flow provides a subsidy of water to plants downslope, which allows different plant species to occupy the toeslopes (Thompson et al., 2011a; Hwang et al., 2012). Perched and deep aquifers can sustain streamflow during dry periods and are thus essential for riparian and aquatic ecosystems. At the catchment scale, the persistence of flow can be analyzed by means of the flow duration curve (a plot that shows the percentage of time that discharge in a stream is likely to equal or exceed some specified value of interest).Climate has a first order control on the annual water balance (Budyko, 1974). The aridity index (ratio of average annual potential evapotranspiration, PET, to average annual rainfall, P) can predict the evaporative fraction and the runoff coefficient at climate time scales (~30 years) in many catchments. Less is known about how catchment ecosystems affect the annual water balance. Huxman et al. (2004) showed that when biomes undergo drought conditions their rainwater use efficiency (Net Primary Production/P) converges to a common maximum value. In a similar context, Troch et al. (2009) introduced the Horton index (the ratio of vaporization (i.e. evapotranspiration) to catchment wetting (the amount of water that infiltrates into the soils and does not runoff superficially) as a measure of rainwater use efficiency of the catchment's ecosystems, and showed that the Horton Index (HI) converges to a value of 1 when catchments undergo drought conditions. Subsequent research about the HI found that the index is controlled by climate (aridity index) and landscape characteristics (catchment slope and elevation) and is a useful first-order predictor for annual and interannual vegetation greenness at the catchment scale (Brooks et al., 2011; Voepel et al., 2011). 3The role of catchment water storage on vegetation response has been acknowledged in several recent ecohydrological studies (Tague, 2009; Miller et al., 2010; Thompson et al., 2011a). Evidence that water storage subsidizes plant water use through lateral hydrological connectivity (driven by topography) has been shown at the plot and the catchment scale and across different biomes (Scott et al., 2008; Thompson et al., 2011b; Hwang et al., 2012). Past research carried out in a temperate climate experimental watershed suggests a direct relationship between landscape position and storage that e...

Hydrologic functioning of the deep critical zone and contributions to streamflow in a high‐elevation catchment: Testing of multiple conceptual models

“…In the second step, assuming a linear reservoir behaviour for MGC, the reservoir constant (K res ; with units days) was estimated using a least‐square optimization approach. The assumption of linear reservoir behaviour for MGC is consistent with other studies that considered catchments of varying scales located in diverse climatic settings (e.g., van Dijk, , Arciniega‐Esparza et al, ; Peña‐Arancibia, van Dijk, Mulligan, & Bruijnzeel, ; Troch et al, ). van Dijk () suggests a linear reservoir assumption provides a better compromise between the filter's simplicity and its performance than does a nonlinear reservoir.…”

“…In the recursive filter equation (Equation ), α is known as the filter parameter. A value of 0.925 was used for α , as suggested by Nathan and McMahon () and used for numerous other studies (Arciniega‐Esparza, Breña‐Naranjo, & Troch, ; Troch et al, ; Voepel et al, ). In the standard procedure of applying the Lyne and Hollick's digital filter (Ladson et al, ), the daily streamflow time series for a given period of record was reflected 30 time steps before and after the period of record using the existing observations, and the filter was passed three times in the following order: forward, backward, and forward.…”

“…van Dijk () suggests a linear reservoir assumption provides a better compromise between the filter's simplicity and its performance than does a nonlinear reservoir. The baseflow at any time ( t ) can be related to the baseflow ( B 0 ) at some previous time, t 0 , using Equation (Arciniega‐Esparza et al, ; van Dijk, ) when there is negligible groundwater recharge. Differentiating Equation leads to Equation .…”

Snow to Precipitation Ratio Controls Catchment Storage and Summer Flows in Boreal Headwater Catchments

“…Assuming a linear reservoir model, that is, a linear storage‐discharge relationship ( β = 1), catchment water storage S C (mm) can be calculated as where K (=1/ α ) (day) is the recession constant (Brutsaert, ) and Q Max ¯ BF (mm/day) is the maximum baseflow (Arciniega‐Esparza et al, ). In order to determine the maximum baseflow, baseflow separation was performed using a digital filter approach (Lyne & Hollick, ), implemented in the R {hydrostats} package.…”

“…where K (=1/α) (day) is the recession constant (Brutsaert, 2008) and Q Max¯BF (mm/day) is the maximum baseflow (Arciniega-Esparza et al, 2017). In order to determine the maximum baseflow, baseflow separation was performed using a digital filter approach (Lyne & Hollick, 1979), implemented in the R {hydrostats} package.…”