Translator Disclaimer
1 January 2017 Ecosystem Water Availability in Juniper Versus Sagebrush Snow-Dominated Rangelands
Author Affiliations +
Abstract

Western Juniper (Juniperus occidentalis Hook.) has greatly expanded in the past 150 years and now dominates over 3.6 million ha of rangeland in the Intermountain Western United States. The impacts of juniper encroachment on critical ecohydrological relationships among snowdistribution, water budgets, plant community transitions, and habitat requirements for wildlife, such as the greater sage grouse (Centrocercus urophasianus), remain poorly understood. The goal of this study is to better understand how juniper encroachment affects water availability for ecohydrologic processes and associatedwildlife habitat in snow-dominated sagebrush (Artemisia spp.) steppe ecosystems. A 6-yr combined measurement and modeling study is conducted to explore differences in snow distribution, water availability, and annual water balances between juniper-dominated and sagebrushdominated catchments. Although there is large interannual variability in both measured weather data and modeled hydrologic fluxes during the study, results indicate that juniper-dominated catchments have greater peak accumulations of snow water equivalent, earlier snow melt, and less streamflow relative to sagebrushdominated catchments. Water delivery is delayed by an average of 9 days in the sagebrush-dominated scenario comparedwith the juniper-dominated scenario as a result of increasedwater storage in snow drifts. The delayed water input to sagebrush-dominated ecosystems in typical water years has wide-ranging implications for available surface water, soil water, and vegetation dynamics associated with wildlife habitat for sagebrush obligates such as sage grouse. Results from this study imply that the retention of high-elevation, sagebrush-dominated landscapes may become crucial for sage grouse habitat management if mid- and low-elevation precipitation continues to transition from snow to rain dominated.

Introduction

Western Juniper (Juniperus occidentalis Hook.) now dominates over 3.6 million ha of rangeland in the Intermountain Western United States (Miller and Tausch, 2001), much of which represents expansion from presettlement habitat (Tausch et al., 1981; Johnson and Miller, 2006; Miller et al., 2008). Juniper (Juniperus spp.) expansion into sagebrush (Artemisia spp.) ecosystems can influence the vegetation community (Bates et al., 2000;Miller and Tausch, 2001;Miller et al., 2005) and the hydrology and soil resources of an area (Pierson et al., 2007, 2010), which all affect wildlife habitat. Disentangling the ecological ramifications of the transition from sagebrush-dominated systems to juniperdominated systems requires a holistic understanding of key interacting ecohydrologic feedbacks (Romme et al., 2009; Williams et al., 2016). These ecohydrologic transformations reduce habitat of key sagebrush indicator species, such as the greater sage grouse (Centrocercus urophasianus) (Connelly et al., 2011; Knick and Connelly, 2011). The effects of the transition from sagebrush-dominated to woodlanddominated landscapes on wildlife habitat have been well established in the literature (Connelly et al., 2011), and critical ecological relationships between vegetation and hydrologic processes have emerged (Petersen et al., 2009; Pierson et al., 2010, 2013; Williams et al., 2014, 2016). However, a coherent understanding of feedback among vegetation, hydrology, and habitat remains tenuous (Blomberg et al., 2012; Guttery et al., 2013; Donnelly et al., 2016).

Sage grouse require a mosaic of mature shrubs and tall grasses for cover and a balanced composition of herbaceous species and sagebrush to provide forage during their annual life cycle stages (Connelly et al., 2000, 2011). Plant community composition, structure, and productivity are all strongly coupledwith the timing and amount of ecosystemwater availability (Flerchinger et al., 1998, 2010; Rodriguez-Iturbe, 2000; Ryel et al., 2010; Chauvin et al., 2011; Roundy et al., 2014b). Sagebrush and herbaceous productivity is strongly tied to soil water recharge deep in the soil profile (Cline et al., 1977; Sturges, 1977; Richards and Caldwell, 1987; Caldwell and Richards, 1989; Ryel et al., 2004; Schlaepfer et al., 2012). Sage grouse also require seasonal high-elevation wet meadows, riparian areas, springs, and the associated forbs for late brood rearing in late summer/early autumn seasons (Connelly et al., 2011; Hammersmark et al., 2008; Loheide et al., 2009; Loheide and Gorelick, 2007). The sustainability of wet meadows, riparian areas, and springs is also intimately linked to the timing and magnitude of water delivery to the landscape (Whiting and Godsey, 2016).

Juniper impacts sage-grouse directly by changing the vegetation structure and creating overhead perches for predators (Casazza et al., 2011) butmight also have indirect effects via altering the hydrologic behavior of an area (Blomberg et al., 2012; Guttery et al., 2013; Donnelly et al., 2016). There is a common perception that juniper encroachment can lead to a loss of surface water, but there is little science to quantify these changes. The ecohydrologic impacts of tree encroachment on soil water dynamics have been described and quantified (Miller et al., 2005; Mollnau et al., 2014; Roundy et al., 2014b). Junipers have competitive advantages over intercanopy herbaceous species in terms of soil water extraction, including extensive root systems capable of soil water redistribution (Breshears et al., 1997, 1998; Miller et al., 2000, 2005; Newman et al., 2010). Most of the woodland encroachment research dealing with impacts on soil water dynamics have focused on changes following juniper removal (Bates et al., 2000; Young et al., 2013; Mollnau et al., 2014; Roundy et al., 2014b), whichmay be influenced by the short-termflush in herbaceous vegetation immediately following tree removal (Miller et al., 2014; Roundy et al., 2014a).

The links between ecosystem water availability and the redistribution of snowfall are critical but understudied components in the literature regarding ecohydrologic impacts of juniper encroachment. Plant community transitions from sagebrush to juniper woodlands in the Intermountain West commonly occur on siteswith snow-dominated precipitation regimes. Wind and topography in these uplands interact to redistribute falling snow, while vegetation reduces wind velocities (Burke, 1989; Marks and Winstral, 2001; Marks et al., 2001, 2002; Winstral and Marks, 2002). The reduction of wind speeds caused by trees both promotes the deposition of snow and decreases turbulent fluxes that melt snow.

Vegetation-snow feedbacks produce consistent ecosystem characteristics such as growing season length, soil temperatures, species composition, and primary production (Hiemstra et al., 2006). Deeper snow accumulation commonly benefits plant productivity by providing greater insulation for, and prolongedwater delivery to, the soil profile (Sturm et al., 2001; Liston et al., 2002).

For snow-dominated semiarid uplands, knowing how and where snow accumulates and melts is crucial in the understanding the distribution of catchment water inputs, available soil water, and the timing and amount of streamflow (Flerchinger et al., 1998; Flerchinger and Cooley, 2000; Seyfried et al., 2009; Williams et al., 2009). The distribution of snow on the landscape may also be the best predictor of spatial and temporal patterns in soil moisture in semiarid snow-dominated sagebrush catchments (Williams et al., 2009). Both the vegetationaffected distribution of snow and the snowmelt-driven water delivery control catchment streamflow initiation, volume, and recession (Flerchinger and Cooley, 2000; Luce et al., 1998; Luce and Tarboton, 2004; Seyfried et al., 2009;Williams et al., 2009). Snow distribution differences between sagebrush- and woodland-dominated landscapes are therefore key links for understanding differences in water availability for plant communities and wildlife.

The goal of this study is to better understand how juniper encroachment affectswater availability for ecohydrologic processes and associated wildlife habitat in snow-dominated sagebrush steppe systems. Our specific research questions are: 1)What are the differences in snowdistribution and water delivery to the landscape between juniperdominated and sagebrush-dominated catchments, 2) What is the effect of those differences on the catchment water balance and streamflow, and 3) What are the implications of tree-induced changes on water availability and those effects on sagebrush steppe ecosystemdynamics? To address these research questions we performed a combined measurement and modeling study to quantify snow distribution, water delivery to the landscape, and streamflow for four rangeland watersheds currently in the late stages of juniper encroachment (juniper-dominated) and for the samewatershedswith amosaic of sagebrush and herbaceous vegetation (sagebrush-dominated).

Study Site and Measured Data

The study area consists of four catchments referred to herein as the South Mountain Experimental Catchments. The catchments are located (-116.90®W,42.67®N) on South Mountain in the Owyhee Mountains just east of the Idaho-Oregon border (Fig. 1). Precipitation at the study area is snow dominated (Fig. 2). The study spans 6 water yrs (WY) (1 October to 30 September) from WY 2008 to WY 2013, which provide a range ofweather conditions typical for this region. The 6-yrwater year average precipitation is 620 mm, with WY 2011 being the wettest year with 867 mm and WY2012 being the driest year with 445 mm (Table 1). Average annual air temperature for the study area is 7.0°C (Table 2). The prevailing wind direction during precipitation is from the west (274°). The drainage areas and relief for the four catchments range from 20 to 70 ha and 1665 to 1898 m, respectively (Table 3). Mean hillslope gradients across the four catchments range from 18% to 23%. The study area spans multiple ecological sites, each with sagebrush as the dominated shrub component for the reference community. Currently, vegetation in each of the catchments is primarily juniperdominated, with significant canopy closure. Soils are classified mostly as well-drained gravelly loams, which is supported by observations (NRCS, 2016).

Four drop box v-notch weirs (Bonta and Pierson, 2003) and six weather stations were installed at the South Mountain Experimental Catchments from 2006 to 2008 with the purpose of studying the ecohydrologic response of juniper-dominated sagebrush steppe systems to juniper removal. Each weather station records precipitation mass, wind speed and direction, air temperature, relative humidity, and incoming solar radiation at an hourly interval. Snow depth is also recorded every 15 minutes at each station using ultrasonic sensors. Manual measurements of snow water equivalent (SWE) were made two to three times each year at snow courses near the six weather stations. In addition, mean densities from four intensive snow surveys were used with automated snow depth values to obtain additional SWE model validation values.

A Lidar dataset acquired in November 2007 before snowfall provides an accurate 1 m2 snapshot of bare earth elevation and vegetation structure for each of the study catchments (Sankey et al., 2013). The Lidar point densitywas 7 points per squaremeter resulting in a vertical accuracy of approximately 3 cm. All Lidar processing was done using tools developed by the Boise Center Aerospace Laboratory (BCAL, 2016). In addition to Lidar-derived vegetation heights, a manual vegetation survey at three randomly located 30 × 30 m plots in each of the four study catchmentswas conducted in August 2008. Heights of all junipers >1 m within each plotwere measured, and the heights of all shrubs and junipers <1 m were recorded along five transects within the plots (Fig. 3).

Figure 1.

Locationmap of the South Mountain Experimental Watersheds showing the location of the four weirs that define catchments E, F, G, and Mand the sixweather stations (E2, F1, G1, G2, M1, and M2). Close views of each weather station include model pixels classified as juniper dominated and the location from which model results were extracted to compare to measurements.

f01_116.jpg

Methods

Distributed snowpack accumulation and melt are modeled for the South Mountain Experimental Catchments with the present latesuccession juniper conditions (juniper-dominated scenario) and for conditions with amosaic of sagebrush vegetation (sagebrush-dominated scenario). Water availability for each study catchment under both conditions is assessed by a combination of time series modeling (water delivery to the soil surface) and the derivation of annual water budgets through a water balance approach.

Figure 2.

Water year precipitation measured at the six weather stations showing the proportion of rain, snow, and mixed events. The dashed area of the rain bar depicts the amount of spring rains (1 March to 30 May) received that year. The mean water year precipitation total for the 6 years of this study was 620 mm.

f02_116.jpg

Table 1

Measured (meas) andmodeled (mod) water year water precipitation (precip) and water balance terms (snowwater input [SWI], streamflow [Q], evapotranspiration [ET]) for the South Mountain Experimental Catchments (SM). Theweighted P value (p-val) frompaired t-tests is included as a performancemeasure of the ETmodeling (Flerchinger et al. in press), indicating that none of the simulationswere significantly different frommeasured values at the α=0.05 level. Precipitation is also shown for Upper Sheep Creek (USC) subcatchment of the nearby Reynolds Creek Experimental Watershed. Juniper-dominated ET is calculated as the residual (resid) of the water balance.

t01_116.gif

Snow Modeling

The iSnobal model (Marks et al., 1999a,b) uses the catchment topography and distributed meteorological forcings to estimate the snow accumulation and melt. iSnobal is a physically based mass and energy model that has been extensively applied to investigate snow physics, processes, and the distributedmelt patterns in similar areas in the Owyhee Mountains (Marks and Winstral, 2001; Marks et al., 2001, 2002; Winstral and Marks, 2002, 2014; Seyfried et al., 2009; Nayak et al., 2011; Kumar et al., 2013; Reba et al., 2014; Rasouli et al., 2015) and forested areas (Marks et al., 1998; Link and Marks, 1999; Garen and Marks, 2005). The model produces estimates of SWE, snow melt, and surface water input (SWI),which is the combined liquidwater drainage leaving the bottomof the snow pack and rain on the ground surface. Input data required by the model include precipitation and available energy. The use of a complex, physically based snow model is necessary to be able to sufficiently parameterize vegetation differences between juniperand sagebrush-dominated catchments. The energy balance of the snowpack at each model pixel is expressed as:

e01_116.gif
where ΔH is change in snowpack energy; and Rnet ,H,LvE,G, and M are net radiative, sensible, latent, conductive, and advective (from precipitation) energy fluxes, respectively (Marks et al., 1992); and Δt is the time step. The model requires spatially distributed net solar, incoming thermal, air temperature, vapor pressure, wind speed, soil temperature, and precipitation. iSnobal represents the snowpack as a two-layer system, with a fixed-thickness surface layer, and a variable thickness lower layer representing the remainder of the snow cover. The model computes outgoing thermal radiation to get Rnet and solves for the rest of the energy balance. If ΔH is negative, the snow will cool, increasing its cold content, or the amount of energy required to bring the snow to 0°C. If ΔH is positive, the snow will warm, reducing its cold content. Once the snow is at 0°C, the cold content is zero and any addition of energy will result in melt. If the addition of liquid water to the snow by either melt, or rain exceeds 1% (Davis et al., 1985), the excess is released to the soil as SWI. If the ground is not snow covered, the model passes precipitation that falls as rain to the soil as SWI.

Table 2

Water year precipitation-weighted wind speed and direction and water year average air temperature.

t02_116.gif

We apply iSnobal at an hourly time step on a 10-m DEM created by resampling the 1-m Lidar-derived bare earth elevation model. This resolution optimizes our ability to capture vegetation and topographic differences while maintaining sufficient computational efficiency. Mean daily SWE and SWI are derived from respective model outputs for the four study catchments and compiled in a time series dataset. The total SWI for the WY is used in conjunction with measured data to develop annual water balance estimates as described in the water balance subsection.

Juniper effects on snow accumulation and melt are accounted for in iSnobal by classifyingmodel pixels as being juniper-dominated, junipersheltered, forest opening, or open. The lidar-derived 1m2 mean vegetation height map was classified as juniper or no juniper using a 1.5-m threshold based on the manual vegetation height surveys (see Fig. 3) (Sankey et al., 2013). This threshold was chosen to minimize the classification of any other vegetation as juniper. Model pixels (10 × 10 m, a 100 m2 area) were classified as juniper dominated if at least one-third (33) of the 1-m2 Lidar cellswere classified as junipers. This conservative threshold is chosen on the basis of a comparison between collocated NAIP imagery and the 10-m pixel classification. Higher thresholds excluded many pixels with large junipers, while lower thresholds overestimated the juniper coverage of the basin. Maximum canopy height is calculated as the mean of the 1-m Lidar-derived maximum vegetation heights classified as juniper. Model pixels are classified as forest openings if at least five of the eight surrounding pixels are classified as juniper-dominated (Winstral and Marks, 2002). Junipersheltered pixels are defined for each of the cardinal wind directions and are on the lee side of a pixel classified as juniper dominated.

Several variables used to run the snow model are affected by junipers, including snow accumulation, surface wind speeds, net solar radiation, and incoming thermal radiation. These effects and the parameterization of model-forcing data to account for them are described later in the associated subsections.

Table 3

Watershed areas, the percent of the pixels classified as juniper-dominated, elevation ranges, and mean slopes. Mean catchment elevations are in parentheses.

t03_116.gif

Figure 3.

Histogram of measured vegetation heights of junipers and shrubs. The vertical black line shows the 1.5 m cutoff used in this study to classify junipers.

f03_116.jpg

Solar Radiation

Clear sky incoming solar radiation and snow albedo are calculated using the Image Processing Workbench (IPW) software package utility stoporad (Frew, 1990;Marks et al., 1999b). Stoporad accounts for the interaction ofmeasured topography, calculated sun angle, and empirically derived snow grain size based on the time from last snowfall (Dozier, 1980; Dozier and Frew, 1981; Dubayah, 1994. Incoming solar radiation and albedo are split into visible (0.28–0.70 µm) and near infrared (0.70–2.8 µm)wavelengths to bettermodel the effects of albedo differences in those wavelengths (Marshall and Warren, 1987). Both visible and near infrared incoming solar radiation are further split into beam and diffuse portions to better account for vegetation shading affects. Incoming solar radiation is corrected for cloud cover using the clear sky fraction, or the ratio ofmeasured incoming solar radiation and modeled clear sky radiation at weather stations E2, G2, and M2 (see Fig. 1). Surrounding vegetation affects incoming solar measurements at the other weather stations. Cloud-corrected incoming solar radiation is used for the sagebrushdominated scenario. Juniper canopy effects are then accounted for in pixels classified as juniperdominated using the maximum vegetation height for each model pixel, an optical transmissivity of 0.16 and an extinction coefficient of 0.074 m-1 equivalent to dense conifer (Link and Marks, 1999). In addition, late-season albedo values are decreased to account for juniper litter accumulation on the snow surface (Marks and Winstral, 2001;Winstral and Marks, 2002; Kormos et al., 2014).

Thermal Radiation

Incoming clear sky thermal radiation is calculated using elevation and humidity data and corrected for surrounding topography (Marks and Dozier, 1979). Those values are corrected for cloud cover using a measured relationship between the clear sky fraction from the nearby Reynolds Mountain East subcatchment, 45.5 km from the study site within the Reynolds Creek Experimental Watershed (RCEW). This relationship is then applied to the South Mountain Experimental Catchments using the locally calculated clear sky fraction (Reba et al., 2011). Cloud corrected incoming thermal radiation is used for the sagebrush-dominated scenario. Juniper canopy effects are represented by assigning a canopy transmissivity of 1 to open areas, 0.16 to pixels classified as juniper, and 0.75 to forest openings (Link and Marks, 1999; Winstral and Marks, 2002). Incoming thermal radiation from the canopy is calculated using the Stefan Boltzman equationwith a canopy emissivity of 0.96 assuming that the measured air temperaturewas equivalent to the canopy temperature.

Precipitation

Measured hourly precipitation at the six weather stations is distributed by detrended kriging if it is classified as rain or mixed phase precipitation events based on dew point temperature (Garen et al., 1994; Garen, 1995; Marks et al., 2013). The distribution of snow events account for wind redistribution as a function of wind speed, wind direction, topography, and vegetation (Winstral et al., 2013).

The degree of topographic or canopy sheltering or exposure, with subsequent scour or deposition and drifting, was achieved using methods developed by Winstral and Marks (2002) that have been successfully applied across many mountain environments in North America (Winstral et al., 2009, 2013; Winstral and Marks, 2014). As discussed in the cited papers, the wind field is derived from the maximum upwind slope parameter. The wind flow separation, or the wind behavior that promotes drifting to occur, is described by the slope break parameter. These parameters are calculated fromthe 10-m digital elevationmodel using the procedures described by Winstral and Marks (2002). Accumulation ratios are calculated as a function of the exposure of each model pixel to define drift and scour zones (Winstral et al., 2013). Accumulation ratios for pixels located in areas defined as drift zones are calculated as a function of measured wind speed (Winstral and Marks, 2014). A minimum accumulation ratio of 1.1 and a maximumof 4.2 is imposed on pixelswithin drift zones. Accumulation ratios in model pixels outside the drift zones vary between 0.0 and 1.0.

Juniper effects on precipitation distribution are accounted for by modifying the accumulation ratios in pixels defined as juniper dominated or juniper sheltered. Juniper-dominated and juniper-sheltered pixels are assigned an accumulation ratio of 1.0 (Winstral and Marks, 2002). Pixels in drift zones are assigned the drift accumulation ratio regardless of vegetation classification. The precipitation distribution is then obtained bymultiplying the distributed accumulation ratio bywind-corrected, measured precipitation at station E2. We chose E2 on the basis of the fact that vegetation influences are minimal and it measured the most precipitation most often (23% of the time).

Wind

Measured wind speeds are distributed across South Mountain for each time step as a function ofmeasuredwind direction and vegetation and topographic sheltering or exposure to get the juniper model wind field (Winstral et al., 2009, 2013). Wind distributions for the sagebrush-dominated scenario are obtained by increasing measured wind speeds using a relationship between wind speeds measured at a semivegetation sheltered site and from a fully wind-exposed ridge site at the nearby RCEW (Fig. 4). The RCEW semisheltered site is similar to weather stations E2, M2, and G2 in that it is on a ridge with a few large trees close to it. We expect this increase in wind speed as a result of a reduction of the near surface roughness for a modeling domain without junipers. The increasedwind speeds are then distributed across South Mountain for each hour as a function of topographic sheltering and exposure.

Water Balance

We used a simple annual water balance to demonstrate the differences in water delivery to juniper- and sagebrush-dominated catchments:

e02_116.gif
where SWI is the modeled surface water input, which includes snowmelt, rain draining through the snow, and rain on snow-free surfaces. ET is evapotranspiration, and Q is stream discharge. SWI values account for evaporation from the snow surface.

ET is estimated for juniper-dominated scenarios as the difference betweenmodeled SWI and measured Q. A range of ET values for the sagebrush model scenario is obtained from a well-validated measurement (eddy covariance and soil moisture) and modeling exercise from low sagebrush (A. arbuscula Nutt. subsp. arbuscula) and mountain big sagebrush (A. tridentata Nutt. subsp. vaseyana [Rydb.] Beetle) vegetation communities in the Upper Sheep Creek subwatershed (USC) of RCEW (Flerchinger, 2016). These ET values are available for the same water years and similar elevation (1837–2023 m), weather, and precipitation conditions as in this study (see Table 1). Simulated weekly ET from mountain big sagebrush was not significantly different from eddy covariance system measurements for any of the six WYs (mean P value 0.546, α = 0.05). Simulated weekly ET from low sagebrush was not significantly different from soil moisture mass balance calculations of ET for 4 of the 6WYs, with WY 2010 and WY 2011 showing significant differences (mean P value 0.396). For simplicity,we present the results of the mass balance portion of this study assuming a vegetation distribution of 20%mountain big sagebrush and 80% lowsagebrush. This distribution is based on 1) mountain big sagebrush occurs where drifts occur; 2) 10% of the catchment is classified as drift zones using a wind direction of 274°, the mean direction during storms; and 3) increases in water inputs from that 10% drift area may support an additional 10% of the catchment to develop mountain big sagebrush. The other 80% of the catchment is assigned ET values of low sagebrush communities that require lesswater. This exercise is supported by observations of vegetation structure and drift locations at USC (Flerchinger and Cooley, 2000).

Figure 4.

Relationship between two wind speed (WS) measurement sites at Reynolds Creek Experimental Watershed used to account for increased wind speeds in the sagebrush-dominated scenario. WS Exposed is a wind measurement site on an exposed ridge with no surrounding vegetation. WSTree is awindmeasurement site on a ridge and partially sheltered by trees. Contour lines depict point density based on a 0.5 ms-1 square grid.

f04_116.jpg

Q for the juniper-dominated scenario is measured at the four weirs. Q for the sagebrush-dominated scenario is calculated as the difference between sagebrush-dominated modeled SWI and ET values from USC with the assumed vegetation distribution. Runoff ratios are calculated as the ratio of measured or calculated Q to modeled total basin SWI from juniper-dominated and the sagebrush-dominated scenarios.

Fortuitously, interception-induced losses to evaporation in this region are not a large component of the annual water balance and are not considered for this exercise. Losses from the interception of rain are assumed negligible due to the timing difference between high evaporative demand in the summer and high precipitation in the winter. Though both trees and shrubs intercept snow, the combination of moderate temperatures and generally sunny and/or windy conditions following storms leads to intercepted snow sloughing off and becoming either melt water drip, or accumulations of wet snow at the base of the canopy. Additional discussion and an evaluation of the implications of this assumption are addressed in the discussion section.

The melt-out date of the four catchments is the first day of the WY that the modeled basin averaged SWE is zero after peak accumulation. The timing of water delivery to the four study catchments is evaluated using the WY day when 75% of themodeled SWI enters the catchments. We chose 75% to highlight late-season differences in water availability related to snow distribution. Higher percentages (later differences) are more spring rain dependent, which have identical distributions for the juniper- and sagebrush-dominated scenarios. The difference between thismetric for juniper- and sagebrush-dominated scenarios provides a measure of how much longer water is available for a sagebrushdominated catchment than a juniper-dominated catchment.

Results

Snow Accumulation and Melt Patterns

The modeled accumulation and melt of snow for the juniperdominated scenario reasonably matches SWE measurements near the six weather stations (see Figs. 1 and 5). The snow model over predicts SWE near weather station F1, which has the most juniper cover from WY 2008 to WY 2011 (see Figs. 1 and 5, Table 3). These years are characterized by normal or above normal precipitation (see Fig. 2, Table 1). Modeled SWE nearweather stations E2 and M2, which have sparser juniper cover, are underestimated for all years. Although the data available to assess model performance are limited, results indicate that the snowsystemis representedwell, and that they can help us to better understand howjuniper encroachment affects snowdeposition andwater availability in these systems.

Basin average SWE from the juniper-dominated scenario has a greater accumulation and melts out an average of 18 days earlier than the sagebrush-dominated scenario (Fig. 6). The greater SWE accumulation for the juniper-dominated scenario is a result of modeled differences in precipitation distributions (see Table 1), where pixels classified as juniper dominated or juniper sheltered trap more snow than pixels that are not juniper dominated. The shortest melt-out timing difference across catchments was WY2011, which had a mean of 1.75 days. Catchment F had the lowest difference in melt-out timing across all years with an average of 11 days.

Earlier melt-out and greater accumulation from the juniperdominated scenario are reflected in the modeled cumulative basin SWI time series (Fig. 7). The WY day that 75% of the total SWI has entered a catchment occurs on average 7 May for the juniperdominated scenario and May 16th on sagebrush-dominated scenario, a difference of 9 days. This time varies from -12 to +27 days. Total SWI values are an average of 86 mm more for the juniper-dominated simulations and range from 42 to 150 mm.

Figure 5.

Measured (meas) snow water equivalent (SWE) points (black circles and triangles) andmodeled (mod) SWE (green lines) near the six weather stations in the South Mountain Experimental Catchments.

f05_116.jpg

Figure 6.

Modeled mean basin snow water equivalent (SWE) for juniper- and sagebrush-dominated scenarios showing higher peak accumulation and earlier melt out for juniperdominated scenario in number of days.

f06_116.jpg

Figure 7.

Modeled total basin cumulative surface water input (SWI) from each of the catchments showing the time of delay of water inputs (days) in sagebrush-dominated scenario and the higher magnitude of total SWI (mm) in juniper-dominated scenario.

f07_116.jpg

The SWE distributions on 1 April illustrate how juniper cover plays an important role in the distribution of snow cover (Fig. 8). The difference in snow cover for the juniper- versus sagebrush-dominated scenario varies markedly across years. Catchments for the juniperdominated scenario produce a more uniform snow distribution (as indicated by SWE, Fig. 8A) while the absence of trees produces a more heterogeneous snow distribution (Fig. 8B). Drifts tend to occur in the same places in all years regardless of vegetation cover but are larger in sagebrush-dominated model scenarios.

Figure 8.

Modeled snowwater equivalent (mmof SWE) distributions on 1 April of each water year for the A, juniper-dominated scenario and B, sagebrush-dominated scenario. Late season differences show that the sagebrush-dominated scenario results in more SWE stored in drifts opposed to the juniper-dominated scenario, which results in a more even snow distribution. Yellow circles indicate the location of weather stations.

f08_116.jpg

Water Balance

The four catchments behave similarly in terms of snow dynamics, even though there is large interannual variability in precipitation and weather conditions. This is demonstrated by the small range in SWI within years but the large variability between years (see Figs. 7 and 9). Average estimated ET from all years is 531 mm and 348 mm for the juniper- and sagebrush-dominated scenario, respectively (see Figs. 9 and 10, see Table 1). Mean annual measured Q for the existing juniper vegetation community is 115 mm. Estimated mean annual streamflow for the sagebrush-dominated scenario is 212 mm (see Fig. 9, Table 1). The estimated mean runoff ratio from the sagebrushdominated scenario is 0.38, compared with the juniper-dominated scenario value of 0.16 (Fig. 11).

Figure 9.

Water yearmass balance for the South Mountain Experimental Catchments showing total modeled surface water input (SWI) for each catchment as total bar height. That total input is divided into evapotranspiration (ET) depicted as the gray bar height, and streamdischarge depicted as the colored bar height. ET for the sagebrush-dominated scenario is estimated as a weighted average with 20% mountain big sagebrush and 80% low sagebrush, which is the assumed vegetation distribution.

f09_116.jpg

Discussion

The results of this study suggest that the conversion of grassland and shrub-steppe communities to juniper woodlands within the Intermountain Western United States can significantly alter local water balances by altering both patterns of snow deposition and the timing and magnitude of melt and the delivery of water to the soil.

Spatiotemporal Differences in Snow Distribution and Water Availability

Model results indicate that mean catchment peak SWE accumulations are greater for all years for all catchments in the juniperdominated scenario (see Fig. 6). The differences in peak snow accumulation result from a balance between widespread vegetation sheltering in the juniper-dominated scenario and higher wind speeds forming larger drifts in the sagebrush-dominated scenario (see Fig. 8). Both processes result in increased snow accumulation, but widespread vegetation sheltering results in a more homogenous increase in snow accumulation compared with increased snow storage in drifts induced by higher wind speeds. The lower peak basin accumulations in the sagebrush-dominated scenario occur because of increased snow scour in the absence of trees (Hiemstra et al., 2002). The snow distribution under the sagebrush-dominated scenario is concentrated in topographically sheltered snow drifts. The 1 April snow distributions from WY2012 and WY2013, which were the lowest snow years of this study, illustrate these marked differences (see Fig. 8). Similar sheltering effects on snow distribution are observed from 21 detailed snow surveys over 11 years in RCEW (Winstral and Marks, 2014), in arctic tundra (Essery et al., 1999), and in alpine and boreal forests (Pomeroy et al., 1999).

Figure 10.

Estimated evapotranspiration (ET) differences between juniper- and sagebrush-dominated scenarios.

f10_116.jpg

Model results also indicate that the snow melts out earlier in the juniper-dominated scenario (see Figs. 6 and 7). The earlier snow meltout times are a result of a balance between the previously discussed accumulation differences. The lower WY 2011 melt-out timing difference results fromlowwinds during storms for this year (see Table 2) leading to smaller drift enhancement. Smaller differences in melt-out timing from Catchment F result from high juniper cover leading larger treesheltering enhancement of snow compared with drift enhancement (see Table 3). Increased late-season melt dynamics also affect meltout timing. Net all-wave radiation increases quickly in the presence of junipers as a result of increased incoming solar radiation, decreased albedo from litter accumulation, and warming trees emitting more thermal radiation to the snowpack (Koivusalo and Kokkonen, 2002). The snow albedo in drifts decays more slowly in the sagebrush-dominated scenario due to the lack of tree litter accumulation (Link and Marks, 1999; Cristea et al., 2014). This allows isolated snow drifts to persist later into the growing season than the more evenly distributed snow cover that results from the juniper-dominated scenario.

Figure 11.

Runoff ratios (Q/SWI) showing the difference between the juniper-dominated and sagebrush-dominated scenarios for the 6 water years of this study.

f11_116.jpg

Catchment Water Balances and Streamflow Estimates

Catchment estimated ET is similar from year to year in the juniperdominated scenario. The exception is WY 2011, in which ET is 155 mm higher than the 6-yr average (see Fig. 10, Table 1). This year is characterized by a large magnitude of total precipitation but also a large proportion of spring rainfall, when ET demands increase (see Fig. 2). For this reason it is the year that we have the least confidence that loss due to canopy interception is not an important term in the water balance and, in turn, the least confidence in the sagebrush-dominated Q and juniper-dominated ET estimates. This resultmay skew streamflow estimates and runoff ratios from juniper-dominated scenarios slightly higher, but this does not affect the overall implications of the study.

Annual total catchment SWI magnitudes are higher for the juniperdominated scenario than for the sagebrush-dominated scenario as depicted by the total bar height in Figure 9. However, due to greater estimated ET fromjunipers, mean annual streamflowis often larger in the sagebrush-dominated scenarios, as indicated by the height of the colored bars in Figure 9. These modeled estimates suggest that sagebrush-dominated catchments produce substantially more streamflow than juniper-dominated catchments, about 100 mm in this study (see Figs. 9 and 11, Table 1). This result is dependent on the assumed amounts of mountain big sagebrush and low sagebrush (20% and 80%) in the catchment, as well as precipitation and weather conditions. The anomaly in this pattern is WY 2010, which has slightly below average measured precipitation and generates less streamflow for the sagebrush-dominated scenario (see Table 1). The lower stream discharge is likely the result of USC modeled ET from WY2010 being too high of an estimate to be used for the South Mountain Experimental Catchments based on precipitation-ET relationships for the other 5WYs.

Over large areas, (>500 km2)where drift and scour zones essentially average out, total snow deposition and SWI magnitudes for any given year would be expected to be similar between juniper- and sagebrush-dominated catchments. However, these catchments are relatively small comparedwith the dominant scale of precipitation patterns. There is also a large area to the east of these watersheds that is less vegetation-sheltered, which we expect to act as a snow source area. Winstral et al. (2009, 2013) observed that, in areas designated as topographic scour zones, snowaccumulates to about 80% of the surrounding vegetation height (Walker et al., 2001; Essery and Pomeroy, 2004). Additional deposition is simply scoured away. Low vegetation and shrubs (i.e., sagebrush) are filled and then become sources of snow for downwind deposition sites. In general junipers are taller and unlikely to be filled to within 80% of height and therefore able to capture snow throughout the season. For these reasons, we believe that the differences in annual snow deposition and SWI values between the juniperand sagebrush-dominated scenarios are reasonable.

Snowy, sagebrush-dominated catchments are more efficient at translating precipitation and snow melt to streamflow as is indicated by considerably higher mean runoff ratios (see Fig. 11). The difference in runoff ratios is a result of the sagebrush-dominated scenario having lower total catchment SWI values and higher mean annual discharge values, a function of estimated ET (see Table 1). Differences between runoff ratios in the juniper- and sagebrush-dominated scenarios are greater for windy years, when the snow distribution differences are dominated by drift zones (see Figs. 8 and 11, Table 2).

Model Performance and Evaluation of Assumptions

Modeled SWE values near weather stations reasonably match measured values from snow courses and snow surveys, especially considering the shallow and ephemeral nature of the snowpack in this environment. The current best-practice methods employed in this study limit errors in the precipitation distribution by 1) maintaining a realisticmass balance between measured andmodeled precipitation totals (see Table 3), 2) accounting for vegetative sheltering of blowing snow, 3) accounting for topographic induced drifting, and 4) accounting for topographic-induced decreases in accumulation. Deviations between measured and modeled SWE highlight the difficulty in parameterizing a precipitation distribution in catchments with complex vegetation and topography. Modeling errors in the parameterization of juniper effects on precipitation distribution (i.e., increased accumulation from vegetation sheltering) would have systematic effects on our mass balance results, given the extent of juniper cover in the catchments (see Fig. 1, Table 3). It is important to note that any overestimation in juniper-dominated precipitation will result in an overestimate of the same magnitude in ET from the juniper-dominated scenario (see Table 1, Eq. (2)). For this reason, the juniper-dominated ET estimates presented in Table 1 are an upper bound estimate. This would also translate into a lower bound estimate of the presented runoff ratios for the juniper-dominated scenario (see Fig. 11) and a decrease in the discrepancies between the juniper- and sagebrush-dominated scenario peak SWE accumulations (see Fig. 6) and total basin SWI (see Fig. 7).

Interception-induced losses are assumed to be negligible in this environment across both the juniper- and sagebrush-dominated scenarios. We acknowledge that in colder environments where intercepted snow remains in the canopy longer (Musselman et al., 2008; Bradford et al., 2014), or evaporative demand is higher following storms (Owens et al., 2006), direct sublimation losses can be important. In this environment, losses due to the interception of rain are minimized by the offset between high precipitation and high evaporative demand. Minimal loss due to snowinterception is supported by a careful lysimeter measurement study from a maritime site in Oregon, which showed 1) snow intercepted by a forest canopy rapidly sluffed from the trees, 2) subsequent sublimation losses were generally close to zero, and 3) sublimation was approximately 5% of winter precipitation (Stork et al., 2002). That site was wetter (average winter precipitation of 2000 mm), lower (1200 m), and more densely forested (30- to 50-m canopy heights, 65%-80% canopy closure). Their meticulous measurement effort found that up to 60% of snowwas intercepted by themature forest canopy. However, their lysimeter data show the primary interception removal mechanism was melt water drip and mass release to the ground below the canopy.

The South Mountain Experimental Catchments are located in the interior Western United States, which is generally cooler, drier, and more open in terms of canopy compared with the Oregon study site (Stork et al., 2002). Lower vegetation generally becomes covered with snow early in the winter. Although we assume negligible loss of mass to sublimation of intercepted snow,we carefully calculatewind-driven turbulence, vapor gradient, and sublimation/condensation from or to snow on the ground. Over the six snow seasons of this study, sublimation losses averaged -3.5% of precipitation, with condensation gains of +0.5%, for a mean flux of -3%. Furthermore, if direct loss from intercepted snowwas significant,wewould expect a systematic overestimate of measured snow within the juniper-dominated South Mountain. However, our model results indicate that measured and simulated snow over the six snow seasons of this study are well matched, and that,while thematch is not perfect, there is not a systematic bias (see Fig. 5).

On the basis of those results, we believe that negligible interception in this climate and environment is a valid assumption. However, the lack of measured site-specific interception data and a common perception that this component of the water balance may be large warrants more consideration. Interception and subsequent sublimation from a juniper-dominated catchment in any climate is expected to be larger than interception from a sagebrush-dominated catchment because turbulence in the upper regions of a 4- to 5-mjuniper canopywill be greater than in a submeter shrub canopy. A systematic reduction in the precipitation or SWI magnitudes in the juniper-dominated scenario compared with the sagebrush-dominated scenario would 1) decrease the difference in annual catchment precipitation/SWI totals, 2) enhance the melt-out times that 75% of the SWI enters the juniper-dominated catchments (see Fig. 7), 3) decrease the difference between ET magnitudes (see Fig. 10), and 4) decrease the difference between runoff ratios (see Fig. 11). The implications of increased interception from junipers could reduce the difference in total peak SWE and SWI between model scenarios but would result in an earlier meltout within juniperdominated catchments, thereby enhancing the extended delivery of water input into the summer for sagebrush-dominated catchments.

Effects of Water Availability on Vegetation Dynamics and Habitat

Water fromsnowdrifts is known to providemuch needed soilwater for vegetation and often sustains streamflow into the late spring/early summer (Luce et al., 1998; Luce and Tarboton, 2004; Kumar et al., 2013; Wang et al., 2013). The combination of delayed water inputs to and increased streamflow from sagebrush-dominated catchments may lead to substantial habitat benefits for sagebrush obligates when compared with the hydrology of juniper-dominated catchments. Delayed water inputs to a catchment, resulting from increased snow drifting, may result in more distributed surface water sources such as springs and wet meadows, which would improve sage grouse habitat conditions. Because the drifts delay the delivery of SWI to the soil until well into the growing season, there is water available for shrubs and forbs that would otherwise be desiccated by a summer generally without precipitation. The extension of water availability from winter precipitation provided by snow drifts is known to be important for sage grouse during late brood-rearing times (Connelly et al., 2011). Following treatment of western juniper by fire or cutting, forbs often increase in cover and production, indicating the importance of increased available water and soil nutrients (Bates et al., 2000; Roundy et al., 2014a,b).

Interacting Effects of Climate on Snow Distribution and Water Availability

The 6 yrs of this study period encompass sufficient interannual variability inweather conditions to discuss the hydrologic implications in a broader climate context. In general, model results indicate that sagebrush-dominated catchments produce more streamflow, later into the summer than juniper-dominated catchments (see Fig. 9 and 11, Table 3). Snow melt-out times are similar between juniper- and sagebrush-dominated scenarios on years with lower wind speeds during storms (WY 2010 and WY 2011) and in catchmentswith higher juniper density (catchments F and M). Under these conditions, increases in energy balance components related to the juniper canopy are outweighed by the shear magnitude of the snowpack increases by wind sheltering from trees. This suggests that there are combinations of lowwinds, high precipitation amounts, and cold air temperature conditions during storms where streamflow from juniper-dominated systems may extend longer into the summer compared with sagebrushdominated systems.

Regional shifts in precipitation from snow to rain in low- to midelevations highlight the importance of maintaining high-elevation summer sage grouse habitat (Knowles et al., 2006; Nayak et al., 2010). Because drift formation is dependent on snowfall, the extended release of meltwater into the late spring or early summer fromdriftsmay be declining for lower to middle elevations. Sage grouse habitat at raindominated lower elevations has already been substantially reduced due to extensive cheatgrass (Bromus tectorum L.) invasion and an associated increase in annual area burned (Balch et al., 2013; Miller et al., 2011). Blomberg et al. (2012) suggested that sage grouse populations across low- to mid-elevations are strongly influenced by climatedriven variation in resource availability, primarily precipitation and soil water availability for food production. The study further found sage grouse populations were negatively impacted by cheatgrass invasion and suggested that more frequent drought and increased spread of exotic grasses under warming climate conditions would further reduce sage grouse habitat. Guttery et al. (2013) likewise concluded that decreased snowfall with warming climate conditions would reduce late season soil moisture and the quality of habitat for sage grouse brood rearing. The retention of sagebrush steppe vegetation at higher elevations and the associated spatiotemporal patterns of snowaccumulation and melt may be critical with respect to maintaining sage grouse summer habitat with changing climate (Blomberg et al., 2012; Guttery et al., 2013). Sage grouse diets depend on forbs during this summer season, and they will travel a long distance, often to higher elevations, to meet this need (Connelly et al., 2011). Although this stage represents a fraction of sage grouse life cycle, the loss of high-elevation sagebrush steppe areas due to woodland encroachment may have disproportionately adverse effects on sage grouse as precipitation continues to shift toward a rain-dominated regime (Guttery et al., 2013).

Conclusions and Ecological Implications

Vegetation and topography impart distinct signatures on the distribution of snow and surface water availability that, in turn, strongly influence the diversity of ecosystem properties including habitat for sagebrush obligates like the greater sage grouse. Results from this study generally indicate that snow water equivalent peaks higher, snow melts out earlier, and more water is lost to evapotranspiration in catchments dominated by juniper as compared with sagebrush steppe vegetation. Our results also indicate that snow drifts are larger and persist longer (due to higher wind speeds) in sagebrush-dominated catchments. The prolonged snow retention in drifts generates greater annual and prolonged summer-season streamflow with respect to juniperdominated catchments. Our results suggest that catchments in snowdominated climates with sagebrush vegetation effectively capture, store, and deliver water that may be used for sustaining vegetation diversity associated with critical sage grouse habitat. In contrast, juniper-dominated systems in similar climates likely generate greater overall water input, but the timing ofwater availability is less beneficial to late season shrub and herbaceous productivity and habitat recruitment. We did not specifically assess the succession point at which juniper encroachment imparts the previously noted impacts, but our results clearly demonstrate ecohydrologic benefits associatedwith retention of sagebrush mosaics on snow-dominated rangelands. Our results imply that retention of high-elevation sagebrush vegetation in snowdominated uplands may be particularly important in sustaining sage grouse habitat under warming climate conditions. Decreased snowfall associated with warming winter temperatures at lower elevations is likely to reduce late summer water availability critical for sustaining sage grouse habitat,which is already diminished due to invasive annual grasses and increasing wildfire activity.

Acknowledgments

We thank Professor Nancy Glenn at Boise State University for assistance in the acquisition and processing of the South Mountain Lidar dataset; the Lowry and Stanford families for cooperation and property access; the Bureau of Land Management Boise District and Owyhee Field Offices, specifically Skip Nyman, Tony Runnels, John Wilford, Zane Cram, and Barry Caldwell for field support and data collection; Drs. Gerald Flerchinger and Adam Winstral for thoughtful review and suggested revisions; and Dr. Rupesh Shrestha and Alex Boehm for Lidar processing and the development of bare earth and canopy height models.

References

1.

Balch, JK, Bradley, BA, D'Antonio, CM, Gómez-Dans, J, 2013. Introduced annual grass increases regional fire activity across the arid western USA (1980–2009). Global Change Biology 19 (1), 173–183. Google Scholar

2.

Bates, JD, Miller, RF, Svejcar, TJ, 2000. Understory dynamics in cut and uncut western juniper woodlands. Journal of Range Management 53 (1), 119–126. Google Scholar

3.

BCAL LiDAR Tools ver 1.5.3 [computer program]. Idaho State University, Department of Geosciences, Boise Center Aerospace Laboratory, Boise, ID, USA. Google Scholar

4.

Blomberg, EJ, Sedinger, JS, Atamian, MT, Nonne, DV, 2012. Characteristics of climate and landscape disturbance influence the dynamics of greater sage-grouse populations. Ecosphere 3 (6), 1–20. Google Scholar

5.

Bonta, J, Pierson, F, 2003. Design, measurement, and sampling with drop-box weirs. Applied Engineering in Agriculture 19 (6), 689–700. Google Scholar

6.

Bradford, JB, Schlaepfer, DR, Lauenroth, WK, 2014. Ecohydrology of adjacent sagebrush and lodgepole pine ecosystems: the consequences of climate change and disturbance. Ecosystems 17 (4), 590–605. Google Scholar

7.

Breshears, DD, Myers, OB, Johnson, SR, Meyer, CW, Martens, SN, 1997. Differential use of spatially heterogeneous soil moisture by two semiarid woody species: Pinus edulis and Juniperus monosperma. Journal of Ecology 85 (3), 289–299. Google Scholar

8.

Breshears, DD, Nyhan, JW, Heil, CE., Wilcox, BP, 1998. Effects of woody plants on microclimate in a semiarid woodland: soil temperature and evaporation in canopy and intercanopy patches. International Journal of Plant Sciences 159 (6), 1010–1017. Google Scholar

9.

Burke, IC, 1989. Control of nitrogen mineralization a sagebrush steppe landscape. Ecology 70 (4), 1115–1126. Google Scholar

10.

Caldwell, MM, Richards, JH, 1989. Hydraulic lift: water efflux from upper roots improves effectiveness of water uptake by deep roots. Oecologia 79 (1), 1–5. Google Scholar

11.

Casazza, ML, Coates, PS, Overton, CT, 2011. Linking habitat selection and brood success in greater sage-grouse. In: Sandercock, B.K., Segelbacher, G. (Eds.), Ecology, conservation, and management of grouse. No. 39. University of California Press Berkeley, Berkeley, CA, USA, pp. 151–167. Google Scholar

12.

Chauvin, GM, Flerchinger, GN, Link, TE, Marks, D, Winstral, AH, Seyfried, MS, 2011. Longterm water balance and conceptual model of a semi-arid mountainous catchment. Journal of Hydrology 400 (1–2), 133–143. Google Scholar

13.

Cline, JF, Uresk, DW, Rickard, WH, 1977. Comparison of soil water used by a sagebrushbunchgrass and a cheatgrass community. Journal of Range Management 30 (3), 199–201. Google Scholar

14.

Connelly, JW, Schroeder, MA, Sands, AR, Braun, CE, 2000. Guidelines to manage sage grouse populations and their habitats. Wildlife Society Bulletin (1973–2006) 28 (4), 967–985. Google Scholar

15.

Connelly, J, Rinkes, E, Braun, C, 2011. Characteristics of greater sage-grouse habitats: a landscape species at micro and macro scales. In: Knick, S.T, Connelly, J.W. (Eds.), Greater sage-grouse: ecology and conservation of a landscape species and its habitats Vol. 38 of studies in avian biology. University of California Press, Berkeley, CA, USA, pp. 69–83. Google Scholar

16.

Cristea, NC, Lundquist, JD, Loheide, SP, Lowry, CS, Moore, CE, 2014. Modelling how vegetation cover affects climate change impacts on streamflow timing and magnitude in the snowmelt-dominated upper Tuolumne Basin, Sierra Nevada. Hydrological Processes 28 (12), 3896–3918. Google Scholar

17.

Davis, RE, Dozier, J, LaChapelle, ER, Perla, R, 1985. Field and laboratory measurements of snow liquid water by dilution. Water Resources Research 21 (9), 1415–1420. Google Scholar

18.

Donnelly, JP, Naugle, DE, Hagen, CA, Maestas, JD, 2016. Public lands and private waters: scarce mesic resources structure land tenure and sage-grouse distributions. Ecosphere 7 (1), 1–15. Google Scholar

19.

Dozier, J, 1980. A clear-sky spectral solar radiation model for snow-covered mountainous terrain. Water Resources Research 16 (4), 709–718. Google Scholar

20.

Dozier, J, Frew, J, 1981. Atmospheric corrections to satellite radiometric data over rugged terrain. Remote Sensing of Environment 11, 191–205. Google Scholar

21.

Dubayah, RC, 1994. Modeling a solar radiation topoclimatology for the rio grande river basin. Journal of Vegetation Science 5 (5), 627–640. Google Scholar

22.

Essery, R, Pomeroy, J, 2004. Vegetation and topographic control of wind-blown snow distributions in distributed and aggregated simulations for an arctic tundra basin. Journal of Hydrometeorology 5 (5), 735–744. Google Scholar

23.

Essery, R, Li, L, Pomeroy, J, 1999. A distributed model of blowing snow over complex terrain. Hydrological Processes 13 (14–15), 2423–2438. Google Scholar

24.

Flerchinger, GN, Cooley, KR, 2000. A ten-year water balance of a mountainous semi-arid watershed. Journal of Hydrology 237 (1–2), 86–99. Google Scholar

25.

Flerchinger, GN, Cooley, KR, Hanson, CL, Seyfried, MS, 1998. A uniform versus an aggregated water balance of a semi-arid watershed. Hydrological Processes 12 (2), 331–342. Google Scholar

26.

Flerchinger, GN, Marks, D, Reba, ML, Yu, Q, Seyfried, MS, 2010. Surface fluxes and water balance of spatially varying vegetation within a smallmountainous headwater catchment. Hydrology and Earth System Sciences 14 (6), 965–978. Google Scholar

27.

Flerchinger, G.N., Seyfried, M.S., Hardegree, S.P., 2016. Hydrologic Response and Recovery to Prescribed Fire and Vegetation Removal in a Small Rangeland Catchment. Ecohydrol  http://dx.doi.org/10.1002/eco.1751Google Scholar

28.

Frew, J, 1990. The image processing workbench [Ph. D. thesis] Department of Geography, University of California, Santa Barbara, CA, USA. Google Scholar

29.

Garen, DC, 1995. Estimation of spatially distributed values of daily precipitation in mountainous areas. Mountain hydrology: peaks and valleys in research and applications. Canadian Water Resources Association, Cambridge, Ontario, Canada, pp. 237–242. Google Scholar

30.

Garen, DC, Marks, D, 2005. Spatially distributed energy balance snowmelt modelling in a mountainous river basin: estimation of meteorological inputs and verification of model results. Journal of Hydrology 315 (1–4), 126–153. Google Scholar

31.

Garen, DC, Johnson, GL, Hanson, CL, 1994. Mean areal precipitation for daily hydrologic modeling in mountainous regions. JAWRA Journal of the American Water Resources Association 30 (3), 481–491. Google Scholar

32.

Guttery, MR, Dahlgren, DK, Messmer, TA, Connelly, JW, Reese, KP, Terletzky, PA, Burkepile, N, Koons, DN, 2013. Effects of landscape-scale environmental variation on greater sage-grouse chick survival. PLoS ONE 8 (6), e65582. Google Scholar

33.

Hammersmark, CT, Rains, MC, Mount, JF, 2008. Quantifying the hydrological effects of stream restoration in a montane meadow, northern California, USA. River Research and Applications 24 (6), 735–753. Google Scholar

34.

Hiemstra, CA, Liston, GE, Reiners, WA, 2002. Snowredistribution by wind and interactions with vegetation at upper treeline in the Medicine Bow Mountains, Wyoming, USA. Arctic, Antarctic, and Alpine Research 34 (3), 262–273. Google Scholar

35.

Hiemstra, CA, Liston, GE, Reiners, WA, 2006. Observing, modelling, and validating snow redistribution by wind in a Wyoming upper treeline landscape. Ecological Modelling 197 (1–2), 35–51. Google Scholar

36.

Johnson, DD, Miller, RF, 2006. Structure and development of expanding western juniper woodlands as influenced by two topographic variables. Forest Ecology and Management 229 (1–3), 7–15. Google Scholar

37.

Knick, S, Connelly, J.W. (Eds.), 2011. Greater sage-grouse: ecology and conservation of a landscape species and its habitats Vol. 38 of studies in avian biology. University of California Press, Berkeley, CA, USA. Google Scholar

38.

Knowles, N, Dettinger, MD, Cayan, DR, 2006. Trends in snowfall versus rainfall in the western United States. Journal of Climate 19 (18), 4545–4559. Google Scholar

39.

Koivusalo, H., and T. Kokkonen. 2002. Snow processes in a forest clearing and in a coniferous forest. Journal of Hydrology 262(1–4) :145–164. Google Scholar

40.

Kormos, PR, Marks, D, McNamara, JP, Marshall, HP, Winstral, A, Flores, AN, 2014. Snow distribution, melt and surface water inputs to the soil in the mountain rain-snow transition zone. Journal of Hydrology 519 (A), 190–204. Google Scholar

41.

Kumar, M, Marks, D., Dozier, J, Reba, M, Winstral, A, 2013. Evaluation of distributed hydrologic impacts of temperature-index and energy-based snow models. Advances in Water Resources 56, 77–89. Google Scholar

42.

Link, TE, Marks, D, 1999. Point simulation of seasonal snow cover dynamics beneath boreal forest canopies. Journal of Geophysical Research - Atmospheres 104 (D22), 27841–27857. Google Scholar

43.

Liston, GE, McFadden, JP, Sturm, M, Pielke, RA, 2002. Modelled changes in arctic tundra snow, energy andmoisture fluxes due to increased shrubs. Global Change Biology 8 (1), 17–32. Google Scholar

44.

Loheide, SP, Gorelick, SM, 2007. Riparian hydroecology: a coupled model of the observed interactions between groundwater flow and meadow vegetation patterning. Water Resources Research 43 (7), W07414. Google Scholar

45.

Loheide, SP, Deitchman, RS, Cooper, DJ, Wolf, EC, Hammersmark, CT, Lundquist, JD, 2009. A framework for understanding the hydroecology of impacted wet meadows in the Sierra Nevada and Cascade Ranges, California, USA. Hydrogeology Journal 17 (1), 229–246. Google Scholar

46.

Luce, CH, Tarboton, DG, 2004. The application of depletion curves for parameterization of subgrid variability of snow. Hydrological Processes 18 (8), 1409–1422. Google Scholar

47.

Luce, CH, Tarboton, DG, Cooley, KR, 1998. The influence of the spatial distribution of snow on basin-averaged snowmelt. Hydrological Processes 12 (10–11), 1671–1683. Google Scholar

48.

Marks, D, Dozier, J, 1979. A clear-sky longwave radiation model for remote alpine areas. Archiv für Meteorologie, Geophysik und Bioklimatologie, Serie B 27 (2), 159–187. Google Scholar

49.

Marks, D, Winstral, A, 2001. Comparison of snow deposition, the snow cover energy balance, and snowmelt at two sites in a semiarid mountain basin. Journal of Hydrometeorology 2 (3), 213–227. Google Scholar

50.

Marks, D, Domingo, J, Frew, J, 1999b. Software tools for hydro-climaticmodeling and analysis: image processing workbench, ARS-USGS Version 2. ARS Technical Bulletin 99, 1. Google Scholar

51.

Marks, D, Domingo, J, Susong, D, Link, T, Garen, D, 1999a. b. A spatially distributed energy balance snowmelt model for application in mountain basins. Hydrological Processes 13 (12–13) , 1935–1959. Google Scholar

52.

Marks, D, Dozier, J, Davis, RE, 1992. Climate and energy exchange at the snow surface in the Alpine Region of the Sierra Nevada: 1. Meteorological measurements and monitoring. Water Resources Research 28 (11), 3029–3042. Google Scholar

53.

Marks, D, Kimball, J, Tingey, D, Link, T, 1998. The sensitivity of snowmelt processes to climate conditions and forest cover during rain-on-snow: a case study of the 1996 Pacific Northwest flood. Hydrological Processes 12 (10–11), 1569–1587. Google Scholar

54.

Marks, D, Link, T, Winstral, A, Garen, D, 2001. Simulating snowmelt processes during rainon- snow over a semi-arid mountain basin. Annals of Glaciology 32 (1), 195–202. Google Scholar

55.

Marks, D, Winstral, A, Reba, M, Pomeroy, J, Kumar, M, 2013. An evaluation ofmethods for determining during-stormprecipitation phase and the rain/snowtransition elevation at the surface in a mountain basin. Advances in Water Resources 55, 98–110. Google Scholar

56.

Marks, D, Winstral, A, Seyfried, M, 2002. Simulation of terrain and forest shelter effects on patterns of snow deposition, snowmelt and runoff over a semi-arid mountain catchment. Hydrological Processes 16 (18), 3605–3626. Google Scholar

57.

Marshall, SE, Warren, SG, 1987. Parameterization of snow albedo for climate models. In: Goodison, B.E, Barry, R.G, Dozier, J. (Eds.), Large scale effects of seasonal snow cover. IAHS-AIHS 166, pp. 44–50. Google Scholar

58.

Miller, RF, Tausch, RJ, 2001. The role of fire in pinyon and juniper woodlands: a descriptive analysis. In: Galley, KEM, Wilson, TP (Eds.), Proceedings of the invasive species workshop: the role of fire in the control and spread of invasive species. Fire Conference 2000: the first national congress on fire ecology, prevention, andmanagement. Miscellaneous Publication No. 11. Tall Timbers Research Station, Tallahassee, FL, pp. 15–30. Google Scholar

59.

Miller, RF, Bates, JD, Svejcar, TJ, Pierson, FB, Eddleman, LD, 2005. Biology, ecology, and management of western juniper (Juniperus occidentalis). Technical report. Agricultural Experiment Station, Oregon State University, Corvallis, OR, USA. Google Scholar

60.

Miller, RF, Knick, ST, Pyke, DA, Meinke, CW, Hanser, SE, Wisdom, MJ, Hild, AL, 2011. Characteristics of sagebrush habitats and limitations to long-term conservation. In: Knick, S.T, Connelly, J.W. (Eds.), Greater sage-grouse: ecology and conservation of a landscape species and its habitats Vol. 38 of studies in avian biology. University of California Press, Berkeley, CA, USA, pp. 145–184. Google Scholar

61.

Miller, RF, Ratchford, J, Roundy, BA, Tausch, RJ, Hulet, A, Chambers, J, 2014. Response of conifer-encroached shrublands in the Great Basin to prescribed fire and mechanical treatments. Rangeland Ecology & Management 67 (5), 468–481. Google Scholar

62.

Miller, RF, Svejcar, TJ, Rose, JA, 2000. Impacts of western juniper on plant community composition and structure. Journal of Range Management 53 (6), 574–585. Google Scholar

63.

Miller, RF, Tausch, RJ, McArthur, ED, Johnson, DD, Sanderson, SC, 2008. Age structure and expansion of pinon-juniper woodlands: a regional perspective in the intermountain west. Research Paper RMRS-RP-69. U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station, Fort Collins, CO, USA. Google Scholar

64.

Mollnau, C, Newton, M, Stringham, T, 2014. Soil water dynamics and water use in awestern juniper (Juniperus occidentalis) woodland. Journal of Arid Environments 102, 117–126. Google Scholar

65.

Musselman, KN, Molotch, NP, Brooks, PD, 2008. Effects of vegetation on snow accumulation and ablation in a mid-latitude sub-alpine forest. Hydrological Processes 22 (15), 2767–2776. Google Scholar

66.

Nayak, A, Marks, D, Chandler, DG, Seyfried, M, 2010. Long-term snow, climate, and streamflow trends at the Reynolds Creek Experimental Watershed, Owyhee Mountains, Idaho, United States. Water Resources Research 46 (6), w06519. Google Scholar

67.

Nayak, A, Marks, D, Chandler, D, Winstral, A, 2011. Modeling interannual variability in snow-cover development and melt for a semiarid mountain catchment. Journal of Hydrologic Engineering 17 (1), 74–84. Google Scholar

68.

Newman, BD, Breshears, DD, Gard, MO, 2010. Evapotranspiration partitioning in a semiarid woodland: ecohydrologic heterogeneity and connectivity of vegetation patches. Vadose Zone Journal 9 (3), 561–572. Google Scholar

69.

NRCS, U, 2016. Web soil survey. Google Scholar

70.

Owens, MK, Lyons, RK, Alejandro, CL, 2006. Rainfall partitioning within semiarid juniper communities: effects of event size and canopy cover. Hydrological Processes 20 (15), 3179–3189. Google Scholar

71.

Petersen, SL, Stringham, TK, Roundy, BA, 2009. A process-based application of state-andtransition models: a case study of western juniper (juniperus occidentalis) encroachment. Rangeland Ecology & Management 62 (2), 186–192. Google Scholar

72.

Pierson, FB, Bates, JD, Svejcar, TJ, Hardegree, SP, 2007. Runoff and erosion after cutting western juniper. Rangeland Ecology & Management 60 (3), 285–292. Google Scholar

73.

Pierson, FB, JasonWilliams, C, Hardegree, SP, Clark, PE, Kormos, PR, Al-Hamdan, OZ, 2013. Hydrologic and erosion responses of sagebrush steppe following juniper encroachment, wildfire, and tree cutting. Rangeland Ecology & Management 66 (3), 274–289. Google Scholar

74.

Pierson, FB, Williams, CJ, Kormos, PR, Hardegree, SP, Clark, PE, Rau, BM, 2010. Hydrologic vulnerability of sagebrush steppe following pinyon and juniper encroachment. Rangeland Ecology & Management 63 (6), 614–629. Google Scholar

75.

Pomeroy, J.W, Hedstrom, N, Parviainen, J, 1999. The snow mass balance of Wolf Creek. In: Pomeroy, J, Granger, R (Eds.), Wolf Creek Research Basin: Hydrology, Ecology, Environment (National Water Research Institute). Minister of Environment, Saskatoon, pp. 15–30. Google Scholar

76.

Rasouli, K, Pomeroy, JW, Marks, DG, 2015. Snowpack sensitivity to perturbed climate in a cool mid-latitude mountain catchment. Hydrological Processes 29 (18), 3925–3940. Google Scholar

77.

Reba, ML, Marks, D, Link, TE, Pomeroy, J, Winstral, A, 2014. Sensitivity of model parameterizations for simulated latent heat flux at the snow surface for complex mountain sites. Hydrological Processes 28 (3), 868–881. Google Scholar

78.

Reba, ML, Marks, D, Seyfried, M, Winstral, A, Kumar, M, Flerchinger, G, 2011. A long-term data set for hydrologic modeling in a snow-dominated mountain catchment. Water Resources Research 47 (7). Google Scholar

79.

Richards, JH, Caldwell, MM, 1987. Hydraulic lift: substantial nocturnal water transport between soil layers by Artemisia tridentata roots. Oecologia 73 (4), 486–489. Google Scholar

80.

Rodriguez-Iturbe, I, 2000. Ecohydrology: a hydrologic perspective of climate-soilvegetation dynamies. Water Resources Research 36 (1), 3–9. Google Scholar

81.

Romme, WH, Allen, CD, Bailey, JD, Baker, WL, Bestelmeyer, BT, Brown, PM, Eisenhart, KS, Floyd, ML, Huffman, DW, Jacobs, BF, Miller, RF, Muldavin, EH, Swetnam, TW, Tausch, RJ, Weisberg, PJ, 2009. Historical and modern disturbance regimes, stand structures, and landscape dynamics in piñon-juniper vegetation of the western United States. Rangeland Ecology & Management 62 (3), 203–222. Google Scholar

82.

Roundy, BA, Miller, RF, Tausch, RJ, Young, K, Hulet, A, Rau, B, Jessop, B, Chambers, JC, Eggett, D, 2014a. a. Understory cover responses to piñon-juniper treatments across tree dominance gradients in the Great Basin. Rangeland Ecology & Management 67 (5), 482–494. Google Scholar

83.

Roundy, BA, Young, K, Cline, N, Hulet, A, Miller, RF, Tausch, RJ, Chambers, JC, Rau, B, 2014b. Piñon-juniper reduction increases soil water availability of the resource growth pool. Rangeland Ecology & Management 67 (5), 495–505. Google Scholar

84.

Ryel, RJ, Leffler, AJ, Ivans, C, Peek, MS, Caldwell, MM, 2010. Functional differences in water-use patterns of contrasting life forms in Great Basin steppelands. Vadose Zone Journal 9 (3), 548–560. Google Scholar

85.

Ryel, RJ, Leffler, AJ, Peek, MS, Ivans, CY, Caldwell, MM, 2004. Water conservation in Artemisia tridentata through redistribution of precipitation. Oecologia 141 (2), 335–345. Google Scholar

86.

Sankey, T, Shrestha, R, Sankey, JB, Hardegree, S, Strand, E, 2013. Lidar-derived estimate and uncertainty of carbon sink in successional phases of woody encroachment. Journal of Geophysical Research - Biogeosciences 118 (3), 1144–1155. Google Scholar

87.

Schlaepfer, DR, Lauenroth, WK, Bradford, JB, 2012. Ecohydrological niche of sagebrush ecosystems. Ecohydrology 5 (4), 453–466. Google Scholar

88.

Seyfried, MS, Grant, LE, Marks, D, Winstral, A, McNamara, J, 2009. Simulated soil water storage effects on streamflow generation in a mountainous snowmelt environment, Idaho, USA. Hydrological Processes 23 (6), 858–873. Google Scholar

89.

Stork, P, Lettenmaier, DP, Bolton, SM, 2002. Measurement of snow interception and canopy effects on snow accumulation and melt in a mountainous maritime climate, Oregon, United States. Water Resources Research 38 (11), 1–16. Google Scholar

90.

Sturges, DL, 1977. Soil water withdrawal and root characteristics of big sagebrush. American Midland Naturalist 98 (2), 257–274. Google Scholar

91.

Sturm, M, Holmgren, J, McFadden, JP, Liston, GE, Chapin, FS, Racine, CH, 2001. Snow- shrub interactions in arctic tundra: a hypothesis with climatic implications. Journal of Climate 14 (3), 336–344. Google Scholar

92.

Tausch, RJ, West, NE, Nabi, AA, 1981. Tree age and dominance patterns in Great Basin pinyon-juniper woodlands. Journal of Range Management 34 (4), 259–264. Google Scholar

93.

Walker, D, Billings, W, De Molenaar, J, 2001. Snow-vegetation interactions in tundra environments. Snow Ecology 266–324. Google Scholar

94.

Wang, R, Kumar, M, Marks, D, 2013. Anomalous trend in soil evaporation in a semi-arid, snow-dominated watershed. Advances in Water Resources 57, 32–40. Google Scholar

95.

Whiting, JA, Godsey, SE, 2016. Discontinuous headwater stream networks with stable flowheads, Salmon River Basin, Idaho. Hydrological Processes. Google Scholar

96.

Williams, CJ, McNamara, JP, Chandler, DG, 2009. Controls on the temporal and spatial variability of soil moisture in a mountainous landscape: the signature of snow and complex terrain. Hydrology and Earth System Sciences 13 (7), 1325–1336. Google Scholar

97.

Williams, CJ, Pierson, FB, Al-Hamdan, OZ, Kormos, PR, Hardegree, SP, Clark, PE, 2014. Can wildfire serve as an ecohydrologic threshold-reversal mechanism on juniperencroached shrublands. Ecohydrology 7 (2), 453–477. Google Scholar

98.

Williams, CJ, Pierson, FB, Spaeth, KE, Brown, JR, Al-Hamdan, OZ, Weltz, MA, Nearing, MA, Herrick, JE, Boll, J, Robichaud, PR, Goodrich, DC, Heilman, P, Guertin, DP, Hernandez, M, Wei, H, Hardegree, SP, Strand, EK, Bates, JD, Metz, LJ, Nichols, MH, 2016. Incorporating hydrologic data and ecohydrologic relationships into ecological site descriptions. Rangeland Ecology & Management 69 (1), 4–19. Google Scholar

99.

Winstral, A, Marks, D, 2002. Simulating wind fields and snowredistribution using terrainbased parameters to model snow accumulation and melt over a semi-arid mountain catchment. Hydrological Processes 16 (18), 3585–3603. Google Scholar

100.

Winstral, A, Marks, D, 2014. Long-term snow distribution observations in a mountain catchment: assessing variability, time stability, and the representativeness of an index site. Water Resources Research 50 (1), 293–305. Google Scholar

101.

Winstral, A, Marks, D, Gurney, R, 2009. An efficient method for distributing wind speeds over heterogeneous terrain. Hydrological Processes 23 (17), 2526–2535. Google Scholar

102.

Winstral, A, Marks, D, Gurney, R, 2013. Simulating wind-affected snow accumulations at catchment to basin scales. Advances in Water Resources 55, 64–79. Google Scholar

103.

Young, KR, Roundy, BA, Eggett, DL, 2013. Tree reduction and debris from mastication of utah juniper alter the soil climate in sagebrush steppe. Forest Ecology and Management 310, 777–785. Google Scholar

Notes

[1] This research was funded in part by Grant NA05OAR4601137 from the National Oceanic and Atmospheric Administration Earth System Research Laboratory Physical Sciences Division; the Bureau of Land Management Owyhee Uplands Pilot Project (ISU-BLMAgreement #DLA060249 and ARS-BLM Agreement #DLI050018); NSF Idaho EPSCoR Program; NSF under award number EPS-0814387 and CBET-0854553; U.S. Dept of Agriculture (USDA)-NRCS Water and Climate Center-Portland, Oregon (60-5362-4-003); the NSF Reynolds Creek CZO Project (58-5832-4-004); and USDA-ARS CRIS Snow and Hydrologic Processes in the Intermountain West (5362-13610-008-00D). USDAis an equal opportunity provider and employer. Mention of trade names or commercial products in this publication is solely for the purpose of providing specific information and does not imply recommendation or endorsement by the USDA.

Published by Elsevier Inc. on behalf of The Society for Range Management. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Patrick R. Kormos, Danny Marks, Frederick B. Pierson, C. Jason Williams, Stuart P. Hardegree, Scott Havens, Andrew Hedrick, Jonathan D. Bates, and Tony J. Svejcar "Ecosystem Water Availability in Juniper Versus Sagebrush Snow-Dominated Rangelands," Rangeland Ecology and Management 70(1), 116-128, (1 January 2017). https://doi.org/10.1016/j.rama.2016.05.003
Received: 17 February 2016; Accepted: 25 May 2016; Published: 1 January 2017
JOURNAL ARTICLE
13 PAGES


SHARE
ARTICLE IMPACT
Back to Top