- Split View
-
Views
-
Cite
Cite
Mohamed A B Abdallah, Ricardo Mata-González, Jay S Noller, Carlos G Ochoa, Effects of western juniper (Juniperus occidentalis) control on ecosystem nitrogen stocks in central Oregon, USA, Journal of Plant Ecology, Volume 14, Issue 6, December 2021, Pages 1073–1089, https://doi.org/10.1093/jpe/rtab052
- Share Icon Share
Abstract
In the Oregon of USA, the control of western juniper (Juniperus occidentalis) is an accepted rangeland management practice to restore sagebrush steppe habitats of importance to wildlife and livestock. The effects of juniper cutting on ecosystem nitrogen, however, have not been well addressed although woody plant control has important implications for local watershed management and regional nitrogen pools.
We quantified ecosystem nitrogen stocks in two adjacent watersheds, comprised of a treated watershed (most juniper removed) and an untreated watershed (juniper not removed). Thirteen years after juniper removal, we measured aboveground nitrogen stocks for juniper trees, shrubs, grasses and litter in both watersheds. We also measured belowground nitrogen stocks (roots and soil) in both watersheds at two soil depths (0–25 and 25–50 cm).
Aboveground nitrogen stocks were 6.9 times greater in the untreated than in the treated watershed considering the much larger aboveground biomass. However, root nitrogen stocks were 3.1 times greater in the treated one due to the gain of understory root biomass associated with juniper cutting. Soil nitrogen stocks at both 0–25 and 25–50 cm depths were not affected by juniper removal. Overall, total ecosystem nitrogen stocks did not differ between the treated (9536 kg N ha−1) and untreated (9456 kg N ha−1) watersheds. The greatest ecosystem nitrogen accumulation (at least 95% total ecosystem nitrogen) resided belowground (soil 0–50 cm and roots) in both watersheds. This study provides evidence that the benefits of juniper removal can be attained without significantly affecting the capacity of ecosystem nitrogen storage.
摘要
在美国俄勒冈州,控制西美圆柏(Juniperus occidentalis)的数量是一种公认的牧场管理措施,该做法 有助于恢复蒿草草原(sagebrush steppe)生境,该生境对野生动物和家畜都十分重要。然而,尽管控制木本 植物数量会对当地的流域管理和区域性氮库造成重要影响,但砍伐西美圆柏对生态系统中氮元素的影响问题尚未得到很好的解决。本文定量研究了两个相邻流域生态系统中的氮储量,其中一个流域经过处理(流域内的大部分圆柏已被清除掉),而另一个流域未处理(圆柏未被清除)。在圆柏被移除13年后,我们测定了两个流域里圆柏树林、灌木丛、草丛和枯枝落叶层的地上氮储量,以及两个流域中两个土层(0–25和25–50 cm)内的地下氮储量(根系和土壤氮储量)。研究结果表明,未处理流域的地上氮储量是处理流域 的6.9倍,因为未处理流域的地上生物量要大得多。然而,由于砍伐圆柏导致林下植被的根系生物量增加,所以处理流域的根系氮储量是未处理流域的3.1倍。0–25和25–50 cm土层氮储量没有受到圆柏砍伐的影响。总体而言,生态系统总氮储量在处理流域(9536 kg N ha−1)和未处理流域(9456 kg N ha−1)之间并没有显著差异。在两个流域里,生态系统中最大的氮积累量(至少95%的生态系统总氮量)均存在于地下(0–25 cm深的土壤以及根系)。这项研究证明,清除圆柏并不会显著影响生态系统的储氮能力。
INTRODUCTION
The widespread invasion of grasslands by woody species has occurred in many ecosystems (Barger et al. 2011; Li et al. 2016; Stevens et al. 2017; Wang et al. 2019). This geographically worldwide vegetation change, woody encroachment, has been shown to influence the ecosystem structure by altering the vegetation composition, hydrology, as well as the spatial distribution and fluxes of nutrients (Archer 2010; Chen et al. 2015; Michaelides et al. 2012). As woody species establish and replace herbaceous species, the quantities and qualities of aboveground and belowground litter inputs, and the soil microbial biomass are altered (Barger et al. 2011; Godoy et al. 2010; Li et al. 2017; Liao and Boutton 2008; Zhou et al. 2017), with subsequent implications on carbon and nitrogen pools (Hughes et al. 2006; McKinley and Blair 2008; Throop and Archer 2008). Nitrogen is considered the growth-limiting nutrient in arid and semi-arid regions (Gebauer and Ehleringer 2000) and almost in all western ecosystems of the USA (Rau et al. 2009), yet the influence of woody plants on nitrogen accumulation requires further research. Woody plant proliferation resulted in increases (Archer et al. 2004; Boutton and Liao 2010; Liao et al. 2006; Springsteen et al. 2010), declines (Li et al. 2012; Yusuf et al. 2015), as well as no significant change (McCarron et al. 2003) in soil nitrogen storage. Moreover, the total nitrogen pool in grasslands experiencing woody plant encroachment tended to increase (Boutton and Liao 2010; Liao et al. 2008) and decrease (Jackson et al. 2002). Realizing the impacts of transition from grassland to woodland on total nitrogen has significant implications for global change, whole ecosystem dynamics, land resource use and ecosystem management (McKinley et al. 2008; Wei et al. 2009; Yusuf et al. 2015).
Over the past 150 years, juniper (Juniperus spp.) and piñon (Pinus spp.) coniferous woodlands have greatly expanded their range across the Intermountain region of the western USA (Omernik 1987; Romme et al. 2009). Almost, 95% of the expansion has occurred in the sagebrush (Artemisia spp.) steppe communities (Miller et al. 2011). Among the range of encroaching juniper species, western juniper (Juniperus occidentalis spp. occidentalis Hook.) is estimated to cover approximately four million hectares in semi-arid ecosystems of the inland northwest USA (Azuma et al. 2005). The greatest concentrations of western juniper are found in eastern Oregon and northeast California (Bates 1996). Several factors are associated with juniper encroachment, including a rise of atmospheric carbon dioxide, climate change, overgrazing and reduced fire frequencies (Miller et al. 2005; Soulé et al. 2004). The encroachment of juniper is reported to cause major declines in understory productivity and diversity (Miller et al. 2014a, 2014b; Roundy et al. 2014), change the spatial distribution of nutrients litter and soils beneath juniper canopies (Klemmedson and Tiedemann 2000; Miwa and Reuter 2010), increase soil erosion (Miller et al. 2005; West 1984) and increase interception of precipitation (Eddleman et al. 1994; Larsen 1993). Juniper control has not resulted in a net change in the total ecosystem carbon stocks in the treated area with respect to the intact juniper area (Abdallah et al. 2020a). Contrary to the area dominated by juniper, control of juniper stands resulted in significant hydrological improvements (Abdallah et al. 2020b; Mata-González et al. 2021; Ochoa et al. 2018; Ray et al. 2019).
Although studies pertaining the western juniper ecosystem are increasing, quantitative evidence regarding the effects of juniper encroachment and removal on ecosystem nitrogen pools is lacking. Studies indicated that succession to western juniper woodland increased nitrogen and other soil nutrients in juniper biomass, litter mats and canopy influenced soils (up to 30 cm depth), but understory vegetation such as grasses and shrubs, were not counted for the determination of total nitrogen pools (Bates et al. 2002; Doescher et al. 1987; Klemmedson and Tiedemann 2000; Tiedemann and Klemmedson 1995). The greater aboveground production of woody encroachers in semiarid rangelands results in an increase in litter biomass with its greater inherent biochemical resistance to decomposition, contributing to nitrogen input to the soil (Liao et al. 2006; Mogashoa et al. 2021; Sandhage-Hofmann et al. 2020). However, control of woody plants with deep and fibrous roots commonly results in a decline in soil nitrogen concentrations (Eldridge and Ding 2020; John et al. 2012). Control of western juniper (cut treatment) increased grass nitrogen biomass to 6.19 kg N ha−1 compared with 0.59 kg N ha−1 in woodland treatment in the second year post-cutting, but neither the aboveground juniper biomass nor belowground biomass was determined (Bates et al. 2000). Information on nitrogen stocks as affected by western juniper removal does not exist.
This research aimed to provide quantitative estimates of ecosystem nitrogen stocks in juniper dominated systems following juniper cutting in a semiarid rangeland ecosystem site in central Oregon in the USA. Previously, vegetation dynamics at eastern Oregon were described three years (Dittel et al. 2018) and at this site ten years after juniper cutting (Ray et al. 2019). These evaluations indicated that juniper trees cutting was effective at increasing understory biomass and cover, and regrowth juniper trees. Since the implementation of juniper reduction treatments has changed the vegetation composition, the potential effect of this management practice on nitrogen accumulation is hard to predict. Assessing changes in comprehensive ecosystem-level nitrogen pools (e.g. aboveground, root, litter and soil nitrogen) require an understanding of how both aboveground pools (including understory vegetation) and belowground pools are affected by with and without juniper woodlands. Such information is important for enhancing our knowledge of woody encroachment and managed removal influence on local, regional and global nitrogen cycling.
Therefore, the objective of this study was to quantify and compare major pools of nitrogen in an encroached juniper watershed and an adjacent watershed where juniper removal took place 13 years prior to quantification. We hypothesized that 13 years after cutting mature western juniper, aboveground and belowground nitrogen stocks would have decreased in the treated watershed compared with the untreated watershed, implying lower soil nitrogen pools due to juniper removal. This research was unique because it involved a paired study approach to monitoring changes in ecosystem nitrogen pools post western juniper removal. Prior to this study, vegetation cover (including juniper, shrubs and grasses) did not differ significantly between the adjacent watersheds (Fisher 2004). Thus, the vegetation differences between the watersheds reported in this study are due to juniper clearing in the treated watershed.
MATERIALS AND METHODS
General site description
This study was carried out in the Camp Creek-Paired Watershed Study (CCPWS) site, located 27 km northeast of Brothers, Oregon (43.96° N, 120.34° W). The CCPWS site covers an area of around 212 ha and includes paired watersheds, one treated (116 ha) and the other untreated (96 ha) (Supplementary Fig. S1) with elevations ranging from 1370 to 1524 m. The untreated watershed is dominated by western juniper. In the treated watershed, about 90% of the western juniper stands were removed in the fall of 2005 using chainsaws, leaving only old-growth trees intact and a big sagebrush (Artemisia tridentata spp. vaseyana) as dominant overstory vegetation. Juniper canopy cover averaged 31.5% and tree density averaged 327 trees/ha in the untreated watershed (Abdallah et al. 2020a). The average interspace cover area is 68.5% in the untreated watershed (Abdallah et al. 2020a). After cutting, juniper cover was estimated to account for 1% in the treated watershed with alterations in vegetation composition including western juniper regrowth and a significant increase of shrubs and grasses (Ray et al. 2019). The average slope for each watershed is approximately 25% with similar distribution of aspects (Fisher 2004).
The understory is dominated by Sandberg bluegrass (Poa secunda), Idaho fescue (Festuca idahoensis), prairie junegrass (Koeleria macrantha), bluebunch wheatgrass (Pseudoroegneria spicata) and Thurber’s needlegrass (Achnatherum thurberianum). In addition to A. tridentate, more shrubs such as rubber rabbitbrush (Ericameria nauseosa), antelope bitterbrush (Purshia tridentata) and green rabbitbrush (Chrysothamnus viscidiflorus) were common in the whole study site. The average annual precipitation (2009–17) at the study site was 358 mm (Ochoa et al. 2018). Soils in both watersheds are classified as Westbutte very stony loam, Madeline Loam and Simas gravelly silt loam, where Westbutte and Madeline define around 70%–74% of the study site and Simas makes up the rest with more soil series occupying <1% (Fisher 2004). Both Westbutte and Madeline series are formed of colluvium derived from basalt, tuff and andesite, whereas Simas is formed of loess and colluvium derived from tuffaceous sediments (Fisher 2004). The Westbutte series is classified as loamy-skeletal, mixed, superactive, frigid Pachic Haploxerolls. The Madeline series is classified as clayey, smectitic, frigid Aridic Lithic Argixerolls. The Simas series is classified as fine, smectitic, mesic Vertic Palexerolls. The juniper-dominated watershed is mainly made of 48% Madeline, 26% Westbutte and 21% Simas series, whereas the treated watershed consists of 50% Westbutte, 20% Madeline and 3% Simas series (Fisher 2004).
Experimental design
The two management practices (treatments) used for this study were: (i) untreated = juniper trees and all vegetation species were uncut and maintained in the untreated watershed and (ii) treated = juniper trees were manually removed with chainsaws to ground level in 2005 and other vegetation species were uncut in the adjacent treated watershed. The felled trees and debris because of juniper removal were scattered and left on the ground. Both watersheds were used for cattle grazing before and after juniper control. Twenty replicate plots, each of 20 m × 20 m were established in each watershed. The 20 plots were systematically randomized in a 4 × 5 grid trying to represent varying characteristics of slope and aspect within each watershed (Supplementary Fig. S1). The coordinates of the sampling locations were randomly generated, located on a digital map, and then the plots were found on the terrain with the aid of a GPS unit. The distance between plots within the predefined grid was 130 m among columns and 180 m among rows. The 20 m × 20 m plots were established to sample juniper trees. Subsequently, a 10 m × 10 m sub-plot within each juniper sampling plot (20 m × 20 m plot) was established to estimate shrub biomass (Supplementary Fig. S2). Additionally, four sub-sub-plots of 2 m × 2 m within each juniper sampling plot (20 m × 20 m plot) were established for grass and litter biomass determinations. In total, eighty 2 m × 2 m sub-sub-plots (4 sub-sub-plots × 20 plots) were established for grass and litter biomass evaluations in each watershed.
Nitrogen stocks: aboveground biomass and litter
Aboveground biomass for western juniper trees (mature in the untreated watershed and regrowth in the treated watershed), shrubs, grasses and litter in both watersheds was estimated (Abdallah et al. 2020a). Allometric equations were used to estimate the biomass (kg) of aboveground juniper trees based on canopy area data in each 20 m × 20 m plot of the untreated watershed. Aboveground biomass for regrowth juniper trees was determined by counting the number of present trees in each 20 m × 20 m plot of the treated watershed, clipping and collecting a representative tree sample, obtaining its dry weight at the laboratory, and extrapolating biomass weight by area. The biomass estimation procedure for regrowth juniper trees was also applied to estimate shrub biomass in both watersheds, except that for shrubs one representative individual was collected for each shrub species found within the 10 m × 10 m sub-plot. Grass aboveground biomass was estimated by harvesting all live standing tissue for dry matter analysis in all eighty 2 m × 2 m sub-sub-plots of each watershed. Litter biomass was estimated from the same sub-sub-plots that were used for grass sampling.
The nitrogen concentrations (%) for aboveground biomass were determined in five samples for western juniper, grasses, the main shrub species found in the area and litter. Each sample was dried at 60 °C until constant weight, finally ground, and analyzed for total nitrogen using a CNS automatic analyzer (Elementar Vario MMARCO CNS, Elementar Analysen Systeme GmbH, Hanau, Germany) at the Central Analytical Laboratory of the Crop and Soil Science Department at Oregon State University (Corvallis, OR). Proportions of aboveground biomass and nitrogen concentration for shrubs (foliage and main stem components) were estimated. Also, proportions of aboveground biomass and nitrogen concentration were estimated for different western juniper tree parts; boles, foliage, live branches >3 cm, live branches <3 cm and dead branches (Tiedemann and Klemmedson 2000).
Nitrogen stocks: root biomass
Root biomass using a trench method was measured for random stands of regrowth juniper, remaining tree stumps, shrubs and grasses in the treated watershed, while in the untreated watershed, it was measured for random stands of mature western juniper trees, shrubs and grasses (Abdallah et al. 2020a). For each vegetation type in each watershed, three trenches were made to a depth of 50 cm using an excavator (Bobcat Inc., West Fargo, North Dakota, USA). The trench width was 61 cm while the trench length was about 3 m (the exact length for each trench was measured and recorded). Root biomass was measured for the first 25 cm depth and then the subsequent 25 cm depth to evaluate the root variation. The roots from each trench and soil depth were carefully separated from the soil using a sieve, rinsed, oven-dried and weighed. Since it was hard to analyze shrub roots by species, root nitrogen for shrub was determined without separating shrubs species. Total nitrogen of roots was determined in a similar manner as aboveground biomass and with the same analyzer. For aboveground and belowground biomass, percent of total nitrogen was multiplied by each fraction’s mass to obtain the mass of total nitrogen per area.
Nitrogen concentrations: soil sampling
Soil sample collection and analysis procedures were conducted according to Abdallah et al. (2020a). Soil samples were obtained from areas 50 cm to the trunk (under the canopy of mature western juniper trees in the untreated watershed) and the stump (near stumps of cut western juniper trees in the treated watershed), and interspaces of both watersheds in all established plots (20 plots in each watershed) (Supplementary Fig. S1). Therefore, there were two sampling contexts in each plot, under-canopy/near stumps and interspace. Two soil samples were collected from each sampling context of each plot at two soil depths (0–25 cm and 25–50 cm). Each sample was sieved to avoid gravel or rocks, oven-dried at 40 °C for 48 h, weighed and analyzed using the same analyzer used for the vegetation pools.
Nitrogen stocks: soil bulk density and nitrogen calculation
A soil core ring with a radius of 2.5 cm was used to determine bulk density from four plots selected from the middle of the 20 plots in each watershed. One core sample was obtained in each zone of under-canopy of mature western juniper trees in the untreated watershed, near-stump of cut western juniper trees in the treated watershed, and in interspaces of both watersheds at two soil depths (0–25 cm and subsequent 25–50 cm). Soil core samples were oven-dried at 105 °C for 48 h and weighed. Bulk density was calculated as the ratio of the mass of oven-dried soil sample to core volume (g cm–3).
The soil nitrogen mass per area (kg N ha–1) was calculated using the following formula: Soil N (kg N ha–1) = % N/100 × BD (g cm–3) × D (cm) × 100 000 (conversion factor units from g cm-2 to kg ha-1), where % N = percentage nitrogen content of the sample, BD = soil bulk density in g cm–3 and D = soil depth (cm). Zonal areas of juniper canopy cover and interspaces were measured in a previous study (Abdallah et al. 2020a) in the same site. Zone-cover specific soil nitrogen stocks were estimated by multiplying mean soil nitrogen stock for each plot zone cover (under-canopy/near-stump and interspace zones for both watersheds) by the total areas of the relevant plot zone-cover. Then, the values for both zones were summed for each plot to obtain soil nitrogen stocks per plot. All nitrogen values for aboveground and belowground biomass were presented in units, kg N ha–1.
Statistical analysis
Differences in each analyzed variable were determined by single factor analysis of variance (ANOVA) with two-sample t-tests (treated vs. untreated). All analyses were performed with the R Statistical Software (R Core Team 2019). The number of sample replications by treatment varied by analyzed variable; for aboveground biomass of juniper and shrubs (n = 20), for grasses and litter (n = 80), for root biomass by plant type (n = 3), for soil nitrogen (n = 20), for soil bulk density (n = 4).
RESULTS
Nitrogen concentration of biomass samples
The purpose of obtaining the nitrogen concentrations of biomass samples was to calculate nitrogen stocks and they were not meant to test for nitrogen content differences among species. Nevertheless, mature juniper and shrubs (average) had (P < 0.05) more aboveground nitrogen concentration, and shrubs (average) and grasses had (P < 0.05) more root nitrogen concentration than the other plant types and/or litter (Table 1). For all shrubs, foliage including branches comprised on average about 73.0% of the aboveground nitrogen concentration and the main stems comprised the rest.
Biomass samples . | Aboveground nitrogen (%) . | Root nitrogen (%) . |
---|---|---|
Mature western juniper | 0.97(0.03)a | 0.41(0.09)b |
Regrowth western juniper | 0.79(0.05)b | 0.55(0.07)b |
Shrubs (average) | 1.03(0.11)a | 0.85(0.12)a |
Artemisia tridentata | 1.06(0.12) | |
Purshia tridentata | 1.31(0.10) | |
Eriogonum fasciculatum | 0.82(0.08) | |
Ericameria nauseosus | 0.92(0.03) | |
Grasses | 0.61(0.03)c | 0.97(0.09)a |
Western juniper stumps | — | 0.60(0.12)b |
Litter (mainly western juniper debris) | 0.62(0.12)c | — |
Biomass samples . | Aboveground nitrogen (%) . | Root nitrogen (%) . |
---|---|---|
Mature western juniper | 0.97(0.03)a | 0.41(0.09)b |
Regrowth western juniper | 0.79(0.05)b | 0.55(0.07)b |
Shrubs (average) | 1.03(0.11)a | 0.85(0.12)a |
Artemisia tridentata | 1.06(0.12) | |
Purshia tridentata | 1.31(0.10) | |
Eriogonum fasciculatum | 0.82(0.08) | |
Ericameria nauseosus | 0.92(0.03) | |
Grasses | 0.61(0.03)c | 0.97(0.09)a |
Western juniper stumps | — | 0.60(0.12)b |
Litter (mainly western juniper debris) | 0.62(0.12)c | — |
Root nitrogen analysis by species was not conducted on shrubs. Different lowercase letters (a, b, c) along columns indicate significant differences (P < 0.05) in nitrogen concentration among plant types and litter. For all live species (juniper, shrubs and grasses), the aboveground nitrogen analysis was made in leaves and stems or twigs. The number of sample replications by each analyzed variable (n = 5).
Biomass samples . | Aboveground nitrogen (%) . | Root nitrogen (%) . |
---|---|---|
Mature western juniper | 0.97(0.03)a | 0.41(0.09)b |
Regrowth western juniper | 0.79(0.05)b | 0.55(0.07)b |
Shrubs (average) | 1.03(0.11)a | 0.85(0.12)a |
Artemisia tridentata | 1.06(0.12) | |
Purshia tridentata | 1.31(0.10) | |
Eriogonum fasciculatum | 0.82(0.08) | |
Ericameria nauseosus | 0.92(0.03) | |
Grasses | 0.61(0.03)c | 0.97(0.09)a |
Western juniper stumps | — | 0.60(0.12)b |
Litter (mainly western juniper debris) | 0.62(0.12)c | — |
Biomass samples . | Aboveground nitrogen (%) . | Root nitrogen (%) . |
---|---|---|
Mature western juniper | 0.97(0.03)a | 0.41(0.09)b |
Regrowth western juniper | 0.79(0.05)b | 0.55(0.07)b |
Shrubs (average) | 1.03(0.11)a | 0.85(0.12)a |
Artemisia tridentata | 1.06(0.12) | |
Purshia tridentata | 1.31(0.10) | |
Eriogonum fasciculatum | 0.82(0.08) | |
Ericameria nauseosus | 0.92(0.03) | |
Grasses | 0.61(0.03)c | 0.97(0.09)a |
Western juniper stumps | — | 0.60(0.12)b |
Litter (mainly western juniper debris) | 0.62(0.12)c | — |
Root nitrogen analysis by species was not conducted on shrubs. Different lowercase letters (a, b, c) along columns indicate significant differences (P < 0.05) in nitrogen concentration among plant types and litter. For all live species (juniper, shrubs and grasses), the aboveground nitrogen analysis was made in leaves and stems or twigs. The number of sample replications by each analyzed variable (n = 5).
Western juniper components were just estimated to show the proportion of each tree component to the total aboveground biomass or nitrogen concentration, and they were not meant to test for nitrogen content differences between watersheds (Table 2). For mature western juniper, the biomass of boles and live Branches <3 cm were not significantly different (t = 1.11, P = 0.28) and comprised the greatest parts of the tree biomass (32.2% and 26.4%, respectively). The biomass of dead branches in mature juniper was the lowest component, accounting for 2.7% of the total aboveground biomass in it. Although it was not significantly different (P > 0.05) than live Branches <3 cm and boles, foliage was the major tree biomass component for regrowth western juniper. Concentrations of nitrogen were significantly (P < 0.05) greater in foliage than in other western juniper tree components in both watersheds (Table 2).
Tree component . | Mature western juniper . | . | Regrowth western juniper . | . |
---|---|---|---|---|
. | Biomass (t ha−1) . | Nitrogen (%) . | Biomass (t ha−1) . | Nitrogen (%) . |
Boles | 13.83(1.74)a | 0.16(0.01)c | 0.53(0.36)a | 0.08(0.0)c |
Foliage | 6.61(0.83)b | 0.42(0.01)a | 0.85(0.58)a | 0.52(0.03)a |
Live branches >3 cm | 7.55(0.95)b | 0.09(0.0)d | 0.0(0.0)c | 0.0(0.0)d |
Live branches <3 cm | 11.34(1.43)a | 0.18(0.01)b | 0.58(0.39)a | 0.14(0.01)b |
Dead branches | 1.15(0.15)c | 0.02(0.0)e | 0.02(0.01)b | 0.01(0.0)d |
Tree component . | Mature western juniper . | . | Regrowth western juniper . | . |
---|---|---|---|---|
. | Biomass (t ha−1) . | Nitrogen (%) . | Biomass (t ha−1) . | Nitrogen (%) . |
Boles | 13.83(1.74)a | 0.16(0.01)c | 0.53(0.36)a | 0.08(0.0)c |
Foliage | 6.61(0.83)b | 0.42(0.01)a | 0.85(0.58)a | 0.52(0.03)a |
Live branches >3 cm | 7.55(0.95)b | 0.09(0.0)d | 0.0(0.0)c | 0.0(0.0)d |
Live branches <3 cm | 11.34(1.43)a | 0.18(0.01)b | 0.58(0.39)a | 0.14(0.01)b |
Dead branches | 1.15(0.15)c | 0.02(0.0)e | 0.02(0.01)b | 0.01(0.0)d |
Different lowercase letters (a, b, c, d, e) along columns indicate significant differences (P < 0.05) in biomass or nitrogen concentration among tree components.
Tree component . | Mature western juniper . | . | Regrowth western juniper . | . |
---|---|---|---|---|
. | Biomass (t ha−1) . | Nitrogen (%) . | Biomass (t ha−1) . | Nitrogen (%) . |
Boles | 13.83(1.74)a | 0.16(0.01)c | 0.53(0.36)a | 0.08(0.0)c |
Foliage | 6.61(0.83)b | 0.42(0.01)a | 0.85(0.58)a | 0.52(0.03)a |
Live branches >3 cm | 7.55(0.95)b | 0.09(0.0)d | 0.0(0.0)c | 0.0(0.0)d |
Live branches <3 cm | 11.34(1.43)a | 0.18(0.01)b | 0.58(0.39)a | 0.14(0.01)b |
Dead branches | 1.15(0.15)c | 0.02(0.0)e | 0.02(0.01)b | 0.01(0.0)d |
Tree component . | Mature western juniper . | . | Regrowth western juniper . | . |
---|---|---|---|---|
. | Biomass (t ha−1) . | Nitrogen (%) . | Biomass (t ha−1) . | Nitrogen (%) . |
Boles | 13.83(1.74)a | 0.16(0.01)c | 0.53(0.36)a | 0.08(0.0)c |
Foliage | 6.61(0.83)b | 0.42(0.01)a | 0.85(0.58)a | 0.52(0.03)a |
Live branches >3 cm | 7.55(0.95)b | 0.09(0.0)d | 0.0(0.0)c | 0.0(0.0)d |
Live branches <3 cm | 11.34(1.43)a | 0.18(0.01)b | 0.58(0.39)a | 0.14(0.01)b |
Dead branches | 1.15(0.15)c | 0.02(0.0)e | 0.02(0.01)b | 0.01(0.0)d |
Different lowercase letters (a, b, c, d, e) along columns indicate significant differences (P < 0.05) in biomass or nitrogen concentration among tree components.
Aboveground nitrogen stocks
The greater aboveground biomass for the trees and grasses in the untreated watershed compared with the treated watershed (t = 7.6, P < 0.0001 and t = 2.9, P < 0.05, respectively) resulted in aboveground nitrogen stocks around 26 and 1.5 times more in trees and grasses of the untreated than the treated watershed (Table 3). However, the aboveground biomass for total shrubs (average) and litter was greater (t = 5.78, P < 0.0001 and t = 6.66, P < 0.0001, respectively) in the treated watershed than in the untreated watershed. Shrubs and litter had 6.5 and 6.2 times more nitrogen in the treated than the untreated watershed, respectively. All the main shrub species had greater (P < 0.05) aboveground nitrogen stocks in the treated watershed than in the untreated watershed, except for the shrub, E. nauseosus where aboveground nitrogen stocks were the same (t = 0.41, P = 0.68) between the two areas (Table 3). Since it was the dominant shrub species in both watersheds, A. tridentata stored 51% and 80% of the total shrub nitrogen in the untreated and treated watersheds, respectively.
Parameters/various components . | Biomass (t ha−1) . | . | P . | Biomass nitrogen (kg N ha−1) . | . | P . |
---|---|---|---|---|---|---|
. | Untreated . | Treated . | . | Untreated . | Treated . | . |
Trees | 43.0(5.41)a | 2.06(1.39)b | <0.0001 | 417.5(52.5)a | 16.3(11.0)b | <0.0001 |
Shrubs (average) | 0.25(0.05)b | 1.72(0.25)a | <0.0001 | 2.68(0.57)b | 17.4(2.62)a | <0.0001 |
Artemisia tridentata | 0.13(0.03)b | 1.31(0.24)a | <0.0001 | 1.37(0.30)b | 13.90(2.54)a | <0.0001 |
Purshia tridentata | 0.08(0.03)b | 0.18(0.03)a | <0.05 | 1.05(0.34)b | 2.36(0.39)a | <0.05 |
Eriogonum fasciculatum | 0.02(0.01)b | 0.12(0.02)a | <0.0001 | 0.16(0.16)b | 0.97(0.17)a | <0.0001 |
Ericameria nauseosus | 0.01(0.005)a | 0.01(0.003)a | 0.68 ns | 0.09(0.05)a | 0.07(0.02)a | 0.68 ns |
Grasses | 0.16(0.01)a | 0.11(0.01)b | < 0.05 | 0.98(0.09)a | 0.66(0.06)b | < 0.05 |
Litter | 0.69(0.11)b | 4.28(0.53)a | <0.0001 | 4.27(0.71)b | 26.51(3.26)a | <0.0001 |
Parameters/various components . | Biomass (t ha−1) . | . | P . | Biomass nitrogen (kg N ha−1) . | . | P . |
---|---|---|---|---|---|---|
. | Untreated . | Treated . | . | Untreated . | Treated . | . |
Trees | 43.0(5.41)a | 2.06(1.39)b | <0.0001 | 417.5(52.5)a | 16.3(11.0)b | <0.0001 |
Shrubs (average) | 0.25(0.05)b | 1.72(0.25)a | <0.0001 | 2.68(0.57)b | 17.4(2.62)a | <0.0001 |
Artemisia tridentata | 0.13(0.03)b | 1.31(0.24)a | <0.0001 | 1.37(0.30)b | 13.90(2.54)a | <0.0001 |
Purshia tridentata | 0.08(0.03)b | 0.18(0.03)a | <0.05 | 1.05(0.34)b | 2.36(0.39)a | <0.05 |
Eriogonum fasciculatum | 0.02(0.01)b | 0.12(0.02)a | <0.0001 | 0.16(0.16)b | 0.97(0.17)a | <0.0001 |
Ericameria nauseosus | 0.01(0.005)a | 0.01(0.003)a | 0.68 ns | 0.09(0.05)a | 0.07(0.02)a | 0.68 ns |
Grasses | 0.16(0.01)a | 0.11(0.01)b | < 0.05 | 0.98(0.09)a | 0.66(0.06)b | < 0.05 |
Litter | 0.69(0.11)b | 4.28(0.53)a | <0.0001 | 4.27(0.71)b | 26.51(3.26)a | <0.0001 |
The management practices are (i) Untreated (western juniper left uncut) and (ii) Treated (western juniper removed). The trees in the Treated management watershed are regrowth western juniper after 13 years of juniper removal. Different lowercase letters (a, b) within the same parameter along rows indicate significant differences (P < 0.05) between management practices for a given plant group or litter. ns = not significant. t ha−1= ton of biomass per hectare. kg N ha−1= kilogram of nitrogen per hectare. The number of sample replications by each analyzed variable; n = 20 for aboveground biomass of juniper and shrubs and n = 80 for grasses and litter.
Parameters/various components . | Biomass (t ha−1) . | . | P . | Biomass nitrogen (kg N ha−1) . | . | P . |
---|---|---|---|---|---|---|
. | Untreated . | Treated . | . | Untreated . | Treated . | . |
Trees | 43.0(5.41)a | 2.06(1.39)b | <0.0001 | 417.5(52.5)a | 16.3(11.0)b | <0.0001 |
Shrubs (average) | 0.25(0.05)b | 1.72(0.25)a | <0.0001 | 2.68(0.57)b | 17.4(2.62)a | <0.0001 |
Artemisia tridentata | 0.13(0.03)b | 1.31(0.24)a | <0.0001 | 1.37(0.30)b | 13.90(2.54)a | <0.0001 |
Purshia tridentata | 0.08(0.03)b | 0.18(0.03)a | <0.05 | 1.05(0.34)b | 2.36(0.39)a | <0.05 |
Eriogonum fasciculatum | 0.02(0.01)b | 0.12(0.02)a | <0.0001 | 0.16(0.16)b | 0.97(0.17)a | <0.0001 |
Ericameria nauseosus | 0.01(0.005)a | 0.01(0.003)a | 0.68 ns | 0.09(0.05)a | 0.07(0.02)a | 0.68 ns |
Grasses | 0.16(0.01)a | 0.11(0.01)b | < 0.05 | 0.98(0.09)a | 0.66(0.06)b | < 0.05 |
Litter | 0.69(0.11)b | 4.28(0.53)a | <0.0001 | 4.27(0.71)b | 26.51(3.26)a | <0.0001 |
Parameters/various components . | Biomass (t ha−1) . | . | P . | Biomass nitrogen (kg N ha−1) . | . | P . |
---|---|---|---|---|---|---|
. | Untreated . | Treated . | . | Untreated . | Treated . | . |
Trees | 43.0(5.41)a | 2.06(1.39)b | <0.0001 | 417.5(52.5)a | 16.3(11.0)b | <0.0001 |
Shrubs (average) | 0.25(0.05)b | 1.72(0.25)a | <0.0001 | 2.68(0.57)b | 17.4(2.62)a | <0.0001 |
Artemisia tridentata | 0.13(0.03)b | 1.31(0.24)a | <0.0001 | 1.37(0.30)b | 13.90(2.54)a | <0.0001 |
Purshia tridentata | 0.08(0.03)b | 0.18(0.03)a | <0.05 | 1.05(0.34)b | 2.36(0.39)a | <0.05 |
Eriogonum fasciculatum | 0.02(0.01)b | 0.12(0.02)a | <0.0001 | 0.16(0.16)b | 0.97(0.17)a | <0.0001 |
Ericameria nauseosus | 0.01(0.005)a | 0.01(0.003)a | 0.68 ns | 0.09(0.05)a | 0.07(0.02)a | 0.68 ns |
Grasses | 0.16(0.01)a | 0.11(0.01)b | < 0.05 | 0.98(0.09)a | 0.66(0.06)b | < 0.05 |
Litter | 0.69(0.11)b | 4.28(0.53)a | <0.0001 | 4.27(0.71)b | 26.51(3.26)a | <0.0001 |
The management practices are (i) Untreated (western juniper left uncut) and (ii) Treated (western juniper removed). The trees in the Treated management watershed are regrowth western juniper after 13 years of juniper removal. Different lowercase letters (a, b) within the same parameter along rows indicate significant differences (P < 0.05) between management practices for a given plant group or litter. ns = not significant. t ha−1= ton of biomass per hectare. kg N ha−1= kilogram of nitrogen per hectare. The number of sample replications by each analyzed variable; n = 20 for aboveground biomass of juniper and shrubs and n = 80 for grasses and litter.
Belowground nitrogen stocks
Root nitrogen stocks
Root nitrogen stocks for the uncut mature western juniper did not differ significantly (t = 1.24, P = 0.28) by depth (Table 4). However, root nitrogen stocks for western juniper stumps and regrowth trees in the treated watershed were 2.4 and 3.0 times, respectively, greater (t = 3.45, P < 0.05 and t = 4.19, P < 0.05) in the topsoil layer (0–25 cm depth) than in the bottom soil layer (25–50 cm depth). Root nitrogen stocks for shrubs in treated and untreated watersheds were 3.5 and 10.3 times, respectively, greater (t = 9.10, P < 0.001 and t = 15.77, P < 0.0001) in the topsoil layer (0–25 cm depth) than in the bottom soil layer (25–50 cm depth). Additionally, root nitrogen stocks for grasses in treated and untreated watersheds were about 19 and 35 times, respectively, greater (t = 5.11, P < 0.05 and t = 5.28, P < 0.05) in the top than in the bottom soil layer.
Parameters/various components . | Biomass (t ha−1) . | . | Biomass nitrogen (kg N ha−1) . | . |
---|---|---|---|---|
. | Untreated . | Treated . | Untreated . | Treated . |
0–25 cm soil depth | ||||
Intact mature juniper/stumps | 1.54(0.24)Ab | 3.02(0.37)Aa | 6.3(1.0)Ab | 18.1(2.3)Aa |
Regrowth juniper trees | — | 2.15(0.15)A | — | 11.8(0.8)A |
Shrubs | 3.51(0.17)Aa | 2.53(0.16)Ab | 29.8(1.5)Aa | 21.5(1.3)Ab |
Grasses | 1.51(0.28)Ab | 12.02(2.23)Aa | 14.6(2.7)Ab | 116.6(21.6)Aa |
25–50 cm soil depth | ||||
Intact mature juniper/stumps | 1.84(0.04)Aa | 1.25(0.35)Ba | 7.6(0.2)Aa | 7.5(2.08)Ba |
Regrowth juniper trees | — | 0.71(0.31)B | — | 3.9(1.7)B |
Shrubs | 0.34(0.10)Ba | 0.72(0.12)Ba | 2.9(0.9)Ba | 6.1(1.1)Ba |
Grasses | 0.04(0.01)Bb | 0.63(0.13)Ba | 0.4(0.1)Bb | 6.2(1.3)Ba |
Total: 0–50 cm soil depth | ||||
Intact mature juniper/ stumps | 3.38(0.20)a | 4.27(0.52)a | 13.9(0.8)b | 25.6(3.1)a |
Regrowth juniper trees | — | 2.85(0.44) | — | 15.7(2.4) |
Shrubs | 3.85(0.25)a | 3.25(0.22)a | 32.7(2.1)a | 27.6(1.9)a |
Grasses | 1.55(0.27)b | 12.66(2.24)a | 15.0(2.6)b | 122.8(21.7)a |
Parameters/various components . | Biomass (t ha−1) . | . | Biomass nitrogen (kg N ha−1) . | . |
---|---|---|---|---|
. | Untreated . | Treated . | Untreated . | Treated . |
0–25 cm soil depth | ||||
Intact mature juniper/stumps | 1.54(0.24)Ab | 3.02(0.37)Aa | 6.3(1.0)Ab | 18.1(2.3)Aa |
Regrowth juniper trees | — | 2.15(0.15)A | — | 11.8(0.8)A |
Shrubs | 3.51(0.17)Aa | 2.53(0.16)Ab | 29.8(1.5)Aa | 21.5(1.3)Ab |
Grasses | 1.51(0.28)Ab | 12.02(2.23)Aa | 14.6(2.7)Ab | 116.6(21.6)Aa |
25–50 cm soil depth | ||||
Intact mature juniper/stumps | 1.84(0.04)Aa | 1.25(0.35)Ba | 7.6(0.2)Aa | 7.5(2.08)Ba |
Regrowth juniper trees | — | 0.71(0.31)B | — | 3.9(1.7)B |
Shrubs | 0.34(0.10)Ba | 0.72(0.12)Ba | 2.9(0.9)Ba | 6.1(1.1)Ba |
Grasses | 0.04(0.01)Bb | 0.63(0.13)Ba | 0.4(0.1)Bb | 6.2(1.3)Ba |
Total: 0–50 cm soil depth | ||||
Intact mature juniper/ stumps | 3.38(0.20)a | 4.27(0.52)a | 13.9(0.8)b | 25.6(3.1)a |
Regrowth juniper trees | — | 2.85(0.44) | — | 15.7(2.4) |
Shrubs | 3.85(0.25)a | 3.25(0.22)a | 32.7(2.1)a | 27.6(1.9)a |
Grasses | 1.55(0.27)b | 12.66(2.24)a | 15.0(2.6)b | 122.8(21.7)a |
The management practices are (i) Untreated (western juniper [Juniperus occidentalis] left uncut) and (ii) Treated (western juniper removed). The comparison of mature trees included root determination near existing juniper trees (untreated watershed) and near juniper tree stumps (treated watershed). The evaluation of regrowth trees was only performed in the treated watershed. Different capital letters (A, B) within the same parameter along columns indicate significant differences (P < 0.05) between soil depths (0–25 vs. 25–50 cm) for a given management practice and plant type. Different lowercase letters (a, b) within the same parameter along rows indicate significant differences (P < 0.05) between management practices for a given plant type. ns = not significant. t ha−1 = ton of biomass per hectare. kg N ha−1 = kilogram of nitrogen per hectare. The number of sample replications by each analyzed variable (n = 3).
Parameters/various components . | Biomass (t ha−1) . | . | Biomass nitrogen (kg N ha−1) . | . |
---|---|---|---|---|
. | Untreated . | Treated . | Untreated . | Treated . |
0–25 cm soil depth | ||||
Intact mature juniper/stumps | 1.54(0.24)Ab | 3.02(0.37)Aa | 6.3(1.0)Ab | 18.1(2.3)Aa |
Regrowth juniper trees | — | 2.15(0.15)A | — | 11.8(0.8)A |
Shrubs | 3.51(0.17)Aa | 2.53(0.16)Ab | 29.8(1.5)Aa | 21.5(1.3)Ab |
Grasses | 1.51(0.28)Ab | 12.02(2.23)Aa | 14.6(2.7)Ab | 116.6(21.6)Aa |
25–50 cm soil depth | ||||
Intact mature juniper/stumps | 1.84(0.04)Aa | 1.25(0.35)Ba | 7.6(0.2)Aa | 7.5(2.08)Ba |
Regrowth juniper trees | — | 0.71(0.31)B | — | 3.9(1.7)B |
Shrubs | 0.34(0.10)Ba | 0.72(0.12)Ba | 2.9(0.9)Ba | 6.1(1.1)Ba |
Grasses | 0.04(0.01)Bb | 0.63(0.13)Ba | 0.4(0.1)Bb | 6.2(1.3)Ba |
Total: 0–50 cm soil depth | ||||
Intact mature juniper/ stumps | 3.38(0.20)a | 4.27(0.52)a | 13.9(0.8)b | 25.6(3.1)a |
Regrowth juniper trees | — | 2.85(0.44) | — | 15.7(2.4) |
Shrubs | 3.85(0.25)a | 3.25(0.22)a | 32.7(2.1)a | 27.6(1.9)a |
Grasses | 1.55(0.27)b | 12.66(2.24)a | 15.0(2.6)b | 122.8(21.7)a |
Parameters/various components . | Biomass (t ha−1) . | . | Biomass nitrogen (kg N ha−1) . | . |
---|---|---|---|---|
. | Untreated . | Treated . | Untreated . | Treated . |
0–25 cm soil depth | ||||
Intact mature juniper/stumps | 1.54(0.24)Ab | 3.02(0.37)Aa | 6.3(1.0)Ab | 18.1(2.3)Aa |
Regrowth juniper trees | — | 2.15(0.15)A | — | 11.8(0.8)A |
Shrubs | 3.51(0.17)Aa | 2.53(0.16)Ab | 29.8(1.5)Aa | 21.5(1.3)Ab |
Grasses | 1.51(0.28)Ab | 12.02(2.23)Aa | 14.6(2.7)Ab | 116.6(21.6)Aa |
25–50 cm soil depth | ||||
Intact mature juniper/stumps | 1.84(0.04)Aa | 1.25(0.35)Ba | 7.6(0.2)Aa | 7.5(2.08)Ba |
Regrowth juniper trees | — | 0.71(0.31)B | — | 3.9(1.7)B |
Shrubs | 0.34(0.10)Ba | 0.72(0.12)Ba | 2.9(0.9)Ba | 6.1(1.1)Ba |
Grasses | 0.04(0.01)Bb | 0.63(0.13)Ba | 0.4(0.1)Bb | 6.2(1.3)Ba |
Total: 0–50 cm soil depth | ||||
Intact mature juniper/ stumps | 3.38(0.20)a | 4.27(0.52)a | 13.9(0.8)b | 25.6(3.1)a |
Regrowth juniper trees | — | 2.85(0.44) | — | 15.7(2.4) |
Shrubs | 3.85(0.25)a | 3.25(0.22)a | 32.7(2.1)a | 27.6(1.9)a |
Grasses | 1.55(0.27)b | 12.66(2.24)a | 15.0(2.6)b | 122.8(21.7)a |
The management practices are (i) Untreated (western juniper [Juniperus occidentalis] left uncut) and (ii) Treated (western juniper removed). The comparison of mature trees included root determination near existing juniper trees (untreated watershed) and near juniper tree stumps (treated watershed). The evaluation of regrowth trees was only performed in the treated watershed. Different capital letters (A, B) within the same parameter along columns indicate significant differences (P < 0.05) between soil depths (0–25 vs. 25–50 cm) for a given management practice and plant type. Different lowercase letters (a, b) within the same parameter along rows indicate significant differences (P < 0.05) between management practices for a given plant type. ns = not significant. t ha−1 = ton of biomass per hectare. kg N ha−1 = kilogram of nitrogen per hectare. The number of sample replications by each analyzed variable (n = 3).
The root biomass (0–50 cm soil depth) for mature western juniper trees (intact vs. stumps) and shrubs were not significantly different (P > 0.05) between untreated and treated watersheds (Table 4). Although the root biomass for the mature trees was the same (t = 1.58, P = 0.19) in both watersheds, the stump roots in the treated watershed stored about 1.8 more nitrogen compared with the uncut tree roots in the untreated watershed. The stump roots in the treated watershed had 1.5 times more nitrogen concentration compared with the intact tree roots in the untreated watershed (Table 1). It was clear that the root nitrogen stock for regrowth juniper on the treated watershed was a little higher (13%) than in mature juniper. Root nitrogen stocks for the shrubs were not significantly different (t = 1.80, P = 0.15) between the watersheds (Table 4). The greater grass biomass in the treated than untreated watershed resulted in grass root nitrogen stocks eight times more (t = 4.92, P < 0.05) in the treated in comparison to the untreated watershed.
Soil nitrogen stocks
There were significant soil bulk density changes by depth (0–25 cm vs. 25–50 cm) among the corresponding zones of the watersheds, except for the near-stump zone in the treated watershed where soil bulk density was not significantly different (t = 0.21, P = 0.84) by depth (Fig. 1a and b). Across all zones, soil bulk density was significantly greater (P < 0.05) in the bottom soil layer of interspace zone in the treated watershed. In both soil layers, soil nitrogen concentrations (%) were significantly greater (P < 0.05) under mature western juniper tree canopies of the untreated watershed and near western juniper tree stumps of the treated watershed than the corresponding interspaces in each watershed (Fig. 1c and d).
Soil nitrogen stocks in under-canopy and near-stumps zones of western juniper were approximately 50% and 45% significantly greater (P < 0.05) in the top (0–25 cm) than in the bottom soil layer (25–50 cm) for untreated and treated watersheds, respectively (Fig. 2a and b). Also, the interspace areas of both watersheds contained significantly more (P < 0.05) soil nitrogen stocks in the top than in the lower soil layer. On the whole soil depth (0–50 cm), under-canopy zones of mature western juniper and near-stump zones of cut western juniper contained approximately 1.7 times more soil nitrogen than the interspaces (Fig. 2c). Soil nitrogen stocks were not significantly (t = 0.42, P = 0.68) different under the canopies of mature western juniper and near-stump zones of cut western junipers. Similarly, soil nitrogen stocks did not change (t = 0.32, P = 0.75) in interspaces of both management practices. Considering the areas of zone cover, total soil nitrogen stocks were not different (P > 0.05) between the watersheds or across the zones (Fig. 3).
Total nitrogen stocks by management practice
Total aboveground nitrogen and root nitrogen stocks differed by watershed (P < 0.05), with the value from untreated watershed was 6.9 times more and 3.1 times less than that from treated one (Fig. 4). Soil nitrogen stocks were not significantly (t = 0.30, P = 0.76) different between both watersheds. Contrary to the total aboveground nitrogen stocks, total belowground nitrogen stocks (root and soil) did not differ (t = 0.43, P = 0.67) by watershed. In addition, total belowground nitrogen stocks were 21 and 153 times greater than their corresponding total aboveground nitrogen stocks in the untreated and treated watersheds, respectively. Total nitrogen stocks, including both total belowground and aboveground nitrogen, did not differ significantly (t = 0.08, P = 0.94) between the management practices of untreated and treated watersheds.
DISCUSSION
Effects of management practices on aboveground nitrogen stocks
After 13 years of juniper removal in the treated watershed, we estimated the mean aboveground biomass in the untreated watershed to be 44.1 t ha−1, which was approximately 5.4 times greater than that of the adjacent treated watershed (8.17 t ha–1). The total aboveground biomass is comparable to that of oak-pine forest which was estimated to be about 61.9 t ha–1 including both overstory and understory biomass (Hubbard et al. 2004). Western juniper trees comprised approximately 97% of the total aboveground biomass in the untreated watershed and 25% in the treated watershed. This resulted in the aboveground nitrogen to be substantially greater in the untreated watershed compared with the treated watershed, supporting our hypothesis. Cutting mature western juniper in the treated watershed caused the aboveground nitrogen to decline considerably compared with the untreated watershed where the uncut mature juniper accumulated the largest size of aboveground nitrogen. Thirteen years after control, the regrowth juniper trees were already many on the treated watershed, but their aboveground nitrogen accumulation was 16.3 kg N ha–1 and only 4% of that of mature western juniper. Cutting juniper caused shrub species mainly A. tridentate, P. tridentate and E. fasciculatum to establish and increase in biomass in the treated than the untreated watershed. The greater shrub biomass in the treated watershed translated into greater biomass nitrogen compared with the untreated watershed. Other studies reported a rapid increase in shrub presence following juniper cutting in different areas of Oregon (Bates et al. 2017; Dittel et al. 2018; Ray et al. 2019).
Additionally, litter accumulation in the treated watershed, contained on average 26.5 kg N ha–1, was substantially greater than the untreated watershed because of the juniper cut-and-leave operation. Aboveground juniper litter in the cut woodland is of low quality (high carbon:nitrogen ratio, 77:1 and 240:1 for leaves, and the twigs and branches, respectively) and decomposes very slowly (Miller et al. 2005; Wall et al. 2001). Litter nitrogen in the treated watershed was the greatest pool, accountable for 43% of the total aboveground nitrogen in it. Unpredictable results were that grass biomass and nitrogen stocks were lower in the treated than in the untreated watershed. In western juniper and pinyon-juniper woodlands, increases in grass presence following juniper cutting have been reported (Bates et al. 2016, 2017; Dittel et al. 2018; Ray et al. 2019). It is likely that the decreased grass biomass accumulation at the treated watershed was attributed to uneven cattle grazing, excessive on the treated watershed, during the year of measurements. The total aboveground nitrogen in the treated watershed, estimated to be 61.9 kg N ha–1, is around 50% higher than 30 kg N ha–1 measured for grassland area adjacent to Juniperus virginiana forest encroached area (McKinley and Blair 2008).
Mature western juniper-dominated area, untreated watershed stored relatively considerable quantities of aboveground nitrogen, even with the positive response of shrubs and the litter accumulation resulting from juniper cutting in the treated watershed. The greater total aboveground nitrogen stocks found in the untreated watershed were within the ranges of aboveground nitrogen stocks (370–870 kg N ha–1) measured in areas encroached by the shrub Cornus drummondii (Lett et al. 2004); and J. virginiana (McKinley and Blair 2008; Norris et al. 2007). The contribution of shrubs, grasses and litter to total aboveground nitrogen stocks in the untreated watershed was insignificant in comparison to mature western juniper trees. Combined, the total aboveground nitrogen stock of those three pools was 7.93 kg N ha–1, representing 1.9 % of the tree contribution in the untreated watershed.
Effects of management practices on belowground nitrogen stocks
Distribution of shrub and grass root biomass was not evenly along soil depths (0–25 vs. 25–50 cm), with significantly greater biomass distribution in the topsoil layer than the bottom soil layer. Thus, there is a decline in understory (grasses and shrubs) root nitrogen stocks with increasing soil depth in both watersheds. While shrubs have a greater root depth in the upper 20–30 cm of the soil profile, grasses also have a dense fibrous root system of this limited depth, well suited to exploit soil resources, where moisture and nutrients are at peak concentrations (Briske 2017). Contrary to shrubs and grasses, root nitrogen stocks for mature juniper trees in the untreated watershed were more consistently observed with depth down to 50 cm. Western juniper trees have extensive root systems (Mollnau et al. 2014) with largest distribution reaching as deep as 50 cm soil depth (Young et al. 1984). In the treated watershed, root nitrogen for the stump and regrowth juniper did not distribute evenly along depths with concentrations in the upper soil layer, approximately 70%–75% greater than the bottom soil layer. Juniper root biomass in the treated watershed was not equally proportioned along soil depths, with substantially greater biomass proportion in the topsoil layer compared with the bottom soil layer.
Regardless of soil depth, this study demonstrated a considerable increase (3.1-fold) in root nitrogen stocks 13 years after juniper cutting in the treated watershed. The increase was predominately attributed to the proliferation in grass root biomass after juniper cutting, even with the small decline in grass aboveground biomass. Prior research has shown greater grass biomass production following juniper cutting (Bates et al. 2017; Dittel et al. 2018; Ray et al. 2019), yet grass root responses have not been investigated before. Within the treated watershed, grass roots compared with the combined juniper and shrub roots were the greatest nitrogen pools, accounting for 64% of the total root nitrogen stocks in it. Because of their high root/shoot ratio and greater volume of rhizosphere, perennial grasses can promote microbial activity, enhance nitrogen mineralization and accumulate more nitrogen belowground (Gupta et al. 2014; Sainju et al. 2017). The direct availability of the fixed nitrogen to the roots contributes to reduced nitrogen losses from leaching (Crews and Peoples 2005; Gupta et al. 2014) and reduces the risk of nitrous oxide emissions by denitrification (IPCC 2013). In an earlier study in the same site, 4760 kg carbon stocks per hectare for grass roots were calculated from the treated watershed (Abdallah et al. 2020a). In grassland ecosystems, the root carbon-to-nitrogen (C/N) ratio is an important parameter regulating root decomposition (Fornara et al. 2020; Vivanco and Austin 2006). The higher C/N ratio of 38.8 for grass roots as calculated from the present study and Abdallah et al. (2020a) may result in a slower rate of root decomposition, leading to lower carbon losses and greater rates of soil carbon accumulation through time.
Soil nitrogen was considerably held in the topsoil layer in both under-canopy and near-stumps zones of western juniper. The increase in soil nitrogen at the topsoil layer (0–25 cm depth) was likely due to the fact that the greatest concentrations of soil organic matter inputs from litterfall are allocated in the top layer (0–30 cm soil depth) (Jobbágy and Jackson 2000). Soil nitrogen in the topsoil layers could increase because of more intensive exploration of the soil by woody plant roots (Boutton and Liao 2010). The deep-rooted woody plants can uplift and concentrate nutrients from deeper to shallow soil horizons (Jobbagy and Jackson 2000, 2004). Thus, the effect of western juniper trees on soil nitrogen is more pronounced in the topsoil layer. The interspace areas in both watersheds also showed difference even though small in nitrogen stocks with soil depth. Our analysis of total nitrogen stocks as impacted by juniper cutting covered the 0–50 cm soil layer since juniper roots may control changes at that depth (Young et al. 1984). Nevertheless, there were no differences in soil nitrogen stocks because of juniper clearing (between the management practices of treated and untreated watersheds) at either the 0–25 cm or the 25–50 cm soil layers.
Regardless of zonal cover areas, nitrogen was more concentrated (1.7-fold) in soils under-canopy and near-stump zones of cut western juniper than their corresponding interspace zones in both watersheds. Greater soil fertility in juniper mounds compared with interspaces (Davenport et al. 1996; Kramer and Green 1999) may contribute to the higher soil nitrogen under-canopy zones of western juniper. Young et al. (2014) found that tree mounds of Utah juniper (Juniperus osteosperma Torr.) had greater concentrations of total nitrogen than adjacent interspaces. Our findings agree with previous studies that determined greater concentrations of total nitrogen in soils under woody canopies relative to intercanopy soils (Law et al. 2012; Michaelides et al. 2012; Miwa and Reuter 2010; Turnbull et al. 2010; Turpin-Jelfs et al. 2019; Wheeler et al. 2007). In addition, our findings show that soil nitrogen near-stump zones of cut western juniper stayed raised even 13 years post tree cutting and did not differ from soil nitrogen under-canopy zones of mature western juniper of the untreated area. Western juniper resource islands can persist for 15 years following canopy removal because of deep mounds (Miwa and Reuter 2010). A study measuring nutrient availability for western juniper in Oregon had found that soil inorganic nitrogen in canopy zones, while decreasing overtime, remained elevated with small differences between cut and uncut treatments lasting into the sixth-year posttreatment (Bates and Davies 2017).
Total soil nitrogen stocks at 50 cm depth, including both under-canopy and interspace areas, remained unchanged 13 years post juniper control with an average of 9126 kg N ha-1. Our findings did not support our hypothesis of lower soil nitrogen post juniper cutting. Unchanged total soil nitrogen between forest and grazing lands was reported in Ethiopia (Yimer et al. 2006, 2007). Our results contrasted with observations that reported lower soil nitrogen in un-encroached grasslands relative to woody-encroached sites (Berihu et al. 2017; Dahl et al. 2020; Li et al. 2019; Mureva et al. 2018; Thomas et al. 2018). Although not statistically different, soil nitrogen in the interspace zone of the treated watershed was about 1.3 times greater than that of under-canopy/near-stump zones, which may be explained by the increased plant uptake of nitrogen in mature western juniper compared with the interspaces. Juniper removal reduces competition for belowground resources and releases additional soil water (Abdallah et al. 2020b; Mata-González et al. 2021; Zhong et al. 2020) and nitrogen for uptake by understory plants (Bates et al. 2000), which was indicated by the higher aboveground biomass nitrogen of understory (combined grasses and shrubs, excluding juniper) and higher root nitrogen in the treated compared with the untreated watershed.
Total nitrogen stocks and management practices
Both management practices (untreated and treated watersheds) had the same total nitrogen accumulations with an average nitrogen level of 9496 kg N ha–1. Despite substantial increases in aboveground nitrogen with western juniper encroachment in the untreated watershed, we did not find significant changes in belowground nitrogen pools (0–50 cm soil depth) between both watersheds. These findings agree with our hypothesis partially, indicating that the treated watershed contained less aboveground nitrogen compared with the untreated watershed. According to Hughes et al. (2006), the stand development of Prosopis glandulosa (honey mesquite) increased total aboveground nitrogen including P. glandulosa biomass, herbaceous biomass and litter (~1860 kg N ha–1), but with no significant change in the upper 10 cm soil depth accompanying the encroachment in a temperate savanna. Our findings contrast with other observations where less ecosystem nitrogen stocks were stored in native compared with invaded ecosystems (Liao et al. 2008; Norris et al. 2007). In this study, the increase of understory plant roots resulted from western juniper removal contributed to a partial offset to the losses of aboveground nitrogen in the treated watershed. Greater root biomass enhances nitrogen cycling and accumulation (Fornara and Tilman 2008).
In both management practices, most of the total nitrogen pool was contained belowground (roots and soil). In this study, 95% and 99% of the total nitrogen stocks in the untreated and treated watershed, respectively, are allocated belowground (0–50 cm soil depth) as compared with the aboveground nitrogen pools (Fig. 4). Our findings are approximate to those of Sharrow and Ismail (2004) who showed that about 99% of the total nitrogen stocks were contained belowground (0–45 cm depth) for both agroforest and grasses-dominated pastures. Moreover, an encroached area by J. virginiana stands was reported to contain 89% of total ecosystem nitrogen belowground (0–10 cm soil depth) (McKinley and Blair 2008). Our results indicated that the single largest pool of nitrogen in the CCPWS site was the soil, accounting for 95% (untreated watershed) and 97% (treated watershed) of the total ecosystem nitrogen stocks. Therefore, any anthropogenic activities that might have adverse effects on soils will have significant implications in reducing nitrogen stocks in western juniper systems. Degradation of the belowground nitrogen pool will eventually lead to decline of total nitrogen accumulation in the system.
Many encroaching woody plants fix nitrogen symbiotically (Eldridge et al. 2011; Sitters et al. 2013) and thus have the capacity to accumulate more nitrogen into the ecosystem (Boutton and Liao 2010). Because of their deep root systems (Miller et al. 2005) and high transpiration rates (Abdallah et al. 2020b; Mata-González et al. 2021) juniper trees may promote hydraulic redistribution, thereby favoring nitrogen uptake as observed in other woody plants (Sardans and Peñuelas 2014). When woody plants are removed, nitrogen mineralization and immobilization increase immediately, and plant uptake decreases (Vitousek 1981). According to Bates et al. (2002), western juniper removal led to nitrogen loss from soils because of the buildup of potassium chloride (KCl) extractable nitrogen levels during the first-year post juniper removal. However, by the second-year post removal there were few differences in either extractable nitrogen pools or net nitrogen mineralization among zonal treatments (cut intercanopy, cut debris, uncut intercanopy and uncut canopy). In this study, therefore, the potential for nitrogen losses because of western juniper control may be counterbalanced by the ability of the understory to respond to cutting.
Juniper removal is associated with several ecological advantages like habitat restoration for native wildlife, released understory plants, greater soil moisture and restoration of watershed hydrological functions (Abdallah et al. 2020b; Baruch-Mordo et al. 2013; Dittel et al. 2018; Ochoa et al. 2018; Ray et al. 2019). This study demonstrates that the advantages of juniper control can be achieved without considerably disturbing nitrogen pools of these systems.
CONCLUSIONS
This study assessed changes in ecosystem nitrogen stocks in response to western juniper control in a semiarid rangeland ecosystem of central Oregon, USA. We compared the ecosystem biomass and nitrogen allocation in an uncut juniper area (untreated watershed) with that in an adjacent cut juniper area (treated watershed) where most mature juniper trees were removed 13 years ago. Thirteen years post juniper removal, we observed that the aboveground biomass was greater in the untreated watershed and the belowground biomass was greater in the treated watershed. As a result, the aboveground nitrogen stock declined in the treated western juniper area, but the belowground nitrogen stock did not, resulting in no significant changes in total nitrogen stocks (aboveground and belowground). A greater root nitrogen accumulation in the treated area than in the untreated area partially compensated the losses in aboveground nitrogen caused by juniper cutting.
Ecosystem nitrogen was dominated by the belowground pool in both management practices. The ecosystem nitrogen stock substantially accumulated belowground (over 95%). Thus, changes in the 5% aboveground biomass are not as important in the short term. However, this 13-year post treatment research, is still of short duration to explore soil nitrogen changes. More research of longer duration about effects of western juniper encroachment and control on soil nitrogen is needed to further support findings from this study. Protecting belowground nitrogen source is of great importance. This study concludes that western juniper removal did not impact belowground nitrogen pools, at least 13 years post control. Thus, it is evident that western juniper removal can help to restore understory plants and improve hydrological functions of systems while maintaining the potential for ecosystem nitrogen accumulation.
Supplementary Material
Supplementary material is available at Journal of Plant Ecology online.
Figure S1: Location of the study site, showing both watersheds and their plots layout used in this study.
Figure S2: Plot sizes established in the study for biomass evaluations.
Funding
This research was supported by Oregon State University and the Oregon Agricultural Experiment Station.
Acknowledgements
We thank David Prado-Tarango and Keira Mitchell for their assistance with the fieldwork. We are grateful to Dr Rick Martinson for facilitating the use of heavy equipment for root sampling. We gratefully acknowledge the continuous support of the Hatfield High Desert Ranch in this research effort.
Conflict of interest statement. The authors declare that they have no conflict of interest.
REFERENCES
Abdallah MAB, Durfee N, Mata-González R, et al. (