Effects of post oak (Quercus stellata) and smooth brome (Bromus inermis) competition on water uptake and root partitioning of eastern redcedar (Juniperus virginiana)

Abstract

Eastern redcedar Juniperus virginiana is encroaching into new habitats, which will affect native ecosystems as this species competes with other plants for available resources, including water. We designed a greenhouse experiment to investigate changes in soil moisture content and rooting depths of two-year-old J. virginiana saplings growing with or without competition. We had four competition treatments: 1) none, 2) with a native tree (Quercus stellata), 3) with an invasive grass (Bromus inermis), and 4) with both Q. stellata and B. inermis. We measured soil moisture content over two years as well as root length, total biomass, relative water content, midday water potential, and mortality at the end of the experiment. When J. virginiana and B. inermis grew together, water depletion occurred at both 30–40 cm and 10–20 cm. Combined with root length results, we can infer that J. virginiana most likely took up water from the deeper layers whereas B. inermis used water from the top layers. We found a similar pattern of water depletion and uptake when J. virginiana grew with Q. stellata, indicating that J. virginiana took up water from the deeper layers and Q. stellata used water mostly from the top soil layers. When the three species grew together, we found root overlap between J. virginiana and Q. stellata. Despite the root overlap, our relative water content and water potential indicate that J. virginiana was not water stressed in any of the plant combinations. Regardless, J. virginiana saplings had less total biomass in treatments with B. inermis and we recorded a significantly higher mortality when J. virginiana grew with both competitors. Root overlap and partitioning can affect how J. virginiana perform and adapt to new competitors and can allow their co-existence with grasses and other woody species, which can facilitate J. virginiana encroachment into grasslands and woodlands. Our data also show that competition with both Q. stellata and B. inermis could limit establishment, regardless of water availability.

Introduction

Plants have mechanisms to avoid and minimize competition with other plant species [1, 2]. Much competition between plants occurs belowground [3, 4]. An important belowground factor affecting plant distribution is water availability [3]. For example, plants with different root-distribution patterns (e.g., root length) might be able to coexist in the same habitat if their roots can occupy different spaces and are able to exploit resources from distinct areas [5, 6]. A major mechanism to avoid water stress and enhance the competitive advantage of a plant is its ability to exploit deeper stores of soil water [7]. Water availability and the competitive abilities of plants for water uptake has long been recognized as a fundamental driver of vegetation and landscape structure [8, 9], as it has the potential to shift community composition based on the water uptake strategies by different species.

The differentiation of root systems has been thought of as a mechanism that reduces competition and facilitates species coexistence and can be used to understand the spatial distribution of plants and plant coexistence [10–12]. Competition for water is considered to occur by water limitation or increase in water demand [13]. However, competition for water has been less frequently studied than other factors such as nutrients or light [13]. Water availability is a particularly important factor in the survival and establishment of seedlings and saplings [14, 15]. The impact of soil-moisture depletion on plants is expressed in diminished plant size, decreased biomass [16], and lower water potential [17]. Moreover, plants growing in water-limited soils are more prone to hydraulic failure and drought-induced mortality [17–19]. Water limitation can also affect a plant’s nutrient uptake [20]. This will result in a decline in available resources for the different parts of the plant, including leaves and roots [21].

Niche differentiation, predicted by the limiting similarity hypothesis [6], is well known to occur when plant species occupy different environmental niche axes, including soil moisture and root depth [6, 22–24]. In one of the most prominent theories of niche separation between trees and grasses, Walter (1939) [10] proposed the two-layer hypothesis as an equilibrium explanation for the coexistence of savanna trees and grasses. Trees and grasses have very distinct root lengths and different water-use strategies [25]. Walter (1939) [10] proposed that grasses predominate over trees in savannas because grasses have shallow roots, allowing them to use water efficiently and take advantage of the available rainfall [25]. In both savannas and grasslands, grasses are considered superior competitors for water in the upper soil profile [23] due to their shallow, dense rooting system with high surface area. The presence of trees in a system can alter soil water content, affecting all associated plants. Trees have access to deeper water because of their extensive rooting systems [8, 24, 26]. In grasslands, however, trees and grasses may use overlapping areas in the soil profile, especially while seedlings and saplings are growing their roots through the soil profile [22, 27]. One aspect that Walter’s (1939) two-layer hypothesis [10] did not focus on is how the roots of two or more tree species interact to allow these trees to coexist [12]. Woody plants can have different water uptake patterns [28]. A few studies comparing the depth of water uptake by co-occurring woody species reported some tree species use only deep soil water, while others can use available water from both shallow and deep soil layers [11, 24, 29].

Eastern redcedar (Juniperus virginiana) is an example of a tree encroaching into grasslands and savannas, specifically in eastern and central US, including grasslands in the Great Plains (grassland located in the interior of North America) [30, 31], and midwestern prairies as far west as Nebraska [32, 33]. J. virginiana is the most widespread conifer in the eastern United States [33]. J. virginiana is known to be drought-tolerant [34] and withstands extremes of drought due to its extensive root system [35], which is both deep and highly branched, potentially giving this tree a competitive advantage in water acquisition over other species [29]. The advantage in terms of drought tolerance of J. virginiana makes it a stronger competitor because it needs to take up less water to maintain the same productivity compared to less drought-tolerant species [31, 36–38]. This has important implications for regional ecohydrology [39]. J. virginiana appears to be suited for survival in a drought-prone regions [40, 41], provided sufficient soil water while growing deep roots in early life stages which could fundamentally alter the soil ecosystem water-use [39, 40]. However, it is not just woody plants but also grasses [41] may outcompete trees for shallow soil water when their rooting zones overlap [42]. Bromus inermis is an invasive deeply rooting rhizomatous perennial grass that is found in many grasslands and old fields [43] and is a potential competitor for J. virginiana. An additional potential competitor for J. virginiana are Quercus spp. J. virginiana is encroaching in areas where oak (Quercus) trees, such as Quercus stellata, are dominant [31, 34]. For example, J. virginiana encroached into the Cross Timbers (western edge of the eastern deciduous forest of the U.S.), transforming Q. stellata-dominated forests to J. virginiana and Q. stellata co-dominant woodlands [31]. Q. stellata may be a strong competitor for J. virginiana because Q. stellata also has taproots that can grow throughout the upper several meters of soil in search of moisture and nutrients [44].

Identifying spatial patterns of soil water depletion when plants are growing alone or in competition with other species can help us understand species coexistence and competition for water. Especially during early stages of establishment, J. virginiana should transition from shallow roots to deeper roots, which should affect its ability to compete with grasses compared to woody plants [29]. B. inermis should rely on water from the top soil layers, thus, J. virginiana should experience more water stress when establishing alongside B. inermis. However, J. virginiana could avoid low water potentials by developing an extensive root system that gives it access to deeper soil moisture. Similarly, water stress for establishing J. virginiana should increase with additional competition with Q. stellata, but the potential of root partition to ameliorate competition with woody plants is more difficult to predict because they also develop deep root systems. In this study we sought to determine the rooting depth and changes in water availability and water status for J. virginiana both growing alone and in competition with B. inermis and/or Q. stellata. We conducted a greenhouse experiment and used frequency domain reflectometry (FDR) to measure water availability at different depths. We also recorded J. virginiana root length, total biomass, and mortality and examined J. virginiana water status by measuring midday leaf water potential and relative water content. We further tested Walter’s (1939) two-layer hypothesis [10] by focusing on asking whether differentiation in the rooting depth of co-occurring species can allow them to co-exist. We made four predictions based on Walter’s (1939) two-layer hypothesis [10]:

1. J. virginiana and B. inermis interactions will lead to water uptake at different depths.

2. There will be niche separation between J. virginiana and Q. stellata roots [31].

3. Soil moisture will be consistently low throughout the soil profile when all three species (J. virginiana, Q. stellata, B. inermis) co-occur, leading to lower water potential for J. virginiana.

4. We expect that the water requirements for J. virginiana, Q. stellata, and B. inermis will increase with time as the plants grow, leading to more intense water stress in the second year of the experiment.

Materials and methods

Ethics statement

The plant materials used in this experiment were purchased from commercial nurseries. J. virginiana saplings were obtained from Pinelands Nursery & Supply, New Jersey, Q. stellata saplings were obtained from Mossy Oak Nativ Nurseries, Mississippi, and B. inermis seeds were obtained from Great Basin Seed company, Utah. The experiment did not involve any endangered or protected species, so no specific permissions were required for the collection and use of plant materials. The experimental research on plants, including the collection of plant material, complied with relevant institutional, national, and international guidelines and legislation. All methods were performed in accordance with the relevant guidelines and regulations.

Experimental design

We conducted a two-year greenhouse experiment, from September 2017 to August 2019, to analyze changes in soil moisture content, rooting depths, total biomass, water stress, and mortality for J. virginiana growing alone and in competition with Q. stellata, and/or B. inermis. We used 95 L containers (55 cm diameter x 70 cm height) (hereafter pots) filled with PRO-MIX^® All-Purpose Professional Grower’s Potting Mix (BFG Supply). PRO-MIX^® has a medium-high moisture retention and a moisture content of 30–50% by weight. We used a completely randomized design in which we grew J. virginiana in four treatments (no competition, hereafter JUVI, Q. stellata competition, hereafter QUST, B. inermis competition, hereafter BRIN, and Q. stellata and B. inermis competition, hereafter QUST+BRIN). We used 16 pots for each treatment. J. virginiana saplings (two-year-old, one sapling in each pot, mean initial height ± S.E. = 226±6.6 mm) were planted first, followed by the Q. stellata (one-year-old, one sapling in each of the QUST and QUST+BRIN treatments, mean initial height ± S.E. = 319±8.3 mm) and B. inermis two weeks after J. virginiana was planted, for a total of 64 pots. In the treatments with B. inermis, we planted 200 grass seeds/pot with a germination rate of 97%.

We grew the plants in a controlled environment (average temperature = 25°C and four hours of supplemental light in winter). All pots were well watered and irrigated for approximately 10 min every other day through a drip-irrigation system at a rate of 40 ml/min. The same amount of water was supplemented in both years and for all treatment combinations. However, in winter, the plant’s growth is slower, and they needed less water to survive. As a result, we irrigated the pots in winter for 5 min every other day through a drip-irrigation system. We monitored soil moisture throughout the experiment to ensure proper drainage to the deeper soil layers. Prior to planting the saplings, we installed a 1 m-long 50 mm-diameter polycarbonate access tube in the middle of every pot to measure moisture content in the soil profile using frequency-domain reflectometry (see below).

Soil moisture content

We used a Diviner 2000^® soil-moisture probe (Sentek Pty Ltd, Stepney, Australia) to obtain moisture content readings every 10 cm in the soil profile [22]. The Sentek system uses frequency-domain reflectometry (FDR). FDR measures soil water content at different depths indirectly by measuring the frequency variations of an electromagnetic pulse propagated into the soil. The difference between the output wave and the return wave frequency is measured to determine moisture content every 10 cm in profile. Factory settings and calibration, to the potting mix used in this experiment and following Sentek recommendations, were used to derive percentage water (content θ%) at each depth. The access tubes, installed before planting, were used to measure moisture content in the soil profile. Soil moisture content was recorded every other week.

Plant overall performance

Throughout the experiment, we recorded J. virginiana mortality in the four treatments. At the end of the experiment (24 months), all plant material was harvested. J. virginiana and Q. stellata roots were washed with water to remove soil, extended to their full length, and measured with a ruler to record the maximum root length. We did not measure root length of B. inermis due to root breakage during harvesting. To account for the effect of competition on the performance of J. virginiana, plants were dried in the oven at 60 ºC for 48 h, then weighed to measure total plant biomass, including root biomass, for J. virginiana, Q. stellata, and B. inermis (S1 Fig).

Water status of J. virginiana saplings

J. virginiana leaf relative water content (RWC) was determined for the four treatments at the end of the experiment (August 2019). J. virginiana leaf samples were weighed immediately after harvesting (fresh weight), then placed in a beaker with distilled water for 18 h at room temperature and then weighed to determine the saturated weight. The samples were then dried in an oven at 60°C for 24 h to obtain their dry weights. The RWC was calculated using the following formula [45, 46]:(1)

We also measured J. virginiana leaf-water potential, at the end of the experiment, between 11:00 am and 1:00 pm using a Scholander pressure chamber, Model 1505D (PMS Instrument Company, New York, USA). Midday water potential represents the time period when the plants are the most active, most water stressed, and have the greatest Vapor Pressure Deficit (VPD) [47].

Statistical analysis

Data were checked and passed the normality and homogeneity of variance using the Shapiro–Wilk and Levene’s test, respectively. A univariate repeated-measures analysis of variance (ANOVA), with α = 0.05, was used to analyze differences in soil moisture content in the four J. virginiana treatments among different soil depths and different measurement dates (Sep. 2018, Nov. 2018, Jan. 2019, May 2019, and Aug. 2019).

To determine if the plants used different amounts of water over 2018 and 2019, we also performed an ANCOVA on the soil moisture data for each treatment, with depth as the covariate (continuous variable) and month and J. virginiana treatments (JUVI, QUST, BRIN, and QUST+BRIN) as the independent variables. Time periods were defined as the first year for soil moisture data collected in September 2018, and second year for soil moisture data collected in August 2019. To examine possible time shifts in soil moisture content, we used a linear and a cubic regression spline. We also used a LOESS regression to visualize possible trends in soil moisture content between the first (2018) and second year of the experiment (2019) in the four J. virginiana treatments.

We ran a univariate analysis of variance (ANOVA) followed by Scheffé post hoc tests to analyze differences in total biomass, root length, J. virginiana RWC, and midday leaf-water potential in the four J. virginiana treatments. Pots with dead J. virginiana were excluded from the analysis. J. virginiana sapling mortality in the different treatments was analyzed using a Chi-square (χ²) test. The LOESS regression and figures were all done in R version 3.6.0 (R Development Core Team, 2019). All other statistical analyses were done in SPSS 26.0 statistical package (IBM SPSS, 2019).

Results

Soil moisture content changes in the four J. virginiana treatments

Our observations showed that water from the drip irrigation system was percolating down the soil profile and was able to reach the bottom of the pots (70 cm). The soil moisture content varied depending on the treatment (F = 29.07; p < 0.001; Fig 1) and soil depths (F = 270.62; p < 0.001; Fig 1). We did not find an overall shift in soil moisture content among the different measurement periods (Sep. 2018, Nov. 2018, Jan. 2019, May 2019, and Aug. 2019) (F = 1.66; p = 0.14; Fig 1). Soil moisture content among the different measurement periods, however, was dependent on J. virginiana treatment combination (Table 1). In addition, we ran both linear regressions and cubic splines to examine possible time shifts in soil moisture content. All regression lines were significant (p < 0.001). Cubic splines had higher coefficients of determination than linear regressions in all cases (S1 Table).

[Figure omitted. See PDF.]

Volumetric soil water content at different depths in 2018 (18-Sep) and 2019 (19-Aug) for (a) JUVI, (c) QUST, (e) BRIN, and (g) QUST+BRIN. Volumetric soil water content at different depths for the different measurement periods for (b) JUVI, (d) QUST, (f) BRIN, and (h) QUST+BRIN. In panels a, c, e, and g the circles and lines represent moisture content and LOESS regressions for 2018 (black) and 2019 (grey). In panels b, d, f, and h, the square symbols with dashed line represent water moisture data at 10 cm, the triangles and black line represent water moisture data at 20 cm, the circles and dotted line (..) represent water moisture content at 30 cm, and the and diamond long-dashed line (––) represent water moisture data at 40 cm. JUVI = J. virginiana only, QUST = J. virginiana growing with Q. stellata, BRIN = J. virginiana growing with B. inermis, and QUST+BRIN = J. virginiana growing with both Q. stellata and B. inermis.

[Figure omitted. See PDF.]

JUVI = J. virginiana only, QUST = J. virginiana growing with Q. stellata, BRIN = J. virginiana growing with B. inermis, and QUST+BRIN = J. virginiana growing with both Q. stellata and B. inermis.

Water uptake pattern of J. virginiana at different depths.

After comparing our soil moisture data to pots without any plants, J. virginiana growing without competitors (JUVI) used more water from the deeper layers of the soil (30 and 40 cm) compared to the shallow depths (10 and 20 cm) (Fig 1a and 1b). We found time progression of soil moisture content between the two years of the experiment (2018 and 2019; Fig 1a). For example, there was significantly more moisture in the soil profile in the first year (2018) than in the second year (2019) (F = 10.29; p = 0.002; Fig 1a; S1 Table), indicating that J. virginiana used more water in the second year of the experiment and followed the same water uptake pattern throughout the experiment’s timeframe. However, we did not find a significant change in water uptake among the different measurement periods at all depths for JUVI treatment (F = 0.63; p = 0.82; Table 1 and Fig 1b).

Water uptake pattern of Q. stellata at different depths.

When J. virginiana grew with the Q. stellata (QUST treatment), we found lower soil moisture content at 10 cm and 20 cm depths compared to JUVI treatment (F = 4.47; p = 0.001). Contrastingly, there was no significant change in water uptake at 30 cm and 40 cm compared to the JUVI treatment (F = 0.007; p = 0.8; Fig 1c and 1d). In addition, we found that the soil moisture content during the first year of the experiment (2018) was greater than in 2019 (F = 64.94; p <0.001; Fig 1c), indicating that the J. virginiana and Q. stellata used more water later in the experiment (S1 Table). In addition, the repeated measures analysis indicated significant changes in moisture content across the different measurement periods. Specifically, there was less soil moisture in May 2019 and Aug 2019 than on other measurement dates at 10, 20, and 40 cm depths (F = 1.96; p = 0.029). However, there was no significant difference in moisture content at 30 cm during the different measurement periods (F = 1.63; p = 0.176; Table 1, Fig 1d).

Water uptake pattern of B. inermis at different depths.

When J. virginiana occurred with B. inermis (BRIN treatment), there was a significant decline in soil moisture content at all depths compared to JUVI treatment (Fig 1e and 1f). In contrast to the results of JUVI treatment, the soil moisture content in the BRIN treatment was significantly lower during the first year (2018) than the year of the experiment (2019) (F = 83.89; p < 0.001; Fig 1e). In addition, we found significantly more moisture in the soil profile (the plants took up less water) later in the experiment (May 2019 and Aug 2019) at all depths (F = 2.22; p = 0.011; Table 1, Fig 1f).

Water uptake pattern of Q. stellata and B. inermis at different depths.

In the treatment with the most competitors, Q. stellata and B. inermis (QUST+BRIN treatment), the soil moisture content varied across the experiment. The QUST+BRIN treatment had significantly more moisture in the profile than in the JUVI treatment throughout the experiment at all depths (F = 20.69; p < 0.001; Fig 1g). Moreover, we found that the soil moisture content in 2018 was lower than in 2019 (F = 412.88; p < 0.001; Fig 1g and 1h), indicating that J. virginiana, Q. stellata, and B. inermis used less water later in the experiment (S1 Table). In addition, the repeated-measures analysis indicated a significant change in soil moisture content throughout the experiment at all depths (Table 1).

Examining water uptake at the different depths, our results indicate that B. inermis mostly took moisture from the shallow depths (10 cm and 20 cm) due to their shallow roots, whereas J. virginiana and Q. stellata from the deeper soil layers. There was an overlap in water uptake at the greater depths (30 cm and 40 cm) between J. virginiana and Q. stellata (Fig 1g and 1h).

Rooting depths and total biomass

J. virginiana root depths and total biomass were dependent on the treatment (F = 35.80; p < 0.001 and F = 12.30; p < 0.001 respectively; Figs 2 and 3). The maximum J. virginiana root length was longer than the pot depths (70 cm). However, our observations and data showed no sign that J. virginiana saplings were root-restricted or rootbound as they were exploring the full volume of the soil. The rooting depth of J. virginiana saplings was not significantly affected by Q. stellata (F = 0.45; p = 0.83; Fig 2). However, in treatments containing B. inermis (BRIN and QUST+BRIN), J. virginiana saplings had significantly shorter roots than J. virginiana saplings growing in the JUVI or with the QUST treatments (F = 39.34; p < 0.001; Fig 2). Moreover, we found an overlap between J. virginiana and Q. stellata root lengths in the QUST+BRIN treatment (Fig 2).

[Figure omitted. See PDF.]

B. inermis negatively affected J. virginiana root length but there was not an effect of Q. stellata competition. Note that we did not measure B. inermis root length due to root breakage during harvesting. Different letters represent significant differences (Scheffe´ post hoc test). JUVI = J. virginiana only, QUST = J. virginiana growing with Q. stellata, BRIN = J. virginiana growing with B. inermis, and QUST+BRIN = J. virginiana growing with both Q. stellata and B. inermis.

[Figure omitted. See PDF.]

B. inermis negatively affected J. virginiana total biomass. Different letters represent significant differences (Scheffe´ post hoc test). JUVI = J. virginiana only, QUST = J. virginiana growing with Q. stellata, BRIN = J. virginiana growing with B. inermis, and QUST+BRIN = J. virginiana growing with both Q. stellata and B. inermis.

The presence of Q. stellata did not have a significant effect on J. virginiana total biomass (F = 0.01; p = 0.93; Fig 3). However, J. virginiana in the JUVI treatment had higher mean total biomass than when grown in the BRIN or QUST+BRIN treatments (F = 25.20; p < 0.001; Fig 3). Overall, B. inermis negatively affected J. virginiana’s performance.

J. virginiana water status

There was no significant difference in J. virginiana RWC (F = 1.64; p = 0.195) between the four J. virginiana treatment combinations (Fig 4). However, there was a significant difference in J. virginiana water potential between the QUST and BRIN treatments (F = 4.18; p = 0.011; Fig 5).

[Figure omitted. See PDF.]

There was no significant difference in RWC across treatments, suggesting that it was not water stress that negatively affected the plants in the treatments with B. inermis. Different letters represent significant differences (Scheffe´ post hoc test). JUVI = J. virginiana only, QUST = J. virginiana growing with Q. stellata, BRIN = J. virginiana growing with B. inermis, and QUST+BRIN = J. virginiana growing with both Q. stellata and B. inermis.

[Figure omitted. See PDF.]

There was a significant difference in water potential only between the QUST and BRIN treatments. Different letters represent significant differences (Scheffe´ post hoc test). JUVI = J. virginiana only, QUST = J. virginiana growing with Q. stellata, BRIN = J. virginiana growing with B. inermis, and QUST+BRIN = J. virginiana growing with both Q. stellata and B. inermis.

J. virginiana mortality

J. virginiana sapling mortality was affected by the treatment combination (χ² = 10.49, df = 3; p = 0.01; Fig 6). We did not find a significant change in J. virginiana mortality in JUVI, QUST (χ² = 0; p = 1), or with BRIN (χ² = 0.8; p = 0.37). There was, however, significantly higher J. virginiana mortality when grown in the QUST+BRIN treatment combination (χ² = 6.79; p = 0.009; Fig 6).

[Figure omitted. See PDF.]

B. inermis negatively affected J. virginiana survival. The only significant difference was in the QUST+BRIN treatment. JUVI = J. virginiana only, QUST = J. virginiana growing with Q. stellata, BRIN = J. virginiana growing with B. inermis, and QUST+BRIN = J. virginiana growing with both Q. stellata and B. inermis.

Discussion

In order to accurately model and forecast species dynamics and the impact of climate change on ecosystem processes, understanding how plants compete for water is vital. Our experimental design allowed us to investigate changes in J. virginiana water uptake and rooting depths to competition with an invasive grass (B. inermis) and a native tree (Q. stellata). We found separation of soil-moisture depletion zones by the different treatment combinations, but our study suggests that factors other than water impact the ability of J. virginiana to co-exist with grasses and other woody species. Furthermore, our data suggest that ecohydrological niche separation among coexisting plant species may not occur in the early life stages of woody species and competition for water may change as the plants adopt different plant water-use strategies [48].

J. virginiana water uptake pattern and rooting depth without competitors

J. virginiana is known to expand its range and survive in different environments due to its ability to alter its root morphology, allowing for enhanced water uptake, especially during periods of low moisture [39]. As predicted, we found that J. virginiana growing without competitors used water mostly from the deeper layers of the soil (30 and 40 cm) compared to the shallow depths (10 and 20 cm; Fig 1a and 1b) due to their extensive rooting system (Fig 2). It has been reported that J. virginiana roots can reach up to 7 m deep for mature trees [29]. Because we used saplings in this experiment, however, J. virginiana roots were not as extensive. Sapling establishment can be strongly suppressed by other plant species, including grasses, than larger trees [49], which makes the sapling stage critical in developing and understanding demographic and encroachment models [49, 50].

Effect of competition on J. virginiana water uptake and rooting depth

One main mechanism in which J. virginiana tolerates drought conditions and facilitates its encroachment could be niche partitioning. For instance, differences in rooting depths reduce water competition between trees and grasses [8] and among different tree species [32]. We found that plant interspecific competition had a significant effect on J. virginiana water uptake, root lengths, total biomass, and mortality (Figs 1–3 and 6, respectively). Consistent with Walter’s two-layer hypothesis (1939, 1979) [10, 51], our results suggest root separation between J. virginiana and B. inermis roots. B. inermis potentially used up more water than J. virginiana saplings at shallow depths (10 cm and 20 cm), due to their shallower rooting system. J. virginiana likely used the most water at greater depths (30 cm and 40 cm). For instance, it was reported that B. inermis roots can reach 1.4 m in depth [52], which is significantly shorter than J. virginiana roots. Eggemeyer et al. (2009) [29] and Awada et al. (2013) [53] have shown that J. virginiana competes for soil moisture with grasses in the top layers when water is available. when surface moisture decreases, however, J. virginiana shift their water uptake to deeper layers [54]. The FDR method we used to characterize soil moisture depletion does not allow us to quantify the total amount of water taken up at each depth, rather it gives us a relative measure of soil water depletion across the different depths. Given the interesting patterns described above, water consumption of J. virginiana competition should be addressed in future studies to better understand the full impact of water competition on establishment.

We also found root partitioning between the J. virginiana and Q. stellata, where J. virginiana had deeper roots than the Q. stellata (Fig 2). As a result, we may infer that J. virginiana took up the water from the deeper layers of the soil than Q. stellata. Torquato et al. (2020) [31] found that J. virginiana and Q. stellata may be using water from different depths in the oak-dominated Cross Timbers reserve in Oklahoma. In their experiment, Torquato et al. (2020) [31] found that J. virginiana’s roots occupied the shallower soil depths (45 cm). These results are not consistent with what we found in our experiment because in our study J. virginiana had deeper roots than Q. stellata. We suggest that the differences between our study and Torquato et al. (2020) [31] may be due to ontogenetic differences because they used mature trees whereas we used saplings. The trees may reverse partitioning strategies as a result of ontogeny, but more research is needed to understand the relationship between tree age, size, and water uptake behavior. Plant size, age, and ontogeny may alter the outcome of interspecific interactions and competition for available resources, with seedlings and saplings establishment benefited more than larger individuals [55]. A few studies showed a shift from facilitation during establishment to a competitive effect on adult plant growth and performance [55–59]. This can be related to the fact that larger plants often have greater resource requirements, which increase their competitive impacts. In addition, larger plants are reported to have a lower mortality because they have more extensive root systems [60, 61].

We also examined water uptake at the different depths when J. virginiana grew with Q. stellata and B. inermis (QUST+BRIN). Our results indicate that there was an overlap in water uptake at the greater depths between J. virginiana and Q. stellata. B. inermis has laterally extensive rooting system that mostly took moisture from the shallow depths, potentially leaving less water for J. virginiana and Q. stellata at the 10 cm and 20 cm depths. As a result, we suggest that the two woody species, J. virginiana and Q. stellata, invested in belowground development and shared water uptake from the 30 cm and 40 cm layers (Fig 1g and 1h). We also found a decline in J. virginiana rooting depths in the QUST+BRIN combination compared to the JUVI treatment (Fig 2) and an overlap between J. virginiana and Q. stellata roots. This suggests that Q. stellata can potentially be considered a major competitor affecting J. virginiana performance but perhaps only in the presence of a shallow-rooted species such as B. inermis.

Changes in water uptake over time

When comparing water uptake between the two years for all treatment combinations, we found that J. virginiana in the JUVI treatment used significantly more water in the second year (2019) than the first year (2018). As evidenced by the fact that the same amount of water was supplemented in both years but there was more moisture in the soil profile in 2018. This might be attributed to the increase in J. virginiana sapling size as they grew older. Similarly, we found that the J. virginiana plants competing with Q. stellata also used significantly more water in 2019 compared to 2018 (there was a decline in water uptake by the J. virginiana and Q. stellata in 2019). Unexpectedly, however, in the treatments with B. inermis (BRIN and QUST+BRIN), the water uptake by the plants declined in 2019 compared to 2018 (higher soil moisture in 2019 than 2018; S1 Table). This could be related to the negative effect of B. inermis on J. virginiana and Q. stellata survival, growth rate, and total biomass (aboveground and belowground biomass). Consequently, J. virginiana and Q. stellata had slower growth in 2019 compared to 2018. Another possible cause for the lower water uptakes in the second year of the experiment could be related to the plants, mainly B. inermis, requiring more water to establish their roots and growing at the beginning of the experiment [11], which resulted in lower soil moisture content in the profile. In 2019, J. virginiana and B. inermis were already established, and their growth rate declined, which ultimately resulted in reduced water requirements and uptake.

Other factors influencing J. virginiana competitive abilities

Grasses can affect woody plant establishment directly, through competition for water, light, or nutrients. Grasses can also have an indirect effect as grass-fueled fires can destroy saplings [62, 63]. Furthermore, sapling emergence, survival, and growth of woody plants are reported to be reduced by grasses [27, 64, 65]. Our results suggest that B. inermis had a significant negative impact on J. virginiana performance (Fig 3), but the lack of water stress for J. virginiana indicates that this was not due to the lack of water availability. Similarly, other studies by Ward (2020, 2021) [41, 66] showed that shade is the most important factor affecting the growth rate and biomass of J. virginiana. Interestingly, we found no evidence of the negative effect of Q. stellata competition on J. virginiana performance even though previous work might suggest that shade by Q. stellata would be detrimental for J. virginiana. Thus, we suggest our results are likely due to the root partitioning between J. virginiana and Q. stellata. Furthermore, Torquato et al. (2020) [31] found that the soil moisture in pure J. virginiana stands was more depleted than in J. virginiana-Q. stellata stands, suggesting that J. virginiana is a better competitor for water than Q. stellata, which may facilitate J. virginiana encroachment into oak woodlands. Lastly, we need to recognize that grasses have the opposite effect and control J. virginiana encroachment. Thus, the relative importance and effects of grasses and woody species competition needs to be better understood.

Supporting information

S1 Fig. B. inermis total biomass (g) (above and below-ground).

https://doi.org/10.1371/journal.pone.0280100.s001

(TIF)

S1 Table. Linear regression and cubic splines equations of the soil moisture content at different depths for the four treatments in the first (2018) and second (2019) year of the experiment.

Cubic splines had higher coefficients of determination than the linear models. All regression lines were significant (p < 0.001).

https://doi.org/10.1371/journal.pone.0280100.s002

Acknowledgments

We would like to thank Christian Combs, John Christakis, and Caitlyn Webb for their help in setting up this experiment.

Citation: Hamati S, Medeiros JS, Ward D (2023) Effects of post oak (Quercus stellata) and smooth brome (Bromus inermis) competition on water uptake and root partitioning of eastern redcedar (Juniperus virginiana). PLoS ONE 18(2): e0280100. https://doi.org/10.1371/journal.pone.0280100

About the Authors:

Samia Hamati

Roles: Data curation, Formal analysis, Supervision, Writing – original draft, Writing – review & editing

E-mail: [email protected]

Current address: Department of Biology, University of North Dakota, Grand Forks, North Dakota, United States of America

Affiliation: Department of Biological Sciences, Kent State University, Kent, Ohio, United States of America

ORICD: https://orcid.org/0000-0002-0045-0824

Juliana S. Medeiros

Roles: Funding acquisition, Writing – review & editing

Affiliation: Holden Arboretum, Kirtland, Ohio, United States of America

David Ward

Roles: Conceptualization, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Supervision, Writing – original draft, Writing – review & editing

Affiliation: Department of Biological Sciences, Kent State University, Kent, Ohio, United States of America

References

1. Schwinning S, Weiner J. Mechanisms determining the degree of size asymmetry in competition among plants. Oecologia. 1998; 113: 447–55. pmid:28308024

2. Wright JS. Plant diversity in tropical forests: a review of mechanisms of species coexistence. Oecologia. 2002; 130: 1–4. pmid:28547014

3. Casper BB, Jackson RB. Plant competition underground. Annual Review of Ecology and Systematics. 1997; 1: 545–570.

4. Sardans J, Grau O, Chen HY, Janssens IA, Ciais P, Piao S, et al. Changes in nutrient concentrations of leaves and roots in response to global change factors. Global Change Biology. 2017; 23: 3849–3856. pmid:28407324

5. Li H, Liu B, McCormack ML, Ma Z, Guo D. Diverse belowground resource strategies underlie plant species coexistence and spatial distribution in three grasslands along a precipitation gradient. New Phytologist. 2017; 216:1140–1150. pmid:28758691

6. Fargione J, Tilman D. Niche differences in phenology and rooting depth promote coexistence with a dominant C4 bunchgrass. Oecologia. 2005; 143: 598–606. pmid:15791430

7. King J, Gay A, Sylvester-Bradley RO, Bingham IA, Foulkes J, Gregory P, et al. Modelling cereal root systems for water and nitrogen capture: towards an economic optimum. Annals of Botany. 2003; 91: 383–390. pmid:12547691

8. Walter H, Breckle SW. The Law of Relative Constancy of Habitat—Ecotypes and Ecoclines. In Ecological Systems of the Geobiosphere. Springer, Berlin, Heidelberg: 1985. pp. 194–202.

9. Farrior CE, Dybzinski R, Levin SA, Pacala SW. Competition for water and light in closed-canopy forests: a tractable model of carbon allocation with implications for carbon sinks. The American Naturalist. 2013; 181: 314–330. pmid:23448882

10. Walter H. Grassland, savanne und busch der arideren teile Afrikas in ihrer ökologischen bedingheit. Jahrbucher für Wissenschaftliche Botanik. 1939; 87: 750–860.

11. Ward D, Wiegand K, Getzin S. Walter’s two-layer hypothesis revisited: back to the roots!. Oecologia. 2013; 172: 617–630. pmid:23266712

12. Mazzacavallo MG, Kulmatiski A. Modelling water uptake provides a new perspective on grass and tree coexistence. PLoS One. 2015; 10: e0144300. pmid:26633177

13. Craine JM, Dybzinski R. Mechanisms of plant competition for nutrients, water and light. Functional Ecology. 2013; 27: 833–840.

14. Mayor MD, Bóo RM, Peláez DV, Elıa OR. Seasonal variation of the soil seed bank of grasses in central Argentina as related to grazing and shrub cover. Journal of Arid Environments. 2003; 53: 467–477.

15. Epron D, Dreyer E. Long-term effects of drought on photosynthesis of adult oak trees [Quercus petraea (Matt.) Liebl. and Quercus robur L.] in a natural stand. New Phytologist. 1993; 125: 381–389. pmid:33874494

16. Jose S, Gillespie AR, Seifert JR, Biehle DJ. Defining competition vectors in a temperate alley cropping system in the midwestern USA: 2. Competition for water. Agroforestry systems. 2000; 48: 41–59.

17. Sperry JS, Love DM. What plant hydraulics can tell us about responses to climate-change droughts. New Phytologist. 2015; 207: 14–27. pmid:25773898

18. Liu X, Zhang Q, Song M, Wang N, Fan P, Wu P, et al. Physiological Responses of Robinia pseudoacacia and Quercus acutissima Seedlings to Repeated Drought-Rewatering Under Different Planting Methods. Frontiers in Plant Science. 2021;12. pmid:34938307

19. Bréda N, Huc R, Granier A, Dreyer E. Temperate forest trees and stands under severe drought: a review of ecophysiological responses, adaptation processes and long-term consequences. Annals of Forest Science. 2006; 63: 625–644.

20. Aung K, Jiang Y, He SY. The role of water in plant–microbe interactions. The Plant Journal. 2018; 93: 771–780. pmid:29205604

21. Barber NA, Soper Gorden NL. How do belowground organisms influence plant–pollinator interactions?. Journal of Plant Ecology. 2015; 8: 1–11.

22. Kambatuku JR, Cramer MD, Ward D. Overlap in soil water sources of savanna woody seedlings and grasses. Ecohydrology. 2013; 6: 464–673.

23. Aschehoug ET, Brooker R, Atwater DZ, Maron JL, Callaway RM. The mechanisms and consequences of interspecific competition among plants. Annual Review of Ecology, Evolution, and Systematics. 2016; 47: 263–281.

24. Zhou Y, Wigley BJ, Case MF, Coetsee C, Staver AC. Rooting depth as a key woody functional trait in savannas. New Phytologist. 2020; 227: 1350–1361. pmid:32306404

25. Holdo RM, Nippert JB, Mack MC. Rooting depth varies differentially in trees and grasses as a function of mean annual rainfall in an African savanna. Oecologia. 2018; 186: 269–280. pmid:29143883

26. Belsky A. J. Influences of trees on savanna productivity: tests of shade, nutrients, and tree-grass competition. Ecology. 1994; 75: 922–932.

27. Ward D, Esler KJ. What are the effects of substrate and grass removal on recruitment of Acacia mellifera seedlings in a semi-arid environment?. Plant Ecology. 2011; 212: 245–250.

28. Rossatto DR, Silva LD, Villalobos-Vega R, Sternberg LD, Franco AC. Depth of water uptake in woody plants relates to groundwater level and vegetation structure along a topographic gradient in a neotropical savanna. Environmental and Experimental Botany. 2012; 77: 259–266.

29. Eggemeyer KD, Awada T, Harvey FE, Wedin DA, Zhou X, Zanner CW. Seasonal changes in depth of water uptake for encroaching trees Juniperus virginiana and Pinus ponderosa and two dominant C4 grasses in a semiarid grassland. Tree physiology. 2009; 29: 157–169. pmid:19203941

30. Tomiolo S, Ward D. Species migrations and range shifts: A synthesis of causes and consequences. Perspectives in plant ecology, evolution and systematics. 2018; 33: 62–77.

31. Torquato PR, Zou CB, Adhikari A, Adams HD, Will RE. Drought tolerance and competition in eastern redcedar (Juniperus virginiana) encroachment of the oak-dominated Cross Timbers. Frontiers in Plant Science. 2020; 11: 59. pmid:32117395

32. Limb RF, Engle DM, Alford AL, Hellgren EC. Tallgrass prairie plant community dynamics along a canopy cover gradient of eastern redcedar (Juniperus virginiana L.). Rangeland Ecology & Management. 2010; 63: 638–644.

33. Lawton RO, Cothran P. Factors influencing reproductive activity of Juniperus virginiana in the Tennessee Valley. Journal of the Torrey Botanical Society. 2000; 127: 271–279.

34. Volder A, Briske DD, Tjoelker MG. Climate warming and precipitation redistribution modify tree–grass interactions and tree species establishment in a warm-temperate savanna. Global Change Biology. 2013; 19: 843–57. pmid:23504841

35. Lawson ER. Juniperus virginiana L. eastern redcedar. Silvics of North America. 1990; 1:131–140.

36. Volder A, Tjoelker MG, Briske DD. Contrasting physiological responsiveness of establishing trees and a C4 grass to rainfall events, intensified summer drought, and warming in oak savanna. Global Change Biology. 2010; 16: 3349–3362.

37. Zou CB, Twidwell D, Bielski CH, Fogarty DT, Mittelstet AR, Starks PJ, et al. Impact of eastern redcedar proliferation on water resources in the Great Plains USA—current state of knowledge. Water. 2018; 10: 1768.

38. Bahari ZA, Pallardy SG, Parker WC. Photosynthesis, water relations, and drought adaptation in six woody species of oak-hickory forests in central Missouri. Forest Science. 1985; 31: 557–569.

39. Caterina GL, Will RE, Turton DJ, Wilson DS, Zou CB. Water use of Juniperus virginiana trees encroached into mesic prairies in Oklahoma, USA. Ecohydrology. 2014; 7: 1124–1134.

40. Ormsbee P, Bazzaz FA, Boggess WR. Physiological ecology of Juniperus virginiana in oldfields. Oecologia. 1976; 23: 75–82. pmid:28309214

41. Ward D. Shade is the most important factor limiting growth of a woody range expander. PLoS One. 2020; 15: e0242003. pmid:33264310

42. Daly C, Bachelet D, Lenihan JM, Neilson RP, Parton W, Ojima D. Dynamic simulation of tree–grass interactions for global change studies. Ecological Applications. 2000; 10: 449–469.

43. Anderson RC. The historic role of fire in the North American grassland. In: Collins S.L. and Wallace L, editors. Fire in North American tallgrass prairies. University of Oklahoma Press, Norman: 1990. pp. 8–18.

44. Stransky JJ. Quercus stellata Wangenh. Silvics of North America. 1990; 2: 738–43.

45. Barrs H. Determination of water deficits in plant tissues. In: Kozowski T, editor. Water Deficits and Plant Growth. Academic Press, New Delhi: 1968. Pp. 235–368.

46. Saura-Mas S, Lloret F. Leaf and shoot water content and leaf dry matter content of Mediterranean woody species with different post-fire regenerative strategies. Annals of Botany. 2007; 99: 545–554. pmid:17237213

47. Williams LE, Araujo FJ. Correlations among predawn leaf, midday leaf, and midday stem water potential and their correlations with other measures of soil and plant water status in Vitis vinifera. Journal of the American Society for Horticultural Science. 2002; 127: 448–454.

48. Illuminati A, Querejeta JI, Pías B, Escudero A, Matesanz S. Coordination between water uptake depth and the leaf economic spectrum in a Mediterranean shrubland. Journal of Ecology. 2022; 110: 1844–1856.

49. Riginos C. Grass competition suppresses savanna tree growth across multiple demographic stages. Ecology. 2009; 90: 335–340. pmid:19323216

50. Cohen D. The expected efficiency of water utilization in plants under different competition and selection regimes. Israel Journal of Botany. 1970; 19: 50–54.

51. Walter MF, Steenhuis TS, Haith DA. Nonpoint source pollution control by soil and water conservation practices. Transactions of the ASAE. 1979; 22: 834–840.

52. Dibbern JC. Vegetative responses of Bromus inermis to certain variations in environment. Botanical Gazette. 1947; 109: 44–58.

53. Awada T, El-Hage R, Geha M, Wedin DA, Huddle JA, Zhou X, et al. Intra-annual variability and environmental controls over transpiration in a 58-year-old even-aged stand of invasive woody Juniperus virginiana L. in the Nebraska Sandhills, USA. Ecohydrology. 2013; 6: 731–740.

54. Verweij RJ, Higgins SI, Bond WJ, February EC. Water sourcing by trees in a mesic savanna: responses to severing deep and shallow roots. Environmental and Experimental Botany. 2011; 74: 229–236.

55. Le Roux PC, Shaw JD, Chown SL. Ontogenetic shifts in plant interactions vary with environmental severity and affect population structure. New Phytologist. 2013; 200: 241–250. pmid:23738758

56. Miriti MN. Ontogenetic shift from facilitation to competition in a desert shrub. Journal of Ecology. 2006; 94: 973–979.

57. Reisman-Berman O. Age-related change in canopy traits shifts conspecific facilitation to interference in a semi-arid shrubland. Ecography. 2007; 30: 459–470.

58. Lortie CJ, Turkington R. Species-specific positive effects in an annual plant community. Oikos. 2008; 117: 1511–1521.

59. Armas C, Pugnaire FI. Ontogenetic shifts in interactions of two dominant shrub species in a semi-arid coastal sand dune system. Journal of Vegetation Science. 2009; 20: 535–546.

60. Kleier C, Rundel PW. Microsite requirements, population structure and growth of the cushion plant Azorella compacta in the tropical Chilean Andes. Austral Ecology. 2004; 29: 461–470.

61. Haussmann NS, McGeoch MA, Boelhouwers JC. Contrasting nurse plants and nurse rocks: the spatial distribution of seedlings of two sub-Antarctic species. Acta Oecologica. 2010; 36: 299–305.

62. Bond WJ. What limits trees in C4 grasslands and savannas? Annual Review of Ecology, Evolution, and Systematics. 2008; 39: 641–659.

63. O’connor TG, Puttick JR, Hoffman MT. Bush encroachment in southern Africa: changes and causes. African Journal of Range & Forage Science. 2014; 31: 67–88.

64. Vadigi S, Ward D. Shade, nutrients, and grass competition are important for tree sapling establishment in a humid savanna. Ecosphere. 2013; 4: 1–27.

65. Kraaij T, Ward D. Effects of rain, nitrogen, fire and grazing on tree recruitment and early survival in bush-encroached savanna, South Africa. Plant Ecology. 2006; 186: 235–246.

66. Ward D. Shade affects fine-root morphology in range-encroaching eastern redcedars (Juniperus virginiana) more than competition, soil fertility and pH. Pedobiologia. 2021; 84: 150708.