Skip to main content
NCBI home page
Springer logoLink to Springer
. 2026 Oct 27;208(11):160. doi: 10.1007/s00442-026-05935-0

Selective mortality leads to competitive release of a foundational perennial grass in an aridifying grassland

Sean L Hoy-Skubik 1,2,✉, David L Hoover 3, Alix A Pfennigwerth 2, Anna C Knight 2, Michael C Duniway 2
PMCID: PMC13714021  PMID: 42944897

Abstract

Vegetation dynamics in dryland ecosystems are defined by competition for water, a limiting and increasingly variable resource. As aridification creates warmer and drier conditions, with more frequent and intense droughts, the nature of competitive relationships may be altered, leading to changes in community composition. Reduced water availability is expected to decrease the cover and abundance of cool-season C3 grasses on the Colorado Plateau, with unknown impacts on other plant functional types, such as drought-tolerant C4 species. Here, we explore the effects of competitive release on Pleuraphis jamesii, a dominant C4 grass with rooting and photosynthetic traits that may enable it to capitalize on the mortality of less drought tolerant C3 species. To accomplish this, we used herbicide to create three removal treatments (C3 grass and forb removal, C4 grass removal, and control (no removal)) surrounding P. jamesii focal individuals and measured the subsequent impacts on water availability and plant water relations and growth. The experimentally induced mortality resulted in greater water availability for focal individuals in the C3 removal treatment. Increased water availability in C3 removal plots translated to less water stress and greater biomass production of focal P. jamesii plants, suggesting that competitive release can serve as the basis for shifts in community composition towards drought tolerant C4 plants observed regionally. These responses were not observed in the C4 removal treatment despite removing roughly equivalent amounts of total plant cover. Our results suggest that reductions in C3 competitors may benefit drought-tolerant C4 species through increased resource availability. These findings contribute to a mechanistic understanding of competitive interactions in increasingly water-limited systems.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1007/s00442-026-05935-0.

Keywords: Competitive release, Drylands, Drought, Community shifts, Physiology

Introduction

Dryland ecosystems are defined by limited water availability due to low and variable precipitation regimes and high atmospheric demand, which may lead to high competition within plant communities for this scarce resource (Noy-Meir 1973; Safriel et al. 2005). However, drylands also display remarkable biodiversity with organisms enduring water stress using a range of evolutionary strategies (Safriel et al. 2005; Ward 2016). Finite resources ensure that competitive interactions play a vital role in community composition, particularly in the context of competition between species with significant niche overlap. Consequently, changes to the nature of these competitive interactions, such as altered competitive fitness or fluctuations in resource availability, can have strong effects on community composition and structure.

Global climate models predict that water availability in drylands will be altered in myriad ways, with potential impacts to plant competitive interactions. The rate of warming in drylands has exceeded average global warming rates over the past 100 years, and this trend is expected to continue (Huang et al. 2017; Karl 2011). Higher temperatures increase evaporative demand, leading to increased aridification and lower water availability (Collins et al. 2013). Such climatic shifts will serve to increase plant hydraulic stress in resource-scarce dryland ecosystems.

While dryland plant species have adaptations to cope with climatic extremes and variability, predicted changes in hydrological regimes may exceed the adaptive capacity of individual species (Schwalm et al. 2010; Smith 2011; Anderegg et al. 2013). Reduced soil moisture can lead to species-specific mortality, shift competitive interactions between plants, and alter community composition (Schwinning and Kelly 2013; Collins et al. 2014; Hoover et al. 2017). Stabilizing mechanisms of coexistence, such as temporal and spatial niche partitioning, may allow for the maintenance of species diversity by reducing interspecific competition for resources (Chesson 2000; Reynolds et al. 2007; Reich 2014; Silvertown et al. 2015). However, overlap in niche space still occurs. Species diversity is therefore thought to reflect a balance between competitive exclusion, wherein competitive interactions inhibit abundance of a given species within a local community and competitive release, wherein a ‘releasing factor’ reduces competitive pressure on a given species, allowing an increase in abundance (Segre et al. 2016). Forecasting future ecosystem dynamics under climate change requires a deeper understanding of how the interaction between shifts in community composition and individual-level water competition dynamics will affect the balance between competitive exclusion and release.

Plants in dryland ecosystems tend to have unique spatiotemporal hydrological niches, specializing in utilizing water from specific soil layers at certain times of year, rather than converging on a generalist phenotype (Schwinning and Ehleringer 2001). Specializations include differences in rooting depths (i.e., spatial niche partitioning) and phenological timing (i.e., temporal niche partitioning), which allow plants to reduce direct competition for water resources (Gulmon et al. 1983; Breshears et al. 2009; Walter 2012). In drylands, differing photosynthetic pathways are a major form of temporal niche partitioning that enables species co-existence (Kemp 1983). Species that utilize the C3 photosynthetic pathway are the most productive and use the most water during spring and early summer (i.e., cool-season species), while species employing the C4 photosynthetic pathway are more active and use more water in the warmer months of summer (i.e., warm-season species). However, climate change will likely alter how cool- and warm-season plant species partition water-use through space and time, thereby upsetting the balance of competition and coexistence.

Several studies have suggested that C4 grasses have higher drought tolerance than their C3 counterparts due to higher water-use efficiency and differences in growth and biomass allocation patterns (Pearcy and Ehleringer 1984; Sage 2004). For example, in the Colorado Plateau, a semi-arid region of southwestern USA that has become increasingly dry over the past several decades (Williams et al. 2020), C3 grasses have decreased in cover and increased in mortality, with little change in C4 grass cover or mortality (Munson et al. 2011; Witwicki et al. 2016; Winkler et al. 2019; Finger Higgens et al. 2025). While C3 and C4 grasses have different optimal temperatures for photosynthesis, the hydrological regime of the Colorado Plateau can create phenological overlap, with both functional types able to utilize consistently high soil moisture levels in spring and monsoon-driven moisture pulses in summer (Comstock and Ehleringer 1992). Where C3 and C4 niches overlap, a reduction of C3 cover could initiate an expansion of C4 cover via competitive release. For example, during the Miocene, C4 grasses increased in dominance relative to C3 grasses in response to decreased rainfall and increased seasonality of water availability (Dettman et al. 2001; Sepulchre et al. 2006). Studies of modern ecosystems have shown that ecological release from interspecific competition can increase the width of a species’ realized niche (Bolnick et al. 2010). However, it is still uncertain how alterations to competitive relationships will affect future community composition.

Here, we examine whether experimental mortality of C3 or C4 species alters competitive interactions in a Colorado Plateau grassland. Specifically, we investigate the impacts of reduced competition on Pleuraphis jamesii Torr. (James’ Galleta Grass, synonym Hilaria jamesii), a warm-season, C4, dominant perennial rhizomatous grass on the Plateau (Duniway et al. 2022) that has been shown to have traits conferring drought resistance such as a deep rooting profile, high root to shoot ratio and high turgor loss point (Hoover et al. 2019), and phenological flexibility in response to drought (Finger Higgens et al. 2025). Previous research suggests that drought may increase moisture availability to P. jamesii by reducing cover of less drought tolerant competitors (Hoover et al. 2021). Our study consists of two removal treatments (C3 grass and forb removal, C4 grass removal) and a control (no removal). We applied herbicide to induce mortality of each subset of the plant community, to investigate community dynamics following selective mortality events while separating the effects of species competition from the effects of drought. We tested three hypotheses: (1) removing a subset of the neighboring plant community would increase soil moisture during the late summer monsoon season; (2) such increases in soil moisture would increase water availability to P. jamesii; and (3) increased water availability due to competitive release would increase P. jamesii biomass and cover of green tissue. By exploring the effects of competitive release on ecohydrology in this context, our work provides insight into future vegetation dynamics in the face of climate change.

Methods

Site description

We conducted this study in a 3.2-ha fenced study area just outside of Canyonlands National Park in southeastern Utah, USA (38.19’ N, 109.75’ W; 1505 m elevation; Fig. 1). The plant community at the site is a native grass-shrubland, dominated by native perennial grasses, annual and perennial forbs, and scattered shrubs (Table S2). The site has a history of light winter cattle grazing before being fenced in 2015. The site is characterized by deep (> 1.5 m) loamy fine sand soils, with little variability in soil texture across space and depth. The region’s climate is marked by cool winters (mean low temperature = -6.5 °C), warm summers (mean high temperature = 32.7 °C) and low precipitation (mean annual precipitation = 221 mm; Gremer et al. 2015). Seasonal fluctuations in temperature and evaporative demand lead to variable water availability, despite consistent average monthly precipitation levels (Comstock & Ehleringer 1992; Gremer et al. 2015). The C3 and C4 species in this region are able to initiate growth and often complete all phenological stages using cool-season soil moisture recharge owing to more reliable precipitation and low potential evaporation from November to April (Comstock & Ehleringer 1992, Finger Higgens et al. 2025). The warm season (July through September), characterized in this region by monsoonal precipitation patterns, can allow for a second growth phase, with C4 species being better able to respond with renewed growth, though monsoon intensity has high inter-annual variability (Comstock and Ehleringer 1992; Finger Higgens et al. 2025).

Fig. 1.

Fig. 1

Open in a new tab

Site location and experimental design. (a) Map of study site. The study site (black dot) is situated on the Colorado Plateau (grey-shaded area on inset map) just outside the Needles District of Canyonlands National Park (NP) in southeastern Utah, USA. (b) Schematic of a sample plot. Eighteen circular plots with a radius of 1 m and centered on a focal P. jamesii individual were situated in an enclosed 3.2-ha native grassland. Each plot was subjected to one of three removal treatments: C3 grass and forb removal, C4 grass removal, or control

Experimental design

We applied three different experimental treatments: C3 grass and forb removal, C4 grass removal, and control (no removal) during the growing season of 2019 (3 treatments × 6 blocks, N = 18 plots; Fig. 1). The C3 grass and forb removal simulates the loss of C3 grass species that has been observed with ongoing climate change (Munson et al. 2011) and drought (Hoover et al. 2015; Finger Higgens et al. 2025). We combined C3 grasses and forbs into a single category because they had similar total cover amounts as C4 grasses prior to treatments (fall 2018, data not shown); this allowed us to avoid the potential confounding effect of differences in total cover removed by each treatment. We used the C4 removal treatment to explore whether the effect of removing competition varies by plant functional group.

To select plots, we first tagged 46 P. jamesii individuals that did not have any shrubs in their immediate surroundings, used morphological measurements to estimate their biomass, and performed ocular cover estimates for a 1-m radius circle surrounding the individual (P. jamesii has been found to have a lateral root spread of 0.5 m, and other grasses in the study site have root spreads from 0.4–0.5 m (Hoover et al. 2019); we therefore do not expect the focal plant to directly compete with individuals outside of the 1-m radius circle). We then selected 18 of the 46 focal individuals that had similar biomass and similar surrounding cover compositions and assigned them to six experimental blocks by location. The three individuals in each block were then randomly assigned to one of the three treatments. Due to the rhizomatous, clump-forming nature of P. jamesii, identifying distinct individuals can be difficult. To collect individual-level data consistently through time, we define an ‘individual’ as all P. jamesii biomass within a 20-cm radius of the plot center and use this definition for data collected below.

On 29-May-2019, we applied an herbicide mixture containing 2% glyphosate (Ranger PRO, Monsanto Company, St. Louis, MO, USA), 2% methylated seed oil (Alligare MSO West, Alligare, LLC, Opelika, AL, USA), and 0.5% temporary blue pattern indicator (Signal, Precision Laboratories, Waukegan, IL, USA) by spraying and brushing the mixture onto all species of the given treatment type (C3 grasses and forbs, C4 grasses, or none) within the 1-m radius of the focal P. jamesii individuals. Upon entering the plant, glyphosate is translocated to regions of active growth, thereby generating mortality of both above- and below-ground tissues (Kanissery et al. 2019). Plastic bags were briefly placed over plants not receiving herbicide to avoid accidental treatment of non-target species. On 19-Jun-2019, we reapplied herbicide to new and surviving plants as needed within each removal treatment group, representing a small fraction of individuals treated in the initial application. We returned on 3-Jul-2019 to visually ensure that the treatments had been effective. By applying herbicide treatments in early summer when plants were actively growing, we ensured that the removal treatments closely simulated drought-induced mortality, wherein mortality is exhibited across tissues, and the biomass and nutrients of dead plants remain in the system. Furthermore, the precipitation patterns of the study period likely resulted in nutrient release dynamics similar to those associated with drought-induced mortality wherein soil rewetting after drought results in a burst of N- and C-mineralization (Birch 1964; Gessler et al. 2017). To determine the effectiveness of herbicide treatments, ocular cover estimates were performed prior to application (28-May-2019) as well as at the end of the experiment (10-Sep-2019).

Soil moisture and meteorological data

Within each plot, we recorded shallow soil volumetric water content (VWC, 5–25 cm depth) at hourly intervals using soil moisture and temperature sensors (30 cm, CS650, Campbell Scientific, Logan, UT, USA) inserted diagonally (45°) in the center of each plot (N = 18 sensors). Precipitation was measured on site using an hourly tipping rain gauge (TE25MM, Texas Electronics, Dallas, TX; precipitation data available from Grote et al. 2023).

Leaf water potential

We measured predawn leaf water potential (Inline graphic) on six separate days over the course of the drying period following a series of three monsoon events from 23-Jul-2019 to 27-Jul-2019, totaling 25.6 mm of precipitation. Inline graphic is assumed to be in hydrostatic equilibrium with soil water potential and thus reflects the ‘supply’ side of the hydrological niche. The first sampling occurred on 29-Jul-2019, while all plots were still saturated with water. The following five samplings occurred as the soil began to progressively dry (6-Aug-2019, 9-Aug-2019, 15-Aug-2019, 22-Aug-2019, 12-Sep-2019). Inline graphic measurements were made with a pressure chamber (Scholander et al. 1965) equipped with a digital pressure gauge (Model 600D, PMS Instruments Co., Albany, NY, USA; accuracy 0.001 MPa). Observations were made using one leaf cut from each focal individual (N = 18). Sampling order was randomized by block, with tissue samples collected and measured from only one block at a time. Tissue samples were kept in plastic bags containing wet paper towels prior to measurements to prevent any desiccation.

Carbon isotope

As an indicator of accumulated plant water availability, we measured P. jamesii leaf δ13C values. These values serve as a proxy for plant hydrologic status accumulated over the growing season, rather than at a single point in time, as with Inline graphic. C4 plants, such as P. jamesii, have been shown to have lower leaf δ13C values in water-limited plants compared to well-watered plants, with the magnitude of the shift dependent on drought severity (Buchmann et al. 1996; Ghannoum et al. 2002). While the response of δ13C to drought in C4 plants can vary according to bundle-sheath leakiness, under most environmental conditions, δ13C is expected to decrease with decreasing water availability (Cernusak et al. 2013; Ellsworth and Cousins 2016). Tissue samples were collected on 23-Aug-2019, transported to the lab, dried for 48 h at 60 °C, and weighed to the nearest 0.01 g. The samples were then sent to the Cornell Isotope Laboratory where they were analyzed using a Delta V isotope ratio mass spectrometer (Thermo Fisher Scientific, Germany) interfaced to a NC2500 elemental analyzer (Carlo Erba Instruments, Italy).

Phenology

We measured plant phenology biweekly over the course of the growing season. We assessed each tagged individual for percent green by visually estimating the amount of green tissue relative to total standing biomass (including brown tissue) of the individual (0–100%) as in Hoover et al. (2021).

Biomass

We destructively harvested all aboveground biomass from focal individuals in November 2019. We estimated biomass at experiment’s onset (May) using a combination of (1) non-destructive morphological measurements within plots and (2) morphological measurements with paired destructive harvests along nearby transects conducted in a previous study at the same site (n = 8 individuals × 6 blocks × 6 sampling periods = 288 biomass samples, Hoover et al. 2021). For each individual, we measured the following morphological traits: plant height, length, width, and basal circumference, as well as the number of live culms, green leaves, and green inflorescences of all ‘clumps’ within the individual’s 20 cm circular area. If groupings of culms had > 3 cm of soil between biomass, they were considered separate clumps. For biomass harvests along the transects and within the plots at the experiment’s end, we clipped green biomass at the ground level and transported the samples to the lab, where they were dried for 48 h at 60 °C and weighed to the nearest 0.01 g. Weights of tissue samples collected for carbon isotope measurements were added back to these weights as needed. Initial biomass values were estimated using allometric models developed by Hoover et al. (2021), predicting green biomass for P. jamesii individuals based on plant volume, number of green culms, number of green leaves, and number of green inflorescences.

Statistical analysis

All analyses were conducted in R (v4.0.2; R Core Team, 2025) using the lme4 (Bates et al. 2022), emmeans (Lenth et al. 2022) and glmmTMB (Brooks et al. 2022) packages. Soil moisture and plant data used in this study are published in (Knight et al. 2026).

We analyzed VWC using a generalized linear mixed model using a Beta error distribution and logit linked function, with treatment, date and their interaction as fixed effects and block as random. We used a pairwise estimated marginal means comparison to determine specific days on which treatment values diverged. Inline graphic was plotted against VWC to identify the onset of hydraulic stress. The segmented (Muggeo 2025) package was used to perform a breakpoint analysis, identifying a critical soil moisture threshold (θcrit) of 3.56% VWC, beyond which Inline graphic begins to drop (dry) appreciably (Fig. 3a). We used this 3.56% VWC threshold to determine the average number of hours each treatment stayed above θcrit during the post-treatment growing season (29-May-2019 – 1-Nov-2019). We analyzed days above θcrit and δ13C using linear models with treatment as a fixed effect and block as a fixed factor to account for spatial structure. Final biomass was analyzed using a linear model with treatment as a fixed effect, block as a fixed factor and initial biomass included as a co-variate. We analyzed Inline graphic and mean percent green using linear mixed-effects models to account for repeated measurements through time. In these models, treatment, date, and their interaction were included as fixed effects, and plot was included as a random intercept to account for non-independence of repeated observations within plots. Block was included as a fixed factor. Inline graphic was transformed using the orderNorm function in the bestNormalize (Peterson 2021) package to increase conformance to normality. Significance of fixed effects was assessed using analysis of variance, and post hoc comparisons among treatments were conducted using estimated marginal means with Tukey adjustment for multiple comparisons.

Fig. 3.

Fig. 3

Open in a new tab

(a): Plot of Leaf Water Potential against Volumetric Water Content for focal P. jamesii individuals across all YPD sampling dates. Each point represents an observation of an individual on a given sampling date during the dry down period. VWC was found by averaging soil moisture values over the three-hour period in which the pre-dawn YPD measurements were recorded. θcrit was determined by estimating the breakpoint, which was identified as 3.56% VWC (represented by vertical dashed line). (b): Days soil moisture was above θcrit by treatment. Statistical significance is derived from a linear mixed effects model predicting total days using removal treatment and block as fixed effects and plot as a random effect (Table 1)

Results

Treatment effectiveness

The herbicide applications resulted in shifts in plot-level plant community composition. Post-treatment cover data indicate that the herbicide application reduced percent cover of the target plant functional types relative to their pre-treatment values and the post-treatment values between treatments (Fig. S1). Mean fall cover values for C3 grasses and forbs in the C3 removal plots were 0.006% and 0.013%, respectively; mean fall cover for C4 grasses in C4 removal plots was 0.002%. The C3 removal plots exhibited a nominal reduction of total cover, decreasing from 3.40% to 2.75% (p = 0.07). The C4 removal and control plots exhibited no detectible change in total cover (from 3.54% to 3.94%, p = 0.51; from 3.30 to 3.86%, p = 0.23; respectively).

Soil moisture

VWC exhibited significant growing season differences between treatments during the study period (Table 1). Weather during the study period (29 May 2019–1 Nov 2019) was unusually dry, with the site receiving only 29.4 mm of precipitation, which is 72% below average (55 year average for this period = 106.0 mm; Menne et al. 2012). This unusually dry growing season was preceded by a wetter than average winter, with precipitation from Nov 2018 – May 2019 greater than 150% of normal (Finger Higgens et al. 2025), suggesting deep moisture recharge leading into the study period. Despite the season-long dry conditions, there were two key moisture pulses in May and July, which led to divergence in soil moisture between the treatments (Fig. 2). VWC in C3 removal plots was significantly higher than in C4 removal plots from Jun 11 to Jun 23 and from Aug 5 to Aug 17 and significantly higher than in control plots from Jun 9 to Jun 29 and from Aug 5 to Aug 19 (p < 0.05, Fig. 2).

Table 1.

Chi-square (χ2) values and associated degrees of freedom (df) and p-values (P) for mixed-effects models predicting soil moisture responses

Soil moisture metric Effect χ2 df P
Volumetric Water Content (%) Treatment 448 2  < 0.001*
Date 15127 95  < 0.001*
Treatment:Date 109 190 0.749
Days above θcrit Treatment 8.085 2 0.018*
Open in a new tab

Fig. 2.

Fig. 2

Open in a new tab

Soil moisture dynamics over the course of the growing season (May-Nov). Daily Volumetric Water Content (VWC) means (5–25 cm depth) by treatment are plotted with colors representing treatment. Solid vertical line represents the date the treatments commenced, dashed horizontal line represents θcrit determined using a regression of the relationship between YPD and VWC (Fig. 3a). Brackets indicate the results of a pairwise emmeans comparison predicting differences between treatments by date. Bracket indicated by ( +) signifies dates where C3/Forb Removal differed from control and bracket indicated by (*) signifies dates where C3/Forb Removal differed from C4 Removal (p < 0.05)

During the study period growing season (29-May-2019 – 1-Nov-2019), VWC in the C3 removal plots averaged 54.3 days above the estimated θcrit (3.56%), which is the VWC threshold at which P. jamesii could no longer consistently maintain turgor (Fig. 3a). This was significantly higher than the number of days above θcrit in both C4 removal (40.3 days) and control (40.2 days) plots (Table 1, Fig. 3b). These differences in days accumulated from two separate drying periods, one following higher levels of spring moisture and one following the Jul 23-Jul 27 monsoon events (Fig. 2).

Physiological responses

A linear mixed effects model predicting variation in Inline graphic of focal P. jamesii individuals across all sampling dates shows a significant effect of date, but no effect of treatment and no treatment x date interaction (Table 2). While P. jamesii focal individuals in C3 removal plots had greater Inline graphic values (mean Inline graphic = -1.03, Fig. 4a) than focal individuals in both C4 removal (mean Inline graphic = -1.51, Fig. 4a) and control plots (mean Inline graphic = -1.40, Fig. 4a), these differences were non-significant (p = 0.414). Observed δ13C values for individuals in C3 removal plots were 0.26‰ greater than individuals in C4 removal (p = 0.153, Fig. 4b) and 0.48‰ greater than in control plots (p = 0.009, Fig. 4b). The higher δ13C values in the C3 removal plots are consistent with less accumulated water stress (Cernusak et al. 2013).

Table 2.

F stats, degrees of freedom (df) and p-values (P) for linear mixed effects models predicting biological responses of focal P. jamesii individuals

Biological response Effect F df P
Inline graphic Treatment 0.96 2 0.414
Date 39.05 5  < 0.001*
Treatment:Date 0.57 10 0.833
Block 0.88 5 0.526
δ13C Treatment 7.24 2 0.011*
Block 0.63 5 0.682
Final Biomass Treatment 4.34 2 0.048*
Initial Biomass 38.54 1  < 0.001*
Block 2.59 5 0.102
Absolute % Green Treatment 6.74 2 0.014*
Date 60.69 11  < 0.001*
Treatment:Date 1.92 22 0.012*
Block 3.17 5 0.057
Open in a new tab

Fig. 4.

Fig. 4

Open in a new tab

(a) Leaf water potentials of P. jamesii measured periodically through the post-precipitation event (Jul 28) dry down. (b) δ13C values of target P. jamesii tissue samples measured against a primary reference scale (Vienna Pee Dee Belemnite). Greater values indicate greater water availability. (c) Differences across treatments for final biomass values. Pre-treatment values were estimated via allometric equations derived from morphology data, and final biomass values were determined by harvesting and weighing aboveground biomass of focal individuals. (d) Differences across treatments for mean percent green over the course of the post-treatment growing season. Statistical significance is derived from linear mixed effects models and pairwise emmeans comparisons to determine differences between treatments (Table 2)

Biomass and phenology

P. jamesii end of season biomass was 21.2% greater in C3 removal plots than in control plots (p = 0.039, Fig. 4c) while biomass in the C4 removal plots was intermediate and not significantly different from either control (p = 0.325) or C3 removal (p = 0.395) when controlling for initial biomass (Fig. 4c). Biweekly phenology observations revealed differences in treatments and dates and a treatment-by-date interaction (Table 2). Pairwise comparisons among treatments averaged across dates suggest focal P. jamesii individuals in C3 removal plots had on average 29.3% higher percent green than individuals in control plots (p = 0.011, Fig. 4d, Fig. S2). Again, percent green was intermediate in C4 removal plots and was not different than control (p = 0.350) or C3 removal (p = 0.122 Fig. 4d, Fig. S2).

Discussion

In this study, we tested whether the removal of subsets of the neighboring plant community would increase soil water availability and improve performance of the dominant C4 perennial grass, P. jamesii. Specifically, we asked whether reductions in neighbor cover (simulating drought-induced mortality, Finger Higgens et al. 2025) impacted soil water availability, alleviated leaf-level water stress, and subsequently enhanced markers of plant physiology, growth, and biomass. Following the experimental reduction of plant cover in the neighboring plant community, we observed an increase in soil moisture with the removal of C3 grasses and forbs but not with the removal of C4 grasses (Table 1; Fig. 3b). The removal treatments also impacted plant physiology (Table 2; Fig. 4a and b), phenology (Table 2; Fig. 4d) and biomass (Table 2; Fig. 4c) of the focal C4 grass, P. jamesii. Thus, we observed treatment effects in biomass and phenology that generally paralleled patterns in soil moisture and water availability, suggesting that differences in water availability may scale up to influence P. jamesii productivity over the course of a growing season. These responses are consistent with competitive release following reductions in C3 neighbors.

Overall, these findings support predictions that selective removal through mechanisms such as functional group-specific mortality can result in an increase in soil moisture, the most limiting resource, and may promote lower water stress and higher plant growth of surviving individuals. Although our experimental manipulation does not directly simulate climate change, it provides insight into how plant competitive interactions may shift following mortality events, including high mortality of C3 perennial grasses observed in drought experiments on the Colorado Plateau (Hoover et al. 2015, Winkler et al. 2019, Finger Higgens et al. 2025). Notably, we saw these changes despite there being only three soil wetting events and just 15 days of divergence in soil moisture among the treatments during the study period growing season. Even during a relatively dry year, C3 removal plots remained above θcrit (i.e. onset of drought stress) for 32% longer than C4 removal plots and 36% longer than control plots underscoring the importance of pulse events as has been shown in previous work (Collins et al. 2014; Hoover et al. 2015). These results also suggest that even a brief divergence in water availability following a precipitation event can have amplified effects on plant productivity. As global warming is expected to increase variability of monsoons, with less frequent but heavier rainfall (Polade et al. 2014; Harp and Horton 2022) these brief divergences will become increasingly important.

While several response variables showed patterns consistent with increased water availability following C3 removal, these differences were not uniform across metrics. Inline graphic did not differ significantly between treatments, although focal individuals in the C3 removal plots tended to maintain higher Inline graphic values during dry-down periods. δ13C values, biomass, and percent green were generally highest in the C3 removal treatment, with the C4 removal treatment exhibiting intermediate responses that were not statistically distinguishable from either the control or C3 removal treatments. The relatively weak separation among treatments may reflect the short duration of the study and the unusually dry growing season, which likely constrained overall plant growth. Previous work has shown that responses of P. jamesii biomass to altered water availability can strengthen over multiple years of drought exposure (Hoover et al. 2021). Additionally, only a small number of precipitation pulses generated divergence in soil moisture among treatments during the study period, potentially limiting the magnitude of downstream physiological and growth responses. Taken together, these findings suggest that competitive release following reductions in C3 cover may influence P. jamesii performance, though the strength and persistence of these effects likely depend on both climatic context and longer-term shifts in community composition.

The competitive release patterns we observed in the C3 removal plots are indicative of patterns that would emerge in response to drought during the cool season, when C3 grasses complete the majority of their phenological stages and forbs have high water requirements associated with establishment and rapid growth (Comstock and Ehleringer 1992; Flores and Briones 2001; Lloret et al. 2004). While current models vary in their predictions of future monsoon-season precipitation (Cook and Seager 2013; Luong et al. 2017; Pascale et al. 2017), the consensus is that cool-season precipitation will decrease in conjunction with increasing evaporative demand (Seager et al. 2007; Seager and Vecchi 2010). This increasing cool-season aridity stands to have appreciable outcomes for Colorado Plateau ecosystems, where cool-season precipitation determines the majority of total annual primary productivity (Caldwell 1985). Following cool-season drought-induced mortality, warm-season moisture pulses may serve to further shift community composition towards C4 species, which exhibit greater phenological flexibility and are better equipped than C3 species to capitalize on warm-season precipitation events (Finger Higgens et al. 2025).

Competitive release with drought-induced mortality

This study supports previous work suggesting that competitive release is an important process structuring dryland communities (Segre et al. 2016). In our study system, the releasing factor is herbicide-induced mortality of C3 species, mimicking patterns of morality expected under climate change (Winkler et al. 2019; Finger Higgens et al. 2025). Several studies have shown that rising temperatures, increased water scarcity and temporal variability of precipitation events correspond with reductions in C3 grass cover (Wertin et al. 2015; Witwicki et al. 2016), yet it is not well-understood how that reduction in cover may affect community dynamics. Although elevated atmospheric CO₂ is often expected to enhance C₃ productivity relative to C₄ specie, this response is highly context dependent (Reich et al. 2018; Monson et al. 2025). Under well-watered and nutrient-rich conditions, C₃ plants generally benefit from higher internal CO₂ availability. However, under the hotter, drier conditions projected for many dryland systems, this CO₂ advantage can be offset by increased stomatal regulation and limited hydraulic conductance, reducing the potential for sustained carbon gain (Hamim 2005; Taylor et al. 2014). Meanwhile, C₄ grasses maintain higher water-use efficiency and can continue assimilating carbon even as drought stress intensifies. Thus, rising CO₂ may not necessarily promote C₃ expansion in water-limited environments and may instead reinforce or accelerate shifts toward C₄ dominance.

Our study has shown that a reduction in C3 cover via selective mortality can increase water availability and uptake for P. jamesii, translating into increased biomass production. Meanwhile, similar experimental reductions in C4 cover did not result in the same increases in water availability and biomass production. While these results initially seem counterintuitive, as it might be expected that the removal of other C4 grasses, which presumably share greater niche overlap with the focal C4 species, would have had a greater impact than the removal of C3 grasses and forbs, we propose that this finding is due to C4 plants having greater water use efficiency and lower transpiration rates than C3 plants. Therefore, removal of the less-efficient C3 plants led to a larger increase in soil moisture levels and the subsequent observed competitive release. Additionally, C3 species have lower nitrogen use efficiency (Monson et al. 2025), such that removal of C3 neighbors may have had greater effects on nitrogen availability than C4 removal. Taken together, these results suggest that, with ongoing drought, surviving C4 grasses may be able to widen their realized niche space by capitalizing on the mortality of less drought-resistant competitors, creating a shift in community structure towards a greater C4 grass dominance.

On the Colorado Plateau, a shift towards greater relative abundance of C4 grasses is predicted in response to climate projections (Gremer et al. 2015; Havrilla et al. 2023) and will likely have cascading effects on other aspects of community structure and ecosystem function. For example, a shift towards a community dominated by warm-season grasses would affect timing and quality of forage for wildlife and livestock (Schwinning et al. 2008). Additionally, a competitive release-driven community shift could serve as the mechanistic basis for shifts in nutrient cycling and even climatic processes (Miki and Kondoh 2002)—models suggest that shifts in community structure can significantly affect climate at regional and continental scales due to species differences in stomatal conductance and rooting depth (Foley et al. 2003; Parry et al. 2007). Understanding how climate change may result in ‘winners’ and ‘losers’ through mechanisms such as competitive release provides a novel pathway for exploring future ecosystem dynamics.

Conclusion

Climate models predict lower and increasingly variable levels of water availability in dryland systems globally. Some species may be more vulnerable to these changes than others because of differences in physiology, including photosynthetic pathway. C3 grasses are expected to suffer reductions in cover and increased mortality in the face of climate change. Here, we show that even during a relatively dry summer, mortality of C3 species could increase water availability to surviving C4 grasses due to competitive release. Competitive release may be one mechanism through which C4 grasses expand their niche space as drylands become warmer and drier. However, there may be a threshold where conferred benefits of this mechanism are outweighed by other resource scarcities, and more work is required to determine how this competitive release mechanism functions at varying drought severities. Further understanding of both the limits of this mechanism and the ability of C4 grasses to expand their niche space will be crucial in informing predictions of future climate-induced changes in vegetation dynamics.

Supplementary Information

Below is the link to the electronic supplementary material.

Supplementary file1 (DOCX 1323 KB) (1.3MB, docx)

Acknowledgements

The authors thank Emily Brooke and Jessica Mikenas for assisting in project establishment and data collection. They thank Ed Grote for technical support and installation of soil moisture sensors and Kelli Quinn for guidance in herbicide application. Any use of trade names is for descriptive purposes only and does not imply endorsement by the US Government.

Authors’ contributions

SH, AP, DH, and MD conceived and designed the experiment. SH, AP, and AK collected and analyzed the data. AK curated data for release. SH wrote the original manuscript draft, with review and edits from all other authors.

Funding

This work was supported by the US Geological Survey Ecosystems Mission Area.

Data availability

Data generated during this study will be available from the USGS ScienceBase-Catalog (Knight et al., 2026).

Code availability

Not applicable.

Declarations

Conflicts of interest

The authors declare that they have no conflict of interest.

Consent to participate

Not applicable.

Consent for publication

Not applicable.

Ethics approval

Not applicable.

Footnotes

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

References

  1. Anderegg WRL, Kane JM, Anderegg LDL (2013) Consequences of widespread tree mortality triggered by drought and temperature stress. Nat Clim Change 3:30–36 [Google Scholar]
  2. Bates D, Maechler M, Bolker B et al (2022) lme4: linear mixed-effects models using ‘Eigen’ and S4.
  3. Birch HF (1964) Mineralisation of plant nitrogen following alternate wet and dry conditions. Plant Soil 20:43–49 [Google Scholar]
  4. Bolnick DI, Ingram T, Stutz WE, Snowberg LK, Lau OL, Paull JS (2010) Ecological release from interspecific competition leads to decoupled changes in population and individual niche width. Proc R Soc B Biol Sci 277:1789–1797 [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Breshears DD, Whicker JJ, Zou CB, Field JP, Allen CD (2009) A conceptual framework for dryland aeolian sediment transport along the grassland–forest continuum: effects of woody plant canopy cover and disturbance. Geomorphology 105:28–38 [Google Scholar]
  6. Brooks M, Bolker B, Kristensen K et al (2022) glmmTMB: generalized linear mixed models using template model builder.
  7. Buchmann N, Brooks JR, Rapp KD, Ehleringer JR (1996) Carbon isotope composition of C4 grasses is influenced by light and water supply. Plant Cell Environ 19:392–402. 10.1111/j.1365-3040.1996.tb00331.x [DOI] [Google Scholar]
  8. Caldwell M (1985) Cold desert. In: Chabot BF, Mooney HA (eds) Physiological ecology of North American plant communities. Springer Netherlands, Dordrecht, pp 198–212 [Google Scholar]
  9. Cernusak LA, Ubierna N, Winter K, Holtum JAM, Marshall JD, Farquhar GD (2013) Environmental and physiological determinants of carbon isotope discrimination in terrestrial plants. New Phytol 200:950–965 [DOI] [PubMed] [Google Scholar]
  10. Chesson P (2000) Mechanisms of maintenance of species diversity. Annu Rev Ecol Syst 31:343–366 [Google Scholar]
  11. Collins SL, Belnap J, Grimm NB et al (2014) A multiscale, hierarchical model of pulse dynamics in arid-land ecosystems. Annu Rev Ecol Evol Syst 45:397–419 [Google Scholar]
  12. Collins M, Knutti R, Arblaster J et al (2013) Long-term climate change: projections, commitments and irreversibility. Clim Change 2013 - Phys Sci Basis Contrib Work Group Fifth Assess Rep Intergov Panel Clim Change 1029–1136.
  13. Comstock J, Ehleringer J (1992) Plant adaptation in the Great Basin and colorado plateau. Gt Basin Nat 52:195–215 [Google Scholar]
  14. Cook BI, Seager R (2013) The response of the North American Monsoon to increased greenhouse gas forcing. J Geophys Res Atmos 118:1690–1699. 10.1002/jgrd.50111 [DOI] [Google Scholar]
  15. Dettman DL, Kohn MJ, Quade J, Ryerson FJ, Ojha TP, Hamidullah S (2001) Seasonal stable isotope evidence for a strong Asian monsoon throughout the past 10.7 m.y. Geology 29:31–34 [Google Scholar]
  16. Duniway MC, Benson C, Nauman TW et al (2022) Geologic, geomorphic, and edaphic underpinnings of dryland ecosystems: Colorado Plateau landscapes in a changing world. Ecosphere 13:e4273. 10.1002/ecs2.4273 [DOI] [Google Scholar]
  17. Ellsworth PZ, Cousins AB (2016) Carbon isotopes and water use efficiency in C4 plants. Curr Opin Plant Biol 31:155–161 [DOI] [PubMed] [Google Scholar]
  18. Finger Higgens RA, Hoover DL, Knight AC et al (2025) Flexible phenology of a C4 grass linked to resiliency to seasonal and multiyear drought events in the American Southwest. Ecol Evol 15:e71435. 10.1002/ece3.71435 [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Flores J, Briones O (2001) Plant life-form and germination in a Mexican inter-tropical desert: effects of soil water potential and temperature. J Arid Environ 47:485–497 [Google Scholar]
  20. Foley JA, Costa MH, Delire C, Ramankutty N, Snyder P (2003) Green surprise? How terrestrial ecosystems could affect earth’s climate. Front Ecol Environ 1:38–44 [Google Scholar]
  21. Gessler A, Schaub M, McDowell NG (2017) The role of nutrients in drought-induced tree mortality and recovery. New Phytol 214:513–520. 10.1111/nph.14340 [DOI] [PubMed] [Google Scholar]
  22. Ghannoum O, von Caemmerer S, Conroy JP (2002) The effect of drought on plant water use efficiency of nine NAD–ME and nine NADP–ME Australian C4 grasses. Funct Plant Biol 29:1337–1348. 10.1071/fp02056 [DOI] [PubMed] [Google Scholar]
  23. Gremer JR, Bradford JB, Munson SM, Duniway MC (2015) Desert grassland responses to climate and soil moisture suggest divergent vulnerabilities across the southwestern United States. Glob Change Biol 21:4049–4062. 10.1111/gcb.13043 [DOI] [PubMed] [Google Scholar]
  24. Grote E, Duniway MC, Reed SC et al (2023) Meteorological measurements from five weather stations in Grand and San Juan Counties in Southeastern Utah
  25. Gulmon SL, Chiariello NR, Mooney HA, Chu CC (1983) Phenology and resource use in three co-occurring grassland annuals. Oecologia 58:33–42 [DOI] [PubMed] [Google Scholar]
  26. Hamim (2005) Photosynthesis of C3 and C4 Species in response to increased CO2 concentration and drought stress. HAYATI J Biosci. 10.1016/S1978-3019(16)30340-0 [DOI] [Google Scholar]
  27. Harp RD, Horton DE (2022) Observed changes in daily precipitation intensity in the United States. Geophys Res Lett. 10.1029/2022GL099955 [DOI] [Google Scholar]
  28. Havrilla CA, Bradford JB, Yackulic CB, Munson SM (2023) Divergent climate impacts on C3 versus C4 grasses imply widespread 21st century shifts in grassland functional composition. Divers Distrib 29:379–394. 10.1111/ddi.13669 [DOI] [Google Scholar]
  29. Hoover DL, Duniway MC, Belnap J (2015) Pulse-drought atop press-drought: unexpected plant responses and implications for dryland ecosystems. Oecologia 179:1211–1221 [DOI] [PubMed] [Google Scholar]
  30. Hoover DL, Duniway MC, Belnap J (2017) Testing the apparent resistance of three dominant plants to chronic drought on the Colorado Plateau. J Ecol 105:152–162 [Google Scholar]
  31. Hoover DL, Koriakin K, Albrigtsen J, Ocheltree T (2019) Comparing water-related plant functional traits among dominant grasses of the Colorado Plateau: implications for drought resistance. Plant Soil 441:207–218 [Google Scholar]
  32. Hoover DL, Pfennigwerth AA, Duniway MC (2021) Drought resistance and resilience: the role of soil moisture–plant interactions and legacies in a dryland ecosystem. J Ecol 109:3280–3294 [Google Scholar]
  33. Huang J, Li Y, Fu C et al (2017) Dryland climate change: recent progress and challenges. Rev Geophys 55:719–778 [Google Scholar]
  34. Kanissery R, Gairhe B, Kadyampakeni D, Batuman O, Alferez F (2019) Glyphosate: its environmental persistence and impact on crop health and nutrition. Plants 8:499 [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Karl TM, National Academies of Sciences, Engineering, and Medicine (2011) Climate stabilization targets: emissions, concentrations, and impacts over decades to millennia. The National Academies Press, Washington, DC. 10.17226/12877 [DOI] [Google Scholar]
  36. Kemp PR (1983) Phenological patterns of Chihuahuan Desert plants in relation to the timing of water availability. J Ecol 71:427–436 [Google Scholar]
  37. Knight AC, Hoy-Skubik SL, Hoover DL, Pfennigwerth AA, Duniway MC (2026) Vegetation and soil moisture data from 2019 selective mortality experiment in perennial grassland on the Colorado Plateau, Utah, USA: U.S. Geological Survey data release. 10.5066/P97GVNBG. [DOI]
  38. Lenth RV, Buerkner P, Giné-Vázquez I et al (2022) emmeans: estimated marginal means, aka least-squares means.
  39. Lloret F, Peñuelas J, Estiarte M (2004) Experimental evidence of reduced diversity of seedlings due to climate modification in a Mediterranean-type community. Glob Change Biol 10:248–258 [Google Scholar]
  40. Luong TM, Castro CL, Chang H-I et al (2017) The more extreme nature of North American Monsoon precipitation in the southwestern United States as revealed by a historical climatology of simulated severe weather events. J Appl Meteorol Climatol. 10.1175/JAMC-D-16-0358.1 [DOI] [Google Scholar]
  41. Menne MJ, Durre I, Korzeniewski B et al (2012) Global Historical Climatology Network - Daily (GHCN-Daily), Version 3. Canyonlands - the Needles, 05/29/1990–11/01/2019.
  42. Miki T, Kondoh M (2002) Feedbacks between nutrient cycling and vegetation predict plant species coexistence and invasion. Ecol Lett 5:624–633. 10.1046/j.1461-0248.2002.00347.x [DOI] [Google Scholar]
  43. Monson RK, Li S, Ainsworth EA et al (2025) C4 photosynthesis, trait spectra, and the fast-efficient phenotype. New Phytol 246:879–893. 10.1111/nph.70057 [DOI] [PubMed] [Google Scholar]
  44. Muggeo VMR (2025) Segmented: Regression Models with Break-Points / Change-Points Estimation (with Possibly Random Effects)
  45. Munson SM, Belnap J, Schelz CD, Moran M, Carolin TW (2011) On the brink of change: plant responses to climate on the Colorado Plateau. Ecosphere 2:art68 [Google Scholar]
  46. Noy-Meir I (1973) Desert ecosystems: environment and producers. Annu Rev Ecol Syst 4:25–51 [Google Scholar]
  47. Parry ML, Canziani OF, Palutikof JP, Linden PJ van der, Hanson CE (2007) Climate Change 2007: Impacts, Adaptation and Vulnerability. Contribution of Working Group II to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. S Solomon (ed). Cambridge ; New York: Cambridge University Press.
  48. Pascale S, Boos WR, Bordoni S et al (2017) Weakening of the North American monsoon with global warming. Nat Clim Change 7:806–812. 10.1038/nclimate3412 [DOI] [Google Scholar]
  49. Pearcy RW, Ehleringer J (1984) Comparative ecophysiology of C3 and C4 plants. Plant Cell Environ 7:1–13 [Google Scholar]
  50. Peterson RA (2021) Finding optimal normalizing transformations via bestNormalize. R J 13:310. 10.32614/RJ-2021-041 [DOI] [Google Scholar]
  51. Polade SD, Pierce DW, Cayan DR et al (2014) The key role of dry days in changing regional climate and precipitation regimes. Sci Rep 4:4364. 10.1038/srep04364 [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. R Core Team (2025) R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing https://www.r-project.org/
  53. Reich PB (2014) The world-wide ‘fast–slow’ plant economics spectrum: a traits manifesto. J Ecol 102:275–301 [Google Scholar]
  54. Reich PB, Hobbie SE, Lee TD, Pastore MA (2018) Unexpected reversal of C3 versus C4 grass response to elevated CO2 during a 20-year field experiment. Science 360:317–320. 10.1126/science.aas9313 [DOI] [PubMed] [Google Scholar]
  55. Reynolds HL, Mit℡bach GG, Darcy-Hall TL, Houseman GR, Gross KL (2007) No effect of varying soil resource heterogeneity on plant species richness in a low fertility grassland. J Ecol 95:723–733 [Google Scholar]
  56. Safriel U, Adeel Z, Niemeijer D et al (2005) Dryland Systems. Ecosyst Hum Well- Curr State Trends 1.
  57. Sage RF (2004) The evolution of C4 photosynthesis. New Phytol 161:341–370 [DOI] [PubMed] [Google Scholar]
  58. Scholander PF, Bradstreet ED, Hemmingsen EA, Hammel HT (1965) Sap pressure in vascular plants. Science 148:339–346 [DOI] [PubMed] [Google Scholar]
  59. Schwalm CR, Williams CA, Schaefer K et al (2010) Assimilation exceeds respiration sensitivity to drought: a FLUXNET synthesis. Glob Change Biol 16:657–670 [Google Scholar]
  60. Schwinning S, Ehleringer JR (2001) Water use trade-offs and optimal adaptations to pulse-driven arid ecosystems. J Ecol 89:464–480 [Google Scholar]
  61. Schwinning S, Kelly CK (2013) Plant competition, temporal niches and implications for productivity and adaptability to climate change in water-limited environments. Funct Ecol 27:886–897 [Google Scholar]
  62. Schwinning S, Belnap J, Bowling D, Ehleringer J (2008) Sensitivity of the Colorado Plateau to change: climate, ecosystems, and society. Ecol Soc. 10.5751/ES-02412-130228 [DOI] [Google Scholar]
  63. Seager R, Vecchi GA (2010) Greenhouse warming and the 21st century hydroclimate of southwestern North America. Proc Natl Acad Sci USA 107:21277–21282. 10.1073/pnas.0910856107 [DOI] [PMC free article] [PubMed] [Google Scholar]
  64. Seager R, Ting M, Held I et al (2007) Model projections of an imminent transition to a more arid climate in southwestern North America. Science 316:1181–1184. 10.1126/science.1139601 [DOI] [PubMed] [Google Scholar]
  65. Segre H, DeMalach N, Henkin Z, Kadmon R (2016) Quantifying competitive exclusion and competitive release in ecological communities: a conceptual framework and a case study. PLoS ONE 11:e0160798 [DOI] [PMC free article] [PubMed] [Google Scholar]
  66. Sepulchre P, Ramstein G, Fluteau F, Schuster M, Tiercelin J-J, Brunet M (2006) Tectonic uplift and Eastern Africa aridification. Science 313:1419–1423 [DOI] [PubMed] [Google Scholar]
  67. Silvertown J, Araya Y, Gowing D (2015) Hydrological niches in terrestrial plant communities: a review. J Ecol 103:93–108 [Google Scholar]
  68. Smith MD (2011) An ecological perspective on extreme climatic events: a synthetic definition and framework to guide future research. J Ecol 99:656–663 [Google Scholar]
  69. Taylor SH, Ripley BS, Martin T et al (2014) Physiological advantages of C4 grasses in the field: a comparative experiment demonstrating the importance of drought. Glob Change Biol 20:1992–2003. 10.1111/gcb.12498 [DOI] [PMC free article] [PubMed] [Google Scholar]
  70. Walter H (2012) Vegetation of the earth and ecological systems of the geo-biosphere. Springer Science & Business Media [Google Scholar]
  71. Ward D (2016) The biology of deserts. Oxford University Press [Google Scholar]
  72. Wertin TM, Reed SC, Belnap J (2015) C₃ and C₄ plant responses to increased temperatures and altered monsoonal precipitation in a cool desert on the Colorado Plateau, USA. Oecologia 177:997–1013 [DOI] [PubMed] [Google Scholar]
  73. Williams AP, Cook ER, Smerdon JE et al (2020) Large contribution from anthropogenic warming to an emerging North American megadrought. Science 368:314–318 [DOI] [PubMed] [Google Scholar]
  74. Winkler DE, Belnap J, Hoover D, Reed SC, Duniway MC (2019) Shrub persistence and increased grass mortality in response to drought in dryland systems. Glob Change Biol 25:3121–3135 [DOI] [PubMed] [Google Scholar]
  75. Witwicki DL, Munson SM, Thoma DP (2016) Effects of climate and water balance across grasslands of varying C3 and C4 grass cover. Ecosphere 7:e01577 [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary file1 (DOCX 1323 KB) (1.3MB, docx)

Data Availability Statement

Data generated during this study will be available from the USGS ScienceBase-Catalog (Knight et al., 2026).

Not applicable.

Articles from Oecologia are provided here courtesy of Springer

RESOURCES