Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Open Access

Peer-reviewed

Research Article

Plant community re-organization and increased productivity due to multi-year nutrient enrichment of a coastal grassland

  • Joseph K. Brown ,

    Contributed equally to this work with: Joseph K. Brown, Ashley Moulton, Julie C. Zinnert

    Roles Conceptualization, Data curation, Formal analysis, Investigation, Visualization, Writing – original draft

    Affiliations Department of Biology, Virginia Commonwealth University, Richmond, Virginia, United States of America, Department of Biological Sciences, Minnesota State University, Mankato, Minnesota, United States of America

    ⨯
  • Ashley Moulton ,

    Contributed equally to this work with: Joseph K. Brown, Ashley Moulton, Julie C. Zinnert

    Roles Conceptualization, Investigation, Methodology, Writing – review & editing

    Affiliation Department of Biology, Virginia Commonwealth University, Richmond, Virginia, United States of America

    ⨯
  • Julie C. Zinnert

    Contributed equally to this work with: Joseph K. Brown, Ashley Moulton, Julie C. Zinnert

    Roles Conceptualization, Data curation, Funding acquisition, Investigation, Methodology, Project administration, Resources, Supervision, Writing – review & editing

    jczinnert@vcu.edu

    Affiliation Department of Biology, Virginia Commonwealth University, Richmond, Virginia, United States of America

    ⨯

Abstract

Nutrient enrichment alters plant community structure and function at a global scale. Coastal plant systems are expected to experience increased rates of nitrogen and phosphorus deposition by 2100, caused mostly by anthropogenic activity. Despite high density of studies investigating connections between plant community structure and ecosystem function in response to nutrient addition, inconsistencies in system response based on the ecosystem in question calls for more detailed analyses of nutrient impacts on community organization and resulting productivity response. Here, we focus on nutrient addition impacts on community structure and organization as well as productivity of different lifeforms in a coastal grassland. We established long-term nutrient enrichment plots in 2015 consisting of control (C), nitrogen (N), phosphorus (P), and nitrogen + phosphorus (NP) treatments. In 2017 we collected graminoid and forb productivity, root productivity, and community composition for each plot. We found no N x P interaction, but N enrichment was a significant main effect on productivity, highlighting N limitation in coastal systems. Importantly, nutrient enrichment treatments did not alter root productivity. However, all treatments caused significant differences in community composition. Using rank abundance curves, we determined that community composition differences were driven by increased dominance of nitrophilous graminoids, re-organization of subordinate species, and species absences in N and NP plots. Results of this study highlight how coastal grassland communities are impacted by nutrient enrichment. We show that community re-organization, increased dominance, and absence of critical species are all important mechanisms that reflect community-level impacts of nutrient enrichment in our coastal grassland site.

Citation: Brown JK, Moulton A, Zinnert JC (2022) Plant community re-organization and increased productivity due to multi-year nutrient enrichment of a coastal grassland. PLoS ONE 17(7): e0270798. https://doi.org/10.1371/journal.pone.0270798

Editor: Daehyun Kim, Seoul National University, REPUBLIC OF KOREA

Received: February 1, 2022; Accepted: June 19, 2022; Published: July 28, 2022

Copyright: © 2022 Brown et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: Data for this project are available on the Virginia Coast Reserve data portal: Zinnert, J., Brown, J., and A. Moulton. 2018. Long-Term Fertilization Plots in Swale Vegetation on Hog Island, VA 2015. Virginia Coast Reserve Long-Term Ecological Research Project Data Publication knb-lter-vcr.311.3 (http://doi.org/10.6073/pasta/eeb59ddf840c51b93c4bc7b765d4d59f).

Funding: National Science Foundation, grant numbers DEB-1237733 and DEB-1832221 to JCZ https://www.nsf.gov/div/index.jsp?div=DEB Virginia Commonwealth University Dean’s Funds to JCZ https://www.vcu.edu The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The authors have declared that no competing interests exist.

Introduction

As global N and P enrichment becomes more common, researchers are investigating the effects nutrients have on plant communities around the world [1–3]. Increased nutrient enrichment can alter plant community composition, plant functional trait expressions, and ecosystem functionality [1, 3–8]. Nutrient enrichment influences species composition as well as community structure and organization (i.e., species diversity, richness, evenness, and dominance), often at large spatial scales [4, 7, 9–15]. However, scientific literature reflects a lack of consistency regarding community response to nutrient enrichment. For example, some studies show increased N and P enrichment results in decreased species diversity, evenness, and richness, with species disappearance identified as a common occurrence [12, 16–21], while others do not [22–24]. Nutrient enrichment commonly impacts primary productivity levels [1, 2, 5, 25], but some studies have shown no impact on forb or graminoid species [26]. Increased productivity after nutrient enrichment can result in higher competition for space and light, mechanisms recognized as drivers of altered community composition and structure [5, 11, 12, 18, 19, 27, 28]. Given tight connections between ecosystem productivity and community structure, it is important that existing variabilities are addressed to understand how nutrients impact plant communities in a variety of ecosystems [29].

The unpredictability of system response to nutrient enrichment highlights how differences in regional species pools and/or pre-existing environmental conditions may influence the final community structure and ecosystem functionality as a system responds to multiple nutrient enrichment treatments. Thus, it is critical to investigate how N and P (both individually and synergistically) influence plant community dynamics in coastal grassland ecosystems which have different abiotic stressors than other grasslands. Coastal grasslands occurring behind dune systems on barrier islands are uniquely impacted by environmental factors like sea-spray, varying access to freshwater, and sand burial that could influence plant community response to nutrient enrichment [30]. Furthermore, barrier islands are present globally in the same areas that are expected to increase in nutrient deposition [31]. Barrier islands occur along 30% of U.S. coastlines with more than half existing on the Atlantic coast where nutrient enrichment is expected to spike [31, 32]. Barrier island dynamics including disturbance response and resilience are impacted by feedbacks between dominant physical processes and extant plant communities [30, 33–35]. By investigating barrier island response to nutrient enrichments, it is possible to better understand how increased nutrient deposition impacts ecosystems experiencing multiple disturbances over short temporal scales. High nutrient enrichment via run-off and atmospheric deposition onto barrier islands can play a substantial role in driving plant community differences in composition of species and expressed functional traits. Impacts of nutrient enrichment may be heightened in coastal grasslands given the prominence of N limitation in sandy soils that serve as the foundation of barrier islands [8, 33, 36–38]. As a result, hypotheses have been established that attempt to explain how plant communities altered by nutrient enrichment leads to subsequent changes in barrier island ecology, specifically disturbance response and successional processes in grasslands developing behind dunes [13, 39].

In this study we focus on a mid-Atlantic coastal grassland, as this geographic region is likely to experience dramatic increases in nutrient loading in the next 80–100 years [31]. Previous nutrient enrichment models suggest coastal areas in the northern hemisphere have been dramatically impacted by N deposition in the last 30 years, with recent work highlighting the role chemically reduced forms of N has on global N flux [40–42]. Projections of future change show that, in the continental U.S., the Atlantic coast is expected to see the largest regional increase in N enrichment by 2100 [31]. Past nutrient enrichment studies in coastal systems have shown that N enrichment can cause long-term plant community change on sand dunes [37, 43, 44]. Day et al. [37] found that after N enrichment on a dune system, species re-ordering ensues with longstanding shifts in species dominance and species disappearance. Aboveground and belowground biomass were higher just one year after N fertilization treatment with effects lasting nearly 10 years [43]. Similar conclusions have been made in European dune systems where elevated nutrient deposition increases productivity and alters plant community structure, specifically reducing species richness [14].

Previous research in calcareous European sand dunes also shows that decalcified acidic sand dunes are more sensitive to N deposition due to the increased availability of P, highlighting the importance of both N and P for plant community response to expected nutrient enrichment [45]. Increases in P enrichment through run-off and deposition has accrued less focus but research has found that coastal systems are some of the most at-risk ecosystems for P enrichment [46]. For example, Mahowald et al. [47] suggests a net gain in atmospheric total P in terrestrial systems located near oceans, while most other terrestrial systems show a net loss of total P. Although it has been proposed P is not a limiting nutrient in terrestrial coastal systems [48–50], Kooijman et al. [45] showed sensitivity of dune vegetation to N deposition when P is abundantly available in the soil. Such findings justifies investigating N, P, and N + P (NP) treatments in coastal systems. Furthermore, N and P can influence plant communities synergistically, such that productivity is higher in NP treatments than either N or P treatments individually [1, 51]. Despite work on coastal sand dunes, knowledge gaps remain in understanding how non-dune coastal grassland systems will respond to expected nutrient enrichment.

We address these knowledge gaps by conducting a 3-year nutrient enrichment study to understand how prolonged N, P, and NP treatments affect coastal grassland communities. Specifically, we aim to understand how nutrient enrichment alters lifeform productivity and plant community composition after 3 years of nutrient treatments. In this study we focus on using plant community structure (evenness, richness, and diversity) and reorganization to explain altered community composition. We used visualization of rank abundance curves (RACs) to explore whether systematic plant community re-organization contributes to community differences [52]. We hypothesize that 1) both aboveground and belowground graminoid and forb productivity will increase with N and NP treatments, 2) that N and NP additions will cause differences in community structure metrics (species diversity, richness, and evenness), and 3) compared to control plots (serving as a reference of a natural community), N and NP, enrichment will cause differences in community composition. Based on previous findings, we do not expect P to have major impacts on biomass or community structure unless N enrichment is also present.

Methods

Study site, plot establishment, and nutrient application

Barrier islands can vary in morphological characteristics, but most have a foredune or series of dunes proceeding a marsh ecosystem on the backside with grasslands developing in lower lying areas between dunes (Fig 1). Access to freshwater for extant plant communities on barrier islands are reliant on a freshwater lens that is fed entirely by precipitation (Fig 1). Hog Island (37.417 N, 75.686 W) is a barrier island that is part of the Virginia Coast Reserve (VCR) Long-term Ecological Research site along the U.S. Atlantic coast. Nutrient fertilization plots were established on the southern end of the island in a grassland environment composed of perennial graminoid and annual/biennial forb species. The site consists of sandy, well-drained soils and has been described as stable due to presence of a linear dune ridge which provides protection from coastal disturbances [35].

thumbnail
Download:
Fig 1. Cross-section schematic of barrier island.

https://doi.org/10.1371/journal.pone.0270798.g001

Cross-section of a traditional barrier island morphology. Prominent landforms and habitats characteristic of a barrier island include a primary and secondary dune, with grasslands colonizing low lying areas between dune ridges. Plant communities in terrestrial barrier island habitats are dependent on a rain-fed freshwater lens. Terrestrial habitats eventually reach an ecotone where marsh species dominate, and semidiurnal tidal cycling is present.

Plots were established on the Hog Island grassland in 2015 following a modified Nutrient Network approach [53]. Experimental set-up followed a completely randomized block design with nutrient applications twice per growing season (May and June) from 2015–2017, we administered split fertilization application to enhance uptake efficiency by C3 and C4 species. Each nutrient treatment was administrated at 5 g m-2 yr-1, totaling 10 g m-2 yr-1 by the end of each season. The result of the experimental set-up was four treatments; nitrogen (N10 P0 or N), phosphorus (N0 P10 or P), nitrogen + phosphorus (N10 P10 or NP), and control (N0 P0 or C) with each treatment replicated five times (n = 5). Blocks were a total of 6 m2, treatment units within each block were 3 m2 with 1 m walkways, which were not recognized as usable plot space. Treatment units were further subdivided into four 1 m2 plots to allow for separate spaces for destructive productivity sampling and long-term species composition data collection. Phosphorus was applied as triple super phosphate, while N was applied as ammonium nitrate in particulate solid form.

Lifeform productivity sampling

Productivity was defined as all aboveground vegetative biomass at the end of the growing season (September 2017). Productivity was sampled from one subplot within a haphazardly placed 0.1 x 1 m frame and extrapolated to represent plot level productivity (g m-2). Samples were sorted by lifeform (graminoid or forb), oven dried separately for 72 h at 60°C, and weighed (g). Belowground productivity was also collected in 2017. We used a single root core taken in a haphazardly selected spot in the same subplot where aboveground biomass was collected. For each root core, large roots were identified visually and collected from each soil core before the remainder of the cores were washed in a series of sieve stacks to collect smaller roots. All roots were rinsed with water before samples were oven dried for 72 h at 60°C and weighed (g).

Species composition and community structure

In July of 2017 we recorded aerial species cover to the nearest 1% using an undisturbed subplot designated for species composition data collection. We calculated species evenness, richness, and diversity using the codyn package in R [54]. Species richness was computed as the number of different species in each replicate while species evenness was calculated as Evar for each replicate, this evenness index is recommended for general use compared to other evenness indices [55]. Evar produces an index between 0–1, where 0 represents minimum evenness within a replicate and 1 represents maximum evenness within a replicate [55]. Species diversity was calculated as Shannon’s diversity index for each replicate (H’).

Abundance differences were calculated for each species using control plots as a reference treatment. We pooled species abundance across all replicates to create a single species pool for our study site [54]. Developing a species pool allowed us to investigate how abundances of each species present at our sites differed between control plots and each treatment group. For example, a positive value indicated that a species had higher abundance in the treatment community than in the control community, while negative values indicated that a species had lower abundance in the treatment community than in the control community. Averaged species abundances were ranked in order of most abundant to least abundant to visualize the RACs for each treatment. RACs have been utilized in previous studies to inform differences in species compositions among treatment groups [7, 52].

Statistical analysis

All analyses were conducted using R statistical program (R Core Team 2019). We used a two-way analysis of variance (ANOVA) to determine whether mean (± SE) graminoid and forb productivity were significantly impacted by N x P interaction (α = 0.05). Two-way ANOVAs were performed on log transformed productivity data to meet assumptions of normality (checked visually using QQ-plots) and homogeneity of variance (assessed using Levene’s test). Tukey HSD was used as a post-hoc test to determine significant pairwise differences among specific nutrient treatments (α = 0.05). A two-way ANOVA was used to test for a significant N x P interaction for community structure metrics (species diversity, richness, and evenness), after confirming assumptions of normality and homogeneity of variance. Tukey HSD was used as a pairwise post-hoc test for each community structure metric (α = 0.05).

Non-metric multidimensional scaling (NMDS) ordination was used to visualize community composition of nutrient enrichment treatments. Resulting NMDS visualization identifies replicate similarity based on distance in ordination space. Points (representing experimental replicates) closer together are more similar in community composition than points that are further apart. Prior to conducting the NMDS, species abundances were standardized to represent a relative abundance for each replicate. We conducted the NMDS in vegan R package using Bray-Curtis distance measure to calculate a distance matrix of relative species abundances for each treatment replicate [56]. The NMDS was performed using 3-dimensions to a minimized stress value (maximum iteration = 999). We used Monte Carlo randomization test to determine whether the final NMDS solution had a lower stress than would be expected by chance (α = 0.05). A principal components rotation was applied to the finalized NMDS to aid interpretation. This technique rotates the finalized plot such that NMDS axis 1 and 2 represent maximum variation of the data allowing for calculation of species correlations in ordination space.

We used the envfit function in vegan to calculate species correlations in ordination space [56]. Significance of fitted species correlations was assessed within the envfit function using permutation test (permutation = 999). Species that produced significant p-values are more correlated in ordination space than would be expected by chance and thus explain a large proportion of variation in the data. These species were highlighted in RACs to examine the impact of species rank order shuffling on community composition dissimilarity.

Centroids for each treatment group were calculated by aggregating site scores. Centroids can be interpreted as mean community composition for each treatment group. We used a permutational multivariate-ANOVA (PERMANOVA) to determine significant differences among community compositions of nutrient enrichment treatments (permutation = 999, α = 0.05) [57]. A pairwise post-hoc test was conducted to determine which pairwise treatment groups had different community compositions (FDR adjusted α = 0.05). We calculated multivariate homogeneity of group dispersion to determine within-group variation among treatment replicates using the betadisper function in vegan [56, 58]. Testing homogeneity of multivariate group dispersion has been identified as a distance-based method of identifying differences in beta-diversity among groups [59]. We used an ANOVA on mean distance between each point and their respective median centroid to determine significant differences of beta-dispersion among treatment groups (α = 0.05). A Tukey HSD test was used as a post-hoc test for pairwise differences in mean beta-dispersion (α = 0.05).

Results

Lifeform and belowground productivity

There was no significant NP interaction on graminoid or forb productivity (Table 1). However, N addition was a significant main effect whereby graminoid and forb productivity were higher in plots fertilized with N compared to plots where N addition was absent (Fig 2). Graminoid productivity averaged 966 ± 152 g m-2 when N enrichment was present (N: 744 ± 102 and NP: 1187 ± 262 g m-2) and 229 ± 42 g m-2 (C: 186 ± 46 and P: 271 ± 71 g m-2) when N enrichment was absent (Fig 1). Forb productivity followed a similar pattern, when N addition was present productivity averaged 152 ± 41 g m-2 (N: 92 ± 51 and NP: 169 ± 64 g m-2) and was much lower when N fertilization was absent, averaging 12 ± 5 g m-2 (C: 5 ± 3 and P: 20 ± 8 g m-2) (Fig 1).

thumbnail
Download:
Fig 2. Graminoid and forb biomass production across nutrient enrichment plots.

A, B Effect of nitrogen and phosphorus on productivity of graminoid and forb lifeforms in coastal mesic grassland (mean ± SE). Brackets and asterisks indicate significant main effect of N treatments on productivity (‘*’ = p-value < 0.05, ‘***’ = p-value < 0.0001). C, D Effect each nutrient treatment had on productivity of graminoids and forbs (mean ± SE). Hollow points represent productivity values of individual replicates in each treatment. Treatments include control (N0 P0), phosphorus (N0 P10), nitrogen (N10 P0), and nitrogen + phosphorus (N10 P10). Nitrogen treatments are represented on the x-axis while P treatments are represented by color. Scale of the productivity axes are adjusted to improve visualization of variation in the data and are not directly comparable across lifeform type.

https://doi.org/10.1371/journal.pone.0270798.g002

thumbnail
Download:
Table 1. Effects of nitrogen, phosphorus, and nitrogen + phosphorus additions on graminoid and forb productivity compared to control based on two-way ANOVA.

https://doi.org/10.1371/journal.pone.0270798.t001

Despite differences in graminoid and forb aboveground productivity across N enriched treatments, there was no significant interaction or main effects found in belowground productivity among nutrient enrichment treatments (S1 Fig).

Community structure, composition, and organization

There were 22 different plant species identified across all plots consisting of 10 graminoid species and 12 forb species (Table 2). No significant N x P interaction, nor N or P main effects were identified for species diversity or species richness (Table 3). Nitrogen had a significant effect on species evenness with no evidence of an N x P interactions or main effect of P (Table 3). Mean species evenness was significantly higher in N treated communities compared to communities without N treatments (ANOVA: F1, 16 = 9.98, p-value < 0.05).

thumbnail
Download:
Table 2. Species information with NMDS correlations along axis 1 and 2 and results from envfit randomization test.

https://doi.org/10.1371/journal.pone.0270798.t002

thumbnail
Download:
Table 3. Effects of nutrient additions on plant community metrics (richness, evenness, and Shannon’s diversity index [H′]).

https://doi.org/10.1371/journal.pone.0270798.t003

The NMDS analysis on community composition of treatment groups reached a final 3-dimensional solution with a minimized stress = 0.08 (Fig 3). Nutrient enrichment caused all treatments to differ in mean community composition (PERMANOVA: F3, 16 = 4.52, p-value = 0.001) (Table 4). Community composition of all treatment groups clearly varied along NMDS1 (Fig 3). Within-group dispersion did not vary, suggesting similar beta-diversity among nutrient treated communities (ANOVA: F3, 16 = 2.05, p-value > 0.05).

thumbnail
Download:
Fig 3. Non-metric multidimensional scaling ordination of species composition across nutrient enrichment treatments.

Non-metric multidimensional scaling (NMDS) ordination of species composition grouped into convex hulls by nutrient treatment. Points represent within treatment replicates with stars indicating centroids of each treatment group. Centroids can be viewed as a representation of mean community composition. Colors and symbols are matched to nutrient treatment (C = control, P = phosphorus, N = nitrogen, NP = nitrogen + phosphorus).

https://doi.org/10.1371/journal.pone.0270798.g003

thumbnail
Download:
Table 4. Pairwise comparisons of plant community composition between nutrient treatments based on PERMANOVA post-hoc test.

https://doi.org/10.1371/journal.pone.0270798.t004

Species correlated with NMDS1 and/or NMDS2 were used to disentangle species specific contributions to the dissimilarity of communities (Table 1). Nutrient treated communities (N, P, and NP) correlated with perennial graminoid species (Ammophila breviligulata, Spartina patens, Cyperus esculentus, Panicum amarum, and Setaria parviflora), suggesting higher abundance of these species in treated communities compared to control communities (Table 1). Alternatively, control communities were correlated with forb species, including Conyza canadensis, Gnaphalium purpureum, and Linum virginianum as well as a perennial graminoid, Andropogon virginicus (Table 1).

RACs and abundance differences

Differences in species rank order, as well as species absences, existed between treatment and control communities (Fig 4). This is especially true for species that non-randomly correlated with NMDS axes. Differences in species rank order were directly informed by abundance differences between treatment and control communities. Species similar in growth form and life history (e.g., S. patens, A. breviligulata, S. parviflora, and C. esculentus) had higher abundance across treatment communities compared to control communities (Fig 5). For some species (i.e., S. patens and A. breviligulata) higher abundance in treatment groups resulted in increased dominance, as these species did not differ in rank position across treatments (Fig 4). For example, in P treated communities S. patens abundance was on average ~6% higher and A. breviligulata was 2% higher than in control communities (Fig 5A). In N treated communities we found that A. breviligulata had abundance >10% higher compared to control communities, while S. patens was still ~6% higher (Fig 5B). In NP treated communities, we again found both A. breviligulata and S. patens had a higher abundance compared to control communities, but by larger percentages (11% and 8%, respectively) (Fig 5C).

thumbnail
Download:
Fig 4. Rank abundance curves of species across nutrient enrichment treatments.

Rank abundance curves (RACs) for all species in each treatment group. A, B) RACs for species in control and phosphorus plots, respectively. C, D) RACs for species in nitrogen and nitrogen + phosphorus plots, respectively. Species that were found to significantly correlate with the spread on plots in the NMDS analysis (Fig 2) are highlighted with a colored halo.

https://doi.org/10.1371/journal.pone.0270798.g004

thumbnail
Download:
Fig 5. Abundance differences of species between control and each nutrient enrichment treatment.

Abundance difference of each species present in A) control and phosphorus (P) plots; B) control and nitrogen (N) plots; and C) control and nitrogen + phosphorus (NP) plots. Control plots are used as a reference to the natural coastal grassland community. Negative values (yellow bars) indicate species that are less abundant in nutrient treated plots compared to control plots. Positive values (colors corresponding to nutrient treatment) indicate species that are more abundant in nutrient treated plots compared to control plots.

https://doi.org/10.1371/journal.pone.0270798.g005

In other cases, abundance differences were coupled with re-ordering of species ranks. For example, S. parviflora abundance was between 7–11% higher and shifted to a higher rank in all treatment communities compared to controls (Figs 4 and 5). In NP treated communities, a similar pattern was exhibited for C. esculentus, with abundance 11% higher in NP communities compared to controls, which coupled with a shift to higher species rank order (Figs 4A, 4D and 5A, 5C).

Conversely, similar species existed at lower abundance across all treatment communities compared to control communities. For example, C. canadensis and G. purpureum (annual forbs) had lower abundance in treatment communities compared to control communities (Fig 5). Gnaphalium purpureum abundance was consistently ~1% lower in nutrient treated communities compared to control communities. For G. purpureum this small negative difference in abundance was substantial because it caused G. purpureum to decrease between 3–5 rank positions across treatment communities compared to controls (Fig 4). Conyza canadensis abundance was only 1% lower in P treated communities compared to controls and resulted in a similar rank in both communities (Figs 4A, 4B and 5A). However, in N and NP treated communities, C. canadensis abundance was between 2–5% lower compared to controls which decreased the species by 4–5 rank positions (Figs 4 and 5).

Certain abundance differences, and associated species rank order, were treatment dependent. Phosphorus and NP treated communities had depressed abundance of A. virginicus compared to control communities (-4% and -7%, respectively), causing this species to decrease in rank (Figs 4 and 5). However, A. virginicus existed at the same rank in control and N treated communities and increased in abundance by ~3% in N plots (Figs 4 and 5).

Many species fluctuated in presence/absence among communities, most of which could be caused by plot spatial differences (i.e., pre-treatment distribution differences) and thus were not significantly correlated with community composition dissimilarity (Fig 4, Table 1). However, some notable species were systematically absent in N and NP treated communities. We found L. virginianum (a low-ranking species that contributes to community dissimilarity) was absent in both N and NP treated communities (Fig 4C, 4D, Table 1). In NP communities, F. castanea was absent (Fig 4D). Fimbristylis castanea also significantly contributed to community divergence in ordination space (Table 1).

Discussion

Aboveground and root productivity

Our study shows N enrichment, regardless of the P fertilization level, increases graminoid and forb productivity in our coastal grassland system. This finding was expected given the level of N limitation in coastal grassland soils [30, 33, 36]. Our results complement studies in coastal dune systems that found increased aboveground productivity following N fertilization [37, 43]. Phosphorus enrichment did not have a significant impact on graminoid productivity. These findings support those identifying N as a more limiting nutrient than P in coastal systems [45, 48–50]. We show that mid-Atlantic coastal grassland productivity is more impacted by N enrichment than P enrichment. These findings suggest that within a 3-year enrichment window, coastal grassland systems do not experience the same synergistic effects that are seen in at the global scale [1]. As a result, local scale response to nutrients should be a focal research initiative to determine how N rather than P interact with plant communities of specific habitats to influence ecosystem functioning. While our results do not show P as having significant impacts increasing productivity across vegetative lifeforms, there is a non-significant trend showing NP enrichment causing increased productivity compared to N enrichment alone. This response may reflect similarities with European coastal system response to nutrient enrichment [45], but more data is required to determine a definitive response.

Unexpectedly, we did not find difference among treatments in root productivity. We expected root productivity to increase in response to nutrient enrichment based on previous research indicating that N can have long-lasting effects on belowground biomass in coastal dune systems [43]. However, given that our study was conducted in a coastal grassland, interactions that influence plant success differ from those dominating in dune plant communities, such as increased competitive interactions and more frequent access to freshwater [30, 35]. Similar to our findings, global scale research shows that aboveground and belowground productivity patterns do not always correlate in the presence of nutrient enrichment [60]. Cleland et al. [60] suggest belowground biomass increases with nutrient enrichment (N addition), but only when coupled with high light. However, as light competition increases at the soil surface, N additions decreased belowground biomass [60]. We suggest a similar response in our grassland system. Enrichment of nutrient poor, sandy soils likely increase aboveground productivity enough to effectively decrease light at the soil surface, causing decreased belowground production. Increased aboveground biomass, resulting in competition for light and space, has also been shown to impact multiple aspects of community structure, composition, and organization which was found in our study as well.

Community structure

Multiple aspects of plant community structure can be altered as competition for light and space increases [2, 7, 10, 11, 19]. Our results show that in a coastal grassland system, nutrient enrichment did not have impacts on species diversity or species richness. However, we did find that species evenness differed between specific nutrient enrichment treatments. Plots fertilized with N had significantly lower species evenness compared to plots that received no N treatment. This pattern is likely driven by increased dominance of the top ranked species in the community (A. breviligulata and S. patens), a well-documented response to N additions in coastal dunes [37, 61].

Nutrient enrichments can still impact community composition and organization without evident changes in common community structure metrics [4]. Here, we found nutrient enrichment treatments drive significant differences in community composition, causing distinct communities associated with each treatment. We found that when N enrichment is present productivity increases and community composition differs compared to controls, while P enrichment only influences community composition. Nutrient impacts on community composition but not productivity highlights benefits of understanding both species and functional trait response to nutrient enrichment, as both can inform different community-ecosystem relationships [8]. For studies focusing on species response, investigating abundance differences and rank order reshuffling in each community provides an outlet for detailing how increased dominance, species re-ordering, and absences of specific species ignite community differences among treatment groups.

Community re-organization

Our data shows that N, P, and NP enrichment all result in species re-ordering of a few key species and increased dominance of top-ranking species. These findings expand on previous dune research showing N fertilization increases the dominance of certain species over time [37, 61]. Ammophila breviligulata and S. patens increased abundance across treatment groups but did not contribute to species re-organization, as they remained the top ranked species across treatment groups. These two graminoid species have naturally high abundance in mid-Atlantic coastal grasslands. The increase in abundance and rank order of graminoid species compared to other lifeforms has been found in many systems, as graminoids (especially those that are nitrophilous) tend to infiltrate N fertilized plots rapidly [13]. As a result, literature describes nitrophilous graminoids as “winners” in nutrient driven competitive scenarios [13]. Increased dominance is an important consideration because these species can impact ecological interactions that contribute to eliminating species that are normally able to colonize in coastal grasslands. These findings are similar to those found in dune systems [37]. Additionally, other graminoid species displayed systematic increase in our grassland site across nutrient enrichment plots. For example, S. parviflora increased abundance across all treatment groups, which caused significant re-organization of species ranks within nutrient treated communities. Setaria parviflora became one of the top ranked species in each nutrient enrichment treatment, shifting from 9th ranked species in C plots to 3rd-4th ranked species in nutrient enrichment plots. Cyperus esculentus also increased abundance across all nutrient enrichment treatments, with a major increase in NP plots, where it became the 3rd most dominant species.

We also found some annual forbs and perennial graminoids systematically decrease or become entirely absent in response to nutrient enrichment, contributing to community re-organization and community composition differences. Andropogon virginicus decreased abundance in P and NP plots, impacting its within-community rank position which contributed to differences in community composition. Many presence/absence differences exist among low-ranking species in each treatment, however only F. castanea and L. virginianum were both significantly correlated with community divergence and also absent in N and NP plots. These results show nutrient enrichment (N, P, or NP) can cause differences among communities driven by dominant species re-ordering, while N and NP also can cause absence of key species in coastal grasslands. We posit mechanisms leading to responses found here are caused by interactions between increased productivity and characteristics of successful vs. unsuccessful species, which we highlight below.

Previous research suggests long-standing community change on barrier island dunes by altered N flux is primarily driven by competitive exclusion [37, 61]. We expand upon these findings by showing similar patterns in coastal grasslands and propose that certain characteristics of successful species promote increased abundance in nutrient enrichment plots thus altering community organization and composition. Gross and Mittelbach [62] show tall clonal species have substantial impacts on community structure in fertilized grasslands. Clonal species were found to increase in abundance after fertilization and correlated with decreased species richness driven by the ability to spread vegetatively as decreased light limited seedling recruitment [62]. In our study, species that increased dominance and rank order in fertilized plots (i.e., S. patens, A. breviligulata, S. parviflora, and C. esculentus) are clonal and do not rely on seedling recruitment in competitive environments. Furthermore, we also show lower abundance and/or rank order of G. purpureum (annual forb) and C. canadensis (annual forb) across all treatments compared to controls. As annual forbs, abundance of these species can be limited by competitive exclusion, primarily through decreased dispersal ability and seedling recruitment in densely vegetated areas.

Competitive exclusion mechanisms can also explain lower abundance and rank order of A. virginicus (short-lived, ruderal, non-rhizomatous graminoid) in NP treated communities. We suggest increased productivity in NP treated plots influences A. virginicus recruitment. Previous studies have identified decreased A. virginicus abundance as competition increases [63, 64]. Peters and Lowance [65] found A. virginicus abundance decreased in N fertilized plots, eventually leading to species replacement by other dominant graminoids. Interestingly, we found A. virginicus conserved its rank order in N treated communities, compared to controls, suggesting that abundance may not be consistently impacted across treatments. Alternatively, A. virginicus abundance may be impacted by other biotic/abiotic interactions in our system that were not measured in the current study.

Species absence was a significant factor influencing community composition differences and was most common in N and NP treated communities. While previously mentioned sub-optimal recruitment conditions can explain the absence of L. virginianum in N and NP plots, it does not explain the absence of F. castanea in NP plots. The Fimbristylis genera are often clonal and grow in dense caespitose tufts, which are traits that would indicate success in highly competitive environments. However, F. castanea is also an early successional species that pioneers newly developed swales and is eventually replaced by other mid- to late-successional species after 4–7 years [66]. Nutrient enrichment has been shown to increase successional rates in a number of terrestrial systems around the world including coastal dune and grassland systems [13]. Here, NP enrichment may alter rates of succession by the third year of enrichment.

Conclusions and implications for barrier island systems

Impacts of nutrient enrichment on barrier island grassland systems has not accrued much focus despite previous work indicating community composition and productivity changes in coastal dune systems [37, 43, 61]. We show that some patterns are conserved across dune and grassland habitats such as nutrient enrichment increases aboveground graminoid biomass. However, belowground productivity has also been shown to increase in response to nutrient enrichment on barrier island dune systems [43], which we did not find in our coastal grassland site. This suggests that adjacent habitats may be influenced by nutrients in different ways. Increased aboveground growth response could decrease light levels at the soil surface in grasslands, which has been shown to stagnate N driven belowground productivity at a global scale [60].

Similar to previous dune studies, we found nutrients impact community composition and organization by increasing dominance of S. patens and A. breviligulata, enhancing our understanding of their role in grassland competitive interactions and influence over plant community organization and future trajectory. We show that previously dominant species can become more dominant after fertilization, resulting in an exclusionary effect whereby they prevent colonization and/or success of neighboring species. These types of mechanistic responses to nutrient addition disrupt coexistence within native plant communities, which elicits potential for community instability. Such effects are not normally seen in coastal grasslands because nutrient limitation serves as a stressor maintaining levels of coexistence [67]. Community composition differences without major changes in commonly measured community structure metrics (i.e., species diversity and richness) are evidence that large impacts on community composition can occur from community re-organization of species ranks without clear changes in basic community structure metrics [4, 52]. Community responses like these highlight a critical avenue for future research to use community re-organization analyses when obvious changes in species diversity, richness, or evenness are not apparent.

By investigating abundance and RAC differences in nutrient enrichment plots compared to control plots, we found increased abundance and rank order of specific species inform how nutrient enrichment alters coastal grassland community organization. Species exhibiting clonal growth strategy may be better competitors as nutrient additions increase productivity and decrease light availability, a mechanism found to decrease success of dispersal-driven annual species in other grassland systems [62]. By utilizing RAC analyses more frequently, future research can understand more nuanced mechanisms that may be impacting plant community response to nutrient enrichment.

Supporting information

S1 Fig. Rooting mass across nutrient enrichment treatments.

Rooting biomass among nutrient treatments in coastal mesic grassland. Nitrogen treatment levels are labeled on the x-axis while phosphorus treatment levels are indicated by color. Treatments include control (N0P0), phosphorus (N0P10), nitrogen (N10P0), and nitrogen + phosphorus (N10P10). Dots represent individual replicates of each treatment to visualize the spread of the data.

https://doi.org/10.1371/journal.pone.0270798.s001

(TIF)

Acknowledgments

We extend thanks to The Nature Conservancy, Virginia Coast Reserve, the University of Virginia Coastal Research Center, and the Nutrient Network for permitting, logistical, and research support. Valuable field assistance and editorial feedback was provided by Rima Franklin, J. Anthony Stallins, Donald R. Young, and the students and post-docs of the Coastal Plant Ecology Lab at Virginia Commonwealth University.

References

  1. 1. Fay PA, Prober SM, Harpole WS, Knops JMH, Bakker JD, Borer ET, et al. Grassland productivity limited by multiple nutrients. Nat Plants. 2015;1. pmid:27250253
  2. 2. Harpole WS, Sullivan LL, Lind EM, Firn J, Adler PB, Borer ET, et al. Addition of multiple limiting resources reduces grassland diversity. Nature. 2016;537: 93–96. pmid:27556951
  3. 3. Firn J, McGree JM, Harvey E, Flores-Moreno H, Schütz M, Buckley YM, et al. Leaf nutrients, not specific leaf area, are consistent indicators of elevated nutrient inputs. Nat Ecol Evol. 2019;3: 400–406. pmid:30718853
  4. 4. Avolio ML, Koerner SE, La Pierre KJ, Wilcox KR, Wilson GWT, Smith MD, et al. Changes in plant community composition, not diversity, during a decade of nitrogen and phosphorus additions drive above-ground productivity in a tallgrass prairie. J Ecol. 2014;102: 1649–1660.
  5. 5. Borer ET, Seabloom EW, Gruner DS, Harpole WS, Hillebrand H, Lind EM, et al. Herbivores and nutrients control grassland plant diversity via light limitation. Nature. 2014;508: 517–520. pmid:24670649
  6. 6. La Pierre KJ, Smith MD. Functional trait expression of grassland species shift with short- and long-term nutrient additions. Plant Ecol. 2015;216: 307–318.
  7. 7. Koerner SE, Avolio ML, La Pierre KJ, Wilcox KR, Smith MD, Collins SL. Nutrient additions cause divergence of tallgrass prairie plant communities resulting in loss of ecosystem stability. J Ecol. 2016;104: 1478–1487.
  8. 8. Brown JK, Zinnert JC. Trait-based investigation reveals patterns of community response to nutrient enrichment in coastal mesic grassland. Diversity. 2021;13: 1–15.
  9. 9. Carson WP, Barrett GW. Succession in Old-Field Plant Communities: Effects of Contrasting Types of Nutrient Enrichment. Ecology. 1988;69: 984–994.
  10. 10. Clark CM, Tilman D. Loss of plant species after chronic low-level nitrogen deposition to prairie grasslands. Nature. 2008;451: 712–715. pmid:18256670
  11. 11. Harpole WS, Sullivan LL, Lind EM, Firn J, Adler PB, Borer ET, et al. Out of the shadows: multiple nutrient limitations drive relationships among biomass, light and plant diversity. Funct Ecol. 2017;31: 1839–1846.
  12. 12. Hautier Y, Niklaus PA, Hector A. Competition for light causes plant biodiversity loss after eutrophication. Science. 2009;324: 636–638. pmid:19407202
  13. 13. Bobbink R, Hicks K, Galloway J, Spranger T, Alkemade R, Ashmore M, et al. Global assessment of nitrogen deposition effects on terrestrial plant diversity: A synthesis. Ecol Appl. 2010;20: 30–59. pmid:20349829
  14. 14. Soons MB, Hefting MM, Dorland E, Lamers LPM, Versteeg C, Bobbink R. Nitrogen effects on plant species richness in herbaceous communities are more widespread and stronger than those of phosphorus. Biol Conserv. 2017;212: 390–397.
  15. 15. Zhang Y, Cheng L, Li K, Zhang L, Cai Y, Wang X, et al. Nutrient enrichment homogenizes taxonomic and functional diversity of benthic macroinvertebrate assemblages in shallow lakes. Limnol Oceanogr. 2019;64: 1047–1058.
  16. 16. Wilson SD, Tilman D. Interactive effects of fertilization and disturbance on community structure and resource availability in an old-field plant community. Oecologia. 1991;88: 61–71. pmid:28312732
  17. 17. Clark CM, Cleland EE, Collins SL, Fargione JE, Gough L, Gross KL, et al. Environmental and plant community determinants of species loss following nitrogen enrichment. Ecol Lett. 2007;10: 596–607. pmid:17542938
  18. 18. Ceulemans T, Merckx R, Hens M, Honnay O. Plant species loss from European semi-natural grasslands following nutrient enrichment—is it nitrogen or is it phosphorus? Glob Ecol Biogeogr. 2013;22: 73–82.
  19. 19. Demalach N, Kadmon R. Light competition explains diversity decline better than niche dimensionality. Funct Ecol. 2017;31: 1834–1838.
  20. 20. De Schrijver A, de Frenne P, Ampoorter E, van Nevel L, Demey A, Wuyts K, et al. Cumulative nitrogen input drives species loss in terrestrial ecosystems. Glob Ecol Biogeogr. 2011;20: 803–816.
  21. 21. Flores-Moreno H, Reich PB, Lind EM, Sullivan LL, Seabloom EW, Yahdjian L, et al. Climate modifies response of non-native and native species richness to nutrient enrichment. Philos Trans R Soc B Biol Sci. 2016;371: 20150273. pmid:27114575
  22. 22. Sundqvist MK, Liu Z, Giesler R, Wardle DA. Plant and microbial responses to nitrogen and phosphorus addition across an elevational gradient in subarctic tundra. Ecology. 2014;95: 1819–1835. pmid:25163116
  23. 23. Yang X, Yang Z, Tan J, Li G, Wan S, Jiang L. Nitrogen fertilization, not water addition, alters plant phylogenetic community structure in a semi-arid steppe. J Ecol. 2018;106: 991–1000.
  24. 24. Read QD, Henning JA, Classen AT, Sanders NJ. Aboveground resilience to species loss but belowground resistance to nitrogen addition in a montane plant community. J Plant Ecol. 2018;11: 351–363.
  25. 25. Zhao Y, Yang B, Li M, Xiao R, Rao K, Wang J, et al. Community composition, structure and productivity in response to nitrogen and phosphorus additions in a temperate meadow. Sci Total Environ. 2019;654: 863–871. pmid:30448675
  26. 26. Li W, Zhang R, Liu S, Li W, Li J, Zhou H, et al. Effect of loss of plant functional group and simulated nitrogen deposition on subalpine ecosystem properties on the Tibetan Plateau. Sci Total Environ. 2018;631–632: 289–297. pmid:29525708
  27. 27. Hautier Y, Seabloom EW, Borer ET, Adler PB, Harpole WS, Hillebrand H, et al. Eutrophication weakens stabilizing effects of diversity in natural grasslands. Nature. 2014;508: 521–525. pmid:24531763
  28. 28. DeMalach N, Zaady E, Kadmon R. Light asymmetry explains the effect of nutrient enrichment on grassland diversity. Ecol Lett. 2017;20: 60–69. pmid:27933739
  29. 29. Tang Z, Deng L, An H, Yan W, Shangguan Z. The effect of nitrogen addition on community structure and productivity in grasslands: A meta-analysis. Ecol Eng. 2017;99: 31–38.
  30. 30. Young DR, Brantley ST, Zinnert JC, Vick JK. Landscape position and habitat polygons in a dynamic coastal environment. Ecosphere. 2011;2.
  31. 31. Sinha E, Michalak AM, Balaji V. Eutrophication will increase during the 21st century as a result of precipitation changes. Science. 2017;357: 1–5. pmid:28751610
  32. 32. Stutz ML, Pilkey OH. A Review of Global Barrier Island Distribution. J Coast Res. 2001; 15–22.
  33. 33. Ehrenfeld JG. Dynamics and processes of barrier island vegetation. Rev Aquat Sci. 1990;2: 437–480.
  34. 34. Zinnert JC, Via SM, Nettleton BP, Tuley PA, Moore LJ, Stallins JA. Connectivity in coastal systems: Barrier island vegetation influences upland migration in a changing climate. Glob Chang Biol. 2019;25: 2419–2430. pmid:30932269
  35. 35. Brown JK, Zinnert JC. Topography and disturbance influence trait-based composition and productivity of adjacent habitats in a coastal system. Ecosphere. 2020;11.
  36. 36. Art HW, Bormann FH, Voigt GK, Woodwell GM. Barrier island forest ecosystem: role of meteorlogic nutrient inputs. Science (80-). 1974;184: 60–62. pmid:17734631
  37. 37. Day FP, Conn C, Crawford E, Stevenson M. Long-term effects of nitrogen fertilization on plant community structure on a coastal barrier island dune chronosequence. J Coast Res. 2004;20: 722–730.
  38. 38. Brantley ST, Bissett SN, Young DR, Wolner CWV, Moore LJ. Barrier island morphology and sediment characteristics affect the recovery of dune building grasses following storm-induced overwash. PLoS One. 2014;9. pmid:25148028
  39. 39. Zinnert JC, Stallins JA, Brantley ST, Young DR. Crossing scales: The complexity of barrier-island processes for predicting future change. Bioscience. 2017;67: 39–52.
  40. 40. Seitzinger SP, Kroeze C, Bouwman AF, Caraco N, Dentener F, Styles R V. Global patterns of dissolved inorganic and particulate nitrogen inputs to coastal systems: Recent conditions and future projections. Estuaries. 2002;25: 640–655.
  41. 41. Ackerman D, Millet DB, Chen X. Global Estimates of Inorganic Nitrogen Deposition Across Four Decades. Global Biogeochem Cycles. 2019;33: 100–107.
  42. 42. Li Y, Schichtel BA, Walker JT, Schwede DB, Chen X, Lehmann CMB, et al. Increasing importance of deposition of reduced nitrogen in the United States. Proc Natl Acad Sci U S A. 2016;113: 5874–5879. pmid:27162336
  43. 43. Heyel SM, Day FP. Long-term residual effects of nitrogen addition on a barrier island dune ecosystem. J Torrey Bot Soc. 2006;133: 297–303.
  44. 44. Aggenbach CJS, Kooijman AM, Fujita Y, van der Hagen H, van Til M, Cooper D, et al. Does atmospheric nitrogen deposition lead to greater nitrogen and. Biol Conserv. 2017;212: 416–422.
  45. 45. Kooijman AM, van Til M, Noordijk E, Remke E, Kalbitz K. Nitrogen deposition and grass encroachment in calcareous and acidic Grey dunes (H2130) in NW-Europe. Biol Conserv. 2017;212: 406–415.
  46. 46. Peñuelas J, Poulter B, Sardans J, Ciais P, Van Der Velde M, Bopp L, et al. Human-induced nitrogen-phosphorus imbalances alter natural and managed ecosystems across the globe. Nat Commun. 2013;4. pmid:24343268
  47. 47. Mahowald N, Jickells TD, Baker AR, Artaxo P, Benitez-Nelson CR, Bergametti G, et al. Global distribution of atmospheric phosphorus sources, concentrations and deposition rates, and anthropogenic impacts. Global Biogeochem Cycles. 2008;22: 1–19.
  48. 48. Osgood DT, Zieman JC. Factors controlling aboveground Spartina alterniflora (Smooth cordgrass) tissue element composition and production in different-age barrier island marshes. Estuaries. 1993;16: 815–826.
  49. 49. Kachi N, Hirose T. Limiting Nutrients for Plant Growth in Coastal Sand Dune Soils. J Ecol. 1983;71: 937.
  50. 50. Kooijman AM, Dopheide JCR, Sevink J, Takken I, Verstraten JM. Nutrient limitations and their implications on the effects of atmospheric deposition in coastal dunes; lime-poor and lime-rich sites in the Netherlands. J Ecol. 1998;86: 511–526.
  51. 51. Elser JJ, Bracken MES, Cleland EE, Gruner DS, Harpole WS, Hillebrand H, et al. Global analysis of nitrogen and phosphorus limitation of primary producers in freshwater, marine and terrestrial ecosystems. Ecol Lett. 2007;10: 1135–1142. pmid:17922835
  52. 52. Avolio ML, Carroll IT, Collins SL, Houseman GR, Hallett LM, Isbell F, et al. A comprehensive approach to analyzing community dynamics using rank abundance curves. Ecosphere. 2019;10.
  53. 53. Moulton A. Short-term Effects of Nutrients on a Barrier Island Grassland Community By. Virginia Commonwealth University. 2017. Available: https://scholarscompass.vcu.edu/etd/4859/
  54. 54. Hallett LM, Jones SK, MacDonald AAM, Jones MB, Flynn DFB, Ripplinger J, et al. codyn: An r package of community dynamics metrics. Methods Ecol Evol. 2016.
  55. 55. Smith B, Wilson JB. A Consumer’s Guide to Evenness Indices. Oikos. 1996.
  56. 56. Oksanen J, Blanchet FG, Friendly M, Kindt R, Legendre P, Mcglinn D, et al. vegan: Community ecology package. R Packag version 25–4. 2019. Available: https://cran.r-project.org/web/packages/vegan/vegan.pdf
  57. 57. Anderson MJ. Permutational Multivariate Analysis of Variance (PERMANOVA). Wiley StatsRef: Statistics Reference Online. 2017. pp. 1–15.
  58. 58. Anderson MJ. Distance-based tests for homogeneity of multivariate dispersions. Biometrics. 2006;62: 245–253. pmid:16542252
  59. 59. Anderson MJ, Ellingsen KE, McArdle BH. Multivariate dispersion as a measure of beta diversity. Ecol Lett. 2006;9: 683–693. pmid:16706913
  60. 60. Cleland EE, Lind EM, DeCrappeo NM, DeLorenze E, Wilkins RA, Adler PB, et al. Belowground Biomass Response to Nutrient Enrichment Depends on Light Limitation Across Globally Distributed Grasslands. Ecosystems. 2019;22: 1466–1477.
  61. 61. Day FP, Adams EC, Gibala-Smith LA, Graziani DJ, McMillan B, Sedghi N, et al. Determining change in coastal barrier island dune vegetation following a decade of nitrogen fertilization. J Coast Res. 2018;34: 1100–1104.
  62. 62. Gross KL, Mittelbach GG. Negative effects of fertilization on grassland species richness are stronger when tall clonal species are present. Folia Geobot. 2017;52: 401–409.
  63. 63. Bazzaz FA. Succession on Abandoned Fields in the Shawnee Hills, Southern Illinois. Ecology. 1968;49: 924–936.
  64. 64. Bazzaz FA. The response of natural ecosystems to the rising global CO2 levels. Annu Rev Ecol Syst. 1990;21: 167–196.
  65. 65. Peters EJ, Lowance SA. Fertility and Management Treatments to Control Broomsedge in Pastures. Weed Sci. 1974;22: 201–205.
  66. 66. Johnson AF. Rates of vegetation succession on a Coastal dune system in Northwest Florida. J Coast Res. 1997;13: 373–384.
  67. 67. Shmida A, Ellner S. Coexistence of Plant Species with Similar Niches. Vegetatio. 1984;58: 29–55.