Palacio-Lopez K, Hovick SM, Mattingly KZ, Weston LM, Hofford NP, Finley L, Tayal A, Reinartz JA. Suppression of reed canarygrass by assisted succession: A sixteen-year restoration experiment. Journal of Applied Ecology.

https://doi.org/10.5061/dryad.r7sqv9sks

Summary

1. Assisted succession could enable long-term restoration where successional trajectories stall due to competition from invasive plants. Many invasives are shade-intolerant, therefore interventions reducing light availability should suppress invasion and re-establish successional processes. However, given how ubiquitous nonlinearities are in ecology, restoration success also depends on identifying critical system thresholds, e.g., invader abundances below which regeneration of desired species is possible. We report the successful use of assisted succession to restore a swamp forest invaded by Phalaris arundinacea (reed canarygrass; hereafter Phalaris), initiated by a high-density planting of woody species to outcompete Phalaris by reducing light availability.

2. We established five pre-planting treatments in a Phalaris near-monoculture in Wisconsin, USA: herbicide-only, herbicide+plow, herbicide+burn, mow+herbicide, and control. In 2003 we planted 22 tree and shrub species at high densities, then in 2019 we censused the site to: (1) screen for long-term differences among treatments, (2) evaluate long-term effects of our interventions on community composition, and (3) characterize the critical thresholds that enable invader suppression and restoration success.

3. Vegetation responses and light availability across our four pre-planting invader removal treatments did not differ. Late fall glyphosate application suppressed Phalaris long enough that a dense canopy of native woody species could establish and eventually out-shade it. Overstory tree and shrub densities of 0.071/m2 suppressed Phalaris to 50% cover, but, due to nonlinearities, much higher densities were needed to reduce light availability and thus Phalaris cover enough to shift the system from being invader-dominated. Compositional similarities between juvenile woody species and the overstory suggest a long-term restoration success.

4. Synthesis and applications. Invasive species management and the restoration of target plant communities can be aided by assisting successional trajectories that have stalled. We document a restoration strategy for forests invaded by shade-intolerant invaders that is both effective and economical, as only a simple site preparation and single planting effort is required. Establishing a dense canopy of woody species in this way can break the feedbacks maintaining invader dominance and re-introduce feedbacks enabling long-term ecosystem recovery. We also illustrate the value of identifying critical thresholds influencing the abundance and impact of key invasive species.

Description of the data and file structure

“_df_diversity_richness_051320.csv” - 2019 richness and diversity of woody species calculated on a per-plot basis.

• plot - Experimental plot. See Figure S1 for experimental design layout.

• block - Experimental block. See Figure S1 for experimental design layout.

• area - Plot area in sq m.

• treat - Phalaris removal treatment. (P)–herbicide+plow; (B)–herbicide+burn; (M)–mow+herbicide; (H)–herbicide-only; and (C)–control.

• diversity.all - Inverse Simpson’s index for all woody species (trees and shrubs).

• richness.all - Number of woody species (trees and shrubs).

• diversity.trees - Inverse Simpson’s index for trees.

• richness.trees - Number of tree species.

• diversity.shrubs - Inverse Simpson’s index for shrubs.

• richness.shrubs - Number of shrub species.

“_df_sapling_quad_all_sites_050820.csv” - 2019 quadrat-based sampling of woody species in the sapling layer (woody species >1 m tall and <2.5 cm DBH).

• site - Fellenz Woods Park (data not reported in this manuscript, see Hovick and Reinartz 2007), Winker Property (data not reported in this manuscript, see Hovick and Reinartz 2007), or Huiras Lake State Natural Area.

• block - Experimental block. See Figure S1 for experimental design layout.

• plot - Experimental plot. See Figure S1 for experimental design layout.

• area - Plot area in sq m.

• elevation_f - Plot elevation in ft.

• elevation_m - Plot elevation in m.

• treat - Phalaris removal treatment. (P)–herbicide+plow; (B)–herbicide+burn; (M)–mow+herbicide; (H)–herbicide-only; and (C)–control.

• quad - Sampling quadrat.

• to_include - Identifier denoting quadrat samples to omit due to being counted twice.

• sp_name - Sapling species common name abbreviation. Blank if the quadrat had no saplings. See Table 1 for additional species details.

• cat - Identifier to denote tree vs. shrub (See Table 1 for additional species details). NA if the quadrat had no saplings.

• sapling - Identifier to denote saplings.

• sapling_count - Count of saplings of that species within the quadrat.

• notes - additional details such as species ID or information about data transcription from physical data sheets.

“_df_small_trees_050520.csv” - 2019 quadrat-based sampling of woody species in the seedling layer (<1 m but >30 cm in height).

• site - Fellenz Woods Park (data not reported in this manuscript, see Hovick and Reinartz 2007), Winker Property (data not reported in this manuscript, see Hovick and Reinartz 2007), or Huiras Lake State Natural Area.

• block - Experimental block. See Figure S1 for experimental design layout.

• plot - Experimental plot. See Figure S1 for experimental design layout.

• area - Plot area in sq m.

• elevation_f - Plot elevation in ft.

• elevation_m - Plot elevation in m.

• treat - Phalaris removal treatment. (P)–herbicide+plow; (B)–herbicide+burn; (M)–mow+herbicide; (H)–herbicide-only; and (C)–control.

• quad - Sampling quadrat.

• sp_name - Seedling species common name abbreviation. NA if the quadrat had no seedlings. See Table 1 for additional species details.

• status - a=alive, d=dead. NA if the quadrat had no seedlings.

• sll_tree_count - Count of seedlings of that species within the quadrat.

• notes - additional details about data transcription from physical data sheets.

“_df_soilmoist_light_rcg_elevation_treat_quad_050320.csv” - 2019 quadrat-based sampling of vegetation and environmental variables.

• site - Fellenz Woods Park (data not reported in this manuscript, see Hovick and Reinartz 2007), Winker Property (data not reported in this manuscript, see Hovick and Reinartz 2007), or Huiras Lake State Natural Area.

• block - Experimental block. See Figure S1 for experimental design layout.

• plot - Experimental plot. See Figure S1 for experimental design layout.

• area - Plot area in sq m.

• elevation_f - Plot elevation in ft.

• elevation_m - Plot elevation in m.

• treat - Phalaris removal treatment. (P)–herbicide+plow; (B)–herbicide+burn; (M)–mow+herbicide; (H)–herbicide-only; and (C)–control.

• quad - Sampling quadrat.

• mean_soil_moist - Per-plot mean soil moisture.

• light - Total light availability above the herbaceous level estimated as percent canopy openness using hemispherical photos.

• v_cover_rcg - Visually estimated Phalaris cover via modified Daubenmire cover classes: <5%, 5-24%, 25-49%, 50-74%, and 75-100% (Daubenmire, 1959), allowing the sum of Phalaris and non-Phalaris cover to exceed 100%. Cover estimates were transformed to the midpoint of each cover class (restricting midpoints to minimum 2.5%, maximum 87.5%).

• v_cover_no_rcg - Visually estimated cover of species other than Phalaris, via modified Daubenmire cover classes: <5%, 5-24%, 25-49%, 50-74%, and 75-100% (Daubenmire, 1959), allowing the sum of Phalaris and non-Phalaris cover to exceed 100%. Cover estimates were transformed to the midpoint of each cover class (restricting midpoints to minimum 2.5%, maximum 87.5%).

• bare - Visually estimated cover of bare ground, via modified Daubenmire cover classes: <5%, 5-24%, 25-49%, 50-74%, and 75-100% (Daubenmire, 1959), constraining the summed estimates of litter, vegetation, and bare soil to equal 100%.

• litter - Visually estimated cover of litter, via modified Daubenmire cover classes: <5%, 5-24%, 25-49%, 50-74%, and 75-100% (Daubenmire, 1959), constraining the summed estimates of litter, vegetation, and bare soil to equal 100%.

• veg - Visually estimated cover of vegetation, via modified Daubenmire cover classes: <5%, 5-24%, 25-49%, 50-74%, and 75-100% (Daubenmire, 1959), constraining the summed estimates of litter, vegetation, and bare soil to equal 100%.

• note - additional information about data transcription from physical data sheets.

• plot_quad - Identifier combining experimental plot and sampling quadrat.

• sap_all - count of saplings of all woody individuals (trees and shrubs) within the quadrat.

• sap_tree - count of all tree species saplings within the quadrat.

• sap_shrub - count of all shrub species saplings within the quadrat.

• sap_ash - count of all ash tree saplings within the quadrat.

• sap_dogwood - count of dogwood shrub saplings within the quadrat.

• sap_elderberry - count of all elderberry shrub saplings within the quadrat.

• sap_blackraspberry - count of all raspberry shrub saplings within the quadrat.

“_df_trees_shrub_no_sapling_050620.csv” - 2019 per-plot census of all large woody individuals (≥2.5 cm diameter at breast height (DBH)).

• page - Physical data sheet source.

• to_include - Identifier denoting quadrat samples to omit due to being counted twice.

• site - Fellenz Woods Park (data not reported in this manuscript, see Hovick and Reinartz 2007), Winker Property (data not reported in this manuscript, see Hovick and Reinartz 2007), or Huiras Lake State Natural Area.

• block - Experimental block. See Figure S1 for experimental design layout.

• plot - Experimental plot. See Figure S1 for experimental design layout.

• area - Plot area in sq m.

• treat - Phalaris removal treatment. (P)–herbicide+plow; (B)–herbicide+burn; (M)–mow+herbicide; (H)–herbicide-only; and (C)–control.

• sp_name - Woody species common name abbreviation. NA if the plot had no large woody individuals. See Table 1 for additional species details.

• cat - Identifier to denote tree vs. shrub (See Table 1 for additional species details). NA if the plot had no large woody individuals.

• sapling - Identifier to denote sapling or not (nosap).

• dbh - Diameter at breast height, cm.

• dbh_old - See note for these observations, error in collection of some DBHs.

• canopy_area - Dimension of canopy area (m2) for shrubs.

• wrong_canopy_area - See note for these observations, error in collection of some areas.

• basal_area - Basal area (m2) for trees calculated from DBH.

• status - a=alive, d=dead, r=resprout. NA if the plot had no large woody individuals.

• note - additional information about data transcription from physical data sheets.

“data_survival_all_years_edited_111320.csv” - per-plot counts of woody individuals at planting, in 2003 census, 2004 census, and as large woody individuals in 2019.

• plot - Experimental plot. See Figure S1 for experimental design layout.

• treat - Phalaris removal treatment. (P)–herbicide+plow; (B)–herbicide+burn; (M)–mow+herbicide; (H)–herbicide-only; and (C)–control.

• cat - Identifier to denote tree vs. shrub (See Table 1 for additional species details).

• identifier - Identifier combining plot and species name.

• sp_name - Woody species common name abbreviation. See Table 1 for additional species details.

• sum_of_n Rows - Data validation column. Column not used in the analysis. Column shows the total rows in which a species appears in a previous file.

• n_planted - number of individuals planted of a species in that plot.

• sum_n_03_S - sum per species found in spring 2003

• sum_n_03_F_original_noused - data not used in the analysis. Sum per species found in fall 2003

• sum_n_03_F - sum per species found in fall 2003

• sum_n_04 - sum per species found in 2004

• n_19_origin - sum per species found in 2019

• n_19 - sum per species found in 2019 omitting dead and resprout

• status_2019 - a=alive, d=dead, r=resprout. NA if the plot had no large woody individuals.

• subs_03F-04F - Data validation column. Column not used in the analysis. Column shows the difference in the number of species planted in fall 2003 and 2004.

• note - notes about calculations.

“ECxs 5_18_23.csv” - percent target calculations. Output from “ECxs 5_18_23.Rmd”

• Percent_target - target/threshold percent of y variable.

• x - x variable that is being manipulated to achieve the target.

• y - y variable that is being targeted.

• Estimate - curve-estimated value of x required to achieve target y.

• Lower 95 - lower 95% confidence bound around estimate.

• Upper 95 - upper 95% confidence bound around estimate.

• figure - manuscript figure plotting the estimates.

“Hill slopes and EC50s 5_18_23.csv” - curve model parameter estimates. Output of “Hill slopes and EC50s 5_18_23.Rmd”

• parameter - curve model parameter: slope or asymptote (e:1:50).

• x - x variable that is being manipulated to achieve the target.

• y - y variable that is being targeted.

• Estimate - estimated value of parameter.

• Std.Error - standard error around parameter estimate.

• Lower 95 - lower 95% confidence bound around estimate.

• Upper 95 - upper 95% confidence bound around estimate.

• t-value - t statistic of estimate.

• p-value - significance of estimate.

• figure - manuscript figure plotting the curve.

“mean_quad_data.csv” - per-plot means of key environmental and vegetation variables.

• plot - Experimental plot. See Figure S1 for experimental design layout.

• block - Experimental block. See Figure S1 for experimental design layout.

• treat - Phalaris removal treatment. (P)–herbicide+plow; (B)–herbicide+burn; (M)–mow+herbicide; (H)–herbicide-only; and (C)–control.

• area - Plot area in sq m.

• elev_m - Plot elevation in m.

• mean_soil_moist - Per-plot mean soil moisture.

• mean_light - per-plot mean light availability above the herbaceous level estimated as percent canopy openness using hemispherical photos.

• mean_v_cover_no_rcg - per-plot mean of visually estimated cover of species other than Phalaris, via modified Daubenmire cover classes: <5%, 5-24%, 25-49%, 50-74%, and 75-100% (Daubenmire, 1959). Within a sampling quadrat, we allowed the sum of Phalaris and non-Phalaris cover to exceed 100%. Cover estimates were transformed to the midpoint of each cover class (restricting midpoints to minimum 2.5%, maximum 87.5%).

• mean_v_cover_rcg - per-plot mean of visually estimated cover of Phalaris, via modified Daubenmire cover classes: <5%, 5-24%, 25-49%, 50-74%, and 75-100% (Daubenmire, 1959). Within a sampling quadrat, we allowed the sum of Phalaris and non-Phalaris cover to exceed 100%. Cover estimates were transformed to the midpoint of each cover class (restricting midpoints to minimum 2.5%, maximum 87.5%).

• mean_sap_all - per-plot mean count of saplings of all woody individuals (trees and shrubs).

• mean_sap_tree - per-plot mean count of all tree species saplings.

• mean_sap_shrub - per-plot mean count of all shrub species saplings.

• mean_sap_ash - per-plot mean count of all ash saplings.

“treatment_reference.csv” - summary of experimental design variables.

• site_num - number to differentiate the 3 sites.

• site - Fellenz Woods Park (data not reported in this manuscript, see Hovick and Reinartz 2007), Winker Property (data not reported in this manuscript, see Hovick and Reinartz 2007), or Huiras Lake State Natural Area.

• plot - Experimental plot. See Figure S1 for experimental design layout.

• treatment - Long descriptive names for Phalaris removal treatments.

• treat - Abbreviated names for Phalaris removal treatments. (P)–herbicide+plow; (B)–herbicide+burn; (M)–mow+herbicide; (H)–herbicide-only; and (C)–control.

• area_m2 - Plot area in sq m.

• block - Experimental block. See Figure S1 for experimental design layout.

“z_abundance_ash_05_27_21.csv” - change in ash abundance over time from planting to 2019 survey.

• time - 0=at planting, 0.5=survey in 2003, 1.5=survey in 2004, 16=2019 survey after 16 years.

• treat - Phalaris removal treatment. (P)–herbicide+plow; (B)–herbicide+burn; (M)–mow+herbicide; (H)–herbicide-only; and (C)–control.

• freq - abundance of ashes.

Sharing/Access information

Other publicly accessible locations of the data will include being linked by the journal, Journal of Applied Ecology.

Code/Software

“DRM plots KZM 5_18_23.R”

This R script loads data and required packages and creates curve and correlation figures (Figure 3, 4, 5, and Supplemental Figure S4). The script was developed in R v. 4.2.1.

“ECxs 5_18_23.Rmd”

This R script loads data and required packages, fits the curve models, then prints to “ECxs 5_18_23.csv” the estimates at the 10, 25, and 50% threshold levels of restoration targets. The script was developed in R v. 4.2.1.

“Figure 2 overstory sapl seedl density 1_30_23.R”

This R script loads data and required packages and creates figure 2 (comparison of woody densities in each vegetation layer). The script was developed in R v. 4.2.1.

“Figure S2 density diff w ANOVA 11_8_23.R”

This R script loads data and required packages and creates figure S2 plus relevant ANOVA (shows changes over time in woody densities, comparing across treatments). The script was developed in R v. 4.2.1.

“Figure S3 woody rich div w ANOVA 11_8_23.R”

This R script loads data and required packages and creates figure S3 plus relevant ANOVA (shows richness and diversity of different layers of woody vegetation, comparing across treatments). The script was developed in R v. 4.2.1.

“Figure S5 ash over time 1_31_23.R”

This R script loads data and required packages and creates figure S5 (change over time in ash density, separated by treatment). The script was developed in R v. 4.2.1.

“Hill slopes and EC50s 5_18_23.Rmd”

This R script loads data and required packages, fits the curve models, then prints to “Hill slopes and EC50s 5_18_23.csv” the curve model parameters. The script was developed in R v. 4.2.1.

“Tukeys among trts Table S1 11_20_23.R”

This R script loads data and required packages and performs analyses reported in table S1 (comparisons across treatments of main focal variables). The script was developed in R v. 4.2.1.