Floristic Quality Research Articles

Recovery of herb‐layer vegetation and soil properties after pile burning in a Midwestern oak woodland

Thinning and removal of woody vegetation is the first step in restoring ecosystem structure to systems altered by woody encroachment. However, pile burning—a common method of eliminating woody residue at restoration sites—can promote the establishment of exotic species and adversely impact soils and native vegetation via extreme soil heating. Despite its widespread use, pile burning effects remain poorly understood in oak woodlands compared to coniferous forests. We examined how pile burning influenced soil properties and herb‐layer recovery in a Midwestern, U.S. oak woodland undergoing restoration via exotic shrub removal, tree thinning, and prescribed burning. We quantified soil properties and passive vegetation recovery inside and adjacent to pile burn scars after 3 years, and tested whether native seed additions in year 1 increased native cover and reduced exotic cover in year 3. Bare ground cover, soil pH, and concentrations of P and Ca remained elevated 3 years after pile burning. In contrast, native cover, native richness, and floristic quality recovered to levels similar to adjacent control locations. Pile burning did not promote exotic invasions, increase inorganic forms of N, or alter patterns of the native cover of most growth forms. However, Carex species failed to reestablish in burn scars. Compared to passive recovery, seed additions increased native cover in burn scars and reduced exotic cover in adjacent control locations. Our results indicate that burn scars can naturally recover in oak woodland, but native seed additions may accelerate this process and improve restoration outcomes.

Determining vegetation metric robustness to environmental and methodological variables.

Land managers need reliable metrics for assessing the quality of restorations and natural areas and prioritizing management and conservation efforts. However, it can be difficult to select metrics that are robust to sampling methods and natural environmental differences among sites, while still providing relevant information regarding ecosystem changes or stressors. We collected herbaceous-layer vegetation data in wetlands and grasslands in four regions of the USA (the Midwest, subtropical Florida, arid southwest, and coastal New England) to determine if commonly used vegetation metrics (species richness, mean coefficient of conservatism [mean C], Floristic Quality Index [FQI], abundance-weighted mean C, and percent non-native species cover) were robust to environmental and methodological variables (region, site, observer, season, and year), and to determine adequate sample sizes for each metric. We constructed linear mixed effects models to determine the influence of these environmental and methodological variables on vegetation metrics and used metric accumulation curves to determine the effect of sample size on metric values. Species richness and FQI varied among regions, and year and observer effects were also highly supported in our models. Mean C was the metric most robust to sampling variables and stabilized at less sampling effort compared to other metrics. Assessment of mean C requires sampling a small number of quadrats (e.g. 20), but assessment of species richness or FQI requires more intensive sampling, particularly in species-rich sites. Based on our analysis, we recommend caution be used when comparing metric values among sites sampled in different regions, different years, or by different observers.

A Floristic Quality Assessment Methodology for Citizen Science

A Floristic Quality Assessment Methodology for Citizen Science

Declining Native Species Richness in Natural Areas in Eastern North America: An Example from Baker Woodlot in Central Michigan

Natural areas are intended to preserve biodiversity and understanding how well they fulfill this role is crucial as pressure from anthropogenic disturbances like habitat loss increases. We re-inventoried the flora of Baker Woodlot in East Lansing, Michigan, 40 years after an initial study to document changes in community composition. Both inventories were based on vascular plant collections vouchered by specimens and deposited in the Michigan State University Herbarium. We compared results with 10 similar floristic change studies in northeastern North America to understand patterns in species turnover, accounting for gains and losses of natives and nonnatives. In Baker Woodlot and across the comparable studies, total richness stability masked a consistent pattern of native losses and nonnative gains. In Baker Woodlot, total richness increased marginally (3%; 11 species), reflecting a net loss of 19 natives and a gain of 30 nonnatives. Across the comparable studies, richness increased by 3.6%, but seven native species were lost per decade on average. Nonnative turnover was higher compared to native turnover and positively correlated with time between inventories. Despite net loss of natives, Baker Woodlot experienced only a slight decline in floristic quality. Baker Woodlot remains a high-quality beech-maple forest, which is a prominent component of Michigan's biodiversity. Species richness alone does not adequately address biodiversity concerns. In Baker Woodlot and comparable forests, natives are being replaced by nonnatives.

Application of IUCN Red List of Ecosystems to assess the ecological status of marine bar ecosystems of Burullus wetland: A Ramsar site

The IUCN Red List of Ecosystems has been employed to assess the status of ecosystems and evaluate the risk of their collapse. The RLE assessment was conducted on the main ecosystems of the marine bar of Burullus wetland; sand plain and salt marshes ecosystems of the marine bar which separates the lake from the Mediterranean Sea. Four criteria were used for conducting the assessment; (A) reduction in distribution, (B) constricted distribution, (C) abiotic components degradation, and (D) disruption of biotic processes and interactions. The integration of remotely-sensed data allowed the application of RLE to study the changes in ecosystems distribution for Criteria A and B. Criterion A showed the same result for sand plain and salt marshes as Critically Endangered (CR). This indicates the distribution of the marine bar is extremely affected by the urban and human activities in the region. It is also expected to be affected by any slight changes in the sea-level and costal erosions in the future in terms of disruption in biotic processes and interactions. Criteria D was applied using floristic quality assessment for current floristic composition and the available historic data. The results show that sand plain and salt marshes are Near Threatened (NT). However, the application of Criteria C was Data Deficient DD due to lack of long term data about the abiotic variables responsible for environmental degradation. The identical results of RLE assessment of sand plain and salt marshes reflect that both ecosystems are equally affected by the ongoing activities in the region, particularly over the last few years. The results indicate that the marine bar ecosystems were disturbed by anthropogenic events such as urban development, grazing and farming. Also natural events such as climate change and its consequences are expected to have severe impacts on the marine bar in the future. Therefore, further studies should focus on the study of the main interactions and processes in the marine bar ecosystems.

Floristic Quality Assessment Index of the Dagwan Stream in Dachigam National Park of Kashmir Himalaya

Floristic Quality Assessment Index of the Dagwan Stream in Dachigam National Park of Kashmir Himalaya

Old-growth attributes in a maturing secondary Indiana state forest: an opportunity for balanced management1

Primary old-growth forests are rare in the state of Indiana and are found only in small, isolated patches of nature preserves. Recent research underscores the values of large tracts of secondary old-growth (referred to in this paper as “older” forests), including carbon sequestration, forest resiliency, mitigation of climate change impacts, and conservation of biological diversity. Due to historical management, many portions of Indiana state forests are maturing, and yet the “older” successional stage is underrepresented. One of these maturing forests, the site of our study, is in south-central Indiana within the Back Country Area (BCA) of the Morgan-Monroe/Yellowwood State Forest (two state forests, herein called a complex, including parcels in three counties with the largest block of forests spanning Brown and Monroe counties). Our study of the forest structure of the BCA addresses the following questions: (a) Is this mature forest developing ecological attributes of old-growth forests, thus making it a good candidate for an older forest reserve? (b) Does this forest support the plant community characteristic of natural areas of Indiana? We analyzed forest stand metrics, coarse woody debris and snags, age of the largest live trees, and the floristic quality of the herbaceous layer. We then compared these data with published accounts from Indiana nature preserves. Metrics such as tree basal area, number of large-diameter trees, and number of snags are within the range of those measured in existing old-growth forests of Indiana. Analysis of tree cores and log cross-sections shows that 78% of the sample trees were over 100 yr old, with 24% over 130 yr, which is also comparable to old-growth forest nature preserves in the state. The metrics of the Floristic Quality Assessment of the broader plant community of the BCA are comparable to or higher than the metrics from published assessments of Indiana nature preserves. Collectively, the attributes measured indicate that the BCA supports a maturing secondary forest with old-growth characteristics. In the absence of further timber harvest, the BCA as a reserve could reach older forest status within 50 yr. Our study demonstrates that the BCA forest is a prime candidate for an older forest reserve. This study of the BCA could also serve as a model for future assessments of potential older forest reserves in other state forests. Reserves of older forests free from harvest are not incompatible with other management goals and can be a valuable part of a broader-scale management plan. These forest reserves should be promoted across the state to increase the representation within Indiana's state forest system of late-successional secondary forest exhibiting ecological attributes of old-growth forests, thereby enhancing the role of state forests in biodiversity protection and climate mitigation.

Plant diversity and conservation value of wetlands along a rural–urban gradient

Background: Wetlands are the most threatened ecosystem in South Africa despite the range of ecosystem goods and services they provide. A significant cause of wetland destruction and degradation is a lack of understanding, by planners, policymakers and developers, of their ecological and socio-economic importance.Objectives: This study assessed the floristic composition and diversity of wetlands in the former Tlokwe Municipal area along a rural–urban gradient.Methods: Fourteen wetland sites were surveyed along an urbanisation gradient. Vegetation surveys were done in quadrats along transects in each wetland recording the cover-abundance of each species. The data were analysed by using ordinations, similarity percentages, and the adjusted Floristic Quality Assessment Index.Results: Overall, the proportional species composition of urban and rural wetlands was mainly similar. Trends indicated that the alpha diversity increased with both habitat size and heterogeneity along a rural–urban gradient. In all wetlands, indigenous species were the most abundant, with the highest score in the largest urban wetland. The floristic quality varied widely along the gradient with none of the sites in pristine condition.Conclusion: The similarity in species composition and floristic quality of the wetlands, as well as the high levels of indigenous species richness, indicated that urban wetlands are worthy of conservation. However, the signs of disturbances and the presence of alien species means that restoration strategies need to be implemented to improve the quality of the wetlands.

Co-Variation among Vegetation Structural Layers in Forested Wetlands

Co-variation among vegetation structural layers occurs in some forests but has been minimally found in forested wetlands. We assessed co-variation in six vegetation layers (three size classes of trees, tree seedlings, shrubs, and herbs) in 39 forested wetlands including in five before and after invasion by the beetle emerald ash borer (Agrilus planipennis; EAB) in northwestern Ohio, USA. Across the 39 wetlands, cover of wetland herbs requiring full sunlight was negatively related to tree density, while herbs tolerant of shade minimally co-varied with tree layers. Several wetlands contained both large trees and regeneration (small trees and seedlings) of the same tree species, but often distributions of large trees and regeneration were disjointed. Variation in plant community quality (e.g., non-native cover, floristic quality) for understory layers was partly modeled (34–75% of variance) using multivariate combinations of tree layers. Low-density pin oak (Quercus palustris) flatwoods, which contained species-rich, high-quality understories, drove much co-variation in vegetation layers, suggesting that coupling of layers might occur on only segments of landscape gradients. Another factor was that nearly every site had a different dominant herbaceous species, producing extreme compositional heterogeneity (94% dissimilarity among plots), limiting possible co-variation, but creating high beta diversity. On the five long-term EAB sites, shrub and herb composition co-varied before EAB invasion, but not 14 years later after shrub cover doubled. High diversity in vegetation layers among sites suggests that conserving forested wetlands, including sites that individually might not have high floristic quality, can be a major contributor to landscape diversity.

Vegetation changes associated with release from cattle grazing in a WI calcareous fen

Although grazing was a part of calcareous fens’ disturbance regimes, few studies have examined the impact of grazing release in these systems, and managers are reticent to experiment with restoring grazing for management purposes. Here I describe vegetation, edaphic, and hydrologic differences in a Wisconsin calcareous fen between areas long-protected (since 1987) from cattle grazing and those protected in 2011, 2.5 years before the study. Vegetation surveys found that the long-protected areas had significantly lower floristic quality, fewer rare and specialist species, and more woody-plant encroachment than areas abandoned in 2011. Interviews with a farmer and air-photo interpretations suggested that vegetation changed in the protected areas around the time of abandonment. Known edaphic and hydrologic drivers of vegetation change in fens (volumetric water content, electrical conductivity, nutrient availability) could not account for the vegetation differences between the two areas. Site histories found no events accounting for these vegetation differences other than grazing. These results suggest that grazing release led to increased shrub encroachment, decline in floristic quality, and-possibly- a decline in specialist and rare species.

Niche ecology in Floristic Quality Assessment: Are species with higher conservatism more specialized?

Floristic Quality Assessment (FQA) metrics are used to assess the conservation value of natural areas based on the composition of their plant communities. FQA is based on coefficients of conservatism (C-values). C-values are regional, expert-assigned values, which range from 0 to 10 to reflect a taxon’s fidelity to high quality natural areas and its relative tolerance to modern human impacts. Approximately half of FQA publications assert without evidence that plant species with lower C-values are more generalist, whereas species with higher C-values are habitat specialists. The connection between specialization and C-values has never been investigated. Our primary objective was to test the claim that ecological specialization is associated with C-values. We first used the “theta (θ) method,” based on species co-occurrences, to estimate niche width for 274 species in 598 wetlands, grasslands, and forests in Illinois, USA. As a secondary proxy for niche width, we counted the habitat preferences of all Illinois vascular taxa listed from the Illinois Plant Information Network (ILPIN). For θ in grasslands and wetlands, and for ILPIN habitat counts, the lowest and highest C-values were associated with greater ecological specialization, whereas species with low to intermediate C-values were the most generalist. In forests, we found weaker, negatively linear relationships between C-values and θ, indicating that species with a high conservation value (i.e. conservative) were the most specialized. However, when shrub and canopy layers were combined with the herbaceous layer data in forests, we did not find an association between C-values and specialization. We also reanalyzed the data to determine the effect of non-native species on the relationship between niche width and C-values because non-native species, by default, receive C-values of 0. Removing non-native species from the analyses did not influence our conclusions. Both methods to measure specialization suggested there is a weak relationship between C-values and ecological specialization. Our findings support the common claim that some conservative species are more specialized than species with intermediate or low C-values. However, we found the general C-value and specialization relationship is more complex than assumed by many FQA publications because if it exists, it is not necessarily linear, i.e., species with low C-values are often specialists. Specialization and niche width could be latent components of C-values, but FQA users should avoid defining species conservatism by niche and specialization.

Taxonomic distinctness poorly reflects floristic quality in a wetland study system

Biomonitoring typically uses taxonomic diversity information while ignoring phylogenetic diversity. Evolutionary relatedness may offer deeper insight to how local assemblages relate with human disturbance and ecological degradation. Degradation of floristic quality may filter species with similar evolutionary traits, whereas intact floristic quality may limit phylogenetic clumping and increase representation of more distantly related taxa. We tested this hypothesis using average taxonomic distinctness (AvTD) and measures of floristic quality (mean ecological conservatism, native species richness, percent exotic species) in vascular plant assemblages of 115 wetlands in the US southern plains. In line with the hypothesis, we observed positive correlations with conservatism (r = 0.28, P = 0.0007) and native richness (r = 0.24, P = 0.0018) and a negative correlation (r = −0.21, P = 0.0169) with exotics, but the plotted relationships looked obscure. A strongly skewed AvTD distribution revealed a clear gradient in lower than expected AvTD. Responses along this gradient covaried with native richness (quadratic model R2 = 0.75) but showed no pattern with conservatism and a weak response to exotics. These results suggest that native richness has potential to predict lower than expected AvTD values. However, attributing these values to degraded floristic quality requires caution when richness is driven by sampling effects such as species-area relationships or has a non-linear relationship to human disturbance. Given the vague correlations and ambiguity of richness, plant taxonomic distinctness may not provide a clear bioindicator for wetlands. More work is needed to elucidate how evolutionary structure may play into bioassessment, which traditionally has been phylogenetically neutral.

Dramatic long-term restoration of an oak woodland due to multiple, sustained management treatments.

We measured 34 years of plant community change in a degraded oak woodland undergoing ecological management. Management included regular prescribed fire, control of white-tailed deer populations, repeated sowing of a diverse seed mix, and removal of invasive plants. We tracked change with several conservation metrics. Time series analysis showed no significant changes over time in either plant species richness or the Shannon-Weiner diversity index. Floristic Quality Assessment measures—the Floristic Quality Index (FQI), Cover-weighted FQI, and the Mean Coefficient of Conservatism (Mean C)—all increased dramatically over time, such that their values now surpass those of the highest quality representative of this habitat in the region. Cover-weighted FQI had the added benefit of being quick to respond (negatively and positively) to short-term management changes during the study. This sensitivity highlights its utility for adaptive management, enabling timely, data-driven changes to ongoing management regimes. Plant community composition showed striking changes during the study period, as species of high conservation value replaced weedier species. As a group, conservative woodland species are notoriously slow to recover from degradation, making this flora’s recovery particularly notable. A mid-study cessation of management immediately stalled the woodland’s recovery according to Floristic Quality metrics, but the restoration quickly returned to its positive trajectory with the resumption of management treatments. These results illustrate that impressive plant biodiversity restoration can be achieved, even in highly degraded contemporary oak ecosystems, if ecological management is comprehensive and if it is sustained over time.

Assessing Conservation Value and Restoration Priorities in an Agroecological Landscape

The integration of agricultural and natural ecological systems is increasingly viewed as an essential step toward achieving conservation goals, from local to global scales. For this study, we assessed the ecological conditions and conservation value of upland habitats on six Wisconsin potato farms participating in an ecolabel program that requires the implementation of ecological management plans on non-crop lands. Our objective was to determine how natural and restorable-to-natural elements of the landscape related to adjacent intensively managed agricultural fields, in order to prioritize restoration targets and activities. We compared the plant species richness, floristic quality, and vegetation structure between three dominant upland habitats: woodlands, pine plantations, and weedy fields. We recorded 205 native plant species across all sites (approximately 25% of the regional native flora), indicating that habitat patches surrounding agricultural lands can offer substantial conservation value. Woodlands had the highest average conservation value (mean of 43.8 native species per site) and weedy fields the lowest (mean of 6.3 native species per site). Habitat edges were characterized by a higher frequency of both exotic and prairie–savanna indicator species, representing a unique assemblage of species warranting special consideration for conservation and restoration. We recommend that restoration efforts on this and similar agroecological systems prioritize woodland edges and weedy corners, where prescribed fire, native plant seeding, and invasive species removal could produce significant conservation gains while reducing agricultural weed colonization of cultivated fields.

Use of Remote Sensing and Field Data to Quantify the Performance and Resilience of Restored Louisiana Wetlands

Typical goals of wetland restoration efforts are to conserve, create, or enhance wetland structure, and to achieve wetland function that approaches or exceeds natural conditions. Measuring wetland establishment, condition, and resilience can be difficult, especially because monitoring wetland function has traditionally been time-intensive, costly, and often required repeat field-based surveys. Remote sensing provides novel collections of data and facilitates rapid assessments of wetland landscapes, land cover, species/habitat composition, change detection, degradation, diversity, as well as system threats and pressures. A combination of remotely collected and in situ vegetation data were used in conjunction with landscape metrics and vegetative indices. These data were used to evaluate and compare changes and trends in condition, function and resilience of restoration sites and reference wetlands in southwest Louisiana, USA. Results of this work show the restored wetlands reached structural and functional equivalency to reference wetlands after approximately three to ten years post-construction. With adequate maturity, the restored wetlands outperformed the reference wetlands, having higher percentage of land, land aggregation, aboveground vegetation productivity and floristic quality. Supplementing traditional field-based methods with remote sensing applications provided enhanced metrics for inventorying and monitoring of wetland resources, forecasting of resource condition and stability, and adaptive management strategies.

Expert‐based measures of human impact to vegetation

AbstractBackgroundHuman impact has had a profound influence on modern vegetation. Expert‐based vegetation indicator systems have been developed to measure and characterize human impacts on vegetation. Floristic Quality Assessment (FQA) is a widely used, primarily North American system for assessing degradation and natural quality through species‐based values called coefficients of conservatism. The hemeroby and naturalness indicator value (NIV) systems were independently developed for users to assess human impacts in Europe. Despite the similarities among these indicator systems, there is no mutual recognition among them, and they have developed and operate in relative isolation from one another.ApproachWe review the FQA, hemeroby, and NIV systems to provide a basic summary of the three systems and to evaluate them, highlighting their nearly identical core mechanics and conceptual commonalities. We also compare and contrast the three systems to less integrative ecological indicator systems that can be used to indirectly measure human impact, notably the Ellenberg system.FindingsWe describe how FQA and naturalness indicator values, in particular, could be considered twin systems. Despite these core similarities, users of these systems do not cite each other, potentially overlooking benefits from applying methods and concepts from separate systems. The FQA, hemeroby, and NIV systems have unique weaknesses, strengths, and primary applications compared to other ecological indicator systems.ConclusionsWe conclude by discussing the future role and utility of the three specialized human impact indicator systems for scientists and practitioners. Human impact indicator values may be valuable for use in basic research, but arguably their most important applications are in the practice of conservation, such as monitoring restored ecosystems.

Floristic and Macroinvertebrate Responses to Different Wetland Restoration Techniques in Southeastern Wisconsin

Assessments of wetland restoration quality, especially comparisons among restoration techniques, are essential to adaptive management, but are rarely reported. We evaluated wetlands in southeastern Wisconsin that were restored using different hydrological techniques along with wetlands in two reference groups: wetlands without hydrological modification and Waterfowl Production Areas. Our objectives were 1) to compare condition of wetlands using indices for plant communities and macroinvertebrates and 2) evaluate provisioning of macroinvertebrates for waterfowl broods as a measure of habitat quality. Overall, floristic quality of sedge meadows and emergent marshes was poor in restored wetlands (weighted mean C-values were 1.55 ± 0.13 (SE) and 1.47 ± 0.25, respectively); however, sedge meadow quality was greater in wetlands without surface modification (weighted mean C-value = 3.43 ± 0.82). On a scale from “excellent” to “very poor”, we classified condition of open water communities as “good” based on the macroinvertebrate index across all restoration categories. Density and abundance of macroinvertebrate taxa preferred by waterfowl broods were similar across restoration categories and indicated that wetland basins provide ample food resources. We recommend that conservation planners and wetland managers state quantifiable, appropriate objectives for restoration ahead of implementation and address conflicting objectives with stakeholders.

Plant Community Response to Hydrologic Restoration in the Great Dismal Swamp

Peatlands in the mid-Atlantic outer coastal plain region contain obligate hydrophyte species which were harvested and replaced by facultative tree species. The Great Dismal Swamp was drained from the colonial era until 1974, when water levels were partially restored. In September 2013, further restoration consisting of two large weirs followed extensive peat-burning fires. This study evaluated depth-to-water-table and vegetation structure both prior to and following weir operation. Wells were installed and depth-to-water-table was recorded continuously from 2013 to 2015 within six of the 15 forested stands where vegetation species dominance was quantified for tree, shrub, herb, and vine strata. Following weir installation in 2013, water tables rose an average of 28.08 cm in 2014 and 32.69 cm in 2015, during the June–July peak of the growing season. Most water levels were too low to meet the federal regulatory indicator of wetland hydrology or the seasonally flooded, saturated hydrologic regime typical of peatlands. After restoration, species dominance and frequency, as well as metrics based on hydrophyte dominance and floristic quality in study plots, were unchanged. Ordinations detected no directional shift in plant community composition among pre- and post-weir periods. Although insufficient time may have passed for shifts in plant communities, additional increases in water level above those reported here appear necessary to restore a pre-disturbance hydrologic regime and plant community structure. However, above some threshold water level, planted trees will exhibit increased mortality and limit reestablishment.

Development of a modified floristic quality index as a rapid habitat assessment method in the northern Everglades.

Floristic quality assessments (FQA) using floristic quality indices (FQIs) are useful tools for assessing and comparing vegetation communities and related habitat condition. However, intensive vegetation surveys requiring significant time and technical expertise are necessary, which limits the use of FQIs in environmental monitoring programs. This study modified standard FQI methods to develop a rapid assessment method for characterizing and modeling change in wetland habitat condition in the northern Everglades. Method modifications include limiting vegetation surveys to a subset of taxa selected as indicators of impact and eliminating richness and/or abundance factors from the equation. These modifications reduce the amount of time required to complete surveys and minimizes misidentification of species, which can skew results. The habitat characterization and assessment tool (HCAT) developed here is a FQA that uses a modified FQI to detect and model changes in habitat condition based on vegetation communities, characterize levels of impact as high, moderate, or low, provide predictive capabilities for assessing natural resource management or water management operation alternatives, and uniquely links a FQI with readily accessible environmental data. For application in the northern Everglades, surface water phosphorus concentrations, specific conductivity, distance from canal, and days since dry (5-year average) explained 67% of the variability in the dataset with > 99.9% confidence. The HCAT approach can be used to monitor, assess, and evaluate habitats with the objective of informing management decisions (e.g., as a screening tool) to maximize conservation and restoration of protected areas and is transferable to other wetlands with additional modification.

Impacts of groundwater extraction on calcareous fen floristic quality.

Groundwater withdrawal has increased over the past several decades throughout the U.S. Upper Midwest, yet impacts of pumping on groundwater-dependent wetlands remain understudied. Here, we compared measures of floristic quality, hydrologic conditions, and nutrient availability in pairs of more-impacted fens and less-impacted fens throughout Wisconsin. Floristic quality was significantly lower in more-impacted fens than in less-impacted fens, the result of the disappearance of rare and specialist species and the increase in richness and cover of non-native and weedy species. Plots within more-impacted fens generally had lower root-zone volumetric water content, greater depth to water table, and higher available nitrogen and phosphorus than within less-impacted fens, although nonuniformly among or within sites. Lower volumetric water content predicted plot-level declines in floristic quality, richness of rare or specialist species, an increase in the number of non-native or problematic species, and an increase in cover of non-native and problematic species. Our results strongly suggest that groundwater withdrawals have substantial negative impacts on nearby fen quality and furthering imperilment of several species they contain.