Relationships between plant functional diversity, species diversity and productivity depend on semi-natural grassland habitat type

Relationships between plant functional diversity, species diversity and productivity depend on semi-natural grassland habitat type | Authorea try { document.documentElement.classList.add('js'); } catch (e) { } var _gaq = _gaq || []; _gaq.push(['_setAccount', 'G-8VDV14Y67G']); _gaq.push(['_trackPageview']); (function() { var ga = document.createElement('script'); ga.type = 'text/javascript'; ga.async = true; ga.src = ('https:' == document.location.protocol ? 'https://ssl' : 'http://www') + '.google-analytics.com/ga.js'; var s = document.getElementsByTagName('script')[0]; s.parentNode.insertBefore(ga, s); })(); Skip to main content Preprints Collections Wiley Open Research IET Open Research Ecological Society of Japan All Collections About About Authorea FAQs Contact Us Quick Search anywhere Search for preprint articles, keywords, etc. Search Search ADVANCED SEARCH SCROLL This is a preprint and has not been peer reviewed. Data may be preliminary. 13 January 2026 V1 Latest version Share on Relationships between plant functional diversity, species diversity and productivity depend on semi-natural grassland habitat type Authors : Eoin Halpin 0009-0006-0819-1261 [email protected] , Oliver Lynch Milner , Samuel Hayes 0000-0002-2710-7305 , Karen Bacon , Fiona Cawkwell 0000-0002-7365-8909 , and Astrid Wingler Authors Info & Affiliations https://doi.org/10.22541/au.176830949.99236894/v1 245 views 132 downloads Contents Abstract Supplementary Material Information & Authors Metrics & Citations View Options References Figures Tables Media Share Abstract Semi-natural grasslands provide many ecosystem services with biomass production for livestock being of major importance. Both plant species diversity (i.e. taxonomic diversity at the species level) and functional diversity can affect grassland productivity and ecosystem services. However, previous research has revealed positive, negative or unimodal diversity–productivity relationships in grasslands. This research focused on three different Irish semi-natural grassland habitats: dry calcareous and neutral grassland (GS1), dry-humid acid grassland (GS3) and wet grassland (GS4). Comparison of functional diversity revealed that GS1 and GS4 grasslands had higher functional richness than GS3. Classification of strategies from leaf traits according to the competitor, stress tolerator, ruderal (CSR) system showed that GS4 had the highest values for competitiveness (C) and GS3 the lowest. GS3 had higher values for stress tolerance (S) than the other two habitats, but ruderality (R) did not differ among the habitats. Based on differences found between the habitats, we tested the hypothesis that diversity-productivity relationships depend on grassland habitat type. For all habitats combined, functional diversity was positively correlated with species richness, Simpson diversity and Simpson evenness, but not for the generally species-rich GS1 grasslands. Relationships between productivity (NDVI derived from UAV surveys) and diversity were overall negative, with the most strongly negative diversity-productivity relationship found in the GS3 grassland habitat. Community-weighted means of leaf dry matter content (CWM–LDMC) and community S strategy were also negatively correlated with productivity in GS3 grasslands, but no relationship between CWM-LA or C strategy was found in any of the grassland habitats. Differences in diversity-productivity relationships in different habitats suggest that management or soil type influence the nature of these relationships, especially in stressful acid grasslands. INTRODUCTION Grasslands provide a range of ecosystem services globally such as climate regulation, erosion control and pollination service, as well as forming the main basis for meat and dairy production (Bengtsson et al., 2019; Zhao et al., 2020). Semi-natural grasslands, which are extensively managed by livestock grazing or hay cutting but without ploughing or sowing, have been shown to be one of the most diverse ecosystems, harbouring some of the highest number of vascular plant species recorded (Wilson et al., 2012; Chytrý et al. 2015). With the global demand for produce from livestock ever increasing (Hemathilake and Gunathilake, 2022), the intensity of management has increased in grasslands and is leading to a loss of semi-natural habitat (Bullock et al., 2011). Increases in management intensity have been shown to increase productivity (Lemaire et al., 2012) while having negative effects on functional diversity and species diversity (Guo, 2007; Harpole et al., 2017; Andraczek et al., 2023: Zhang et al., 2023). High species diversity is important for ecosystem health as it is beneficial for genetic diversity, for pollinator species, and for facilitating species from other trophic levels, while protecting against the invasion of undesirable species (Spehn et al., 2000; Fargione and Tillman, 2005; Bullock et al., 2011). Species diversity may also be important for productivity as complementarity effects can support over-yielding (Loreau et al., 2001). However, species or phylogenetic diversity cannot explain the impact of environmental change on grassland productivity (Xu et al., 2018; Petchey et al., 2004). Instead, the diversity of functional traits (functional diversity) was shown to have a higher explanatory power than species diversity or functional group diversity on grassland productivity (Petchey et al., 2004). The maintenance of high functional diversity is also imperative for the ecological stability of grassland ecosystems in response to rainfall conditions (Hallet et al., 2017). Many studies have analysed the relationship between functional diversity, species diversity and grassland productivity (e.g. Roscher et al., 2012; Polley et al., 2013; Harpole et al., 2016; Zhang et all. 2017). Some of the literature suggests that increased biodiversity in grasslands has positive effects on productivity and yield stability (Tilman et al., 1996; Tilman et al., 2001; Binder et al., 2018; Zhang et al., 2017; Korrell et al., 2024) with the relationship being in part due to an increase in diversity of functional traits that co-occur with increases in biodiversity (Petchey et al., 2004). Other research has found unimodal relationships that are dependent on management intensity, indicating an optimum management strategy that benefits both productivity and diversity (Wang et al., 2022). Such unimodal relationships can shift to negative relationships when grassland management intensity is high (Brun et al., 2019). Recently, it was shown that absolute abundance of the most dominant species predicts productivity, but the relative abundance of these species is negatively related with plant species richness in global grasslands (Zhang et al., 2025). Grassland ecosystem function and productivity depend directly on the functional traits (Lavorel and Garnier, 2002), such as plant height of dominant species (Zhang et al., 2017; Xu et al., 2018). Among functional traits, three leaf traits were identified as indicators of plant functional strategies by Pierce et al. (2013): leaf area (LA) for photosynthetic organ size, specific leaf area (SLA) indicating an acquisitive strategy, and leaf matter content (LDMC) indicating a conservative strategy. It was shown that these traits can be used to classify functional strategies according to Grime’s (1977) competitor, stress tolerator, ruderal (CSR) concept (Pierce et al., 2013; 2017). Research for grass species shows that CSR strategies determined from the leaf traits are also reflective of other functional traits, such as scores for C being correlated with plant height and those for R with specific root length (Wingler and Sandel, 2023). Li et al. (2024) determined CWM (community-weighted mean) values of CSR strategies for grasslands on the Qinghai-Tibetan Plateau, showing that these grasslands are mainly R- or SR-dominated, but with a slight shift to more C strategies upon nitrogen addition. In addition, Wirth et al. (2024) recently demonstrated that integration of CSR strategies into a vegetation model can be used to predict grassland productivity with the C functional type predominantly contributing to biomass in temperate grasslands. We therefore expected communities dominated by more competitive species (characterized by high CWM-LA) to have higher productivity. Diversity-productivity relationships are highly dependent on other factors such as management practices and abiotic conditions (Wang et al., 2022; Ma et al., 2010). Different grassland habitats that vary in management and environment, especially soil conditions, should therefore show different diversity-productivity relationships. In addition, predominant functional strategies are likely to be affected by the grassland habitat. This paper aims to define the relationships between functional strategies, species diversity, functional diversity and productivity in three Irish semi-natural grassland habitat types (Fossitt, 2000); dry calcareous and neutral grassland (GS1), dry-humid acid grassland (GS3) and wet grassland (GS4). The three habitats showed clear differences in functional diversity and community functional strategies. We therefore hypothesized that some of the discrepancies in the literature for diversity-productivity relationships can be explained with grassland habitat type. We further determined the relationship between species diversity and functional diversity metrics and investigated whether functional strategy, species diversity or functional diversity is the main determinant of productivity. MATERIALS AND METHODS Site selection and surveying Twelve semi-natural grassland sites were selected for this study, four sites for each of three semi-natural grassland habitats categorised by Fossitt in Ireland (Fossitt, 2000): GS1 - dry calcareous and neutral grassland, GS3 - dry-humid acid grassland and GS4 - wet grassland. These sites did not exclusively contain one Fossitt habitat type but instead are described based on the dominant habitat type present. Previously studied sites from the 2007-2012 Irish semi-natural grassland survey (ISGS) (O’Neill et al. 2013) were used as a basis for selection of semi-natural grasslands sites in the western part of Ireland with high precipitation (880 mm to 2,712 mm annually; Supporting Information Table S1). The three grassland habitat types differed in pH with GS1 being most alkaline (pH = 6.94), GS4 slightly more acidic (pH = 6.6) and GS3 being most acidic (pH = 5.71). This confirms the Fossitt habitat classification, which includes soil pH in describing habitat type. The categorization of relevés into habitat types was based on the habitat classification assigned to each relevé during a botanical survey in 2023 (Lynch Milner et al., unpublished). This designation was used instead of the overall site level habitat classification (provided by the ISGS survey in 2012) because many relevés, particularly in GS1 sites, were identified as GS4 habitat. Sample collection and storage Leaf samples were collected from the months May to July 2024 with sites from different habitats chosen every week to exclude any seasonal variation in sampling times between habitat types. Leaf samples were taken from each 2 m x 2 m relevé recorded for each of the twelve sites. Five leaf samples were taken per species, each leaf from a different individual. To ensure sample collection was reflective of the relevé’s species composition and abundance, only species with more than 3% abundance were collected so that roughly 85-90% of relative species abundance (for all higher plants) was represented. Leaves were fully rehydrated in the laboratory (Garnier, 2001b). Leaf fresh weight was measured first. To measure leaf area (LA) a flatbed scanner was used to produce high-quality images of the leaves with a ruler to scale. LA was calculated using the ImageJ software. The samples were then placed into a fan-assisted oven at 60 ℃ for 72 hours after which leaf dry weight was measured. Specific leaf area (SLA) was calculated by dividing the leaf area by the dry weight; leaf dry matter content (LDMC) was calculated by dividing the dry weight by the fully hydrated fresh weight, expressed as percent. UAV surveys to determine NDVI Grassland productivity was indirectly measured using Normalized Difference Vegetation Index (NDVI) values, which act as a proxy for above-ground net primary productivity (Gu et al., 2013). A DJI Mavic 3 Multispectral quadcopter (https://ag.dji.com/mavic-3-m) was used to capture high resolution imagery of the study sites. Surveys were conducted in June and July of 2024 at times ranging from 10.50 am to 6.26 pm, depending on location and weather conditions (Table S2). Most flights were pre-planned to capture imagery at a height above ground of 120 m and with 70% side and front overlap. Some sites had highly variable and steep terrain, requiring the UAV to vertically track a digital elevation model to maintain a consistent height above ground. The small mismatches between the actual terrain and the digital elevation models introduced some variability in the overall ground resolution, which ranged from 5.0 cm to 6.1 cm across the sites. Measurements were not taken for every relevé, due to problems with accessibility, so any analysis performed in relation to productivity included a reduced dataset: Sites 6, 7, 8 and 11 were not surveyed, resulting in one GS1, one GS3 and two GS4 grasslands being excluded. Orthomosaics were generated in Agisoft Metashape Professional 1.7.5 and the projection set to Irish Transverse Mercator. NDVI (= (near infrared-red)/(near infrared + red)) maps were then created before being exported as geotiff images for further analysis. Settings and filters used in generating the orthomosaics are available in Supporting Information Table S3. Replication statement Scale of inference Scale at which the factor of interest is applied Number of replicates at the appropriate scale Site Relevé 3 to 6 relevés for traits 2 to 6 relevés for NDVI Grassland habitat type Relevé 17 (GS1), 12 (GS3), 19 (GS4) relevés for traits 13 (GS1), 7 (GS3), 10 (GS4) relevés for NDVI Data analysis Since Fossitt grassland habitat type sometimes varied among relevés within a site, we based analyses at the habitat level on the classification of the relevés, not the site overall. The leaf trait values taken from the five samples of each species were averaged. The resulting values were multiplied by the species abundance (taken from the vegetation survey) as the proportionate effect a functional trait has on its community is influenced by its abundance (Diaz and Cabido 2001). These weighted values were then averaged across all selected species in each relevé to determine the community weighted mean (CWM) values. Competition-stress tolerance-ruderality (CSR) functional strategies were determined from the community weighted mean leaf area (LA), specific leaf area (SLA) and leaf dry matter content (LDMC) using the StrateFy tool (Pierce et al. 2017). To derive metrics for functional diversity the R package FD (Laliberté & Legendre 2010), downloaded through R statistical package version 4.3.3 (R Core Team 2023), was used to analyze the trait data. By providing the dbFD function with individual abundance and trait matrices for each relevé, the dbFD function produces multiple indices associated with functional diversity including: FRic (an index for functional richness), FEve (functional evenness), FDiv, FDis (two indices for functional divergence) (Botta & Dukát, 2005; Villéger et al., 2008; Petchey and Gaston, 2006). These indices were calculated for each relevé and compiled into a single data frame. The rao.diversity() function from the R package SYNCSA was also used to calculate Rao’s functional diversity (FDq) (a measure of functional divergence) as well as functional redundancy (FRed) which is a measure of the difference between species diversity and the trait divergence of species (FDq) (de Bello et al. 2007). To visualize how FD metrics and community strategies varied in relation to each other and across habitats, a non-metric multidimensional scaling (NMDS) ordination plot was produced using dissimilarity matrices derived from species abundance data. The functional diversity metrics and (CSR) functional strategies were plotted onto the NMDS plot using the enfit() function. Species diversity metrics were calculated using the ‘Vegan’ package in R. This analysis provided values for Simpson diversity/evenness, Shannon diversity/evenness and species richness. Correlation tests using the non-parametric Spearman method were performed between the CWM leaf trait values, functional diversity, species diversity metrics and NDVI. To determine differences in functional diversity metrics and NDVI in the three habitat types, a Kruskal-Wallis test was performed. Post hoc, Wilcoxon tests were performed after analysis of variance to assess what sites/habitats differed significantly. Community weighted mean leaf trait values A significant difference was found in community-weighted mean leaf area (CWM-LA) values between habitat types (p < 0.0001) (Fig. 1a). The GS4 habitat had the highest leaf area followed by GS1 and GS3. A Wilcoxon test between habitat types indicated that all habitats differed significantly from each other with the GS4 and GS3 habitats differing most significantly (p <0.0001). While a Kruskal-Wallis test found no significant differences in the grassland habitats for community-weighted mean specific leaf area (CWM-SLA) (Fig. 1b), significant differences were found in community-weighted mean leaf dry matter content (CWM-LDMC) values between habitats (Fig. 1c). A Wilcoxon test between habitat types indicated that GS3 relevés had significantly higher CWM-LDMC values than both GS1 and GS4 relevés, whereas the difference between GS1 and GS4 was insignificant. Figure 1. Differences between habitats in community-weighted mean leaf area (a), specific leaf area (b) and leaf dry matter content (c). The Fossitt habitat is indicated by the colour of boxplots (GS1 = green, GS3 = red, GS4 = blue). The length of the box indicates the interquartile range. The band in the middle of the box represents the median value. The whiskers either side of each box indicate minimum and maximum values with values beyond these points considered outliers. Coloured dots represent individual relevés. The significance of the differences between categorical variables (Kruskal-Wallis) is indicated by the number of red asterisks in the top left corner (p individual habitats (Wilcoxon test) is indicated by the black asterisks above the brackets linking each habitat type. Above-ground net primary productivity (NDVI) NDVI values differed significantly between sites (p <0.01) and between habitat types (p <0.05) (Fig. 2). GS4 relevés had the highest average NDVI followed by GS3 and then GS1. GS4 was significantly different from both GS3 and GS1 habitats (Fig. 2a). Site 3 and Site 12 (GS1 and GS3 respectively) had the lowest NDVI with site 9 and 10 (GS4) having the highest average NDVI values (Fig. 2b). Figure 2. Differences in NDVI values between habitats (a) and sites (b). The Fossitt habitat is indicated by the colouration of boxplots (GS1 = green, GS3 = red, GS4 = blue). The length of the box indicates the interquartile range. The band in the middle of the box represents the median value. The whiskers either side of each box indicate minimum and maximum values with values beyond these points considered outliers. Coloured dots represent data points (these points are scattered for visual purposes). The significance of the difference between categorical variables (Kruskal-Wallis) is indicated by the number of red asterisks in the top left corner (P >0.05, <0.05*, significance of difference between individual habitats (Wilcoxon test) is indicated by the black asterisks above the brackets linking each habitat type. CSR functional strategies The CSR strategies for the vegetation communities were calculated from the CWM traits using the Stratefy Tool (Pierce et al. 2017). It is clear from the ternary diagrams (Fig. 3a-c) that all habitats were dominated by stress tolerant/ruderal species. GS3 had communities that ranged widely on the stress tolerance-ruderality axis; however competitive strategies were not prevalent in these communities (Fig. 3b). GS1 and GS4 grassland communities had similar strategies with GS4 communities being more competitive (Fig. 3a, b). The competitiveness strategy differed significantly between the three habitats (p <0.0001) (Fig. 3d). GS4 communities were the most competitive followed by GS1 and GS3. Significant differences were also found in stress tolerance strategy (p <0.01) with GS3 having the highest scores for stress tolerance followed by GS1 and GS4 (Fig. 3e). Ruderality did not differ significantly between habitat types (Fig. 3f). Figure 3. CSR strategies in each habitat type and variation in community strategies between habitats. The ternary diagrams (Figure 3, a, b & c) illustrate the strategies for all relevés in GS1, GS3 and GS4 habitats respectively. In the boxplots (d, e & f) The Fossitt habitat is indicated by the colour (GS1 = green, GS3 = red, GS4 = blue). The length of the box indicates the interquartile range. The band in the middle of the box represents the median value. The whiskers either side of each box indicate minimum and maximum values with values beyond these points considered outliers. Coloured dots represent individual relevés). The significance of the differences between categorical variables (Kruskal-Wallis) is indicated by the number of red asterisks in the top left corner (p individual habitats (Wilcoxon test) is indicated by the black asterisks above the brackets linking each habitat type. Functional diversity The functional diversity metrics calculated were based on the three leaf trait measurements for each vegetation community. These metrics include functional richness (FRic), functional evenness (FEve), functional divergence (FDiv) and Rao’s functional diversity (FDq). FRic differed significantly between the three grassland habitats (p < 0.05) (Fig. 4a). Post-hoc analysis using a Wilcoxon test found significantly lower FRic for GS3 than the two other grassland habitats; however no significant difference was observed between GS1 and GS4. A Kruskal-Wallis test found no significant difference in FEve values between grassland habitats (Fig. 4b). Similarly, no significant difference was found for FDiv (Fig. 4c) or FDq (Fig. 4d). Figure 4. Differences in community functional richness (FRic) (a), functional evenness (FEve) (b), functional divergence (FDiv) (c) and Rao’s functional diversity (FDq) (d) between habitat types. The Fossitt habitat is indicated by the colour of boxplots (GS1 = green, GS3 = red, GS4 = blue). The length of the box indicates the interquartile range. The band in the middle of the box represents the median value. The whiskers either side of each box indicate minimum and maximum values with values beyond these points considered outliers. Coloured dots represent individual relevés. The significance of the difference between categorical variables (Kruskal-Wallis) is indicated by the number of red asterisks in the top left corner (p individual habitats (Wilcoxon test) is indicated by the black asterisks above the brackets linking each habitat type. Association of functional strategies and diversity metrics with habitat type FDiv increased opposingly to the other functional diversity metrics with GS4 and GS3 habitats having higher values (Fig. 5). FRic was associated with the GS1 and GS4 habitats, while FRed, FEve, FDq were highest in the GS1 habitat. Scores for S were highest in the GS3 habitat, the GS4 habitat had the highest C scores, while R was associated with the GS1 habitat. Figure 5. Relationship between functional diversity metrics, CSR strategies and the three grassland habitats. The ordination plot is based on non-metric multidimensional scaling (NMDS) where patterns in species abundances in each relevé are analysed. The two axes that explain most of the variation in abundances are plotted (NMDS1 & NMDS2). The data points represent individual relevés and are coloured based on their Fossitt habitat. The area of multidimensional space taken up by each habitat is represented by the convex hulls, each of which are also coloured based on habitat. The dotted lines represent the functional diversity metrics with the direction of the line indicating the directionality of the vector and with the length of each line indicating how strongly the FD metric correlates with the ordination axes. FRic (functional richness), FEve (functional evenness), FDiv (functional divergence), FDq (and Rao’s quadratic entropy), C (competitiveness), S (stress tolerance), R (ruderality). Relationship between functional richness and species diversity Since FRic differed between the habitats (Fig. 4a), we analysed the relationship between this functional diversity metric and species diversity. For all habitats combined, FRic showed significant positive correlations with species diversity metrics, including species richness (Fig. 6a), Simpson diversity (Fig. 6b) and Simpson evenness (Fig. 6c), with Simpson diversity being the most strongly correlated (p = 0.0001). When the habitats were analysed separately, GS3 communities had a positive correlation between FRic and all species diversity indices. GS4 communities showed a significant positive correlation of FRic with Simpson diversity and evenness, while the correlation with species richness was marginal (p = 0.054). In contrast, for GS1 communities no significant relationship was found between FRic and species diversity and, in the case of species richness, there appeared to be a negative trend (Fig. 6a). Overall positive relationships were found between functional evenness and Rao’s functional diversity with species diversity (Supporting Information Figs. S1 and S2). Figure 6. Relationship between functional richness and species diversity. Relationship between FRic and species richness (a), Simpson diversity (b) and Simpson evenness (c). The overall regression between the two variables is indicated by the black line and the regression based on habitat type is indicated by coloured lines (green for GS1, red for GS3 and blue for GS4). The statistics are derived from Spearman-Rank tests with the same colours. Data points, representing individual relevés, are also coloured based on habitat. Relationship between species diversity and productivity The species diversity metrics showed a range of patterns with NDVI values (Fig. 7). Species richness (p = 0.03) and Simpson diversity (p = 0.008) showed significant negative correlation with NDVI (Fig. 7a,b). For GS3 communities these negative correlations were also significant. While the overall correlation between Simpson evenness and NDVI was not statistically significant (p = 0.07), GS3 habitats showed significant declines in NDVI with higher species evenness (p = 0.002) (Fig. 7c). Figure 7. Relationship between species diversity metrics and grassland productivity. Relationship between grassland yield (NDVI) with species richness (a), Simpson diversity (b) and Simpson evenness (c). The overall linear regression between the two variables is indicated by the black line and the regression based on habitat type is indicated by coloured lines (green for GS1, red for GS3 and blue for GS4). The statistics are derived from Spearman-Rank tests. Data points, representing individual relevés, are also coloured based on habitat. Relationship between functional diversity and productivity To test if functional diversity drives productivity, we determined the relationship between the functional diversity metrics and productivity assessed from NDVI. Overall, FRic showed a significant negative correlation (p = 0.04) with NDVI (Fig. 8a). There was also a significant negative correlation between FRic and NDVI when the relationship was analysed for GS3 grassland separately (p = 0.049). FEve also showed a significant negative correlation with NDVI values when including all relevés (p = 0.03) but no significant correlation existed within individual habitats (Fig. 8b). Rao’s functional diversity showed a significant negative correlation p = 0.038) with NDVI values (Fig. 8c), However, this was not statistically significant for individual habitat types. CWM-LDMC showed a significant negative correlation (p = 0.002) with NDVI values (Fig. 8d). This negative relationship was significant in GS3 grasslands (p = 0.034) but not in GS1 or GS4. No correlation was found between CMW-LA or CMW-SLA and NDVI (Supporting Information Fig. S3). Figure 8. Relationship between functional diversity metrics and grassland productivity. Relationship between grassland yield (NDVI) and FRic (a), FEve (b), FDq (c) and CWM-LDMC (d). The overall linear regression between the two variables is indicated by the black line and the regression based on habitat type is indicated by coloured lines (green for GS1, red for GS3 and blue for GS4). The statistics are derived from Spearman-Rank tests. Data points, representing individual relevés, are also coloured based on habitat. Relationship between CSR strategies and productivity Despite our initial hypothesis that competitiveness drives productivity, and the higher productivity and C values in GS4 (Figs. 3 and 4), CWM values for competitiveness (C) showed no correlation with NDVI values (Fig. 9a). However, CWM values for stress tolerance (S) showed a significant negative correlation with NDVI (Fig. 9b). GS3 grasslands were the only habitat to also show this significant negative correlation when habitats were analysed separately. The CWM values for ruderality (R) showed an overall significant positive correlation with NDVI however there was no significant pattern present within the individual habitats (Fig. 9c). Figure 9. Relationship between CSR strategy and grassland productivity. Relationship between grassland yield (NDVI) with competitive strategy (a), stress tolerance strategy (b) and ruderal strategy (c). The overall linear regression between the two variables is indicated by the black line and the regression based on habitat type is indicated by coloured lines (green for GS1, red for GS3 and blue for GS4). The statistics are derived from Spearman-Rank tests. Data points, representing individual relevés, are also coloured based on habitat. DISCUSSION Significant differences in functional traits, CSR strategy, productivity, and functional richness were found between the habitat types (Figs. 1-4). Habitat type was associated with functional strategy and functional diversity (Figs. 4 and 6). While neither CWM values of LA nor competitiveness drove productivity (Fig. 9, Supporting Information Fig. S3), CWM values for stress tolerance and LDMC (Fig. 8) were negatively correlated with productivity. Relationships between functional diversity and species diversity differed depending on the habitat type (Fig. 6). Neither species diversity nor functional diversity supported increased productivity (Figs. 8 and 9). Instead, we detected negative diversity-productivity relationships, especially in GS3 (dry-humid acid grassland). Grassland communities from different habitat types differ in their functional strategies We determined leaf functional traits by field sampling of plants growing in the semi-natural grassland communities at each site. Other research on grassland functional diversity has relied on database values (Rolo et al., 2016; Anderegg et al., 2024). While database values reflect species composition, they do not represent intraspecific genetic variation or phenotypic plasticity. The functional strategies of grassland communities presented here are therefore reflective of the plant community composition as well as the response of the individuals in the community to the surrounding environment (Moog et al., 2005; Yu et al., 2022). The semi-natural grassland habitats monitored in this research differed significantly in their community-weighted mean functional trait values (Fig. 1) and their vegetation functional strategy (Fig. 3, Fig. 5). Li et al. (2024) also determined CSR strategies using the StrateFy tool for grassland communities. Compared to their research in high-altitude grasslands, where competitiveness was almost absent, the scores for competitiveness were higher in the grasslands studied here (Fig. 3). GS4 grasslands had the highest competitiveness scores, reflecting high CWM-LA values. As competitive species grow in the absence of stress and disturbance (Grime, 1977), reduced disturbance by lower grassland management increases competitiveness (Moog et al. 2005). High competitiveness of the GS4 habitat in our study could thus be related to under-management or abandonment. The GS4 habitat was also the most productive (Fig. 2), but contrary to our initial hypothesis, no positive correlation was found between scores for competitiveness and productivity for any of the habitats (Fig. 9). The CSR stress tolerance strategy was highest in GS3 grasslands (Fig. 3). The stress factors in the GS3 environment likely arise from grazing, from exposure to heavy winds or high precipitation (between 1570 and 2712 mm annually; Supporting Information Table S1). Low soil pH can also be a strong stress factor, negatively influencing plant growth in this habitat (Foy, 1984). The high CWM-LDMC (Fig. 1) indicates a resource-conservative strategy (Wright et al., 2004) associated with lower productivity than in GS4 (Fig. 2), and also suggests lower palatability (Dostálek et al., 2020), which is not favourable for livestock management (Bumb et al., 2016). Many GS3 relevés were dominated by low-palatability species such as Nardus stricta which had an average S-strategy score of 96% suggesting poor grazing potential. While ruderality was associated with GS1 (Fig. 5), it was mostly below 50% for all habitats, in contrast to high ruderality found in high-altitude grasslands (Li et al., 2024). This indicates good grassland management (Moog et al., 2005) as grasslands that become more degraded (mainly from overgrazing) tend to show a shift in strategy from stress tolerance to ruderality (Zhou et al., 2021). Some outlier relevés with high ruderality in our data (Fig. 3) may therefore indicate overgrazing. Habitat types differ in functional richness and in the relationship between functional richness and species diversity FRic differed significantly between the grassland habitats (Fig. 4) and was strongly correlated with Simpson Diversity (Fig. 6). The high FRic values found at some sites in the GS4 habitat are likely indicative of a more favourable environment where no environmental filtering is occurring and so a variety of niche spaces can be occupied. Such a relaxation of environmental constraints may also benefit the competitiveness (Fig. 3) and productivity (Fig. 2) of the GS4 habitat. FRic was also significantly higher in GS1 habitats compared to GS3. The spatial heterogeneity in GS1 grasslands, with highly variable soil depths and its exposed karst landscape, likely had an influence on the high FRic values by increasing the variation in microhabitats (Schippers et al., 2022). FRed (a measure for functional redundancy) was also significantly higher in GS1 than GS3 communities (Supporting Information Fig. S4). FRed is as important as FRic in relation to ecosystem health as higher FRed minimizes the impact that species diversity collapse will have on ecosystem functions (Fonseca and Ganada, 2001). The absence of differences in the other functional diversity metrics between habitats, especially in FDq which is closely related to complementarity effects, indicates that there is no difference in niche partitioning between the grassland habitat types (Roscher et al., 2012). It was expected that a greater diversity of species should come with a greater amount of trait variation (Cadotte et al., 2011). However, relationships between functional diversity and species diversity have been shown to vary along environmental gradients such as disturbance levels and nutrient availability (Cadotte et al., 2011; Mayfield et al., 2010; Biswas and Mallik, 2010). Previous research has thus shown that functional and taxonomic diversities can be uncoupled (Carmona et al., 2012), and such uncoupling of functional and phylogenetic diversity is common in European grasslands (Večeřa et al., 2023). In contrast to the positive correlations in GS3 and GS4 habitats, GS1 showed no correlation between FRic and species diversity (Fig. 6). This is surprising because, in contrast to other functional diversity indices, FRic is not independent from species richness (Villéger et al., 2008). For GS1, this may be due to the consistently high species diversity present in this habitat. The greater the number of species in a community and the greater the evenness of species abundances (Simpson evenness), the less functionally dissimilar species are as there are only a finite number of functional strategies (Mayfield et al., 2010). Therefore, the functional redundancy (amount of redundant species with functionally identical traits) will increase, but FRic will not increase further with increased species number. Another reason why the functional diversity-species diversity relationship is not apparent in GS1 grasslands could be due to increases in intraspecific functional diversity, which allows greater niche overlap (allowing for more species to coexist) without influencing community functional diversity metrics (Mason et al., 2011). This pattern has been observed in other dry limestone grasslands (Le Bagousse‐Pinguet et al., 2014). In contrast, in GS3 communities, where species richness is low, higher numbers of species can lead to increased functional diversity as more species will lead to greater functional trait variation. FRed (a measure for functional redundancy) was significantly lower in GS3 than GS1 communities which supports this explanation (Supporting Fig. S4). Higher species and functional diversity are associated with lower, not higher productivity, especially in acid grasslands We hypothesized, based on previous observations, that grassland productivity could be predicted from functional diversity and species diversity (Roscher et al., 2012, Tillman, 1996 Zhang et al., 2017; Korrell et al., 2024). Previous research that found strong correlations between functional diversity and productivity has been performed in manipulated grassland systems where efforts are exerted to control for all variations in management and abiotic conditions (Roscher et al., 2012). Such experiments may thus not reflect the effect of diversity on (semi-)natural ecosystems where other variables differ considerably. In contrast to experimental grasslands, diversity-productivity relationship for (semi-)natural grasslands can be unimodal or, with increasing management intensity, negative (Brun et al., 2019; Wang et al., 2022). The results from this study suggest that higher diversity (both taxonomic and functional) does not support greater productivity, and no unimodal relationships were found (Fig. 7, Fig. 8). Instead, the functional and species diversity-productivity relationships were overall negative. In contrast to the negative relationship between FDq and productivity shown here (Fig. 8), research on experimental grasslands has found a positive correlation between FDq and productivity (Roscher et al., 2012). Low FDq, along with low FEve, indicates a community where the dominant species are functionally similar. This suggests that strong niche partitioning had no beneficial effect on productivity, going against the “diversity hypothesis” (Tillman, 1996). FRic also showed a negative relationship with productivity (Fig. 8). Similar to FDq and FEve, a low FRic indicates trait convergence with very little differences in functional traits between the most dissimilar species in a community. Both functional and species diversity varied in their relationship with productivity in the different grassland habitat types (Fig. 7, Fig. 8). Other studies that have analysed manipulated grassland ecosystems have found the diversity-productivity relationships to shift along gradients of management intensity (Nie et al., 2024) and soil nutrients (Guo et al., 2023). The communities with the lowest species diversity (GS3 communities) showed the most strongly negative correlation between diversity and productivity (Fig. 7, Fig 9). These grasslands were species poor and, based on their significantly higher CWM-LDMC (Fig. 1) and stress tolerance scores (Fig. 3), plants were adapted to more stressful environments. This goes against previous studies that suggest that, in lower-resource environments, diversity has a stronger positive effect on productivity (Wang et al., 2022). The observed differences between habitats in the diversity-productivity relationship may also be due to differences in the most limiting resource in each habitat type. Some resources, such as water availability, positively influence species diversity, functional diversity and productivity (causing a positive correlation) (Ma et al., 2010) whereas other resources such as soil nutrient availability can cause decreases in species diversity and functional diversity while increasing productivity (causing a negative correlation) (Harpole et al., 2016; Zhang et al., 2023). This is in agreement with the high precipitation but low pH and low nitrogen content (unpublished) at the GS3 sites. The GS3 grasslands had a wide range of management intensities ranging from heavy grazing to near abandonment. Different management intensity could have resulted in the strongly negative diversity-productivity relationship in GS3, with more intense land use at the more productive GS3 sites resulting in decreased diversity, whereas under-management of less productive sites may have led to increased biodiversity (Pakeman et al., 2017; Andraczek et al., 2023). Conversely, overgrazing may have increased the diversity of some GS3 sites (Hallett et al., 2017) and simultaneously reduced their productivity. Grassland communities with stronger stress tolerance strategies are less productive Instead of being determined by functional diversity, functional traits of dominant species, such as plant height, may determine productivity (Xu et al., 2017; Zhang et al. 2018; Yang et al., 2023). CWM-SLA was also shown to drive productivity in dry grasslands in response to fertilization (Yang et al., 2023). However, we did not find a positive correlation of CWM-SLA or CWM-LA with productivity here (Supporting Information Fig. S3). The absence of a positive relationship between competitiveness (Fig. 9) and productivity was also unexpected. Instead, our work showed an overall positive relationship between R strategies and productivity (Fig. 9), but only for all habitats combined, and this was not reflected in significant correlations with CWM-SLA, indicating that the relationship between acquisitive strategies and productivity is weak. In contrast to Blaix et al. (2023), our research shows a negative relationship between CWM-LDMC and productivity (Fig. 8), in line with predictions by Lavorel and Garnier (2002). LDMC is indicative of a resource-conservative strategy (Garnier et al., 2001) and an indicator or stress tolerance strategy in the CSR classification (Pierce et al., 2013), which was also negatively correlated with productivity here (Fig. 9). The stronger predictive quality of LDMC compared to LA or SLA may be due to its lower intraspecific variation (Smart et al., 2017) and less impact of other factors such as light availability (Hodgson et al., 2011). Analysis of intraspecific variation in leaf traits in this study also found that LDMC was the least variable trait (Supporting Information Fig. S5). Conclusion Semi-natural grasslands are an important source of biodiversity as well as biomass for livestock farming. Our research shows a trade-off between these two ecosystem services in Irish semi-natural grasslands. We hypothesized that measurements of functional diversity would predict the productivity of semi-natural grassland ecosystems as the vegetation function should directly influence ecosystem services. It was also hypothesized that functional diversity would have a stronger predictive power on ecosystem productivity than species diversity. Instead, relationships between productivity and both species diversity and functional diversity were overall negative. The different grassland habitats differed significantly in functional diversity, functional strategy and also in their productivity-diversity relationships. Competitiveness or higher LA did not drive productivity; instead, productivity was negatively correlated with stress tolerance and LDMC. This may suggest that productivity is more strongly influenced by ecosystem characteristics (such as soil pH), in addition to management practices, than by diversity. REFERENCES Akhzari, D., Pessarakli, M. & Eftekhari Ahandani, S. (2015). Effects of grazing intensity on soil and vegetation properties in a Mediterranean rangeland. Communications in Soil Science and Plant Analysis , 46, 2798-2806. https://doi.org/10.1080/00103624.2015.1089272 Anderegg, W.R., Martinez-Vilalta, J., Mencuccini, M. & Poyatos, R. (2024). Community assembly influences plant trait economic spectra and functional trade-offs at ecosystem scales. Proceedings of the National Academy of Sciences , 121, e2404034121. https://doi.org/10.1073/pnas.2404034121 Andraczek, K., Weigelt, A., Bottero Cantuarias, C. J., Fischer, M., Hinderling, J., Prati, D., Rauwolf, E.M. N. & van der Plas, F. (2023). Relationships between species richness and biomass production are context dependent in grasslands differing in land‑use and seed addition. Scientific Reports , 13, 19663. https://doi.org/10.1038/s41598-023-47020-z Bengtsson, J., Bulllock, J.M., Egoh, B., Everson, C., Everson, T., O’Connor, T., O’Farrell, P. J. Smith, H.G., & Lindborg, R. (2019). Grasslands—more important for ecosystem services than you might think. Ecosphere , 10, e02582. https://doi.org/10.1002/ecs2.2582 Binder, S., Isbell, F., Polasky, S., Catford, J.A. & Tilman, D. (2018). Grassland biodiversity can pay. Proceedings of the National Academy of Sciences, 115(15), 3876-3881. https://doi.org/10.1073/pnas.1712874115 Biswas, S.R. & Mallik, A.U. (2010). Disturbance effects on species diversity and functional diversity in riparian and upland plant communities. Ecology 91, 28-35. https://doi.org/10.1890/08-0887.1 Blaix, C., Chabrerie, O., Alard, D., Catterou, M., Diquelou, S., Dutoit, T., Lacoux, J., Loucougaray, G., Michelot-Antalik, A., Pacé, M., Tardif, A., Lemauviel-Lavenant, S. & Bonis, A. (2023). Forage nutritive value shows synergies with plant diversity in a wide range of semi-natural grassland habitats. Agriculture Ecosystems & Environment , 347, 108369. https://doi.org/10.1016/j.agee.2023.108369 Botta‐Dukát, Z. (2005). Rao’s quadratic entropy as a measure of functional diversity based on multiple traits. Journal of Vegetation Science , 16, 533-540. https://doi.org/10.1111/j.1654-1103.2005.tb02393.x Brun, P., Zimmermann, N.E., Graham, C.H., Lavergne, S., Pellissier, L., Münkemüller, T. & Thuiller, W. (2019). The productivity-biodiversity relationship varies across diversity dimensions. Nature Communications , 10, 5691. https://doi.org/10.1038/s41467-019-13678-1 Bullock, J.M., Jefferson, R.G., Blackstock, T.H., Pakeman, R.J., Emmett, B.A., Pywell, R.J., Grime, J.P. & Silvertown, J. (2011). Semi-natural grasslands. Cambridge, UK, UNEP‐WCMC. In: Technical Report: The UK National Ecosystem Assessment, pp162‐195. Bumb, I., Garnier, E., Bastianelli, D., Richarte, J., Bonnal, L. & Kazakou, E. (2016). Influence of management regime and harvest date on the forage quality of rangelands plants: the importance of dry matter content. AoB Plants , 8, plw045. https://doi.org/10.1093/aobpla/plw045 Cadotte, M.W., Carscadden, K. & Mirotchnick, N. (2011). Beyond species: functional diversity and the maintenance of ecological processes and services. Journal of Applied Ecology , 48, 1079-1087. https://doi.org/10.1111/j.1365-2664.2011.02048.x Carmona, C.P., Azcárate, F.M., de Bello, F., Ollero, H. S., Leps, J. & Peco, B. (2012). Taxonomical and functional diversity turnover in Mediterranean grasslands: interactions between grazing, habitat type and rainfall. Journal of Applied Ecology , 49, 1084-1093. https://doi.org/10.1111/j.1365-2664.2012.02193.x Chytrý, M., Dražil, T., Hájek, M., Kalníková, V., Preislerová, Z., Šibík, J., Ujházy, K., Axmanová, I., Bernátová, D., Blanár, D., Dančák, M. & Dřevojan, P. (2015). The most species-rich plant communities in the Czech Republic and Slovakia (with new world records). Preslia , 87, 217-278. De Bello, F.D., Lepš, J., Lavorel, S. & Moretti, M. (2007). Importance of species abundance for assessment of trait composition: an example based on pollinator communities. Community Ecology , 8, 163-170. https://doi.org/10.1556/ComEc.8.2007.2.3 Díaz, S. & Cabido, M. (2001). Vive la différence: plant functional diversity matters to ecosystem processes. Trends in Ecology & Evolution , 16, 646-655. https://doi.org/10.1016/S0169-5347(01)02283-2 Fonseca, C.R. & Ganade, G. (2001). Species functional redundancy, random extinctions and the stability of ecosystems. Journal of Ecology , 89, 118-125. https://doi.org/10.1046/j.1365-2745.2001.00528.x Fossitt, J.A. (2000). A guide to habitats in Ireland. The Heritage Council of Ireland Series. ISBN 1901137279 Foy, C.D., 1984. Physiological effects of hydrogen, aluminum, and manganese toxicities in acid soil. Soil acidity and liming , 12 , pp.57-97. https://doi.org/10.2134/agronmonogr12.2ed.c2 Garnier, E., Laurent, G., Bellmann, A., Debain, S., Berthelier, P., Ducout, B., Roumet, C. & Navas, M.L. (2001). Consistency of species ranking based on functional leaf traits. New Phytologist 152, 69-83. https://doi.org/10.1046/j.0028-646x.2001.00239.x Gossner, M.M., Lewinsohn, T.M., Kahl, T., Grassein, F., Boch, S., Prati, D., Birkhofer, K., Renner, S.C., Sikorski, J., Wubet, T. & Arndt, H. (2016). Land-use intensification causes multitrophic homogenization of grassland communities. Nature , 540, 266-269. https://doi.org/10.1038/nature20575 Grime, J.P. (1977). Evidence for existence of three primary strategies in plants and its relevance to ecological and evolutionary theory. American Naturalist , 111, 1169–1194. Guo, X., Zuo, X., Medina-Roldán, E., Guo, A., Yue, P., Zhao, X., Qiao, J., Li, X., Chen, M., Wei, C. & Yang, T. (2023). Effects of multi-resource addition on grassland plant productivity and biodiversity along a resource gradient. Science of The Total Environment , 857 , 159367. https://doi.org/10.1016/j.scitotenv.2022.159367 Hallett, L.M., Stein, C. & Suding, K.N. (2017). Functional diversity increases ecological stability in a grazed grassland. Oecologia , 183, 831-840. https://doi.org/10.1007/s00442-016-3802-3 Harpole, W.S., Sullivan, L.L., Lind, E.M., Firn, J., Adler, P.B., Borer, E.T., Chase, J., Fay, P.A., Hautier, Y., Hillebrand, H. & MacDougall, A.S. (2016). Addition of multiple limiting resources reduces grassland diversity. Nature , 537 ), 93-96. https://doi.org/10.1038/nature19324 Hemathilake, D.M.K.S. & Gunathilake, D.M.C.C. (2022). Agricultural productivity and food supply to meet increased demands. In Future foods (pp. 539-553). Academic Press. https://doi.org/10.1016/B978-0-323-91001-9.00016-5 Hodgson, J.G., Montserrat-Marti, G., Charles, M. et al. (2011) Is leaf dry matter content a better predictor of soil fertility than specific leaf area? Annals of Botany , 108, 1337–1345. https://doi.org/10.1093/aob/mcr225 Kemmitt, S.J., Wright, D., Goulding, K.W. & Jones, D.L. (2006). pH regulation of carbon and nitrogen dynamics in two agricultural soils. Soil Biology and Biochemistry , 38, 898-911. https://doi.org/10.1016/j.soilbio.2005.08.006 Korell, L., Andrzejak, M., Berger, S., Durka, W., Haider, S., Hensen, I., Herion, Y., Höfner, J., Kindermann, L., Klotz, S. & Knight, T.M. (2024). Land use modulates resistance of grasslands against future climate and inter‐annual climate variability in a large field experiment. Global Change Biology 30, e17418. https://doi.org/10.1111/gcb.17418 Laliberté, E. & Legendre, P. (2010). A distance‐based framework for measuring functional diversity from multiple traits. Ecology , 91, 299-305. https://doi.org/10.1890/08-2244.1 Lavorel, S. & Garnier, E. (2002) Predicting changes in community composition and ecosystem functioning from plant traits: revisiting the Holy Grail. Functional Ecology , 16, 545-556. https://doi.org/10.1046/j.1365-2435.2002.00664.x Le Bagousse‐Pinguet, Y., de Bello, F., Vandewalle, M., Leps, J. & Sykes, M.T. (2014). Species richness of limestone grasslands increases with trait overlap: evidence from within‐and between‐species functional diversity partitioning. Journal of Ecology , 102, 466-474. https://doi.org/10.1111/1365-2745.12201 Lemaire, G. (2012). Intensification of animal production from grassland and ecosystem services: a trade-off. CABI Reviews , 1-7. https://doi.org/10.1079/PAVSNNR20127012 Li, X.Q., Li, Y., Shen, H., Li, S., Zhao, Z.Z., Xiao, J.N., Zhang, R., Shi, H., Zuo, H., Tu, D.J., Chen, G.M., Zhou, X.L. & Dong, S. K. (2024). Different responses of individuals, functional groups and plant communities in CSR strategies to nitrogen deposition in high-altitude grasslands. Science of the Total Environment , 953, 176051. https://doi.org/10.1016/j.scitotenv.2024.176051 Ma, W., He, J.S., Yang, Y., Wang, X., Liang, C., Anwar, M., Zeng, H., Fang, J. & Schmid, B. (2010). Environmental factors covary with plant diversity–productivity relationships among Chinese grassland sites. Global Ecology and Biogeography , 19 , 233-243. https://doi.org/10.1111/j.1466-8238.2009.00508.x Mason, N.W., de Bello, F., Doležal, J. & Lepš, J. (2011). Niche overlap reveals the effects of competition, disturbance and contrasting assembly processes in experimental grassland communities. Journal of Ecology , 99, 788-796. https://doi.org/10.1111/j.1365-2745.2011.01801.x Mayfield, M.M., Bonser, S.P., Morgan, J.W., Aubin, I., McNamara, S. & Vesk, P.A. (2010). What does species richness tell us about functional trait diversity? Predictions and evidence for responses of species and functional trait diversity to land‐use change. Global Ecology and Biogeography 19, 423-431. https://doi.org/10.1111/j.1466-8238.2010.00532.x Moog, D., Kahmen, S. & Poschlod, P. (2005). Application of CSR-and LHS-strategies for the distinction of differently managed grasslands. Basic and Applied Ecology 6, 133-143. https://doi.org/10.1016/j.baae.2005.01.005 Nie, Y., Xu, L., Xin, X. & Ye, L. (2024). Long-term grassland diversity-productivity relationship regulated by management regimes in northern China. Science of The Total Environment , 949, 175084. https://doi.org/10.1016/j.scitotenv.2024.175084 O’Neill, F.H., Martin, J.R., Devaney, F.M. & Perrin, P.M. (2013). The Irish semi-natural grasslands survey 2007-2012. Irish Wildlife Manuals, No. 78. National Parks and Wildlife Service, Department of Arts, Heritage and the Gaeltacht, Ireland. Pakeman, R.J., Hewison, R.L., Riach, D., Fisher, J.M., Hurskainen, S., Fielding, D.A. & Mitchell, R.J. (2017). Long-term functional structure and functional diversity changes in Scottish grasslands. Agriculture Ecosystems & Environment , 247, 352-362. https://doi.org/10.1016/j.agee.2017.06.033 Petchey, O.L., Hector, A. & Gaston, K.J. (2004). How do different measures of functional diversity perform? Ecology , 85, 847-857. https://doi.org/10.1890/03-0226 Pierce, S., Brusa, G., Vagge I. & Cerabolini, B.E.L. (2013). Allocating CSR plant functional types: the use of leaf economics and size traits to classify woody and herbaceous vascular plants. Functional Ecology , 27, 1002–1010. https://doi.org/10.1111/1365-2435.12095 Pierce, S., Negreiros, D., Cerabolini, B.E., Kattge, J., Díaz, S., Kleyer, M., Shipley, B., Wright, S.J., Soudzilovskaia, N.A., Onipchenko, V.G. & Van Bodegom, P.M. (2017). A global method for calculating plant CSR ecological strategies applied across biomes world‐wide. Functional Ecology, 31, 444-457. https://doi.org/10.1111/1365-2435.12722 Polley, H.W., Isbell, F.I. & Wilsey, B.J. (2013). Plant functional traits improve diversity‐based predictions of temporal stability of grassland productivity. Oikos , 122, 1275-1282. https://doi.org/10.1111/j.1600-0706.2013.00338.x Rolo, V., Rivest, D., Lorente, M., Kattge, J. & Moreno, G. (2016). Taxonomic and functional diversity in Mediterranean pastures: insights on the biodiversity–productivity trade‐off. Journal of Applied Ecology , 53, 1575-1584. https://doi.org/10.1111/1365-2664.12685.or Roscher, C., Schumacher, J., Gubsch, M., Lipowsky, A., Weigelt, A., Buchmann, N., Schmid, B. & Schulze, E.D. (2012). Using plant functional traits to explain diversity–productivity relationships. PloS One , 7, e36760. https://doi.org/10.1371/journal.pone.0036760 Schippers, P., Ozinga, W.A. & Pouwels, R. (2022). Factors affecting functional diversity of grassland vegetations. Ecological Modelling , 472, 110078. https://doi.org/10.1016/j.ecolmodel.2022.110078 Searchinger, T.D., James, O., Dumas, P., Kastner, T. & Wirsenius, S. (2022). EU climate plan boosts bioenergy but sacrifices carbon storage and biodiversity. Nature 612, 27-30. https://doi.org/10.1038/d41586-022-04133-1 Smart, S.M., Glanville, H.C., Blanes, M.D.C., Mercado, L.M., Emmett, B.A., Jones, D.L., Cosby, B.J., Marrs, R.H., Butler, A., Marshall, M.R. & Reinsch, S. (2017). Leaf dry matter content is better at predicting above‐ground net primary production than specific leaf area. Functional Ecology , 31, 1336-1344. https://doi.org/10.1111/1365-2435.12832 Spehn, E.M., Joshi, J., Schmid, B., Alphei, J. & Körner, C. (2000). Plant diversity effects on soil heterotrophic activity in experimental grassland ecosystems. Plant and Soil , 224, 217-230. https://doi.org/10.1023/A:1004891807664 Stevens, C.J., Duprè, C., Dorland, E., Gaudnik, C., Gowing, D.J., Bleeker, A., Diekmann, M., Alard, D., Bobbink, R., Fowler, D. & Corcket, E. (2010). Nitrogen deposition threatens species richness of grasslands across Europe. Environmental Pollution , 158, 2940-2945. https://doi.org/10.1016/j.envpol.2010.06.006 Tilman, D., Wedin, D. & Knops, J. (1996). Productivity and sustainability influenced by biodiversity in grassland ecosystems. Nature , 379, 718-720. https://doi.org/10.1038/379718a0 Tilman, D., Reich, P.B., Knops, J., Wedin, D., Mielke, T. & Lehman, C. (2001). Diversity and productivity in a long-term grassland experiment. Science 294, 843-845 https://doi.org/10.1126/science.1060391 Večeřa, M. et al. (2023). Decoupled phylogenetic and functional diversity in European grasslands. Preslia , 95, 413-445. https://doi.org/10.23855/preslia.2023.413 Villéger, S., Mason, N.W. & Mouillot, D. (2008). New multidimensional functional diversity indices for a multifaceted framework in functional ecology. Ecology 89, 2290-2301. https://doi.org/10.1890/07-1206.1 Wang, Y., Du, J., Pang, Z., Liu, Y., Xue, K., Hautier, Y., Zhang, B., Tang, L., Jiang, L., Ji, B. & Xu, X. (2022). Unimodal productivity–biodiversity relationship along the gradient of multidimensional resources across Chinese grasslands. National Science Review , 9 , nwac165. https://doi.org/10.1093/nsr/nwac165 Wilson, J.B, Peet, R.K., Dengler, J. & Pärtel, M. (2012). Plant species richness: the world records. Journal of Vegetation Science , 23, 796–802. https://doi.org/10.1111/j.1654-1103.2012.01400.x Wingler, A. & Sandel, B. (2023) Relationships of the competitor, stress tolerator, ruderal functional strategies of grass species with lifespan, photosynthetic type, naturalization and climate. AoB Plants 15, plad021. https://doi.org/10.1093/aobpla/plad021 Wirth, S.B., Poyda, A., Taube, F., Tietjen, B., Müller, C., Thonicke K., Linstädter, A., Behn, K., Schaphoff, S., von Bloh, W., & Rolinski, S. (2024). Connecting competitor, stress-tolerator and ruderal (CSR) theory and Lund Potsdam Jena managed Land 5 (LPJmL 5) to assess the role of environmental conditions, management and functional diversity for grassland ecosystem functions. Biogeosciences , 21, 381-410. https://doi.org/10.5194/bg-21-381-2024 Wright, I.J., Reich, P.B., Westoby, M., Ackerly, D.D., Baruch, Z., Bongers, F., Cavender-Bares, J., Chapin, T., Cornelissen, J.H., Diemer, M. & Flexas, J. (2004). The worldwide leaf economics spectrum. Nature , 428, 821-827. https://doi.org/10.1038/nature02403 Xu, Z., Li, M.H., Zimmermann, N.E., Li, S.P., Li, H., Ren, H., Sun, H., Han, X., Jiang, Y. & Jiang, L. (2018). Plant functional diversity modulates global environmental change effects on grassland productivity. Journal of Ecology , 106, 1941-1951. https://doi.org/10.1111/1365-2745.12951 Yang, Y.T., Chen, Z.F., Xu, B.C., Ghanizadeh, H., Li, W., Ding, C.Q., Zhou, R.L. & Wen, Z.M. (2023). Contrasting patterns of community-weighted mean traits and functional diversity in driving grassland productivity changes under N and P addition. Frontiers in Plant Science ,14, 1145709. https://doi.org/10.3389/fpls.2023.1145709 Yu, J., Hou, G., Zhou, T., Shi, P., Zong, N. & Sun, J. (2022). Variation of plant CSR strategies across a precipitation gradient in the alpine grasslands on the northern Tibet Plateau. Science of the Total Environment , 838, 156512. https://doi.org/10.1016/j.scitotenv.2022.156512 Zhang, Q., Buyantuev, A., Li, F.Y., Jiang, L., Niu, J.M., Ding, Y., Kang, S.& Ma, W.J. (2017). Functional dominance rather than taxonomic diversity and functional diversity mainly affects community aboveground biomass in the Inner Mongolia grassland. Ecology and Evolution , 7, 1605-1615. https://doi.org/10.1002/ece3.2778 Zhang, M., Li, G., Wang, Y., Pan, D., Sun, J. & Wang, L. (2023). Land use intensification alters the relative contributions of plant functional diversity and soil properties on grassland productivity. Oecologia, 201, 119-127. https://doi.org/10.1007/s00442-022-05288-4 Zhang, P. et al. (2025). Dominant species predict plant richness and biomass in global grasslands. Nature Ecology and Evolution . https://doi.org/10.1038/s41559-025-02701-y Zhao, Y., Liu, Z. & Wu, J. (2020) Grassland ecosystem services: a systematic review of research advances and future directions Landscape Ecology , 35, 793-814. https://doi.org/10.1007/s10980-020-00980-3 Zhou, T., Hou, G., Sun, J., Zong, N. and Shi, P., 2021. Degradation shifts plant communities from S-to R- strategy in an alpine meadow, Tibetan Plateau. Science of the Total Environment , 800, 149572.https://doi.org/10.1016/j.scitotenv.2021.149572 Supporting information Table S1. Location and climatic details for the study sites. The climatic details were taken from Met Éireann monthly temperature and rainfall grids (https://www.met.ie/monthly-rainfall-and-temperature-grids/) at a resolution of 1km 2 . Values are annual rainfall and annual averages of monthly temperatures for 2024. Figure S1. Relationship between functional evenness and species diversity. Relationship between FEve and species richness (a), Simpson diversity (b) and Simpson evenness (c). The overall regression between the two variables is indicated by the black line and the regression based on habitat type is indicated by coloured lines (green for GS1, red for GS3 and blue for GS4). The Statistics are derived from Spearman-Rank tests with the same colour associations as the linear model lines. Data points, representing individual relevés, are also coloured based on habitat. Figure S2. Relationship between Rao’s functional diversity and species diversity. Relationship between FDq and species richness (a), Simpson diversity (b) and Simpson evenness (c). The overall regression between the two variables is indicated by the black line and the regression based on habitat type is indicated by coloured lines (green for GS1, red for GS3 and blue for GS4). The Statistics are derived from Spearman-Rank tests with the same colour associations as the linear model lines. Data points, representing individual relevés, are also coloured based on habitat. Figure S3. Relationship between community weighted mean leaf traits and grassland productivity. Relationship between grassland yield (NDVI) and CWM-LA (a), CWM-SLA (b). The overall linear regression between the two variables is indicated by the black line and the regression based on habitat type is indicated by coloured lines (green for GS1, red for GS3 and blue for GS4). The statistics are derived from Spearman-Rank tests. Data points present individual relevés also coloured based on habitat. Figure S4. Differences functional redundancy (FRed) between habitats and sites. The Fossitt habitat is indicated by the colouration of boxplots (GS1 = green, GS3 = red, GS4 = blue). The length of the box indicates the interquartile range. The band in the middle of the box represents the median value. The whiskers either side of each box indicate minimum and maximum values with values beyond these points considered outliers. Coloured dots represent data points (these points are scattered for visual purposes). The significance of difference between categorical variables (Kruskal-Wallis) is indicated by the number of red asterisks in the top left corner (p >0.05, <0.05*, significance of difference between individual habitats (Wilcoxon test) is indicated by the black asterisks above the brackets linking each habitat type. Figure S5. Intraspecific variation in leaf area, specific leaf area and leaf dry matter content. Intraspecific variation in leaf area (a), specific leaf area (b) and leaf dry matter content (c). Intraspecific variation is measured by the coefficient of variation (CV) which provides a standardized value for variation within a species across relevés. The ten species present are the species with highest CV values and the eleventh grey value is the average CV value among all species. Species are colour-coded green for graminoid species and yellow for forb species. Supplementary Material File (halpin et al. supporting information.docx) Download 1.09 MB Information & Authors Information Version history V1 Version 1 13 January 2026 Copyright This work is licensed under a Non Exclusive No Reuse License. Keywords community ecology comparative ecological experiment plants terrestrial Authors Affiliations Eoin Halpin 0009-0006-0819-1261 [email protected] University College Cork View all articles by this author Oliver Lynch Milner University of Galway View all articles by this author Samuel Hayes 0000-0002-2710-7305 University College Cork View all articles by this author Karen Bacon University of Galway View all articles by this author Fiona Cawkwell 0000-0002-7365-8909 University College Cork View all articles by this author Astrid Wingler University College Cork View all articles by this author Metrics & Citations Metrics Article Usage 245 views 132 downloads .FvxKWukQNSOunydq8rnd { width: 100px; } Citations Download citation Eoin Halpin, Oliver Lynch Milner, Samuel Hayes, et al. Relationships between plant functional diversity, species diversity and productivity depend on semi-natural grassland habitat type. Authorea . 13 January 2026. DOI: https://doi.org/10.22541/au.176830949.99236894/v1 If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download. For more information or tips please see 'Downloading to a citation manager' in the Help menu . Format Please select one from the list RIS (ProCite, Reference Manager) EndNote BibTex Medlars RefWorks Direct import Tips for downloading citations document.getElementById('citMgrHelpLink').addEventListener('click', function() { popupHelp(this.href); return false; }); $(".js__slcInclude").on("change", function(e){ if ($(this).val() == 'refworks') $('#direct').prop("checked", false); $('#direct').prop("disabled", ($(this).val() == 'refworks')); }); View Options View options PDF View PDF Figures Tables Media Share Share Share article link Copy Link Copied! Copying failed. Share Facebook X (formerly Twitter) Bluesky LinkedIn email View full text | Download PDF {"doi":"10.22541/au.176830949.99236894/v1","type":"Article"} Now Reading: Share Figures Tables Close figure viewer Back to article Figure title goes here Change zoom level Go to figure location within the article Download figure Toggle share panel Toggle share panel Share Toggle information panel Toggle information panel Go to previous graphic Go to next graphic Go to previous table Go to next table All figures All tables View all material View all material xrefBack.goTo xrefBack.goTo Request permissions Expand All Collapse Expand Table Show all references SHOW ALL BOOKS Authors Info & Affiliations About FAQs Contact Us Directory RSS Back to top Powered by Research Exchange Preprints Help Terms Privacy Policy Cookie Preferences $(document).ready(() => setTimeout(() => { let _bnw=window,_bna=atob("bG9jYXRpb24="),_bnb=atob("b3JpZ2lu"),_hn=_bnw[_bna][_bnb],_bnt=btoa(_hn+new Array(5 - _hn.length % 4).join(" ")); $.get("/resource/lodash?t="+_bnt); },4000)); (function(){function c(){var b=a.contentDocument||a.contentWindow.document;if(b){var d=b.createElement('script');d.innerHTML="window.__CF$cv$params={r:'9fe1c6104ff509d6',t:'MTc3OTE3OTc4Mw=='};var a=document.createElement('script');a.src='/cdn-cgi/challenge-platform/scripts/jsd/main.js';document.getElementsByTagName('head')[0].appendChild(a);";b.getElementsByTagName('head')[0].appendChild(d)}}if(document.body){var a=document.createElement('iframe');a.height=1;a.width=1;a.style.position='absolute';a.style.top=0;a.style.left=0;a.style.border='none';a.style.visibility='hidden';document.body.appendChild(a);if('loading'!==document.readyState)c();else if(window.addEventListener)document.addEventListener('DOMContentLoaded',c);else{var e=document.onreadystatechange||function(){};document.onreadystatechange=function(b){e(b);'loading'!==document.readyState&&(document.onreadystatechange=e,c())}}}})();