GrassPlot is the EDGG-affiliated database of multi-scale plant diversity in Palaearctic grasslands. The database started as a repository for the data collected at the Research Expeditions/Field Workshops of the Eurasian Dry Grassland Group (EDGG) and similar multi-scale sampling schemes. It formerly was named "Database Species-Area Relationships in Palaearctic Grasslands". GrassPlot is registered under the code EU-00-003 in the Global Index of Vegetation-Plot Databases (GIVD;

Species rich pasture in Val di susa, N Italy. FN
Sandy grassland - R.P.

Left: A species rich pasture in Val di Susa, Northern Italy (photo: F. Napoleone). Middle: Nested-plot sampling (EDGG Biodiversity Plot) in a sandy dry grassland in NE Poland (photo: J. Dengler). Right: Flower rich sandy grassland at the roadside in western Poland (photo: R. Pielech).

GrassPlot is looking for high-quality phytodiversity data sampled on plots of the following standard areas: 0.0001 m², 0.001 or 0.0009 m², 0.01 m², 0.1 or 0.09 m², 1 m², 10 or 9 m², 100 m², and 1000 or 900 or 1024 m².

We preferentially include nested-plot multi-scale data, but we also welcome data for single grain sizes, provided they were carefully sampled with the aim of complete species lists. We request that plots have been precisely delimited in the field, usually with metal pins in the corners and a measuring tape on the perimeter, which typically is not the case for conventional phytosociological relevés. Nested-plot data with at least two different plot sizes are also accepted when plot sizes deviate from our standards.

Any type of grassland s.l. from the whole Palaearctic biogeographic realm (Europe, North Africa, West, Central and North Asia) are welcome, including dwarf-shrub communities, deserts and semi-deserts, rocks and screes, saline habitats, wetlands, dunes and ruderal communities. Aquatic and segetal vegetation are not included.

Data of vascular plants and/or terricolous non-vascular plants (bryophytes, lichens and macro-algae) can be provided. Although richness counts per plot (together with metadata, such as plot size, coordinates, grassland type) are sufficient, we encourage to provide even more valuable data with species composition and potentially cover and selected environmental data. For more details on our requirements, see here.


Left: Thorn-cushion steppe in Armenia. Middle: Fen vegetation in the Biebrza National Park, NE Poland. Right: Dune vegetation in Southern Sicily, Italy. Photos by  J. Dengler

GrassPlot is a highly selective database for specific purposes that complements the existing "all-purpose" supranational databases European Vegetation Archive (EVA) and "sPlot", which we are collaborating with. 

More details about the history and content of GrassPlot can be found here.


Join the GrassPlot Consortium

If you have data that meet the criteria mentioned above and agree with the GrassPlot Bylaws, please contact Idoia Biurrun. As member of the GrassPlot Consortium you will have the possibility to opt-in as co-author to various paper projects or to propose your own projects.

Foundation: 2010



How to use GrassPlot data

You are welcome to use data contained in the GrassPlot database for research purpose. You can apply for your paper project using the GrassPlot Paper proposal form.

Note that at least one consortium member needs to be involved in your project to be eligible. A list over all consortium members can be found in the GrassPlot database reports (Dengler et al. 2018; Biurrun et al. 2019).

Content of the GrassPlot database (state: 1 October 2020):

  • 225 datasets
  • 309 data owners
  • 49 countries
  • 202,579 plots of different grain sizes across 32,105 independent plots, among them 6,623 with data also for non-vascular plants
  • 25,441 individual plots and 6,664 nested-plot series including at least two grain sizes and 5,431 with at least four grain sizes
  • 2,904 0.0001-m2 plots; 4,341 0.001 (or 0.0009)-m2 plots; 70,548 0.01-m2 plots; 5,756 0.1 (or 0.09)-m2 plots; 25,432 1-m2 plots; 11,035 10 (or 9)-m2 plots; 6,321 100-m2 plots; 187 1,000 (or 900 or 1,024)-m2 plots; 76,055 non-standard plot sizes 


Distribution of independent plots across vegetation types

Natural grasslands: Alpine grasslands (14.1%), Xeric grasslands and steppes (8.5%), Rocky grasslands (3.3%), Alpine steppes (0.4%).

Secondary grasslands: Mesic grasslands (11.7%), Meso-xeric grasslands (16.3%), Sandy dry grasslands (4.8%), Wet grasslands (4.5%), Mediterranean grasslands (2.9%). 

Azonal habitats: Saline communities (9.3%), Wetlands (8.7%), Dunes (3.0%), Saline steppes and semi-deserts (1.2%), Rocks and screes (1.2%).

Dwarf-shrublands: Arctic-alpine heathlands (1.6%), Garrigues and Thorn cushion communities (1.7%), Lowland heathlands (0.4%).

Tall forb and ruderal communities: Ruderal communities (1.5%), Tall forb and fringe communities (1.0%).

Deserts and semi-deserts: Cold deserts and semi-deserts (2.1%), Warm deserts and semi-deserts (0.1%), Alpine deserts (<0.1%).

Not assigned (1.8%).

Distribution of independent plots across phytosociological classes

Festuco-Brometea (22.9%), Molinio-Arrhenatheretea (11.0%), Juncetea maritimi (6.4%), Scheuchzerio palustris-Caricetea fuscae (6.2%), Juncetea trifidi (4.9%), Koelerio-Corynephoretea canescentis (4.3%), Stipo-Trachynietea distachyae (1.9%), Helichryso-Crucianelletea maritimae (1.5%), Ammophiletea (1.3%), Elyno-Seslerietea (1.2%), Festuco-Puccinellietea (1.2%), Phragmito-Magnocaricetea (1.2%), Nardetea strictae (0.9%), Oxycocco-Sphagnetea (0.9%), Festuco hystricis-Ononidetea striatae (0.8%), Festucetea indigestae (0.7%), Cleistogenetea squarrosae (0.7%), Spartinetea maritimae (0.7%), Epilobietea angustifolii (0.5%), Lygeo sparti-Stipetea tenacissimae (0.5%), Therosalicornietea (0.5%), Sedo-Scleranthetea (0.4%), .

Other classes (3.6%).

Not assigned (25.7%).


More details about the history and content of GrassPlot can be found here.


Figure: Spatial distribution of independent plots (upper) and nested plots (at least four grain sizes; lower) shown as plot density in equally-sized grid cells of 10,000 km2. From: Biurrun et al. 2019, Palaearctic Grasslands 44, 26-47



Responsibilities within GrassPlot:

Outreach (including EVA and sPlot)


Coordination of opt-in authorships







IDOIA BIURRUN (Deputy Custodian)


Responsibilities within GrassPlot:

Database manager, specifically for meta-, header and richness data

Contact person for potential data providers



Department of Plant Biology and Ecology
University of the Basque Country
Apdo. 644
E-48080 Bilbao



Idoia Biurrun



Responsibilities within GrassPlot:

Preparation and harmonization of compositional data



P.le Aldo Moro 5
00185 Rome




Sabina Burrascano



Responsibilities within GrassPlot:

GrassPlot Diversity Benchmarks

Handling of EDGG Field Workshop data



Institute of Environmental Biology, Faculty of Biology, 
University of Warsaw, ul. Żwirki i Wigury 101, 
02-089, Warsaw



Iwona Dembicz



Responsibilities within GrassPlot:

Syntaxonomic assignment of plots





R. Guarino



Responsibilities within GrassPlot:

Management of GrassPlot webpage



NIBIO - Norwegian Institute of Bioeconomy Research
Holtvegen 66
9016 Tromsø




J. Kapfer



Responsibilities within GrassPlot:

Management of the database structure

Online platform of GrassPlot data (GrassPlot Diversity Explorer)



Department of Forest Biodiversity, Faculty of Forestry
University of Agriculture in Kraków



Remigiusz Pielech


Former members of the Governing Board 2017-2019 

Viktoria Wagner (CA), Timo Conradi (DE), Alireza Naqinezhad (IR)

Opt-out project #02:  Benchmarking plant diversity of Palaearctic grasslands

Start: 18 January 2019

Project leader(s): Idoia Biurrun (ES)

Summary: Dengler et al. (2016, Bulletin of the Eurasian Dry Grassland Group 31: 12-26) published a first comparative overview of plant diversity in Palaearctic grasslands; they provided mean, maximum and minimum richness values for seven grain sizes, from 1 cm2 to 100 m2 across 20 datasets. In the meanwhile, we have developed the GrassPlot database, with more than 201,000 plots, and thus we are ready to provide benchmarking richness data for these grasslands. We will calculate several stats for our eight standard grain sizes, from 1 cm2 to 1000 m2, and for the complete vegetation as well as for vascular plants, bryophytes and lichens. These richness data will be compared among vegetation types, biomes and biogeographic regions.


Opt-in project #03: How do environmental factors shape the diversity of vascular plants, bryophytes and lichens in Palaearctic grasslands at multiple scales?

Start: 11 February 2020

Project leader(s): Iwona Dembicz (PL) & Jürgen Dengler (CH)

Summary: Understanding the factors governing alpha diversity patterns is a central theme in ecology and biogeography. However, results from regional studies are often not consistent, which can be attributed to idiosyncrasies of the regions and to the fact that different studies analysed different grain sizes as their alpha diversity, while a strong scale dependence of drivers of alpha diversity is generally acknowledged. Here we want to derive a general model that describes alpha-diversity in Palaearctic grasslands s.l. We aim to overcome the idiosyncrasies of previous studies by (a) combining many different regions and grassland types and (b) conducting our analyses with nested-plot data for different grain sizes simultaneously. Moreover, compared to previous regional studies with member datasets of GrassPlot (e.g. Löbel et al. 2006, Turtureanu et al. 2014, Kuzemko et al. 2016, Polyakova et al. 2016), we will include not only the local environmental variables (topography, soil, land use) and climate, but also parameters related to landscape configuration (amount of grassland in the surrounding, heterogeneity of the landscape), history (human impact, glaciations,…) and possibly species pool. Our results will be beneficial both from the point of view of basic ecological knowledge (advancing the understanding how drivers of biodiversity interact across spatial scales) and for practical conservation (e.g., identification of plant biodiversity hotspots; identification of human-related disturbance levels most benefiting biodiversity). Among others, we hope that our resulting model will be able to explain why a few small regions of the Palaearctic grasslands have such an extraordinary small-scale richness (“world record holders”) compared to other similar grasslands in other regions, and possibly predict other, not yet known such hotspots (which then could be sampled).

In the analyses, we will use nested-plot series with richness data of vascular plants, bryophytes and lichens. The richness data of each selected grain size will be used as response variables and a wide range of topographic, climatic, edaphic, anthropogenic and historical parameters (taken from GrassPlot and from other databases) as explanatory variables in our models (e.g. generalized linear mixed models and/or boosted regression trees). The importance of particular predictors will be assessed with multimodel inference or other similar approach. Details of the modelling approach (e.g. how to account for the spatial and ecological clustering at multiple scales) will be discussed with the approved co-authors.


Opt-in papers #04B & #04B.2: Patterns and drivers of small-scale beta-diversity in Palaearctic grasslands

Start: 20 January 2019

Project leader(s): Jürgen Dengler (CH/DE) & Iwona Dembicz (PL/CH)

Summary: In Dengler et al. 2019, we demonstrated that in Palaearctic grasslands the power function is generally by far the best model to describe species-area relationships (SARs) in continuous vegetation at grain sizes up to 1000 m². Based on this finding and the general notion that the exponent of the power function SARs (z-value) is an appropriate measure of standardized multiplicative beta-diversity, we conducted an extensive study on the relationship of z-values to a wide array of factors based on more than 4500 nested plot series.

Status: One paper dealing with various ecological predictors at biogeographic and site scale has been submitted to a macroecological journal in July 2020 where it is now under evaluation (#04B). A follow-up paper is currently in preparation that will analyse how z-values vary between and within biomes, broad and narrow vegetation types (#04B.2).


Opt-in paper #04C: How do environmental factors shape the scale dependence of local z-values in Palaearctic grasslands?

Start: 11 February 2020

Project leader(s): Jinghui Zhang (CH & CN) & Jürgen Dengler (CH)

Summary: Background: Species–area relationships (SAR) are fundamental in the understanding of biodiversity patterns. Dengler et al. (2019) found that the “normal” power function was the most suitable function across the taxonomic groups spanning a wide range of grassland types throughout the Palaearctic, meaning that the exponent z (= slope in the double-log representation) of the power function is relatively constant across spatial scales. However, when having a closer look, some studies found significant changes of the z-values across spatial scales (e.g. Crawley & Harral 2001; Turtureanu et al. 2014; Polyakova et al. 2016), but the position of the peaks (steepest slopes) was not consistent, while in other regional studies no scale-dependence occurred (Kuzemko et al. 2016). Aims: To answer the following two questions: 1) Is there a general pattern of scale dependence of z-values in Palaearctic grasslands? 2) Can different relationships of “local” z-values to grain sizes (no relationship, i.e. constant z; increasing; decreasing, unimodal with different peaks; u-shaped with different peaks) be explained by environmental factors, e.g. disturbance regimes or measures of small-scale heterogeneity? Methods outline: Our target variable are the so-called “local z-values” (i.e. local derivative of SARs in log-log representation between two subsequent grain sizes of a nested-plot series; any better name than “local z-values” is welcome). We will model their relationship to the scale, for which we use. If Ai and Si are the grain size and the species richness of a particular grain size, the local z between grain sizes A1 and A2 is simply defined as:

local z = (log S2 – log S1) / (log A2 – log A1)

log (local grain size) = (log A2 + log A1) / 2

We intend to assign each local z value to the mean of the logarithms of the two grain sizes (= logarithm of the geometric mean). Doing so, local z-values are independent on how grain sizes increase in a particular dataset (in EDGG Field Workshops it is always a 10fold increase, but not necessarily in other datasets).

We will calculate local z-values for any grain size transition in any nested-plot series for the three taxonomic groups (vasc. plants, bryophytes, lichens), provided both S-values are >0 (otherwise local z-values are not defined). Then we will relate the scale-dependence of the local z-values to a wide range of potential predictors (topographic, climatic, edaphic, anthropogenic and historical parameters, but particularly measures of disturbance and heterogeneity) and their interactions.


Opt-in paper #15: Environmental drivers and spatial scaling of species abundance distributions in Palaearctic grassland vegetation

Start: 19 December 2020

Project leader(s): Werner Ulrich

Summary: We ask 1) whether and how species abundance distributions in temperate grasslands change in shape and parameter values across spatial scales, 2) whether these changes follow predictable trends, 3) whether scaling regions in SAD shapes exist, and 4) whether and how SAD parameters and scaling can be attributed to respective gradients in community size, important species traits, and environmental factors?

We will fit important SAD models (lognormal, log-series, Weibull) to the cover data. We will also rely on model independent parameters of SADs, the variance as a measure of the range in abundance, the skewness as a measure of an excess of relatively rare or abundant species, and the kurtosis that quantifies the proportion of species with relatively intermediate abundances. Particularly, the latter parameter has gained too few attention.


Opt-in paper #16: Components of beta-diversity across different sampling grains in Eurasian grasslands

Start: 11 February 2019

Project leader(s): Sabina Burrascano (IT) & Salza Palpurina (BG)

Summary: Sampling grain may strongly influence patterns of beta diversity. In general, it can be hypothesized that by analyzing any given dataset encompassing different grains, higher beta-diversity will be found when considering small sampling units as compared to large ones. Small sampling units, in fact, contain fewer individuals and a smaller part of the local heterogeneity and species diversity than larger ones. It is now widely accepted that beta-diversity can be partitioned into two components deriving from two different ecological processes: species replacement (also called turnover) and richness difference or nestedness (species gain and loss). Species replacement refers to the fact that species replace each other along ecological gradients, with the replacement rate being a function of gradients’ length and of the species niche breadth. Richness difference refers to the fact that one community may include a larger number of species than another. It may reflect the quantity and heterogeneity of the available resources, or differences in the species pools across the sampled communities (Baselga, 2010). Most studies addressing patterns of beta-diversity and its components in grasslands have been based on single-grain sampling units. It is therefore not known to what extent these patterns are sensitive to the grain at which the sampling is performed. Especially, it is yet to be investigated to what extent the grain-related changes in patterns of beta-diversity are driven by species replacement or richness difference. The general hypothesis would be that replacement is more relevant than richness differences at small grains as compared to larger ones and vice versa. However, this general hypothesis was never properly tested and quantitatively assessed leaving a great degree of uncertainty in the study and the interpretation of the scale-dependency of beta-diversity patterns. The aim of this study will be to examine the effect of sampling grain on patterns of beta-diversity components in grasslands across the Eurasian continent. Since the spatial extent at which the sampling is performed is likely to strongly influence the variation of replacement and richness differences across grains, we will also analyse the distance decay of these two components of beta-diversity at different spatial grains. Also the degree of environmental and structural heterogeneity is expected to strongly influence beta-diversity patterns across different spatial scales, in relation to the dominance by different species and to their response to ecological drivers operating at different spatial scales (Tuomisto et al., 2017). For this reason, components of beta-diversity across spatial grains will be investigated within and across different phytosociological classes. We will compute both the multiple-site β-diversity (i.e. variation across multiple sites), and the beta-diversity based on pairwise dissimilarities following the framework of Legendre (2014) and analyse how these differ across seven spatial grains. Pairwise dissimilarity values obtained for different spatial grains will be modeled against spatial distances in order to investigate also the effect of spatial extent. Analyses will be run within and across major phytosociological classes.


Opt-in paper #17: RECALL – Revisiting CriticAL Loads of atmospheric nitrogen deposition

Start: 7 January 2020

Project leader(s): Tobias Ceulemans (BE), Maarten van Geel (BE) & Carly Stevens (UK)

Summary: In a recent European survey of semi-natural grasslands and in Atlantic European bogs, we found thatcommunities of arbuscular and ericoid mycorrhizal fungi showed the highest negative change at levels of atmospheric nitrogen deposition of 7 to 8 kg N ha-1 year-1 (Global Ecology and Biogeography Ceulemans et al. 2019, Ceulemans et al. in preparation). The results are remarkably consistent with the work of Sietse Van der Linde et al. (Nature 2018) in a European survey of ectomycorrhizal communities of a range of tree species that also showed negative change points at 5 to 6 kg N ha-1 year-1. Similarly, the late Richard Payne found that the critical loads for higher plant species in grasslands may be set too high (PNAS, Payne et al. 2013). These results call into question the current critical loads that are used in European environmental policy. Following the troubles in the Netherlands with wide-spread street protests of farmers against the nitrogen policy of the Kingdom, we feel it would be an excellent time to step in as scientists and revisit the critical loads with a critical eye. We already have access to datasets with hundreds of plots of bryophytes in European grasslands, mycorrhizae (orchid and arbuscular) in European grasslands (several types), heathlands and bogs (ericoid), European forests (ecto-), but we want to pull our resources. Next to additional bryophytes, higher plants and mycorrhizae; we aim to include lichens and arthropods such as butterflies. We then propose to analyse the data calculating average change points in taxon occurrence across an environmental gradient and integrates these at the community level to identify points in environmental change coinciding with maximal community change. This will allow to verify the critical loads with largescale empirical evidence across taxonomic groups and habitat types. The proposed procedure should lead to a straightforward paper to revisit (comfirm or reject) the environmental policy.


Opt-in paper #19: Biases in species richness data in large phytosociological databases

Start: 19 December 2020

Project leader(s): Remigiusz Pielech, Jürgen Dengler, Idoia Biurrun, Iwona Dembicz, Anna Kuzemko, Borja Jiménez-Alfaro, Florian Jansen

Summary: A recently submitted manuscript from the GrassPlot database (Biurrun et al. subm.: Benchmarking plant diversity of Palaearctic grasslands and other open habitats) found unexpectedly strong differences in vascular plant species richness between mean values from GrassPlot and the few published mean values from large EVA member databases, where for nearly all vegetation class x region x grain size combinations the GrassPlot data had much higher richness. Given the wide and increasing usage of EVA and GrassPlot in many research projects, we consider it important for users to be aware of potential biases (incomplete richness records, preferential recording of particular species-rich or species-poor stands) in both databases. This would allow users to take counter-measures, e.g. excluding some data or regions or applying correction factors. With this study we thus aim to determine how often there are significant differences in mean richness estimates derived from both databases, and if so in which direction and how strong. We would quantify these for all combinations of countries and vegetation classes that are represented in both databases, but possibly aggregate at a higher level (country group or group of related vegetation classes). While GrassPlot directly contains richness data for the standard grain sizes 1 m², 10 m² and 100 m², in the case of EVA generalised additive models (GAMs) would be modelled to get values for 10 m², 100 m² and, where appropriate, also for 1 m². To account for potential reasons of incomplete records, we possibly will carry out the comparisons also for subsets of both databases that are expected to have higher data quality (EVA: the more recent plots; GrassPlot: only the plots of the EDGG nested plot sampling). Moreover, to assess other potential factors that could lead to biased richness estimates at the country or country group level, we will compare spatial aggregation patterns in both databases
and possibly apply resampling to avoid undue impacts of oversampled regions. Specifically for GrassPlot data, where for each dataset the sampling strategy (random, systematic, preferential) is already available (but will be further refined during the project), we will test how strong the biasing effects are.

Opt-out papers #01A, #01B: “Database papers”

Start: March 2017

Project leader(s): Jürgen Dengler (CH) & Idoia Biurrun (ES)

Summary: The GrassPlot Consortium published an initial paper that the describes the scope and the content of the database in 2018. When there are major updates, we are publishing follow-ups of this database paper.


Dengler, J., Wagner, V., Dembicz, I., García-Mijangos, I., Naqinezhad, A., Boch, S., Chiarucci, A., Conradi, T., Filibeck, G., (…) & Biurrun, I. 2018. GrassPlot – a database of multi-scale plant diversity in Palaearctic grasslands. Phytocoenologia 48: 331–347. []

Biurrun, I., Burrascano, S., Dembicz, I., Guarino, R., Kapfer, J., Pielech, R., Garcia-Mijangos, I., Wagner, V., Palpurina, S., (…) & Dengler, J. 2019. GrassPlot v. 2.00 – first update on the database of multi-scale plant diversity in Palaearctic grasslands. Palaearctic Grasslands 44: 26-47. [pdf]

#1B figure


Opt-in paper #04A: “Species-area relationships”

Start: January 2019

Project leader: Jürgen Dengler (CH)

Summary: Patterns and drivers of small-scale beta-diversity in Palaearctic grasslands


Dengler, J., Matthews, T.J., Steinbauer, M.J., Wolfrum, S., Boch, S., Chiaruzzi, Al, Conradi, T., Dembicz, I., Marcenò, C., (…) & Biurrun, I. 2020. Species-area relationships in continuous vegetation: Evidence from Palaearctic grasslands. Journal of Biogeography 47: 72-86. [pdf] [supplements]

#4A figure

Thanks to funding from BayIntAn program of the Bavarian Research Alliance (BayFor) and co-funding from BayCEER, an international workshop on "Scale-dependent phytodiversity patterns in Palaearctic grasslands" took place in Bayreuth from 6-10 March 2017.

The main aims were the organisation of the further development of the database, planning of overarching analyses and papers of the data as well as third-party grant proposals based on them.

The 13 workshop participants from nine countries combined representative of the main data contributors with specialists in ecological theory and state-of-the-art statistical analyses of such huge macroecological datasets. During one intensive week the participants planned about one dozen papers based on the common database, but also coined a name for this database (GrassPlot) and developed bylaws that balance the rights and interests of data contributors and data users.

During the Workshop one of the participants, Prof. Dr. David Storch (Charles University of Prague), gave a public lecture on "Biodiversity scaling: between biology and geometry", closely related to some of the main research topics that are going to be addressed with the GrassPlot data.

More information about the workshop can be found here.

Host: Jürgen Dengler (BayCEER, DE)

Participants: Idoia Biurrun (ES), Timo Conradi (DK/DE), Iwona Dembicz (BayCEER, DE/PL), Goffredo Filibeck (IT), Itziar García-Mijangos (ES), Riccardo Guarino (IT), Elisabeth Hüllbusch (BayCEER, DE), Monika Janišová (SK), Alireza Naqinezhad (IR), Santiago Soliveres (CH/ES), Manuel J. Steinbauer (DK/DE), David Storch (CZ), Viktoria Wagner (CZ/AT)

Remote participants: Steffen Boch (CH/DE), Alessandro Chiarucci (IT), Francesco de Bello (CZ), Swantje Löbel (SE/DE), Werner Ulrich (PL)