Assessing biodiversity change in the Anthropocene poses a significant challenge; therefore, the development of accessible databases is crucial to support ecological research and guide effective conservation strategies, addressing global biodiversity loss (Meyer et al. 2015; Dornelas et al. 2018).

A large amount of available vegetation data, encompassing extensive coverage in terms of space, habitats, and time, derive from phytosociological relevés (Chytrý et al. 2016). These data were collected over the past decades thanks to a long-standing tradition of botanists and represent an essential data source for fundamental research and nature conservation applications, including monitoring the occurrence and abundance of plant species in different habitats (Chytrý et al. 2016; Sabatini et al. 2021; Bagella et al. 2024; Gholizadeh et al. 2025). In this context, the development of specialised databases focused on temporal dynamics is particularly crucial, as they provide the empirical foundation for understanding long-term ecological shifts (Costello et al. 2013).

A key methodological approach underpinning such databases is the resurvey of historical sites i.e. revisiting previously studied locations to resample historical relevés. This approach offers opportunities to study temporal vegetation changes with acceptable relocation accuracy and quantify long-term biodiversity dynamics (Kapfer et al. 2017; Knollová et al. 2024).

Vegetation resurvey is considered a valid approach for assessing how species composition, abundance, and distribution have evolved in response to anthropogenic pressures, climate change, and habitat alterations over recent decades (Kapfer et al. 2017; Dornelas et al. 2018). This approach not only highlights temporal trends but may also help identify drivers of change, thereby guiding targeted interventions to mitigate further declines (Perring et al. 2016). Nevertheless, even if it presents methodological shortcomings, such as relocation errors, it still represents one of the best methods currently available for monitoring vegetation changes (Verheyen et al. 2018; Douda et al. 2023).

In Europe, the ecological importance of coastal dune systems is formally recognized under the Habitats Directive (EEC 1992), which designates several dune habitats of Community interest. The Directive requires Member States to ensure their favourable conservation status through monitoring, protection, and restoration measures.

Sardinia, a Mediterranean hotspot of plant diversity (Médail and Quezél 1999), is internationally renowned for its coasts and beaches, many of which are in protected areas. In some northeastern and southeastern zones, access to approximately 20 beaches has been restricted; however, growing tourist flows urgently call for a regional plan to identify areas at risk of overtourism and promote sustainable tourism practices (Balletto et al. 2025).

Despite their conservation and economic importance, most Sardinian dune systems, although supported by a considerable amount of historical data, have not been recently monitored to assess long-term changes in vegetation.

Resurveying Sardinian dunes is thus essential for tracking long-term temporal trends in plant biodiversity and supporting successful conservation and restoration measures.

This article introduces the first regional database, Resurvey Sardinian Dunes (ReSarDu), compiled from resurveys of coastal dune systems across Sardinia (Italy). The database includes data from historical surveys dating back to the 1970s and from resurveys carried out at the same sites in 2023–2025.

Specifically, this paper aims to: (i) fill the knowledge gap relative to biodiversity changes in Mediterranean dune vegetation through the last decades; (ii) provide a further contribution to ReSurveyDunes (Acosta et al. 2025), the database of resurveyed coastal dune vegetation in Italy.

For the resurvey, we selected 188 historical phytosociological relevés (hereafter old surveys) mainly distributed along the northeastern and southwestern Sardinian coasts (Fig. 1) and conducted between 1975 and 2005 (Fig. 2a). These old surveys were mainly extracted from VegDunes database (Prisco et al. 2012), which includes data from several studies on Sardinian dune vegetation (De Marco and Mossa 1975; Brambilla et al. 1982; Valsecchi and Bagella 1991; Filigheddu and Valsecchi 1992; Mossa 1992; Mossa and Biondi 1992; Bartolo et al. 1992; Arrigoni 1996; Biondi et al. 2001; Biondi et al. 2004) integrated with other old published (Farris et al. 2013) and unpublished surveys. Each old survey was provided with standard information, plot size, survey date and assigned to the corresponding Annex I habitat according to Habitats Directive (92/43/EEC).

The old plots to be resurveyed were selected based on several attributes, such as location, to ensure maximum spatial distribution, and the possibility of identifying their position as precisely as possible. To achieve this objective, for example, all past surveys carried out by the surveyors involved in the resurvey campaign were included.

Among the 188 selected old surveys, 73 exhibited low relocation uncertainty and were consequently considered as performed in quasi-permanent plots. The remaining 115, lacking detailed location information, were considered as performed in non-traceable plots (Kapfer et al. 2017). For identifying plot locations when coordinates were unavailable, we relied on alternative sources, including locality descriptions, reference points provided in the original publication, or support from the surveyor who conducted the initial sampling.

For each non-traceable plot, following Ross et al. (2010), we carried out three resurveys in three different plots (hereafter called new plots). In total, we conducted 418 resurveys: 73 in quasi-permanent plots and 345 in the new plots, all accurately georeferenced with a Garmin Montana 680.

Resurveying was conducted between 2023 and 2025 considering the same plot area and season of the old surveys. The shape was mainly rectangular, following the distribution of vegetation along transects. The occurrence of all vascular plant species was recorded, and species abundance was visually estimated using the Braun-Blanquet seven-degree scale (Braun-Blanquet 1964). Taxonomic nomenclature was standardized according to the Euro+Med PlantBase (Euro+Med PlantBase 2025). To each new survey, the following additional information were associated: elevation, exposure, and slope.

Each new survey was assigned to a specific habitat type according to the EU Habitats Directive (92/43/EEC), based on the presence/abundance of typical or characteristic species (Biondi et al. 2009). The following habitat types were identified: 1210 – annual vegetation of drift lines, 2110 – embryonic shifting dunes, 2120 – shifting dunes along the shoreline with Ammophila arenaria, 2210 – Crucianellion maritimae fixed beach dunes and 2230 – Malcolmietalia dune grasslands. The surveys not classifiable to any habitat type were labelled as NC (Not Classifiable as habitat). The most represented habitat is 2120 (122 surveys), followed by 2110 (108 surveys). The largest share of old surveys derive from the 1981–1990 period (Fig. 2a). The most frequent survey sizes range from 11 to 50 m² (Fig. 2b).

The ReSarDu database serves as a valuable resource for understanding long-term vegetation changes in Mediterranean coastal dune systems. By integrating historical and newly collected phytosociological data, we provide a comprehensive tool for assessing biodiversity trends and supporting effective conservation and restoration strategies. Being composed of georeferenced data, ReSarDu has the potential to help address existing knowledge biases regarding the flora of Sardinia (Bagella et al. 2020; Melis et al. 2025).

The database constitutes a reservoir of georeferenced plots that can be repeatedly monitored in the coming years, offering unique opportunities to investigate temporal trajectories of plant communities, assess the impacts of anthropogenic pressures and climate change, and evaluate the success of management actions.

The standardized structure of ReSarDu ensures compatibility and interoperability with national and international vegetation databases, such as ReSurveyDunes and ReSurveyEurope, thereby promoting data sharing and facilitating large-scale ecological analyses across Mediterranean coastal systems.

Bagella S., Caria M.C., Denaro A., Malavasi M.: Conceptualization. Bagella S., Caria M.C., Denaro A., Farris E., Malavasi M.: Methodology. Denaro A., Rivieccio G., Maccioni A., Farris E., Bagella S., Caria M.C., Fadda M., Bruno A., Malavasi M.: Investigation. Denaro A., Sperandii M.G., Malavasi M.: Data Curation. Bagella S., Denaro A.: Writing – Original Draft. All authors: Writing – Review & Editing. Farris E., Bagella S.: Funding Acquisition.

The authors declare that they have no conflict of interest. Simonetta Bagella is an Associate Editor, Alicia Teresa Rosario Acosta, Emmanuele Farris, and Maria Carmela Caria are part of the Editorial Review Board in Vegetation Ecology and Diversity but took no part in the peer review or decision-making process for this manuscript.

By the end of the project, we plan to integrate ReSarDu into VegItaly database (GIVD: EU-IT-001). The data sharing will be fully accessible upon specific request to the database custodians (Agnese Denaro adenaro@uniss.it; Simonetta Bagella sbagella@uniss.it).