Programma della conferenza

Improving forests health and resilience is one of the goals of the New EU 2030 Forest Strategy and of the EU 2030 Biodiversity Strategy because forests provide many ecosystem services. In Mediterranean areas, wildfires are recurrent anthropogenic disturbances causing forest ecosystems degradation. This is of particular concern in areas of conservation interest that are increasingly affected by wildfires. A transdisciplinary and multiscale approach is adopted in the PRIN Project 20222CT8J3 FLER_MeCoFor to evaluate wildfires legacies through specific structural and functional indicators, based on field, ancillary, and satellite data. The project aims to assess the effect of recurrent wildfires in the Natura 2000 (N2K) site IT9130006—Pinewoods of the Ionian Arch (Apulia region, Southern Italy) which hosts the Habitat of Priority Interest 2270* (Wooded dunes with Pinus pinea and/or Pinus pinaster). At the landscape level the role of fire refugia in post-fire ecological succession is being investigated at two spatial scales (N2K forest landscape and the 9 individual stands comprised). This is based on the reconstruction, presented here, of the 1980 to 2020 fire regime and fire refugia spatio-temporal pattern and recovery time, within burned forest patches. Over the 40 years interval 22 (frequency 0.55) wildfires were mapped and characterised in terms of fire severity using ancillary and LANDSAT imageries. More than half (55.66%) of the total landscape area (1893.60 ha) was disturbed (mean fire size 47.91±46.98 ha) and 7.26 % of the burned area was classified as fire refugia (mean size 3.48±4.19 ha). Individual stands exhibit different fire regimes and fire refugia (permanent vs non-permanent) recovery space-time pattern. This confirms fire regime scale-dependency and underlies differential resilience dynamics and pathways of fire refugia and their neighbouring areas. These aspects are being further investigated at the ecosystem level by considering the effects of different fire severities on vegetation and soil properties.

Coastal areas, placed between land and sea, host a highly specialized biodiversity that is consistently impinged by human activities. Remote sensing (RS) offers sound support for describing and modelling landscape dynamics.

Based on multi-temporal landscape analysis we explored the main dynamic processes shaping landscape changes in the Central Adriatic coasts over the last 70 years. We specifically compared dynamic processes and landscape changes inside and outside coastal LTER (Long Term Ecological Research) sites, which offer a perfect playground for studying and modelling ecological processes.

We focused on a coastal tract (Molise region) that hosts two LTER sites (IT20-003-T: Foce Saccione-Bonifica Ramitelli and IT20-002-T: Foce Trigno–Marina di Petacciato) that are also included in the N2K network (IT7222217 and IT7228221).

Based on fine-scale (1:5000) multi-temporal land cover maps (1954, 1986, 2022), we calculated transition matrices and identified the main dynamic processes (e.g. urban expansion, seashore erosion, etc.) occurring on two-time steps (1954-1986; 1986-2022). We compared landscape processes inside and outside LTER sites implementing a Random Forest model (RF).

Major changes occurred in the first time step (1954-1986) with Agriculture expansion and forestation processes dominating inside LTER sites and urbanization outside. In the second time step (1986-2022) coastal landscape resulted more dynamic inside LTER sites with naturalization being the main process. This second period registered minor changes with curtailed expansion of agriculture and urban areas on both landscapes (inside and outside LTER sites) which do not align with the global trend of coastal urbanization.

The increased understanding of the main dynamic processes shaping coastal landscapes provides new elements and updated information useful for managing and conserving coastal areas.

Human activities such as urbanization, tourism, and invasive species increasingly threaten coastal dune ecosystems, which provide essential ecosystem services. Effective conservation and management require understanding dune vegetation’s structure and function, focusing on eco-functional characteristics. This study analyzes and maps eco-functional plant classes in Molise, southern Italy, using high-resolution remote sensing and plant functional traits. Forty-six taxa were identified, including diagnostic species for various EUNIS habitats, during vegetation sampling of 67 sites in the "Foce Trigno-Marina di Petacciato" LTER area. Ecological indicator values for soil moisture, nitrogen, reaction, light, and temperature were extracted, and traits such as life form, height, seed mass, and dispersal distance were analyzed. High-resolution remote sensing data was acquired using unmanned aerial vehicles with RGB, multispectral, and LiDAR sensors.

Model calibration involved evaluating multicollinearity and retaining critical vegetation and remote sensing variables. Vegetation clusters were identified using K-means clustering based on scaled vegetation variables. A model with four classes and five attributes (seed mass, dispersal distance class, ecological indicator value for temperature, Berger-Parker, and Simpson dominance index for life forms) was chosen from 2544 potential models for its ecological relevance. This model illustrated distinct eco-functional classes associated with specific dune habitats. Significant taxa for classification included Eryngium maritimum and Thinopyrum junceum, recognized for their ecological and diagnostic value.

The final eco-functional maps revealed heterogeneity in both dunes, with Class 1 prevalent near the coast and Class 4 more isolated. One dune exhibited distinct horizontal zonation with repeating bands, while the other showed a subtle mosaic pattern. This comprehensive approach, combining field sampling and advanced remote sensing, provides a robust framework for studying coastal dune vegetation. Insights into ecosystem resilience can guide strategies to mitigate environmental threats and enhance conservation efforts. Future research should refine these models and extend their applicability to other coastal regions.

Invasive alien plants negatively affect biodiversity and ecosystem services. The EU Copernicus Mission delivers free remote sensing data, facilitating cost-effective and timely monitoring of invaded areas. This study deploys multispectral (Sentinel-2) and thermal (Sentinel-3) satellites to characterize ecophysiological traits of vegetation patches invaded by Ailanthus altissima in Sardinia, and analyzes seasonal ecophysiological changes between highly invaded and native vegetation classes. A total of 176 invaded patches and their non-invaded buffer areas were identified on aerial orthophotos, digitized and rasterized at the resolution of 20 m². These cells were classified to the second level of the regional vegetation map (Carta della Natura) in A. altissima and native vegetation classes. A. altissima dominance was found in six vegetation classes: Mediterranean maquis, Mediterranean sub-nitrophilous grassland, deciduous woody vegetation, evergreen woody vegetation, agricultural herbaceous areas, and agricultural woody areas. After, we calculated a set of spectral indices as proxies of leaf chlorophyll and carotenoid content (CVI, SIPI3), productivity and canopy biomass (EVI, LAI), leaf water content (NMDI, MSI), soil features (CI), and daily evapotranspiration. We analyzed the monthly trends of these indices in invaded patches and buffer areas and their seasonal differences between invaded and not invaded cells, using linear mixed models (LMMs), two-way ANOVA, and Estimated Marginal Means. Our results highlighted the potential of Copernicus mission in capturing the temporal trends of ecophysiological spectral traits in invaded areas, as the high conditional R² values of LMMs ranged from 0.522 of CVI to 0.776 of LAI. The greatest significant differences between invaded and not invaded cells were observed during summer, i.e. higher productivity and canopy biomass, greater leaf water content, lower leaf carotenoid content, and lower bare soil presence. These results confirmed that A. altissima might have a competitive advantage over native vegetation, especially during the summer drought period of the Mediterranean basin.

The interactions among individuals, either vertical and horizontal, occur in space, and the space units can be conceptually described by the individual home range. However, while the home range concept is clearly defined as the spatial area required by an individual to satisfy its overall needs over the life cycle, its operationality and comparability can be hindered due to the complexity of individual behavior, such as sociality, and life cycle, as well as the absence of a standardized individual tracking system for most, if not all, species. Thus, the concept of home range can only be fully operational and comparable when the temporal and spatial scales are appropriately defined, based on the specific scientific questions being addressed. Here, we present a few study cases, covering both terrestrial and aquatic species, at local and global scale, using field and laboratory data to analyze the actual concept consistent with the data used, which allows assessment of home-range size and gives insights into the underlying mechanisms. When the spatial area required by an individual to satisfy its overall needs is contextualized to the competitive interactions between individuals of the same or different species, a time frame and an individual tracking system can be adapted to describe and analyze the spatial extent where interactions occur as well as the resource-mediated interaction among individuals. As all interactions occur in space, a proper, modular, definition of home range seems suitable to assign a spatial extent to intra and interspecific interactions among competing individuals.

The study of food webs represents a valuable tool for understanding the ecological dynamics that shape energy flows and predator-prey relationships. Because of their relative stability, food webs are also a good proxy for assessing anthropogenic impacts at the ecosystem level. On this basis, within the implementation of the Marine Strategy Framework Directive, Member States are required to assess the status of the marine food web in the context of Descriptor 4, specifically considering the diversity and productivity of trophic guilds. One of the most widely used approaches for assessing the structure and functioning of marine trophic networks is based on the use of stable isotopes of carbon and nitrogen (δ13C and δ15N), which are particularly useful for understanding the trophic ecology of a wide range of taxa whose diet composition may be impractical to assess through traditional approaches (e.g., stomacal content analysis). However, such data are scarce and have not been collected systematically, often making it difficult to assess the status of marine trophic webs. Here we present ISOMED, a georeferenced database of published isotopic values of δ13C and δ15N and elemental carbon and nitrogen composition of basal organic matter sources and consumers collected in the Mediterranean Sea. The reported information includes data for more than 5000 records at the species/taxa level. ISOMED provides a unique tool to study interactions and energy fluxes for components of the Mediterranean food web and contribute to the operationalization of ecosystem indicators for the assessment of Good Environmental Status.