This was a field protocol and observational mapping study rather than an experiment measuring the ecological effects of grazing. Its authors focused on an area encompassing two corridors in central Italy: Corridor 1 links the Sirente-Velino Regional Natural Park with the Abruzzo, Lazio and Molise National Park, while Corridor 2 connects the latter with the Majella National Park. The corridors support Marsican bears and also provide habitat for mammals including roe deer, red deer, wild boar and porcupine.

To compile the data, the researchers requested 2023 livestock records from municipal offices in the 15 municipalities wholly or partly within the study area. The records included the number of animals assigned to farms using municipal pasture under Italy’s annual or seasonal “fida pascolo” grazing-concession system. This system is governed by provisions dating from 1927 and 1928, while the cadastral framework for land parcels derives from an 1886 law. Municipal documents identified the relevant cadastral sheets and parcels and the livestock numbers associated with each farm, although completeness and detail varied among municipalities.

Farmers were interviewed anonymously to fill gaps in animal-category classifications and records of where pastures were located. On-site visits supplied pasture assignments by cadastral sheet and animal numbers by species and age class. The resulting database recorded the municipality, an anonymised farm code, cadastral sheet, area allocated in hectares, and numbers of cattle older than 24 months, cattle aged 6–24 months, cattle younger than six months, sheep and goats older than 12 months, and horses older than six months. Only aggregate information was published to protect farms’ identities and locations.

For mapping, the team selected cadastral sheets from the Italian national cadastral cartography and combined them with livestock data in QGIS. Where animal numbers were available by farm and parcel, the parcels’ areas were proportionally allocated to the relevant cadastral sheet, under an assumption that animals were distributed evenly within it. For Gagliano Aterno, where the records identified grazed sheets but not animal numbers for each one, the calculation used the municipality’s total livestock divided by its total grazed area. The protocol then associated livestock counts and densities with the mapped sheets.

Grazing pressure was expressed in livestock units, or LSUs, to combine different animal categories. Under European Commission Implementing Regulation 2016/669 conversion factors used by the municipalities, cattle and equines over the specified age thresholds counted as 1 LSU, cattle aged six months to two years as 0.6 LSU, cattle below six months as 0.4 LSU, and sheep and goats as 0.15 LSU. The resulting shapefile contained eight mapped variables: LSU, LSU density, equines, equines density, cattle, cattle density, sheep and goats combined, and their density.

Gagliano Aterno had the highest municipal LSU count, at 461.8, followed by Ortona dei Marsi with 373.2, Anversa degli Abruzzi with 350.7 and Scanno with 295.2. Gagliano Aterno also recorded the most cattle, with 366, compared with 204 in Ortona dei Marsi and 166 in Cocullo. Cattle density was highest in Ortona dei Marsi at 2.03 cattle per hectare of grazed cadastral sheets, followed by Scanno at 0.96. Anversa degli Abruzzi had the largest combined sheep-and-goat count, 1,173 animals, and the highest density, 8.21 individuals per hectare; Scanno ranked third in that category’s density at 5.38. Ortona dei Marsi had both the largest equine population, 117, and the highest equine density, 1.62 per hectare.

The municipal LSU estimates were generally higher than values from two larger-scale studies by Malek and colleagues, although some municipalities showed higher comparison values in one of those datasets. After horses were excluded because the comparison datasets omitted them, the protocol’s municipal values showed a moderate correlation with Malek et al. (2024a), although the result was not statistically significant at the 0.05 threshold (Spearman’s rho 0.47, p=0.08). Correlation with Malek et al. (2024b) was weak (Spearman’s rho 0.14, p=0.02). The authors attributed differences to methodological variation and the use of different LSU coefficients.

Several limitations constrain the maps. They exclude livestock on private pastureland because those data were unavailable, which may cause some underestimation even though private grazing was a small part of the mapped region. The method assigns livestock to cadastral sheets rather than locating individual animals, and its homogeneous-distribution assumption may not reflect unequal forage availability or herds concentrating in particular places. Results also have limited temporal validity because municipal grazing concessions are renewed annually and may change each year. The protocol nevertheless identifies mapped areas for assessing potential overgrazing and the interactions between pasture activity, habitat conditions, wildlife and possible pathogen transmission in bear corridors.

Subject of Research: Standardised mapping of livestock grazing pressure in two Central Apennines ecological corridors

Article Title: Map of livestock density in Central Appenines: a standardised protocol

Article References: Dragonetti, C., Masiello, G., Villa, F., Von Kempis, S. R., Cipollone, M., & Di Marco, M. (2025). Map of livestock density in Central Appenines: a standardised protocol. Nature Conservation, 59, 335-349. https://doi.org/10.3897/natureconservation.59.155735

Image Credits: AI Generated

DOI: 10.3897/natureconservation.59.155735

Keywords: livestock density, grazing pressure, Central Apennines, Marsican brown bear, municipal pastures, livestock units, GIS mapping, biodiversity conservation, livestock, density, Central, Appenines