Canton d'Aurignac - ZA eLTER Pyrénées Garonne © Luc Barbaro
PARMENIDE - Results

The recorded acoustic diversity is linked to the natural character of the landscapes - PARMENIDE

PARMENIDE shows that the degree of naturalness of a landscape promotes acoustic diversity. In particular, landscape mosaics that are not highly fragmented and contain a high proportion of permanent grassland, as well as landscapes where management practices are becoming more extensive (particularly organic farming), very clearly promote biophony.

Canton d'Aurignac - ZA eLTER Pyrénées Garonne
© © Luc Barbaro

Temperate rural landscapes have undergone profound changes since the post-war period due to the combined effects of climate and land-use changes. To better quantify the underlying socio-ecological dynamics, there is a need for integrative metrics that capture the complexity of the relationships between agricultural practices, landscape heterogeneity and biodiversity. The acoustic component of biodiversity has seen significant conceptual and methodological advances with the emergence, over the past decades of landscape ecoacoustics and the use of acoustic diversity metrics that allow for the simultaneous quantification of sounds of biological origin (biophony) and anthropogenic origin (anthropophony). In the PARMENIDE project, we tested the hypothesis that the acoustic diversity of rural landscapes would be correlated to the compositional and configurational heterogeneity of landscape mosaics and enhanced by production systems tending towards the extensification of management practices (notably organic farming). 

Approaches

To test this hypothesis, we gathered a large-scale sample of long-term study sites in France drawn from the networks of INRAE sites (Sebiopag, UREP, UE St Laurent de la Prée), CNRS (Armorique, Arc Jurassien and Pyrénées-Garonne LTERs) and Regional Natural Parks (Baronnies Provençales, Pyrénées Ariégeoises). By combining approaches from ecoacoustics and landscape ecology, we tested the effect of the proportion of different types of land use cover (biological or non-biological, perennial or non-perennial) in the landscape on indices of acoustic diversity, as well as the effect of the degree of naturalness of the surrounding landscape at multiple spatial scales extending up to 1 km around the recorders. 

Results

At the national level, our results show that landscape heterogeneity and the shift toward large-scale extensification of farming practices are key factors in acoustic diversity. In particular, the amount of permanent grassland, the extent of organic farming in the landscape, and the degree of naturalness promote biophony, whereas techno-anthropophony dominates in the most fragmented rural landscapes. 

At the regional level, we observe that the effects of organic farming practices and landscape heterogeneity can vary depending on the local composition of bird communities—as measured by the acoustic complexity of bird choruses—in response to agricultural practices stemming from the locally dominant production systems within the landscape. 

These results open new perspectives for the study of the acoustic diversity of mosaic farmland landscapes, within which the multiplicity of sound sources makes their analysis both complex and integrative of socio-ecological dynamics. 

The perspectives of the PARMENIDE project are currently being explored within the framework of its follow-up Biosefair project, FARMSOUND, in which we are expanding the response metrics to include those derived from bird acoustic communities identified via deep learning (BirdNET) and the geomatics variables to include those describing historical landscapes before and immediately after post-war agricultural land consolidation.

 

These two approaches also make it possible to significantly expand the interdisciplinary scope of the research undertaken within PARMENIDE by emphasizing the integration of concepts and methods from geomatics and artificial intelligence with those already employed from landscape ecology, ecoacoustics, and agroecology.

 

Participants

INRAE units involved

Partners

  • LADYSS - Laboratoire dynamiques sociales et recomposition des espaces - CNRS
  • SETE - Station d’Écologie Théorique et Expérimentale - CNRS
  • University of Stirling (Écosse)
  • ISPRA - Italian Institute for Environmental Protection and Research (Italie)

Contacts - coordination