1. www.irstea.fr Structure And Biodiversity In Managed And Unmanaged Mixed Beech Forests: A Comparison Based On The Strict Forest Reserves Network In France Symposium for Research in Protected Areas 10 - 12 June 2013, Mittersill (Austria) Nicolas Debaive, Yoan Paillet, Coryse Pernot, Vincent Boulanger, Nicolas Drapier, Olivier Gilg, Patrice Hirbec, Frédéric Gosselin

2. 2 Unmanaged forests in Europe Primeval forests < 1% of the European forest area 13% - US west coast 40-52% - Canada Reference state for forest management and biodiversity Global effect of management on forest dwelling species?

3. 3

4. 4 ? ? No real “primeval” forests SFR: 0.3% of the forest area Most of the French strict forest reserves are recent No structure / biodiversity comparisons between reserves and managed forests

5. 5 Aims and scopes Hypotheses -Higher levels of old-growth structural attributes and biodiversity in unmanaged forests -Increasing “old-growthness” and biodiversity with time since last harvesting

6. 6 Materials and methods Study sites 15 lowland and mountains beech dominated forests, 213 plots Plots chosen at random in and around forest reserves, controlled forest site Time since last harvesting - MAN: 9 ±12 years - UNM: 46 ± 38 years

7. 7 Stand structure characterization Combined fixed angle and surface techniques Living wood DBH>7.5cm Snags DBH>7.5cm Logs D>5cm

8. 8 Biodiversity sampling: 7 taxa Vascular plants, bryophytes 1000m², 2 observers, 35min Epixylic fungi all living and dead trees Birds and bats 5 and 40 min point counts Insects saproxylic and carabid beetles

9. 9 Analyses Stand structure Non-linear generalized model, confidence intervals re-estimated by bootstrapping In practice Multiplication coefficient assessed against a null hypothesis Biodiversity (total species richness) Generalized linear models with Poisson distribution Explanatory variables - management: MAN vs. UNM - elevation: LWL vs. MON - Time since last harvesting 32 % *

10. 10 Results: stand structure 145 % 360 % Total deadwood volume (m 3 /ha) Basal area of Very large Trees (m 2 /ha)

11. 11 Most of the results significant in lowland forests - Number and basal area of large trees - Deadwood volumes Far less in mountain forests - Number and basal area of Very Large Trees - Total deadwood volume Higher levels in mountains both in MAN and UNM forests Results: stand structure

12. 12 Results: stand structure Effects of the time since last harvesting

13. 13 Managed forestsUnmanaged reserves TaxanEstimated MeanSEEstimated MeanSEp Fungi998.61.19212.31.190*** Vascular plants19732.51.10032.71.100ns Bryophytes8619.61.11123.91.110*** Carabids1213.31.2933.11.294ns Saproxylic beetles16926.02.03224.22.032ns Birds18511.11.07511.91.075(*) Bats1014.81.3525.71.351ns Results: biodiversity Managed forests vs. strict reserves

14. 14 Results: biodiversity Total richness vs. Time since last harvesting

15. 15 Discussion Strong effects of forest management on forest structure Most structural attributes Are higher in unmanaged forests Increase with time since last harvesting: Forest harvesting tends to Shorten forets silvigenic cycle Eliminate aged and senescent phases Reserves in aggradation phase Slow recovery of old-growth characteristics

16. 16 Only total richness of Fungi, Bryophytes and Birds showed significant differences Species that depend on deadwood and large trees during their life cycle Surprisingly no response of saproxylic beetles Deadwood volume = main driver? Discussion Small differences in terms of biodiversity

17. 17 Lack of response for some groups despite structural differences French reserves very recent: no colonisation of typical species An extinction debt already been paid ? Sustainable management in progress Greater role of other spatial and time scales on biodiversity than management per se ? Discussion Small differences in terms of biodiversity

18. 18 Thanks for listening !