Using a proxy of plant productivity (NDVI) to find key periods for animal performance: the case of roe deer


  • Nathalie Pettorelli,

  • Jean-Michel Gaillard,

  • Atle Mysterud,

  • Patrick Duncan,

  • Nils Chr. Stenseth,

  • Daniel Delorme,

  • Guy Van Laere,

  • Carole Toïgo,

  • Francois Klein

N. Pettorelli, A. Mysterud and N. C. Stenseth, Centre for Ecological and Evolutionary Synthesis (CEES), Dept of Biology, Univ. of Oslo, PO Box 1066 Blindern, NO-0316 Oslo, Norway. – J.-M. Gaillard, Unité Mixte de Recherche No. 5558 « Biométrie et Biologie Evolutive », Bâtiment 711, Univ. Claude Bernard Lyon 1, 43 Boulevard du 11 novembre 1918, FR-69622 Villeurbanne Cedex, France ( – P. Duncan, Centre d'Etudes Biologiques de Chizé, Unité Propre de Recherche No. 1934, Villiers-en-Bois, FR-79360 Beauvoir-sur-Niort, France. – D. Delorme, G. Van Laere, C. Toigo and F. Klein, Office National de la Chasse de la Faune Sauvage, Centre National d'Etudes et de Recherches Appliquées Cervidés-Sangliers, 85bis Avenue de Wagram, FR-75017 Paris, France.


Animals in seasonal environments are affected by climate in very different ways depending on season and part of the climatic effects operates indirectly through the plants. Vegetation conditions in spring and summer are regarded as decisive for the reproductive success and the offspring's condition of large herbivores, but objective ways to determine key periods during the growing season have not been done often due to limitations in plant data. Using satellite data (NDVI), we determined how plant productivity from birth to fall influences the following winter body mass of roe deer fawns. We do this in two populations, the first inhabiting the low productive Chizé reserve in south western France with an oceanic climate and the second from Trois Fontaines, a highly productive forest with continental climate in east France. The effect of plant productivity was similar for male and female fawn mass, as expected from the weak intensity of sexual selection in roe deer life history traits. We found contrasting results between sites, with a strong effect of plant productivity in spring (April-May) in the Chizé population, but no effect in the Trois Fontaines population. The relatively low variability in winter fawn body mass could account for the absence of NDVI effects at Trois Fontaines. However, such results might also point to a limitation in the use of the NDVI, since the relationship between the canopy and the plant productivity at the ground level might be weak in the highly productive forest of Trois Fontaines.