The forage maturation hypothesis [FMH] assumes that herbivores cope with the trade-off between digestibility and biomass in forage by selecting vegetation at intermediate growth. The green wave hypothesis [GWH] extends the FMH to suggest how spatiotemporal heterogeneity in plant quality shapes migratory movements of herbivores. Growing empirical support for these hypotheses mainly comes from studies in vast landscapes with large-scale habitat heterogeneity. It is unclear, however, to what extent ungulates surf green waves in human-altered landscapes with small-scale heterogeneity in terms of land use and topography. We used plant phenological proxies derived from Sentinel 2 satellite data to analyse the habitat selection of 93 collared red deer [Cervus elaphus] in montane and alpine habitats. Using a step selection analysis, we investigated how plant phenology, i.e. the instantaneous rate of green-up [IRG] and normalised difference vegetation index [NDVI], and a set of variables describing topography and human presence influenced red deer resource selection in open habitats. We learned that red deer selected areas with high biomass at green-up and avoided habitats with possible exposure to human activity. Additionally, landscape structure and topography strongly influenced spatial behaviour of red deer. We further compared cumulative access to high-quality forage across migrant strategies and found migrants gained better access than residents. Many migratory individuals surfed the green wave, their surfing behaviour, however, became less pronounced with decreasing distance to settlements. Within the constraints of topography and human land use, red deer track spring green-up on a fine spatiotemporal scale and follow the green wave across landscapes in migration movements. Thus, they benefit from high quality forage even in human-dominated landscapes with small-scale heterogeneity and vegetation emerging in a heterogenic, dynamic mosaic. --

We used spatial relocation data for female [58] and male [35] red deer that were captured and marked in a period from 2015 to 2018. Individuals were darted and immobilized at night in order to collar them [GPS telemetry collars; Vectronic Aerospace GmbH]. They were then monitored over a period of 1-2 years, except for early drop-offs due to fatalities or technical issues. Capturing, marking and collaring was performed in the winter home ranges by authorities and game wardens of the respective cantons and was in line with Swiss animal welfare laws and approved by the appropriate authorities [permissions SG13-12, GR2014-07F, GR2015-09, VS07-17].

We used plant phenological proxies derived from Sentinel 2 open satellite data to analyse the habitat selection of the 93 red deer in montane and alpine habitats based on the Greenwave hypothesis (GWH). With a step selection analysis, we investigated how plant phenology, i.e. the instantaneous rate of green-up (IRG), normalised difference vegetation index (NDVI) and a set of variables describing topography and human presence contributed to explaining red deer resource selection in open habitats. Additionally, we assessed the animals’ migration strategies and investigated which strategy might be more beneficial in terms of access to high quality forage.

 

ArcGIS [version 10.5, ESRI], R version 3.5.1 (R Core Team, 2018)