COS 43-5 - Landscape or local? Understanding relative roles of macro- and site-level conditions on long-term patterns in phenology for desert grassland plant species

Tuesday, August 8, 2017: 9:20 AM
D139, Oregon Convention Center
Dawn M. Browning1, Sheri Spiegal1 and Theresa M. Crimmins2, (1)Jornada Experimental Range, USDA - Agricultural Research Service, Las Cruces, NM, (2)National Coordinating Office, USA National Phenology Network, Tucson, AZ

Phenology is highly responsive to antecedent and immediate environmental conditions. Temperature and daylength are widely documented to strongly influence phenology in many species. Site-level conditions such as soil available water-holding capacity (AWC), degree of canopy closure, and aspect may also exert influence, especially in environments where moisture is limiting. However, the degree to which broad-scale environmental variables drive phenology relative to site-level or microclimate factors is not well known. We evaluated the relative influences of abiotic drivers that varied by site, such as soil depth and AWC, and abiotic drivers that varied inter-annually, but exhibited less variation across sites, such as air temperature and daylength, using a long-term dataset (1993 – 2016) from the southwestern United States. We hypothesized that site-level conditions would play a greater role in influencing the timing of leaf-out and fruiting than landscape-scale forcings. Using data for 16 species with a high degree of replicability across 15 sites that range in landscape position, soil texture and soil depth, we used agglomerative hierarchical cluster analysis to identify individual species and groups of species showing high within-year concordance in timing of leaf-out and fruiting across variable site conditions and rainfall years.


Membership in three of six first leaf clusters were species-specific including all sites where C3 shrubs Prosopis glandulosa, Flourensia cernua, and Gutierrezia sarothrae are monitored; this indicates that variable site-level conditions are less important than landscape factors. For first leaf, C4 grasses occurred either in one large cluster with members occurring at 14 of 15 sites, a smaller cluster including grasses at playa sites, and a third cluster with two Muhlenbergia species including all sites where they are monitored. For first fruit, there were two large clusters with C4 grasses – one with species that occurred at sites with sandy and gravelly soils and another with species at sites on alluvial flats. C3 shrubs occurred in fruit clusters as groups of species spanning multiple sites. Larrea tridentata and Prosopis glandulosa demonstrated similar patterns in fruit production to occur in one cluster while Gutierrezia sarothrae demonstrated a unique species-specific pattern in fruiting over the of the 23-year period. These data indicate that site-level factors play a greater role than broad-scale drivers in the timing of first fruit for C4 grasses while timing for C3 shrubs is consistent across sites. Understanding factors driving dryland plant phenology can increase efficiency of conservation-based phenology monitoring and refine development of predictive models.