High dissimilarity within a multiyear annual record of pollen assemblages from a North American tallgrass prairie.
Ecol Evol
; 6(15): 5273-89, 2016 Aug.
Article
en En
| MEDLINE
| ID: mdl-27551382
Grassland vegetation varies in composition across North America and has been historically influenced by multiple biotic and abiotic drivers, including fire, herbivory, and topography. Yet, the amount of temporal and spatial variability exhibited among grassland pollen assemblages, and the influence of these biotic and abiotic drivers on pollen assemblage composition and diversity has been relatively understudied. Here, we examine 4 years of modern pollen assemblages collected from a series of 28 traps at the Konza Prairie Long-Term Ecological Research Area in the Flint Hills of Kansas, with the aim of evaluating the influence of these drivers, as well as quantifying the amount of spatial and temporal variability in the pollen signatures of the tallgrass prairie biome. We include all terrestrial pollen taxa in our analyses while calculating four summative metrics of pollen diversity and composition - beta-diversity, Shannon index, nonarboreal pollen percentage, and Ambrosia:Artemisia - and find different roles of fire, herbivory, and topography variables in relation to these pollen metrics. In addition, we find significant annual differences in the means of three of these metrics, particularly the year 2013 which experienced high precipitation relative to the other 3 years of data. To quantify spatial and temporal dissimilarity among the samples over the 4-year study, we calculate pairwise squared-chord distances (SCD). The SCD values indicate higher compositional dissimilarity across the traps (0.38 mean) among all years than within a single trap from year to year (0.31 mean), suggesting that grassland vegetation can have different pollen signatures across finely sampled space and time, and emphasizing the need for additional long-term annual monitoring of grassland pollen.
Texto completo:
1
Colección:
01-internacional
Base de datos:
MEDLINE
Idioma:
En
Revista:
Ecol Evol
Año:
2016
Tipo del documento:
Article
Pais de publicación:
Reino Unido