Surveying historical patterns in vegetation change (1929-2019) in the upper East River Valley using RMBL's archival herbarium records

Abstract

One of the ongoing challenges in studying the effects of global change on biodiversity is finding credible baseline data of prior conditions. The Rocky Mountain Biological Laboratory in Gothic, Colorado is located in a valley with a wide range of plant species and communities due to its diversity of microclimates, and sharp gradients of precipitation and elevation (ca. 8000 – 14000 feet ASL). There have been a number of studies conducted near Gothic documenting changes in plant community composition, population distributions and flowering phenology, but the herbarium has been little used by researchers as a source of historical plant distribution and phenology data. Herbarium data provide a novel tool to assist biologists when conducting floristic studies to trace vegetation distribution patterns and studies of long-term phenological change. With the data available in the RMBL’s online herbarium database, potential study sites can be located, and target species lists can be generated from a long list of historical specimens (dating back to 1929) to conduct modern surveys and make qualitative comparisons of species presence or absence. Here we test the adequacy of archival herbarium data as a baseline to study historical changes in species richness and abundance from three plant communities from montane to alpine habitats. We collected 380 new specimens from Kettle Ponds (montane meadow and riparian), 113 specimens from Lower Virginia Basin (subalpine meadow) and 125 new specimens from Upper Virginia Basin (alpine meadow). We added a total of 161 new species, unrepresented by historical records, for these sites; 76 at Kettle Ponds, 64 at Lower Virginia Basin and 21 at Upper Virginia Basin. In addition, we estimated the relative abundance of the dominant species at each site using quadrat samples of percent vegetation cover. We conclude that using RMBL’s herbarium database is useful for finding potential study sites with a diversity of species, but these species lists are incomplete due to a variety of sampling biases. Species lists do not reflect relative species abundance; historical collections seldom included the common, dominant species identified from quadrat sampling. We concluded that blanket sampling is the most effective method to document species richness so long as every species, common or rare, are collected with the intention of building a complete species list. Finally, after compiling merged species lists of both historical and recent specimen data, we found that the rate of accumulation of new species declined asymptotically as we approached a complete species list for each site we sampled throughout the summer.

Local Knowledge Graph (9 entities)

Related Works

Items connected by shared entities, co-authorship, citations, or semantic similarity.

References (6)

1 in Knowledge Hub, 5 external