LAND - Terrestrial biodiversity through time - novel methods and their applications

The ultimate goal for the project is to develop further methods to be able to estimate terrestrial biodiversity through time. It is important in the project to use and develop new numerical methods, as well as to evaluate the more traditional methods used earlier. To reach this goal it is important to study the present-day relationship between diversity in the vegetation and the pollen it produces and deposits. This relationship is not 1:1, but dependent on factors such as pollen production, pollen dispersal, and taxonomic resolution. To be able to study the present-day relationship between plants and pollen we have selected Setesdalen as our study area. The valley has an elevational gradient from 0 - 1200 meters above sea level, and covers six main vegetation types. From 53 lakes pollen has been counted from surface sediment samples and the surrounding vegetation has been mapped. To be able to compare todays vegetation and pollen it has been necessary to transpose plants into their pollen types so that the taxonomic resolution is the same. This is done in BioDiverse based on pollen identification as in Beug (2004), Fægri et al. (1964, 1989), Moore at al. 1991, and Birks and Peglar (upublisert) and plant names as given in Lid & Lid (2005). The table is published at https://www.uib.no/rg/EECRG/artikler/2012/03/vascular-plants-and-their-pollen-or-spore-type-innorway. Based on the collected data expected pollen richness, plant richness, and pollen and plant richness against elevation have been calculated. Results indicate that the number of plants has a unimodal relationship with elevation, and the plant richness is highest in the boreo-nemoral zone in the south, and decreases with increasing elevation. Pollen richness has a opposite unimodal relationship and expected richness is high both in the boreo-nemoral and the low-alpine zones. After the conversion of plants into pollen the relationship between plants and pollen in Setesdalen is linear (see Felde et al 2014, Holocene). The results shows that the main vegetation zones are better represented in the pollen data than in the vegetation data, probably a result of pollen data giving a more regional signal than the vegetation data. This is important to take into account when estimating plant and pollen richness for the past. In the final phase of the project we are working on the further development of methods to calculate richness and diversity with different methods of correcting the pollen data based on available information on pollen production and representation and how to implement this when estimating past (last 10 000 years) richness.

There is an urgent need for long-term (> 50 years) biodiversity data to assist in the prediction of future biodiversity changes in response to global change. Long-term ecological data such as pollen assemblages preserved in lake sediments can provide quan titative estimates for the last 8000 years of richness, evenness, composition, and turnover, the major components of biodiversity that determine the functional traits of ecosystems. Such estimates are biased, and this project develops and tests new method s for the unbiased estimation of these four components of biodiversity at sites situated in the major vegetation zones in Fennoscandia. It also attempts to calibrate pollen richness and plant-species richness and to synthesise, interpret, and model the bi odiversity patterns in space and time.