TREELIM – Biogeography and climatology

A2 True climate: Microclimatology

Climate data expressed as seasonal mean temperature at the leading edge differ by species, but not by region. We are able to link next weather station data with our onsite 2-m air temperature readings linearly. Absolute minimum temperatures at tree top are ca. 2 K lower than air temperature recorded at 2 m above ground. Presumably decisive extremes are rare, but we recorded temperatures below -20 °C. These data are currently used to hindcast long term occurrences/probabilities of absolute minima in situ.

The distributional limits of tree species at high latitude or high elevation are likely associated with some manifestations of temperature. In the present section we explain in detail the cross continental datalogging campaign that we conducted to derive at further insights of temperature regimes that govern tree species distribution. In order to better isolate these temperatures the experiment was designed to encompass not only tree species elevational transects (Switzerland) but also a latitudinal transect (south Scandinavia).

We first conducted an extensive database research and identified two regions in Switzerland, where we found the highest occurrences of the studied species: One near Martigny (46°6′ N, 7°4′ E) and one near Chur (46°51′ N, 9°32′ E). In autumn 2009 during several field trips we placed 65 miniature dataloggers in those regions. For that we surveyed more than 5 elevational transects per region and extracted the highest occurrence of every species. At every species limit we mounted 4 sensors to hourly record the temperature profile of the tree. The same approach was used in southern Sweden, where the studied species find their northern limits along a latitudinal transect. The devices were recovered in autumn 2011 providing us with an immense dataset of 2 years hourly temperature values (regions Western and Eastern Switzerland) and 1 year of data from Southern Sweden.

This data is now being analysed to provide at least two major outcomes. (1) we aim at deriving at vertical temperature profiles along the trees and (2) we are working on recovering trees temperatures by relating our measured data to close long-term weather stations. In both subjects we concentrate on minimum temperatures since we believe these to be most important for tree species establishment at the edge of their distributions.

The data analysed so far revealed that by using weather station data to predict minima and maxima of 2 m air temperature inside the forest prediction errors of up to 4 K may accumulate (Fig. 4). Furthermore crown and understory minimum and maximum temperatures may deviate more than 1.8 K from 2 m air temperature inside the forest.