Prof. I.D. Hodkinson Dr. S.J. Coulson School of Biological & Earth Sciences, Liverpool John Moores University, Byrom St., Liverpool L3 3AF, UK (Contact details: Tel. 0151 2312030 Fax. 0151 207 3224 email email@example.com; firstname.lastname@example.org) Prof. N.R. Webb NERC Centre for Ecology & Hydrology, Winfrith Technology Centre, Dorchester, Dorset, DT2 8ZD, (Contact details: email email@example.com) Objectives and Hypotheses Our main objectives are to: describe, measure and model patterns and rates of invertebrate community development and succession following glacial retreat in the high Arctic using known chronosequences. cross-relate rates of community change to known climatic shifts. relate invertebrate community development to rates of key ecological processes such as decomposition of organic matter. evaluate the potential for more southerly species successfully to invade existing Arctic invertebrate communities. develop descriptive and predictive models of community development under conditions of climatic amelioration. We are testing the following hypotheses: that dispersal of particular functional groups of invertebrates in response to climate warming is a rate-limiting factor for invertebrate succession and community development in the high Arctic. that invertebrate community development in response to climatic warming is deterministic and directional, and therefore predictable. that the magnitude and stability of key ecosystem processes, such as decomposition, in the high Arctic are linked to biotic complexity, which can be suitably characterised by the invertebrate community composition. that natural succession provides a useful model for predicting rates of invasion by colonising species following climatic amelioration. Study sites Studies on two contrasting but complementary chronosequences on west Spitsbergen commenced in June 2000, an oligotrophic succession on t he glacial foreland of Midtre Lovénbre and a relatively eutrophic succession on Lovénøyene, a series of islands in Kongsfjord. A 1.5 km transect was established, extending from the foot of the Midtre Lovénbre to the terminal moraines and across the sandur. Seven equally spaced sampling sites (approx 20 x 40 m) were established at right angles to the main transect line). Each site was chosen to represent the most mature vegetation type present at each point. By contrast, each Lovénøy was viewed as a separate sample site. The chronology of glacial 'retreat' was established from vertical and oblique aerial and ground based photographs held by the Norsk Polarinstitutt Archive, Tromsø, from historical records and ground photographs and, for the oldest site, by radiocarbon dating of the soil. Results Ages of sites: The ages of the sites from the Midtre Lovénbre sequence vary between 2 years (site one) to 1900 (site seven), while the islands vary between 100 (Leirholmen) to 1800 (Storholmen). Plant community description and soil formation A detailed description has been made in the changes in the plant community (18 taxa) from site 1-7 on the Lovénbre - from unconsolidated parent to almost 100% ground cover. The presence, abundance and dynamics of each species have been described. Species have been characterised as early, mid or late successional. Parallel trends occur in soil characteristics including increasing depth, increasing organic matter and water content, decreasing clast size and a lowering of pH. Animal community description The soil fauna comprise primarily Collembola, mites, Enchytraeidae and chironomid larvae. Herbivores (one aphid and sawfly larvae) are few but hymenopteran parasitoids and predators (spiders and gamasid mites) are abundant. The distribution patterns of species and their abundances have been quantified for both the Lovénbre and Lovénøyene chronosequnces. The very first colonisers of bare moraines are Linyphiid spider species (predators). Other early soil colonisers are generally the surface active species such as the collembolan Isotoma anglicana. The poorest colonisers are the deep soil dwelling species. Experiments are thus underway examining wind blown dispersal and survival on seawater. A cellular automaton model, using absolute density and pitfall trap is being used to simulate diffusion dispersal of soil animals. A set of unusual weather conditions in late July produced a mass immigration of a small moth Plutella xylostella into Svalbard. This chance event has allowed us to track in detail the movement of associated weather systems and to reconstruct the direction and source of immigrants. Such events are rare but may become increasingly frequent as climate changes, opening a closed gateway for animals from further south to move into the Arctic. Continuing work Current visit (late July/early August) is aimed at collecting supporting information on the plant cover and microhabitat characteristics for manuscripts in preparation.