Integrative analysis of ecological patterns in an untended temperate woodland utilising standard and customised software

Abstract

The integration of a comprehensive monitoring programme conducted in a temperate woodland through statistical analysis and simulation modelling is presented here. The Heron Wood Reserve in Peebleshire, Scotland is a 7.5-ha wood, left untended to facilitate investigation into a natural Scottish Woodland Ecosystem. Data on fungal fruiting have been collected since 1994, and since November 2000 physico-chemical and biotic experiments on soil and woodland litter have also been conducted with the aim to construct and integrate a number of simulation submodels of the various ecological systems present in an untended Scottish woodland. Important systems for modelling include patterns of fungal succession, habitat characteristics and fungal biochemical patterns. A large volume of data has been collected so computer software is used for the storage and handling of data. Microsoft Access was the database used for storage; integration of functions available in Microsoft Excel with a customised software written in Microsoft Visual Basic and Microsoft VBA allowed various statistical modelling tests (e.g. Scheffé analysis of contrast) to be carried out, and the quick display of data in a suitable manner. Model Maker software is used to construct and test simulation models formulated from the data and ultimately to facilitate the running of various simulations. Integration of the separate lines of investigation is demonstrated on the example of a case study describing differences in the dynamics of decomposers in beech- and birch-dominated habitats. Integration of the information obtained allowed us to put forward the following hypothesis. The leaf litter produced by beech trees is of lower quality compared to birch, and decomposition in beech-dominated habitats is delayed (i.e. in comparison to birch-dominated) because of the differences in the extent to which these plants create and maintain a litter and humus layer (sensu Wardle, Communities and Ecosystems. Linking the Aboveground and Belowground Components, Princeton University Press, Princeton, 2002). Hence, the decomposition in beech-dominated habitats mainly occurs somewhat deeper in the soil profile, and may be carried out with some considerable involvement of ectomycorrhizal fungi. The correct understanding of the site-specific peculiarities of the ecological patterns is important for the improvement of biodiversity of woodland and forest ecosystems both in Scotland and worldwide.

Introduction

Complex integrative modelling studies help to enhance our knowledge of natural environment and are, therefore, an important prerequisite for sustainable management of ecosystems. Here we present an overview of a Scottish temperate woodland ecosystem study, integrating a comprehensive monitoring programme with the examination of data obtained through analysis of a database, and statistical and simulation modelling.

Half of Britain’s forest and woodland area is located in Scotland with forest and woodland covering an area of 1.2 million ha—15% of Scotland’s land area. Over 17,000 ha of land are planted with trees every year to replace those felled and to establish new woodlands, while employment in the forestry and primary wood processing sectors exceeds 10,000. Around 40% of Scotland’s woodlands are managed for recreation and include important conservation habitats such as the Caledonian Forest Reserves (Forestry Commission, http://www.forestry.gov.uk).

Analyses of biological interaction patterns and modelling of key ecosystem dynamics within an untended woodland may be used to improve management, forecasting, sustainability and conservation of Scotland’s woodlands. To obtain suitable and substantial data required for such analyses, a number of ecosystem variables and processes are under investigation, including, e.g., bacteria and fungal dynamics, decomposition of plant material, abiotic factors such as pH and soil composition, fungal fruiting and local weather conditions.

To analyse the dynamics of a complex ecosystem, the information obtained in different lines of investigation may be integrated by means of statistical and simulation modelling. Particularly useful in this respect may be a multiple simultaneous application of the Scheffé analysis of contrast presented here. In particular, Scheffé’s analysis of contrasts is used to compare site plots, which can be characterised using many different attributes. Scheffé’s analysis of contrasts is a very flexible technique that can be used in a variety of investigations (e.g. see Fernie and Staines, 2001, where this technique was used to form a taxonomy of European grocery markets). However, it is not widely used in ecology, perhaps because it is not available in the most commonly used software (e.g. Excel, Minitab). For example, in one application found (Danielson and Visser, 1989) this technique was used for pairwise comparisons only when Tukey’s method would have been more powerful. The software developed in this investigation allows ad hoc, and a library of regularly used contrasts to be tested very efficiently.

Standard techniques used to analyse samples have been adapted and may be beneficial in future woodland ecosystem studies. Computer software has been utilised to integrate the information obtained throughout this research, from data storage, calculations for different dynamic systems, displaying of data, statistical calculation and ultimately constructing simulation models. These may all be adapted for studies of other temperate woodlands.