Modelling and monitoring: examining the utility of dynamic landscape metapopulation models for sustainable forest management

Abstract Reliable, practical and affordable means of assessing the sustainability of forest management remain elusive. Monitoring of biological indicators is an important element, but sufficiently powerful monitoring strategies are expensive and monitoring alone may not provide answers in time to avloid irreversible environmental or ecological damage. We propose a model-based approach to assessing sustainability using bio-indicators of ecosystem condition to provide timely feed-back to managers about the sustainability of current and alternative forest management options, but acknowledge that assessment of biological indicators is only one component of the assessment of sustainability. This model-based approach can also support the development of better-targeted and more relevant monitoring systems. Dynamic landscape meta-population (DLMP) models integrate spatial models of forest change (also known as landscape dynamic models or forest succession models) with meta-population models. which describe demographic and biological attributes of species and the dynamic consequences of migration and habitat change. DLMP models may be used to predict the meta-population level consequences of management actions on bio-indicator species, and as such. may provide a relatively efficient and inexpensive approach to assessing an important aspect of the potential sustainability of forest management. We review some of the criticisms of monitoring for trend in indicators and the advantages and problems associated with using DLMP models of bio-indicator species to evaluate the sustainability of forest management options. We draw on results of a case study of the brown creeper (Certhis americana) in northern Ontario, Canada, that explored the sustainability of competing forest management scenarios. Based on case study results, DLMP models of bio-indicator species appear to be useful for (1) assessing and ranking the sustainability of management options, (2) quantifying the stresses placed on ecosystems by particular management activities, (3) targeting future research and data collection, and (4) dealing explicitly with both environmental and model uncertainties.