A review of models of landscape change

Abstract

Models of landscape change may serve a variety of purposes, from exploring the interaction of natural processes to evaluating proposed management treatments. These models can be categorized as either whole landscape models, distributional landscape models, or spatial landscape models, depending on the amount of detail included in the models. Distributional models, while widely used, exclude spatial detail important for most landscape ecological research. Spatial models require substantial data, now more readily available, via remote sensing, and more easily manipulated, in geographical information systems. In spite of these technical advances, spatial modelling is poorly developed, largely because landscape change itself is poorly understood.To facilitate further development of landscape models I suggest (1) empirical multivariate studies of landscape change, (2) modelling of individual landscape processes, (3) explicit study of the effect of model scale on model behavior, and (4) 'scaling-up' results of studies, on smaller land areas, that have landscape relevance.

Is research keeping up with changes in landscape policy? A review of the literature

“…Mediterranean Europe, for example, is heavily shaped not only by interactions with northern European neighbours but also by biogeographical, social and economic flows across the Mediterranean Basin to the North African shore (Lutterbeck, 2006;Ribas-Mateos, 2005). Perhaps in due course the ELC should be expanded to an international level, given that landscape change is also a product of global change (Baker, 1989;Lavorel et al, 1998;Thompson et al, 1998); indeed, proposals for a Global Landscape Convention have already been put forward (Ibrahim, 2010).…”

Modeling the relationships between land use and land cover on private lands in the Upper Midwest, USA

“…Models of land-use and land-cover change have been developed to address when, where and why land-use and land-cover change occurs (for good summaries of these models see Baker, 1989;Riebsame et al, 1994;Lambin, 1997;Theobald and Hobbs, 1998). They usually involve empirically fitting the models to some historical pattern of change, then extending those patterns into the future for prediction.…”

“…that it is constant in both time and space; (2) the assumption of spatial independence of transitions; and (3) the difficulty of ascribing causality within the model, i.e. the transition probabilities are often derived empirically from multi-temporal maps with no description of the process (Baker, 1989). This third limitation is particularly acute when land-cover changes are under investigation, for example, from remotely sensed imagery, and when those changes are driven by social and economic processes (Turner, 1987).…”

“…This third limitation is particularly acute when land-cover changes are under investigation, for example, from remotely sensed imagery, and when those changes are driven by social and economic processes (Turner, 1987). To address limitations 1 and 3 above, Baker (1989) suggested setting state transition probabilities as a function of exogenous or endogenous variables, which vary in space and time. Equation (1) then becomes…”

Markov chain models for vegetation dynamics

“…Transition probabilities are modelled as linearly dependent on 126 ecological factors by some authors (Usher, 1987;Baker, 1989). If the time intervals 127 between observations are not equal, specific estimation techniques are available 128 (Frydman, 1992;1995a;1995b).…”