Coupling Landscape Patterns and Ecological Processes - Landscape Ecology for Sustainable Environment and Culture - page 3

1.1 Introduction
As an interdisciplinary science linking ecology and geography, landscape ecology
has been applied widely for its theories and approaches. The link between land-
scape patterns and ecological processes forms the foundation of landscape ecol-
ogy, understanding the relationship between them is key to further promoting the
study of landscape ecology (Fortin and Agrawal 2005 ; Fu et al. 2001 ;Wu 2007 ;
Wu and Hobbs 2002 ). Landscape pattern generally refers to the spatial structural
characteristics of a landscape, i.e., the spatial arrangement and deployment of
landscape units in various sizes and shapes. Ecological processes relate to the flow
and transfer of matter, energy, water, biota, and information within or among
ecosystems, with special emphasis on the dynamic characteristics of the occur-
rence and development of the ecological events or phenomena. Landscape patterns
are associated closely with ecological processes; the interactions between them
demonstrate some ecological functions tinted by scale dependence. Theoretically,
landscape patterns and ecological processes are inseparable. To make the research
easier, some researchers focus on the features of landscape patterns, while others
focus on the dynamics of ecological processes.
Spatial statistical analysis, landscape pattern metrics, and dynamic models
constitute the three primary methods for analyzing landscape patterns. As a basic
method, spatial statistical analysis explores the areas/percentages and geometries
of different landscapes and their variations at different stages (Odland 1988 ; Tobler
1970 ). Landscape pattern metrics are designed to quantify the landscape pattern by
analyzing landscape structural composition and spatial configuration. Dynamic
models of landscape pattern consist primarily of spatial Markov (Aaviksoo 1995 ;
Li 1995 ), cellular automata (Wolfram 1984 ;Wu 2002 ), and agent-based models
(Bithell and Brasington 2009 ; Ligtenberg et al. 2004 ; Matthews et al. 2007 ).
Landscape pattern metrics are used widely in analyzing landscape spatial
configurations (Haines-Young and Chopping 1996 ). To date, a variety of indices
have been developed, including patch area, edge/shape, proximity, diversity and
lacunarity indices, aggregation, and location-weighted landscape contrast metrics
(Chen et al. 2008 ). Unfortunately, most existing landscape metrics fail to capture
the fundamental patterns that can be used to identify or explain ecosystem pro-
cesses, due to lack of ecological implications; e.g., landscape metrics dealing with
land cover composition and connectivity/fragmentation fail to capture the linear
features or the hub-and-corridor spatial patterns (Jones et al. 2012 ). Current
landscape analyses based on pattern metrics do not go any further than analyzing
the geometrical characteristics of the land cover pattern, let alone linking them
with relevant ecological processes (Chen et al. 2008 ; Gustafson 1998 ; Lü et al.
2007 ). Thus, landscape pattern models are stagnating in defining the land cover
transfer rules and dynamic simulations by analyzing landscape transfer probability.
The absence of ecological implications makes it difficult to explain the underlying
mechanism of the landscape pattern change.