Abstract
Even with considerable attention in recent decades, measuring and working with patterns remains a complex task due to the underlying dynamic processes that form these patterns, the influence of scales, and the many further implications stemming from their representation. This work scrutinizes binary classes mapped onto regular grids and counts the relative frequencies of all first-order configuration components and then converts these measurements into empirical probabilities of occurrence for either of the two landscape classes. The approach takes into consideration configuration explicitly and composition implicitly (in a common framework), while the construction of a frequency distribution provides a generic model of landscape structure that can be used to simulate structurally similar landscapes or to compare divergence from other landscapes. The technique is first tested on simulated data to characterize a continuum of landscapes across a range of spatial autocorrelations and relative compositions. Subsequent assessments of boundary prominence are explored, where outcomes are known a priori, to demonstrate the utility of this novel method. For a binary map on a regular grid, there are 32 possible configurations of first-order orthogonal neighbours. The goal is to develop a workflow that permits patterns to be characterized in this way and to offer an approach that identifies how relatively divergent observed patterns are, using the well-known Kullback–Leibler divergence.
Highlights
Satellites are continually imaging the surface of the Earth and producing representations of real landscapes
It has previously been extensively documented that composition and configuration are the dominant descriptors of a spatial pattern [14,27]; while exceptions exist [28], these attributes are generally described by different metrics and in differing units of measurement
This study provides a means by which binary landscapes can be summarized by their 32 most primitive pattern elements
Summary
Satellites are continually imaging the surface of the Earth and producing representations of real landscapes These representations have characteristics based on sensor optics and processing decisions that were formally incorporated into the design of the imaging systems. At this stage, the infinite complexity of any imaged landscape is generalized, and subsequent classification of these images into categorical representations even further simplifies the reality of these landscapes through depiction. The exact characteristics of these categorical maps will be determined by numerous decisions leading to their construction and the limitations of the systems used to acquire and process the data Regardless, such landscape maps and representations of reality are at the core of many landscape ecological studies (e.g., [1,2,3])
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
