[Paper Review] The landscapemetrics and motif packages for measuring landscape patterns and processes

Jakub Nowosad, Maximilian H. K. Hesselbarth|arXiv (Cornell University)|May 10, 2024

Land Use and Ecosystem ServicesEnvironmental Science3 citations

TL;DR

This paper introduces the R packages landscapemetrics and motif for quantifying landscape patterns using categorical raster data, such as land use/land cover (LULC) maps. It demonstrates how landscape metrics and pattern-based spatial analysis enable researchers to compute, visualize, and compare spatial patterns across three European regions, with applications in ecological monitoring and pattern-process relationship studies.

ABSTRACT

This book chapter emphasizes the significance of categorical raster data in ecological studies, specifically land use or land cover (LULC) data, and highlights the pivotal role of landscape metrics and pattern-based spatial analysis in comprehending environmental patterns and their dynamics. It explores the usage of R packages, particularly landscapemetrics and motif, for quantifying and analyzing landscape patterns using LULC data from three distinct European regions. It showcases the computation, visualization, and comparison of landscape metrics, while also addressing additional features such as patch value extraction, sub-region sampling, and moving window computation. Furthermore, the chapter delves into the intricacies of pattern-based spatial analysis, explaining how spatial signatures are computed and how the motif package facilitates comparisons and clustering of landscape patterns. The chapter concludes by discussing the potential of customization and expansion of the presented tools.

Motivation & Objective

• To demonstrate the application of landscape metrics and pattern-based spatial analysis in ecological research using categorical raster data.
• To introduce the landscapemetrics R package for computing composition and configuration metrics across multiple spatial levels (patch, class, landscape).
• To present the motif R package for comparing, clustering, and detecting similar landscape patterns using spatial signatures.
• To provide practical workflows for metric computation, visualization, sub-region sampling, and moving window analysis in real-world European LULC datasets.
• To support open-source extensibility of both tools for custom metric and signature integration in ecological and spatial research.

Proposed method

• Utilizes the landscapemetrics R package to compute 45+ landscape metrics (e.g., patch density, contagion, aggregation) across patch, class, and landscape levels.
• Applies the motif package to compute spatial signatures based on local neighborhoods of cells, capturing both composition and configuration of LULC patterns.
• Employs hierarchical clustering and similarity measures (e.g., cosine distance) to group spatial units with similar landscape patterns.
• Uses reclassified LULC data (5 general classes: urban, agriculture, vegetation, marshes, water) from Copernicus 2018 for three European regions (France, Netherlands, Sweden).
• Implements moving window analysis and sub-region sampling to assess spatial heterogeneity and local pattern variation.
• Visualizes results using thematic maps and cluster diagrams to interpret spatial patterns and metric distributions.

Experimental results

Research questions

• RQ1 How can landscape metrics computed via the landscapemetrics package effectively quantify the composition and configuration of land use/land cover patterns in diverse European landscapes?
• RQ2 To what extent can the motif package detect and group spatially coherent landscape patterns using spatial signatures derived from local cell neighborhoods?
• RQ3 How do landscape metrics and pattern-based analysis contribute to understanding the pattern-process link in ecological systems?
• RQ4 Can the integration of moving window and sub-region sampling enhance the detection of local spatial variations in landscape structure?
• RQ5 What are the practical workflows for applying these tools to real-world LULC datasets in ecological and environmental research?

Key findings

• The landscapemetrics package successfully computes a comprehensive set of landscape metrics across multiple spatial levels, enabling detailed quantification of LULC pattern structure in the three European regions.
• The motif package effectively identifies spatial clusters of similar landscape patterns, with results showing coherent grouping of regions based on spatial signatures derived from local cell configurations.
• Moving window analysis reveals localized variations in landscape metrics, highlighting areas with high spatial heterogeneity or abrupt changes in land cover structure.
• Patch-level and class-level metric extraction enables targeted analysis of specific land cover types, such as urban or water bodies, across the study regions.
• The combination of landscapemetrics and motif allows for robust comparison of landscape patterns over time and across regions, supporting temporal and spatial ecological monitoring.
• The open-source nature of both packages facilitates customization, with support for user-defined metrics and spatial signatures, enhancing their adaptability to diverse research needs.