Presentation Open Access

Describing landscapes by statistics of local pattern features: application to landscape regionalization, change, and search

Jakub Nowosad; Tomasz Stepinski

MARC21 XML Export

<?xml version='1.0' encoding='UTF-8'?>
<record xmlns="">
  <datafield tag="653" ind1=" " ind2=" ">
    <subfield code="a">local pattern features</subfield>
  <datafield tag="653" ind1=" " ind2=" ">
    <subfield code="a">pattern similarity</subfield>
  <datafield tag="653" ind1=" " ind2=" ">
    <subfield code="a">landscape regionalization</subfield>
  <datafield tag="653" ind1=" " ind2=" ">
    <subfield code="a">change detection</subfield>
  <datafield tag="653" ind1=" " ind2=" ">
    <subfield code="a">landscape search</subfield>
  <controlfield tag="005">20200120150453.0</controlfield>
  <controlfield tag="001">1209418</controlfield>
  <datafield tag="700" ind1=" " ind2=" ">
    <subfield code="a">Tomasz Stepinski</subfield>
  <datafield tag="856" ind1="4" ind2=" ">
    <subfield code="s">3416470</subfield>
    <subfield code="z">md5:873d31faa5e150563e041fc86e9dd0f0</subfield>
    <subfield code="u"></subfield>
  <datafield tag="542" ind1=" " ind2=" ">
    <subfield code="l">open</subfield>
  <datafield tag="260" ind1=" " ind2=" ">
    <subfield code="c">2018-03-28</subfield>
  <datafield tag="909" ind1="C" ind2="O">
    <subfield code="p">openaire</subfield>
    <subfield code="o"></subfield>
  <datafield tag="100" ind1=" " ind2=" ">
    <subfield code="0">(orcid)0000-0002-1057-3721</subfield>
    <subfield code="a">Jakub Nowosad</subfield>
  <datafield tag="245" ind1=" " ind2=" ">
    <subfield code="a">Describing landscapes by statistics of local pattern features:  application to landscape regionalization, change, and search</subfield>
  <datafield tag="540" ind1=" " ind2=" ">
    <subfield code="u"></subfield>
    <subfield code="a">Creative Commons Attribution Share Alike 4.0 International</subfield>
  <datafield tag="650" ind1="1" ind2="7">
    <subfield code="a">cc-by</subfield>
    <subfield code="2"></subfield>
  <datafield tag="520" ind1=" " ind2=" ">
    <subfield code="a">&lt;p&gt;Landscape metrics quantify various aspects of a pattern and are well-suited for comparing landscapes from a focused perspective, for example, a landscape fragmentation.&lt;br&gt;
However they are ill-suited for an overall assessment of similarity between landscapes because it is unclear how to combine semantically different metrics into a single measure of patterns congruity. &amp;nbsp;&lt;/p&gt;

&lt;p&gt;To perform analyses based on an overall similarity between landscapes, we represent their patterns by a histogram of co-occurrence pattern features. &amp;nbsp;&lt;br&gt;
A co-occurrence feature is a pair of classes assigned to two neighboring raster cells.&lt;br&gt;
For a landscape with k classes the histogram has (k^2+k)/2 bins which together quantify composition and configuration of the pattern.&lt;br&gt;
Note that all bins have the same semantic meaning &amp;ndash; they are counts of co-occurrence features &amp;ndash; so there is no problem of how to combine them to assess an overall similarity between two landscapes.&lt;br&gt;
For such assessment we use the Jensen-Shannon divergence (JSD) between two co-occurrence histograms.&lt;br&gt;
The result is a single number between 0 (patterns are identical) and 1 (none of the classes found in one landscape is found in the other).&lt;/p&gt;

&lt;p&gt;Similarity-based analysis allows working with patterns in large (regional, continental, global) datasets.&lt;br&gt;
Possible tasks include: pattern regionalization, change detection, and search.&lt;br&gt;
We have developed an open-source software GeoPAT 2.0 (Geospatial Pattern Analysis Toolbox), downloadable from&amp;nbsp;, which performs those tasks.&lt;/p&gt;

&lt;p&gt;Presented examples include global delineation of ecophysiographic land units (utilizes patterns of land cover, land forms, soils, and climate), global assessment of local landscape change between 1992 and 2015 (utilizes CCI-LC global land cover data), and landscape search over the US (utilizes NLCD).&lt;br&gt;
Landscape search is illustrated by an online interactive application ( where user selects a landscape and the app produces a map of US indicating locations with an overall similar landscape.&lt;/p&gt;</subfield>
  <datafield tag="773" ind1=" " ind2=" ">
    <subfield code="n">doi</subfield>
    <subfield code="i">isVersionOf</subfield>
    <subfield code="a">10.5281/zenodo.1209417</subfield>
  <datafield tag="024" ind1=" " ind2=" ">
    <subfield code="a">10.5281/zenodo.1209418</subfield>
    <subfield code="2">doi</subfield>
  <datafield tag="980" ind1=" " ind2=" ">
    <subfield code="a">presentation</subfield>
All versions This version
Views 306306
Downloads 143143
Data volume 488.6 MB488.6 MB
Unique views 294294
Unique downloads 126126


Cite as