The use of local indicators of spatial association to improve LiDAR-derived predictions of potential amphibian breeding ponds

James Thomas Julian, John A. Young, John W. Jones, Craig D. Snyder, C. Wayne Wright

Research output: Contribution to journalArticle

6 Scopus citations

Abstract

We examined whether spatially explicit information improved models that use LiDAR return signal intensity to discriminate in-pond habitat from terrestrial habitat at 24 amphibian breeding ponds. The addition of Local Indicators of Spatial Association (LISA) to LiDAR return intensity data significantly improved predictive models at all ponds, reduced residual error by as much as 74%, and appeared to improve models by reducing classification errors associated with types of in-pond vegetation. We conclude that LISA statistics can help maximize the information content that can be extracted from time resolved LiDAR return data in models that predict the occurrence of small, seasonal ponds.

Original languageEnglish (US)
Pages (from-to)89-106
Number of pages18
JournalJournal of Geographical Systems
Volume11
Issue number1
DOIs
StatePublished - Jan 1 2009

All Science Journal Classification (ASJC) codes

  • Geography, Planning and Development
  • Earth-Surface Processes

Fingerprint Dive into the research topics of 'The use of local indicators of spatial association to improve LiDAR-derived predictions of potential amphibian breeding ponds'. Together they form a unique fingerprint.

  • Cite this