Evaluating multiple bioclimatic risks using Bayesian belief network to support urban tree management under climate change

Yoonjung Kim, Chan Park, Kyung Ah Koo, Myung Kyoon Lee, Dong Kun Lee*

*Corresponding author for this work

Research output: Contribution to journalJournal articleResearchpeer-review

Abstract

Understanding the vulnerability of trees affected by climate change is a key requirement for identifying management priorities and suggesting suitable urban tree species. To measure such vulnerability under changing climate conditions, indicators of bioclimatic characteristics should be identified and evaluated using past and current geographic growth ranges. However, although climate events often occur simultaneously (e.g., frost and drought), and management issues in this regard need to be clarified, it is challenging to consider multiple risks in a climate change vulnerability assessment. Therefore, we applied a Bayesian belief network (BBN) to interlink the bioclimatic requirements of species and seasonal climate risk of the study site to comprehensively assess the multiple risks. In particular, we integrated expert knowledge and supporting evidences from relevant studies to construct the BBN. The developed BBN revealed vulnerability to frost considering occurrences of cascading and co-occurring climatic risks such as warmer winters and droughts throughout the phenological cycle. As a case study, two tree species, Zelkova serrata and Camellia japonica from Seoul, Republic of Korea, were evaluated. Among the climatic risks considered, the BBN revealed that shortened frost hardening and the occurrence of spring frost right after an extraordinarily warm winter would mainly affect vulnerability to frost of the two species. In particular, C. japonica had high vulnerability due to its high susceptibility to coldness, though growing temperature will be perfectly satisfied under climate change. Generally, this study provides insights to consider multiple bioclimatic risks for guiding urban tree management under climate change.

Original languageEnglish
Article number126354
JournalUrban Forestry & Urban Greening
Volume43
Number of pages9
ISSN1618-8667
DOIs
Publication statusPublished - 2019

Keywords

  • Bayesian belief network
  • Bayesian network
  • Climate change vulnerability assessment
  • Multiple bioclimatic risk
  • Seasonal climate impact
  • Urban tree management

Cite this