Publication Date:
2024-05-18
Description:
Climate change and land cover change often interactively affect plant species distributions. This study addresses
the vulnerability of lowland and upland orchids to climate change and land cover change. Endemic orchids of
New Guinea were grouped into four classes (lowland epiphyte, lowland terrestrial, upland epiphyte, upland
terrestrial) based on their life form and elevation range. Forty occurrence records of endemic orchids were
selected for each class, totaling 160 occurrence records. Ensemble modelling combining two machine learning
algorithms was used to generate predictive current and future suitable areas for orchid classes. Model performance was evaluated using the AUC and TSS metrics. Suitable areas for both lowland and upland orchids
(epiphyte and terrestrial) were predicted decrease in the future due to climate change and land cover change.
The loss of suitable areas for upland terrestrial orchids was predicted to be most significant in the worst-case
climate change scenario (SSP 5–8.5). Both lowland and upland orchids (epiphyte and terrestrial) tend to shift
to higher elevation ranges from the present distributions. The predictive models have AUC values 〉0.90 and TSS
value 〉0.80, indicating the models have excellent potential for predicting the impact of climate change and land
cover change on orchid distributions.
Keywords:
Ensemble model
;
Climate change
;
Species distribution model
;
Orchids
;
Lowland
;
Upland
;
New Guinea
Repository Name:
National Museum of Natural History, Netherlands
Type:
info:eu-repo/semantics/article
Format:
application/pdf
Permalink
