2021
DOI: 10.1111/ddi.13303
|View full text |Cite
|
Sign up to set email alerts
|

Predicting current and future global distribution of invasive Ligustrum lucidum W.T. Aiton: Assessing emerging risks to biodiversity hotspots

Abstract: This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
2

Citation Types

0
11
0

Year Published

2022
2022
2024
2024

Publication Types

Select...
9

Relationship

0
9

Authors

Journals

citations
Cited by 13 publications
(11 citation statements)
references
References 86 publications
0
11
0
Order By: Relevance
“…For each projected pixel, the MOP analysis allows for restricting the portion of the training data used to compute the Euclidean distance (Owens et al, 2013). Within species accessible area, we computed environmental similarity between each projected pixel and the nearest 10% of training data points (Montti et al, 2021), and then filtered habitat suitability estimates for projected pixels showing very high (MOP values ≥0.9), high (MOP ≥0.8) and moderate (MOP ≥0.7) environmental similarity with the training data. To minimise issues with unlimited dispersal, we restricted all projections to the respective calibration area defined for each species.…”
Section: Methodsmentioning
confidence: 99%
“…For each projected pixel, the MOP analysis allows for restricting the portion of the training data used to compute the Euclidean distance (Owens et al, 2013). Within species accessible area, we computed environmental similarity between each projected pixel and the nearest 10% of training data points (Montti et al, 2021), and then filtered habitat suitability estimates for projected pixels showing very high (MOP values ≥0.9), high (MOP ≥0.8) and moderate (MOP ≥0.7) environmental similarity with the training data. To minimise issues with unlimited dispersal, we restricted all projections to the respective calibration area defined for each species.…”
Section: Methodsmentioning
confidence: 99%
“…To account for the impact of model extrapolation on each species projection, we computed the mobility‐oriented parity (MOP) metric (Owens et al, 2013) within the calibration area of each species. We calculated the MOP metric by measuring the Euclidean distance between environmental conditions of the projected pixel and the nearest 10% training data observations (Montti et al, 2021). The MOP metric was further normalized to 1 and subtracted from 1 to reflect environmental similarity (Owens et al, 2013).…”
Section: Methodsmentioning
confidence: 99%
“…The final models were generated by ensemble using the average of the best algorithms (those with Jaccard over the average for a single model) for current and future conditions, avoiding the inclusion of poor predictions in the final model. Ensemble models were binarized using a threshold that maximised the Jaccard metric (0.545), which showed the best values of sensitivity (0.984) and specificity (0.991) (Montti et al, 2021; Velazco et al, 2019) (Supplementary file S3). The ensemble is a commonly used method to deal with uncertainty caused by different algorithms (Diniz‐Filho et al, 2009; Thuiller et al, 2019).…”
Section: Methodsmentioning
confidence: 99%