[article]
| Titre : |
Rapid spatial risk modelling for management of early weed invasions: Balancing ecological complexity and operational needs |
| Type de document : |
Électronique |
| Auteurs : |
Jens G. Froese ; Alan R. Pearse ; Grant Hamilton |
| Année de publication : |
2019 |
| Article en page(s) : |
2105-2117 |
| Langues : |
Anglais (eng) |
| Catégories : |
[CBNPMP-Thématique] Plantes subspontanées, naturalisées, envahissantes
|
| Résumé : |
1. When an invasive alien ‘weed’ emerges in a previously uninhabited landscape, land managers must respond quickly to facilitate effective eradication or containment, and minimize long‐term negative impacts. However, on‐ground management decisions are often made under time, knowledge and capacity constraints. Spatially explicit tools for assessing invasion risk rapidly and under uncertainty would help land managers to better target interventions.
2. We developed a generic methodology that integrates (a) interactions between ecological risk factors and invasion processes affecting both the potential suitability for population growth and the actual susceptibility to propagule introduction from source populations with (b) spatially explicit data in (c) a probabilistic Bayesian network modelling framework. Our methods focused on the operational needs of land mangers responding to weed incursions, streamlining data and knowledge collection, simplifying model calibration, and facilitating adoption via a collection of user‐friendly web apps called riskmapr.
3. We tested the generality of our methodology on two contrasting weeds (the rainforest tree Cecropia spp. and the cactus Cylindropuntia rosea) that are targeted for containment and local eradication in Queensland, Australia. Case study models were calibrated from published knowledge about abiotic and biotic factors affecting suitability for, and susceptibility to, weed invasion. Validation of annual risk maps against weed detections in subsequent years showed that models accurately predicted the field‐observed progression of each invasion to date.
4. We developed a rapid spatial risk modelling methodology that is theoretically comprehensive and practically simple. Our streamlined methods and open access implementation using riskmapr facilitate adoption by land managers. Models and risk maps can be used to target interventions or improve spatially explicit understanding of the risk factors and processes driving early weed invasions. |
| Lien pérenne : |
DOI : 10.1111/2041-210X.13284  |
| Permalink : |
https://biblio.cbnpmp.fr/index.php?lvl=notice_display&id=147660 |
in Methods in ecology and evolution > 10 (2019) . - 2105-2117
Froese, Jens G., Pearse, Alan R., Hamilton, Grant
2019
Rapid spatial risk modelling for management of early weed invasions: Balancing ecological complexity and operational needs.
Methods in ecology and evolution, 10: 2105-2117.
|
Rapid spatial risk modelling for management of early weed invasions: Balancing ecological complexity and operational needs in Methods in ecology and evolution, 10 (2019)
