Rapid spatial risk modelling for invasion management under uncertainty

作者
Jens Froese,Grant S. Hamilton
标识
DOI:10.36334/modsim.2017.g8.froese
摘要

Across the globe, invasive alien plant species are a major threat to biodiversity and agricultural industries.Invasive plant spread is an ongoing process that plays out at different spatiotemporal scales.When an invasive 'weed' emerges in a new country, region or landscape, there is often a distinct lag phase during which population increase and invasive spread is slow.Rapid response by biosecurity managers to new incursions may bring the greatest opportunities with regard to facilitating eradication or effective containment as well as minimizing long-term control costs and negative impacts.An important consideration for allocating finite management resources is the spatial distribution of invasion risk.However, hard data is rarely available to assess risk, and decisions are consequently made under considerable uncertainty.We developed a flexible, spatially-explicit methodology for rapidly modelling invasion risk using Bayesian networks and Geographic Information Systems (GIS).In this well-established modelling framework, the key processes of invasion and the major risk variables affecting these processes are represented as a Bayesian network.GIS analysis is used to link each risk variable contained in the model to one or several spatial data proxies and to map model predictions.Here, we present a simplified implementation that can be used rapidly and under limited knowledge, using the case study of Mexican bean tree (Cecropia spp.) in North Queensland.We modelled two aspects of invasion risk: (a) the potential, time-independent environmental suitability of a site for plant establishment and persistence, and (b) the actual susceptibility of suitable sites to propagule introduction from source infestations recorded during annual 'detection periods'.We derived risk variables and spatial data proxies via desktop research and used simple mathematical functions to model the interplay between risk variables, invasion processes (establishment, persistence and propagule pressure), and invasion risk (suitability and susceptibility).We also evaluated the ability of our model to accurately predict the observed spatial progression of the Cecropia spp.invasion in North Queensland by validating mapped annual susceptibility against subsequently observed infestations.The majority of North Queensland's Wet Tropics was predicted to be potentially highly or moderately suitable for Cecropia spp.invasion.Actual susceptibility to invasion in a given fruiting season is more restricted due to dispersal constraints from source infestations.In each evaluated detection period, modelled susceptibility was a good predictor of actual Cecropia spp.detections in subsequent years.Annual susceptibility maps may be used for allocating surveillance and containment resources around existing infestations.Low-frequency long-distance dispersal or human-mediated translocations were not captured in the model.Further automation of risk factor / spatial proxy selection and data pre-processing may make our methodology for rapidly modelling invasion risk accessible to biosecurity agencies that typically have significant field knowledge and GIS capacities but may have limited expertise in applying complex modelling methods.

科研通智能强力驱动
Strongly Powered by AbleSci AI
科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
iitj发布了新的文献求助10
刚刚
余香的应助被yy采纳,获得10
刚刚
yuchuncheng完成签到,获得积分0
1秒前
gycao2025发布了新的文献求助10
1秒前
lzl007完成签到 ,获得积分0
3秒前
3秒前
Yolo完成签到 ,获得积分20
5秒前
特特特特特完成签到,获得积分10
5秒前
哈基米这一块完成签到,获得积分10
5秒前
Fly完成签到,获得积分10
5秒前
5秒前
胡杨树2006完成签到,获得积分10
6秒前
Fan完成签到 ,获得积分0
6秒前
椰子云的应助被邢一采纳,获得10
6秒前
冷cool完成签到,获得积分10
7秒前
Lucas完成签到,获得积分10
7秒前
7秒前
7秒前
dd完成签到,获得积分10
7秒前
毛啊毛完成签到,获得积分10
9秒前
9秒前
森林木发布了新的文献求助10
9秒前
qhebdb完成签到 ,获得积分10
10秒前
米豆完成签到 ,获得积分10
11秒前
123456qqqq发布了新的文献求助50
11秒前
dd发布了新的文献求助10
11秒前
11秒前
锦诗完成签到,获得积分10
12秒前
贪玩的水壶完成签到 ,获得积分10
12秒前
762323661完成签到,获得积分10
12秒前
wtian1221完成签到,获得积分10
13秒前
Nidhogg完成签到,获得积分10
13秒前
deer完成签到,获得积分10
13秒前
leave完成签到,获得积分10
14秒前
CarryLJR发布了新的文献求助10
14秒前
会盟发布了新的文献求助10
14秒前
14秒前
zs关闭了zs的文献求助
15秒前
DavidSun完成签到,获得积分10
15秒前
wwwww完成签到,获得积分10
15秒前
高分求助中
(应助此贴封号)【重要!!请各用户(尤其是新用户)详细阅读】【科研通的精品贴汇总】 10000
拉伸高分子熔体中缠结行为的模拟和小角中子散射研究 1500
The World Atlas of Wine 9th edition 1000
奈特基础组织学:配套组织病理学(第4版)(Netter's Essential Histology With Correlated Histopathology) 800
John McDowells Bildungstheorie 800
八旗制度の研究 600
Analyses de reconnaissances de dettes passées devant les échevins d'Ypres (1249-1291) 600
热门求助领域 (近24小时)
化学 材料科学 医学 生物 纳米技术 计算机科学 工程类 有机化学 化学工程 生物化学 物理 复合材料 内科学 催化作用 光电子学 人工智能 细胞生物学 无机化学 物理化学 数学
热门帖子
关注 科研通微信公众号,转发送积分 8159151
求助须知:如何正确求助?哪些是违规求助? 9554617
关于积分的说明 21719982
捐赠科研通 7748814
什么是DOI,文献DOI怎么找? 3427149
关于科研通互助平台的介绍 2552091
邀请新用户注册赠送积分活动 2471436