An integrated risk-assessment framework for multiple threats to floodplain values in the Kakadu Region, Australia, under a changing climate

Abstract

The internationally important river–floodplains of the Kakadu Region in northern Australia are at risk from invasive species and future sea-level rise–saltwater inundation (SLR–SWI), requiring assessments of multiple cumulative risks over different time frames. An integrated risk-assessment framework was developed to assess threats from feral animals and aquatic weeds at three SLR-scenario time frames (present-day, 2070 and 2100) to natural (magpie goose habitats), cultural (indigenous hunting–fishing sites) and economic (tourism revenue less invasive species control costs) values. Probability density functions (pdfs) were fitted to spatial data to characterise values and threats, and combined with Monte Carlo simulation and sensitivity analyses to account for uncertainties. All risks were integrated in a Bayesian belief network to undertake ‘what if’ management-scenario analyses, and incorporated known ecological interactions and uncertainties. Coastal landscapes and socio-ecological systems in the region will be very different by 2100 as a result of SLR; freshwater ecosystems will transform to marine-dominated ecosystems and cannot be managed back to analogue conditions. In this context, future invasive-species risks will decrease, reflecting substantial loss of freshwater habitats previously at risk and a reduction in the extent of invasive species, highlighting the importance of freshwater refugia for the survival of iconic species.

A framework for complex climate change risk assessment

“…For instance, low-income workers are often employed outdoors and live in poorly ventilated housing, spend a greater portion of their income on healthcare, and lose relatively more from missing a day of work, all making them more vulnerable and exposed to morbidity and mortality from heat waves. 52 Although integrating quantitative and qualitative knowledge of interactions between physical, ecological, and social systems remains challenging, knowledge co-production approaches to complex risk assessment that use integrated risk assessment models, 53,54 storylines, and scenario planning can highlight interactions across system boundaries that generate risk not evident from more conventional climate impact projections. 31,55,56 Second, responses to risk are often excluded as drivers of risk even though they play a key role in driving potential outcomes, including inaction, and are well recognized in financial and policy domains.…”

Kakadu's wetlands: more change is afoot

“…Using a fairly conservative approach based on a 25% frequency of sea-water (over 2 PSU) inundation over a monthly tidal cycle, the findings suggests that although freshwater habitat loss will be marginal (,4%) by 2030, there is a high probability of a threshold effect potentially resulting in an estimated loss of 42% by 2070, increasing to 78% by 2100. The findings of Bayliss et al (2018a) also suggest that the extent of loss varies among the floodplains, reaching 80% by 2100 in the most pronounced case. However, it should be noted that the model relies on a 'bathtub-like' SLR effect, meaning that the variable hydraulic friction of the floodplain was not directly incorporated into the model.…”

“…For KNP's Indigenous communities, this is likely to include the loss of hunting and fishing areas, and sites of cultural significance. The decline in tourism will also have financial implications for tourism operators, and the revenue used to support the Park's management activities (Bayliss et al 2018a;Dutra et al 2018).…”

“…maintaining water chestnuts and the Magpie Geese colonies they support), presently, such events occur 6-12 times per year in environments that are also subjected to seasonal and often sustained freshwater inundation events (Woodroffe et al 1986;Bayliss et al 2018a). However, the proposed future scenarios suggest a very different pattern in the timings between saltwater and freshwater exposure, ultimately resulting in increasingly sustained periods of seawater exposure, creating conditions that exceed the physiological limits of a large proportion of the biota (Nelson et al 2015Humphrey et al 2018;Pettit et al 2018;Stephenson et al 2018).…”

Multi-risk assessment of agrometeorological disasters based on a three-level hierarchical framework

“…法正在为越来越多的人所熟知。Kloos 等 [28] 提出基于系统动力学的多风险评估框架，在系 统动力学的基础上，考虑到多种灾害之间的相互作用，并考虑到脆弱性水平的相互作 用。Bayliss 等 [29] 将风险和风险管理概念结合在一起，描述了在灾害层面上环境与社会系 统之间的相互作用和耦合过程。Bathrellos [30] 基于多标准分析和 GIS 生成了多灾种地图， 用来分析不同来源的数据，开发灾害分析模型以及估算研究区域的脆弱性和风险。尽管 上述框架与模型方法都可以运用到多风险评估中来，但是由于未能定量分析灾害间的相 互作用过程、触发级联效应以及耦合机制，始终不能清晰把握多灾害之间的相互作用。 因此，本文通过联合分析与量化所有影响承灾体的气象风险因素，比较不同来源的 灾害，对灾害间的级联效应与耦合机制进行定量分析，对于制定风险管理战略与农户适 应性管理对策具有重大意义。文中运用多风险评估框架探讨了农业气象灾害的特征及变 化趋势，结果表明：陕西省洪涝的灾害率和灾害强度呈显著下降趋势，冷冻的灾害率呈显 著上升趋势，干旱和风雹的灾害率与灾害强度无显著变化趋势。张丽艳等 [31] 研究表明在 引起陕西省洪涝灾害发生的所有因素中，较重要的降水量因素整体呈现出下降趋势；屈 振江等 [32] 认为陕西省部分地区处于花期冻害的易发区，冻害发生频率相对较高。本文对 四种灾害的特征分析很好地佐证了前人研究结论。此外，研究发现干旱和洪涝灾害对于其 他灾害的影响显著。这与于小兵等提出的旱涝灾害对生态环境影响巨大的结论相符 [33] 。 在多风险评估部分中，对实地调研数据进行分析后可以发现霜冻和干旱灾害对于农 业生产的巨大影响，这是由于陕北和关中地区近些年来旱灾和霜冻频繁发生，农户未能 表 3 BHM 模型结果 [34] 。多重脆弱 性的分析表明在各类作物中，仅仅只有大豆对洪涝与干旱灾害的脆弱性受到选定生计资 本的显著影响。这与杨浩等 [35] 关于提高教育水平、劳动人数和固定资产等生计资本可以 减少农户因气象灾害而带来的脆弱性不符，可能的原因是，旱涝灾害对农业生产的影响 巨大，而仅仅增加部分生计资本很难降低整体脆弱性。然而，粮食、小麦、谷子和大豆 对风雹与冷冻的脆弱性却受到选定生计资本的显著影响。风雹和冷冻灾害尽管对作物的 影响很大，但由于各种防冻、防雹措施的开发与应用，鼓励农户采用新技术并适当提高 生计资本能够很好地降低脆弱性。孙特生等 [36] 研究表明，针对性地来提高农户的生计资 本，能够避免或减轻灾害所造成的损失。此外，根据不同生计资本对脆弱性的显著影 响，应尽可能地增加有效灌溉面积率、农作物种植面积、农村用电量和农用磷肥施用折 纯量等指标来降低脆弱性。 本文探讨农业多风险评估的现实意义在于：第一，多风险评估是进行多风险管理的 首要前提，对于进行更详细的风险分析和策划农业气象灾害的防治方案至关重要；第 二，多风险评估能够定量分析灾害间的相互作用关系，为更加有针对性的综合治理提供 依据；第三，多风险评估考虑到脆弱性在不同受灾要素中的动态变化，为制定适应性管 理措施提供参考。此外，本文提出的三级分层多风险评估框架为一些基于指标或过程的 灾害风险评估和管理工作指出了实践方向。 参考文献(References):…”