Modeling the Dynamics of Community Resilience to Coastal Hazards Using a Bayesian Network

Abstract

Studies on how variables of community resilience to natural hazards interact as a system that affects the final resilience (i.e., their dynamical linkages) have rarely been conducted. Bayesian network (BN), which represents the interdependencies among variables in a graph while expressing the uncertainty in the form of probability distributions, offers an effective way to investigate the interactions among different resilience components and addresses the natural–human system as a whole. This article employs a BN to study the interdependencies of ten resilience variables and population change in the Lower Mississippi River Basin (LMRB) at the census block group scale. A genetic algorithm was used to identify an optimal BN where population change, a cumulative resilience indicator, was the target variable. The genetic algorithm yielded an optimized BN model with a cross-validation accuracy of 67 percent over a period of 906 generations. Six variables were found to have direct impacts on population change, including level of threat from coastal hazards, hazard damage, distance to coastline, employment rate, percentage of housing units built before 1970, and percentage of households with a female householder. The remaining four variables were indirect variables, including percentage agriculture land, percentage flood zone area, percentage owner-occupied house units, and population density. Each variable has a conditional probability table so that its impacts on the probability of population change can be evaluated as it propagates through the network. These probabilities could be used for scenario modeling to help inform policies to reduce vulnerability and enhance disaster resilience.

有关社区面对自然灾害的回復力变因, 如何作为一个系统相互作用(例如其间的动态连结)并影响最终回復力之研究仍为数甚少。贝叶斯网络 (BN) 以图表呈现变因间的相互依赖性, 并以概率分布的形式表达不确定性, 提供了探讨不同回復力组成元素之间的互动的有效方法, 并将自然—人类系统视为一个整体。本文运用 BN, 研究十大回復力变因与密西西比下游流域 (LMBR) 人口调查街廓群体尺度中的人口变迁之间的相互依赖性。本研究运用基因演算法来指认最适 BN, 其中人口变迁此一积累回復力指标是针对的变因。基因演算法生产出最适的 BN 模型, 并在九百零六个生产週期中, 具有百分之六十七的交叉验证准确率。本研究发现六大变因对人口变迁具有直接影响, 包括沿海灾害的威胁程度、灾害损失、离海岸线的距离、就业率、1970 年前建造的房屋比率, 以及女性家户长的比率。其馀的四大变因是间接变因, 包括农业用地比率、粮食区域比率、由所有人居住的房屋单位比率, 以及人口密度。每项变因各自具有条件式概率表, 因此可对这些变因在网络中传播时对人口变迁的可能性带来的影响进行评估。这些可能性能够用来作为情境模式化, 以协助拟定降低脆弱性和增进灾害回復力的政策。

Rara vez se han acometido estudios sobre el modo como las variables de resiliencia comunitaria ante amenazas naturales interactúan como un sistema que afecta la resiliencia final (esto es, sus vínculos dinámicos). La red bayesiana (BN), que representa las interdependencias entre variables en un gráfico de incertidumbre en forma de distribuciones de probabilidad, ofrece una manera efectiva de investigar las interacciones entre diferentes componentes de la resiliencia y aboca el sistema natural-humano como un todo. Este artículo emplea una BN para estudiar las interdependencias entre diez variables de resiliencia y el cambio de población en la Cuenca del Bajo Río Misisipi (LMRB) a escala de grupo de manzanas censales. Se usó un algoritmo genético para identificar una BN óptima donde el cambio poblacional, un indicador de resiliencia acumulativo, era la variable objetivo. El algoritmo genético dio lugar a un modelo de BN optimizada con una exactitud de validación cruzada del 67 por ciento en un período de 906 generaciones. Se halló que seis variables tenían impactos directos sobre el cambio de la población, incluyendo en ellas el nivel de riesgo de amenazas litorales, el daño por catástrofe, la distancia a la línea costera, la tasa de empleo, el porcentaje de unidades habitacionales construidas antes de 1970, y el porcentaje de hogares con cabeza de familia femenina. Las restantes cuatro variables eran variables indirectas, que incluían el porcentaje de tierra agrícola, el porcentaje espacial de la zona de inundación, el porcentaje de unidades habitacionales ocupadas por el propietario, y la densidad de población. Cada variable tiene una tabla de probabilidad condicional de modo que sus impactos sobre la probabilidad del cambio poblacional pueden evaluarse a medida que se propaga a través de la red. Estas probabilidades pueden ser usadas para modelar el escenario que ayude a informar las políticas que reduzcan la vulnerabilidad y fortalezcan la resiliencia ante el desastre.

Key Words:

• Bayesian network
• community resilience
• coupled natural-human system
• Lower Mississippi River Basin
• population recovery

关键词：:

• 贝叶斯网络
• 社区回復力
• 自然—人类共变系统
• 密西西比河下游流域
• 人口回復。

Palabras clave:

• red bayesiana
• resiliencia de la comunidad
• sistema natural–humano acoplado
• Cuenca del Bajo Río Misisipi
• recuperación de la población

Additional information

Funding

This article was written based on the work supported by two research grants from the U.S. National Science Foundation: one under the Dynamics of Coupled Natural Human Systems (CNH) Program (Award No. 1212112) and the other under the Coastal Science, Engineering and Education for Sustainability (Coastal SEES) Program (Award No. 1427389). Support by a NOAA-Louisiana Sea Grant (Grant No. R/S-05-PD) is also acknowledged.

Notes on contributors

Heng Cai

HENG CAI is a Postdoctoral Research Associate in the Department of Environmental Sciences at Louisiana State University, 2281 Energy, Coast and Environment (ECE) Building, Baton Rouge, LA 70803. E-mail: [email protected]. Her research interests include geospatial modeling, coupled natural–human systems, and the assessment of disaster resilience.

Nina S. N. Lam

NINA S. N. LAM is Professor and E. L. Abraham Distinguished Professor of Louisiana Environmental Studies in the Department of Environmental Sciences at Louisiana State University, Baton Rouge, LA 70803. E-mail: [email protected]. Her research interests include GIScience, remote sensing, environmental health, resilience, and sustainability.

Lei Zou

LEI ZOU is a Postdoctoral Research Associate in the Department of Environmental Sciences at Louisiana State University, Baton Rouge, LA 70803. E-mail: [email protected]. His research interests include GIScience, location-based social media data mining, coupled natural–human system modeling, and disaster resilience.

Yi Qiang

YI QIANG is Assistant Professor in the Department of Geography at University of Hawaii–Manoa, Honolulu, HI 96822. E-mail: [email protected]. His research interests include GIScience in general, geocomputation, visual analytics, and hazard vulnerability in particular.