Assessing the cascading impacts of natural disasters

By Asjad Naqvi and Irene Monasterolo from the IIASA Advancing Systems Analysis Program

Asjad Naqvi and Irene Monasterolo discuss a framework they developed to assess how natural disasters cascade across socioeconomic systems.

© Bang Oland | Dreamstime.com

The 2021 Nobel Prize for Physics, was awarded to the topic of “complex systems”, highlighting the need for a better understanding of non-linear interactions that take place within natural and socioeconomic systems. In our paper titled “Assessing the cascading impacts of natural disasters in a multi-layer behavioral network framework”, recently published in Nature Scientific Reports, we highlight one such application of complex systems.

In this paper, we develop a framework for assessing how natural disasters, for example, earthquakes and floods, cascade across socioeconomic systems. We propose that in order to understand post-shock outcomes, an economic structure can be broken down into multiple network layers. Multi-layer networks are a relatively new methodology, mostly stemming from applications in finance after the 2008 financial crisis, which starts with the premise that nodes, or locations in our case, interact with other nodes through various network layers. For example, in our study, we highlight the role of a supply-side production layer, where the flows are trade networks, and a demand-side household layer, which provides labor, and the flows are migration flows.

Figure 1: A multi-layer network structure

In this two-layer structure, the nodes interact, not only within, but across layers as well, forming a co-evolving demand and supply structure that feeds back across each other. The interactions are derived from economic literature, which also allow us to integrate behavioral responses to distress scenarios. This, for example, includes household coping mechanisms for consumption smoothing, and firms’ response to market signals by reshuffling supply chains. The price signals drive flows, which allows the whole system to stabilize.

We applied the framework to an agriculture-dependent economy, typically found in low-income disaster-prone regions. We simulated various flood-like shock scenarios that reduce food output in one part of the network. We then tracked how this shock spreads to the rest of the network over space and time.

Figure 2: Evolution of vulnerability over time

Our results show that the transition phase is cyclical and depends on the network size, distance from the epi-center of the shock, and node density. Within this cyclical adjustment new vulnerabilities in terms of “food insecurity” can be created. Then, we introduce a new measure, the Vulnerability Rank, or VRank, to synthesize multi-layer risks into a single index.

Our framework can help inform and design policies, aimed at building resilience to disasters by accounting for direct and indirect cascading impacts. This is especially crucial for regions where the fiscal space is limited and timing of response is critical.

Reference:

Naqvi, A. & Monasterolo, I. (2021). Assessing the cascading impacts of natural disasters in a multi-layer behavioral network framework. Scientific Reports 11 e20146. [pure.iiasa.ac.at/17496]

Note: This article gives the views of the author, and not the position of the Nexus blog, nor of the International Institute for Applied Systems Analysis.