Search

Human mobility networks can reveal insights into resilience phenomena, such as population response to, impacts on, and recovery from crises. The majority of human mobility network resilience characterizations, however, focus mainly on macroscopic network properties; little is known about variation in measured resilience characteristics (i.e., the extent of impact and recovery duration) across macroscopic, substructure (motif), and microscopic mobility scales. To address this gap, in this study, we examine the human mobility network in eight parishes in Louisiana (USA) impacted by the 2021 Hurricane Ida. We constructed human mobility networks using location-based data and examined three sets of measures: (1) macroscopic measures, such as network density, giant component size, and modularity; (2) substructure measures, such as motif distribution; and (3) microscopic mobility measures, such as the radius of gyration and average travel distance. To determine the extent of impact and duration of recovery, for each measure, we established the baseline values and examined the fluctuation of measures during the perturbation caused by Hurricane Ida. The results reveal the variation of impact extent and recovery duration obtained from different sets of measures at different scales. Macroscopic measures, such as giant components, tend to recover more quickly than substructure and microscopic measures. In fact, microscopic measures tend to recover more slowly than measures in other scales. These findings suggest that resilience characteristics in human mobility networks are scale-variant, and thus, a single measure at a particular scale may not be representative of the perturbation impacts and recovery duration in the network as a whole. These results spotlight the need to use measures at different scales to properly characterize resilience in human mobility networks. [ABSTRACT FROM AUTHOR]

This paper presents damage assessment using a hierarchical transformer architecture (DAHiTrA), a novel deep‐learning model with hierarchical transformers to classify building damages based on satellite images in the aftermath of natural disasters. Satellite imagery provides real‐time and high‐coverage information and offers opportunities to inform large‐scale postdisaster building damage assessment, which is critical for rapid emergency response. In this work, a novel transformer‐based network is proposed for assessing building damage. This network leverages hierarchical spatial features of multiple resolutions and captures the temporal differences in the feature domain after applying a transformer encoder to the spatial features. The proposed network achieves state‐of‐the‐art performance when tested on a large‐scale disaster damage data set (xBD) for building localization and damage classification, as well as on LEVIR‐CD data set for change detection tasks. In addition, this work introduces a new high‐resolution satellite imagery data set, Ida‐BD (related to 2021 Hurricane Ida in Louisiana) for domain adaptation. Further, it demonstrates an approach of using this data set by adapting the model with limited fine‐tuning and hence applying the model to newly damaged areas with scarce data. [ABSTRACT FROM AUTHOR]