Analysis of disaster-affected population mobility through grid-aggregated mobile location data: The 2017 Jiuzhaigou earthquake, China

Zezhi Lin, Rui Mao, Huaiqun Zhao, Zihui Tang, Saini Yang and Po Pan

PLOS ONE, 2025, vol. 20, issue 10, 1-16

Abstract: In disaster research, individual-level mobile phone location data is considered highly valuable for assessing population mobility and disaster impacts. However, due to privacy regulations in China, only spatially aggregated mobile data with a resolution of 1 km × 1 km are available. These data do not contain explicit population individual population movement, which poses challenges for analyzing population movement patterns in disaster research. To using this grid-based mobile data to describe population movement, we applied an empirical orthogonal function (EOF) method to the post-disaster phase of the 2017 Jiuzhaigou earthquake. The first EOF mode (EOF1) primarily exhibits positive anomalies centered over the Jiuzhaigou Valley. The principal components for the EOF1 show a decreasing trend from midnight to 20:00, indicating a continuous outflow of population from the Jiuzhaigou Valley during this period. The second mode (EOF2) exhibits negative anomalies at the Jiuzhaigou Valley and along the road to the southwest of the Valley, while positive anomalies appear along two roads, i.e., one extending from the Jiuzhaigou Valley to Shuanghe, and the other from the Chuanzhusi Town government square to western Chuanzhusi. The primary components of EOF2 reveal that, from midnight to 10:00, population increased along these two roads while decreasing over the Jiuzhaigou Valley and the road leading southward to the Chuanzhusi Town government square. After 10:00, this population change pattern diminished between 10:00–15:00. Based on the EOF2 results, two evacuation routes were identified: Path 1 extended northwest from the Chuanzhusi Town government square; Path 2 led southeast from Jiuzhaigou Valley through Shuanghe Town. In comparison, the BBAC_I clustering method identifies clusters with similar temporal trends but fails to pinpoint the most affected areas or infer evacuation directions. In contrast, EOF analysis overcomes these limitations by revealing key impact zones and evacuation patterns, even in the absence of trajectory data.

Date: 2025
References: View complete reference list from CitEc
Citations:

Downloads: (external link)
https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0335415 (text/html)
https://journals.plos.org/plosone/article/file?id= ... 35415&type=printable (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:plo:pone00:0335415

DOI: 10.1371/journal.pone.0335415

Access Statistics for this article

More articles in PLOS ONE from Public Library of Science
Bibliographic data for series maintained by plosone ().