Linking Luojia 1-01 nightlight imagery to urban crime

• Published in: Applied geography (Sevenoaks) Vol. 125; p. 102267
• Main Authors: Liu, Lin, Zhou, Hanlin, Lan, Minxuan, Wang, Zengli
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.12.2020
• Subjects: Burglary; Crime; Luojia 1-01; Nightlight; Satellite image; Street robbery; Street robbery; Satellite image; Burglary; Luojia 1-01; Nightlight; Crime
• ISSN: 0143-6228; 1873-7730
• DOI: 10.1016/j.apgeog.2020.102267

Various environmental criminology theories and empirical studies have linked the urban environment to crime. The crime pattern theory, in particular, argues that edges, either social or physical, affect crime. A recent study has combined both social and physical edges to derive composite edges. A composite edge index measured by NPP-VIIRS satellite nightlights at the census tract level is found to be related to street robbery and burglary. Nightlight images of Luojia 1-01, launched in June 2018, have a much higher spatial resolution than that of NPP-VIIRS. This study applies Luojia 1-01 nightlight data to measure composite edges by nightlight gradients at the smaller census block group level. The effects of the composite edges on street robbery and burglary are explored by negative binomial models. Results show that composite edges measured by Luojia 1-01 nightlight data improve the fitness of models noticeably on street robbery but not on burglary. Nightlight gradients make a statistically significant and positive impact on the street robbery rate, but an insignificant and negative impact on the burglary rate. Furthermore, the composite edge effect on street robbery is more substantial than that on burglary. In sum, this study provides evidence that Luojia 1-01 nightlight imagery can help explain crime at the aggregated block group level, but its impact on crime varies by crime type. •Make the first attempt of using Luojia 1-01 nightlight imagery to study urban crime.•Confirm that Luojia 1-01 is applicable for crime research at the census block group level.•Reveal that composite edge effects are more substantial on street robbery than burglary.