Variation characteristics and the impact of urbanization of extreme precipitation in Shanghai

Yali Mou, Xichao Gao, Zhiyong Yang, Tianyin Xu, Jie FengScientific 2022 · DOI

Abstract

With the rapid development of urbanization, the characteristics of extreme precipitation in urban areas have changed significantly. Revealing the spatial and temporal distribution of extreme precipitation under changing environment is the basis of scientific response to the urban flood. Trends of extreme precipitation at 95% and 99% thresholds in Shanghai and the influence of urbanization on them were analyzed. The results show that: (1) The precipitation threshold limit values for each site are 5.7–6.3 mm at 95% and 14.3–17.16 mm at 99%. (2) Extreme precipitation at each site has been increasing over the past 50 years, with growth rates of 95% and 99% extreme precipitation at 8.02–11.46%/10a and 7.11–16.86%/10a, respectively. (3) Extreme precipitation in Shanghai is significantly positively correlated with the urbanization of the area around the site, with more extreme precipitation and greater growth rates in highly urbanized areas.

Introduction

As global mega-cities expand, they fundamentally alter local microclimates. Urban flooding has become a critical threat, necessitating a precise understanding of how extreme rainfall patterns evolve alongside urban growth. While global warming contributes to increased precipitation, the specific role of urbanization—characterized by the replacement of natural vegetation with impervious surfaces and the resulting "urban heat island" effect—is complex. This study addresses how the degree of urbanization around specific locations in Shanghai influences the intensity and rate of change of extreme hourly rainfall.

Methodology and Data

The study utilized 50 years (1971–2020) of hourly precipitation data from 10 meteorological stations across Shanghai. To define "extreme" events, researchers employed the percentile threshold method, identifying the top 5% (95th percentile) and top 1% (99th percentile) of hourly rain events for each station. This approach accounts for local micro-climatic variations.

Urbanization was quantified using satellite-derived land-use maps from five periods between 1980 and 2020. Buffer zone analysis was conducted within a 10km radius of each station to measure the expansion of urban land over time.

Figure 1: Map of Shanghai illustrating urban land expansion from 1980 to 2020 and the locations of the meteorological stations used in the study.

Key Findings

The Intensification of Extremes

A primary insight from the analysis is that extreme precipitation is increasing significantly faster than average rainfall. While the city is getting wetter overall, the heaviest downpours are becoming disproportionately more intense. The growth rate for 95% extreme precipitation was found to be 8.02–11.46% per decade, while 99% events grew by 7.11–16.86% per decade. In contrast, average precipitation only increased by 4.9–6.96% per decade.

Figure 4: Comparison of relative growth rates (% per decade) for 95% and 99% threshold extreme precipitation versus average hourly precipitation across the 10 stations.

Urbanization as a Catalyst

The study reinforces the "urban rain-island" theory. Highly urbanized areas create thermal and dynamic conditions that accelerate water vapor transport and convection, effectively "sucking" moisture into the urban center and triggering violent bursts of rain. A strong positive linear correlation exists between urban land area and extreme annual rainfall. Quantitatively, for every 100 $km^2$ increase in urban land within a station's 10km buffer, annual extreme precipitation (at the 95% threshold) increased by approximately 24.01 mm.

Conclusion and Future Directions

The findings demonstrate that urbanization has a clear intensifying effect on regional extreme precipitation in Shanghai. However, the authors note limitations, including the relatively small sample size of 10 stations and the inherent variability of precipitation. Future research should explore the role of urban aerosols, which may act as cloud condensation nuclei and further complicate rainfall patterns. Disentangling the effects of local urbanization from global climate change remains a priority for future attribution analyses.

Summary updated

Abstract

With the rapid development of urbanization, the characteristics of extreme precipitation in urban areas have changed significantly. Revealing the spatial and temporal distribution of extreme precipitation under changing environment is the basis of scientific response to the urban flood. Trends of extreme precipitation at 95% and 99% thresholds in Shanghai and the influence of urbanization on them were analyzed. The results show that: (1) The precipitation threshold limit value for each site are 5.7 ~ 6.3 mm at 95% with a variation factor of 0.04, and 14.3 ~ 17.16 mm at 99% with a variation factor of 0.06. The precipitation thresholds under 99% conditions were more significantly different among stations. (2) The extreme precipitation at each site has been increasing over the past 50 years, and the growth rates of 95% and 99% extreme precipitation are 8.02~11.46%/10a and 7.11~16.86%/10a, respectively. The growth rate of extreme precipitation is significantly higher than that of average precipitation, while the extreme part of the precipitation probability distribution increases considerably. There is a strong variability in extreme precipitation in this region, while the 99% threshold precipitation varies more. (3) The extreme precipitation in Shanghai is significantly positively correlated with the urbanization of the area around the site. Urbanization has an increasing effect on regional extreme precipitation, with more extreme precipitation and greater growth rate in highly urbanized areas.
BibTeX
@inproceedings{mou2022variation,
  author = {Yali Mou and Xichao Gao and Zhiyong Yang and Tianyin Xu and Jie Feng},
  title = {{Variation characteristics and the impact of urbanization of extreme precipitation in Shanghai}},
  booktitle = {Scientific},
  year = {2022},
  doi = {10.1038/s41598-022-22352-4},
}
Figure