An extreme rainfall event occurred in the Beijing metro area on 21 July 2012 when about 460 mm of rain fell in the Fangshan district in about 18 hours (from 1000 BST 21 to 0400 BST 22 July, 2012). The hourly rainfall rates were higher than 85mm. It caused widespread damage (about 79 people were reported dead and about $2 billion in economic losses, Zhang et al. 2013). The official China Daily newspaper reported Monday that rain and flooding caused damages of at least 10 billion Yuan ($1.6 billion) damage, with 60,000 people evacuated from their homes (http://www.huffingtonpost.com/2012/07/22/china-storms-heavy-rains-20-dead-10-beijing_n_1693380.html). The official Globe Times said that it was the heaviest rainstorm in the capital in 61 years.
We made an attempt to investigate whether and how much the Beijing city itself impacted this heavy rainfall event. In the past, the importance of urban modification of rainfall has been highlighted by many investigators. Due to the increased population density in the cities and also the recent extreme flash flood in the cities, a detailed study of the mechanisms involved in the extreme events over the urban regions are needed so that these extreme events can be predicted with more accuracy in future. Previous studies reported a mixed results regarding the impact of urbanization on precipitation: some pointed out a reduction in precipitation due to the reduction in available moisture, whereas some found that the heavy rainfall in the metropolis were enhanced. Apart from these studies showing the impact (positive or negative) of urbanization on rainfall, another study on the Beijing storm of 2012 found that the operational models gave the right results but for wrong reasons (Zhang et al., 2013). The model predicted the rainfall mainly from topographical lifting and the passage of cold front, whereas the observations showed that rainfall was mostly generated by convective cells that were triggered by local topography and then propagated along a quasi-stationary linear convective system into Beijing. In particular, most of the extreme rainfall occurred in the warm sector far ahead of the cold front (Zhang et al. 2013). Miao et al. (2011) investigated the impact of Beijing urbanized areas on summer precipitation, and found that the urbanization prior to 1980 decreases the maximum rainfall whereas the later urbanization resulted in bifurcating the path of the rainfall. Kusaka et al. (2014) pointed out that convective rainfall responded nonlinearly to initial conditions and model configurations. They examined the urbanization impact on precipitation at Tokyo with the ensemble climate simulation approach. Given the above results and uncertainty in model physics parameterization, it is challenging to predict heavy rainfall in cities. Moreover, uncertainties observing small-scale features of heavy reainfall in heterogeneous cities makes it more difficulty in validating model results.
For the present work, we used WRF-ARW (Weather Research and Forecasting/ARW core) model for the prediction of the Beijing super-storm with the urban modeling system. An evaluation of the WRF-Urban modeling system framework is discussed in Chen et al. (2011). We used different urban modeling options: bulk urban parameterization scheme, single Layer urban canopy model (SLUCM), multi-layer urban canopy model (BEP), and BEP with Building Energy Model (BEM) in order to assess the impact of urban processes on the Beijing super-storm. The preliminary results show that the Beijing super-storm is sensitive to the initial atmospheric conditions as well as the choice of the urban parameterization schemes. The model results are compared with the radar observations, station observations, and satellite-derived data (CMORPH) and CPC climatology data. Fig. 1 shows the area-averaged (over the rectangular box in right panel below) total amount of rainfall (mm) for the 24 hr period starting 21 July 2012, 00Z using different WRF-Urban physics parameterization options: (a) Grass (where all the urban areas are replaced with grassland), (b) WRF/SLUCM (SLUCM without anthropogenic heat effect), (c) WRF/SLUCM (with anthropogenic heat effect), (d) WRF/BEP, and (e) WRF/BEP+BEM.
Fig 1: Area averaged total Accumulated precipitation (mm) over the rectangular box (right panel above)
The spatial distribution (not shown here) of precipitation illustrates that each urban scheme has some positive effects in forecasting rainfall coverage or rainfall amount, which leads to the conclusion that ensemble approach may be the best approach for the prediction of such extreme events. The analysis of the results show that, due to the complex urban processes, proper representation of the urban region is necessary to obtain the best results.
Furthermore, this investigation will be integrated with a risk and vulnerability assessment to analyze urban flood risk and vulnerability in Beijing, and to explore how rapid urban development is impacting the city vulnerability.Urban risk can be defined in many ways –e.g., the likelihood of occurrence of floods and droughts; the possibility of reduced water availability, loss, injury, water and other impacts; and the probability of occurrence of a climate adverse event like more intense precipitation events (Shrader-Frechette, 1991). We define urban risk as the location-specific damage potential, which is a function of flood hazard level, number of exposed elements and their social vulnerability (Müller, 2013; Romero-Lankao et al., 2013).
Building on prior work (Romero-Lankao et al., 2014; Romero-Lankao et al., 2013), we will explore correlations between damage impacts and various elements of a new spatial/temporal risk index (that integrates hazard and vulnerability indices). Quantitative data collection and analysis methods will be used to explore:
– Hazard risk i.e., types of flooding events with their magnitude along the rapidly urbanized Beijing region. Depending on data availability we will use the Standardized Precipitation Index (SPI) and the standardized runoff index to measure runoff excess.
– Exposure: spatio-temporal distribution of people and infrastructures (e.g., built-up area, roads, shopping centers, health facilities, sanitation and water)
– Differential socioeconomic vulnerability within the cities. Census and/or survey data indicators of human capital (education, employment); physical capital (housing quality; green areas per km2, access to drinking water, sanitation, and other high quality services); financial capital (home amenities, insurance) and social capital (social networks, emergency response systems, length of residence).
Chen, F., H. Kusaka, R. Bornstain, J. Ching, C.S.B. Grimmond, S. Grossman-Clarke, T. Loridan, K. Manning, A. Martilli, S. Miao, D. Sailor, F. Salamanca, H. Taha, M. Tewari, X. Wang, A. Wyszogrodzki, and C. Zhang, 2011: The integrated WRF/urban modeling system: development, evaluation, and applications to urban environmental problems. Int. J. of Clim., 31, 273-288. DOI: 10.1002/joc.2158.
Kusaka et al. 2014, Mechanism of Precipitation Increase with Urbanization in Tokyo as Revealed by Ensemble Climate Simulations. J. Appl. Meteor. Clim., 53, 824-839. doi: http://dx.doi.org/10.1175/JAMC-D-13-065.1.
Miao, S., F. Chen, Q. Li, and S. Fan, 2011: Impacts of Urban Processes and Urbanization on Summer Precipitation: A Case Study of Heavy Rainfall in Beijing on 1 August 2006. J Appl. Meteorol. Climatol., 50 (4), 806-825 DOI: 10.1175/2010JAMC2513.1.
Romero-Lankao, P., Hughes, S., Qin, H., Hardoy, J., Rosas-Huerta, A., Borquez, R., and Lampis, A. (2014). Scale, urban risk and adaptation capacity in neighborhoods of Latin American cities. Habitat International 42, 224-235.
Zhang et al 2013, The Beijing extreme rainfall of 21 July 2012: “Right results” but for wrong reasons. Geophys. Res. Lett., 40, 1426-1431, DOI: 10.1002/grl.50304.