Session Overview
UCP9: Impact of cities on precipitations
Tuesday, 21/Jul/2015:
4:30pm - 6:15pm

Session Chair: Robert ("Bob") D. Bornstein, San Jose State University
Location: Caravelle Room


The effect of urban environments on storm evolution through a radar-based climatology of the central United States

Kristin Calhoun1,2, Darrel Kingfield1,2,3, Kirsten de Beurs3, Geoffrey Henebry4

1Cooperative Institute for Mesoscale Meteorological Studies, University of Oklahoma, United States of America; 2NOAA/National Severe Storms Laboratory, United States of America; 3Dept. of Geography and Environmental Sustainability, University of Oklahoma, United States of America; 4Geospatial Sciences Center of Excellence, South Dakota State University, United States of America

The influences urban environments have on storm development and patterns have been understudied to date, particularly for small to mid-sized cities. Using data from the multi-year reanalysis of remotely sensed storms (MYRORSS), climatological aspects of thunderstorms and their associated hazards have been examined in and around urban areas of varying size, shape, and growth in the central United States. The MYRORSS dataset provides a uniform grid of quality-controlled multi-radar data from the Weather Service Radar 1988 Doppler network that can be used to quantify the spatiotemporal characteristics of hazardous weather such as hail and storm rotation. Output from severe weather algorithms (including hail-size estimation, vertically integrated liquid, azimuthal shear and mesocyclone detection) are compared to complex urban areas and changes in these urban areas over three different periods across the past 10 years. Additionally, these climatological patterns are compared to individual storm behavior in and around the urban environments of Minneapolis-St. Paul, Dallas-Ft. Worth, Omaha, and Oklahoma City. Results will examine the differences in storm size, track, and severity prior, during and after crossing each of the different urban domains. We will also examine differences that arise due to diurnal effects, seasonality, and varying the storm type and size relative to urban size and shape.


Impact of Urbanization on Local Circulation and Precipitation over the Leeward Mountain

Hiroyuki Kusaka, Mayumi Mizunari

University of Tsukuba, Japan

Thermally induced local circulation, such as valley circulation, sea breeze circulation, or heat island circulation, is generated on sunny days with weak winds in mountainous, coastal and urban areas. A close relationship is known to exist between such local circulations and diurnal variation in precipitation.

It is conjectured that sea breeze circulation and heat island circulation contribute to the generation of precipitation. In recent years, research has been conducted on the effects of complex systems involving sea breeze circulation and heat island circulation on precipitation occurring in coastal cities.

Tokyo is the world's largest coastal city with mountains on the windward side and the sea on the leeward side of the environmental wind. In this area, convective precipitation typically occurs in the mountainous area around noon on sunny days in the warm season; after which, a precipitation system moves to the coastal area on the leeward side of the mountains in the afternoon, and eventually generates precipitation in the coastal area in the nighttime.

In this study, we evaluate the impact of the sea and city on diurnal variation in precipitation occurring leeward side of the mountain; that is, precipitation observed on the mountain in the daytime and in the coastal urban area in the nighttime.


Impact of urbanization on the Beijing Super-Storm of 21 July 2012 under current and future climatic conditions

Fei Chen1, Mukul Tewari1, Shiguang Miao2, Paty Romero-Lankao1

1NCAR, United States of America; 2Institute of Urban Meteorology, Chinese Meteorological Administration, Beijing, China

The Beijing Super-Storm produced more than 190 mm rainfall during 21-22 July 2012. With the maximum of 460 mm at the precipitation center, it was the heaviest rainfall event over the Beijing Metro area in the last six decades, and it took 77 lives and caused more than $161 million (U.S. Dollars) in property damages. This study employed the Weather Research and Forecasting (WRF) model coupled with urban modeling systems (WRF-Urban) to address two major issues: 1) What are important uncertainties affecting the forecasting this high-impact weather event? To which degree does an ensemble approach could improve the prediction of the location and amount of this urban flood? How does urban expansion contribute to the formation and evolution of this storm? and 2) What would be the impacts of future climate change on this event? These two objectives were achieved through conducting and analyzing a number of WRF-Urban high-resolution hindcast and future-climate sensitivity experiments. Hindcast results indicated that initial atmospheric conditions and time used in WRF-Urban substantially affected the forecast of this storm regarding its initiation, location and intensity. Complex terrains in the western regions outside the Beijing Metro played the most important role in triggering the storm and determining its movement paths. Various properties in the urbanized regions (such as roughness, the representation of multi-layer urban canopy in the model, anthropogenic heating, etc) played secondary and yet not negligible role in modulating the movement of the storm, especially the bifurcation of precipitation, and the amount of rainfall. For future climate-change simulations, we used a pseudo–global warming (PGW) approach. The 11-year-average data from global climate-model simulations were used to generate the climate signals, which were used to perturb the initial conditions based reanalysis data. The perturbed fields include wind speed, geopotential height, relative humidity, sea-surface temperature, and soil temperature for the future climate simulations. Preliminary results from WRF-Urban simulation under future warming climate conditions showed that the Beijing Super-Storm are intensified with more widespread rain in the region.

Numerical simulation of urban influence on summertime precipitation in Tokyo: How does urban temperature rise affect precipitation?

Naoko Seino, Toshinori Aoyagi, Hiroshige Tsuguti

Meteorological Research Institute, Japan

Precipitation modification due to increasing urbanization is one of major topics in the urban climate research, because precipitation amount and spatial distribution are fundamentally important for both water resource management in cities in arid area and disaster prevention in urban areas vulnerable to flash flood. Recently, Fujibe et al. (2009) showed that precipitation amounts observed in central Tokyo had statistically significant increasing trend and positive spatial anomaly to those in surrounding areas in the afternoon of warm season. However, the detected change in precipitation can be influenced by factors other than urbanization, such as global warming and associated regional climate change. For the better understanding of urban impact on precipitation, numerical simulations for recent eight years August have been conducted.

The Non-hydrostatic Model (NHM) of Japan Meteorological Agency (JMA) is utilized in the simulation. Horizontal grid interval is 2km and model domain covers central Japan region. The Square Prism Urban Canopy (SPUC) scheme is incorporated in the model to represent heat and radiation exchanges by urban canopy elements. To investigate how increasing heat island intensity in Tokyo affects precipitation in the metropolitan area, comparative experiments have been done for August from 2006 to 2013. Realistic urban surface condition was given in one of the experiments (SPUC experiment). Less urbanized surface condition was assumed in the other experiment, where only slab surface scheme was used (SLAB experiment). Mean temperature difference between the two experiments was at most one degree.

Simulation results suggest that mean monthly precipitation amount in the central Tokyo area is approximately 10% larger in the SPUC experiment than in the SLAB experiment, which is caused by the urban temperature rise. We also examined daily maximum amount in the urban area. On average, larger daily precipitation amounts were found in the SPUC experiment. However, differences in daily precipitation amounts and spatial distribution between the SPUC and SLAB experiments largely varied case by case. Processes associated with the simulated precipitation difference should be further investigated.


Modeling of the 26 August 2011 extreme precipitation event over Tokyo with Canada’s subkm-scale Global Environmental Multiscale (GEM) model

Stephane Belair1, Sylvie Leroyer1, Lubos Spacek1, Naoko Seino2

1Environment Canada, Canada; 2Meteorological Research Institute, Japan Meteorological Agency, Japan

High-resolution atmospheric modeling systems with grid spacing of a few hundreds of meters are now at the front line of short-range deterministic numerical weather prediction (NWP). In this context, 1-km and 250-m versions of Environment Canada (EC)’s Global Environmental Multiscale (GEM) model are being configured and tested over a few key urban areas. Such a system has been set-up and tested for the 26 August 2011 extreme precipitation case over Tokyo, as part of EC’s participation to the international research and development initiative for the Tokyo Metropolitan Area Convection Study (TOMACS).

It will be shown at the conference that GEM is able to realistically reproduce the intensity and structure of the mesoscale weather system responsible for the flooding event that took place over Tokyo on 26 August 2011. A series of sensitivity tests have been performed to assess the relative importance of urban surface processes, as well as other processes. Results emphasize the need for subkm-scale grid spacing for this event, and evidence the important role that the urban areas plays in sustaining intense precipitation rates over parts of Tokyo.

Cities & Storms: How Land Use, Settlement Patterns, and the Shapes of Cities Influence Severe Weather

Geoffrey M. Henebry1, David J. Stensrud2, Allison L. Steiner3, Christopher Small4, Laura R. Musacchio5, Trisha L. Jackson1, Kirsten M. de Beurs6, Kristin Calhoun6,7, Stacey Kawecki3, Larissa Reames6, Cole P. Krehbiel1

1South Dakota State University, United States of America; 2Pennsylvania State University, United States of America; 3University of Michigan, United States of America; 4Columbia University, United States of America; 5University of Minnesota, United States of America; 6University of Oklahoma, United States of America; 7NOAA/National Severe Storms Laboratory, United States of America

It has long been known that larger urban areas can affect the behavior of severe weather. However, only nine US cities have a population more than 1 million and 16 cities are larger than 800 sq km. The extent to which smaller urban areas influence severe weather has received less attention. This presentation will review findings to date from an interdisciplinary project funded by NASA to investigate the influences of urbanized areas on storm characteristics and behavior. Our research has focused on the Great Plains of the US where an abundance of warm season storms interact with cities of various sizes that are embedded within a matrix of agricultural land uses. Dynamical aspects of city-storm interactions were simulated using WRF-ARW using an inner domain with 500 m resolution. The approach presented actual configurations of various cities to a “well-behaved” supercell to assess various storm characteristics, e.g., updraft helicity, storm-relative helicity, absolute vorticity, following the encounter with the urbanized area. How aerosol loading in the urban dome influences storm behavior was probed using WRF-CHEM simulations at 4 km horizontal resolution, with a focus on cloud fraction, cloud effective radius, simulated reflectivity, and accumulated precipitation inside and outside of plumes of polluted urban air. Scenarios included doubling and halving baseline anthropogenic emissions. In addition to these simulations, we used time series of NEXRAD data series to explore the city-storm interactions from the perspective of weather radar and time series of space-borne remote sensors to evaluation the influence of city size and location on land surface phenology. Finally, we have been exploring how to use the concept of “land architecture” to facilitate knowledge transfer between the science and design communities.

Urban Impacts on Regional Rainfall Climatology

Dev Niyogi

Purdue University, United States of America

This presentation will provide an overview of the recent work that documents the state of the science regarding the urban impacts on regional rainfall climatology. Multiscale datasets, and modeling studies will be presented to develop a comprehensive picture regarding the confounding role of land surface heterogeneity, urban/rural aerosols, regional topography and interactions in assessing and documenting these impacts.

Efforts underway in transferring the knowledge gained into future urban design modules and tools will also be presented.