Suspended Particles and Urban Design Criteria: High Resolution Measurement of Respiratory Health and Geospatial information for Newborns and Young Children in Barcelona and Portland 1 Introduction Unborn, new-born and young children’s physiological development may be significantly affected by the air quality in urban environments. This requires not only knowledge of air quality including suspended particles but small-scale geospatial understanding of public space. Respiratory and neurological diseases in children such as asthma and air born chemicals have been linked together by health specialists in Europe and the United States (Pederson) (Pope). The air quality in the public space where these sensitive members of society pass through, play, eat and learn goes mostly unmeasured at the small-scale. Most US and European cities possess fewer than five active suspended particle PM air quality stations for any single airborne quality. This research investigates the creation of an affordable handheld instrument and geospatial analysis tool at the small scale of suspended particles PM 2.5 and smaller 0.5 micrometers with urban design characteristics. Measurement are taken along sidewalks in Barcelona’s “Eixample” neighborhood, and later in Portland, Oregon: 1) at 33m increments across standard 100m blocks in a 3 x 3 block test area and 2) along the daily route of a typical two-year-old child from home to public realm and back. The study uses expertise in computing, urban design and health science using parametric architectural software and table-based quantitative, qualitative and geospatial information to empower behavioral change.

2 BACKGROUND / CONTEXTUALIZATION Identification of sample air quality criteria The identification of a first air quality criteria to measured and analyze began by researching the following list of US EPA criteria publically accessible via web-based maps of major cities: • CO • CO2 • Lead - TSP (LC) • Lead - PM10 (LC) • NO • NO2 • Ozone • PM10   • PM2.5 • SO2 • PM2.5 (limited) • IMPROVE - Interagency Monitoring of Protected Visual Environments • NATTS - National Air Toxics Trends Stations • NCore - NCore Multi-pollutant Monitoring Network We subsequently interviewed respiratory health expert Jordi Sunyer at the Centre for Research in Environmental in Barcelona. We identified suspended particles PM 2.5 as effective criteria especially contributions from automobile combustion, disc brakes and rubber tires. In 2006 Barcelona’s annual average PM count was 50µg/m3 with OMS recommendation not to exceed 20µg/m3 without sensitivity to asthma and bronchitis in children (WHO).  

[figure 1, Concentrated PM10 observed in world cities, World Health Organization, OMS]

Scale of measured criteria: urban design and computing context Suspended particles go unmeasured at the small-scale of public space related to urban form, parcel use and transportation. In Manhattan for example has only one active PM sensor located below Canal Street. In Portland, Oregon, only one active sensor exists (AirData). One PM sensor exists in the Eixample study area in downtown Barcelona (L’aire). The investigation focuses on the street level space in front of buildings. 100 meter long Eixample blocks are measured at twelve locations in Barcelona and 200 feet long blocks in Portland in future data collection. Measurement in cities using the senses (Lynch) (Vitello) complements top down formal design. Everyday activities over time, phenomenon, may be measured (Norberg Schulz) and urban activities may emerge from ‘loose spaces’ in cities (Franck). Carlo Ratti’s MIT’s SENSEable Cities lab has led the urban computing field to include issues of information access at the small scale (Nabian et al.). We utilized a previous small-scale mapping technique for 3 x 3 block areas in Barcelona, taking measurements at the center of sidewalks in public right-of-ways at 33m increments (author). This mapping method allows variations in urban design characteristic to emerge including building heights, open space, adjacent traffic lanes, bus routes, bicycle lanes, landscaping buffers such as trees and low bushes and urban furniture. Other air pollution tools include a networked of everyday citizens such as “Smart Citizen” (IaaC / Cisco) and a neighborhood study in Portland focusing of registered air pollution contributors by Richard Beckwith of Intel Labs (Wong). 3 METHOD To create a geospatial understanding of PM air pollution at the small-scale a parametric modeling approach is used combining: 1) a PM sensor device measuring geo-locational point-data and 2) a table based mapping organization to analyze and visualize data various types of urban design data. PM Sensor With the objective to produce an affordable point sensory for PM 2.5 using a Sharp Optical Dust Sensor with an Arduino Uno and later USB connected with a Raspberry Pi. A python script was written to process the PM general concentrations and noise sensor. We are targeting a 0.5 micrometer concentration of PM that is more accurate than PM 2.5 US EPA standard. A GPS module was used to add latitude and longitude. An LCD display was added to verification data sent via WiFi module in CSV table format. An SD secure digital card was used. The sensor device is designed to take measurements at the 33m increments first assigned via a My Places Google map with 108 pin locations and exported from KML to CSV format. The device uses a ‘butterfly net’ technique walking along the center of sidewalks within a one-meter tolerance of the 108 points to capture data (UAV for future tests). At each point location ten measurements are taken and averaged by the Arduino. The Raspberry Pi creates a CSV file for both the PM and noise data collection for all108 points. Data measurement is taken at three levels: ground, inhalation level and at 5m above the ground to analyze differences from urban design characteristics and PM level. A helium balloon is used for the sensor 5m above ground connected via a lite-weight, six-wire cable (future wi-fi module possible). A third Sharp X sensor is hung to 10cm above ground level.

[figure 2, PM sensor device assembly] Geospatial Information, GI, Urban Analysis Tool The research data of PM and sound are analyzed and visualize geospatially using Rhino Grasshopper (Rutten) with plugins and custom scripting. Traditional planning geospatial information systems software such as ESRI ArcGIS (ArcGIS) use table based data including latitude and longitude. This model is used in Rhino Grasshopper with two basic plugins and scripting: 1) ELK plugin using open street maps files in OSM format with existing building, highway, bike and rail data and 2) HUMAN plugin using CSV files with data including latitude, longitude and other quantitative and qualitative data (Heumann). Output visualization of this tool includes colored maps, a series of lines in the z-vector, a surface topography and curve defined topography. Criteria to select the 3 x 3 Eixample district test area included:

- Regularity of the city grid (100m square) - Regularity of the urban wall - At least one public open space at block center - Regularity of the public right-of-way - Variations in traffic lanes, bus routes and bus shelters, bike share and bike lanes.

[figure 3, 3 x 3 block study area and 108 data points] The 108 points are uploaded to the Raspberry Pi as a CSV file. Using a GPS module on the handheld device a ‘butterfly net’ script is written to capture data when the device is walked in the 3 x 3 test area and passed within a 1m radius of each of the 108 points. Walking protocol was developed including pace and device orientation.

[figure 4, Data Collection, with device within bus shelter]

4 DATA: COLLECTION AND ANALYSIS Collection Test Area Measurements were taken on March 24, 2014, in the 3 x 3 block Test Area 01 in Barcelona’s Eixample district between Compte de Urgell and Villaroel, and Arago and Gran Via. Pre-test measurements were taken the prior day. Data was taken at one height holding the device. Future data will include ground level, inhalation level and at 5m above grade. Test observations included spring pollen, the influence of wind and the need for various protocols, control tests and observation recording. Collection Limits and Future Prototype Criteria Natural Environment The sensitivity of the Sharp sensor is controlled using a Dylos DC1100 Pro (Dylos). Future tests will use a battery powered Dylos DC1700 Pro for onsite control testing and other control test will use high quality lab equipment of Xavier Querol of the Spanish Research Council, CSIC in Barcelona. The Sharp sensor in the device performed with higher ‘jumps’ in data recording than the Dylos but was judged adequate to identify significant PM differences between locations. Future in-lab control tests will study the effects of observed wind, pollen, distance to source, sensor orientation and time of exposure. Tests will be taken at high and low traffic. A wind damper and improved sound sensor will be added. Additional ambient environment sensors such as anemometer wind sensors may be added.

[figure 5, Data Collection, adjacent private vehicles, taxis, motorcycles] Built Urban Environment Test observations of urban design criteria included possible sources of data variation:

- Effects within or around bus shelters - Ground-level storefront uses (example, car painting shop on Urgell and pizza shop) - Building height - Street slope - Street orientation to sea and mountains - distance to criteria such as traffic lanes, pedestrian medians and vegetation buffers

(example, Gran Via) - Zoning use sensitive to children including schools and parks

The Everyday Route Future tests will track the possible everyday route of a two-year old child across but not limited to Test Area 01 including interior residence locations, schools, shops, parks and other open spaces. These tests will examine the affects of the built design characteristics in plan, section and 3D. Analysis / Visualization

Data in Rhino Grasshopper is 1) analyzed and 2) visualized. Future analysis and visualization will integrate degrees of information such as PM, ambient environment (sound, wind) and urban form and quality. An objective of this visualization is the ability to read both integrated data and the individual degrees/layers of information with graphic differences such as z-vector, color, line weight, line type and transparency. Future diagrams of unit parcel, section and city block will be added as well as diagrams of parametric workflow.

[figure 6, 3 x 3 block study area, PM0.5 data visualization at 108 data points] 5 CONCLUSSIONS AND FUTURE RESEARCH Information Flow and Scale The result of this research-in-progress is a confirmation of the importance of scale of suspended particle PM air quality and urban design characteristics at the small scale within an overall regularized block structure. For example, PM measurement at the sidewalk along the high traffic of Gran Via de Les Corts Catalan was low. This may be a result of broad airflow at this atypically wide street, limited vehicular traffic of two lanes including taxi/bus lanes, adjacent pedestrian and bike lanes and or a low vegetation wall buffering the additional six lanes of vehicle traffic. The accuracy of measuring twelve points around each 100m x 100m Eixample block provides this detail, filling a gap in the research of PM air quality in cities.

[figure 7, Gran Via de Les Corts Catalanes, sidewalk and pedestrian / bike median and vegetation buffer]

Deployment and Attachment The mobility of the device allows a more diverse investigation of urban design characteristics than fixed monitoring stations do not afford. Bruno Latour’s idea of ‘attachment’ suggests that we better understand the environment by capturing its meaning over time (Latour). Information technology allows this new understanding of the built environment. Future data will be collected on foot, on bicycles, via UAV (currently in negotiation with the mayor’s office of Barcelona) or via public infrastructure such as city buses, street cleaners or police vehicles. We also noticed that the relatively low sensitivity of the Sharp sensor ($25) was acceptable with high steps in data readings. Data sensitivity may be enhanced using a more costly Dylos 1700 portable sensors ($400), tuned with high-end devices ($10-20,000) from the Spanish Research Council. Urban Design The relatively fine scale of PM air pollution measurement may require even smaller scaled measurements of typical urban design conditions such as block structure, open space, street width, pedestrian buffers and vegetation. Future tests including the everyday path of a two-year child will provide additional information about known sensitive location such as homes, kindergartens, parks, benches along sidewalks and pedestrian streets with proven connections of indoor and outdoor air quality (Sunyer). The aim of this information is not only to inform city planners and government agencies but more importantly to empower everyday citizens, namely parents of children, with improved decision making for the healthy growth of our next generation.

6 REFERENCES AirData, Interactive Map, US EPA, PM2.5 criteria selection, http://www.epa.gov/airdata/ad_maps.html Pope C, Ezzati M., Dockery D (2009), “Fine-Particulate Air Pollution and Life Expectancy in the United States” The New England Journal of Medicine, n engl j med 360;4 nejm.org January 22, 2009, http://www.uphe.org/wp-content/uploads/2011/12/nejm-improved-life-expectancy-and-air-quality.pdf ArcGIS (1999 1st ed.) Environmental Systems Research Institute author (2012) “Place Branding from the Bottom Up: Strengthening Cultural Identity through Small-Scaled Connectivity.” Conference Proceedings: Roots-Politics-Place. Berlin: inPolis: 237-252. author (2013) Green Design, Materials and Manufacturing Processes, “Parametric Places 22@: Smart Urban Analysis Tools and Place Branding Value.” SIM Conference Proceedings Editors: Helena Bartolo et al. London: Taylor & Francis Group. pp. 701-705 Bell ML (2012). Assessment of the Health Impacts of Particulate Matter Characteristics. Research Report 161. Health Effects Institute, Boston, MA. Dylos DC1100-Pro Sensor, particle sensor, DC 1700-Pro rechargeable, http://www.dylosproducts.com/ornodcproair.html Franck, K and Lepori, R (2007) Architecture from the Inside Out: From the Body, the Senses, the Site and the Community. London: Routledge Hostius D.M,. Pillarisetti A., Smith K.R., and Seto E. (2014), “Field calibrations of a low-cost aerosol sensor at a regulatory monitoring site in California,” Journal of Atmospheric Measuring Technology (AMT), Discuss, 7, 605-632, www.atmos-meas-tech-discuss.net/7/605/2014 http://www.atmos-meas-tech-discuss.net/7/605/2014/amtd-7-605-2014-discussion.html Heumann A (2011 1st ed.). Human (Formerly Heumann Design/Tech), Plugin to Rhinoceros 3D by Robert McNeel & Associates

IaaC / Cisco (2014), Citizen Sensor, citi-sense.nilu.no, http://www.iaac.net/events/citizen-sensors-fablab-barcelona-iaac-launches-a-new-crowdfunding-campaign-for-smart-citizens-269

Nabian, Offenhuber, Vanky and Ratti (2013), “Data dimension: accessing urban data and making it accessible,” Proceedings of the Institution of Civil Engineers Urban Design and Planning 166 February 2013 Issue DP1, Pages 60–75 http://dx.doi.org/10.1680/udap.12.00011 Latour, B & Yaneva, A (2008) “Give Me a Gun and I Will Make All Buildings Move: An Ant’s View of Architecture.” In R. Geiser (ed.), Explorations in Architecture: Teaching, Design, Research. Basel: Birkhäuser. 80. Lippmann M, Chen L-C, Gordon T, Ito K, Thurston GD (2013). National Particle Component Toxicity (NPACT) Initiative: Integrated Epidemiologic and Toxicologic Studies of the Health Effects of Particulate Matter Components. Research Report 177. Health Effects Institute, Boston, MA

Lynch, K (1960) The Images of the City. Cambridge: The MIT Press

Norberg-Schulz, C (1976) “The Phenomenon of Place.” Architectural Association Quarterly 8. No. 4. London: AA. Pg 3-10

Pederson, Marie, et al (2013). “Ambient air pollution and low birth weight: a European cohort study (ESCAPE)” The Lancet Respiratory Medicine, Volume 1, Issue 9, Pages 695 - 704, November 2013, http://www.thelancet.com/journals/lanres/article/PIIS2213-2600%2813%2970192-9/abstract

Rueda, S et al. (2012) El Urbanismo Ecologico: Su Application en el Diseno de un Ecobarrio en Figueres. Barcelona: Agencia de Ecologia Urbana de Barcelona.

Rutten, D (2007 1st ed.) Grasshopper & Galapagos. Seattle: Robert McNeel & Associates.

Sevillano, Elena G (2014). “La contaminación empeora un año agresivo para los alérgicos,” El Pais, 22 March 2014, Society Section. http://sociedad.elpais.com/sociedad/2014/03/22/actualidad/1395521189_792576.html

Sharp Optical Sensor (0.35mg min; 0.1mg/m3). GP2Y1010AU0F

Sunyer, J (2014), Interviews with Jordi Sunyer conducted between January and March of 2014 at the CREAL Biomedical Center, Passeo Maritimo, Barcelona

Vitiello, R and Willcocks, M (2006) The Difference is in the Detail. Place Branding Vol. 2, 3, 248-262.

WHO, World Health Organization (Organització Mundial de la Salud)

Wong, Kristine A. (2013), “Intel squeezes into smart (and crowded) sustainable cities space”, GreenBiz.Com, June 28, http://www.greenbiz.com/news/2013/06/28/intel-squeezes-crowded-and-smart-sustainable-cities-space