MAGIC Cambridge Test Site

Shiwei Fan, Ray Freshwater, Rod Jones

20 September 2018

Outline

• Monitoring– Deployment map– Monitors

New MAGIC monitorsOutdoor temperature sensorsWeather station and pyranometer

• Results– Weather data– Outdoor temperatures– Outdoor pollutants– Indoor CO2

– Controlled test

• Summary

About

Imagine a city with no air pollution or heat island…

– Current HVAC system is carbon intensive

We need to think differently…

– Natural ventilation in buildings

– Diluted air pollution levels

– Increased albedo

– Integrated green and blue spaces

– Public education and policy change

Reference site

MMM

M

MM M

M

M

M

M/T

T

MM/T

M

M/T

TM/T

M

MAGIC Field Study at Cambridge

By Shiwei Fan

First stage on 2 July 2018 (done)

Second stage 17 Aug 2018 (done)

M – MAGIC MonitorT – Temperature Logger

Weather Stations

Test Room

Coe Fen River Cam

Lamppost Column

New MAGIC Monitors

• Doubled the memory size, 3 months@30 seconds and 6 months@1minute

• Improved signal to noise ratio with new firmware

• Improved reliability by removing onboard serial to USB chip

• Features: low-power, battery powered, and portable

CO

NO2

CO2

Outdoor Temperature Sensors

• Reduce impact from solar radiation with Stevenson screens

• Optimized sensor placement to improve performance

17.0 °C

18.0 °C

19.0 °C

20.0 °C

21.0 °C

22.0 °C

23.0 °C

00:00 02:24 04:48 07:12 09:36 12:00 14:24 16:48 19:12 21:36

849428_1min

849409_2min

849427_5min

Averaged TS/N 849428 19.539 °CS/N 849409 19.498 °CS/N 849427 19.523 °C

Max difference 0.04 °C

Lab Test Results

• First stage: 8 MAGIC monitors and 6 temperature monitors on 2 July 2018

• Second stage: 10 MAGIC monitors on 17 August 2018

Outdoor Deployment

Union Rd Park Bridge

Weather Stations and Pyranometer

• Weather station and pyranometer- Transported back from LSBU- Installed on the roof of Baker

building- Close to the test room- Operating from 3 July 2018

• Data access to Chemistry weather data

Baker ChemistryPyranometer @ Jiyun Song

Win

d D

irec

tio

n (°)

Win

d S

pee

d (

m/s

)

• Wind speed at Baker building roof is lower than that at Chemistry building roof

• Wind direction is almost identical at Baker and Chemistry building roofs

Results - Weather Data

Baker Building RoofChemistry Building Roof

Prevailing wind direction is southwest

Results - Weather Data

Te

mp

era

ture

(℃

)

Outdoor Temperatures

Temperatures in the park and on the bridge are lower than others

Sky view factor?

Outdoor TemperaturesTe

mp

era

ture

(℃

)

To

tal S

ola

r R

ad

iatio

n (𝑊/𝑚

2)

Temperature is obviously related to solar radiation – it rises from sunrise, peaks at sunset and then gradually decreases overnight

Outdoor Temperatures

Park L1SAX

Outdoor Hourly Averaged COC

O (

pp

b)

Hours

• CO at different locations have similar trends

• CO looks lower in the park/river however this may be due to sensor calibration

• CO peaks at morning and evening rush hours are observed; the maximum CO occurs in late evening

• Boundary layer height may contribute to low concentration at noon

• Meteorological conditions24 July 2018

Outdoor NO2

Indoor Deployment at Architecture Cambridge

Tru

mp

ingt

on

Str

eet

MAGIC3

MAGIC4

MAGIC1

MAGIC2

MAGIC7

MAGIC5

MAGIC6

MAGIC8

ArrayTemperature Array

Camera

• Multiple sensors to account for non-homogeneous CO2

distribution

• Camera to monitor people number and window positions

Temperature Array @ Megan Davies Wykes

Indoor CO2

• Window opening events

• Higher CO2 than background (~400 ppm) to minimize estimation error

• Number of occupants

Weekend WeekendWeekend Weekend Weekend

CO

2(p

pm

)

Controlled Test

• Controlled tests on single-sided ventilation only

• Ultrasonic anemometer for wind speed/direction (4Hz)

• Decay method to estimate ventilation rates

Ventilation Rate from CO2

𝐶𝑡1 = 𝐶𝑒𝑥 +𝐹

𝑄+(𝐶𝑡0 − 𝐶𝑒𝑥 −

𝐹

𝑄)𝑒−

𝑄

𝑉(𝑡1−𝑡0)

Indoor CO2 at t1

Room volume (known)

Personal flux (estimated, uncertainty?) Occupancy (people counter)

Outdoor CO2

Ventilation rate?

Flux (gCO2n)Indoor CO2 at t0

CO2 generation rate depends on metabolic rate or level of physical activity and skin area

Controlled Test

Two windows open One window open

CO

2(p

pm

)

• Two window open: Q = 185 L/sOne window open: Q = 128 L/s

• Ventilation rate is larger with two windows open than that with one window open

Controlled TestA

ngle

(°C

)V

eloc

ity (

m/s

)

High

Low

Two windows open One window open

Summary

• Completed deployments of indoor and outdoor monitors; interpretation is ongoing

• Significantly lower temperatures are observed at green and blue spaces – sky view factor? meteorology?

• Outdoor pollutant distribution is influenced by street layout; street canyon effect is observed

• A trial controlled test was carried out

Future work

• More controlled tests with traffic monitoring added are planned– Simultaneous indoor and outdoor pollutants monitoring– Controlled indoor tests with window openings to estimate ventilation rates– Indoor and outdoor exchange during window opening events– Collect traffic data for traffic modelling