Pasadena City College Underground Car Parking Ventilation Systems Report

User Generated

ej0708

Engineering

Pasadena City College

Description

I have only an intro here. I need a background section. The background is used to educate the reader so they can understand my project. I will need sections as follows:

1. Air pollutants and health

2. Measurement devices

3. Parking structure types

4. ventilation systems

(if you don’t have access to any article, use sci-hub)

You don’t have to use all the articles. If you find a better replacement, put them on. Follow the example and the guideline I send you.

(air quality proposal draft#1 is mine)

Unformatted Attachment Preview

Research Plan Title ● A concise summary of the proposed work, include appropriate detail Abstract ● ● ● A quick, one paragraph summary of the basic premise of your proposal. Addresses the questions you will seek to answer, your need statement, and a brief description of relevance. Hint: This is the last section you should write. Background / Literature Review Intro: Introduce your topic by identifying the big picture social issue you are addressing and explaining what is currently known about the problem. ● Include a question, statistic, or phrase that will draw the audience’s attention. You need a hook that draws the reader’s/audience’s interest. The hook directly relates to your project, or the open space in your big picture. ● Explain why your topic should be of interest to the reader/audience. It would be best if you explain how your topic has local/national/global implications. ● Include 1-3 figures/schemes if they help convey meaning and support your text (referenced properly, of course). ● Hint: verbally explain the Context for your Project to 2-3 other people. This will help you organize the information into logical bites. Background: Provide the technical background necessary to understand the area you are working in. ● Identify and explain the technical concepts that are involved in your project. This is the most important aspect of your research project. Think about teaching your audience by: ○ Explaining the concepts that show why your project is worth pursuing ○ Defining key terms related to your experiment ○ If helpful, include 2-3 figures or schemes. Be sure to reference the figures properly. ● Be sure that this section flows and does not feel like a laundry list of facts. ● Explain how these concepts relate to your project. Lit Review: Provide a review of the work that has already been done in the field as well as clearly identifying the specific open space for your work. ● For a review of existing work, identify 2-3 published studies closely related to your project area. For each of the related studies, complete these steps in the following order. ○ ○ ○ ○ Briefly describe the experiments findings Describe the experiments’ observables. Give results from the experiment and what those results mean (as they relate to your proposed project area) Include a description of the “open space” that sets up your project. Problem Statement ● ● What specific problem are you addressing or working to solve? This will be your need statement. By this time in the writing, you have told the story Context → Background → Current Work → Open Space → Need Statement and the reader should truly appreciate the scope of the problem and why you want to solve it. What is your proposed solution (device, process, etc.)? This is the overview of your project. The details are in the next section. Project Planning ● Identify and describe the project details that support the completion of your solution. ● ● ● ○ Describe your successful proof of concept ○ What additional hardware, software, integration, etc. will lead to a viable solution. List the resources that you will need to complete the project. ○ This should include hardware, professional support, time, money, etc. ○ This section is the basis for a funding proposal. Provide a work plan (include a task list plus associated timeline). ○ The work plan should be sufficiently comprehensive to engender confidence that significant planning details have not been overlooked. ○ Include a project specific Gantt chart, making sure that it aligns with the course Gantt & timing. ○ Clearly identify the critical path (project component sequencing and which steps are dependent on others) Describe your validation plan. ○ What measurable data will you use to evaluate the efficacy of your project? ○ How (with what timing and what tools) will you acquire this data? ○ What analysis will be required in order to claim a successful project? Works Cited ● ● All citations must be APA format. Be sure that when referencing these works above, proper in-line citations are used. Chalk Talk - on demand A 5-10 minute presentation of your research at this point. The first couple of items should be well defined, and you should be comfortable discussing these conversationally. Items 3-5 will change as the project develops. 1. Project description - 1 sentence 2. What is the social connection 3. What have you done 4. What are you doing now 5. What do you still need to do Introduction The world is quickly urbanizing, and a key problem is that there is inadequate living and working spaces. To deal with the increasing number of automobiles, most cities in China and other crowded cities worldwide tend to save more space by turning outdoor parking or parking structures under the ground. This not only saves places for the government but also saves money for the constructors that they don’t have to buy additional lands for parking anymore. Because of the rapid growth of underground construction, it is essential for people to realize the danger they put themselves into every day, and they can’t really avoid it. Even though some people might live in a house that they can have their car park outside on the street instead of a garage, in commercial centers and their workplaces, owners tend to like their parking underground to use the land most effectively. How often do you go to the underground parking lot or stay in any indoor spaces, like an office without a window or subway stations? And how long you usually stay there? Most people remain inside the building longer than they think, and most of them don’t know about the damage from indoor pollution. “The major pollutants emitted by motor vehicles including CO, NOx, sulfur oxides, (SO), HC, lead (Pb) and suspended particulate matter (SPM), have damaging effects on human health”(Bhandarkar). They can cause harmful diseases like asthma, heart and lung issues, increase the risk of getting cancer, and premature death. All this brings up the importance of ventilation systems in indoor spaces. By modeling the Ventilation system, people can know how sufficient ventilation is and whether there is a better or cheaper way to perform the same performance. (reword) - Next paragraph - discuss parking structure design and ventilation research Problem statement The people who are indoors a lot and park underground need to know the effectiveness of the ventilation system by testing different models exist on the market, so as to avoid the damage brought by indoor pollutants. Reference Bhandarkar, S. (n.d.). Vehicular Pollution, Their Effect on Human Heatlh and Mitigation Measures. Retrieved from https://d1wqtxts1xzle7.cloudfront.net/48231041/VE004_1_2_33_40.pdf?1471849302=&respons e-contentdisposition=inline%3B+filename%3DVehicular_Pollution_Their_Effect_on_Huma.pdf&Expires =1612498834&Signature=RymQIMiAi17m1KcUT9LnhOcIk3p8uvsdH7n9MXD0heYkCXOgI1 TBCgiBOoRx0OGXedaBLVTFk2qAVfyrD7CWUHB~aNetLmoJyTI9cHSDktYO24iUgr7Ckjts X~C974LgUT8sMEsvqHiFxdag6Ra8n7gCMXkubciQ8rFRgt6cU~GIiTQ6RvjzTJhjFpP5J5IPwlhS-rpeZhV6GaKKD71nmm3zvVoabxJwq3BBUhrhIGFUefC0qiwpysh7caOgcFsMTtfUxM7OjDqGL267oEMqV2 ON0iKOGOnNv6fDa7dAqwe9GEqhGybEw1n1u9PV2cRn90J8cY1TNf1xr-MxGRoA__&KeyPair-Id=APKAJLOHF5GGSLRBV4ZA. https://www.pnas.org/content/111/23/8386.abstract?etoc https://www.sciencedirect.com/science/article/abs/pii/S036013231730358X Read article on HVAC in garages http://blog.dwyer-inst.com/2020/03/04/understanding-ventilation-efficiency-and-dcv-in-parkingstructures-and-garages/#sthash.piGTBDu4.dpbs What is building code for China? ● https://www.gbpn.org/sites/default/files/China_Country%20Summary_0.pdf https://scihub.wikicn.top/10.1016/j.apr.2020.08.008 Indoor air quality investigation at air-conditioned and non-air-conditioned markets in Hong Kong https://doi.org/10.1016/j.scitotenv.2003.09.031 2020 Kim Paper --Exposure to respirable particles and TVOC in underground parking garages under different types of ventilation and their associated health effects; Oh, HJ., Sohn, JR., Roh, JS. et al. Exposure to respirable particles and TVOC in underground parking garages under different types of ventilation and their associated health effects. Air Qual Atmos Health 13, 297– 308 (2020). https://doi.org/10.1007/s11869-020-00791-0 I have only intro here. I need a background section. The background is used to educate the reader so they can understand my project. I will need sections as follows: 1. Air pollutants and health 2. Measurement devices 3. Parking structure types 4. ventilation systems (if you don’t have access to any article, use sci-hub) You don’t have to use all the articles. If you find better replacement, put them on. Follow the example and the guideline I send you. Asmar 1 Kristen Asmar Honors Engineering for Social Good Research Proposal Identifying Fatigue in the Vastus Lateralis Muscle in Athletes Recovering from Knee Injury Abstract Athletes overuse their muscles when doing rehabilitation exercises in an attempt to shorten their rehabilitation process and return to their exercises or sports. In order to prevent overusing muscles, this device warned users when his/her muscle is fatigued and therefore to stop rehabilitation for that day. The hinged brace told the user when he/she has reached the desired angle for hamstring curls and when he/she has held his/her knee for the desired amount of time. For validation, the brace was tested on someone who tires easily, someone who was recovering from a knee injury–making sure they were well into their rehab in order to prevent additional damage when testing–or on a healthy person by inducing fatigue. For a control, the knee brace was tested on someone who is not recovering from a knee injury. In the future, the brace can be modified for active rehabilitation, controlling the patient’s knee for him/her and stopping the movements when his/her muscle is fatigued. Intro In an effort to shorten their rehabilitation period and improve their injury faster, athletes over do physical rehabilitation. There is a misconception that more rehabilitation will equal more recovery, so atheltes perform their rehabilitation exercises longer than necessary (Hilliard, Asmar 2 Robert C, et al, 2017). As a result, the athletes overuse their recovering muscles and cause more damage to their injuries by tearing the damaged ligaments more or by halting the progress due to damaging the muscle after having rehabilitated it some. This device will use electromyography, encoders, and LEDs as a warning system for muscle fatigue. The millivolt output of the EMGs, which is based on amplitude, will be inputted into algorithms to calculate the root mean square and median in order to take the raw EMGs and transform them into quantitative data. The EMGs will be taken through skin sensors placed on the vastus lateralis muscle, which is on the outer side of the thigh. Once the root mean square increases and the median frequency decreases to a certain amount, the user will be warned that his/her muscle is fatigued. If time allows, encoders will be added to the brace to let the users know when they have reached the desired angle for hamstring curls. The additions to test all of this will be added to a neoprene knee brace with hinges, making sure to keep the knee brace light and mobile. Background: After talking with the assistant athletic trainer at the University of La Verne, Meg Ryan, it was decided there was a gap in rehabilitation for athletes. It is important to receive feedback about the muscles that athletes are trying to recover because it can lead to more damage to the already hurt muscles. It would be beneficial to receive biofeedback for how much the muscle is firing so as to get a better grasp at how well the muscle is recovering from surgery or injury (Ryan). In a gait training study done on children with cerebral palsy, the group who received biofeedback during the training did statistically better in regards to muscle movement (Dursun, Dursun, Alican, 2003), suggesting that biofeedback is favorable and important in therapy. Asmar 3 Therefore, it would be beneficial to integrate sEMG sensors into the knee brace so that raw EMG signals can be converted to root mean squares (RMS) and median frequencies (MF) in order to muscle firing information. Electromyography, or EMGs, is the technique used to record and analyze the electrical activity of skeletal muscles. EMGs are useful when determining how healthy muscles and motor neurons are. The EMG itself is a measurement of the electrical speed and strength between two points on the body, in response to nerve stimulation, such as contracting the muscle. sEMGs are surface EMG measurements that use electrodes that stick directly onto the skin in order to measure muscle electrical activity, as opposed to needle EMGs, which are inserted directly into the body. EMGs can be used to determine when muscles are fatigued through MF and RMS. A fatigued muscle caused MF, the frequency value that consists of two parts of equal energy that separates the power spectrum, which can be determined through algorithms within code. RMS is value after averaging the amplitudes of the raw EMG signals. When the median frequency lowers and the value of the root mean square increases, the muscle is fatigued. Lit Review #1: Development of an EMG-Controlled Knee Exoskeleton to Assist Home Rehabilitation in a Game Context Mingxing Lyu, Wei-Hair Chen, Xilun Ding, Jianhua Wang, Zhongcai Pei, and Baochang Zhang made an EMG exoskeleton knee rehabilitation device for patients who are in the initial phases of recovery after a stroke. Previous rehab exoskeletons required assistance from a physical therapy and can cost patients a large amount of money to use, so this device was made to be usable at home so as to lower the price of rehab. Also, this device motivates the user to complete rehab exercises because of the game attached. In a study done on stroke patients performing hand rehabilitation through a gamified approach, therapists and patients provided positive feedback to Asmar 4 gamified rehabilitation processes because of an increase of interest and motivation to complete the rehab exercises (Carneiro, et al., 2018). The creators modified the app Flappy Bird so that the patients’ knee movements control the height of the bird. To translate a knee movement to flight of the bird, the device uses a Myo thigh-band, which consists of 16 dry surface EMG sensors and is easily adjustable. The Myo thigh band has a sampling frequency of 200 Hz and sends raw sEMG data to the host computer via Bluetooth Low Energy. A Kalman filter is used to calm down the noisiness of raw EMG data. The exoskeleton itself assists the patients with both hip and knee movements and is made out of aluminum and 3D-printed parts so as to limit the weight of the exoskeleton. Lit Review #2: A Comparative Study of EMG Indices in Muscle Fatigue Evaluation Based on Grey Relational Analysis during All-Out Cycling Exercise Lejun Wang, Yuting Wang, Aidi Ma, Guoqiang Ma, Yo Ye, Ruijie Li, and Tianfeng Lu studied different EMG indices in muscle fatigue in all-out cycling exercises in order to determine which EMG calculation was most reliable for detecting muscle fatigue in cyclists. Muscle fatigue is identified through a reduced capacity to generate force or create power output. Grey relational analysis was used to quantify the change of trends in identified sequences by using grey relational grade. The EMG sensors were placed on the vastus lateralis muscle, as it the muscles most responsible for power in cycling exercises. The indices studied were EMG root mean square (RMS, median (MF), and mean power frequency (MPF) based on Fourier Transform. 10 volunteer cyclists completed a 5 minute warm up, 3 minute rest, then a 30 second high power session. The scientists found that the MF and MPF were the more reliable than EMG RMS when determining when the muscles become fatigued, which was when the MF and MPF showed a significant decreasing tendency. But, the MF and MPF are limited by the Fourier Transform, Asmar 5 wavelet transformation has been observed as an alternate when dealing with more dynamic contractions. Overall, MF derived from wavelet transformation was the most reliable EMG indice when determining muscle fatigue because it had the highest grey relational grade. Lit Review #3: A Muscle Fatigue Monitor Based on the Surface Electromyography Signals and Frequency Analysis Patients completing rehabilitation may experience muscle fatigue or may further damage their injury without knowing. In order to alert the user of these unknown pains, Gutiérrez, Cardiel, and Hernández created a way to monitor when muscle fatigue is present. The two factors to determine when the muscle was fatigued are: a shift of the median frequency in the power spectrum density and the Borg scale criteria related to pain. EMG signal processing was used to determine when the median frequency shifted and the root mean square value was measured within the time domain in order to determine when the muscle is fatigued. Both the values for median frequency and root mean square were determined by using algorithms programmed in LabVIEW. The muscle was determined to be fatigued when the value of the median frequency decreases, while the root mean square values increase, meaning the frequency of signals of sEMGS, but the amplitudes of the signals increased, reflecting the overexertion of the muscle. The rehab exercise I will focus on is hamstring curls. The exercise consists of the patient holding on the back of a chair and bending their knees towards their back as far as they could. The patient should hold their leg for 5 seconds once they reach the angle. The process I will create will make sure the patient completes 3 sets of 10 hamstring curls. After each set, the knee brace will encourage the patient to bend their knee more. Hamstring curls were chosen because it is a popular rehabilitation exercise for recovering athletes, according to the American Academy Asmar 6 of Orthopaedic Surgeons, and it will mostly be used as a proof of concept exercise because it involves angles and movement of the knee. Driving Questions, Hypothesis, Open Space for New Work Driving Question: Will this device prevent the overuse of muscles? Hypothesis: If muscle fatigue is detected in the vastus lateralis muscle of athletes recovering from knee injury, then the athlete may recovery more fully due to limiting further injury during the rehab process. Open Space for New Work: In the future, the knee brace can be improved by adding a mechanical controlling device that raises the knee for the patient when the patient is not ready for active rehabilitation. The knee brace would still detect when the muscle is fatigued and when to stop the knee from bending, but the knee brace would be the device controlling and lifting the knee, rather than the patient. Methods/Materials Methods: The proposed device is a knee brace that detects fatigue in the vastus lateralis and the desired angle for hamstring curls. The myoware sensor and EMG skin sensors will be used to measure the EMG signals of the muscle. For fatigue detection, RMS and MF will be calculated using the algorithms found in the study, A Muscle Fatigue Monitor Based on the Surface Electromyography Signals and Frequency Analysis (Gutiérrez, et al., 2016). The raw data from Asmar 7 the EMGs is the millivolt output of the signals based on amplitude. Those millivolt values will be inputted into the algorithms, which will generate a numerical value that will be analyzed, possibly every second. When the RMS and MF reach a certain point, the RMS increasing and the MF decreasing enough, the brace will notify the user that the muscle is fatigued with an audible noise from an Arduino speaker. If time allows, encoders will be used to detect and control the angle at which the knee is bent. LEDs will be used to warn the user when he or she has reached the desired angle by using green, yellow, and red lights, so that there will be no learning curve. Another LED will be used as a timer, blinking every second, so the user will know how much time has passed since reaching the desired angle. The knee brace will be coded using Arduino and the additional parts will be added to an already functioning knee brace. The knee brace chosen is a combination of a neoprene sleeve and hinges, in order to stabilize the knee enough during the exercise and have hinges available if encoders can be added. If live data acquisition is not possible due to lack of coding proficiency, then a backup plan is to download the data to be later analyzed by a therapist. The injured person will go through the entire rehabilitation protocol while using the knee brace. At the end of the protocol, the data will be downloaded so that the therapist can note when or if fatigue was reached. This knowledge can advise the therapist on if adjustments to the patient’s rehabilitation protocol are necessary. For validation, I will test the device on a person who tires easily, is recovering from a knee injury, or on a healthy person. The recovering athlete and the healthy person are the most plausible clients because they provide their own controls. For the athlete, the control is the noninjured knee, and for the healthy person, the control is data acquisition of the muscle that was not Asmar 8 forced to fatigue. If the healthy person is the chosen client, then in order to test the fatigue detection, the muscle of the client needs to be forced into fatigue through prolonged knee movement or short bursts of exercise. Materials: Knee braces 1. https://www.shockdoctor.com/products/ultra-knee-support-with-bilateralhinges?variant=14354302435381 a. $69.99 Arduino Nano 1. https://store.arduino.cc/usa/arduino-nano a. $22.00 Arduino speaker 1. https://www.adafruit.com/product/1890 a. $1.95 Arduino 1. https://store.arduino.cc/usa/arduino-uno-rev3 a. $22.00 MyoWare sensor (detects EMGs) 1. https://www.adafruit.com/product/2699 a. $37.95 3M Red dot EMG surface sensors (100) Asmar 9 1. https://www.amazon.ca/Red-Dot-Multi-Purpose-MonitoringElectrode/dp/B01AME7YC0/ref=pd_sim_147_1/136-88739542404105?_encoding=UTF8&pd_rd_i=B01AME7YC0&pd_rd_r=c7d59e69-ec62-4e4d8ccd-7e2385b8ca76&pd_rd_w=JxHMQ&pd_rd_wg=f54pn&pf_rd_p=ca0a769f-467a4dd2-b2dc9f16d6aac0a1&pf_rd_r=QR7R55TVXNPB6TT1G0J7&psc=1&refRID=QR7R55TVXN PB6TT1G0J7 2. $23.91 Timeline Works Cited: Asmar 10 American Academy of Orthopaedic Surgeons. Knee Conditioning Program. (n.d.). Retrieved December 19, 2019, from https://orthoinfo.aaos.org/globalassets/pdfs/2017rehab_knee.pdf. Carneiro, F., Tavares, R., Rodrigues, J., Abreu, P., & Restivo, M. T. (2018 ,November). A Gamified Approach for Hand Rehabilitation Device. Retrieved from University of Port website: https://onlinejour.journals.publicknowledgeproject.org/index.php/ijoe/article/viewFile/7793/4788 Cifrek, M., Medved, V., Tonkovíc, S., & Ostojíc, S. (2009, January). Surface EMG based muscle fatigue evaluation in biomechanics. Retrieved from Clinical Biomechanics website: https://www.researchgate.net/profile/Mario_Cifrek2/publication/24200364_Surface_EM G_Based_Muscle_Fatigue_Evaluation_in_Biomechanics/links/59f7759d0f7e9b553ebee4 6b/Surface-EMG-Based-Muscle-Fatigue-Evaluation-in-Biomechanics.pdf Copaci, D., Martín, F., Moreno, L., & Blanco, D. (2019, March). SMA Based Elbow Exoskeleton for Rehabilitation Therapy and Patient Evaluation. Retrieved from IEEE website: https://ieeexplore.ieee.org/document/8658083 Dursun, E., Dursan, N., & Alican, D. (2003, September). Effects of biofeedback treatment on gait in children with cerebral palsy. Retrieved from Taylor and Francis website: https://www.researchgate.net/profile/Erbil_Dursun2/publication/8966208_Effects_of_bio feedback_treatment_on_gait_in_children_with_cerebral_palsy/links/0deec524d23761ca5 a000000Effects-of-biofeedback-treatment-on-gait-in-children-with-cerebral-palsy.pdf Guo, S., Ji, J., & Tao, S. (2014, December). Lower limb rehabilitation robot for gait training. Retrieved from ResearchGate website: Asmar 11 https://www.researchgate.net/profile/Shuai_Guo17/publication/278395290_Lower_limb_ rehabilitation_robot_for_gait_training/links/561335d708aea9fb51c28ff5.pdf Hilliard, R. C., Blom, L., Hankemeier, D., & Bolin, J. (2017, May). Exploring the Relationship Between Athletic Identity and Beliefs About Rehabilitation Overadherence in College Athletes. Retrieved December 19, 2019, from National Center for Biotechnology Information website: https://www.ncbi.nlm.nih.gov/pubmed/27632827 Krebs, H. I., Ferraro, M., Buerger, S. P., Newbery, M. J., Makiyama, A., Sandmann, M., . . Hogan, N. (2004, October). Rehabilitation robotics: pilot trial of a spatial extension for MIT-Manus. Retrieved from BioMed Central website: https://jneuroengrehab.biomedcentral.com/track/pdf/10.1186/1743-0003-1-5 Lyu, M., Chen, W.-H., Ding, X., Wang, J., Pei, Z., & Zhang, B. (2019, August 27). Development of an EMG-Controlled Knee Exoskeleton to Assist Home Rehabilitation in a Game Context. Retrieved December 19, 2019, from Frontier in Neurorobotics website: https://www.frontiersin.org/articles/10.3389/fnbot.2019.00067/full Mayo Clinic. (n.d.). Electromyography (EMG). Retrieved December 19, 2019, from Mayo Clinic website: https://www.mayoclinic.org/tests-procedures/emg/about/pac-20393913 Narang, Y. S., Arelekatti, V.N. M., & Winter, A. G. (2015, July). The Effects of Prosthesis Inertial Properties on Prosthetic Knee Moment and Hip Energetics Required to Achieve Able-Bodied Kinematics. Retrieved from IEEE website: https://ieeexplore.ieee.org/document/7155597 Wang, L., Wang, Y., Ma, G., Ye, Y., & Lu, T. (2018, April 16). A Comparative Study of EMG Indices in Muscle Fatigue Evaluation Based on Grey Relational Analysis during All-Out Asmar 12 Cycling Exercise. Retrieved December 19, 2019, from Hindawi website: https://new.hindawi.com/journals/bmri/2018/9341215/ Wu, Q., Wang, X., Chen, B., & Wu, H. (2011, October). Design and Fuzzy Sliding Mode Admittance Control of a Soft Wearable Exoskeleton for Elbow Rehabilitation. Retrieved from IEEE website: https://ieeexplore.ieee.org/document/8489867 Zhi, Y. X., Lukasik, M., Li, M. H., Dolatabadi, E., Wang, R. H., & Taati, B. (2017, December). Automatic Detection of Compensation During Robotic Stroke Rehabilitation Therapy. Retrieved from IEEE website: https://ieeexplore.ieee.org/document/8214256 Environmental Pollution 247 (2019) 626e637 Contents lists available at ScienceDirect Environmental Pollution journal homepage: www.elsevier.com/locate/envpol On-site assessments on variations of PM2.5, PM10, CO2 and TVOC concentrations in naturally ventilated underground parking garages with traffic volume* Zhijian Liu a, *, Hang Yin b, Shengyuan Ma a, Gaungya Jin a, Jun Gao c, **, Wenjun Ding d a Department of Power Engineering, North China Electric Power University, Baoding, Hebei, 071003, PR China Department of Civil Engineering, Technical University of Denmark, DK-2800, Kgs, Denmark School of Mechanical Engineering, Tongji University, Shanghai, 200092, PR China d Laboratory of Environment and Health, College of Life Sciences, University of Chinese Academy of Sciences, No.19A Yuquan Road, Beijing, 100049, China b c a r t i c l e i n f o a b s t r a c t : Article history: Received 5 November 2018 Received in revised form 14 January 2019 Accepted 23 January 2019 Available online 25 January 2019 There have been an increasing number of automobile vehicles in cities, so that newly developed residential areas are mostly designed with underground parking garages (UPGs). For naturally ventilated UPGs, the ventilation performance may be insufficient to discharge totally vehicle-induced pollutants out of the enclosed underground spaces, which consequently results in threats to residents' health. This study, therefore, aims at examining the patterns of pollutant concentrations in naturally ventilation UPGs as well as their sensitivities to traffic volume. In particular, the naturally ventilated UPGs’ weekday particulate matters (PM2.5 and PM10), CO2 and TVOC concentration as well as their relationships between traffic volume were quantitively evaluated based on field measurements in eight residential areas in Baoding, China. Results indicated that daily average PM2.5, PM10, CO2 and TVOC concentrations in studied UPGs were 105.81 mg/m3, 464.17 mg/m3, 571 ppm and 24 ppb, respectively. The PM2.5 concentrations in UPGs were slightly higher than that in ambient environments, while the PM10 concentrations in UPGs were significantly higher. Furthermore, both PM10 and TVOC concentrations in UPGs were in significant relationships with traffic volume at the p < 0.01 level, while the concentration of UPG PM2.5 generally exhibited a significant correlation (p < 0.01) with that of the ambient. Nevertheless, a combination of traffic volume, the ambient and accumulative effect was much better to explain the hourly PM10 concentration in UPGs. These findings will be conducive to instruct engineers with fundamental knowledge of UPG ventilation design. © 2019 Elsevier Ltd. All rights reserved. Keywords: Underground parking garage Natural ventilation Indoor pollutant concentrations Daily variation characteristics Correlation analysis Multiple regression model 1. Introduction Significant growth of automobile number has been witnessed for the last few decades in China. Therefore, the construction of underground parking garage (UPG) is popular for newly-built living quarters due to the demand for parking space. Due to the severe energy situation and the worldwide wave of urban sustainability (He et al., 2018), most underground garages still use natural ventilation systems. Thus, the air quality of UPGs under natural * This paper has been recommended for acceptance by Dr. Haidong Kan. * Corresponding author. ** Corresponding author. E-mail addresses: zhijianliu@ncepu.edu.cn (Z. Liu), gaojun-hvac@tongji.edu.cn (J. Gao). https://doi.org/10.1016/j.envpol.2019.01.095 0269-7491/© 2019 Elsevier Ltd. All rights reserved. ventilation has become a hot spot. Compared with the temperature, humidity, particulate matter (PM) and biological particles that are often studied in previous research of indoor pollutants (Liu et al., 2018a; Liu et al., 2018b), it is necessary to consider the impact of traffic volume on the indoor pollution of UPGs. The pollutants including carbon monoxide (CO), carbon dioxide (CO2), nitrogen dioxide (NO2), particulate matter (PM), volatile organic compounds (VOCs), etc. generated from automobiles would be concentrated and accumulated in this enclosed or semi-enclosed micro-environments. Although there is no specific guideline or standard on the air quality of UPGs, it is possible to quantify the health risks of a large part of pollutants presented in garages, which including 39 compounds (Glorennec et al., 2008). Those who work in UPGs, such as security staff and cleaners, would stay for a long time in this environment. While car park users only enter the garage 2 to 3 Z. Liu et al. / Environmental Pollution 247 (2019) 626e637 times a day and the duration time of staying in garage is very short, but each time they exposure to high concentrations of air pollutants. Therefore, exposure to traffic-related air pollutants in UPGs would cause adverse health effects to both car park workers and users (Bowatte et al., 2018; Costa et al., 2017; Li et al., 2014; Tham et al., 2018; Zhang et al., 2017). Many researchers have performed studies on the air quality of UPGs, and research methods include numerical simulation and field measurement (Chan and Chow, 2004; Jokl, 2000). For example, Papakonstantinou et al. developed a CFD model for the simulation of CO levels inside a typical central garage under natural and mechanical ventilation, respectively, to investigate the ventilation effectiveness(Papakonstantinou et al., 2003). Song et al. simulated particle concentration distribution under transient natural wind through the entrance of the UPG used three-dimensional numerical model(Song and Zhao, 2017). Zhao et al. investigated the transient air flow, particle dispersion, and particle exposure risk during and after transient vehicle exhaust in an UPG(Zhao and Zhao, 2016). In addition, most of studies on air quality of UPGs were conducted on field measurement. Vukovi et al. compared the instrumental measurements with active moss biomonitoring in terms of the measuring accuracy of heavy metals and PAHs concentrations in PM10 samples collected in four parking garages. The parking facilities selected as the research object were different in size, structure, the number of parking spaces and the type of ventilation system(Vukovic et al., 2014). Demir et al. measured the air pollutant concentrations in terms of exhaust emissions of vehicles in multistory car parks under different ventilation system including natural, ductwork and jet fan ventilation systems, and determined that short-term exposure of humans to the environment of this car park would not cause health problems(Demir, 2015). Mar c et al. measured the BTEX compounds concentrations of a two-level UPG(Marc et al., 2016), while Hun et al. investigated the indoor BTEX and MTBE concentrations in residences with an attached garage, a detached garage or carport for natural ventilation(Hun et al., 2011). The results indicated that gasoline-related VOCs from parked cars especially in attached garages would deteriorate indoor air quality. Batterman et al. also measured the VOC concentrations at 15 residential garages during 4-day measuring period and concluded that VOC concentrations were not strongly correlated to air exchange rates (AERs), but appeared largely dependent on door opening and VOC sources(Batterman et al., 2006). Zhao et al. investigated the seasonal patterns of PM10, PM2.5 and PM1.0 in a naturally ventilated UPG, concluding that the natural ventilation could not guarantee the quality of indoor air in this environment because the daily average PM10 and PM2.5 in the garage generally exceeded the long-term (24 h) exposure limit of the Chinese standard especially in winter(Zhao et al., 2017). Moreover, the Pearson correlation and a mass-balance model were employed to determine the impacts of traffic flow, AER and outdoor particles. The relationship between indoor pollutants and traffic volume has been found in many studies(Ho et al., 2004; Hun et al., 2011; Zhao et al., 2017). Moreover, accumulative effect of indoor pollutants was taken into consideration in the study of Ho et al. (2004). Kim et al. evaluated the effect of traffic volume on air pollution levels used a multiple regression longitudinal model containing temperature, relative humidity, wind speed and traffic activity (Kim et al., 2007). Because of the high density of urban population, there have been an increasing number of high-rise buildings in China. Meanwhile, the quantities of own cars are increasing by a big margin in residential quarters, so that newly developed residential areas are mostly designed with UPGs to solve the problem of parking space shortage in residential areas. Therefore, the phenomenon of widespread UPGs in densely populated areas has become common 627 in China. Moreover, for most residential UPGs in China, natural ventilation is the most popular ventilating mode, considering construction costs and operating costs. Only a few residential UPGs are designed with mechanical ventilation, but the ventilation system dose not operate throughout the year. Additionally, there is no study aimed at Baoding region (Hebei, China), which the value of Air Quality Index (AQI) often ranked the top 5% among 338 cities in China. Therefore, it is the first time for us to investigate the air quality of residential UPGs under natural ventilation in Baoding city, China. The air pollutant concentrations in parking facilities are affected by many factors such as traffic volume, ambient condition and the construction design of UPGs. However, only a few studies have focused on systemically quantifying the influence factors of indoor air quality. In our study, field measurements were conducted to gain various data of air pollutant concentrations and traffic volume in eight naturally ventilated UPGs in the urban area of Baoding, China, aiming at examining the patterns of pollutant concentrations in naturally ventilated UPGs as well as their sensitivities to traffic volume. In particular, the naturally ventilated UPGs’ weekday particulate matters (PM2.5 and PM10), CO2 and TVOC concentration as well as their relationships between traffic volumes were quantitively evaluated by SPSS 16.0. Moreover, the ambient condition and accumulative effect were introduced into multiple regression model to explain the hourly PM10 concentration in UPGs. Furthermore, the over standard rates of indoor pollutant concentrations based on both international and Chinese national standards were also presented in this study. 2. Methods 2.1. Sample description Eight single-layer UPGs with natural ventilation were selected as measurement objects in this study, and all the samples are located in the central of Baoding city (38.85 N, 115.48 E). The size, structure, the design of air vent, the number of parking spaces and the traffic volume were different in each UPGs. General information about all the measured UPGs was listed in Table 1. It is widely known that the air quality of the enclosed environment heavily depends on the effectiveness of its ventilation system. While for naturally ventilated UPGs, the air quality is mainly affected by its characteristics containing the volume, number, size, as well as the position of air vents and entrances/exits. These detailed information was also listed in Table 1. 2.2. Measurements The field measurements were conducted on weekdays from March 1st to April 10th, 2018, and the measuring time of each sampling site was from 0:00 a.m. to 24:00 p.m. All measurements were taken during weekdays because there are apparent morning and evening rush hours in nearly all the residential UPGs. The data of outdoor particle matter concentration (PM2.5 and PM10) were directly collected from local meteorological environment monitoring stations, with the time interval of 1 h. The locations of sampling sites and all the meteorological environment monitoring stations in Baoding city were shown in Fig. 1, and the distance between the sampling site and the nearer monitoring station was listed in Table 1. For indoor parameters, the temperature, relative humidity, PM2.5, PM10, CO2 and TVOC concentrations were measured and the measurement range and accuracy was listed in Table 2. The instruments were installed at position 1.5 m above the floor in the main parking space away from the entrance or exit. All data were collected continuously (during 24 h) in each sampling 628 Z. Liu et al. / Environmental Pollution 247 (2019) 626e637 Table 1 General information about the all the measured underground parking garages. Case No. Area (m2) Height (m) Volume (m3) Parking space Entrance/Exit a N b Air vent Monitoring station S1 (m) d P N c Distance (km) Huadian Ⅱ 2.8 Natatorium 1.0 Huadian Ⅱ 1.7 Huadian Ⅱ 1.7 Huadian Ⅱ 1.2 City Monitoring Station 1.3 Huadian Ⅱ 1.3 Huadian Ⅱ 1.0 S2 (m) 1 8720 3.5 30520 205 1 6.3  3.3 e 1 4.4  4.5 2 5110 4.0 20440 125 2 Same side 0 e 3 5300 3.0 15900 200 1 4.6  3.5 10.0  3.5 7.3  3.0 e 0 e 4 3200 3.2 3200 108 2 Opposite side 0 e 5 1690 4.8 1690 60 2 Opposite side 2 6.4  5.4 6 7 1930 6535 4.5 3.5 1930 6535 60 215 1 2 e Same side 0 0 e e 8 5985 3.5 5985 232 2 Same side 0 e 7.4  3.0 7.4  3.0 3.9  2.3 3.9  2.3 7.0  3.5 7.5  3.3 7.5  3.3 4.6  3.3 4.6  3.3 Abbreviation: aN- Number; bS1-Size, width  height; cS2-Size, Length  width (The height of the side air vent of the ventilation well is 0.4 m); dP- Position. site with a time interval of 1 min. The measurements of traffic volume were carried out manually by two testers standing in each gate of the garage. They recorded the number of cars entering and leaving the garage every 15 min during the period of measurement. tþð D=2 2.3. Statistical analysis CDt ðtÞ ¼ The wastes generated from automobile were classified into evaporative and exhaust emissions based on the types of emissions by the U.S. Department of Transportation (2002) (Kristanto, 2006). The amount of VOCs generated from fuel evaporation when vehicles are parked in the garage could be totally ignored because the indoor temperature of UPGs is stabilized at less than 10  C. Thus, only the impact of exhaust emission is focused in this study because most pollutants (VOCs, PM, CO and NOx) are generated from the tailpipe when the vehicle is in operation (Sawyer et al., 2000). Therefore, the engine operating time of vehicles could directly affect the air quality in the UPG and the hourly engine running time is linearly correlated with traffic volume (Ho et al., 2004; Kristanto, 2006). However, the engine running time may differ from 60 to 180 s due to the different size and layout of the parking garage according to ASHRAE (ASHRAE, 2007), therefore, the effect of the traffic-related indoor pollutant concentration not only depends on the traffic volumes but also depends on the size and ventilation of the parking garage. In our study, the average ratio of occupied parking spaces in eight UPGs is up to 92%, which means the total number of cars in the garage is equal approximately to the total parking space in the garage. Therefore, the average traffic activity (ATA) was defined as the ratio of the total traffic volume (Ntot) to total number of parking spaces (Nps) in the UPG. It is given by equation (1): Ntot ATA ¼ Nps (1) Space utilization ratio (b) was defined as the ratio of the total area of parking space to that of the garage, and it is given by equation (2): b¼ Nps  A0 A in order to analyze the correlation of indoor and outdoor pollutant concentrations, as well as that of outdoor pollutant concentrations and traffic volume, the measured data should be converted to the same time interval calculated by ) (Ho et al., 2004): (2) A0: The average area of a parking space (2.6  5.3 m). The time intervals of the measured data including traffic volume, indoor and outdoor pollutant concentrations were different from each other, 15 min, 1 min and 60min, respectively. Therefore, Cðt 0 Þdt (3) tD=2 The average values over time intervals of 15 min and 60 min were denoted by C15(t) and C60(t), respectively. The impacts of traffic volume and ambient environment on indoor air pollutants were determined by employing Pearson correlation by SPSS 16.0. Due to the different sources and propagation mechanisms of indoor pollutants, their concentration is not only linearly related to one influence factor. Therefore, a multiple regression was performed to correlate hourly indoor pollutant concentration with the traffic volume, ambient level and accumulative effect. The model is given by equation (4):     C in ¼ a C in t t1   þ bTt þ g C out þ ε t (4) where Cin is the hourly average indoor concentration, Cout is the hourly average outdoor concentration and T is the hourly number of cars entering or leaving the garage. The multiple linear regression model of hourly pollutant concentration in garage was also analyzed by SPSS 16.0. 3. Results and discussion 3.1. Comparison of indoor and outdoor pollutant concentrations The summary of the indoor thermal environment and daily average indoor and outdoor pollutant concentrations in each sample was listed in Table 3, and the traffic-related information was shown in Table 4. Compared with indoor and outdoor PM2.5 concentrations, it is found that the indoor PM2.5 concentration is generally slightly higher than the outdoor, with the average I/O value of 1.16. The daily average indoor PM2.5 concentration of Case 2 was the highest among all the samples that was 199.60 mg/m3, followed by Case 1, Case 6, Case 8, Case 7, Case 3, Case 4 and Case 5 that was 187.90 mg/ m3, 113.73 mg/m3, 101.97 mg/m3, 88.92 mg/m3, 57.71 mg/m3, 49.66 mg/ m3 and 46.97 mg/m3, respectively. While the traffic volume of Case 1 Z. Liu et al. / Environmental Pollution 247 (2019) 626e637 629 Fig. 1. The location of sampling sites and monitoring stations in Baoding, China. Notations: the AQI data used in the above figure are derived from the real-time data (6:00 p.m. 12/ 04/2018) provided by the monitoring stations during the survey period. Table 2 The measurement range and accuracy. Measured parameters Instrument Measurement range Accuracy PM2.5/PM10 CO2 Temperature RH TVOC TSI 8520 TSI 7515 TSI 8392A 0.1e10 mm 0e5000 ppm 17.8e93.3  C 0e95% 0e50 ppm ± 0:1% ± 50ppm ± 0:3 C ± 3% þ1.5% Particle Plus 7301-AQM was the highest that was 455 vehicles per day, followed by that of Case 3, Case 8, Case 2, Case 4, Case 7, Case 5 and Case 6 that was only 56 vehicles per day. In conclusion, the indoor PM2.5 concentration was mainly affected by outdoor environment rather than traffic volume, so traffic-related emissions are not the dominant contributors to indoor PM2.5 concentration. However, for PM10, the daily average concentration indoors was much higher than that outdoors, with the average I/O value of PM10 was about 4.0. The average indoor PM10 concentration of Case 3 was the highest among all the samples, and the value was 913 mg/m3 which was 15.741 times of the outdoor (58 mg/m3). While the indoor PM10 concentrations of Case 4 and Case 5 were the lowest which was 238.44 mg/m3 and 246.39 mg/m3, respectively, and the outdoor concentrations were 65 mg/m3 (I/O ¼ 3.668) and 74 mg/m3 (I/O ¼ 3.330). The results indicated that the concentration of PM10 measured at the sampling site was generally far higher than the ambient concentration. In addition, the daily average concentration of PM10 in the garage was correlated to the total number of vehicles entering or leaving this garage, where the Pearson correlation coefficient was 0.605 (p > 0.05). It is concluded that the traffic-related emissions rather than ambient pollutants are the dominant contributors to the PM10 concentration of the UPGs. 630 Z. Liu et al. / Environmental Pollution 247 (2019) 626e637 Table 3 The indoor thermal environment and daily average indoor and outdoor pollutant concentrations in each sample. Case No. Indoor (N ¼ 1441) Temperature ( C) Relative humidity (%) PM2.5 (mg/m3) PM10 (mg/m3) CO2 (ppm) TVOC (ppb) Outdoor (N ¼ 25) PM2.5 (mg/m3) PM10 (mg/m3) 1 2 3 4 5 6 7 8 Avea SDb Ave SD Ave SD Ave SD Ave SD Ave SD 8.94 9.91 40 4.13 187.90 34.43 535.18 197.92 535 45.77 10 21.85 10.82 5.92 54 3.57 199.60 39.14 477.56 138.65 529 31.12 14 20.00 9.71 7.47 28 1.52 57.71 16.18 913.00 782.65 500 43.80 37 72.15 8.66 9.21 36 2.87 49.66 8.84 238.44 164.72 534 56.86 15 27.02 7.99 12.41 24 2.54 46.97 10.41 246.39 213.41 421 23.18 4 13.35 8.58 5.02 45 3.11 113.73 29.14 440.29 257.92 471 23.01 46 79.94 11.70 7.51 65 6.33 88.92 44.34 303.72 167.79 806 79.00 29 54.20 13.29 14.57 48 7.44 101.97 48.58 558.75 187.86 772 83.22 36 63.72 Ave SD Ave SD 194 60.13 277 69.56 197 44.59 281 88.00 32 16.80 58 22.68 38 7.67 65 15.66 43 13.65 74 19.62 116 30.19 178 47.55 85 41.07 181 30.18 94 63.10 400 260.75 0.968 1.932 1.013 1.700 1.804 15.741 1.307 3.668 1.092 3.330 0.980 2.473 1.046 1.678 1.085 1.397 I/O c PM2.5 PM10 Notations: a Average; b Standard derivation; c I/O The ratio of 24 h average indoor and outdoor pollutant concentration. Table 4 The summary of traffic-related information in each sample. Case No. 1 2 3 4 5 6 7 8 Total traffic volume (cars/24 h) Average traffic activity Space utilization ratio(%) 455 2.22 32 169 1.35 34 308 1.54 52 139 1.29 46 81 1.35 49 56 0.93 43 147 0.68 45 263 1.13 53 Among all the samples, the daily average concentration of CO2 was about 500 ppm, and that of Case 8 was the highest which was 806 ppm, followed by Case 7 of 772 ppm. Meanwhile, the daily average TVOC concentration measured in garage was about 24 ppb. The daily average TVOC concentration of Case 6 was the highest which was 46 ppb, while that of Case 5 was the lowest which was only 4 ppb. By comparing the measured data above, it was difficult to find obvious relationship between indoor TVOC concentration and traffic-related data. Nevertheless, Kim et al. had different discoveries in the study about VOCs of parking garages in US. Kim et al. assessed the level of air pollution in urban parking lots and the impact of vehicle activity on their levels. It was found that the traffic volume on the weekend was 12 times lower than that during the weekday, while the median concentration of air pollutants was also reduced by 2 (CO)-7 (pPAH) times. The median values of TVOC during weekday/weekend was 10.6/0.9 mg/m3, respectively, which could not be directly compared with the results of this study due to inconsistency in measurement methods (Kim et al., 2007). In addition, Batterman et al. reported on VOC concentrations and emissions at 15 residential garages in Michigan that varied in type, size, use and other characteristics. Results showed that the mean concentration of TVOC in one garage was 250.67 mg/m3, which was far higher than the concentration measured in this study (converted to 89.7 mg/m3). It was also found that VOC concentrations were not significantly correlated to garage characteristics or meteorological factors, but were largely dependent on VOC sources present and occupant activities (Batterman et al., 2006). Thus, the daily variation characteristics of TVOC concentration in measured UPGs was essential in the following sections. 3.2. Over standard rate There is no specific guideline or standard on the air quality of UPGs. Therefore, several international and Chinese standards were combined in this study to evaluate the air quality and calculate the over standard rate of air pollutants in garage. The daily average concentration of PM2.5 suggested by WHO Air Quality Guideline (AQG) was set at 25 mg/m3 (WHO, 2006; WHO, 2010), while China has set a ceiling limit for PM2.5 concentration of 75 mg/m3 for indoor environment in the residential area (China, M. E. P., 2012). For daily average concentration of PM10, WHO air quality guideline (WHO, 2006) has recommended the 50 mg/m3 as a set value, as well as EU set value (European Union, 2008). Compared with WHO, another threshold value of PM10 concentration was set as 150 mg/ m3 according to the national standards set by both US Environmental Protection Agency (EPA) (U. S., EPA, 1990) and Chinese government (China MEP, 2002). The indoor CO2 concentration at no more than 700 ppm above the ambient condition (ranges from 300 to 500 ppm) would satisfy the majority of occupants suggested by ASHERAE (2010). Therefore, it is common that 1000 ppm was set as the threshold value for indoor CO2 concentration in buildings (China MEP, 2002; ASHERAE, 2010). The total VOC concentration suggested by Chinese standards was set at 0.6 mg/m3 for 8 h (China MEP, 2002; China MOHURD, 2010). In our study, we measured the total VOCs concentration (in ppb) in each UPG. To compare with the standard of TVOCs, the average percentages of the main components of VOCs in parking facilities should be determined firstly, and then convert the data from ppb units into mg/m3 units. According to the previous studies, eight most abundant VOCs emitted in vehicle exhaust have been identified in many parking facilities (Batterman et al., 2006; Batterman et al., 2007; Glorennec et al., 2008; Kristanto, 2006), and benzene, toluene, ethylbenzene and m,p-xylenes (BTEX) and styrene are as aromatic hydrocarbons which categorized as hazardous air pollutants (HAP) by U.S. EPA (U.S.EPA, 1999). The average percentages of target components of VOCs in garages presented in previous studies were shown in Table 5. It can be concluded that the average rates of five main components among different types of parking garages are relatively Z. Liu et al. / Environmental Pollution 247 (2019) 626e637 631 Table 5 Average percentage of target components of VOCs in garages according to previous studies. Type N Residential attached garage Commercial underground parking facility Residential attached garages Residential garages(including attached, detached and carport) (in Los Angeles, CA) (in Elizabeth, NJ) (in Houston, TX) Residential underground car park (level 1, working days) (level 2, working days) (Level 1, weekends) (Level 2, weekends) Residential attached garages (working days) Residential attached garages 10 8 15 38 21 55 20 20 10 10 7 15 Reference Kristanto (2006) Kristanto (2006) Batterman et al. (2006) Hun et al., 2011 Hun et al., 2011 Hun et al., 2011 Marc et al., 2016 Marc et al., 2016 Marc et al., 2016 Marc et al., 2016 Marc et al., 2016 Batterman et al., 2007 Percentages(%) B T E X S 8 16 11 11 8 18 17 17 20 21 15 11 58 40 58 53 54 51 41 43 37 39 49 58 9 9 6 7 8 6 7 7 8 7 8 6 24 20 25 21 22 19 23 22 23 22 20 25 1 15 0 8 7 7 12 12 12 10 9 0 Abbreviation: N: The number of samples; B-Benzene; E-Ethylbenzene; X-m,p-xylene; S-Styrene. Table 6 Molecular weight, conversion factor and estimated percentages of target components of VOCs in parking facilities. Pollutant Molecular weight (Kristanto, 2006) Conversion factor ppb to mg/m3(Kristanto, 2006) Estimated percentage (%) Benzene Toluene Ethylbenzene Xylene Styrene 78.1 92.1 106.2 106.2 104.2 3.194  103 3.509  103 4.349  103 4.349  103 4.225  103 13 52 8 23 4 Table 7 Comparison of target pollutants concentrations and over standard rates in each sample. Case No. Indoor Mean SD Min PM2.5 (mg/m ) Measuring time ¼ 24 h (0:00 a.m.-24:00 p.m.) 1 187.90 34.43 131.99 2 199.60 39.14 152.11 3 57.71 16.18 28.42 4 49.66 8.85 35.81 5 46.97 10.41 20.31 6 113.73 29.14 70.04 7 88.92 44.34 35.05 8 101.97 48.58 41.23 3 PM10 (mg/m ) Measuring time ¼ 24 h (0:00 a.m.-24:00 p.m.)a 1 535.18 197.92 292.50 2 447.56 138.65 250.10 3 913.02 782.65 50.77 4 238.44 164.72 77.23 5 246.39 213.41 91.17 6 440.29 257.92 150.06 7 303.72 167.79 127.52 8 558.75 187.86 250.05 TVOC (ppb) Measuring time ¼ 8 h (6:00 a.m.-2:00 p.m.) 1 20 33.02 0 2 21 26.32 0 3 94 102.13 0 4 16 24.45 0 5 9 19.93 0 6 63 78.33 0 7 50 73.80 0 8 74 88.90 0 CO2 (ppm) Measuring time ¼ 24 h (0:00 a.m.-24:00 p.m.) 1 535 45.77 464 2 529 31.12 495 3 500 43.80 439 4 534 56.86 440 5 421 23.18 390 6 472 23.01 420 7 806 79.00 563 8 772 83.22 650 3 Reference values Over standard rate (%) Max (exposure time) (reference values) 245.39 288.56 144.54 78.82 122.78 173.34 187.00 220.98 AQG ¼ 25 mg/m3 (24 h) GB ¼ 75 mg/m3 (24 h) 652(AQG) 698(AQG) 131(AQG) 99(AQG) 88(AQG) 355(AQG) 256(AQG) 308(AQG) 151(GB) 166(GB) N N N 52(GB) 19(GB) 36(GB) 2358.00 1346.30 4834.29 1117.20 2697.47 1894.80 849.38 1264.60 AQG ¼ 50 mg/m3 (24 h) EPA ¼ 150 mg/m3 (24 h) GB ¼ 150 mg/m3 (24 h) 970(AQG) 795(AQG) 1726(AQG) 377(AQG) 393(AQG) 781(AQG) 507(AQG) 1018(AQG) 257(GB) 198(GB) 509(GB) 59(GB) 64(GB) 194(GB) 102(GB) 273(GB) 209 125 425 132 145 281 437 424 GB ¼ 0.60 mg/m3 (8 h) (approximately 160 ppb) N N N N N N N N 939 713 762 810 582 555 1142 1126 ASHRAE ¼ 1000 ppm(24 h) GB ¼ 0.10% (24 h) N N N N N N N N a Abbreviation: AQG: WHO Air Quality Guideline; GB: Chinese national standard; EPA: U.S. Environmental Protection Agency; ASHRAE: ASHRAE Standard 62.1, Ventilation for Acceptable Indoor Air Quality; N: The air pollutant concentration does not exceed the reference value of the standards or guidelines. 632 Z. Liu et al. / Environmental Pollution 247 (2019) 626e637 similar. Toluene accounts for the highest proportion (37e58%), followed by m,p-xylene(19e25%), benzene(8e21%), ethylbenzene(6e9%), and styrene(0e15%), and the molecular weight, conversion factor and estimated percentages of target components of VOCs were shown in Table 6. Therefore, the conversion factor of TVOCs in parking facilities was 3.757  103 (1 ppb ¼ 3.757  103 mg/m3), and the threshold of 8-h average TVOCs concentration specified in Chinese standards (GB) was set at approximately 160 ppb. The target pollutants concentrations and over standard rates in all the samples were listed in Table 7. Comparison of indoor and outdoor concentrations of PM2.5, PM10 and TVOC in each parking garage was further shown in Fig. 2, Fig. 3 and Fig. 4, respectively. The PM2.5 concentrations of all the samples exceeded the limited threshold value suggested by WHO AQG and the average over standard rate was 323%. WHO Working Group on “health aspects of air pollution”(WHO, 2004) suggested that the relative risk percentage of PM2.5 mortality would increase 0.6% per 10 mg/m3 under the condition of continuous exposure for 1 days at the current exposure level of air pollution in North America. Pope et al. also proposed that the relative risk percentage would increase 6% per 10 mg/m3 when the duration of continuous exposure increased to 16 years (Pope et al., 2002). In addition, the indoor PM2.5 concentration in five garages exceeded the standard value of the Chinese standard, the over standard rate of Case 2 was the highest (166%), followed by that of Case 1(151%), Case 6(52%), Case 8(36%) and Case 7(19%). The 24 h average PM10 concentrations of all the samples were considerably higher than the air quality set value suggested by US EPA and Chinese national standard, not to mention the WHO AQG. The average over standard rate of PM10 measured in all the measured UPGs was 821% based on WHO AQG, while that was 207% based on Chinese national standard. The CO2 concentration of UPGs was generally lower than the threshold specified by standards or guidelines. For TVOC of all the eight cases, the compliance rate looks better than the PM mentioned above. According to the German Federal Environmental Agency (UBA) (Umweltbundesamt, 2007), the TVOC level ranges from 0.3 to 1.0 mg/m3 when the air quality rating is good, although ventilation may still be required to maintain a good indoor environment. In Fig. 4, the average concentration of 8 h in each sample was within the acceptable range, but the maximum concentrations of all the samples were generally above the threshold value specified in GB, with the range from 125 to 437 ppb. Fig. 3. Comparison of indoor and outdoor PM10 concentrations in all the samples. Fig. 4. Comparison of indoor TVOC concentration in all the samples. 3.3. Daily variation characteristics Fig. 2. Comparison of indoor and outdoor PM2.5 concentrations in all the samples. The daily variations of target pollutant concentrations and traffic volume in all the samples were shown in Fig. 5. As shown in Fig. 5, for weekdays, the number of cars entering or leaving the garage fluctuates up and down with two distinct peaks corresponding morning and evening rush hours in all the residential UPGs. The morning rush hour in the garage was from 7:00 a.m. to 9:00 a.m. while the evening rush hour was from 5:00 p.m. to 10:00 p.m. Among all the samples, the traffic volume always reached its peak at about 7:30 a.m. while there is no obvious peak of traffic volume during the evening rush hour due to the time of cars entering or leaving the garage was relatively dispersed. Additionally, there are a small number of vehicles entering or leaving the garage at noon (11:30 a.m.-1:30 p.m.). The variations of the indoor PM10 concentration mainly depend on that of traffic volume, but the variations could be also affected by the UPGs’ structure and the existence of the air vents. For daily variation trend, the indoor PM10 concentration of measured garage was equal to or slightly higher than the outdoor when no car was entering or leaving the garage, while with the growing of traffic volume, the pollutant concentration had a considerable increase, Z. Liu et al. / Environmental Pollution 247 (2019) 626e637 633 Fig. 5. The daily variations of target pollutant concentrations and traffic volume in all the samples. and after the rush hour, the pollutant concentration gradually decreased to a relatively steady value which was close to the ambient. Therefore, it can be concluded that the variation trend of the indoor PM10 concentration depended on that of traffic volume to a large extent. However, the growth range of indoor PM10 concentration was not completely related to the traffic volume. For example, although the maximum of the traffic volume in Case 1 and Case 3 were similar, which was 23 cars per 15min and 22 cars per 15min, respectively, the maximum indoor PM10 concentration of Case 3 (4834.29 mg/m3) was 2 times higher than that of Case 1 (2358.00 mg/m3). This difference may be due to the ventilation effectiveness of the air vent and the different volume of the UPGs. There was an air vent in Case 1, while no one was in Case 3. Moreover, the volume of the garage in Case 1 was 2 times higher than that in Case 3. Furthermore, for the Case 5 and Case 6 with the same volume of garages, although the total traffic volume of Case 5 (81 cars per day) was higher than that of Case 6 (56 cars per day), the daily average PM10 concentration of Case 5 was only a half of that of Case 6. The possible explanation was the difference in the design of air vents and entrance/exit in UPGs. For Case 5, the 634 Z. Liu et al. / Environmental Pollution 247 (2019) 626e637 Fig. 5. (continued). entrance and exit of the garage are on opposite sides of the garage, and the wind in the underground car park blew predominantly from the entrance toward the exit. In addition, there are two air vents in the garage. While for Case 6, there is only one gate for cars entering and leaving the garage and no other air vent is designed in this garage. Thus, the effectiveness of natural ventilation of Case 5 was relatively better. The same with variation characteristics of PM10, the indoor TVOC concentration was completely affected by the traffic volume. The only one difference was that the indoor TVOC concentration was 0 when there was no car entering or leaving the garage, which determined that the only source of TVOC in the garage was vehicle exhaust. Furthermore, the fluctuation ranges of the TVOC concentrations in garages were also influenced by the traffic volume, the existence of the air vent and the structure of the garage which was similar to the PM10. Nevertheless, unlike PM10 and TVOC, the variation trend of the indoor PM2.5 concentration was largely consistent with the ambient PM2.5 concentration, while there were some slightly fluctuations of the indoor PM2.5 concentration during the rush hours according to the data of Case 3 and Case 5. In addition, the indoor CO2 concentration could be stable in a certain range, but that would be influenced by the traffic volume to a small extent. 3.4. Correlation analysis The 15-min and 60-min average data of indoor pollutant Z. Liu et al. / Environmental Pollution 247 (2019) 626e637 Table 8 Result of Pearson's correlation analysis between indoor target pollutant concentrations and traffic volume, as well as ambient environment. Case No. Indoor Case 1 PM2.5 PM10 TVOC Case 2 PM2.5 PM10 TVOC Case 3 PM2.5 PM10 TVOC Case 4 PM2.5 PM10 TVOC Case 5 PM2.5 PM10 TVOC Case 6 PM2.5 PM10 TVOC Case 7 PM2.5 PM10 TVOC Case 8 PM2.5 PM10 TVOC ra pb R2 r p R2 r p R2 r p R2 R p R2 R p R2 R p R2 R p R2 R p R2 R p R2 R p R2 r p R2 R p R2 R p R2 R p R2 R p R2 R p R2 R p R2 r p R2 r p R2 r p R2 R p R2 R p R2 R Traffic volume Nc ¼ 97 Outdoor N ¼ 25 0.423** 0.000 0.179 0.642** 0.000 0.412 0.801** 0.000 0.642 0.337** 0.001 0.114 0.333** 0.001 0.111 0.513** 0.000 0.263 0.483** 0.000 0.233 0.671** 0.000 0.450 0.695** 0.000 0.483 0.213* 0.036 0.045 0.735** 0.000 0.540 0.774** 0.000 0.599 0.074 0.470 0.005 0.463** 0.000 0.214 0.263** 0.009 0.069 0.038 0.715 0.001 0.418** 0.000 0.175 0.454** 0.000 0.206 0.127 0.215 0.016 0.780** 0.000 0.608 0.839** 0.000 0.704 0.199 0.050 0.040 0.495** 0.000 0.245 0.719** 0.945** 0.000 0.893 0.041 0.847 0.017 e ** 0.789 0.000 0.623 0.005 0.981 0.002 e 0.132 0.530 0.017 0.344 0.092 0.118 e 0.530** 0.006 0.281 0.339 0.097 0.115 e 0.956** 0.000 0.914 0.639** 0.001 0.408 e 0.178 0.394 0.032 0.240 0.249 0.058 e 0.874** 0.000 0.764 0.221 0.289 0.049 e 0.835** 0.000 0.697 0.405* 0.044 0.164 e 635 Table 8 (continued ) Case No. Indoor Traffic volume Nc ¼ 97 p R2 Outdoor N ¼ 25 0.000 0.517 ** Correlation is significant at the 0.01 level (2-tailed). *Correlation is significant at the 0.05 level (2-tailed). a Pearson's correlation coefficient. b Probability (2 tails significance). c Number of data. concentration were calculated by equation (3) to analyze the correlation with the influence factors. The impacts of traffic volume and ambient environment on indoor air pollutants were determined by employing Pearson correlation by SPSS 16.0, and the result was listed in Table 8. Apart from the data of Case 3 and Case 6, the indoor PM2.5 concentration generally exhibited a strong correlation with the outdoor PM2.5 concentration (p  0.01) with the average R2 of 0.695, however, the indoor PM2.5 concentration of Case 3 showed a moderate correlation with the traffic volume (p  0.01) with the R2 of 0.233, and for Case 6, the indoor PM2.5 concentration showed inverse associations with both traffic volume and the outdoor pollutant concentration. Five out of eight samples indicated that the indoor TVOC concentrations were strongly correlated with traffic volume (R2: 0.483e0.704, p  0.01), while moderate correlations were found in Case 2 and Case 6, and the weak correlation was found in Case 5(R2 ¼ 0.069, p  0.01). For indoor PM10 concentration, all of the samples showed the existence of moderate relationships between indoor PM10 concentration and traffic volume (R2: 0.111e0.608, p  0.01), while two of these samples also presented the moderate correlations with the ambient environment (R2 ¼ 0.408, p  0.01 for Case 5 and R2 ¼ 0.164, p  0.05 for Case 8). The weakly relationship between PM10 concentration and traffic flow was also found in the results of Li et al. (Li and Xiang, 2013). One possible reason was that the indoor PM10 concentration might not only affect by traffic volume, but also associated with the ambient and accumulative effect. Therefore, a multiple regression was performed to correlate hourly indoor PM10 concentration with the traffic volume, ambient level and accumulative effect, which calculated by equation (4). The multiple regression analysis of hourly indoor PM10 concentration in each sample was listed in Table 9. For the Case 1, the resultant model is  C in  t   ¼ 0:390 C in t1   þ 5:363Tt þ 0:278 C out þ 150:997 t (5) where R is 0.872 and the adjusted R2 is 0.726. Therefore, compared Table 9 Multiple regression analysis of hourly indoor PM10 concentration in each sample. Case No. Case Case Case Case Case Case Case Case 1 2 3 4 5 6 7 8 Unstandardized coefficients a b g ε 0.390 0.423 0.567 0.535 0.401 0.764 0.537 0.766 5.363 4.500 26.324 11.888 17.700 34.251 10.619 4.116 0.278 0.010 0.274 1.338 2.590 0.006 0.441 0.006 150.997 247.425 81.968 42.539 102.602 32.357 156.575 91.986 R Adjusted R2 0.872 0.572 0.933 0.978 0.895 0.944 0.973 0.953 0.726 0.230 0.853 0.950 0.772 0.877 0.940 0.895 636 Z. Liu et al. / Environmental Pollution 247 (2019) 626e637 with the consideration of a single factor, the R value for this multiple regression model could be improved by 35.8%, and the value of R2 could be increased by 76.2%. Overall, the average R2 value of this synthesized model was 0.780, which was 2.7 times that of the previous model. By considering other factors affecting the indoor PM10 concentration, this paper established a synthesized model of relationship between indoor PM10 concentration, traffic volume, the ambient and accumulative effect with the significant improvement of R2. 4. Conclusions Field measurements in eight residential UPGs under natural ventilation were carried out in this study to identify the variation patterns of indoor pollutant in UPGs as well as to examine the determinants of the indoor air pollution. Based on the above analysis, several following conclusions can be drawn: (1) The daily average indoor concentrations of PM2.5, PM10, CO2 and TVOC in naturally ventilated UPGs were 105.81 mg/m3, 464.17 mg/m3, 571 ppm and 24 ppb, respectively. For particle matters, the indoor PM2.5 concentrations were closely related to that outdoors with the average I/O value of 1.6, while the concentrations of indoor PM10 was much higher than that outdoors with the average I/O value of 4.0. (2) The fraction of naturally ventilated UPGs of which the PM2.5 concentration exceeded regulated air quality threshold was 100% (WHO) and 62.5 (China National Standard), and the average over-standard rate is 323% (WHO) and 166%(China National Standard), respectively. Meantime, the daily average PM10 concentration of all UPGs is significantly higher than the air quality guideline value. According to WHO AQG and Chinese national standards, the average daily over-standard rate was 821% and 207%, respectively. The daily average CO2 concentration of all UPGs was not over-standard and the 8 h average TVOC concentration was lower than the threshold specified by Chinese national standards and was within the UBA good air quality range, but its maximum concentration ranging from 125 to 437 ppb, were generally above the threshold value specified by China National Standard. (3) The PM10 and TVOC concentrations of UPGs were moderately correlated with traffic volume (p  0.01), and the variations of them were also affected by the existence of air vent and UPGs' characteristics. In comparison, the concentrations of indoor PM2.5 exhibited a strong correlation with that of ambient PM2.5 (p  0.01). The indoor CO2 concentration could be stable in a certain range, though the CO2 concentration could be influenced by the traffic volume to a small extent especially in the rush hours. Multiple linear regression model of hourly PM10 concentration in garage including traffic volume, the ambient and accumulative effect was established with the average R2 value of 0.780, which was improved by 166%. Acknowledgment Thanks to all the testing members who have worked tirelessly throughout the period of measurements to bring accurate and comprehensive data, and their names are Di Wu, Yuanwei Liu, Haotian Niu, Qingxu Zhou, Haiyang Liu and Wenbin Zhuang from North China Electric Power University. This work was funded by the National Natural Science Foundation of China (No. 51708211), the National Science and Technology Ministry of China (2016YFC1201302), the Opening Funds of State Key Laboratory of Building Safety and Built Environment National Engineering Research Center of Building Technology (BSBE2017-08), Natural Science Foundation of Hebei (No. E2017502051) and Fundamental Research Funds for the Central Universities (No.2018MS103). References ASHRAE, 2007. HVAC applications. In: ASHRAE Handbook. ASHRAE, Atlanta, GA. ASHERAE, 2010. Standard 62.1, Ventilation for Acceptable Indoor Air Quality. American Society of Heating, Refrigerating and Air-Conditioning Engineers, Atlanta, GA. Batterman, S., Hatzvasilis, G., Jia, C.R., 2006. Concentrations and emissions of gasoline and other vapors from residential vehicle garages. Atmos. Environ. 40, 1828e1844. Batterman, S., Jia, C.R., Hatzivasilis, G., 2007. Migration of volatile organic compounds from attached garages to residences: a major exposure source. Environ. Res. 104, 224e240. Bowatte, G., Lodge, C.J., Knibbs, L.D., Erbas, B., Perret, J.L., Jalaludin, B., Morgan, G.G., Bui, D.S., Giles, G.G., Hamilton, G.S., Wood-Baker, R., Thomas, P., Thompson, B.R., Matheson, M.C., Abramson, M.J., Walters, E.H., Dharmage, S.C., 2018. Traffic related air pollution and development and persistence of asthma and low lung function. Environ. Int. 113, 170e176. Chan, M.Y., Chow, W.K., 2004. Car park ventilation system: performance evaluation. Build. Environ. 39, 635e643. China, MEP, 2002. GB/T188832002, Indoor Air Quality Standard. Administration of Quality Supervision, Inspection and Quarantine, Beijing. China, MOHURD, 2010. GB50325-2010, Code for Indoor Environmental Pollution Control of Civil Building Engineering. Ministry of Housing and Urban-Rural Development of the Peoples Rapublic of China, Beijing. Costa, L.G., Cole, T.B., Coburn, J., Chang, Y.C., Dao, K., Roque, P.J., 2017. Neurotoxicity of traffic-related air pollution. Neurotoxicology 59, 133e139. Demir, A., 2015. Investigation of Air Quality in the Underground and Aboveground Multi-Storey Car Parks in Terms of Exhaust Emissions. World Conference on Technology, Innovation And Entrepreneurship, pp. 2601e2611. European Union, 2008. Directive 2008/50/EC of the european parliament and of the council of 21 may 2008 on ambient air quality and cleaner air for europe. Off. J. Euro. Union 152, 16e17. Glorennec, P., Bonvallot, N., Mandin, C., Goupil, G., Pernelet-Joly, V., Millet, M., Filleul, L., Le Moullec, Y., Alary, R., 2008. Is a quantitative risk assessment of air quality in underground parking garages possible? Indoor Air 18, 283e292. He, B.J., Zhao, D.X., Zhu, J., Darko, A., Gou, Z.H., 2018. Promoting and implementing urban sustainability in China: an integration of sustainable initiatives at different urban scales. Habitat Int. 82, 83e93. Ho, J.C., Xue, H., Tay, K.L., 2004. A field study on determination of carbon monoxide level and thermal environment in an underground car park. Build. Environ. 39, 67e75. Hun, D.E., Corsi, R.L., Morandi, M.T., Siegel, J.A., 2011. Automobile proximity and indoor residential concentrations of BTEX and MTBE. Build. Environ. 46, 45e53. Jokl, M., 2000. Evaluation of indoor air quality using the decibel concept based on carbon dioxide and TVOC. Build. Environ. 35, 677e697. Kim, S.R., Dominici, F., Buckley, T.J., 2007. Concentrations of vehicle-related air pollutants in an urban parking garage. Environ. Res. 105, 291e299. Kristanto, G.A., 2006. Assessment of Volatile Organic Compounds (VOCs) in Indooor Parking Facilities at Houston. Texas Southern University, Texas. Li, H.R., Zeng, X.Y., Zhang, D.L., Chen, P., Yu, Z.Q., Sheng, G.Y., Fu, J.M., Peng, P.A., 2014. Occurrence and carcinogenic potential of airborne polycyclic aromatic hydrocarbons in some large-scale enclosed/semi-enclosed vehicle parking areas. J. Hazard Mater. 274, 279e286. Li, Y.X., Xiang, R.B., 2013. Particulate pollution in an underground car park in Wuhan, China. Particuology 11, 94e98. Liu, Z.J., Cheng, K., Li, H., Cao, G., Wu, D., Shi, Y., 2018a. Exploring the potential relationship between indoor air quality and the concentration of airborne culturable fungi: a combined experimental and neural network modeling study. Environ. Sci. Pollut. Control Ser. 25, 3510e3517. Liu, Z.J., Ma, S.Y., Cao, G.Q., Meng, C., He, B.J., 2018b. Distribution characteristics, growth, reproduction and transmission modes and control strategies for microbial contamination in HVAC systems: a literature review. Energy Build. 177, 77e95. Marc, M., Smielowska, M., Zabiegala, B., 2016. Concentrations of monoaromatic hydrocarbons in the air of the underground car park and individual garages attached to residential buildings. Sci. Total Environ. 573, 767e777. Papakonstantinou, K., Chaloulakou, A., Duci, A., Vlachakis, N., Markatos, N., 2003. Air quality in an underground garage: computational and experimental investigation of ventilation effectiveness. Energy Build. 35, 933e940. Pope III, C.A., Burnett, R.T., Thun, M.J., Calle, E.E., Krewski, D., Ito, K., Thurston, G.D., 2002. Lung cancer, cardiopulmonary mortality, and long-term exposure to fine particulate air pollution. Jama 287, 1132e1141. Sawyer, R.F., Harley, R.A., Cadle, S., Norbeck, J., Slott, R., Bravo, H., 2000. Mobile sources critical review: 1998 NARSTO assessment. Atmos. Environ. 34, 2161e2181. Song, X.C., Zhao, Y., 2017. CFD simulation of particle diffusion at the region near the entrance or exit of an underground parking lot. 2017 2nd. International Conference on Advances on Clean Energy Research (Icacer 2017) 118, 184e188. Z. Liu et al. / Environmental Pollution 247 (2019) 626e637 Tham, K.W., Parshetti, G.K., Balasubramanian, R., Sekhar, C., Cheong, D.K.W., 2018. Mitigating particulate matter exposure in naturally ventilated buildings during haze episodes. Build. Environ. 128, 96e106. Umweltbundesamt(UBA), 2007. Beurteilung von Innenraumluftkontaminationen mittels Referenz- und Richtwerten. Bundesgesundheitsblatt - Gesundheitsforschung -Gesundheitsschutz. U.S., EPA, 1990. NAAQS Table. U.S. Environmental Protection Agency. Available at: https://www.epa.gov/criteria-air-pollutants/naaqs-table. U.S., EPA, 1999. Integrated Risk Information System (IRIS). U.S. Environmental Protection Agency. Vukovic, G., Urosevic, M.A., Razumenic, I., Kuzmanoski, M., Pergal, M., Skrivanj, S., Popovic, A., 2014. Air quality in urban parking garages (PM10, major and trace elements, PAHs): instrumental measurements vs. active moss biomonitoring. Atmos. Environ. 85, 31e40. WHO, 2004. Health Aspects of Air Pollution: Results from the WHO Project" Systematic Review of Health Aspects of Air Pollution in Europe". World Health 637 Organization, Geneva. WHO, 2006. WHO Air Quality Guidelines Global Update 2005 for Articulate Matter, Ozone, Nitrogen Dioxide and Sulfur Dioxide. World Health Organization, Geneva. WHO, 2010. WHO Guidelines for Indoor Air Quality: Selected Pollutants. World Health Organization, Geneva. Zhang, Z.H., Khlystov, A., Norford, L.K., Tan, Z.K., Balasubramanian, R., 2017. Characterization of traffic-related ambient fine particulate matter (PM2.5) in an Asian city: Environmental and health implications. Atmos. Environ. 161, 132e143. Zhao, Y., Song, X.C., Wang, Y., Zhao, J.N., Zhu, K., 2017. Seasonal patterns of PM10, PM2.5, and PM1.0 concentrations in a naturally ventilated residential underground garage. Build. Environ. 124, 294e314. Zhao, Y., Zhao, J.N., 2016. Numerical assessment of particle dispersion and exposure risk in an underground parking lot. Energy Build. 133, 96e103. Thomas 1 Kaitlyn Thomas Honors Engineering For Social Good Research Proposal A Detection and Avoidance System to Address Whale Strikes Abstract Boats passing through any area of open water are a risk to whales. Because whales are mammals, they breathe air and in order to do that, they have to surface to the water and breathe through their blowholes. Boats travel along the surface of the water and therefore are at risk of hitting whales. This is a risk to both the boat and the whale. For larger boats, it is more difficult to maneuver around a whale and so their chances of hitting a whale in their path of transit are great. Whales being a sparsely populated marine mammal group need to be preserved for the good of ecosystems around the globe. Boats need to be able to spot them and maneuver around them. By using a series of hydrophones and an AI detection system, my project hopes to solve this problem by relaying information on the whales distance and direction from the boat to allow the captain to maneuver around the whale and not hit it. Intro As boat traffic becomes more common, whale’s are at risk of being hit. Some species of whales have become endangered due to human activity and ship striking is one of those causes. The operation of boats is complicated on its own and spotting whales and other sea life muddles the already difficult problems captains face on voyages. Low cost whale detection systems exist in the research field and avoidance procedures exist in the boating world, but these two fields do Thomas 2 not intersect. The interception of these fields will create a simpler, more effective, more automatic whale detection and avoidance system on large boats so that they do not hit whales, which will prevent both the damaging of boats and endangering people on the boats and the mortality of whales due to ship strikes. Background (/Lit Review) Many species of whales are endangered and more are dying every year. Because of the increase in whale beachings, which is when dead whales wash up on shores, research institutions began conducting research to pinpoint the reason for their mortality, finding that ship-striking is a large reason for whale mortality. (Kowalewski et. al., 2010) Large boats have a difficult time maneuvering through the ocean at high speeds. As whales approach them, there is little to nothing boat operators can do by the time the whale is spotted to the time the whale strikes the boat. Current research is focused on detecting and plotting patterns in whale’s routines as well as their location throughout the day in an attempt to schedule boat routes around densely populated waters as well as to better understand the nature of whales as a whole. Currently there are two major whale detection methods being used in the research field. The first is visual. John Durban, a Marine Mammal Biologist from the Southwest Fisheries Science Center, “[flies] a small hexacopter to hover 100 to 200 feet above whales to take overhead photographs: from these [he] can do photogrammetry (taking measurements from photographs) to monitor growth, assess body condition, and identify pregnancies. These metrics allow [augmentation of] traditional stock assessments to not only focus on abundance trends of whales, but to also understand the underlying causes of population dynamics” (Noaa. (n.d.)). Other researchers are Thomas 3 using satellite data to plot population density of whales. The second method is auditory. Jessica Crance, a Research Biologist at the Alaska Fisheries Science Center, created a small sailboat drone that is able to go into the ocean for long periods of time and record sound. She describes her project by saying, “Our interest in the project was to determine if the Saildrone would be a suitable platform for passive acoustic monitoring of marine mammals, in particular the critically endangered North Pacific right whale. We attached a small, autonomous acoustic recorder called the Acousonde to the keel of each Saildrone, and set it to record continuously up to 4 kHz, which covers the frequency band of most marine mammals in the Bering Sea” (Noaa. (n.d.)). Google is also conducting similar research. They use stationary hydrophones to pick up on sounds passing them and, using an AI system, are able to detect whether or not the sound is a whale. (“Acoustic Detection of Humpback Whales Using a Convolutional Neural Network.” (2018)). Using this research, they have identified patterns in whale calls and in population behavior and density throughout the waters of the pacific. Where this research does not exist is to aid whale detection and avoidance on boats. The research and efforts that do exist on boats have numerous problem areas. A large problem area is the detection phase. The primary reliance for spotting whales is with human vision, usually involving binoculars. In order for this technique to be successful, a person needs to be constantly on the lookout for whales and looking in the right spot to detect the whale with enough time to maneuver out of the way. A group of researchers and engineers designed a process for maneuvering and avoidance in large boats based upon existing technology on boats. It involves a tree of events and steps to take depending upon the circumstance (Gende, S. M. et al. (2019). However the detection technology used in research and the processes for boat maneuvering are not acting together. Thomas 4 Project Whales are dying due to ship striking, which not only is harmful to whales and the whale population, but also to ocean ecosystems as a whole; whales need to survive. In order to achieve this peaceful existence, boats need to employ an automatic detection and manual avoidance system of whales so as to avoid ship strikes and decrease whale mortality therefore increasing whale population size and preserving ocean ecosystems. Methods and Materials By using a similar method to the Saildrones, I will mount hydrophones onto a boat. The microphones will pick up the sound of the whale and an AI system, similar to Google’s system will be able to separate out the non-whale sounds from the whale sounds. After this process, the interpretation of the sound of the whale will be able to tell the distance of the whale by relating the decibels heard from the hydrophones in the array to the distance between the hydrophones. I have to develop math based on the speed of sound underwater, but I can do this through real world application using speakers and microphones. Because the hydrophones will be in a triangular shape, this will also be able to tell general direction using similar math. The hydrophones will pick up sound only near the surface of the water due to the presence of potential convergence zones or sound channels blocking the sound. However, this should not be a major problem due to the position of the boat being on the surface. A whale cannot hit the boat from far below the boat, meaning that the sounds, even with the presence of oddities in the water channel will still be accurately pinpointing the whale when it is in proximity to the boat. Thomas 5 The hydrophones will be constructed using contact microphones and an aluminum can. Based on a design from a blogpost. (Soundfly Partners. (2017)) After this information is relayed to the captain, he or she will be able to employ a maneuver based on the process that was outlined in the article by Scott M. Gende, Lawrence Vose, Jeff Baken, Christine M. Gabriele, Rich Preston, and A. Noble Hendrix. The AI system will be coded with fast.ai and their jupyter notebook through salamander.ai. The data sets for learning and validating will be obtained through Moby Sound: A reference archive for studying automatic recognition of marine mammal sounds. I will use the sound files of one specific endangered whale in the Pacific Ocean: the humpback whale. Depending on the time it takes to train the AI to sense this species and the similarities in sound patterns across various species of whales, I may also take the time to train the AI for other species. However I will start with the humpback. This system will be validated by playing sounds from Moby Sound through an underwater speaker or a speaker in a plastic bag into the pool to hydrophones on the other side and manually feeding the sound into the Ai system. If the system sorts the sounds accurately with the same accuracy as the trained model, the system is successful. Aluminum Can Contact Microphone GPU Processor (Salamander.ai) Arduino Timeline Thomas 6 Timeline Thomas 7 Works Cited Acoustic Detection of Humpback Whales Using a Convolutional Neural Network. (2018, October 29). Retrieved from https://ai.googleblog.com/2018/10/acoustic-detection-ofhumpback-whales.html. Berman Kowalewski, M., Gulland, F. M. D., Wilkin, S., Calambokidis, J. Mate, B., Cordaro, J., … Dover, S. (2010). Association Between Blue Whale (Balaenoptera musculus) Mortality and Ship Strikes Along the California Coast. Aquatic Mammals, 36(1), 59–66. doi: 10.1578/am.36.1.2010.59 Gende, S. M., Vose, L., Baken, J., Gabriele, C. M., Preston, R., & Hendrix, A. N. (2019). Active Whale Avoidance by Large Ships: Components and Constraints of a Complementary Approach to Reducing Ship Strike Risk. Frontiers in Marine Science, 6. doi: 10.3389/fmars.2019.00592 How does sound in air differ from sound in water? (2018, December 18). Retrieved from https://dosits.org/science/sounds-in-the-sea/how-does-sound-in-air-differ-from-sound-inwater/. Mellinger, D. K., & Clark, C. W. (2006). MobySound: A reference archive for studying automatic recognition of marine mammal sounds. Applied Acoustics, 67(11-12), 1226– 1242. doi: 10.1016/j.apacoust.2006.06.002 Noaa. (n.d.). Tracking Technology: The Science of Finding Whales. Retrieved from https://www.fisheries.noaa.gov/feature-story/tracking-technology-science-findingwhales. Thomas 8 Siderius, M. (n.d.). High Frequency Acoustic Channel Characterization for Propagation and Ambient Noise. Retrieved from https://www.onr.navy.mil/en/-/media/Files/AnnualReports/Ocean-Acoustics/FY07/oasider. Siderius, M., & Porter, M. (n.d.). Effects of Sound on the Marine Environment. Retrieved from https://www.onr.navy.mil/en/-/media/Files/Annual-Reports/MMB/FY08/mbsideri . Soundfly Partners. (2017, October 23). How to Build a DIY Hydrophone. Retrieved from https://flypaper.soundfly.com/produce/how-to-build-a-diy-hydrophone/. Sue E. Moore, Randall R. Reeves, Brandon L. Southall, Timothy J. Ragen, Robert S. Suydam, & Christopher W. Clark. (2012). A New Framework for Assessing the Effects of Anthropogenic Sound on Marine Mammals in a Rapidly Changing Arctic. BioScience, 62(3), 289-295. doi:10.1525/bio.2012.62.3.10 US6424596B1 - Method and apparatus for reducing noise from near ocean surface sources. (n.d.). Retrieved from https://patents.google.com/patent/US6424596B1/en.
Purchase answer to see full attachment
User generated content is uploaded by users for the purposes of learning and should be used following Studypool's honor code & terms of service.

Explanation & Answer

Attached. Please let me know if you have any questions or need revisions.

Running head: UNDERGROUND CAR PARKING VENTILATION SYSTEMS AND AIR
QUALITY

How Does Underground Car Parking Ventilation Systems Impact the Air Quality of Living
Spaces Around them and the Health Safety of their Residents in China?

Student’s Name
Institution Affiliation
Instructor’s Name
Course Code and Title
Date

UNDERGROUND CAR PARKING VENTILATION SYSTEMS AND AIR QUALITY 1
Introduction
The world is quickly urbanizing, and a key problem is that there is inadequate living
and working spaces. To deal with the increasing number of automobiles, most cities in China
and other crowded cities worldwide tend to save more space by turning outdoor parking or
parking structures under the ground. This not only saves places for the government but also
saves money for the constructors that don't have to buy additional lands for parking anymore.
Because of the rapid growth of underground construction, it is essential for people to realize
the danger they put themselves into every day, and they can’t really avoid it. Even though
some people might live in a house where they can have their car park outside on the street
instead of a garage, in commercial centers and their workplaces, owners tend to like their
parking underground to use the land most effectively.
How often do you go to the underground parking lot or stay in any indoor spaces, like
an office without a window or subway stations? And how long do you usually stay there?
Most people remain inside the building longer than they think, and most of them don’t know
about the damage from indoor pollution. “The major pollutants emitted by motor vehicles
including CO, NOx, sulfur oxides, (SO), HC, lead (Pb) and suspended particulate matter
(SPM), have damaging effects on human health”(Bhandarkar). They can cause harmful
diseases like asthma, heart and lung issues, increase the risk of getting cancer, and premature
death. All this brings up the importance of ventilation systems in indoor spaces. By modeling
the Ventilation system, people will know what sufficient ventilation is and whether there is a
better or cheaper way to ventilate their underground parking spaces. This paper will provide a
research into the ventilation designs and options available for car users in China that will help
curb the negative health risks associated with underground parking lots.
Background

UNDERGROUND CAR PARKING VENTILATION SYSTEMS AND AIR QUALITY 2
Underground car parking ventilation systems have a direct impact on the quality of air around
them and thus the respiratory health outcomes of residents living nearby. One of the most
notable pollutants that poses a direct threat to human health is NOx, and a variety of sulphur
oxides. All these chemicals are key emissions by products from cars and trucks in China.
According to Walsh (n.d), the levels of the respiratory poison NOx in China have increased
steadily over the past five decades. According to this author, NOx levels in Beijing have risen
past the 59 µg/m3 mark in 1997, which is beyond the recommended threshold of 50 µg/m3 as
stipulated under the Chinese Class II air quality standard. Walsh (n.d.) note that this steady
rise in air pollution increased directly as the number of people owning cars in China started
rising between 1991 and 2000. This level of pollution at the national level is replicated in
underground parking spaces where millions of city dwellers retire their cars after a busy day
in the city.
Vehicle Exhaust Air Pollutants and Human Health
The key concern here is that this level of air pollution that’s beyond the recommended level
by the Chinese Class II standards poses real health threats to consumers without their
knowledge. Evidence from a study by Zhu, et al. (2012) reveal that cluster groups of people
with poorly ventilated garages suffer from lower respiratory diseases similar to those of
patients from areas with high levels of vehicle exhaust emissions. Examples of these
emissions by products include sulfur oxides and NOx. Therefore, it is empirically confirmed
that high concentrations of vehicle emission gases can lead to lower respiratory diseases and
the negative economic costs therein. However, little is known about what can be done to help
consumers know the level of their garage ventilation to know whether they pose a threat to
their health and that of people around them. Thus, there is a need to conduct this study that
will help identify tools and measurement devices that consumers can use to detect the level of
exhaust gas emission in their garages.

UNDERGROUND CAR PARKING VENTILATION SYSTEMS AND AIR QUALITY 3
Air Quality and Pollution Measurement Devices
There exists many product offerings from innovative companies on smart measurement
devices for exhaust emissions. Examples of these measurement tools that can be installed in
the communally owned garages include Accuscan RSD4600, AVL AMA i60 COMBI, AVL
AMA SLTM, AVL AMA i60 Remote / Stand Alone Unit among others (Smit & Kingston,
2019). These tools leverage modern computing technologies combined with machine learning
capabilities to measure the level of air pollution from exhaust gases, recommend ways to
lower the particles in excess in the air and indicate garages that need ventilation
improvement. In the case for garages, there are softwares such as VFD control ventilation
systems that ensures that HVAC systems in garages, specifically in the USA, run
automatically and continuously to supply oxygen and pump out carbon monoxide among
other emission gases from parking lots (Goldschmidt, 2020 March). A combination of
advanced HVAC ventilation systems fitted with VFD controls for autonomous operations can
help Chinese city residents eliminate the health challenges resulting from poor air quality
from underground parking spaces.
Parking structure types
In China, there are three popular types of parking. The first type is known as horizontal. In
the case of a rental apartment, a horizontal parking structure occupies the lowest floor of the
building and is typical in most apartments in China. In this type of parking, only one car can
occupy a single parking lot at a go. For this reason, the horizontal parking spaces are
ineffective in crowded cities of China such as Beinjing. The second type of parking structure
is vertical parking lots. In these types of parking spaces, vehicles occupy different layers of
the parking lot in a chronological order. This type of parking lot is designed to carry the
maximum number of vehicles at a go. A typical vertical parking lot consists of adjustable
aluminium ladders that suspend the cars in a similar manner to a car bazaar. This approach is

UNDERGROUND CAR PARKING VENTILATION SYSTEMS AND AIR QUALITY 4
suitable for use in highly populated cities in China. It saves on both costs and space.
Unfortunately, this type of parking comes with higher levels of emission because it holds
more cars at a go for the same amount of space. It is sensible to hypothesize that the level of
air pollution associated with vertical parking structures is higher than that of horizontal
parking facilities.
The problem with the current literature and governance systems do not have an understanding
of the level of air pollution that these different types of parking spaces cause. Currently, the
authority in China has a generalized structure for managing parking spaces depending on a
number of factors as shown below (Institute for Transportation and Development Policy,
2015).

This table hints on important parking areas of interest for this study that can help reveal the
level of air pollution matched against ventilation systems in residential, office, retail,
restaurant, hospital, cinema, exhibition and theatre spaces.

UNDERGROUND CAR PARKING VENTILATION SYSTEMS AND AIR QUALITY 5

Types and Models of Ventilation Systems
There are two broad types of ventilation systems (HVAC) systems mainly used in parking
spaces. These two systems are zoned and multi-stage/single-stage HVACs. The multistage
are the most reliable types of ventilation systems (Smit & Kingston, 2019). These are
automated to control air circulation in a room given other factors such as humidity, air
temperature and carbon monoxide concentration in a room (Krarti & Ayari, 2001). These
multi-stage systems are able to switch on and off depending on the set room requirements.
Garages fitted with these systems are both energy efficient and effective in air quality control
(HGTV, n.d.). Single-stage systems are ventilation systems that a user switches on and off
depending on their room needs. Unfortunately, these systems are ineffective during crucial
hours such as when people are sleeping. Zoned systems on the other hand are advanced in the
sense that they moderate both air inflow and heating conditions to ensure optimal air
conditioning for a room. Garages suited with zoned systems can be highly effective in air
pollution control.
Literature Review #1: Ventilating enclosed garages
In a study aimed at examining how ventilation differed across parking spaces, Krarti and
Ayari (2001) note that a garage needs to be properly ventilated such that the rate of carbon
monoxide saturation balances out with the volume of clean air that the ventilation systems
pump in to this enclosed space. This study uses a simple parametric study design to make its
conclusion. Gil-Lopez, Sanchez-Sanchez & Gimenez-Molina (2014) also make similar
findings that for a ventilation system to be sufficient, it must draw in enough volume of fresh
air to cancel out the NOx, CO and CO2 particles from vehicle exhaust gases. Gil-Lopez et al.
(2014) defines a method of calculating the ventilation flow of garage HVAC systems and
making necessary adjustments for energy requirements to run these systems. The authors find

UNDERGROUND CAR PARKING VENTILATION SYSTEMS AND AIR QUALITY 6
that their proposed solution decreases CO2 emission from ventilation systems by saving 24%
of total energy consumed and providing a cost saving of up to 19% per annum.
Literature Review #2: Room Air Quality and Unde...


Anonymous
Great study resource, helped me a lot.

Studypool
4.7
Trustpilot
4.5
Sitejabber
4.4

Related Tags