The Silence of Canadian Cities: Seismic Impact of COVID-19 Lockdowns

ARTASH NATH

he/him | age 17 | Toronto, ON

Edited by Emily Lind


INTRODUCTION

On 11 March 2020, the World Health Organization (WHO) declared COVID-19 a pandemic. Countries worldwide, including Canada, enforced periodic lockdowns—closing borders, non-essential travel, and workplaces to restrict human movement (World Health Organisation, 2020). Lockdowns are essential to stop human transmission of the virus but come with huge social, economic, and mental health costs. There was a marked decrease in air travel, a contraction in the Gross Domestic Product (GDP), a sharp rise in unemployment, and a significant hit to the restaurant sector (Figure 1). However, lockdowns are only effective if citizens comply with the restrictions and stay home.

Figure 1: Impact of Lockdowns on Canadian Economy and Society as reported by Statistics Canada, showing trends in Air Travel, GDP, Employment, and Restaurant Sectors

Background Research
Research studies have shown that while natural ground vibrations from earthquakes occur at a lower frequency of 0 Hz to 2 Hz, human-caused vibrations from transportation, construction, and cultural noises occur at a higher frequency of 5 Hz to 20 Hz (Díaz, J. et al. 2020; Lecocq, T. et al. 2020) (Figure 2).

Figure 2: Sensitivity of seismic stations to ground vibrations

Approach
The research built upon findings from previous studies to monitor the effectiveness of pandemic lockdowns in Canadian cities. Expertise in seismic data analysis was achieved through reviewing scientific papers, engaging in virtual meetings with seismologists, and participating in online hackathons. This provided a solid foundation for analyzing human movement at the surface level by utilizing seismic data captured from below the surface. Seismic stations are used to detect earthquakes. They are very sensitive to ground vibrations, so they are purposely built away from cities to isolate them from urban vibrations of trains, traffic, construction, sports, and cultural events. Lockdowns would impact these urban vibrations and should be reflected in the seismic data. Additionally, seismic stations are spread worldwide, and the data is free and open source in most countries. The ubiquity of seismic stations facilitated the study for Canadian cities and holds the potential for its global expansion to measure lockdown effectiveness in managing future pandemics.

Research Hypothesis
Closure of borders, non-essential travel, workplaces, tourism, and cultural events during lockdowns should decrease anthropogenic seismic vibrations recorded by seismic stations near Canadian cities. Anthropogenic seismic vibration levels should increase once the lockdowns are over.

Research Goals
The primary objective of this research was to create an open-source tool that policymakers, health professionals, media, and the public could utilize to receive current updates on the efficacy of lockdown measures. To achieve this, the project first sought to develop a quantitative model capable of measuring changes in human movement in real-time. The model should be engineered to use open data sources to ensure insights on lockdown effectiveness could be gained without compromising individual privacy. The final step would involve applying this model specifically to quantify the reduction in human activity resulting from COVID-19 lockdowns in major Canadian cities. By focusing on this application, the project could provide concrete evidence on which lockdown policies are effective at reducing human activity, thus offering critical insights for shaping both current and future public health strategies.

MATERIALS & METHODS

Data Source
This study used data from the Canadian National Seismograph Network, a network of 150 seismic stations maintained by Natural Resources Canada and the Incorporated Research Institutions for Seismology (IRIS) for open standards on seismological data (Figure 3).

Figure 3: Sources of seismic data

The selection of cities for this study was based on four critical criteria. First, cities had to be within 50 kilometers of an operational seismic station capable of recording human activities, ensuring proximity for accurate data collection. Second, the seismic stations needed to be operational both before and after the lockdown period to provide a continuous data stream. Third, these stations should offer open data pipelines, making the information readily accessible for analysis. Lastly, to ensure a comprehensive view across the country, the cities were chosen to represent different provinces within Canada, thereby offering a diverse geographical representation.

Nine Canadian cities across six provinces and territories fit my selection criteria (Figure 4). These were Calgary (Alberta), Halifax (Nova Scotia), Montréal (Québec), Ottawa (Ontario), Québec City (Québec), Toronto (Ontario), Vancouver (British Columbia), Victoria (British Columbia), and Whitehorse (Yukon Territories).

Figure 4: Seismic stations selected and the selection criteria

Methodology
The methodology for this study followed a structured approach to data handling and analysis. The initial step was the preprocessing of data, where 16 months of raw seismic waveform data ranging from the pre-lockdown period (November 2019 to March 2020) to the during and post-lockdown periods (March 2020 to March 2021) were downloaded for each city. Data was obtained in the form of mini-SEED (Standard for the Exchange of Earthquake Data) files from each seismic station. 

Raw waveform data was resampled at 100 values per second (100 Hz), adhering to the Nyquist theorem, which facilitates frequency analysis up to 50 Hz. This resampling process resulted in the generation of over 8 billion time-series data values. Subsequently, 'Power Spectral Density' (PSD) calculations were performed to determine the intensity of seismic vibrations across different frequencies. This was done by applying Fast Fourier transformations to 15-minute segments of the waveform data.

Finally, higher-frequency vibrations associated with human activities (5Hz - 20Hz) were separated from lower-frequency vibrations associated with natural geological processes (0Hz – 2Hz). By comparing the higher-frequency anthropogenic vibration data to pre-lockdown levels, the study aimed to measure changes in anthropogenic vibration resulting from the lockdowns.

Custom Python functions and modules were developed for data analysis. Libraries used included Obspy for seismic data processing, Matplotlib for visualization, and Sci-Kit Learn for applying various machine learning techniques.

Validation
Results on changes in human movement from my model were compared with ‘Mobility Trends’ data (Apple, 2021) based on the movement of mobile devices during lockdowns.

RESULTS

Result 1: Seismic vibrations decreased during the lockdown periods for all Canadian cities
The seismic vibrations associated with human activities dropped between 12–55% compared to the pre-lockdown levels. In Calgary and Montreal, the drop was over 50%. In Halifax, Ottawa, and Whitehorse the drop was between 20–55%. In Quebec City, Toronto, Vancouver, and Victoria, the drop was between 12–20% (Figure 5).

Figure 5: Decrease in seismic vibrations across different cities during lockdowns

Result 2: Drop in seismic vibrations was higher during weekends compared to weekdays
During the lockdown periods, malls, movie theatres, sports stadiums, musical, and cultural events that attract the most traffic and human activities during the weekend were closed, leading to a sharper decrease in seismic vibrations over weekends. This further validated that my algorithm was measuring seismic vibrations associated with human activities. (Figure 6).

Figure 6: Seismic vibration levels are shown to be lower on weekends compared to weekdays. The x-axis represents time (November 2019 – March 2020), and the y-axis indicates the intensity of anthropogenic vibrations.

Result 3: Creation of MonitorMyLockdown.com web app
I created an open-source tool to monitor real-time changes in human movement during lockdowns in Canadian cities. The app http://www.MonitorMyLockdown.com/ provides information on changes in anthropogenic seismic activity by analyzing seismic data from the Canadian National Seismograph Network (Figure 7). Weekly updates have been published in English and French since February 2021.

Policymakers can use my app to monitor in real-time if lockdown legislation is working and people are staying at home to enforce surgical lockdowns based on time and location. Healthcare workers can use it to predict where pandemic cases are likely to spike and plan for testing kits and personal protection equipment.

Figure 7: MonitorMyLockdown.com interactive webapp to measure the effectiveness of lockdowns

DISCUSSION

The study's findings between mid-March and mid-May 2020 demonstrate a clear link between the enforcement of lockdowns in Canadian cities and a discernible decrease in seismic activity. This decrease in anthropogenic seismic vibrations, observed across all cities, coincided with the implementation of measures such as border closures and restrictions on non-essential travel, workplaces, construction, and cultural events (Figure 8). Similar but smaller drops in seismic vibrations were observed for subsequent lockdowns as they were less restrictive.

This study is not without limitations. Some seismic stations were located further from city centers, such as in Toronto which made them less sensitive to urban noises. In a few cases, seismic stations were taken off for repairs, such as in Yellowknife which led to the unavailability of complete datasets. Seismic data arrives at an extremely high sampling rate (100 observations/second). To calculate the Power Spectral Density on 80 billion data points using a desktop computer, some aggregation of data was done to make it more manageable. This led to some trade-offs between accuracy and speed of calculations.

Figure 8: Comparison of decrease in seismic vibrations across all Canadian cities during first lockdown. Time is indicated on the x-axis from Feb 1 to June 15. Brighter regions indicate higher seismic activity from human activities

Nevertheless, the findings of this paper are externally validated by other seismic activity research. Globally, anthropogenic seismic noise dropped as much as 50% during the lockdown between March and May 2020 in 77 countries (Lecocq, T. et al., 2020). The 12–55% drop measured by my study for Canadian cities was comparable. Additionally, the reduction in human movement obtained from the study was validated by ‘Mobility Trends’ data published by Apple (Apple, 2021) based on the movement of their devices. It showed a large decrease in human movement during the first lockdown period in Canada (Figure 9).

Figure 9: Co-relating decrease in seismic vibrations with human movements

Seismic data's capability to reflect changes in human behavior during the COVID-19 lockdowns in Canadian cities underscores its potential as a monitoring tool in public health policy. It offers policymakers and public health officials a real-time, non-invasive, and privacy-sensitive method to gauge behavioural changes in response to health directives.

Seismic data provides continuous, real-time insights into compliance with lockdown measures, essential for rapid policy adjustments in response to evolving health crises. For example, anthropogenic vibrations were lower during weekends for most cities due to the closure of shopping, entertainment, and cultural venues. Comparisons of weekday vs. weekend data and anthropogenic seismic vibrations over different frequency bands could help identify the sources of human activities. Policymakers can enforce highly precise, real-time lockdowns targeting those sources and reduce the overall costs of lockdowns.

Unlike traditional tracking methods that rely on personal device data from private companies, seismic monitoring offers a non-invasive, privacy-respecting approach. This method captures the collective behavioural response to public health directives without the need for individual opt-in or reliance on device usage patterns.

The ability of seismic monitoring to complement existing data sources, while offering broad coverage and upholding privacy, positions it as a crucial tool for managing not only the COVID-19 pandemic but also future public health emergencies. This approach underscores the potential for innovative, non-traditional data sources to enhance our understanding and management of global health challenges.

CONCLUSION

1. COVID-19 lockdowns led to seismic silences across Canadian cities
Anthropogenic seismic vibrations fell by 12–55% during the tightest lockdown weeks as borders were closed, construction and factories shuttered, aircraft grounded, and non-essential travel reduced. It was comparable to drops of up to 50% observed in other countries during lockdowns.

2. Open seismic data serves as a proxy for measuring real-time changes in human movement without impinging on peoples’ privacy
As soon as the lockdowns were imposed, an immediate drop in anthropogenic seismic vibrations was observed. The findings were validated by “Mobility Trends” data published by Apple for Canada.

3. Different types of human activities register a seismic impact
While the closure of workplaces, schools, and construction had a greater impact on weekdays or economic activities-related vibrations, the closure of the entertainment, hospitality and sports activities affected the weekend or cultural vibrations more.

REFERENCES

Apple. (n.d.). COVID‑19 - Mobility Trends Reports. Retrieved February 23, 2021, from https://covid19.apple.com/mobility

Díaz, J., Ruiz, M., Sánchez-Pastor, P.S. et al. Urban Seismology: on the origin of earth vibrations within a city. Sci Rep 7, 15296 (2017). https://doi.org/10.1038/s41598-017-15499-y

Diaz, J., Schimmel, M., Ruiz, M., & Carbonell, R. (2020). Seismometers within cities: A tool to connect Earth Sciences and Society. Frontiers in Earth Science, 8. https://doi.org/10.3389/feart.2020.00009

Lecocq, T., Hicks, S. P., Van Noten, K. et al. (2020): Global quieting of high-frequency seismic noise due to COVID-19 pandemic lockdown measures. - Science, 369, 6509, 1338-1343. https://science.sciencemag.org/content/369/6509/1338

Natural Resources Canada (2019, March), Seismograph Stations, Monitoring, Data Collection and Data Products, Retrieved from https://earthquakescanada.nrcan.gc.ca/stndon/index-en.php

Riahi, N., and Gerstoft, P. (2015). The seismic traffic footprint: tracking trains, aircraft, and cars seismically. Geophys. Res. Lett. 42, 2674–2681. https://agupubs.onlinelibrary.wiley.com/doi/10.1002/2015GL063558

Nath, A. (2023, May). The silence of global oceans: Acoustics impact of the COVID-19 lockdown. Canadian Science Fair Journal. https://csfjournal.com/volume-5-issue-4-1/2023/1/11/the-silence-of-global-oceans-acoustics-impact-of-the-covid-19-lockdown

United Nations. (2023, March 4). Intergovernmental conference on an international legally binding instrument under the United Nations Convention on the Law of the Sea on the conservation and sustainable use of marine biological diversity of areas beyond national jurisdiction. UN General Assembly. New York. https://www.un.org/bbnj/sites/www.un.org.bbnj/files/draft_agreement_advanced_unedited_for_posting_v1.pdf

World Health Organisation. (2020, March 11). WHO Director-General's opening remarks at the media briefing on COVID-19 - 11 March 2020. World Health Organization (WHO). https://www.who.int/director-general/speeches/detail/who-director-general-s-opening-remarks-at-the-media-briefing-on-covid-19---11-march-2020

ABOUT THE AUTHOR

Artash Nath

Artash enjoys tackling complex inter-generational challenges using algorithms, computations and big data in oceans, space exploration, robotics, artificial intelligence, and quantum computing. He is the Grand Award Winner of the 2022 International Science and Engineering Fair and the 2022 EU Youth for Ocean Award winner. His most recent is MonitorMyOcean.com: which uses hydrophone data to measure the acoustic silence of oceans during the COVID-19 lockdown. The App got endorsed by UNESCO as a UN Decadal Activity, and presented his work at the UN Ocean Conference in Lisbon. In 2021, he created MonitorMyLockdown.com, which used seismic data to measure the effectiveness of COVID-19 lockdowns in real-time for 9 Canadian cities. The project won him the RISE 100 Award of the Schmidt Futures and Rhodes Trust, and was invited for three weeks to South Africa to collaborate with other global winners.