Temporal noise variation during COVID-19
While significant noise variations across the array are detected, here we show the complete temporal noise variations from March 3 to June 10, 2020. Figure 4 shows the time-lapse noise changes recorded by Ch 981 located beneath Curtin Road on the main campus (Ch 204 in Figure S1 and Ch 1491 in Figure S2). As a comparison, seismic noise changes are plotted against the Google mobility data from workplaces and transport across the county (details in Text S2). Although detailed mobility data near the fiber are unavailable, a general validation could be conducted.
First, we can see that noise experienced a slight decrease (up to 10%) in the spring break compared to the regular spring semester in the low-frequency band (0.01–10 Hz). This decrease (<10 Hz) is attributed to the least school activities during spring break (i.e., many students left school and there were few school activities). In 10–50 Hz, the noise changes in both Ch 981 (Figure 4) and Ch 1491 (Figure S2c) remain flat before the stay-at-home order, while the change in Ch 204 (Figure S1) decreases. During spring break, the quiet roads (Ch 204) are more likely to have reduced vehicle traffic while the main road on campus (Ch 981 and 1491) might remain busy (e.g., citizens driving across the campus and regular bus services). In the high-frequency range (50–100 Hz), the noise changes only decrease in channel 1491 (Figure S2d), which is likely caused by stopping machinery noise (from a construction site near Ch 1491).
After the university closure on March 18, a distinct drop (up to 60% daily average) of noise levels falls to the lowest level in the whole period of the stay-at-home phase (Figures 2 and S1 and S2). Moreover, this universal noise reduction in all frequency bands (0.01–100 Hz) reflects the quieter period and the huge reduction of noise sources due to the stay-at-home order. We inferred that the campus had very few human activities.
After Phase Yellow on May 27, the noise level (0.01–10 Hz) still stays flat at the lowest level (50%~60% reduction) until Phase Green. This feature implies that residents continued to follow the stay-at-home guidelines (e.g., working at home). Interestingly, the noise level (10–100 Hz) increases gradually, which is consistent with the mobility data (transport), suggesting the recovery of road traffics and industrial activities (e.g., shopping and construction business). After Phase Green, the noise in all frequency bands gradually increases by a few percentages (1–10 Hz) to 20% (0.01-1 Hz).
We calculated the root mean square error (RMSE) between Google mobility data from two categories and noise changes in four frequency bands discussed above (Figure 4). A smaller RMSE represents a better correlation with the mobility data in a particular category. We interpret noise sources in the frequency bands of 0.01–1, 1–10, 10–50, 50–100 Hz as school activities mixed with bedrock loading of vehicles (Lindsey et al., 2020); school activities; traffic signals; traffic signals mixed with industrial activities, respectively.