dmims Chapter 5: Earthquake Prediction
Chapter 5: Earthquake Prediction
The third floor was quieter than he had expected.
It wasn't the kind of quiet that comes from emptiness; it was heavy, as if the air in this space was denser than elsewhere. As he stepped onto the stairs leading to the third-floor corridor entrance, he sensed a difference in the density of the floor compared to the second floor. There were fewer cavities beneath the floor slabs, and the resonance of his footsteps faded more quickly. He paused for a second on his first step, marking this sensation in his mind, before continuing onward.
He never put down the testing hammer.
He walked this route from the second-floor corridor to the north staircase and then to the third-floor entrance, a distance that took him about three minutes and twenty seconds. During this time, the testing hammer was kept with its conical tip pointing downwards, and he lightly touched the ground every four to five steps to maintain contact between the tool and the concrete, ensuring a continuous transmission of vibration signals. He had done a systematic timekeeping exercise—from the first touch of the testing hammer to feeling the "pre-compression" signal, and then to hearing the actual footsteps of the plant supervisor, the average time difference was 0.31 seconds, with a standard deviation within 0.05 seconds.
Stable and repeatable.
This is no longer "to be verified". It is a data point with a mean and a bias, which can be modeled.
He paused at the entrance of the corridor, lightly touching the ground with the test hammer to sense the current signal. The factory monitor was in the northeast corner, moving at a rhythm of one step every two seconds, with no directional deviation. The error between the predicted step point and the actual step point from the previous "pre-compression" was within twenty centimeters. In the parentheses in his memo, he crossed out the words "preliminary stability" and replaced them with: "Verification passed (n=17, error <20cm)".
The man he met on the second floor stopped next to him and was also looking down the corridor.
The third-floor corridor is narrower than the second-floor corridor, about 2.5 meters wide, about half a meter shorter. The doors on both sides are closer together. On the north side of the corridor, there is a partial collapse of the ceiling, caused by the peeling of the cement mortar layer. It is not a load-bearing structure, but the fragments are scattered on the corridor floor, occupying one-third of the passage width. The largest piece is estimated to be 40 by 60 centimeters, with a thickness of 3 to 4 centimeters. There are also some smaller fragments, ranging in size from the size of a coin to the size of a fist.
This requires a detour. The passage width is reduced from 2.5 meters to about 1.7 meters, and the detour route will be biased towards the center line of the corridor—the floor vibration transmission efficiency is higher at the center line, and if there is a stomping sound, the downward propagation path is more direct.
He marked this risk in his memo, drew a detour route, and noted "stay close to the east side, land on your toes, control the landing point."
Then he started walking inside.
It took him about one minute and thirty seconds to navigate around the debris area, twenty seconds longer than he had estimated. This was because he had misjudged the position of one of the pieces—he had assumed it was stable, but when he stepped on it with his right foot, the piece slipped slightly. Its bottom was made of fine mortar particles, and there was no friction lock between it and the ground; it was a slippery point. He felt that slight loosening before his center of gravity completely shifted, immediately shifted his weight back to his left foot, paused for about two seconds, and then walked around the piece and continued walking.
During the two-second pause, he touched the ground with the testing hammer.
Factory monitoring, northeast corner, rhythm is normal, no change.
it is good.
He continued forward, towards the north end of the corridor, where the function signs were on the wall ahead. He had already confirmed this in Chapter 3: "Third Floor, Control Area, A/B/C," with the arrow pointing upwards, turn right. Turning right leads east, and the control room is on the right side (east side) of the corridor, the arrow pointing to the third door at the north end.
He counted the number of doors, starting from the stairwell, one, two, and then the third.
The control room door was made of iron, unlike the other doors—the other doors were wooden, some rotten, some rusted. This door was made of iron, thicker and heavier, and the rust marks between the door frame and the door panel were reddish-brown, forming a continuous rust layer, indicating that the rusting had been going on for a long time and that the iron had formed a chemical bond.
He didn't push it immediately, but first placed his hand in the center of the door panel to feel the vibration transmission. The iron plate has a high density and better conduction efficiency than concrete. The vibration signal he felt through the door panel was even clearer than that of the detection hammer—the factory supervisor's footsteps traveled through the floor slab, the wall, and then to this iron door, and the signal transmitted carried the high-frequency oscillations unique to metal, which were different from the low frequency of concrete. The two frequencies superimposed to form a complex vibration spectrum.
He paused for about five seconds and noted the diagram down in his memo: "Iron gate vibration guide: high + low frequency composite, clearer factory monitoring signal. If there is a similar metal structure inside the control room, it can be used as a continuous sensing tool."
Then he began to assess what kind of force was needed to open the door.
The chemical adhesion formed by rust requires a stable, slowly increasing shear force to break, not an impact force. An impact force would produce a loud, instantaneous noise, making it the worst choice; a slow, static application of force allows the rust layer to crack gradually from a single point. This cracking process produces sound, but it is diffuse and low-frequency, not a concentrated, high-frequency burst. He had dismantled aging equipment on-site and knew the difference between these two sounds—the former sounds like metal colliding, while the latter sounds like fine friction; their propagation distances differ, and the probability of them being perceived is completely different.
He estimated the rust area and adhesion strength, visually estimating that the rust layer was about two to three millimeters thick, covering an area of about 60 to 70 percent of the hinge region. The required stable shear force was about 30 to 40 Newtons, and the entire process would require about 25 to 30 seconds of continuous force application.
He wrote the data into his memo, then stepped aside and glanced at his watch.
Five minutes and fifty seconds had passed since he left from the second floor. There were approximately eleven minutes remaining at the window.
Eleven minutes.
He recalculated the time allocation: opening the door takes about 30 seconds, confirming the plant supervisor's current location after entering the control room takes about one minute, calculating the optimal timing for pressing the button takes about 30 seconds, activating the button takes about 30 seconds, confirming the route before evacuation takes about 30 seconds, and the evacuation route is third floor → freight elevator shaft → first floor → side door, estimated to take about three to three and a half minutes. The total takes about six to seven minutes, leaving a safety margin of about four to five minutes.
It's feasible, but not lenient.
He turned to look at the player who had followed him all the way. The other player was standing in the middle of the corridor, about three meters away, his feet on the east side of the path the debris had taken, without stepping on any loose spots. Xie Chengzhou glanced at his position; it was standard, completely within the safe zone. He had memorized every detour path, and his ability to judge where to land had improved significantly in the past half hour—from "landing where Xie Chengzhou had walked" in Chapter 3, he had now begun to judge his own landing points, no longer relying entirely on copying routes.
Xie Chengzhou made no comment on the matter, but simply recorded it in his memo: "Fellow players: improved judgment on where to place their bets, increased proportion of independent judgments."
Then he turned around, lightly touched the ground with the testing hammer again, and made a final confirmation of the position.
Factory monitoring, northeast corner, two seconds per step, normal rhythm.
But this time, he sensed something different.
It wasn't that the factory supervisor's position had changed, but rather that the frequency of the "pre-compression" signal had undergone a slight change—the rhythm was still two seconds, but the "pre-contraction" amplitude of the signal was about 20 percent smaller than before. It was as if the factory supervisor's walking speed itself hadn't changed, but the way the center of gravity shifted had been slightly adjusted, and the angle of the step was slightly different.
He had worked here before and had seen this pattern: when a heavy object moving on a fixed route begins to prepare for a change of route, its motion parameters will first undergo a slight adjustment, which is a physical manifestation of inertia—the route change does not happen suddenly at a certain moment, but accumulates from the steps in front of that moment.
The factory supervisor is preparing for the next route change.
He quickly converted this judgment into time points: he had only observed this amplitude change of the "pre-compression" signal once during the entire upstairs process, which meant it was a low-frequency signal that appeared about 18 to 20 minutes before the route change. He started to work backwards, the last route change was at the end of Chapter 3, so how many minutes had passed since then? There were about 10 to 12 minutes left before the next route change.
The change in "pre-compression" magnitude occurs approximately ten minutes before the route change.
He wrote this new discovery into his memo: "Hidden Rule D (Extension): The pre-compression amplitude is reduced by about 20%, appearing about 10 minutes earlier than the route change. It can serve as an early warning signal for route changes. (n=1, to be reproduced and verified)"
This was a conclusion based on only one data point, which he marked with "n=1" to indicate that he wouldn't overly rely on it in the current copy. But he remembered it.
He stood up, put the testing hammer back in his waist bag, and glanced at his watch again.
With a ten-minute change window, he needs to enter the control room before the change is completed, confirm the new plant monitoring route, and then activate it within the coverage blind spot of the new route. If the new route is opposite to the control room, his window is sufficient; if the new route is close to the control room—that requires a different calculation.
In his memo, he wrote: "Optimal solution: Start pressing immediately after the route change. The plant supervisor will be about 30 minutes away from the next change after the change. If the new route is away from the control room, there is enough time to activate the window. If the new route is close to the control room, wait in the control room and use the window's reflective glass to track the plant supervisor's position in real time before making a judgment."
He had calculated everything, no matter the situation.
He stood back in front of the control room door, took the testing hammer out of his pouch, and lightly tapped the rust layer in the hinge area with the flat end—the sound was very soft and decayed quickly; he used it to judge the density of the rust layer, not to produce sound. The rust layer was dense, consistent with his prediction, requiring a stable shearing force, about thirty seconds.
He glanced at the player who had followed him.
"Step back three meters, against the west wall," he said, his voice so low only the other person could hear. "Place your hand on the wall and feel the vibrations. If the supervisor's rhythm changes, signal me with a hand gesture. Two fingers indicate 'alert status,' four fingers indicate 'monitoring status.'"
The person paused for a moment, then nodded, walked to the west wall, and placed his palm flat on it.
Xie Chengzhou placed both hands on the control room door panel, with his right hand about 15 centimeters above the hinge and his left hand near the door handle. He applied force with both hands simultaneously, in a horizontal shearing direction, slowly increasing the force.
In the third second, a faint friction sound began to emanate from the rusted layer, low-frequency, like grains of sand moving between iron plates.
He maintained his momentum and continued.
At the ninth second, the sound frequency slightly increased, indicating that the adhesive chemical layers were beginning to peel off over a wider area. He focused all his hearing on this sound, while simultaneously paying attention to the other part of his feet—the factory supervisor's rhythm was still two seconds per step, unchanged.
At the seventeenth second, there was a slight tearing sound, not the collision of metal, but the cracking of rust, which lasted for about 0.8 seconds before stopping. He didn't stop; he maintained his strength and continued pushing forward.
In the twenty-fourth second, the door moved.
Instead of springing open, the door moved forward slowly with decreasing resistance. He minimized the pushing speed, using minimal force to maintain the door's movement, preventing it from accelerating due to inertia and making an excessively loud noise when fully open once it lost its locking point. After the door moved about thirty centimeters, he switched from pushing to controlling, placing his palm against the inside of the door to further slow its movement until it was almost imperceptible. Then, when the door was about sixty centimeters open, he stopped it.
Sixty centimeters is enough for you to squeeze in sideways.
He turned his head and glanced at the person.