The ground near a mud volcano is not merely shifting upward; it is apparently gaining mass from below. In southwestern Taiwan, that hidden change may increase hazards over time. Mud diapirs and mud volcanoes are widespread geological features driven by liquid and over-pressured mudstone moving or being pushed upward from below.

These formations can emit flammable gases, form mudflows, and release methane into the atmosphere. In southwestern Taiwan, the process is complex, and its contribution to present-day changes in the Earth's crust and to hazard assessments remains unclear.

The study investigates changes in underground mass near a mud volcano there. To understand the risk, we need to examine the driving force behind a mud diapir, as its rise can be compared with a salt diapir. Mud can theoretically rise because expanding and escaping methane makes it buoyant.

But gravity surveys found positive gravity signals offshore, suggesting relatively dense mud beneath the inland area and nearby coast. The study used different ways to monitor possible underground mass redistribution.

These measurements covered both the physical properties of the mud volcano and changes in the shape of the ground. Gravity was measured repeatedly, while vertical land movement was monitored with precise height measurements and continuously tracked positioning stations.

Groundwater and tectonic movement were also considered as factors that could affect gravity. The gravity record shows a significant increase over time, and the fitted linear trend confirms that the increase is significant. The test compared a null hypothesis of zero trend with an alternative hypothesis that the trend was not zero at all.

The statistical test rejected the null hypothesis, so the observed rise is unlikely to reflect random variation alone and is statistically significant. The record also indicates how well a straight-line increase represents the observations, providing support for treating the long-term rise as a meaningful pattern.

The measurements scatter around a steadily rising path, with their uncertainty bars revealing how closely each observation follows that long-term change. After accounting for effects from tides, air pressure, height, tectonic movement, and groundwater replenishment, the gravity record can be attributed to mud entering from below.

The remaining increase is estimated at 2.073 microgals per year. The long-term rise does not mean that mud moves continuously; intrusion may instead occur in separate, discrete episodes over time. The modeled boundary requires an underground mud body with a radius of at least 50 meters and an intrusion rate of at least 0.409 meters per year to create the observed surface gravity increase.

These ranges are consistent with reported values for active mud volcano systems worldwide, so the inferred underground activity is within the range seen in other active systems. The remaining gravity increase cannot be explained by known groundwater or tectonic processes alone, and the study therefore identifies added underground mass as the most plausible explanation, despite measurements coming from a single gravity site.

The simulations narrowed the possible combinations of underground size, depth, and intrusion rate: below the identified boundary, the model cannot produce the observed increase. Future information can test and refine these possibilities. Although underground seismic images are difficult to interpret in this region because of mobile shales, integrated time-lapse gravimetry and geodetic measurements provide evidence of mass increasing beneath WDNV.

Forward modeling in the study quantifies possible intrusion scenarios, offering possible explanations for the activity of a mud diapir in this region. If the inferred mass addition continues at its present rate, the associated hazards may grow in the future.

More measurements from other methods are still needed for a complete understanding. The evidence points to material slowly entering the shallow ground beneath the site, not just water or ordinary tectonic movement. Tracking both weight and surface motion can improve how future hazards are understood.