Morphological and Magnetic Analysis of Nieuwerkerk Volcano, Banda Sea, Indonesia: Preliminary Hazard Assessment and Geological Interpretation

Abstract

Nieuwerkerk Volcano, located in the Banda Sea, Indonesia, is a submarine volcano whose entire edifice lies beneath sea level. Its proximity to several inhabited islands raises significant concerns regarding potential impacts from future volcanic hazards. Despite historical unrest recorded in 1925 and 1927, a comprehensive geological and geophysical understanding of Nieuwerkerk remains notably limited, with the last research expedition being in 1930. This study seeks to advance our understanding of the geomorphological structure and subsurface characteristics of the region, contributing to a preliminary hazard assessment and delineating key directions for future geoscientific investigation. The data were obtained during our most recent expedition conducted in 2022. High-resolution multibeam bathymetry data were analyzed to delineate the volcano’s morphology, while marine magnetic survey data were processed to interpret magnetic anomalies associated with its structure beneath volcano. Our updated morphological analysis reveals the following: (1) Nieuwerkerk Volcano is among the largest submarine volcanic edifices in the Banda Sea (length = 80 km, width = 30 km, height = 3460 m); (2) there is the presence of twin peaks (depth~300m); (3) there are indications of sector collapse (diameter = 10–12 km); (4) there are significant fault lineaments; and (5) there are landslide deposits, suggesting a complex volcanic edifice shaped by various constructive and destructive processes. The magnetic data show a low magnetic anomaly beneath the surface, where one of the indications is the presence of active magma. These findings significantly enhance our understanding of Nieuwerkerk’s current condition and volcanic evolution for an initial assessment of potential hazards, including future eruptions, edifice collapse, and landslides, which could subsequently trigger tsunamis. Further investigation, including comprehensive geophysical surveys covering the entire Nieuwerkerk area, rock sample analysis, visual seafloor observation, and seawater characterization, is crucial for a comprehensive understanding of its magmatic system and a more robust hazard assessment. This research highlights the critical need for detailed investigations of active submarine volcanoes, particularly those with sparse historical records and close proximity to populated areas, within tectonically complex settings such as the Banda Sea.

1. Introduction

Over the past three centuries, approximately 100 volcano tsunamis have occurred in the world’s ocean, causing extensive damage to both lives and properties [1,2]. Since A.D. 1600, volcano-induced tsunamis have resulted in the deaths of over 55,000 people [3]. The deadliest volcano tsunami in recorded history happened in Indonesia in 1883. The explosive eruption of Krakatau Volcano, with a Volcanic Explosivity Index (VEI) of 6 on 26–27 August 1883, generated tsunami waves exceeding 30 m in height, which hit the shores of the Sunda Strait and resulted in more than 36,000 fatalities.

Volcano tsunamis have three main characteristics, namely low-probability, high impact, and low predictability, so they can occur without warning [4,5]. Additionally, the occurrence of volcano tsunamis is relatively low (about 6% from global tsunami) in contrast to those caused by tectonic activity [5,6]. Consequently, the existing research on volcanic tsunamis remains insufficient. According to Paris et al. [7], the ISI Web of Science database contains only 176 references for the term “volcanic tsunami,” in stark contrast to over 15,000 references for “volcano” and more than 45,000 for “tsunami”. As a result, a lot of people are not aware of the volcano tsunami hazard around them, meaning that they have a great impact on people around shores when volcano tsunamis occur suddenly. For example, the fatalities from the Anak Krakatau tsunami in 2018 occurred due to there being no warning received by local communities around shores in Lampung and Banten Provinces [8]. The tsunami happened just 20–30 min after the southwestern part of Anak Krakatau collapsed [4]. The report of [9] showed that the tsunami resulted in 437 deaths, 31,942 injuries, and 10 still missing. The most recent volcano tsunami occurred in the Pacific due to the massive eruption of Hunga Tonga–Hunga Haʻapai Volcano, with the resulting tsunami reaching a maximum height of around 15 m [10]. The eruption scale of this volcano on 15 January 2022 took researchers by surprise, as the volcano did not have a history of eruptions of such extreme magnitude.

The aforementioned points underscore the need for comprehensive investigations of submarine volcanoes. A thorough understanding of their morphology and geological structure, both at the surface and at depth, is essential for evaluating the potential of volcanic activity to generate tsunamis [5,11]. This knowledge is key to understanding hazardous behaviors and forecasting future volcano tsunami risks. Unfortunately, researching submarine volcanoes is challenging due to difficulties accessing them for sampling, high costs, and other logistical obstacles [12,13]. As a result, data on submarine volcanoes remains limited, with one notable example being the Nieuwerkerk Volcano, which is almost devoid of detailed information.

Nieuwerkerk Volcano is the submarine volcano located in Banda Sea, Indonesia, approximately 50 km east of the Emperor of China (EOC) Volcano (Figure 1). The most recent recorded activities of the Nieuwerkerk Volcano occurred in 1925 and 1927 [14]. In September 1925, there was broken water around the location of Nieuwerkerk Volcano. van Padang [14] suggested that this phenomenon could only have been caused by submarine volcano activity. A similar event also occurred in February or March 1927 [14]. As a result, Nieuwerkerk Volcano is classified as a type-A volcano (indicating activity after A.D. 1600) by the Volcanology and Geological Hazard Mitigation Center of Republic Indonesia (CVGHM). This classification indicates that the volcano may still be active, highlighting its potential for future eruptions.

Nieuwerkerk Volcano is located near several inhabited islands, including Alor Island (~160 km), Wakatobi National Park (~140 km), Buton Island (~180 km), and Southwest Maluku (~197 km) (Figure 1). These islands have a population of more than 210,000; 95,000; 119,000; and 45,000, respectively. These populations are at risk of being impacted by tsunamis triggered by volcanic activity at Nieuwerkerk. Additionally, ref. [7] highlighted that the Banda Sea is one of the Southeast Asian areas with the highest potential for future volcano-induced tsunamis. This proximity to densely populated regions further underscores the potential hazard posed by the volcano.

Research on Nieuwerkerk Volcano has been scarce, leaving its morphology and geology poorly understood. The last major expedition prior to 2022 was the 1930 Snellius Expedition, which provided the first bathymetric data on the topography of the volcano [14,21]. Meanwhile, our latest expedition took place between June and July 2022 as part of the Jalacitra II-2022 “Banda” Expedition, initiated by the Indonesian Navy’s Hydro-Oceanographic Center (Pushidrosal). Preliminary findings from this research have been previously published in the authors’ reports [21,22]. These publications provide an interpretation of the volcano’s morphometric features and identify surface (2D) magnetic anomalies in the vicinity of Nieuwerkerk, contributing to a better understanding of its morphological characteristics.

However, the information about subsurface conditions regarding the potential active magma chamber, as well as the detailed characteristics of Nieuwerkerk Volcano’s edifice correlated with the history of volcanic activities and present conditions, was not provided by those previous studies. Understanding these aspects is important for identifying past and present conditions in order to determine hazardous behavior and forecast future potential hazards. Therefore, in this study, magnetic data were modeled to investigate the subsurface conditions in selected key areas surrounding Nieuwerkerk Volcano. In addition, an advanced analysis of bathymetric data was performed to identify potential evidence of past large-scale volcanic collapses, eruption- or tectonically induced landslides, and anomalous seafloor features that may signify recent eruptive activity within the Nieuwerkerk volcanic system. By integrating geophysical data with detailed seafloor morphological analysis, this comprehensive approach aims to improve our understanding of the Nieuwerkerk Volcano’s edifice, its history of volcanic activity, current state, and potential for future hazardous events.

2. Geological Setting

Nieuwerkerk volcano is situated in the Banda Sea, Maluku, Indonesia, is described as a twin volcano with more than 7 km between peaks, and is often associated with the EOC Volcano located to its west (Figure 1). The Banda Sea (Eastern Indonesia) is located within an area where three major plates, namely the Eurasian, Pacific, and Indo-Australian plates, have been converging since Mesozoic times [23,24]. Interactions between these plates have influenced the geometry of the Banda Sea, which has an area of deep basins, surrounded by submerged rims, and tectonic faults.

The Morphology of the Banda Sea can be classified into two main basins; the North Banda Basin (Sula) lies on the north of the Banda Sea, and there is a basin in the south, which is named the South Banda Basin (Wetar and Damar) (Figure 1). All these basins are separated by a complex system of ridges, e.g., Sinta, Rama, Lucipara, Pisang, Tukang Besi, and the Nieuwerkerk–Emperor of China (NEC) Volcanoes [20,25].

Detailed geological studies of Nieuwerkerk Volcano remain limited. To date, the only known study that examines the characteristics of volcanic products in the vicinity of the volcano is presented in the author’s report [20]. A newly published sample reported by Honthaas et al. [20] indicates that two igneous rock samples, collected from sites 219 and 220, were recovered from the slopes of the volcanoes within the Nieuwerkerk–Emperor of China (NEC) Ridges (Figure 1). Those samples have an andesite composition consisting of plagioclase, orthopyroxene, clinopyroxene, and Fe-Ti oxides. The andesites obtained from the two locations typically exhibit comparable geochemical characteristics, characterized by calc-alkaline andesites, similar to volcanic products from the Wetar arc (comprising Atauro, Lirang, and Wetar islands) and the Lucipara ridges. This result implies that those volcanoes could be remnants of a single magmatic arc, potentially fragmented by the formation of the Wetar and Damar basins [20]. Meanwhile, the radiogenic isotopic signatures (based on Sr and Nd isotope) of those andesite samples align with a process of AFC (Assimilation coupled with Fractional Crystallization), which suggests that mantle-derived basaltic magmas have assimilated continental crust [20]. However, sample 220 exhibits a reduced K₂O content, an elevated Na₂O concentration, and a lower enrichment in Large Ion Lithophile Elements (LILEs), Light Rare Earth Elements (LREEs), and Heavy Rare Earth Elements (HREEs), in addition to being less radiogenic compared to sample 219 [20]. These findings indicate that sample 220 reflects less crustal involvement and represents the more mantle-like end of the NEC compositions.

The K-Ar ages of the two samples (219A and 220A) are 7.38 ± 0.17 Ma and 8.10 ± 0.19 Ma, respectively [20]. Honthaas et al. [20] also explained that the activity of the Nieuwerkerk Volcano had stopped at around 8–7 Ma following the next tectonic process, which was the back arc spreading of the south Banda Sea at around 6.5–3 Ma [23,25,26], where the spreading was considered stopped by Australia’s continental collision.

Recent morphological analysis based on new bathymetric datasets by Febriawan et al. [22], along with a magnetic anomaly study by Alodia et al. [21], revealed that Nieuwerkerk is a massive volcanic structure, covering an area of approximately 2416 km². It extends over 80 km in a northeast–southwest (NE–SW) orientation and exceeds 30 km in width. This study also reveals the discovery of three new seamounts with conical shape. The two seamounts were considered as one unit in Nieuwerkerk volcano and one seamount was located in the western part of Emperor of China. It seems that another geological process occurring in these features was an erosional process in some flanks of the Volcano. In terms of magnetic study, there are no magnetic anomalies on the East Ridge, indicating the absence of fault structures and/or magmatic rocks in the area; in addition, they identified the possibility of new or ancient magmatic flows on the main summit of Nieuwerkerk, and possibly also on secondary summits; however, the current state of lava flows and activity at Nieuwerkerk volcano remains unclear.

3. Materials and Methods

3.1. Data Acquisition

High-resolution bathymetry data of Nieuwerkerk Volcano were obtained during the Jala Citra 2 expedition on board a vessel of the Indonesian Navy Hydro-Oceanographic Center (Pushidrosal), KRI RIGEL 302, from 4 to 15 July 2022. The EM302 Multibeam Echo Sounder System with an operating frequency of 30 kHz and featuring 512 beams and a resolution of 1.50 × 1.50 was used in the survey to perform seafloor mapping at the Nieuwerkerk seamount. This echosounder is capable of mapping at a depth ranging between 10 and 7000 m and is fully equipped with the vessel position and motion sensor (heave, pitch, roll, and yaw). During the expedition, sound velocity measurements were performed to correct the depth, and navigational data were recorded via differential GPS with a positional accuracy of ±1 m. The raw bathymetric data were processed using CARIS HIPS and SIPS 11.4 software, licensed to the Indonesian Navy Hydro-Oceanographic Centre (Pushidrosal). Standard procedures for outlier removal were applied to all track lines using the swath editor, and the sound velocity profile (SVP) was incorporated to correct the raw data.

Ship-borne magnetic data were collected on 4 to 6 July and 13 to 15 July 2022 during relatively quiet interplanetary geomagnetic conditions (Kp < 4) using the Geometrics G-882 (Geometric Inc., San Jose, CA, USA) marine magnetometer and the MagLog software. The magnetometer is towed ~150 m behind the vessel. Raw magnetic data were sampled at 1 Hz to provide a synchronous GPS layback-corrected position for the tow-fish. Throughout the survey, the horizontal position of the magnetometer was determined using the offset between the GNSS antenna, the towing point on the vessel deck, the length of the towed cable, and the vessel’s instantaneous bearing. The survey line layout was designed based on the seamount morphology identified in a prior multibeam bathymetric survey conducted during the same expedition (see Figure 1 in the author’s report [21]). To minimize over-extrapolation, the spacing between adjacent survey lines was set to be no less than the average depth of the corresponding profile. Data acquisition was carried out at an average vessel speed of 7–8 knots.

3.2. Bathymetry Data Modeling

During grid generation, the cleaned sounding data were converted into a digital elevation model (DEM) using a cell size of 40 m, selected based on the water depth and the target spatial resolution. Minor data gaps were then interpolated using Swath Angle Weighted, while larger gaps were unfilled to avoid artificially extending the survey coverage. Finally, the completed DEM was exported in GeoTIFF formats and served as the foundation for generating further products such as slope, aspect, and shaded-relief maps in ArcMAP.

3.3. Geomagnetic Data Modeling

3.3.1. Total Magnetic Intensity (TMI) Map Generation

Initial data processing involved merging the magnetic data with the ship’s navigation records. To isolate the crustal magnetic component, the International Geomagnetic Reference Field (IGRF) 2020 was calculated for the epoch of the survey (2022) and subtracted from the observed total field values. This step yielded the preliminary magnetic anomaly.

A magnetic fixed station in Kupang, East Nusa Tenggara, was used for diurnal corrections. The leveled magnetic anomaly point data were interpolated onto a regular grid with a cell size of 50 m using a minimum curvature algorithm [27]. This algorithm was chosen for its effectiveness in producing a smooth surface from irregularly spaced data. The resulting grid was then used to generate the Total Magnetic Intensity (TMI) map, which serves as the primary dataset for subsequent analysis and interpretation.

3.3.2. Analytic Signal Amplitude (ASA) Transformation

To aid in the identification of magnetic source boundaries independent of remanent magnetization direction, the TMI grid was transformed into the ASA. The analytic signal is a complex function whose amplitude is derived from the three orthogonal derivatives of the magnetic field [28]. The calculation of ASA follows Equation (5) in [28]:

$$\left|A\left(x.y\right)\right|=\sqrt{{\left({\displaystyle \frac{\partial M}{\partial x}}\right)}^2+{\left({\displaystyle \frac{\partial M}{\partial y}}\right)}^2+{\left({\displaystyle \frac{\partial M}{\partial z}}\right)}^2}$$

(1)

where M is the TMI data, and ${\displaystyle \frac{\partial M}{\partial x}}$, ${\displaystyle \frac{\partial M}{\partial y}}$, and ${\displaystyle \frac{\partial M}{\partial z}}$ are the TMI derivatives in the x, y, and z directions, respectively. Meanwhile, the vertical magnetic derivative, ${\displaystyle \frac{\partial M}{\partial z}}$, is calculated following Equation in [29].

$${\displaystyle \frac{\partial M}{\partial z}}={\mathcal{F}}^{-1}\left(\left(\sqrt{{\left(k_x\right)}^2+{\left(k_y\right)}^2}\right)\mathcal{F}\left(M\right)\right)$$

(2)

The resulting ASA map highlights the edges of magnetic bodies, with the peaks of the signal amplitude located directly over the boundaries or centers of the sources.

D Magnetic Modeling

During grid creation, the cleaned sounding data were converted into a digital elevation model (DEM). In parallel, the Nieuwerkerk magnetic anomaly data were modeled using 3D inversion techniques implemented in SimPEG, an open-source Python (SimPEG-based inversion (version 0.24)) library for geophysical simulations. This module sought to fit the data using models that minimize the objective function $\varphi$ so that the data misfit ${\varphi }_d\le {\varphi }_d^{*}$ the target misfit [23,24]. Assuming that the data noise is Gaussian and independent, the target misfit ${\varphi }_d^{*}\approx N/2$, which is half the number of observation data used in the modeling [20].

$$\begin{array}{l}\underset{\mathbf{m}}{\min } \varphi \left(m\right) = {\varphi }_d+\beta {\varphi }_m\\ \mathrm{so}\mathrm{that} {\varphi }_d \le {\varphi }_d^{*}.\end{array}$$

(3)

The data misfit ${\varphi }_d$ is the sum of the squared difference between the observed and modeled data divided by the data uncertainty, whilst the model misfit ${\varphi }_m$ is the distance between the inverted and a reference model [30]. Both misfits are unitless [30]. A positive constant $\beta$, known as the Tikhonov parameter, weights the contribution of ${\varphi }_m$ in $\varphi$ and is determined automatically by the program as the inversion progresses [31].

In this study, no reference model was available to constrain the 3D inversion. Consequently, the SimPEG-based inversion (version 0.24) was conducted using only unconstrained smooth and blocky model endmembers. These endmembers are produced by modifying the ${\mathcal{l}}_p$-norm parameters that regulate the smoothness of the change in physical properties; in our case, this was the magnetization within the model [32]. The ${\mathcal{l}}_p$-norm is an array of four values $\left[p_s,p_x,p_y,p_z\right]$ that each range from 0 to 2 and set how compact (or small) the model bodies are (p_s) and the smoothness of physical property change in three directions ($p_x,p_y,p_z$) [32]. A value of zero for p_s shrinks the model bodies’ size, while a value of two enlarges them [31]. On the other hand, letting ($p_x,p_y,p_z$) = (0, 0, 0) sharpens the model body boundaries, while setting ($p_x,p_y,p_z$) = (2, 2, 2) smoothens them [33].

The first step in the Nieuwerkerk magnetic anomaly modeling is to grid the discrete bathymetric data points onto a bathymetric surface, which will act as the upper limit of the subsurface magnetic bodies. The bathymetric data were gridded using the kriging interpolation algorithm implemented in Golden Software Surfer version 29.3 [34], with a grid resolution of 50 × 50 m. No extrapolation beyond the data extent was applied.

As a modeling input, ungridded magnetic data cropped to a 35 × 19 km area covering the central part of Nieuwerkerk edifice were used. Following this, a modeling cube with a dimension of 51.2 × 25.6 × 25.6 km and a core voxel size of 100 × 100 × 100 m was then made. The horizontal extent of the modeling cube is equal to the cropped magnetic anomaly data, with additional 8.1 km extensions on the western and eastern boundaries and 3.3 km extensions on the northern and southern boundaries to accommodate edge effects. For modeling purposes, the magnetic anomaly data were down-sampled from 0.4 m to a 200 m resolution, twice the horizontal resolution of the core voxel size, in accordance with the 3D modeling guidelines proposed by the study’s author [33].

Since the nature of Nieuwerkerk rock magnetization is not fully understood, the 3D inversion is carried out in scalar and vector magnetic modes. The scalar magnetic inversion estimates the distribution of subsurface rock susceptibility, with the final model output expressed in SI units. In contrast, the vector magnetic inversion recovers both the effective susceptibility (in SI units) and the direction of the rocks’ total magnetization. The effective susceptibility ${\kappa }_e$ can be converted to magnetization by multiplying it with the strength of the local geomagnetic field $\parallel {\mathbf{H}}_{\mathbf{0}}\parallel$ [32].

$${\kappa }_e \times \parallel {\mathbf{H}}_{\mathbf{0}}\parallel =M$$

(4)

To collect models that may represent the Nieuwerkerk subsurface, we need to understand the data uncertainty level and the possible upper bound of the magnetic susceptibility value of the model. To collect models that may represent the Nieuwerkerk subsurface, the data uncertainty level and possible upper bound of the magnetic susceptibility value of the model need to be understood. However, since even estimates of those values are not available, they are determined iteratively by testing several candidate values, ranging from 1 to 6 nT for data uncertainty and 0.1 to 0.3 for the upper bound of the model’s magnetic susceptibility, using an ${\mathcal{l}}_p$-norm = [2, 2, 2, 2] (smooth model endmember). From this experiment, it was found that 2 nT is the best data uncertainty level to invert the Nieuwerkerk magnetic anomaly data, with 0.3 SI being the most appropriate upper bound for the model’s magnetic susceptibility. These values produce a model that minimizes both the ${\varphi }_d$ and $\varphi$ closely or slightly lower than the target misfit (Figure 2).

Using the 2 nT data uncertainty value and 0.3 SI maximum model susceptibility, the possible candidates of the blocky model endmember were then explored. Several ${\mathcal{l}}_p$-norms were tested (e.g., [2, 0, 0, 0], [1, 1, 1, 1], [1, 2, 2, 2], [2, 1, 1, 1], and [0, 2, 2, 2]). The [0, 0, 0, 0] norm was not tested as the magnetic anomaly variations are gradual and do not suggest any presence of sharp physical property transitions. From this experiment, it was found that scalar and vector magnetic models made using an ${\mathcal{l}}_p$-norm = [0, 2, 2, 2] both minimized the ${\varphi }_d$ and $\varphi \approx N/2$ (Figure 1) with a magnetization distribution that was consistent with the models produced with ${\mathcal{l}}_p$-norm = [2, 2, 2, 2].

4. Results

4.1. General Morphology of the Nieuwerkerk Volcano Edifice and Surrounding Areas

New high-resolution bathymetric data acquired during the Jalacitra II-2022 expedition reveal a substantially detailed view of the Nieuwerkerk Volcano’s overall shape, summit depth, and flank characteristics (Figure 3). First, the survey indicates that the highest point of the edifice is located around 300 m below sea level (bsl). From this shallow summit to the surrounding seafloor (approximately 3760 m bsl), the volcano exhibits a total relief of about 3460 m. Second, the updated DEM shows that the main volcanic edifice extends approximately 80 km in a NE–SW direction, with a maximum width of around 30 km (Figure 3). Within this footprint, the new topography clearly identifies two primary summits. The summits lie at around 350 and 380 m bsl, forming a twin-peak structure.

The Nieuwerkerk Volcano exhibits an elongated structure, with an approximate length-to-width ratio of 8:3. However, its morphology varies significantly between the western and eastern sectors, particularly in terms of slope gradients, volcanic features, and the presence of structural lineaments. Accordingly, a detailed morphological analysis is presented in two separate sections, each focusing on one part of the edifice.

4.2. Morphology of the Western Part of Nieuwerkerk Volcano

The slope near the rim of this part is notably steep, measuring ~20–35°. In contrast, the central area exhibits a basin-like morphology, characterized by a comparatively gentle and flat slope (<10°). In Figure 4, the dotted line in the central area outlines a broad, circular to elliptical feature that appears to truncate the original summit area—an observation that raises the possibility of a major summit collapse. The dashed outline encloses an area approximately 10–12 km in diameter.

Figure 4 shows topographic cross-sections (A–B, C–D and E–F) across the Western part of Nieuwerkerk volcano. Each profile extends roughly 14–16 km in length and reaches depths of up to about 2000 m below the local summit. Cross-section A–B traverses the volcano’s Western flank, revealing a broad depression at the central area of the profile and steepening slopes toward both ends. The depth increases markedly in the middle section (around 4000–6000 m along the horizontal axis), suggesting a significant topographic low that may correspond to a crater-like depression, a collapsed sector or a large-scale sector failure that removed the original summit. Cross-section C–D (Figure 4D) similarly shows a relatively deep, laterally extensive depression at the center of the profile, with shallower depths (i.e., higher topography) near the edges.

These cross-sections highlight a notable break in slope that is associated with concave-upward topography, commonly observed in collapse structures on volcanic edifices. The difference in elevation between the highest points and the central low can exceed several hundred meters, underscoring the considerable relief on the volcano’s western side. The eastern edge of this depressed region is characterized by a steep slope of about 30–45°, exhibiting a lineament with NW-SE orientation. Furthermore, there are resistant volcanic products (e.g., lava) in the southern part of this collapse structure, where these products are relatively more resistant compared to those in the northern part. We predict that these products are associated with the remains of an older volcanic edifice that was not shattered due to collapse. Meanwhile, the lineaments are glimpsed in this area with dominant NE-SW orientation (Figure 4).

4.3. Morphology of the Eastern Part of Nieuwerkerk Volcano

The slope in the eastern part is more steep than the western part, at approximately 4–45° degrees. The steepest slope is in the center region (30–45°) of this part where there are volcanic centers. A notable twin-peak volcanic feature is present at the volcanic centers; these are clearly visible on the summit area, which are approximately 7–7.5 km apart (Figure 5B). Importantly, one of these peaks appears truncated by a linear structure (likely a fault line), causing a lateral offset. Specifically, the western peak is offset, indicating tectonic displacement likely related to regional fault activity (Figure 5D, line section A–B). In addition, surrounding the twin-peak structure, minor faults are evident, indicating significant tectonic deformation localized around the volcanic summits (Figure 5B). These minor structures further suggest active tectonic control influencing volcano morphology.

The eastern flank also exhibits a tear-like morphology characterized by steep slopes (approximately 55°) (Figure 5A). This steep morphological feature trends primarily in a northeast–southwest (NE-SW) direction. It distinctly contrasts with the southern slope, which has relatively gentler slopes (less than 15°). Deep incisions along fault lines cause slope differences and morphological tears (Figure 5D, line section C–D). The steep eastern flank extends significantly (marked by long fault line), similar in character to a long, steep slope observed to the south, suggesting alignment with major structural lineaments extending southwestward (Figure 5A).

Several channels or gullies radiate outward from the volcanic peaks, representing erosional features likely formed by submarine mass-wasting processes, sediment gravity flows, or volcanic activity. These features suggest ongoing geomorphological processes that continue to shape the volcanic edifice. In addition, a prominent ridge, interpreted as a dike or lava ridge, is observed along the extreme eastern margin of the volcano (Figure 5A,E). This ridge is elongated (~15 km in length, ~3–5 km in width) and trends primarily NE-SW. The slope angles of the ridge are moderately steep (~20°), and its morphology is elongated and somewhat linear, indicating possible tectonic and volcanic origins linked to underlying geological structures.

In the eastern part of Nieuwerkerk Edifice, other volcanoes were also found (Figure 5A,C,E). The morphologies are not full because of collapse and erosion, as shown by the rugged topography and singular feature (e.g., crater or caldera). More than one singular feature was found, which shows that there was potentially more than one volcano or eruption center. Based on the morphology, those volcanoes are relatively older than Nieuwerkerk Edifice.

4.4. Evidence for Potential Landslide Scarps

Figure 6A,B present color-shaded relief maps of the same area from a three-dimensional perspective, emphasizing the rugged morphology on the transition zone between the western and eastern part. Dashed lines in these figures mark the approximate boundaries of what appear to be large arcuate scarps (12 km in width and 13 km in length). The curved geometry and downslope orientation of these scarps are characteristic of landslide headwalls or sector-collapse amphitheaters, where a portion of the volcanic flank detaches and slides outward. The surface within these boundaries exhibits hummocky textures and irregular slopes—features that are often associated with mass-wasting deposits [35,36,37].

In both the plain view (Figure 6A) and oblique perspective (Figure 6B), the scarps appear segmented, which may indicate either a single large-scale event that produced multiple headwall arcs or a series of successive slope failures over time. The broad spatial extent of these features suggests that the events were substantial enough to significantly reshape the eastern flank of the volcano. Tracing this area upward along the scarp reveals a channel-like or debris scar feature, interpreted as the path of mass movement along the flank near the summit. This path aligns with the lineament features observed in the surrounding terrain and is bounded by a linear structure that intersects one of the volcanic peaks.

4.5. Magnetic Anomaly of Nieuwerkerk Volcano

The difference between the observed and calculated magnetic anomaly for each model end-member, henceforth termed as “modeling residual,” was checked. It was found that all models can fit the observed data well without any systematic spatial variations in the modeling residual (Figure 7). Based on both the observed and calculated magnetic anomaly maps (Figure 7), the area surrounding the peak of Nieuwerkerk Volcano exhibits a pronounced negative magnetic anomaly. This finding is consistent with previous work by [32], who reported a Reduced-to-Pole (RTP) anomaly value of approximately –120 nT in the same region. In the present study, the magnetic anomaly at the volcano’s peak ranges from approximately –130 to –175 nT.

Based on the observed magnetic anomaly, the TMI and ASA maps were generated and integrated with bathymetric data (Figure 8). Both positive and negative magnetic anomalies have been identified in the western and eastern parts of Nieuwerkerk Volcano (Figure 8A). In the western part, the positive anomaly is located within the collapse area, whereas the negative anomaly is observed on the western and southern flanks of this part. Conversely, in the eastern part, the area surrounding the summit (twin peaks) predominantly exhibits a negative magnetic anomaly, which is also present on the western, northern, and southern flanks. The sole positive anomaly is noted in the eastern part, specifically in the vicinity of a tear-like morphology and where another group of volcanoes is situated. In addition, in the southern part of Nieuwerkerk’s edifice, the negative magnetic anomaly appears to extend approximately 50 km. Meanwhile, the 3D Analytic Signal Amplitude in the region is generally below 0.05 nT/m. However, near the twin peak, a circular, more continuous area of >0.05 nT/m analytic signal is encountered (Figure 8B). In addition, the higher value (>0.05 nT) is also observed around the collapse area in the western part.

The Subsurface Magnetic Anomaly

For understanding the subsurface condition, the geomagnetic data were modeled. Following the conclusion of the inversion process, the model produced at each iteration was checked, guided by the misfit evolution curve. The model produced at the 25th iteration is found to be the most representative for every end-member of the scalar and vector inversion, as this model is associated with a value of ϕ that is close to the target misfit (Figure 2). Beyond the 25th iteration, the inversion either overfit or slightly underfit the model (Figure 2).

Scalar magnetic model slices consistently show a concentration of magnetic layer and bodies at <5 km depth under the Nieuwerkerk edifice when the magnetic susceptibility is about 0.06–0.1 SI (Figure 9). In these models, the thickness of the magnetic units is variable, although they are never thicker than 10 km (e.g., Profile CC* in Figure 9). The magnetic layer in the Nieuwerkerk scalar magnetic models tends to become thinner at the central part of the profile, which is also located over the center of the Nieuwerkerk edifice (Figure 9). Underlying this magnetic unit is a large weakly, or non-magnetic, zone (marked by DM in Figure 9). Smaller weak or non-magnetic zones are also embedded within, or overlying the magnetic units, in Profile AA* and BB* to the north of the Nieuwerkerk edifice (Figure 9).

Compared to the scalar magnetic models, the vector magnetic models of Nieuwerkerk exhibit a more uniform magnetic body distribution (Figure 10). Shallow magnetic bodies of the scalar magnetic models reappear in Profile AA*, the western part of BB* and CC*, as well as the eastern part of Profile BB* to DD*, with the magnetic susceptibility being about 1.03–2.58 A/m (Figure 10). A difference between the scalar and vector magnetic models is visible at the western part of Profile BB* to DD* and eastern part of Profile BB, where the absence and presence of magnetic bodies are inverted. For example, a low magnetic susceptibility zone in the western part of the scalar magnetic model of Profile BB* is replaced with bodies having > 1 A/m magnetization in the vector magnetic model of the same profile (Figure 10). Another aspect in which the scalar and vector magnetic models are different is the presence of a weakly magnetized body (∼0.4 A/m) at <2 km depth under Profile BB* and CC* in the central part of the Nieuwerkerk edifice (Figure 10). The more uniform distribution of magnetic bodies in the vector magnetic models might come from the inversion’s flexibility in fitting the complex pattern of Nieuwerkerk magnetic anomalies with voxels of variable magnetization direction. Considering the age, volume, and potential longevity of Nieuwerkerk volcanism, vector magnetic modeling might be able to recover deeper magnetized bodies but at the expense of the shallower ones, reflecting the inherent ambiguity of magnetic modeling and the need for a better model constrain.

The magnetization directions of the shallow (<2 km) magnetic units (magnetizations > 0.5 A/m) modeled under Nieuwerkerk are not uniform. These may take any declination (Figure S1) and inclination (Figure S2) angle, and hence are too ambiguous to indicate a particular magnetization direction. Geologically, these intricately magnetized shallow units may be associated with Nieuwerkerk eruptive products, such as lava flows and pyroclastics, which are often magnetized in a complex way [38]. However, the declination and inclination of the deeper weakly magnetized body (“DM” in Figure 8 and Figure 9 and Figures S1 and S2) consistently indicate a declination of ≤90° (Figure S1) with a negative inclination (Figure S2). However, the declination and inclination of the deeper weakly magnetized body (“DM” in Figure 9 and Figure 10 and Figures S1 and S2) consistently indicate a declination of ≤90^∘ (Figure S1) with a negative inclination (Figure S2).

5. Discussion

Our updated morphological analysis identifies the Nieuwerkerk Volcano as one of the largest submarine volcanic edifices in the Banda Sea, with its summit located at a significantly shallower depth (~300 m below sea level) than previously reported—contrasting with the ~2300 m bsl estimate from van Padang’s early work. The 2000+ m discrepancy likely reflects the limited accuracy of the 1930s single-beam or lead-line measurements, rather than rapid tectonic uplift or extreme volcanic growth within the past century. This assertion is supported by studies calculating uplift rates on some islands in the Banda Sea region (e.g., Sumba Island, Alor, Sumbawa, Savu, Kisar, Kupang, and Timor), which indicate that the maximum uplift rate observed is between 2.3 and 5 mm/yr [39,40,41,42,43]. Consequently, it is implausible for the volcanic edifice to have experienced an uplift of approximately 2000 m in 91 years. Furthermore, our recent geomagnetic data provide novel insights into the subsurface conditions beneath Nieuwerkerk Volcano. This finding enhances the earlier analysis by [32], which was limited to revealing magnetic anomalies on the surface.

Those new insights provide a deep understanding of the detailed characteristics of Nieuwerkerk Volcano, which is made up of volcanic features, and their correlation with tectonic and erosive–depositional processes, as well as the potential active magma chamber or hydrothermal system beneath Nieuwerkerk Volcano. A better understanding of those aspects is important for assessing the potential hazard in the future.

5.1. The Interaction of Volcanic Features/Morphology, Volcanism, Tectonic, and Erosive–Depositional Processes

This study presents evidence of significant volcanic morphologies, including two distinct peaks and volcanic products, specifically lava. This information is crucial as it indicates the presence of either currently active or previously active volcanic systems. Observations of lava flow were recorded in both the western and eastern parts of Nieuwerkerk Volcano (refer to Figure 4 and Figure 5). In the western part, the lava flow is prominently observable along the flank of the volcano (see Figure 4B). Similarly, in proximity to the peak in the eastern part, the lava flow is distinctly evident. In the area that has experienced collapse, remnants of volcanic products can be identified, taking the form of a ridge that corresponds to the remains of the former Nieuwerkerk Edifice (see Figure 4B).

However, while volcanic rock is present throughout all regions of Nieuwerkerk Volcano, varying magnetic anomalies were detected, exhibiting both positive and negative characteristics. High (positive) magnetic anomaly in the volcano area is associated with fresh volcanic products, such as lava, volcanic breccia, and intrusion rock (magma body which has cooled until below curie temperature) [44,45]. Meanwhile, research conducted by [46] revealed that the lava in the caldera of Axial Seamount, which was produced from the eruption in 2015, showed a significantly different magnetic anomaly in 2017. The lava was cooler and more magnetized in 2017 than in 2015, which caused the magnetic anomaly to be higher. This means that the old volcanic product, especially lava, can be reflected from the high positive magnetic anomaly. This result is also similar to the conditions at Yellowstone National Park, where the old volcanic products have a positive magnetic anomaly. Meanwhile, the areas with younger and active volcanism have a low magnetic anomaly [47]. The correlation between low magnetic anomaly and the appearance of volcanism was also found by [48,49,50], where the low magnetic anomaly beneath Palinuro and Brothers Volcanoes, southern Mariana Trough, and Daxi Vent Field on the Carlsberg Ridge correlated with magmatism activity.

The explanations above show us that the high positive magnetic anomaly observed in the central part of the collapse-like region is linked to fresh volcanic products, which represent remnants of a volcanic edifice, composed of old volcanic products (e.g., lava). In contrast, the observed low and negative magnetic anomaly, particularly in the vicinity of the twin peak, appears to be associated with either recent volcanic activity or the presence of inactive younger volcanism. This suggests that the collapse region in the western part may have historically served as an eruption center. This is further supported by the ASA results, where high ASA values in both the collapse area and the twin peak indicate a contrasting geological characteristic in these areas compared to their surroundings. However, to ensure this interpretation, further studies are required, especially petrology and geochemistry studies of volcanic product samples from the representative area within Nieuwerkerk Volcano.

The number of geological structures indicates that tectonic processes can be one of the primary factors controlling the morphology and volcanism of Nieuwerkerk Volcano, especially in the eastern sector. NE–SW lineaments are found extensively in the eastern part where the recent eruption centers are located. The orientation of those lineaments is generally similar to the faults in the Banda Sea (Figure 1). The tear-shaped in the NE of the eastern part was also similarly impacted by the fault with the same orientation. This fault (northern major fault) is one of the major faults in Nieuwerkerk Volcano, which is characterized by the relatively weak magnetic anomaly and boundary between the weak magnetic anomaly area in the left side and high magnetic anomaly area in the right side of the fault (Figure 8). A comparative analysis of the depths reveals that the left side of this fault is shallower than its right counterpart, with depths of 2050 and 2200 m (bsl), respectively. This discrepancy suggests that the left side is relatively uplifted and contributes to the steep slope of that part (Figure 5). A major fault is also found in the southern part of Nieuwerkerk’s edifice, which extends for about 50 km in the direction of ENE-WSW, which was also previously reported by [21], showing a negative magnetic anomaly. The fault is probably related to the Banda Ridge rifting in 6–3 Ma [20]. During that period, the Tukang Besi and NEC Ridge were separated from the active Wetar segment, prior to the transition from subduction to collision driven by rifting in the South Banda Sea. This rifting is typically marked by normal faulting due to crustal extension [51].

Minor faults, predominantly situated on the left side of the northern main fault (Figure 5), are inferred to have played a key role in triggering volcanic activity and associated morphological changes in the area. Notably, this zone hosts twin peaks that represent the most recent eruption centers of the Nieuwerkerk Volcano. Furthermore, the magnetic anomaly of both areas is also different, where the magnetic anomaly of the area on the left side is weaker than on the right side (Figure 8).

Additionally, in terms of structural configuration, the northeast–southwest elongation of the volcano aligns with recognized tectonic trends in the central Banda Sea, where arcs and ridges predominantly reflect plate convergence and back-arc extension [20,26]. The lineament of eruption centers at Nieuwerkerk Volcano extends in a similar direction, reaching approximately 30–40 km to the Southwest, where three younger volcanoes are located: Seamount A, Seamount B, and Emperor of China (Figure 3) [21].

Meanwhile, the faults that were accompanied by erosive–depositional processes have created the channel, scarp, debris, and steep slope, especially at the eastern part (Figure 5). The steep southern slopes and apparent breaks in gradient could be consistent with major slope failures or tectonically controlled escarpments. Incised gullies in the southwestern sector may represent channels carved by submarine mass-wasting events or sediment gravity flows, a process noted in other submarine volcanoes such as Lucipara Ridge [52]. We contend that the erosion in Nieuwerkerk Volcano (especially in Eastern Part) reflects a mature stage, considering that the volcano is sufficiently old. According to [20], the age of igneous rock located in the southern part of the study area (Figure 3) is about 7.38–8.10 Ma. These processes contributed to the current complex morphology of Nieuwerkerk’s edifice, especially in the eastern part.

In contrast, there is not major evidence for fault-like lineaments in the western part. However, its traces are still being identified, despite being faint in the central part of the potential collapse zone. This is perhaps caused by the collapse phenomena, where that process wiped out the faults and left little trace of them. The remaining faults suggest that the collapse was prompted by the processes associated with tectonic activity, in addition to magmatism (eruption). The eastern rim of the collapse structure, characterized by a steep slope and aligned with the surrounding lineaments, may further indicate that tectonic forces contributed to the collapse mechanisms observed in the western part of Nieuwerkerk Volcano.

5.2. Subsurface Condition Beneath Nieuwerkerk Volcano

Our magnetic models consistently suggest that magnetic bodies or layers might be distributed at shallow (<2 km) depths under the central part of the Nieuwerkerk edifice, especially under Profile AA* and the eastern parts of Profile BB–DD. However, the magnetic body distribution in the western and deeper parts of the central Nieuwerkerk edifice remains ambiguous. A better understanding of the variation in the magnetization of Nieuwerkerk rocks is required to constrain the distribution of these bodies.

Despite current limitations in data and modeling, the available evidence indicates that the shallow zones of the central Nieuwerkerk edifice are more strongly magnetized than its deeper counterparts. The more magnetic shallow parts of Nieuwerkerk can be unambiguously associated with volcanic cover, whose sample was dredged ∼6 km S of the E end of Profile DD* (Sample 220A from Honthaas et al. [20]) and dated at ∼8 Ma [20]. The sampling site is situated near a region exhibiting a positive magnetic anomaly, which is detected at a depth of about 1 km from the surface (Figure 9 and Figure 10). A comparable condition is identified in the collapse region/structure, located in the eastern segment of Profile CC*, characterized by a positive magnetic anomaly that extends beyond 1.5 km in depth (Figure 9). Bathymetric data indicate the presence of an ancient lava ridge or lava flow in this area (Figure 4).

The less magnetic, deeper part, however, is more enigmatic. This may be attributed to an updomed Nieuwerkerk volcanic basement (as suggested in, e.g., Bischoff et al. [53]), hydrothermal alteration and demagnetization within the edifice (cf. Paoletti et al. [54]), or the presence of a shallow active magma body (as proposed in Alawiyah et al. [44]). Biasi et al. [45] constructed magnetic models of four volcanic systems, i.e., Mount Saint Helens (1980), Axial Seamount (2015–2020), Kīlauea (2018), and Bárðarbunga (2014). They found that the low (negative) magnetic anomaly is associated with an active magma body; cryptodome; and hydrothermal zone. All of them impact the negative magnetic anomaly if they are closer to the surface. This result is relevant to our study, where the maximum negative anomaly in the center part of Nieuwerkerk Volcano (line BB* in Figure 2) is correlated with the DM body, which is at the surface (Figure 9). Meanwhile, the surficial projection of the DM body is delineated with a high ASA anomaly.

It is also possible that the negative magnetic anomaly detected at the central part of Nieuwerkerk is part of a larger Nieuwerkerk magnetic dipole, whose full range is yet to be observed. If this is the case, then the deeper Nieuwerkerk demagnetized zone of our models is a feature that stems from incomplete anomaly coverage and may not correspond to an actual geological object. If observed, the larger Nieuwerkerk magnetic dipole might be sufficiently modeled by a deep magnetic body associated with past Nieuwerkerk volcanic activities surrounded by a non-magnetic medium.

The possible presence of an active magma chamber beneath Nieuwerkerk Volcano is particularly intriguing, especially considering the findings of [20], which indicate that volcanic activity beneath the Lucipara and NEC Ridge has not been present for approximately 3 Ma due to the cessation of subduction. But that phenomenon is still able to occur, like the volcanism at Isla San Esteban and Volcan Las Tres Virgenes, which were still active until 9.5 to 12.5 m.y. after subduction stopped [55]. In addition, water breaking occurred in 1925 and 1927 around the location of Nieuwerkerk Volcano; this is thought to have been caused by the volcanic activity of that volcano [14].

5.3. Preliminary Hazard Assessment

Our newest findings affirm that Nieuwerkerk is classified as a volcano, with the potential for continued activity in the future. This insight is particularly significant given that Nieuwerkerk was previously regarded as a coral reef [14]. The consequences of these results indicate the potential for hazards in Nieuwerkerk Volcano, including volcanic eruptions and landslides or collapses. Collectively, the results above clarify the current morphology of Nieuwerkerk Volcano and provide new insight into its potential for slope failure hazards, as submarine landslides on such steep flanks can generate local tsunamis, particularly in the eastern part, which is characterized by intensive faults and the possibility of active magma. Coupled with the shallow summit (~300 m bsl), even moderate eruptive or mass-wasting events may pose risks to nearby islands.

The active tectonic dynamics within the Banda Sea can also induce the volcanic activities mentioned above. The occurrence of multiple earthquakes from 2014 to 2019 in this region indicates a high level of tectonic activity [56,57]. This phenomenon is comparable to the large landslide in the southern Tyrrhenian Sea, where earthquake-induced shaking is considered the most plausible external trigger [58]. Furthermore, as noted by [59], earthquakes are widely regarded as the most common triggering mechanism for submarine landslides. Additionally, some submarine volcanoes that have a history of collapse and regrowth could potentially collapse again in the future (e.g., Krakatau, Indonesia, and Kikai, Japan) [60,61,62]. Moreover, the dormant period of this volcano is very long, with the last activity predicted in 1927. The accumulation of large energy heightens the risk of catastrophic eruption in the future, such as that in the 1883 catastrophic eruption of Krakatau Volcano [60,61,63], despite the occurrence of an active magma chamber being confirmed further with additional data.

5.4. Recomendations

To further investigate the evidence supporting the preliminary results mentioned above, a continuing expedition is suggested to fill in many blanks and provide context for future studies, with the aim of better predicting comprehensive results. Future studies should be directed toward achieving better geophysical data coverage, which may be accompanied by ROV-based surveys, rock sampling, and water characterization (CTD). Ongoing efforts to integrate those methods could better resolve internal stratigraphy, determine whether active magmatic or hydrothermal processes continue at depth, ensure there is a potential collapse zone in the western part, address whether this feature formed from a single catastrophic event or through multiple smaller collapses, and confirm the potential for landslides around Nieuwerkerk Volcano.

The geophysical data have to be more comprehensive, and not only include magnetic data. We suggest adding another type of data, i.e., gravity, seismic, and SBP. The magnetic and gravity data have to cover the entire area of Nieuwerkerk Volcano, especially the blank area based on the last survey. Meanwhile, the seismic and SBP data can be collected along certain lines that cover only some key areas, e.g., around the summit, landslide, and depositional area, as well as the potential collapse zone. In addition, the petrophysical and geochemistry analysis of samples can also support future magnetic and gravity modeling for Nieuwerkerk, as well as the interpretation of seismic and SBP data. The samples from the remaining volcanic edifice and depression area (caldera-like/collapse structure) in the western part, around the current eruption center, and the ridge in the most eastern part must be collected to confirm our preliminary results. To achieve good visualization, coupled ROV surveys would provide a significant amount of supplementary data beyond what can be obtained from bathymetric analysis alone [64]. Meanwhile, it is imperative to perform water sampling and characterization in regions identified as potential eruption centers or hydrothermal areas using CTD instruments. Volcanic activities are known to alter water characteristics, such as elevating temperatures, changing conductivity, and altering chemistry, as evidenced by recent occurrences of active submarine eruptions off the coast of Mayotte [65].

Moreover, it is advisable to take into account the timing and the specific type of vessel utilized for cruising. The Banda Sea, recognized as one of the Indonesian seas with the most significant wave heights and water currents, presents challenges in this regard. An analysis conducted by [66] over the period from 2000 to 2010 indicates that the average wave height in the Banda Sea ranges between 1.5 and 3 m, with peaks in June and July that can reach up to 5 m in certain circumstances [67]. These conditions adversely impacted the effectiveness of our expedition in June to July 2022, as several activities could not be carried out, such as assessing water properties using a Conductivity–Temperature–Depth (CTD) device, as well as deploying the grab sampler for rock sampling and using the remotely operated vehicle (ROV) in the water. Additionally, our vessel experienced a brief blackout triggered by weather conditions. For future expeditions, we recommend scheduling operations in March to April, when wave and current conditions tend to be more favorable [66,67]. The selection of vessel type is also critical; employing a scientific research vessel with a robust emphasis on advanced scientific instrumentation tailored to deep-sea research across diverse domains, including oceanography, geophysics, and geology, is suggested.

6. Conclusions

Through the analysis of new bathymetry and marine geomagnetic data collected in 2022, this study enhances our understanding of the characteristics of Nieuwerkerk Volcano’s edifice and subsurface conditions. The findings can be summarized as follows:

• The main Nieuwerkerk Volcano edifice extends approximately 80 km in a NE–SW direction, with a maximum width of around 30 km. The volcano exhibits a total relief of about 3460 m, with shallower depths of 350 m below sea level (bsl). The observed discrepancy of over 2000 m between our research and the measurements recorded during Snellius’ expedition in 1930 (which indicated a shallower depth of 2300 m bsl) likely indicates the limitations inherent in the accuracy of the single-beam and lead-line surveys conducted in the 1930s.
• The western part of Nieuwerkerk Volcano consists of what is possibly a crater-like depression or a collapsed sector, or a large-scale sector failure that removed the original summit, measuring between 10 and 12 km in diameter. The traces of lineaments are also revealed within the central part of the depression with a NE–SW orientation. It is proposed that collapse processes could potentially be initiated by a combination of volcanism (eruption) and tectonic factors (including the presence of weak zones or faults).
• The morphology of the eastern part exhibits greater complexity. There are twin peaks, thought to be the most recent eruption centers, that reach depths of 350 and 380 m bsl and are situated approximately 7 to 7.5 km apart. The surrounding slope of these twin peaks ranges from 30° to 45°. This suggests that the latest activities have occurred mainly in the eastern part. Additionally, numerous lineaments are observed in the eastern part, predominantly oriented in a NE–SW direction, with the main fault extending over 20 km.
• The morphological characteristics of Nieuwerkerk Volcano were influenced by the interplay of volcanic activity, tectonic, and erosive–depositional processes. The landscape features include channels, scarps, debris accumulations, and steep slopes, especially in the eastern region, that are associated with geological lineaments such as faults. The largest potential landslide scarp threat measures 12 km in width and 13 km in length. Additionally, tectonic factors, specifically the presence of potential weak zone or faults, played a significant role in the location of the eruption centers.
• The shallow parts (<2 km) of the central Nieuwerkerk edifice are more magnetized than its deeper parts, which can be linked to volcanic deposits, such as lava flows. In contrast, the deeper parts, which exhibit lower magnetization, may indicate the presence of an active magma source. Furthermore, this could also be related to an updomed Nieuwerkerk Volcanic basement or a hydrothermally altered and demagnetized part of the Nieuwerkerk edifice.
• The initial evidence presented herein indicates that Nieuwerkerk Volcano represents a considerable geological hazard, including volcanic eruptions and mass wasting processes (landslides/slope failures/collapses), particularly in the eastern part, which may initiate volcano tsunamis. However, to achieve a more robust interpretation, we suggest conducting a continuing expedition to fill in many gaps and to further investigate the evidence supporting the preliminary results.
• The integration of additional geophysical data, including gravity, seismic, and SBP surveys, must be performed in the future to obtain more robust results. The magnetic survey should be re-conducted to cover the blank areas based on the last survey. These geophysical datasets can be accompanied by ROV-based surveys, rock sampling, and water characterization (CTD) for a more comprehensive study. Furthermore, meticulous planning regarding expedition timing and vessel selection is very important to ensure efficient and successful data acquisition in the Banda Sea.