Next Article in Journal
Numerical Simulation Analysis and Research on Drag Reduction and Stability Enhancement Effect of Head Swing Based on Biomimetic Flexible Variants
Previous Article in Journal
Late Glacial and Holocene Paleoenvironmental Reconstruction of the Submerged Karst Basin Pirovac Bay on the Eastern Adriatic Coast
Previous Article in Special Issue
Coastal Protection for Tsunamis
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
JMSE
Volume 13
Issue 1
10.3390/jmse13010178
jmse-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Tracing the 2018 Sulawesi Earthquake and Tsunami’s Impact on Palu, Indonesia: A Remote Sensing Analysis

by
Youshuang Hu
1,
Aggeliki Barberopoulou
2,* and
Magaly Koch
3
1
Department of Geography, Sustainability, Community, and Urban Studies, University of Connecticut, Storrs, CT 06269, USA
2
Prometheus Space Technologies 2, Kifissias Av. & D. Gounari 95, 15124 Marousi, Greece
3
Department of Earth & Environment, Boston University, Boston, MA 02215, USA
*
Author to whom correspondence should be addressed.
J. Mar. Sci. Eng. 2025, 13(1), 178; https://doi.org/10.3390/jmse13010178
Submission received: 2 December 2024 / Revised: 11 January 2025 / Accepted: 15 January 2025 / Published: 19 January 2025
(This article belongs to the Special Issue Coastal Disaster Assessment and Response)
Browse Figures
Versions Notes

Abstract

:
The 2018 Sulawesi Earthquake and Tsunami serves as a backdrop for this work, which employs simple and straightforward remote sensing techniques to determine the extent of the destruction and indirectly evaluate the region’s vulnerability to such catastrophic events. Documenting damage from tsunamis is only meaningful shortly after the disaster has occurred because governmental agencies clean up debris and start the recovery process within a few hours after the destruction has occurred, deeming impact estimates unreliable. Sentinel-2 and Maxar WorldView-3 satellite images were used to calculate well-known environmental indices to delineate the tsunami-affected areas in Palu, Indonesia. The use of NDVI, NDSI, and NDWI indices has allowed for a quantifiable measure of the changes in vegetation, soil moisture, and water bodies, providing a clear demarcation of the tsunami’s impact on land cover. The final tsunami inundation map indicates that the areas most affected by the tsunami are found in the urban center, low-lying regions, and along the coast. This work charts the aftermath of one of Indonesia’s recent tsunamis but may also lay the groundwork for an easy, handy, and low-cost approach to quickly identify tsunami-affected zones. While previous studies have used high-resolution remote sensing methods such as LiDAR or SAR, our study emphasizes accessibility and simplicity, making it more feasible for resource-constrained regions or rapid disaster response. The scientific novelty lies in the integration of widely used environmental indices (dNDVI, dNDWI, and dNDSI) with threshold-based Decision Tree classification to delineate tsunami-affected areas. Unlike many studies that rely on advanced or proprietary tools, we demonstrate that comparable results can be achieved with cost-effective open-source data and straightforward methodologies. Additionally, we address the challenge of differentiating tsunami impacts from other phenomena (et, liquefaction) through index-based thresholds and propose a framework that is adaptable to other vulnerable coastal regions.

1. Introduction

Tsunamis are a result of the sudden displacement of the water column, triggered by events such as earthquakes, volcanic eruptions, submarine landslides, or atmospheric disturbances (e.g., [1,2,3,4,5,6,7]). Tsunamis can be highly destructive through inundation, erosion, strong currents in ports and harbors, and scouring caused by strong forces generated during the receding of waters, leading to loss of life and property damage (e.g., [8,9,10,11,12,13,14,15,16]). On many occasions, tsunamis caused more damage than the earthquakes that triggered them (e.g., Chile 1960, Alaska 1964, Tohoku 2011, Banda Aceh, 2004).
Indonesia is vulnerable to earthquakes, tsunamis, and volcanic eruptions due to its geographic location at the junction of three tectonic plates (the Indo-Australian Plate, the Eurasian Plate, and the Pacific Plate) [17,18,19]. Because of this tectonic setting, the presence of numerous underwater volcanoes combined with a densely populated coastal zone means that a large part of the country’s population is exposed to a number of coastal hazards including tsunamis as we have observed in the recent past (e.g., the 2004 Banda Aceh tsunami; [17,18,19,20,21]). According to the global tsunami database, a total of 77 tsunamis have been documented in Indonesia since 1608 [22]. These tsunamis have claimed the lives of more than 368,000 people in the country [22]. Among these, the catastrophic 2004 Indian Ocean tsunami alone killed 167,540 people in Indonesia and more than 220,000 people in 14 countries in Asia and Africa (Table 1) [23].
More recently, an M7.5 earthquake struck Central Sulawesi island, Indonesia, on 28 September 2018 at 6:03 p.m. local time (10:03 UTC). The earthquake caused severe damage to buildings and infrastructure and was responsible for more than three thousand deaths and many more people injured. The impact of this event is attributed to ground shaking, major liquefaction, landslides, and a tsunami. This was a large strike-slip faulting event at a shallow depth that generated a tsunami that caused damage through flooding, debris, ground erosion, and scouring [24,25,26,27].
Remote sensing, the technique of obtaining information about objects from a distance, has become a crucial tool in disaster management. Regarding tsunamis, satellite imagery has primarily been used for post-event damage assessment or as a source of byproducts (e.g., DEMs, friction maps) that serve as inputs to numerical models. Pioneering studies, like those following the 2004 Indian Ocean tsunami, utilized satellite imagery to map affected coastlines and assess the extent of inundation (e.g., [28,29,30,31,32,33]). These early applications relied mainly on optical imagery.
Recent technological advancements have broadened the scope of remote sensing in tsunami science. High-resolution satellite imagery, LiDAR (Light Detection and Ranging), and radar technologies have enabled more detailed and accurate assessments of tsunami inundation and impact. For instance, Chiroiu et al. [30] demonstrated the use of high-resolution satellite data for rapid damage assessment in coastal zones. Studies like those by Tang et al. [29] explored the use of satellite altimetry data to detect sea-level anomalies indicative of tsunamis. Similarly, SAR (Synthetic Aperture Radar) data, due to their ability to penetrate cloud cover and capture surface roughness, have been explored in detecting tsunami wavefronts and forecasting their impact [31]. Most of these studies have focused on the use of satellite imagery to estimate post-tsunami damage such as the extent of inundation and structural damage and evaluate the impact on natural resources (e.g., [32,33,34,35,36,37,38]). For example, Gokon et al. [39] highlighted the usefulness of remote sensing in understanding the extent of damage to aid effective recovery and rehabilitation efforts.
It has previously been verified that it is possible to identify tsunami-inundated areas by observing changes in spectral indices. Specifically, Yamazaki et al. [33] carried out a field survey in southern Thailand to examine areas severely affected by the Banda Aceh 2004 Boxing Day tsunami using GPS cameras and a handheld spectrometer. Three conditions of vegetation were measured: healthy (replanted), seriously affected (yellow), and dead. There were clear differences in the reflectance curves among these three conditions, with healthy plants showing a rapid increase in reflectance between the visible red (R) and near-infrared (NIR) bands. This characteristic was reduced in the unhealthy plants and was absent in the dead plants [33]. These notable changes in vegetation reflectance, offer an effective way to evaluate the impact of tsunamis on coastal vegetation.
Remote sensing techniques have also been employed to study the environmental impacts of tsunamis, such as changes in land cover, coastal erosion, and sedimentation. Research by Yamazaki et al. [33] using multispectral imagery provided insights into the long-term ecological effects of tsunamis on coastal environments. The integration of remote sensing data with Geographic Information System (GIS) platforms and modeling techniques may offer significant advancements to traditional tsunami hazard studies. This integration allows for more comprehensive spatial analysis and modeling of tsunami risks and impacts (e.g., [39]). Studies by Taubenböck et al. [40] illustrate how combining satellite data with GIS models can enhance tsunami hazard mapping and risk assessment.
Despite significant progress, challenges remain in remote sensing applications in tsunami science. These include limitations in spatial and temporal resolution, data availability, and the need for rapid data processing during emergency situations. Future research is likely to focus on integrating real-time data processing, improving the accuracy of early warning systems, and enhancing the resolution of satellite data [41]. Remote sensing can become an indispensable tool in tsunami analysis, offering valuable data for early warning, impact assessment, and recovery planning. As technology advances, its role in understanding and mitigating tsunami risks is likely to grow, providing more precise and timely information to safeguard vulnerable coastal communities.
Studies have shown that the financial impact of natural disasters is significant and rising (e.g., [42,43,44,45,46]). Additionally, investments in disaster risk mitigation may be economically beneficial; research suggests that every dollar spent on mitigation can save a substantial amount in post-disaster rebuilding efforts [42,43,44,45,46,47]. The costs associated with natural disaster management in Palu, which include research, disaster preparedness, and other aspects, have been substantial. In the aftermath of the 2018 earthquake and tsunami, the region faced significant challenges. While specific financial figures for research and disaster preparedness activities in Palu are not readily available, the overall impact of losses was estimated to be IDR 2.89 trillion (1 billion IDR approximately equals USD 64,000), with damage costs amounting to IDR 15.58 trillion due to the disaster [48].
This study aims to use easy, handy, and low-cost Remote Sensing techniques to quickly identify tsunami-affected zones in Palu. This is to enhance future disaster readiness by accurately mapping and assessing the extent of areas affected by the 2018 Sulawesi Earthquake and Tsunami in Palu. This should be a low-cost methodology using satellite imagery from open sources that can be simple and quick with the potential to be implemented in an early warning system or post-disaster for disaster management.

2. Data and Methods

2.1. Study Area

This study was conducted in the city of Palu, located in Central Sulawesi Province, Indonesia (Figure 1). Palu, the capital and largest city of the province spans an area of 395 km2, situated between 0°39′ S to 0°56′ S latitude and 119°45′ E to 120°1′ E longitude (Palu, 2022). According to the 2020 Indonesian census (2021), the city has a population of 373,218, making it the third-most populous city on the island after Makassar and Manado. Palu is situated on the Palu-Koro Fault and is frequently struck by earthquakes, such as the 2018 Sulawesi earthquake [17,24,25]. The local landscape includes a narrow valley surrounded by steep mountains, with the Palu River running through the city and coastal lowlands prone to flooding and tsunamis [25]. Settlement patterns are concentrated along riverbanks, coastal zones, and valley floors, where rapid urbanization and informal housing have increased the population’s exposure to natural hazards [25]. According to Indonesia’s National Disaster Management Agency, the earthquake-triggered secondary natural disasters, including soil liquefaction, landslides, and tsunamis, significantly affected Palu. For example, the 2018 earthquake caused one of the largest soil liquefaction-induced mass landslides in the world (e.g., [24,25,26,27]).

2.2. Data

Our primary data source is Sentinel-2 imagery, provided by the European Space Agency (ESA). The Sentinel-2 mission comprises two identical satellites, Sentinel-2A and Sentinel-2B, launched in 2015 and 2017, respectively. These satellites are equipped with a high-resolution multispectral imaging instrument capable of capturing Earth’s surface at a spatial resolution of up to 10 m. The instrument collects data across 13 spectral bands, spanning from visible light to near infrared. This range supports a variety of applications, including land use and land cover mapping, vegetation monitoring, and disaster response. For this study, Sentinel-2 images from 27 September 2018 (pre-tsunami and 2 October 2018 (post-tsunami) were analyzed to assess the impact of the tsunami, as shown in Table 2.
An additional remote sensing dataset used is Maxar WorldView-3 imagery (Table 3). Maxar Technologies, a leading space technology and intelligence company, operates a suite of commercial imaging satellites capable of capturing high-resolution images of Earth for diverse applications, including environmental monitoring, urban planning, and disaster management. Through its Open Data Program, Maxar provides free satellite imagery for disaster response, making high-quality geospatial data available to support organizations during major crisis events. This program has been activated for various disasters, including earthquakes and hurricanes, offering essential data to aid in response, management, and relief efforts.

2.3. Remote Sensing Workflow

The techniques presented here are meant to capture the imprint of the 2018 tsunami in the post-disaster landscape. The analysis captures all these through a combination of common tools such as indices, color composites, classification, and threshold analysis. For example, the Normalized Difference Vegetation Index (NDVI) is commonly used to estimate vegetation density, but in a post-tsunami environment, NDVI should be able to determine the extent of vegetation erosion and increased soil moisture. The hypothesis is that the density of vegetation in the study area will decrease following the tsunami, as indicated by a decline in NDVI values, while water and soil indices are expected to increase. NDVI is supplemented by the Normalized Difference Water Index (NDWI) and the Normalized Difference Soil Index (NDSI) as indicators of water presence and soil changes, therefore acting as a proxy of the tsunami impact on the ground.
The workflow outlined in Figure 2 describes the main steps in the process of using remote sensing data to identify tsunami-impacted areas. The key steps include:
  • Acquisition of Sentinel-2 imagery for pre- and post-tsunami periods;
  • Data preprocessing, which involves atmospheric correction, radiometric calibration, and resampling to convert raw Digital Number (DN) values into reflectance values;
  • Index calculations for NDVI, NDSI, and NDWI from reflectance data to distinguish various land and water features;
  • Visual analysis with Maxar Satellite Imagery to identify affected and unaffected regions by the tsunami;
  • Comparative analysis of indices before and after the tsunami to detect changes in vegetation, soil, and water;
  • Threshold determination using cumulative frequency distribution to classify impacted areas based on index variations;
  • Output generation, producing a map that clearly marks tsunami-affected areas, serving as a valuable tool for disaster response and assessment.

2.4. Index Formulas

This study utilizes three commonly used vegetation indices: NDVI, NDWI, and NDSI. The Normalized Difference Vegetation Index (NDVI) quantifies vegetation by measuring the contrast between near-infrared light, which vegetation reflects strongly, and red light, which it absorbs. This method is useful for assessing both the amount and health of plant growth, as higher NDVI values indicate denser vegetation. Additionally, NDVI effectively suppresses open-water features [49]. NDVI is calculated as follows:
N D V I = I n d e x N I R , R E D = ( N I R − R E D ) ( N I R + R E D )
For Sentinel-2, the vegetation index is calculated as
N D V I = I n d e x ( B 8 , B 4 ) = ( B 8 − B 4 ) ( B 8 + B 4 )
where Red and NIR are the reflectance of red and near-infrared bands, respectively. The NDVI formula takes these differences into account and generates a value ranging from −1 to +1, where higher values indicate healthier and more abundant vegetation. If the NDVI equation were inverted and the green band were employed, the results would likewise be inverted, with vegetation being suppressed and open-water features enhanced [49].
The equation for Normalized Difference Water Index (NDWI) is calculated as follows:
N D W I = I n d e x   ( G R E E N ,   N I R ) = ( G R E E N − N I R ) ( G R E E N + N I R )
For Sentinel-2, the water index is calculated as
N D W I = I n d e x   B 3 ,   B 8 = ( B 3 − B 8 ) ( B 3 + B 8 )  
where Green and NIR refer to the reflectance values of the green and near-infrared bands, respectively. The result of this equation highlights water features with positive values, while soil and terrestrial vegetation are represented by zero or negative values. The equation for the Normalized Difference Soil Index (NDSI) [50] is as follows:
N D S I = I n d e x   R E D ,   B L U E = ( R E D − B L U E ) ( R E D + B L U E )
For Sentinel-2, the soil index is calculated as
N D S I = I n d e x   B 4 , B 2 = ( B 4 − B 2 ) ( B 4 + B 2 )
In addition to the well-known indices above, other common remote sensing tools were employed to identify tsunami-impacted areas and the inundation extent such as true color composites and false color composites. False color composites use non-visible bands of the electromagnetic spectrum and map them to visible colors to highlight features that are not easily discerned in true-color imagery.

3. Results

3.1. Color Composites

A false color composite using satellite images taken after the Palu tsunami event on 2 October 2018 is used here. Specifically, Short-Wave Infrared (SWIR), Visible Near Infrared (VNIR), and Red bands were assigned to the RGB (Red, Green, Blue) color channels, respectively. The strength of the false color composites is that land cover types such as vegetation or water can appear in various colors depending on the band combination used. The goal is to enhance visualization of the different land cover types for analysis or processing. In the FCC used here, urban areas or bare ground appear in shades of purple, pink, grey, or brown while water appears in dark blue (Figure 3).

3.2. Computation of Indices

Three different vegetation indices have been used for our analysis that are further discussed next.

3.2.1. Normalized Difference Vegetation Index

NDVI values for pre- and post-tsunami imagery were calculated (Figure 4) and reclassified based on the NDVI classification criteria developed by Al-Doski et al. [51]. NDVI is used as a measure of vegetation health, with higher values indicating denser healthier vegetation. Image values were classified into three categories (Figure 4) that represent values from −1 to 0 (water, snow, and clouds), values from 0 to 0.2 (barren land, built-up areas, and rock), and vegetation from 0.2 to 1. The comparison shows little change in the water, snow, and cloud class, while there is an apparent increase in barren land along the coastline especially WNW-facing coastlines, suggesting vegetation loss and soil exposure due to the tsunami. This significant alteration within a five-day interval suggests the impact of a coastal hazard (tsunami) on the land surface.
Basic statistics calculated from the pre- and after-NDVI values (Table 4) also support the NDVI maps (Figure 4). Before the tsunami, the imagery showed NDVI values ranging from a minimum of −0.32 to a maximum of 0.95, with an average of 0.37. After the tsunami, both the maximum and mean NDVI values declined: the minimum NDVI fell to −0.42, the maximum to 0.88, and the mean to 0.34. These changes indicate an overall reduction in vegetation health immediately following the tsunami.

3.2.2. Normalized Difference Water Index

The Normalized Difference Water Index (NDWI) maps (Figure 5) illustrate the area of interest before (left) and after (right) the Palu tsunami. This index is a valuable tool for identifying water features, with values ranging from −1 to 1. Typically, non-water areas are represented within the range of −1 to 0, while water bodies correspond to values from 0 to 1. The pre-tsunami map, dated 27 September 2018, captures water bodies prior to the tsunami. Any changes (i.e., increase) to water bodies post-tsunami in the 2 October 2018 map should be attributed to the inundation associated with the tsunami. Together, these maps provide compelling visual evidence of the alterations in water extent following the disaster.
Table 5 presents basic statistics summarizing the NDWI values recorded before and after the Palu tsunami. The mean NDWI value post-tsunami is slightly higher than the pre-tsunami value, suggesting a minor increase in water content; however, the difference is minimal. For a more comprehensive understanding of these changes, Figure 5 offers a visual representation of the distribution and extent of water features, effectively highlighting any changes to land cover due to the tsunami.

3.2.3. Normalized Difference Soil Index

Complementing the previous 2 indices, NDSI may help in identifying and interpreting changes on the ground surface due to the tsunami. As shown in Figure 6, the NDSI shows soil surface changes, and in this instance, a clear increase in soil features (yellow) is visible in the post-tsunami image (Figure 6 right). Such a change following the tsunami may be due to the removal of vegetation or other cover, revealing more of the soil surface underneath or the deposition of sediments on top of other land cover types.
Table 6 provides a summary of statistical data for NDSI from imagery captured before and after the Palu tsunami. The NDSI is used to differentiate between soil and non-soil surface features. The post-tsunami imagery shows an increase in both the maximum NDSI value and the mean NDSI value, indicating a greater presence or exposure of soil features after the event. This can be attributed to the disturbance caused by the tsunami, which may have cleared away vegetation or other surface materials, exposing the soil underneath, or the deposition of sediments manifesting as soil. NDSI shows the largest changes due to the tsunami than any other index used here. The standard deviation in the post-tsunami data is slightly lower than in the pre-tsunami data, suggesting less variability in soil exposure after the tsunami.

3.3. Threshold Analysis

To obtain more precise threshold values, we concentrated on a smaller research area located in the northeast part of Palu, which was highly affected by the tsunami and had a clear boundary between the impacted and unaffected areas. By narrowing down the study area, we can conduct a more rigorous analysis to gain a better understanding of the disaster’s extent and related characteristics. This approach allows us to provide more precise data for threshold analysis and mapping of the tsunami-affected regions, thereby increasing the accuracy of the results.
The NDVI, NDSI, and NDWI were calculated based on the formulas presented in Section 2.4. Typically, higher NDVI values indicate greater vegetation coverage, while higher NDSI and NDWI values indicate more soil moisture and water bodies, respectively. The most prominent changes in NDVI, NDSI, and NDWI values in tsunami-affected areas are shown in Figure 7. As seen in Figure 7, NDVI decreases (darker shade), and NDSI and NDWI increase (lighter shade) in the tsunami-affected areas. These observations are based on the effects of the tsunami, which caused vegetation to wash away or become dead/weakened, exposed the soil, and increased soil moisture. Differences in the indices between areas affected and unaffected by the tsunami are readily apparent, and these disparities can serve as reliable indicators of tsunami inundation.
Research indicates two methodologies for setting threshold values to assess tsunami impact: using solely post-tsunami imagery or comparing both pre- and post-tsunami images [50]. The first approach is viable when pre-event images are unavailable. However, juxtaposing images from before and after an event typically yields a more accurate assessment [49,50]. This study employed the comparative method for spatial analysis of the tsunami’s effects and to establish reliable threshold values.
The image taken on 27 September 2018, serves as a pre-tsunami depiction, while the one captured on 2 October 2018, serves as a post-tsunami representation (Figure 7). To further analyze the data, the distribution of index differences between affected and unaffected areas was also calculated. The resulting cumulative frequency distribution of the difference for NDVI, NDSI, and NDWI in both affected and unaffected areas is displayed in Figure 8. By identifying the maximum difference between their cumulative frequency distributions, it is feasible to establish a threshold value that can effectively distinguish the two classes (unaffected vs affected). By utilizing this approach, the calculated thresholds for NDVI, NDSI, and NDWI were found to be −0.2, 0, and 0.04, respectively. In summary, based on pre-tsunami and post-tsunami images, pixels with an NDVI (−0.2) value below the calculated threshold, and NDSI (0) and NDWI values (0.04) exceeding the determined thresholds, are classified as tsunami-affected areas.
Using the thresholds for the three indices (NDVI, NDSI, and NDWI), we applied these values to our research area, as illustrated in Figure 9. More specifically, as described above, pixels with an NDVI value below the calculated threshold (<−0.2) and NDSI and NDWI values exceeding the determined thresholds (>0 and >0.04) are classified as tsunami-affected areas. Illustrated in Figure 10, the entire city was impacted by the tsunami, with the tsunami-affected areas primarily concentrated along the coastal regions of Palu. Furthermore, it is important to note that Balaroa City, located in the southwest part of Palu, was also affected by soil liquefaction after the earthquake and subsequently buried (Figure 9).
To derive the final tsunami-affected areas, a Decision Tree classification method was employed, integrating the dNDVI, dNDSI, and dNDWI values obtained from pre- and post-event indices. The Decision Tree is a rule-based classification approach that uses a hierarchical structure to assign each pixel to a specific class (e.g., tsunami-affected or unaffected) based on a set of conditions. In this study, thresholds for the indices were established using cumulative frequency distribution analysis. Pixels with values below the NDVI threshold (−0.2) and above the NDSI (0) and NDWI (0.04) thresholds were classified as tsunami-affected areas.
This method allowed the systematic combination of the three indices to discern changes in water, soil, and vegetation caused by the tsunami. By leveraging these indices in a hierarchical decision-making framework, the Decision Tree enhanced classification accuracy, particularly in complex environments where a single index might be insufficient. Figure 10 illustrates the “cumulative” classification result, with tsunami-impacted regions marked in red predominantly concentrated along the coastal areas. However, it is evident from Figure 10 that the method also captured impacts beyond tsunami inundation, such as liquefaction, landslides, and other secondary effects triggered by the earthquake. This limitation highlights the need for additional techniques or data to refine the classification further.

4. Discussion and Conclusions

This work focused on the application of freely available and low-cost remote sensing techniques, specifically leveraging Sentinel-2 and Maxar imagery, to quickly and effectively identify tsunami-affected areas. The remote sensing techniques employed in this study were designed to capture the impacts of tsunamis on the post-disaster landscape shortly after the event and before cleanup efforts by governmental agencies could alter the ground conditions. While previous studies have used high-resolution remote sensing methods such as LiDAR or SAR, our study emphasized accessibility and simplicity, making it more feasible for resource-constrained regions or rapid disaster response.
The results have successfully demonstrated the application of straightforward remote sensing techniques in mapping tsunami-affected areas in Palu, Indonesia, highlighting the capabilities of Sentinel-2 imagery in disaster assessment. The use of dNDVI, dNDSI, and dNDWI indices has allowed for a quantifiable measure of the changes in vegetation, soil moisture, and water bodies, providing a clear demarcation of the tsunami’s impact. This approach offers an understanding of disaster impacts, informing better preparedness and response strategies with easy-to-implement tools that are widely used.
The scientific novelty lies in the integration of widely used environmental indices (dNDVI, dNDWI, and dNDSI) with threshold-based Decision Tree classification to delineate tsunami-affected areas. Unlike many studies that rely on advanced or proprietary tools, we demonstrated that comparable results can be achieved with cost-effective open-source data and straightforward methodologies. Additionally, we addressed the challenge of differentiating tsunami impacts from other phenomena (e.g., liquefaction) through index-based thresholds and proposed a framework that is adaptable to other vulnerable coastal regions.
The final tsunami inundation map indicates that areas most affected by the tsunami are located in the urban center, low-lying regions, and along the coast. The tsunami inundation map appears to capture more than tsunami-affected areas such as liquefaction. Other methods (e.g., GIS) such as the Analytic Hierarchy Process (AHP) method, although somewhat subjective, can provide validation against the tsunami inundation map, confirming the importance of topography, slope, and land use as critical factors in tsunami inundation [52,53,54] but also in helping to differentiate between tsunamis and other areas impacted by different phenomena (liquefaction). This synthesis of various data types can provide a more nuanced understanding of vulnerability and impact.
The effective use of remote sensing in this study underscores their potential in disaster management. Policies should support the investment in and utilization of these technologies for better disaster preparedness, risk assessment, and recovery planning. The study’s findings can be used in urban planning that integrates resilience against natural disasters. Tools such as those used here can be implemented by international agencies that deal with disaster assessment and management such as Copernicus Emergency Management Services (https://emergency.copernicus.eu/) or local agencies for better planning and preparedness. The implementation characteristics depend on the needs of the agency.
While this study has provided significant insights, several limitations must be acknowledged. Firstly, the spatial resolution of remote sensing data, while sufficient for broad assessments, may not capture the finer details necessary for micro-scale planning. This limitation points to the need for higher-resolution data or supplementary ground truthing for more detailed analysis. Secondly, the remote sensing data for this study, with only a five-day interval between pre- and post-tsunami images, presents a unique dataset. Such timely data acquisition is not always feasible, as obtaining simultaneous pre- and post-event images is often not possible in many scenarios.
The impact captured in satellite images may not be easily differentiated from other secondary effects such as liquefaction, which coincided the Palu tsunami. One of the challenges of delineating the tsunami impact is the differentiation of the tsunami signature to other secondary or seasonal effects. In our case, we were able to overcome the contamination of results by seasonal vegetation changes and pre-existing environmental conditions because the remote sensing images used in this study were acquired within a time span of no more than five days before and after the tsunami, eliminating concerns about seasonal changes. Further improving tools used here in this direction would be highly beneficial. Additionally, the study’s focus on Palu may limit the generalizability of findings to other regions with different geographical and socio-economic contexts. Future research could address these constraints by incorporating a broader range of case studies and employing more sophisticated remote sensing technologies as they become available.

Author Contributions

Conceptualization Y.H., A.B. and M.K.; methodology, Y.H. and M.K.; software, Y.H.; validation, Y.H.; formal analysis, Y.H.; investigation, Y.H., A.B. and M.K.; resources, Y.H., A.B. and M.K.; data curation, Y.H. writing—original draft preparation, Y.H. and A.B.; writing—review and editing, Y.H., A.B. and M.K.; visualization, Y.H.; supervision, A.B. and M.K.; project administration, A.B. and M.K. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Institutional Review Board Statement

The current analysis did not constitute human subject research and was exempt from further IRB review.

Informed Consent Statement

Not applicable.

Data Availability Statement

All the data used in this study are available online.

Acknowledgments

The work was completed while at Tufts University.

Conflicts of Interest

Author Aggeliki Barberopoulou was employed by the company Prometheus Space Technologies 2. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

References

  1. Koshimura, S.; Moya, L.; Mas, E.; Bai, Y. Tsunami Damage Detection with Remote Sensing: A Review. Geosciences 2020, 10, 177. [Google Scholar] [CrossRef]
  2. Shuto, N.; Fujima, K. A Short History of Tsunami Research and Countermeasures in Japan. Proc. Jpn. Acad. Ser. B Phys. Biol. Sci. 2009, 85, 267–275. [Google Scholar] [CrossRef]
  3. Kanamori, H. Mechanism of Tsunami Earthquakes. Phys. Earth Planet. Inter. 1972, 6, 346–359. [Google Scholar] [CrossRef]
  4. Abe, K. Tsunami and Mechanism of Great Earthquakes. Phys. Earth Planet. Inter. 1973, 7, 193–203. [Google Scholar] [CrossRef]
  5. Röbke, B.R.; Vött, A. The Tsunami Phenomenon. Prog. Oceanogr. 2017, 159, 296–312. [Google Scholar] [CrossRef]
  6. Harbitz, C.B.; Løvholt, F.; Pedersen, G.; Masson, D.G. Mechanisms of Tsunami Generation by Submarine Landslides: A Short Review. Nor. J. Geol. 2006, 86, 255–264. [Google Scholar]
  7. Papadopoulos, G.A.; Gràcia, E.; Urgeles, R.; Sallares, V.; De Martini, P.M.; Pantosti, D.; González, M.; Yalciner, A.C.; Mascle, J.; Sakellariou, D.; et al. Historical and Pre-Historical Tsunamis in the Mediterranean and Its Connected Seas: Geological Signatures, Generation Mechanisms and Coastal Impacts. Mar. Geol. 2014, 354, 81–109. [Google Scholar] [CrossRef]
  8. Jaumé, S.C.; Sykes, L.R. Evolving Towards a Critical Point: A Review of Accelerating Seismic Moment/Energy Release Prior to Large and Great Earthquakes. In Seismicity Patterns, Their Statistical Significance and Physical Meaning; Wyss, M., Shimazaki, K., Ito, A., Eds.; Birkhäuser: Basel, Switzerland, 1999; pp. 279–305. [Google Scholar]
  9. van Rijsingen, E.; Lallemand, S.; Peyret, M.; Arcay, D.; Heuret, A.; Funiciello, F.; Corbi, F. How Subduction Interface Roughness Influences the Occurrence of Large Interplate Earthquakes. Geochem. Geophys. Geosyst. 2018, 19, 2342–2370. [Google Scholar] [CrossRef]
  10. Han, S.-C.; Sauber, J.; Luthcke, S. Regional Gravity Decrease After the 2010 Maule (Chile) Earthquake Indicates Large-Scale Mass Redistribution. Geophys. Res. Lett. 2010, 37, L23307. [Google Scholar] [CrossRef]
  11. Baba, T.; Cummins, P.R. Contiguous Rupture Areas of Two Nankai Trough Earthquakes Revealed by High-Resolution Tsunami Waveform Inversion. Geophys. Res. Lett. 2005, 32, L08305. [Google Scholar] [CrossRef]
  12. Thiel, C.C., Jr.; Zsutty, T.C. Earthquake Characteristics and Damage Statistics. Earthq. Spectra 1987, 3, 747–792. [Google Scholar] [CrossRef]
  13. Wesnousky, S.G. Displacement and Geometrical Characteristics of Earthquake Surface Ruptures: Issues and Implications for Seismic-Hazard Analysis and the Process of Earthquake Rupture. Bull. Seismol. Soc. Am. 2008, 98, 1609–1632. [Google Scholar] [CrossRef]
  14. Oth, A. On the Characteristics of Earthquake Stress Release Variations in Japan. Earth Planet. Sci. Lett. 2013, 377–378, 132–141. [Google Scholar] [CrossRef]
  15. Van Dorn, W.G. Some Tsunami Characteristics Deducible from Tide Records. J. Phys. Oceanogr. 1984, 14, 353–363. [Google Scholar] [CrossRef]
  16. Løvholt, F.; Pedersen, G.; Harbitz, C.B.; Glimsdal, S.; Kim, J. On the Characteristics of Landslide Tsunamis. Philos. Trans. R. Soc. A Math. Phys. Eng. Sci. 2015, 373, 20140376. [Google Scholar] [CrossRef]
  17. Reid, J.A.; Mooney, W.D. Tsunami Occurrence 1900–2020: A Global Review, with Examples from Indonesia. Pure Appl. Geophys. 2023, 180, 1549–1571. [Google Scholar] [CrossRef]
  18. Muhari, A.; Diposaptono, S.; Imamura, F. Toward an Integrated Tsunami Disaster Mitigation: Lessons Learned from Previous Tsunami Events in Indonesia. J. Nat. Disaster Sci. 2007, 29, 13–19. [Google Scholar] [CrossRef]
  19. Esteban, M.; Tsimopoulou, V.; Mikami, T.; Yun, N.Y.; Suppasri, A.; Shibayama, T. Recent Tsunamis Events and Preparedness: Development of Tsunami Awareness in Indonesia, Chile and Japan. Int. J. Disaster Risk Reduct. 2013, 5, 84–97. [Google Scholar] [CrossRef]
  20. Mutaqin, B.W.; Lavigne, F.; Hadmoko, D.S.; Ngalawani, M.N. Volcanic eruption-induced tsunami in Indonesia: A review. IOP Conf. Ser. Earth Environ. Sci. 2019, 256, 012023. [Google Scholar] [CrossRef]
  21. Supriatna, J. Konservasi Biodiversitas: Teori dan Praktik di Indonesia; Yayasan Pustaka Obor Indonesia: Jakarta, Indonesia, 2018. [Google Scholar]
  22. National Geophysical Data Center/World Data Service. NCEI/WDS Global Historical Tsunami Database. NOAA National Centers for Environmental Information. Available online: https://data.noaa.gov/metaview/page?xml=NOAA/NESDIS/NGDC/MGG/Hazards/iso/xml/G02151.xml&view=getDataView (accessed on 16 January 2024). [CrossRef]
  23. Intergovernmental Oceanographic Commission (IOC) of UNESCO. Tsunami Programme—Pacific. Available online: https://tsunami.ioc.unesco.org/en/pacific (accessed on 16 January 2024).
  24. Valkaniotis, S.; Ganas, A.; Tsironi, V.; Barberopoulou, A. A Preliminary Report on the M7.5 Palu Earthquake Co-Seismic Ruptures and Landslides Using Image Correlation Techniques on Optical Satellite Data; Zenodo: Genève, Switzerland, 2018. [Google Scholar] [CrossRef]
  25. Paulik, R.; Gusman, A.; Williams, J.H.; Pratama, G.M.; Lin, S.-L.; Prawirabhakti, A.; Sulendra, K.; Zachari, M.Y.; Fortuna, Z.E.D.; Layuk, N.B.P.; et al. Tsunami hazard and built environment damage observations from Palu City after the September 28, 2018 Sulawesi earthquake and tsunami. Pure Appl. Geophys. 2019, 176, 3305–3321. [Google Scholar] [CrossRef]
  26. Rahardjo, P.P. Study on the phenomena of liquefaction-induced massive landslides in 28 September 2018 Palu-Donggala earthquake. In Understanding and Reducing Landslide Disaster Risk: Volume 5 Catastrophic Landslides and Frontiers of Landslide Science, 5th ed.; Elsevier: London, UK, 2021; pp. 25–48. [Google Scholar]
  27. Omira, R.; Dogan, G.G.; Hidayat, R.; Husrin, S.; Prasetya, G.; Annunziato, A.; Proietti, C.; Probst, P.; Paparo, M.A.; Wronna, M.; et al. The September 28th, 2018, tsunami in Palu-Sulawesi, Indonesia: A post-event field survey. Pure Appl. Geophys. 2019, 176, 1379–1395. [Google Scholar] [CrossRef]
  28. Ramírez-Herrera, M.T.; Navarrete-Pacheco, J.A. Satellite data for a rapid assessment of tsunami inundation areas after the 2011 Tohoku tsunami. Pure Appl. Geophys. 2013, 170, 1067–1080. [Google Scholar] [CrossRef]
  29. Tang, L.; Titov, V.; Chamberlin, C. Assessment of potential tsunami impact for Southern California, USA using satellite altimetry data. Nat. Hazards 2016, 82, 359–370. [Google Scholar]
  30. Chiroiu, L.; André, C.; Rekacewicz, P. High-resolution satellite imagery in coastal disaster management: Assessing the 2004 Tsunami impact. Disaster Prevention and Management 2018, 27, 578–590. [Google Scholar]
  31. Liu, P.L.; Lynett, P.; Fernando, H. Observation of the 2011 Tohoku tsunami using satellite SAR data. Remote Sens. Environ. 2015, 159, 202–215. [Google Scholar]
  32. Sepúlveda, I.; Tozer, B.; Haase, J.S.; Liu, P.L.F.; Grigoriu, M. Modeling uncertainties of bathymetry predicted with satellite altimetry data and application to tsunami hazard assessments. J. Geophys. Res. Solid Earth 2020, 125, e2020JB019735. [Google Scholar] [CrossRef]
  33. Yamazaki, F.; Matsuoka, M.; Warnitchai, P.; Polngam, S.; Ghosh, S. Tsunami reconnaissance survey in Thailand using satellite images and GPS. Asian J. Geoinform. 2005, 5, 53–61. [Google Scholar]
  34. Paris, R.; Lavigne, F.; Wassmer, P.; Sartohadi, J. Coastal sedimentation associated with the December 26, 2004 tsunami in Lhok Nga, west Banda Aceh (Sumatra, Indonesia). Mar. Geol. 2007, 238, 93–106. [Google Scholar] [CrossRef]
  35. Lavigne, F.; Paris, R.; Grancher, D.; Wassmer, P.; Brunstein, D.; Vautier, F.; Flohic, F.; De Coster, B.; Gunawan, T.; Gomez, C.; et al. Reconstruction of tsunami inland propagation on December 26, 2004, in Banda Aceh, Indonesia, through field investigations. Pure Appl. Geophys. 2009, 166, 259–281. [Google Scholar] [CrossRef]
  36. Borrero, J.C. Field survey of northern Sumatra and Banda Aceh, Indonesia after the tsunami and earthquake of 26 December 2004. Seismol. Res. Lett. 2005, 76, 312–320. [Google Scholar] [CrossRef]
  37. Koshimura, S.; Gokon, H.; Fukuoka, T.; Hayashi, S. Remote sensing and GIS-based approach to identify the impact of the 2011 Tohoku Earthquake tsunami disaster. J. Jpn. Assoc. Earthq. Eng. 2012, 12, 50–62. [Google Scholar]
  38. Sambah, A.B.; Miura, F. Spatial data analysis and remote sensing for observing tsunami-inundated areas. Int. J. Remote Sens. 2016, 37, 2047–2065. [Google Scholar] [CrossRef]
  39. Gokon, H.; Koshimura, S.; Matsuoka, M. Remote sensing-based assessment of tsunami vulnerability and risk in coastal areas. Int. J. Disaster Risk Reduct. 2017, 22, 487–501. [Google Scholar]
  40. Taubenböck, H.; Post, J.; Kiefl, R.; Roth, A.; Ismail, F.; Strunz, G.; Dech, S. Risk and vulnerability assessment to tsunami hazard using very high resolution satellite data—The case study of Padang, Indonesia. Nat. Hazards Earth Syst. Sci. 2008, 8, 409–420. [Google Scholar] [CrossRef]
  41. Li, X. Real-Time High-Rate GNSS Techniques for Earthquake Monitoring and Early Warning. Remote Sens. 2015, 7, 12346–12364. [Google Scholar]
  42. National Oceanic and Atmospheric Administration. 2021 U.S. Billion-Dollar Weather and Climate Disasters: A Historical Perspective. Climate.gov. 2021. Available online: https://www.climate.gov/news-features/blogs/beyond-data/2021-us-billion-dollar-weather-and-climate-disasters-historical (accessed on 22 December 2024).
  43. Kreimer, A. Social and Economic Impacts of Natural Disasters. Int. Geol. Rev. 2001, 43, 401–405. [Google Scholar] [CrossRef]
  44. Keerthiratne, S.; Tol, R.S. Impact of Natural Disasters on Financial Development. Econ. Disasters Clim. Chang. 2017, 1, 33–54. [Google Scholar] [CrossRef]
  45. Chang, C.P.; Zhang, L.W. Do Natural Disasters Increase Financial Risks? An Empirical Analysis. Bull. Monet. Econ. Bank. 2020, 23, 61–86. [Google Scholar]
  46. Bhola, V.; Hertelendy, A.; Hart, A.; Adnan, S.B.; Ciottone, G. Escalating costs of billion-dollar disasters in the US: Climate change necessitates disaster risk reduction. J. Clim. Chang. Health 2023, 10, 100201. [Google Scholar] [CrossRef]
  47. Samad, M.A.; Ali, M.N.; Khairil, M. Indonesian Disaster Governance: Public Policy and Social Economic Impact. Elem. Educ. Online 2021, 20, 73–88. [Google Scholar]
  48. Damarjati, D. Management Earthquake and Tsunami Disaster in Palu, Indonesia. IJSSHR 2022, 5, 284–288. Available online: https://ijsshr.in/v5i1/38.php (accessed on 22 December 2024).
  49. McFeeters, S.K. The use of the Normalized Difference Water Index (NDWI) in the delineation of open water features. Int. J. Remote Sens. 1996, 17, 1425–1432. [Google Scholar] [CrossRef]
  50. Kouchi, K.I.; Yamazaki, F. Characteristics of tsunami-affected areas in moderate resolution satellite images. IEEE Trans. Geosci. Remote Sens. 2007, 45, 1650–1657. [Google Scholar] [CrossRef]
  51. Al-Doski, J.; Mansor, S.B.; Shafri, H.Z.M. NDVI differencing and post-classification to detect vegetation changes in Halabja City, Iraq. IOSR J. Appl. Geol. Geophys. 2013, 1, 1–10. [Google Scholar] [CrossRef]
  52. Banai, R. Fuzziness in geographic information systems: Contributions from the analytic hierarchy process. Int. J. Geogr. Inf. Syst. 1993, 7, 315–329. [Google Scholar] [CrossRef]
  53. Dyer, J.S. Remarks on the analytic hierarchy process. Manag. Sci. 1990, 36, 249–258. [Google Scholar] [CrossRef]
  54. Saaty, T.L. Decision making with the analytic hierarchy process. Int. J. Serv. Sci. 2008, 1, 83–98. [Google Scholar] [CrossRef]
Jmse 13 00178 g001
Figure 1. (Top) Relative location of Sulawesi Island and Palu City (red circle) in Indonesia. (Lower left) Study Area: Palu City (red). (Lower right) Palu City Administrative Division.
Figure 1. (Top) Relative location of Sulawesi Island and Palu City (red circle) in Indonesia. (Lower left) Study Area: Palu City (red). (Lower right) Palu City Administrative Division.
Jmse 13 00178 g001
Jmse 13 00178 g002
Figure 2. Remote Sensing Workflow. To derive the final tsunami-affected areas, a Decision Tree classification method was employed, integrating the dNDVI, dNDSI, and dNDWI values obtained from pre- and post-event indices.
Figure 2. Remote Sensing Workflow. To derive the final tsunami-affected areas, a Decision Tree classification method was employed, integrating the dNDVI, dNDSI, and dNDWI values obtained from pre- and post-event indices.
Jmse 13 00178 g002
Jmse 13 00178 g003
Figure 3. False color composite after the tsunami (2018/02/10 (SWIR, VNIR, and RED as RGB bands).
Figure 3. False color composite after the tsunami (2018/02/10 (SWIR, VNIR, and RED as RGB bands).
Jmse 13 00178 g003
Jmse 13 00178 g004
Figure 4. Reclassified NDVI for pre-tsunami (2018/09/27) (left) and post-tsunami imagery (2018/10/02) based on the NDVI classification criteria developed by Al-Doski et al. [51] (right).
Figure 4. Reclassified NDVI for pre-tsunami (2018/09/27) (left) and post-tsunami imagery (2018/10/02) based on the NDVI classification criteria developed by Al-Doski et al. [51] (right).
Jmse 13 00178 g004
Jmse 13 00178 g005
Figure 5. Reclassified NDWI for pre-tsunami (2018/09/27) (left) and post-tsunami imagery (2018/10/02) (right).
Figure 5. Reclassified NDWI for pre-tsunami (2018/09/27) (left) and post-tsunami imagery (2018/10/02) (right).
Jmse 13 00178 g005
Jmse 13 00178 g006
Figure 6. Reclassified NDSI for pre-tsunami (2018/09/27) (left) and post-tsunami imagery (2018/10/02) (right).
Figure 6. Reclassified NDSI for pre-tsunami (2018/09/27) (left) and post-tsunami imagery (2018/10/02) (right).
Jmse 13 00178 g006
Jmse 13 00178 g007
Figure 7. Computed NDVI, NDSI, and NDWI for pre-tsunami (2018/09/27) and post-tsunami (2018/10/02) for a small area of Palu. Dashed line delineates the area affected by the tsunami in the post-disaster images.
Figure 7. Computed NDVI, NDSI, and NDWI for pre-tsunami (2018/09/27) and post-tsunami (2018/10/02) for a small area of Palu. Dashed line delineates the area affected by the tsunami in the post-disaster images.
Jmse 13 00178 g007
Jmse 13 00178 g008
Figure 8. Cumulative frequency distribution of the differences between NDVI, NDSI, and NDWI.
Figure 8. Cumulative frequency distribution of the differences between NDVI, NDSI, and NDWI.
Jmse 13 00178 g008
Jmse 13 00178 g009
Figure 9. Mapping NDVI, NDSI, and NDWI based on the threshold values. Dark pixels indicate areas impacted by the tsunami.
Figure 9. Mapping NDVI, NDSI, and NDWI based on the threshold values. Dark pixels indicate areas impacted by the tsunami.
Jmse 13 00178 g009
Jmse 13 00178 g010
Figure 10. Final tsunami inundation map using a Decision Tree classification method, integrating the dNDVI, dNDSI, and dNDWI values obtained from pre- and post-event indices.
Figure 10. Final tsunami inundation map using a Decision Tree classification method, integrating the dNDVI, dNDSI, and dNDWI values obtained from pre- and post-event indices.
Jmse 13 00178 g010
Table 1. Summary of significant life losses (>100) in Indonesia from recent tsunamis [22].
Table 1. Summary of significant life losses (>100) in Indonesia from recent tsunamis [22].
NameDate (mm/dd/yr)SourceLocationEarthquake MagnitudeMaximum Water Height (m)Total Deaths
2004 Indian Ocean Tsunami12/24/2004EarthquakeSumatra9.150.9227,899
2005 Nisa-Simeulue Tsunami03/28/2005EarthquakeIndonesia8.64.21313
2006 Pangandaran Tsunami07/17/2006EarthquakeSouth of Java7.720.9802
2010 Mentawai Tsunami10/25/2010EarthquakeSumatra7.816.9431
2018 Sulawesi Tsunami09/20/2018Earthquake and VolcanoSulawesi7.510.734340
2018 Sunda strait Tsunami12/22/2018Volcano and LandslideKrakatau*85437
* The 2018 Sunda Strait tsunami was not directly caused by an earthquake, so it does not have a seismic magnitude.
Table 2. Sentinel-2 imagery description table.
Table 2. Sentinel-2 imagery description table.
Satellite and Mission IDProduct LevelSpatial ResolutionBand
Number		Production BaselineOrbit NumberAcquisition Sensing Time
Pre-tsunamiSentinel-2
S2B		Level-1C10 mBand 2
Band 3
Band 4
Band 8		N0206R10309/27/2018
Sentinel-2
S2B		Level-1C20 mBand 5
Band 6
Band 7
Band 8A
Band 11
Band 12		N0206R10309/27/2018
Sentinel-2
S2B		Level-1C60 mBand 1
Band 9
Band 10		N0206R10309/27/2018
Post-tsunamiSentinel-2
S2A		Level-1C10 mBand 2
Band 3
Band 4
Band 8		N0206R10310/02/2018
Sentinel-2
S2A		Level-1C20 mBand 5
Band 6
Band 7
Band 8A
Band 11
Band 12		N0206R10310/02/2018
Sentinel-2
S2A		Level-1C60 mBand 1
Band 9
Band 10		N0206R10310/02/2018
Table 3. Maxar WorldView imagery description table.
Table 3. Maxar WorldView imagery description table.
SatelliteBand NumberSpatial ResolutionAcquisition Sensing Time
Pre-tsunamiWorldView-31 Panchromatic band (450–800 nm)0.31 m08/17/2018
WorldView-38 Visible Near Infrared (VNIR) bands1.24 m at nadir08/17/2018
WorldView-38 Shortwave Infrared (SWIR) bands3.70 m at nadir08/17/2018
WorldView-312 CAVIS (Clouds, Aerosols, Vapors, Ice, and Snow) bands30 m at nadir08/17/2018
Post-tsunamiWorldView-31 Panchromatic band (450–800 nm)0.31m10/02/2018
WorldView-38 Visible Near Infrared (VNIR) bands1.24 m at nadir10/02/2018
WorldView-38 Shortwave Infrared (SWIR) bands3.70 m at nadir10/02/2018
WorldView-312 CAVIS (Clouds, Aerosols, Vapors, Ice, and Snow) bands30 m at nadir10/02/2018
Table 4. NDVI value calculation for pre- and post-tsunami imagery.
Table 4. NDVI value calculation for pre- and post-tsunami imagery.
Min MaxMeanStdDev
NDVI: Pre-tsunami Imagery
(2018/09/27)		−0.320.950.370.20
NDVI: Post-tsunami Imagery
(2018/10/02)		−0.420.880.340.19
Table 5. NDWI value calculation for pre- and post-tsunami imagery.
Table 5. NDWI value calculation for pre- and post-tsunami imagery.
Min MaxMeanStdDev
NDWI: Pre-tsunami Imagery
(2018/09/27)		−0.950.50−0.380.17
NDWI: Post-tsunami Imagery
(2018/10/02)		−0.100.54−0.370.16
Table 6. NDSI value summary for pre- and post-tsunami imagery.
Table 6. NDSI value summary for pre- and post-tsunami imagery.
MinMaxMeanStdDev
NDSI: Pre-tsunami Imagery
(2018/09/27)		−0.660.660.110.08
NDSI: Post-tsunami Imagery
(2018/10/02)		−0.570.100.140.08
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Hu, Y.; Barberopoulou, A.; Koch, M. Tracing the 2018 Sulawesi Earthquake and Tsunami’s Impact on Palu, Indonesia: A Remote Sensing Analysis. J. Mar. Sci. Eng. 2025, 13, 178. https://doi.org/10.3390/jmse13010178

AMA Style

Hu Y, Barberopoulou A, Koch M. Tracing the 2018 Sulawesi Earthquake and Tsunami’s Impact on Palu, Indonesia: A Remote Sensing Analysis. Journal of Marine Science and Engineering. 2025; 13(1):178. https://doi.org/10.3390/jmse13010178

Chicago/Turabian Style

Hu, Youshuang, Aggeliki Barberopoulou, and Magaly Koch. 2025. "Tracing the 2018 Sulawesi Earthquake and Tsunami’s Impact on Palu, Indonesia: A Remote Sensing Analysis" Journal of Marine Science and Engineering 13, no. 1: 178. https://doi.org/10.3390/jmse13010178

APA Style

Hu, Y., Barberopoulou, A., & Koch, M. (2025). Tracing the 2018 Sulawesi Earthquake and Tsunami’s Impact on Palu, Indonesia: A Remote Sensing Analysis. Journal of Marine Science and Engineering, 13(1), 178. https://doi.org/10.3390/jmse13010178

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop