The Evolution of Anak Krakatoa Based on Landsat Imagery

Sartika, Yohanes Fridolin Hestrio, Kiki Winda Veronica and Galdita Aruba Chulafak

Aeronautics and Space Research Organization, National Research and Innovation Agency (BRIN), Jakarta, Indonesia

Keywords:

Anak Krakatoa, NDWI, Landsat, Image Processing.

Abstract:

The Anak Krakatoa is a small volcano located in the Sunda Strait, speciﬁcally between the islands of Java

and Sumatra, Indonesia. The Anak Krakatoa is currently still an active volcano. This mountain has erupted

many times and has undergone various changes in its height, shape, and area, with the last eruption in July

2022. This study shows the changes in Anak Krakatoa, especially its shape and size, based on Landsat remote

sensing data from 1990 to 2022. Landsat imageries used in this study are Landsat 5,7, and 8. To obtain

information on the site of Anak Krakatoa, it begins by separating the sea and land around the mountain using

the Normalized Difference Water Index (NDWI). After the land is detected, then the area is calculated. The

volcano reached its minimum area in 1988 and its maximum area in 2022.

1 INTRODUCTION

Anak Krakatoa is a volcano located in the Sunda

Strait between the islands of Java and Sumatra, In-

donesia (Giachetti et al., 2012). The Krakatoa vol-

cano erupted violently in 1883 and caused a tsunami

to run 41 m and caused up to 36000 fatalities (Mu-

taqin et al., 2019), and destroyed 295 settlements

around the Sunda Strait (Bruins et al., 2008). The

eruption also formed a caldera with a depth of 200

to 280 meters below sea level (Umbgrove, 1926).

Krakatoa Island initially consisted of several moun-

tains, namely Mount Rakata, Mount Danan, and

Mount Perboewatan, wherein the incident in 1883,

Mount Danan and Perboewatan erupted (Umbgrove,

1926; Whittaker et al., 1989). After Krakatoa erupted

in 1883, a new volcano appeared in 1927, later re-

ferred to as the Anak Krakatoa (Gardner et al., 2013).

Since Anak Krakatoa ﬁrst appeared, it has been an

active volcano. Anak Krakatoa has been growing and

developing through volcanic activity. It emerged from

the sea in the same location as the original Kraka-

toa volcano. The emergence of Anak Krakatoa is

believed to have resulted from the same geological

processes that led to the formation of the original

Krakatoa volcano, namely the subduction of the Indo-

Australian Plate beneath the Eurasian Plate. The on-

going eruptions and lava ﬂows from Anak Krakatoa

contribute to the island’s growth. The lava cools and

solidiﬁes, creating new land that expands the size of

the island. Over time, this can signiﬁcantly grow the

island’s area. However, it’s important to note that vol-

canic activity can also lead to sudden and destructive

events, such as explosive eruptions, landslides, and

tsunamis. Anak Krakatoa erupted 80 times through

effusive and explosive eruptions from 1927 until 2006

(Igan, 2006) with Surtseyan and Strombolian eruption

types (Ginting et al., 2021). In 2018 Anak Kraka-

toa erupted again and was the ﬁrst eruption after the

1883 eruption, which also resulted in a tsunami (Grilli

et al., 2019). This tsunami also caused damage and

loss of life in the coastal areas of Lampung and Ban-

ten, Indonesia (Grilli et al., 2019). From a volcano

under the sea, Anak Krakatoa grew to about 300 m

above sea level before erupting in 2018 (Ye et al.,

2020). One of the remote sensing data that can be

used in observing the earth’s urface, and the data is

continuous, is Landsat. Landsat was launched in 1972

(Chen et al., 2020) and entered the ninth series in

2022 (Gross et al., 2022). However, not all regions

were recorded in its early mission due to its limita-

tions. For the Anak Krakatoa area, the data that can

be used is data from 1988. Observations and research

on Anak Krakatoa have certainly been very much

done, such as about the tsunami that caused it (Grilli

et al., 2019; Ye et al., 2020), the landscape (Ginting

et al., 2021), its growth (Igan, 2006), the vegetation

(Whittaker et al., 1989), the coral reef (Ferdiansyah

et al., 2019), and many more. This study was car-

ried out to observe the extensive development of the

volcanic island of Krakatoa. To obtain information

on the site of Anak Krakatoa, it begins by separating

Sartika, ., Hestrio, Y., Veronica, K. and Chulafak, G.

The Evolution of Anak Krakatoa Based on Landsat Imagery.

DOI: 10.5220/0012442500003848

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 3rd International Conference on Advanced Information Scientiﬁc Development (ICAISD 2023), pages 60-64

ISBN: 978-989-758-678-1

the sea and land around the mountain using the Nor-

malized Difference Water Index (NDWI). Then, use

the Otsu method to monitor the Anak Krakatoa area.

This method can assist in the automatic tracking of

Anak Krakatoa in the future without human interven-

tion. Usually, monitoring is done using visual inter-

pretation.

2 MATERIALS AND

METHODOLOGY

2.1 Location

The research location is on the Anak Krakatoa Vol-

cano Island in the Sunda Strait, Indonesia, as shown

in Fig. 1. Mount Anak Krakatoa itself is administra-

tively located in the South Lampung Regency, Lam-

pung Province, Indonesia (Suwarsono et al., 2019).

Figure 1: The location of the study area, with the upper

inset showing the location of Sunda Strait in Indonesia and

the lower inset showing the location of Anak Krakatoa in

Sunda Strait.

2.2 Methodology

Figure 2 shows the ﬂowchart of the method used for

the study.

Figure 2: Flowchart.

The method can be divided into several parts: data

selection, calculation of the NDWI, Delineation, and

Compute the area of Anak Krakatoa.

• Data Selection

The data used is Landsat data which consists

of several Landsat series, namely Landsat-5,

Landsat-7, and Landsat-8. These three types of

Landsat imagery have a spatial resolution of 30

meters but have a different numbers of bands. The

data selected is cloud and smoke-free, which can

show a clear boundary between Anak Krakatoa Is-

land and the ocean. The data consists of 5-year

intervals from 1988 to 2018 and 1-year intervals

from 2018 to 2022. The reason for choosing 5-

year data intervals is because the eruptions be-

tween these ranges were not too signiﬁcant until

2018, while the 1-year data is to see more spe-

ciﬁc changes per year that occurred after the ma-

jor eruption in 2018. The data used is shown in

Table 1.

Table 1: List of dates and types of Landsat satellites used.

Date Landsat

Type

Date Landsat

Type

1988-03-04 Landsat-5 2018-12-20 Landsat-8

1993-09-26 Landsat-5 2019-11-05 Landsat-8

1998-12-13 Landsat-5 2020-04-29 Landsat-8

2003-05-09 Landsat-7 2021-08-06 Landsat-8

2008-05-30 Landsat-5 2022-07-08 Landsat-8

2013-09-01 Landsat-8

• NDWI

Normalize Difference Water Index (NDWI) is a

method that enhances the water’s presence while

eliminating the presence of vegetation and soil in

remote sensing data (McFeeters, 1996). NDWI

utilizes the NIR channel, which is strongly ab-

sorbed by water and strongly reﬂected by vege-

tation and soil (McFeeters, 1996). The formula of

NDWI can be seen in equation (1).

NDWI =

GREEN − NIR

GREEN + NIR

(1)

For Landsat-5 and 7, GREEN and NIR are located

in Band 2 and 4, respectively (Chander et al.,

2004; Markham et al., 2004). While in Landsat-

8, the GREEN and NIR are located in Band 3

and 5, respectively (Wang et al., 2017). Figure

3 shows the comparison between the data in RGB

and NDWI.

• Delineation

The delineation is done using manual rather than

automatic delineation or thresholding techniques.

The manual delineation is used because the condi-

tion of Anak Krakatoa itself is pretty active, so in

addition to considering cloud conditions, it must

also consider smoke conditions. In NDWI, it can

eliminate soil and vegetation presence but also

The Evolution of Anak Krakatoa Based on Landsat Imagery

Figure 3: The comparison between the RGB (left) and

NDWI (right) shows that the soil has a darker area than the

water.

eliminate clouds and smoke so that when auto-

matic delineation is carried out, clouds and smoke

will also be detected as land (Choi et al., 2022).

Because of this, it is necessary to select data en-

tirely free of smoke and clouds if using an auto-

mated method, especially the one that intersects

with the coast. Indeed, this will signiﬁcantly limit

the data that can be used. This deﬁciency due to

cloud and smoke can be reduced by using SAR

data such as Sentinel-1, but the data availability

is not as much as Landsat data, which Sentinel-

1 was only available in 2014 (Wegn

uller et al.,

2016).

3 RESULTS AND DISCUSSION

Changes in the Anak Krakatoa area from 1988 to

2018 as shown in Fig. 4. As can be observed from

1988, the lava ﬂow was more directed to the north

with a signiﬁcant addition of area from 1988 to 1993.

From 1998, the additional area was more inclined to-

wards the west and south of the island, although not

as wide as the change in previous years.

Figure 4: The change in the area of Anak Krakatoa from

1988 to 2018.

Figure 5 shows the changes that occurred before

and after the eruption in 2018. Here, it can be ob-

served that a region vanishes in the western portion

of the island while there is an expansion of the region

in the eastern part. The increase in the area in the

east region could be caused by ocean currents carry-

ing eruptive material in that direction.

Figure 5: The change of Anak Krakatoa before and after the

eruption on December 2018. The increase and loss of area

have happened in several locations.

Meanwhile, Fig.6 illustrates the shift in the Anak

Krakatoa region between 2019 and 2022. The west is

where the changes are more pronounced. Naturally,

the lava ﬂow entering the region and causing changes

cannot be separated. In other areas, not too much

change has happened. If there are changes such as

abrasion and accretion in other areas of Anak Kraka-

toa, it is most likely due to the inﬂuence of ocean cur-

rents.

Figure 6: The change of Anak Krakatoa from 2019 to 2022.

The changes in the area that occurred in Anak

Krakatoa can be seen in Table 2.

Table 2: The area of Anak Krakatoa.

Year Area (ha) Year Area (ha)

1988 251.727 2019 325.936

1993 264.571 2020 328.266

1998 293.334 2021 322.320

2003 295.039 2022 333.550

2008 296.609

2013 300.691

2018 306.829

It is clear that between 1988 and 2018, the area

expanded to 55.102 ha, and from 2019 and 2022, the

area increased to 7.16 ha. The most signiﬁcant in-

ICAISD 2023 - International Conference on Advanced Information Scientiﬁc Development

crease in area is from 2021 to 2022 at 11.23 ha. Al-

though the increase in area from 1988 to 1993 was

greater (12,844 ha), it occurred for ﬁve years or had

an average rise of 2.59 ha/year. The fact is that Anak

Krakatoa’s condition has increased yearly in the area

even though there was an eruption in 2018. However,

there is also a decrease in the area from 2020 to 2021,

probably due to soil or rock conditions that are still

unstable and easily carried away by ocean currents.

The growth of the detected vegetation area from 1988

to 2018 also experienced an increase in area. The veg-

etation in Anak Krakatau is on the east side. After the

2018 eruption until 2022, no vegetation has been seen

through Landsat imagery on Anak Krakatoa. Figure

7 shows some of the vegetation conditions of Anak

Krakatoa through true composite color. Table 3 is an

estimate of the area of vegetated areas in Anak Kraka-

toa.

Table 3: Estimated area of vegetation in Anak Krakatoa.

Year Area (ha) Year Area (ha)

1988 23.388 2019 0

1993 36.232 2020 0

1998 52.657 2021 0

2003 57.484 2022 0

2008 61.033

2013 75.951

2018 76.091

Compared with its elevation, Anak Krakatoa loses

more in elevation than in area. In (Saputra et al.,

2021), it was stated that Anak Krakatoa’s height was

338 m to 110 m after the eruption. The condition

of the area certainly, in contrast, has increased from

around 306,829 ha to approximately 325,936 ha.

Figure 7: 1988 and 1998.

Figure 8. Vegetation condition (green area) of

Anak Krakatoa that shows the expansion until the

eruption in 2018.

Figure 8: 2008 and 2022.

4 CONCLUSION

The condition of the Anak Krakatoa continues to de-

velop from year to year in terms of its area. Despite

the eruption, the volcanic island’s area increased from

around 306.829 ha to approximately 325.936 ha while

its height decreased. During observations using Land-

sat data, the area of Anak Krakatoa had a minimum

size of 251,727 ha in 1988 (the earliest data used) and

a maximum area of 333,550 ha in 2022 (the latest data

that was used). The most signiﬁcant increase in the

area occurred from 2021 to 2022, with an additional

area of 11.23 ha. For further research, a study will be

carried out using multi-sensor remote sensing data to

monitor Anak Krakatoa’s development.

ACKNOWLEDGEMENT

The authors would like to thank the Research Center

for Remote Sensing for providing the data used in this

study and the anonymous reviewers for their valuable

comments to improve this article.

REFERENCES

Bruins, H., MacGillivray, J., Synolakis, C., Benjamini,

C., Keller, J., Kisch, H., Kl

ugel, A., and Plicht,

J. (2008). Geoarchaeological tsunami deposits at

palaikastro (crete) and the late minoan ia eruption of

santorini. J. Archaeol. Sci, 35:191–212.

Chander, G., Helder, D., Markham, B., Dewald, J., Kaita,

E., Thome, K., Micijevic, E., Ruggles, T., The, A.,

and Mapper (2004). L.-t.: Landsat-5 tm reﬂective-

band absolute radiometric calibration. IEEE Trans.

Geosci. Remote Sens, 42:2747–2760.

Chen, F., Fan, Q., Lou, S., Yang, L., Wang, C., Claverie, M.,

Wang, C., Junior, J., Goncalves, W., and Li, J. (2020).

Characterization of mss channel reﬂectance and de-

rived spectral indices for building consistent landsat

1-5 data record. IEEE Trans. Geosci. Remote Sens,

58:8967–8984.

Choi, Y.-J., Ban, H.-J., Han, H.-J., and Hong (2022). S.:

A maritime cloud-detection method using visible and

The Evolution of Anak Krakatoa Based on Landsat Imagery

near-infrared bands over the yellow sea and bohai sea.

Remote Sens, 14:793.

Ferdiansyah, M., Widiastuti, E., Ismail, T., and Susanto

(2019). G.n.: Inventarization of coral reefs in the wa-

ters of rakata island, krakatau islands. AQUASAINS,

8:775.

Gardner, M., Troll, V., Gamble, J., Gertisser, R., Hart, G.,

Ellam, R., Harris, C., and Wolff (2013). J.a.: Crustal

differentiation processes at krakatau volcano. Indone-

sia. J. Petrol, 54:149–182.

Giachetti, T., Paris, R., Kelfoun, K., and Ontowirjo (2012).

B.: Tsunami hazard related to a ﬂank collapse of anak

krakatau volcano, sunda strait, indonesia. Geol. Soc.

London, Spec. Publ, 361:79–90.

Ginting, F., Gusnidar, N., M., R., Minasny, B., and Fiantis,

D. (2021). Changes in anak krakatau landscape after

december 2018 eruption. IOP Conf. Ser. Earth Envi-

ron. Sci, 708:012088.

Grilli, S., Tappin, D., Carey, S., Watt, S., Ward, S., Grilli,

A., Engwell, S., Zhang, C., Kirby, J., Schambach,

L., and Muin, M. (2019). Modelling of the tsunami

from the december 22, 2018 lateral collapse of anak

krakatau volcano in the sunda straits, indonesia. Sci.

Rep, 9:11946.

Gross, G., Helder, D., Begeman, C., Leigh, L., Kaewmanee,

M., and Shah (2022). R.: Initial cross-calibration of

landsat 8 and landsat 9 using the simultaneous under-

ﬂy event. Remote Sens, 14:2418.

Igan, S. (2006). Pertumbuhan gunung api anak krakatau

setelah letusan katastroﬁs 1883. Indonesia. J. Geosci,

1:143–153.

Markham, B., Thome, K., Barsi, J., Kaita, E., Helder, D.,

and Barker, J. (2004). Scaramuzza, p.l.: Landsat-

7 etm+ onorbit reﬂective-band radiometric stability

and absolute calibration. IEEE Trans. Geosci. Remote

Sens, 42:2810–2820.

McFeeters, S. (1996). The use of the normalized difference

water index (ndwi) in the delineation of open water

features. Int. J. Remote Sens, 17:1425–1432.

Mutaqin, B., Lavigne, F., Hadmoko, D., and Ngalawani

(2019). M.n.: Volcanic eruption-induced tsunami in

indonesia: A review. iop conf. ser.earth environ. Sci,

256:012023.

Saputra, I., Armijon, A., and Fadly (2021). R.: Anali-

sis perubahan topograﬁ gunung anak krakatau pasca

erupsi tanggal 22 desember 2018 menggunakan data

foto udara dan demnas. J. Geod. Geomatics, 1:43–55.

Suwarsono, N., Prasasti, I., Nugroho, J., Sitorus, J., and

Triyono (2019). D.: Detecting the lava ﬂow deposits

from 2018 anak krakatau eruption using data fusion

landsat-8 optic and sentinel-1 sar. Int. J. Remote Sens.

Earth Sci, 15:157.

Umbgrove, J. (1926). The ﬁrst days of the new subma-

rine volcano near krakatoa. Leidse Geol. Meded,

2:325–328.

Wang, Q., Blackburn, G., Onojeghuo, A., Dash, J., Zhou,

L., Zhang, Y., and Atkinson (2017). P.m.: Fusion

of landsat 8 oli and sentinel-2 msi data. IEEE Trans.

Geosci. Remote Sens, 55:3885–3899.

Wegn

uller, U., Werner, C., Strozzi, T., Wiesmann, A.,

Frey, O., and Santoro, M. (2016). Sentinel-1 sup-

port in the gamma software. Procedia Comput. Sci,

100:1305–1312.

Whittaker, R., Bush, M., and Richards, K. (1989).

Plant recolonization and vegetation succession on

the krakatau islands, indonesia. Ecol. Monogr,

59:59–123.

Ye, L., Kanamori, H., Rivera, L., Lay, T., Zhou, Y., Sianipar,

D., and Satake, K. (2020). The 22 december 2018

tsunami from ﬂank collapse of anak krakatau volcano

during eruption. Sci. Adv, 6:1–10.

ICAISD 2023 - International Conference on Advanced Information Scientiﬁc Development