1 Introduction

Indonesia is a country prone to natural disasters. It lies among the Pacific tectonic plates, the Eurasian and Indo-Australian, based on the scope of geography and geology. These conditions make Indonesia vulnerable to natural disasters such as earthquakes and volcanoes (Sutikno 2007). Indonesia has a history of volcanic eruptions since 1974, about 52 events, which have caused 18,271 deaths and economic losses of US$ 344 million [Centre for Research on the Epidemiology of Disasters (CRED) (2012)]. Merapi volcano is an active volcano of the type stratovolcano, located in Central Java province, with a height of about 2,965 m above sea level. It was formed as a result of the activity of the Indo-Australia plate zone and has caused volcanic activities along the central part of Java island. There have been at least 61 recorded eruptions before the year 2000 (Lavigne et al. 2000).

Merapi eruption events have caused various effects of loss, including environmental degradation, loss of life, property, and others. The impact of the eruption events has been recorded since 1822. Eruption events in 1900 destroyed five villages, killing more than 100 people. In 1920, 1932, 1961, 1996, 1973, 1974, 1975, 1976, 1994, 1995, 1996, and 1998, there has been more than 300 people killed and 500 houses damaged by the eruptions (Lavigne et al. 2000, 2008). The Indonesian National Agency for Disaster Management (BNPB) has noted the eruption event in 2010, which destroyed the settlement’s sector, infrastructure, telecommunications, and caused disruptions to electricity and fresh water supplies.

The need for the rapid mapping of volcanic hazards is necessary for emergency response activities. Volcanic hazards mapping can be used to estimate the impact of disaster loss assessment relatively precisely and efficiently, and to monitor the progress in emergency response and recovery. Remote sensing technology can be used to map the impact of disasters and monitor the progress of recovery and reconstruction. Several researches have been carried out for mapping the impact of the damages caused by volcanic eruptions. This research involved the use of high-resolution optical satellite imagery, as has been carried out by Wang et al. (2003), Fu et al. (2004), Tang et al. (2006), Song et al. (2006), Suzuki and Yamazaki (2008), and Guo et al. (2009). An active sensor, such as a synthetic aperture radar (SAR), was also utilized, as has been demonstrated by Matsuoka and Yamazaki (2000), Stramondo et al. (2006), Trianni and Gamba (2008), and Dong et al. (2010), and light detection and ranging (LIDAR) was also investigated, as researched by Dash et al. (2004).

SAR data can provide useful information for quick damage assessment. As an early stage when the field survey is impossible due to bad weather (cloudy and rainy) and field conditions after a serious disaster, SAR data can penetrate clouds and give the surface information clearly. However, optical remote sensing cannot show the surface condition when it is covered by clouds. SAR is strongly sensitive to the surface changes caused by damage. The damage detection from SAR images requires pre- and post-disaster acquisition (Dong et al. 2010). Change detection could also be applied to detect and assess other natural disasters, such as earthquakes, floods, mud-rock flows, and forest fires. Several other applications have been applied by Le Toan et al. (1997), Quegan et al. (2000), Nonaka et al. (2007), Tzeng et al. (2007), Wulder et al. (2008), Bayarsaikhan et al. (2009), and Gao and Liu (2010) for the monitoring and survey of marine resources, environmental changes and pollution, obtaining a dynamic picture of urban landuse growth, evaluation of plant growth, geographic database updates, etc. SAR data characteristics are poorer visual display, a lower resolution, and the data processing strategies are more complex when compared with the data of optical remote sensing. Utilization of the data is limited to a quick application for monitoring and disaster management (Dong et al. 2010). Geographic information system (GIS) technology can be used as a tool for modeling and assessing the hazards, as has been done by Kresch et al. (2002), Marfai and King (2008), and Ward et al. (2011). A scheme to map and model the volcano eruption and its effects, including landuse assessment based on the development of remote sensing ability and GIS, is proposed in this current research. Several researches have been conducted with different methods of approach used for the hazard mapping and modeling of volcanic eruptions. For hazard mapping, these include, among others, mapping tephra fallout risk (Barberi et al. 1990; Connor et al. 2001), hazard prediction for volcanic explosions (Saito et al. 2001), and the impact of pyroclastic density current (PDC) surges on building damage (Baxter et al. 2005). For modeling, these include, among others, lava flow hazard mapping and using computer simulations (Wadge et al. 1994), GIS application for lava flow estimation (Kauahikaua et al. 1995), mapping of pyroclastic flows using numerical modeling simulations (Rossano et al. 2004), hazard modeling of volcanic ash falls using the Monte Carlo method (Hurst and Smith 2004), GIS method for hazard mapping PDCs (Toyos et al. 2007), and approaches and modeling of lava flow hazards (Felpeto et al. 2007).

This research focuses on providing information related to the damaging effects caused by the 2010 eruption of Merapi volcano based on SAR data (Fig. 1). The objectives of this research are: (a) to generate a map of pyroclastic deposits based on activities pre- and post-volcano eruption of 2010 in the research area, (b) to investigate the impact of volcano eruption on the environment, and (c) to assess the impact of volcano eruption on landuse.

Fig. 1
figure 1

Research area at Merapi volcano, Central Java, Indonesia

Full size image

2 Methodology

In this research, the data processing and analysis were divided into three steps: (1) generate a map of pyroclastic deposits, (2) identification of the environment damage due to volcano eruption, (3) assessment of the volcano eruption effect on landuse. ALOS PALSAR remote sensing data pre-disaster (April 27, 2007) and post-disaster (November 5, 2010) together with the SAR system was used within this research for mapping the pyroclastic deposits. Environment damage in this study focused on the region affected by the distribution of pyroclastic deposits from Merapi eruption events; this could include: sedimentation, air pollution, volcanic ash, etc.. Topographic and geomorphologic maps are considered for profiling and field orientation. SPOT 4 satellite images were used in this research for updating landuse information from the topographic map. The result of the landuse updated data was used for assessment of the volcano eruption’s effect on landuse with the GIS raster environment. The types of spatial data used for this research are shown in Table 1 and detailed descriptions of the methodology steps undertaken in this study are given in the following sections.

Table 1 Types of spatial data used in this research
Full size table

2.1 Generate a map of pyroclastic deposits

The SAR images obtained in the same area at different times (pre- and post-disaster) were analyzed in order to discover the change of land cover (Huang et al. 2011). In this research, a SAR image change detection technique was used to generate a map of pyroclastic deposits. SAR images were obtained that showed a change of information of the ground objects (Le Toan et al. 1997; Quegan et al. 2000; Nonaka et al. 2007; Tzeng et al. 2007; Wulder et al. 2008; Bayarsaikhan et al. 2009; Gao and Liu 2010). The ALOS PALSAR data were used for this study. A map of pyroclastic deposits was constructed based on the differences in back-scattering intensity from the pre- and post-volcano eruption ALOS PALSAR data. To generate a map of pyroclastic deposits, several processes and stages of data processing were required. The ALOS PALSAR product used for this research has a pixel value in the form of a digital number. To obtain a back-scattering coefficient value sigma nought (σ0) at each pixel value of the data, conversion of the absolute radiometric calibration of the digital number value was required. The conversion results illustrated the value of the pixel digital number as a measure of the strength of the radar signal’s back-scattering from a distributed target, usually expressed in dB. More profitable SAR data were used for this study because the system can penetrate clouds, smoke, dust, and distribution of particle volcano material particles, whereas such capabilities are not possessed by remote sensing optical sensors. In a situation of post-eruption, where volcanic activity can cause the scattering of cloud, smoke, dust, and volcano material particles, radar data are very helpful for investigating how much spatial damage occurs due to a volcanic eruption.

2.1.1 SAR data radiometric calibration and processing methods

ALOS PALSAR level 1.5 products used for this research give the image pixel value [or digital number (DN)]. Absolute radiometric calibration was performed on these SAR data to derive the back-scattering coefficient sigma nought from the DN value. Sigma nought is the measure of the strength of a radar signal’s back-scattering from a distributed target, usually expressed in dB. The back-scattering coefficient sigma nought distributed over PALSAR level 1.5 products were derived from Eqs. (1) and (2) (Dong et al. 2010):

$$ \sigma_{i,j}^{0} = K \cdot {\text{DN}}_{i,j}^{2} \quad (i = 1,2, \ldots ,L;\,j = 1,2, \ldots ,M) $$
(1)

Equation (2) was used to obtain sigma nought in dB:

$$ \sigma_{i,j}^{0} ({\text{dB}}) = 10\log_{10} \left( {\sigma_{i,j}^{0} } \right) = 20\log_{10} ({\text{DN}}) + K({\text{dB}}) $$
(2)

where K is an absolute calibration constant, which is provided by the Japan Aerospace Exploration Agency (JAXA), DN i,j is a pixel intensity value at image line i and column j, \( \sigma_{i,j}^{0} ({\text{dB}}) \) is a back-scattering coefficient sigma nought at image line i and column j in dB, and L and M are the numbers of lines and columns in the product.

If there is change to a target’s surface roughness or geometric and structural characteristics, there will be a change in the back-scattering intensity recorded in the SAR image. Comparison between the pre- and post-disaster SAR intensity images will reveal the impact information caused by volcanic eruption. With the change in back-scattering intensity recorded at the time of pre- and post-eruption, the detection of change in the impact disaster can be quickly done. In this research, differential and false-color composition methods were used to highlight the pyroclastic deposits. The differential back-scattering intensity in images was calculated using Eq. (3) (Dong et al. 2010):

$$ d_{i,j} ({\text{dB}}) = \sigma_{{i,j\,{\text{post}}}}^{0} ({\text{dB}}) - \sigma_{{i,j\,{\text{pre}}}}^{0} ({\text{dB}})\quad (i = 1,2, \ldots ,L;\,j = 1,2, \ldots ,M) $$
(3)

where d i,j is a pixel value in dB of the differential back-scattering intensity of an image at image line i and column j, \( \sigma_{{i,j\,{\text{post}}}}^{0} \) is the sigma nought value of the post-disaster back-scattering intensity image at image line i and column j, \( \sigma_{{i,j\,{\text{pre}}}}^{0} \) is the sigma nought value of the pre-disaster back-scattering intensity image at image line i and column j, and L and M are the numbers of lines and columns in the product.

In this research, statistical analysis of covariance and correlation is used to show the trend of a linear relationship among two or more random variables. Two variables are used for correlation and covariance analysis, which are the x-variable as a distribution of pixel values of the back-scattering coefficients pre-volcano eruption and the y-variable as a distribution of pixel values of the back-scattering coefficients post-volcano eruption. For the two random variables x and y having means E{x} and E{y}, the covariance is defined in Eq. (4) (Weisstein 1999):

$$ {\text{Cov}}(x,y) = E\left\{ {\left[ {x - E(x)} \right]\left[ {y - E(y)} \right]} \right\} $$
(4)

The covariance calculation begins with pairs of x and y values. We used the differences from their mean values and multiplied these differences with each other. For instance, if for x1 and y1 this product is positive, then for that pair of data points, the values of x and y has varied with each other in the same direction from their means. If the product is negative, then they have varied in opposite directions. The larger the magnitude of the product, the stronger the strength in the relationship. The covariance is the mean value of this product, calculated using each pair of data points x1 and y1. If the covariance is zero, then the cases in which the product was positive were offset by those in which it was negative, and there is no linear relationship between the two random variables. Computationally, it is more efficient to use Eq. (5) to calculate the covariance (Weisstein 1999):

$$ {\text{Cov}}(x,y) = E\{ xy\} - E\{ x\} E\{ y\} $$
(5)

The value of the covariance is interpreted as follows: (a) positive covariance shows that higher than mean values of one variable tend to be paired with higher than mean values of the other variable, (b) negative covariance shows that higher than mean values of one variable tend to be paired with lower than mean values of the other variable, (c) zero covariance occurs if the two random variables are self-directed, resulting in the zero covariance. A covariance of zero does not necessarily mean that the variables are self-directed, as a non-linear relationship can exist that would still result in a covariance value of zero.

2.2 Identification of the environment damage due to volcano eruption

This research has applied method of cross profile and rapid field observation to investigate the impact of volcano eruption on the environment. Cross profiles in this study were used to describe the cross section of the vertical and horizontal position of a region. By looking at the vertical and horizontal position of an area, the condition of the region can be described well. This could be beneficial to helping in the analysis of the study. Comprehensive conditions of the region may include variations in geomorphology, landuse, topography, and other environmental physical parameters. This method is used by Marfai (2011) to investigate the impact of inundation on coastal environments and ecology. This method is best used to describe and represent the environmental impacts of disasters in a region. The topographic map and geomorphological and landuse units are factors to be considered for profiling and field orientation. Physical conditions as well as environmental damages due to volcanic eruption are the main focuses to be observed during the cross profiling and field work activity. In addition, geomorphological characteristics (landforms) and landuse have been used as a basis unit for the field observation. Topographic maps have also been used to delineate the research boundary and design field routes and mappings.

2.3 Assessment of the volcano eruption impact on landuse

Assessment of the impact of disasters can be done by overlaying the map of the disaster with the landuse map in order to determine the number of damages occurring, as has been done by Marfai and King (2008), Ward et al. (2011), and Marfai (2011). In this research, assessment of the impact of the 2010 volcano eruption on landuse requires a landuse map. Up-to-date information for landuse in the research area is required. This is because the existing landuse information is derived from topographic maps of the year 1998. The data used to update the landuse information are from the SPOT 4 satellite image of the year 2009. Satellite image technologies such as SPOT 4 are very useful for landuse assessment and mapping. The landuse map has been generated using a visual interpretation technique and field observation. Spectral analysis was used to determine the band on SPOT 4 images that have the most channels in response to objects on the earth’s surface. This will be useful to determine the composite color combinations most appropriate for refining the classification of objects. The best color combination results describing the composite object classification is used as the input and the combination for visual interpretation. A topographic map scale of 1:25,000 was obtained by BAKOSURTANAL in 1998 (medium-scale maps), while the SPOT 4 images had a spatial resolution of 20 m in 2009. The input levels on the scale of mapping of SPOT 4 images are some of the input data that can be used for medium-scale mapping (here, other data may be used, such as LANDSAT, ASTER, ALOS AVNIR, etc.). Landuse information in the 1998 topographic maps can be updated by the interpretation of data on the SPOT 4 images from 2009. Visual interpretation techniques on SPOT 4 images of the on-screen digital views are used to update the information on the topographic landuse map from 1998. Topographic maps in the vector data format can be displayed as an image/overlay on the SPOT image, which is used to simplify the process of updating information data on landuse in the study area. Information points of the building site topographic map and results from the field survey by BAKOSURTANAL updated in 2009 are used for this research to detail the landuse information that was affected by the volcano eruption, in particular, settlement landuse. To calculate the impact of the volcano eruption on landuse, an overlay superimposed method in a GIS raster environment on the landuse map, information points of building areas, and a map of pyroclastic deposits were applied.

3 Results and discussion

3.1 Map of pyroclastic deposits in the research area

The overall results of the volcano eruption event during the process of making maps for the research area are presented in Fig. 2. Data from pre-volcano eruption are presented in Fig. 2a, whereas post-volcano eruption data are presented in Fig. 2b. Differences in back-scattering intensity from the pre- and post-volcano eruption data are presented in Fig. 2c, and the results of the map of pyroclastic deposits are presented in Fig. 2d. The false-color composition images of the pre- and post-volcano eruption back-scattering intensity images of ALOS PALSAR are shown, where the pre-eruption image of ALOS PALSAR was assigned to the red channel, and the post-eruption image of ALOS PALSAR was assigned to the green and blue channels. The distribution of pyroclastic deposits from the Merapi eruption in 2010 can be clearly mapped. Differences in the values of back-scattering in the data pre-and post-disaster are helpful for identifying the land surface changes due to the volcanic eruption. Sediment deposits composed of pyroclastic material, i.e., volcanic ash and sand, covered the surface of the land in the southern area of the studied zone. This resulted in data on post-disaster appearing darker when compared with the data pre-disaster. Darkness of the land surface to the data post-disaster is caused by the back-scattering of objects away from the satellite sensor. The distribution of these materials can be mapped by the administrations led by districts Pakem, Cangkringan, Ngemplak, Depok, and Kalasan.

Fig. 2
figure 2

Location of the damage area caused by the 2010 eruption of Merapi revealed by ALOS PALSAR images. a Pre-volcano eruption value of back-scattering coefficient sigma nought image of PALSAR acquired on April 27, 2007. b Post-volcano eruption value of back-scattering coefficient sigma nought image of PALSAR acquired on November 05, 2010. c Differential value of back-scattering image of PALSAR between pre- and post-volcano eruption. d False-color composite RGB image pre- and post-volcano eruption. R pre-volcano eruption image of PALSAR acquired on April 27, 2007, G post-volcano eruption image of PALSAR acquired on November 05, 2010, B post-volcano eruption image of PALSAR acquired on November 05, 2010. Result of the false-color composite: red color is a damage area and material from the volcano eruption in 2010; black to grey to white is material not from the volcano eruption 2010

Full size image

To show the changes between the pre- and post-volcano eruption conditions, the pixels in each of the cross profile areas were then statistically analyzed by covariance and correlation methods. Statistical analysis of the covariance and correlation is used to show the trend of a linear relationship among two or more random variables. Two variables are used for the analysis of correlation and covariance: x as a distribution of pixel values of the back-scattering coefficients sigma nought pre-volcano eruption and y as a distribution of pixel values of the back-scattering coefficients sigma nought post-volcano eruption. The results of the statistical analysis of covariance and correlation calculations are shown in Fig. 3 and Table 2. Figure 3 shows the cross profile and distribution values of back-scattering coefficients sigma nought (dB) pre- and post-volcano eruption from cross profiles 1–5. Table 2 shown the statistical analysis of the covariance and correlation values of back-scattering coefficients sigma nought (dB) pre- and post-volcano eruption in the research area.

Fig. 3
figure 3

1 Cross profile location with background false-color composite image PALSAR data pre- and post-volcano eruption. 2 Distribution of values of back-scattering coefficients sigma nought dB in cross profiles 1–5 (1a–5a). Cross profiles 1–5 back-scattering coefficients sigma nought pre- and post-volcano eruption (1b–5b). The scatter plots and statistical analysis of covariance and correlation values of back-scattering coefficients sigma nought in profiles 1–5, with the x variable being pre-volcano eruption and the y variable being post-volcano eruption

Full size image
Table 2 Statistical analysis of the covariance and correlation values of back-scattering coefficients sigma nought (dB) pre- and post-volcano eruption in the research area
Full size table

The cross profile 1 has the highest covariance and correlation values among all the profiles studied. The means that, in the cross profile 1 area, the pixel values in the pre-volcano eruption has mostly the same values as in the post-volcano eruption condition. There were no significant changes that occurred in cross profile 1. However, the other cross profile areas showed some changes among the pre- and post-volcano eruption conditions based on the covariance and correlation values. The largest changes occurred in the cross profile 4 area, which showed the lowest covariance and correlation values among the cross profile areas, i.e., 0.84 and 0.22, respectively. The lower covariance and correlation values of two random sets of data indicated a less prominent relationship between those data. In this research, the statistical analysis and correlation covariance were not used to describe the extent of damage caused by volcano eruptions. They were, however, used to show a change in the values of back-scattering of ALOS PALSAR pixels in the data, which may show changes in the land surface due to volcano eruptions. The existence of covariance and low correlation values may reflect differences about the value of back-scattering pixel data on the condition of pre- and post-disaster that may show a change on the land surface. The higher correlation and covariance values, may illustrate that there is no change to the value of back-scattering and shows no signs of changing land surfaces due to the disaster. Changes and differences in the value of back-scattering in the data pre- and post disaster are used as input data to generate a map of pyroclastic deposits.

An illustration of the pyroclastic deposits due to the 2010 volcano eruption in the research area is presented in Fig. 4.

Fig. 4
figure 4

Distribution of pyroclastic deposits due to the 2010 volcano eruption in the research area

Full size image

The distribution of pyroclastic deposits accumulate and flow from the peak of Merapi to the south. Material flows into the three-way main rivers, that is: Gendol, Opak, and Kuning. The greatest impact of the flow of the three rivers is Gendol River. The length of the pyroclastic deposits is ±16 km to the south. Figure 4a in this research can be described as Zone I. This zone is the area closest to the peak of Merapi and is located at a distance of ±4 km. Material pyroclastic deposits are scattered on three main streams, that is: Gendol, Opak, and Kuning. The impact of the distribution of pyroclastic deposits in Zone I includes the villages of Glagaharjo, Kepuhharjo, Umbulharjo, and Balerante. Figure 4b of this research is called Zone II, with a distance of ±8 km from the peak of Merapi. Most of the pyroclastic material is spread via Gendol River. The impact of the distribution of material covers an area encompassing Wukirsari village. Figure 4c is Zone III, which has a distance of ±12 km from the peak of Merapi. The impact of the distribution of pyroclastic deposits appear to flow toward Gendol River in the village of Argomulyo.

3.2 Preliminary assessment of the impact of volcano eruption on the environment

Cross profiling has been designed for five transect routes representing entire research areas (Fig. 5). Figure 5 shows cross profiles 1–5 of the geomorphological map and landuse map in 2009. During the investigation, meticulous assessment and observation in each geomorphology and landuse unit was carried out. Information from local communities during the profiling and interview activity was very valuable in order to obtain a description of the environmental problems associated with volcano eruptions. Determination of the sample five transect routes in this study was carried out at random. However, there are some points to consider in making the samples, including the variations in slope, landuse, and geomorphological units. In addition, a transect route was also identified as the area of land surface changes due to the impact of the distribution of pyroclastic deposits.

Fig. 5
figure 5

Cross profiles 1–5 of the geomorphological map (1) and landuse map (2) in 2009

Full size image

Cross profile 1 (Fig. 6) reveals that the impact of volcano eruptions cause several environmental problems, such as effecting the distribution of volcanic ash and material eruption sedimentation. Landuse units consist of forest and material eruption lahars/lava. Impact events of the 2010 volcano eruption appear at a distance ±0.5 km from point A to ±3 km. Environmental problems and damage due to the volcano eruption include affected sedimentation of material eruption and forests being destroyed. The impact on fields of volcano eruption events can be seen in Fig. 7a, b. Conditions of avalanche of the cone of Merapi to the south resulted in material eruptions sliding downwards, as shown in Fig. 7a.

Fig. 6
figure 6

Cross profile 1

Full size image
Fig. 7
figure 7

Impact of volcano eruption in cross profile 1 (a, b). Impact of volcano eruption in cross profile 2 (c, d). Photo field work taken on July, 2011

Full size image

The impact of the volcano eruption area of cross profile 2 (Fig. 8) damaged paddy fields, settlement, bare land, and forests. The impact appeared at a distance of ±5 km from point A to ±3.75 km. Sedimentation of pyroclastic deposits caused environmental problems in this area. The impact on fields from volcano eruption event and sedimentation on river flow conditions around cross profile 2 was the result of a large amount of material being thrown out and burying the damaged residential areas by the eruption of the material presented in Fig. 7c, d.

Fig. 8
figure 8

Cross profile 2

Full size image

The impact of the 2010 Merapi eruption on cross profile 3 (Fig. 9) appeared at a distance of ±1.5 km from point A to ±3 km. Environmental problems that occur in this area include sedimentation, the spread of volcanic ash, and damage to paddy fields, settlements, and dry farming. The impact on fields from the volcano eruption event can be seen in Fig. 10a, b.

Fig. 9
figure 9

Cross profile 3

Full size image
Fig. 10
figure 10

Impact of volcano eruption in cross profile 3 (a, b). Impact of volcano eruption in cross profile 4 (c, d). Photo field work taken on July, 2011

Full size image

Cross profile 4 (Fig. 11) underwent environmental problems such as material eruption sedimentation, damage to several areas of paddy fields, settlements, and dry farming, and the spread of volcano ash. Impact events of the 2010 volcano eruption appeared at the distance ±1.75 km from point A to ±3 km. Building damage occurred in residential areas around the cross profile 4 area due to the material thrown out by the volcano eruptions and bursts of hot clouds (awan wedus gembel) shown in Fig. 10a–d.

Fig. 11
figure 11

Cross profile 4

Full size image

Environmental problems and damage due to volcano eruptions affected the distribution of volcanic ash, sedimentation of material eruption, and some paddy fields, settlements, dry farming were destroyed. The impact on fields of volcano eruption events and sabo dam conditions around area profile 5 (Fig. 12) functioned as a buffer for material flow eruption and the amount of sedimentation material deposited by rivers that are not depicted can be seen in Fig. 13a, b. Impact events of the 2010 volcano eruption appeared at the distance ±1.75 km from point A to ±2.5 km.

Fig. 12
figure 12

Cross profile 5

Full size image
Fig. 13
figure 13

Impact of volcano eruption in cross profile 5 (a, b). Photo field work taken on July, 2011

Full size image

3.3 Impact of volcano eruption on landuse and building areas

The result of calculations between the map of pyroclastic deposits overlayed with the landuse map for the research area are presented in Table 3. Table 3 shows landuse area damage effects of the volcano eruption in the research area while the results calculation between the map of pyroclastic deposits overlayed with the information point of the building site is presented in Table 4. Table 4 shows the building damage area effects of the volcano eruption in the research area.

Table 3 Damage impact on landuse area of the volcano eruption in the research area
Full size table
Table 4 Damage impact on building areas of the volcano eruption in the research area
Full size table

The volcano eruption is estimated to have an impact of 133.31 ha for settlements, 92.32 ha for paddy fields, 235.60 ha for dry farming, 570.98 ha for plantations, 380.86 ha for bare land, and 0.12 ha for forest areas. In this study, estimates of the number of houses destroyed was done by cross profiling or overlaying the map of pyroclastic deposits with the information point of the building sites from the topographic map source of BAKOSURTANAL. ALOS PALSAR imagery with a resolution of 11 m is only used to produce a map of pyroclastic deposits as one of the input data sets in the overlay to the information point of the buildings to take into account the number of damaged houses and cannot be used to estimate the damage to the houses directly. An estimated 4,464 housing units in the village of Glagaharjo were affected, 3,966 units in the village of Kepuharjo, 1,695 units in the village of Argomulyo, 1,225 units in the village of Wukirsari, 865 units in the village of Umbulharjo, and 60 units in the village of Balerante. The total number of buildings damaged is estimated to be around 12,276 units.

The results from the impact of volcano eruptions on landuse obtained for the research area are probably not that useful on ground level, but the results from these calculations can be used as an initial step and rapid method for estimating the magnitude of damage caused by disasters. So, the amount of losses and risk due to the impact of disasters on life and property can be taken into account. As mentioned before, this research used the GIS technique. A superimposed technique between a landuse map and a map of pyroclastic deposits provides an advantage to monitor the impact of volcano eruption in a spatial context. By using a large scale of spatial data input, the damage landuse area can be identified effectively. Given the ability of the GIS technology for monitoring the impact of volcanic eruption, for the next future research, the assessment of the risk and damage in terms of economic losses can be calculated based on the current market value in that area.

4 Conclusion

Distribution of the activities following the eruption of a volcano can be effectively carried out using remote sensing SAR data, such as ALOS PALSAR with the same temporal location. In this paper, we present ALOS PALSAR data with data acquisition conditions pre- and post-volcano eruption. Cross profiling and rapid field work was carried out to describe the environmental conditions within the research area. This is effectively done to investigate the impact of volcano eruption on the environment, given the availability of time and cost of research. The topographic map and geomorphological and landuse factor units are to be considered for profiling and field orientation. The result of calculations of the overlay between the eruption hazard maps with landuse using GIS tools can describe the impact of damage caused by volcano eruptions. Although the result from this calculation is an estimate, it can be used as a rapid method for calculating the volume of disaster damage in the research area. The next future research can be done by calculating the risk of disasters on life and property losses.