, Volume 11, Issue 4, pp 491–501 | Cite as

The Psychosocial Impact of the Environmental Damage Caused by the MT Merapi Eruption on Survivors in Indonesia

  • Sri Warsini
  • Petra Buettner
  • Jane Mills
  • Caryn West
  • Kim Usher
Original Contribution


The eruption of Indonesia’s Mount Merapi volcano in 2010 caused extensive environmental degradation. Settlements and hundreds of hectares of farmlands were buried under volcanic ash. Until now, there has been no research on the psychosocial impact of living in an environment damaged by a volcanic eruption. We studied and compared the psychosocial impact of environmental damage on volcano survivors from two subdistricts—Cangkringan and Pakem. Cangkringan survivors affected by the 2010 eruption continue to live in a damaged environment. The Pakem subdistrict was damaged by eruptions of Mt Merapi in the 1990s but there is no recent damage to their environment. The Indonesian-Environmental Distress Scale (I-EDS), a translated revision of the original Environmental Distress Scale (EDS), was used to collect data. Exploratory statistical methods and multivariate linear regression analyses were performed to examine the relative contributions of demographic variables on the psychosocial impact of living in an environment damaged by volcanic eruption. A total of 348 survivors of the Mt Merapi eruption participated in the survey. The mean I-EDS score for Cangkringan district was 15.8 (SD 1.6; range 11.8–19.8) compared to 14.6 (SD 1.3; range 11.8–18.3) for Pakem district (P < 0.001). This result was confirmed by multiple linear regression analysis showing further that older respondents (P < 0.001), unemployed and retired respondents (P = 0.007), and respondents with no formal school education (P = 0.037) had lower I-EDS scores compared to the respective reference groups. Survivors of the Mt Merapi eruption who continue to live in the environment damaged by the 2010 volcanic eruption experience environmental distress. Relevant interventions should target those from low sosioeconomic groups to deal with the distress.


disasters environmental impact Indonesia psychosocial aspects volcanic eruptions 


Volcano eruptions occur frequently in the world. According to Feldman and Tilling (2011, p. 314), a volcanic eruption occurs five or six times a month somewhere in the world. Today, there are more than 500 active volcanoes in the world; these volcanoes may erupt periodically or continuously. Indonesia has a large number of volcanoes and frequently experiences volcanic eruptions because it lies between two continents and is surrounded by the Pacific Ring of Fire, Indonesia. Indonesia is ranked number four for the largest number of casualties resulting from volcanic eruptions and accounts for 11% of the total number of victims worldwide during the twentieth century; it is ranked first in term of injured and homeless survivors (Witham 2005).

Volcanic eruptions not only result in large numbers of victims (deaths or injuries), but also damaged ecosystems. Several sources also claim that volcanic eruptions may cause climate change as well as environmental degradation (Chester 1988; Sadler and Grattan 1999; Rampino 2002; Oppenheimer 2003). For example, sulphur dioxide gas from volcanic eruptions settles in the stratosphere within several weeks to months, and this is said to cause anomalies in the earth’s air temperature (Rampino 2002; Oppenheimer 2003). In addition, lava re-sedimentation from a volcanic eruption may cause damage to the environment for many years (Self 2006). Even though the number of fatalities caused by volcanic eruptions is often significantly lower than that of other types of natural disasters (Feldman and Tilling 2011), the effects of environmental damage may be an ongoing burden for survivors. 2 years after the last eruption of Mount Merapi in Indonesia in 2010, residential areas, rice fields, farms, roads, bridges, and rivers remained extensively damaged because of materials discharged from the volcano. The cold lava flood continues even today because of heavy rainfall levels in the area (Nugroho 2013).

Environmental degradation has potential to affect the survivors who live in the Mt Merapi area, adding to their burden after the disaster. However, to the best of our knowledge no research has explored the experience of living in an environment damaged by a volcanic eruption. Previous studies on the impact of eruptions have focused on the psychosocial impact on survivors as a result of the perceived loss of their possessions, property or loved ones, or about their experiences living in an evacuation or relocation shelter.

Very few studies have explored the psychosocial impact of environmental degradation. We do know, however, that people living in damaged environments may feel distressed, anxious and saddened by the changes they see. Several international researchers have studied this psychosocial phenomenon. For example, Van Haaften and Van De Vijver (1996) investigated the impact of environmental change on people in Africa. Crighton et al. (2003) explored the somatic symptoms due to drying up the sea ground the coastal area of Uzbekiztan, and Cunsolo Willox et al. (2012) reported the impact of environmental change on the Nunatsiayut in Canada. Australian researchers have also studied the psychosocial impact of environmental damage. Connor et al. (2004) undertook a qualitative study of the impact of open cut coal mining in the Hunter valley area of New South Wales, Australia (Connor et al. 2004) that led on the work undertaken by Higginbotham et al. (2007) who developed the EDS to measure the impact of environmental distress. In addition, others have explored the impact of living on drought-affected farmland (Sartore et al. 2008), and salinity-affected land (Speldewinde et al. 2009).

The researches revealed psychosocial problems such as depression (Speldewinde et al. 2009), loss of identity (Sartore et al. 2008), changes sense of place (Cunsolo Willox et al. 2012) and psychosocial and physical disturbances resulting from environmental distress (Connor et al. 2004; Higginbotham et al. 2007; Cunsolo Willox et al. 2012).

All these studies support an assertion, first made by Albrecht in 2002, that people can feel ‘homesick’ at home when the environment around them is significantly changed or damaged. Albrecht coined the term ‘solastalgia’ to describe the condition. He defined ‘solastalgia’ as the distress or sadness a person feels when his or her home environment is negatively changed. Solastalgia is a kind of longing for return of the beauty and security their home environment once gave. People experiencing solastalgia may describe feeling unfamiliar in their own environment, even though they are living in the place in which they have always been. This distinguishes solastalgia from nostalgia; that is, survivors experience a type of homesickness even though they have not moved from their homes (Warsini et al. 2014).

Albrecht (2005) claimed that solastalgia may also be seen as a result of natural disasters such as floods, earthquakes and tsunamis and may not be experienced only by people who suffer from man-made disasters such as earth exploration, deforestation, drought, and war. Nevertheless, the hypothesis that solastalgia may be caused by natural disasters has never been tested.

The aim of this cross-sectional study was to determine the psychosocial impact on survivors of the environmental degradation from the 2010 Mt Merapi volcanic eruption. In particular, the study compared the feelings of environmental distress experienced by respondents from the Cangkringan subdistrict who directly experienced the last eruption (2010) and continue to live in a damaged environment, with respondents from the Pakem subdistrict who experienced several eruptions prior to 2010 but live in an environment that was not damaged by the last eruption of Mt Merapi.


Setting and Sample

The sample was collected from people living in the areas affected by the 2010 Mt Merapi eruption and areas affected by previous eruptions; Cangkringan and Pakem areas. A total of 350 people completed the survey. Based on the Higginbotham et al. (2007) study, a target sample size of 175 people from Cangkringan and 175 from Pakem was deemed sufficient to detect a mean difference of 1 unit in the EDS scale with a power in excess of 80% and an overall level of significance of 0.05.

A systematic stratified sampling approach was utilised. The sample was divided into two groups based on the subdistricts. Two villages were selected in both Cangkringan and Pakem; Cangkringan was most severely affected by the 2010 Mt Merapi eruption. The systematic data collection process was conducted by visiting every fifth house in each village, starting at the first house in every street, alternating between adult and elderly respondents and data collection continued until the required sample was reached. When the research assistant knocked at the door, they asked for a male or female to answer the questions if possible depending on the gender of the previous participant.

Inclusion criteria for respondents were as follows: (1) adult and older than 18 years; (2) experienced the 2010 eruption of Mt Merapi and now living in a damaged environment (Cangkringan people; only); (3) experienced the 2010 eruption of Mt Merapi but living in an area not damaged by the latest eruption (Pakem people only); (4) able to communicate in Indonesian; and (5) willing to consent to participate in the study. Respondents were excluded if they suffered from or had a history of severe mental illness such as schizophrenia or depression and/or reported difficulty hearing or speaking.

Data Collection Procedure

The study was conducted in two villages of the Cangkringan subdistrict, which were affected by 2010 Merapi eruption, namely Argomulyo (five hamlets) and Glagaharjo (three hamlets). In the Pakem subdistrict, two villages were identified as being the most severely affected by the 1994 and 1997 eruptions—Pakembinangun (two hamlets) and Hargobinangun (two hamlets). Data collection was conducted 2 years after the 2010 Merapi eruption from January to February 2013.

During the survey, 14 research assistants who were fluent in the local language were paid to collect data. Research assistants were given 2 days training to ensure quality and consistency in data collection. All respondents were approached in their homes. The survey was presented in Bahasa Indonesia but read out in Javanese for respondents who could not read the Bahasa version and for respondents who could not read or write.

The Instrument

The first section of the questionnaire included questions about demographics (age, gender, marital status, level of education, occupation, current living arrangement), living experience in Merapi area (length of stay, how many generation, have children and grandchildren who stay in the Mt Merapi area) and volcano-related experiences of respondents (experiences of relocation, family deaths and injuries, loss of property and frequency of eruption and evacuation).

The original EDS consists of 117 items (including demographic background questions) and was developed by Higginbotham et al. (2007) to measure the environmental distress experienced by people who lived in an environment damaged by open-cut mining industries. The tool was tested with populations in the Upper Hunter Valley, Australia, and demonstrated strong internal consistency (Cronbach’s alpha = 0.79–0.96) and test retest reliability [intra-class correlation coefficients (ICC) = 0.67–0.73] (Higginbotham et al. 2007). The authors suggest that this tool could be adapted to measure distress arising in other contexts such as climate change, and natural disaster (Higginbotham et al. 2007). The original scale has eight components (place attachment, frequency, observation, threat, impact, solastalgia, action and reliability of environmental information (trustworthy) (Higginbotham et al. 2007), while the Indonesian EDS (I-EDS) has seven components only. The trustworthy component was removed in the Indonesian version of the EDS as it was not appropriate for the context of the study and number questions were modified for a volcano disaster setting and added. As a result, the original Indonesian version of the EDS tool has a total of 127 items.

The tool was translated into Bahasa Indonesia following the Brislin translation model (Jones et al. 2001), and pilot tested and retested for reliability and validity. Testing of the tool indicated high internal consistency as Cronbach alpha was 0.91 for test and 0.92 for retest. Intra-class correlation also had high results (r = 0.81, 95% confidence interval = 0.72–0.87) (Warsini et al., under review). According to results of the psychometric testing, some questions were revised and nine questions about environmental hazards were deleted, leaving 118 questions in the final tool; the I-EDS.

The I-EDS has seven components excluding the demographic section: place attachment, frequency, observation, threat, impact, solastalgia and action. Place attachment measures the people’s bonding to their environment, in this case is volcano and solastalgia assesses people’s emotional feeling toward environmental destruction. Both subscales consist of ten questions (five options from strongly agree–strongly disagree). Observation, frequency and threat elements determine what kind of changes, how often and threatening the damage experienced by respondents. These subscales consist of 15 items (observation: yes or no options; frequency: three options from not anymore to nearly every day and threat : three options from unthreatening to very threatening). The impact subscale consists of 27 items with 5 options (agree–disagree) that investigate people’s feeling toward the impact of environmental changes due to eruption. The action element describes kinds of activities performed by respondents to overcome the ecological problems and consists of nine items (yes or no options).

An overall (total) environmental distress score was calculated for each individual by combining their subtotal scores from the seven components (place attachment, observed, frequency, threat, impact, solastalgia and action). However, because the seven components varied in the number of questionnaire items that comprised them, their subtotal scores had to be standardised first before they could be added together. This was done by dividing the component subtotal score for each person by the number of items in that component. All weighted component scores were then added up to reach the total score (Higginbotham et al. 2007).

Data Analysis

All data analysis was performed using Statistical Package for the Social Sciences (SPSS) version 21 software package (IBM SPSS, Chicago, IL, USA). Numerical data were described using mean values and standard deviations (SD) when symmetrically distributed and median values and inter-quartile ranges (IQR) when skewed. Parametric tests (t test) and non-parametric tests such as Mann Whitney test, Chi-square and Fischer tests were performed to compare the I-EDS score and demographic variables of respondents in both locations. Parametric tests were used for analysing continuous and normally distributed data, while non-parametric tests were used to analyse categorical data and not normally distributed data. Prior to the conduct of bivariate analysis, some recategorization of categorical data was performed.

Multiple linear regression analyses were performed to predict the risk factors of environmental distress experienced by respondents 2 years after disaster. The I-EDS score was used as the dependent characteristic. The following independent variables were considered: age, gender, marital status, occupation, level of education, location of residence, house ownership, and loss of property experienced during the 2010 eruption. Age was considered as a continuous variable while all categorical variables were dummy coded in preparation for the modelling process. Stepwise forward and backward selection procedures were applied to reach a final stable model. All characteristics not in the final model were checked for potential confounders. All possible two-way interactions were assessed in hierarchical models. A P value of less than 0.05 was considered statistically significant throughout the analysis.

Ethic Approval

The University Human Research Ethics Committee (HREC) at James Cook University Australia and Institutional Review Board of Faculty of Medicine in Universitas Gadjah Mada, Indonesia approved the study. Prior to the interviews, written informed consent was obtained from all respondents. Information about where to seek psychological support should participation in the survey cause distress was provided to respondents.


Two of the original respondents were excluded due to their condition. One of respondents had a chronic physical illness and the other had resided in the Mt Merapi area for less than 10 years. Therefore, responses were analysed from 348 respondents. Overall, 175 (50.3%) respondents resided in Cangkringan. Mean age of participants from Cangkringan was 58.0 years (SD 18.3; range 20–100) while respondents from Pakem were 56.3 years (SD 17.3; range 20–95; P = 0.369) (Table 1). Age and gender were almost equally distributed in both locations. In both groups,the majority of respondents were married (72–78%), worked as farmers (40–41%), and had less than 9 years education (33–40%). As outlined in Table 1, there were no large differences in the demographic characteristics observed between the respondents in the two locations indicating that respondents in both subdistricts share similar characteristics.
Table 1

Demographic Characteristics and Previous Experiences of Volcanic Eruptions Stratified by Residence (Cangkringan and Pakem districts) of 348 Survivors of the Mt Merapi Eruption, Indonesia, October 2010.




P value

(n = 175)

(n = 173)

Demographic characteristics

 Mean age (SD) [years]

58.0 (18.3)

56.3 (17.3)


 Female participants

89 (50.9%)

89 (51.4%)


 Marital status


126 (72.0%)

135 (78.0%)



44 (25.1%)

31 (17.9%)


5 (2.9%)

7 (4.0%)

 Level of school education

  No formal education

55 (31.4%)

57 (32.9%)


  Less than 9 years

59 (33.7%)

70 (40.5%)

  10–12 years

50 (28.6%)

44 (25.4%)


11 (6.3%)

2 (1.2%)



11 (6.3%)

6 (3.5%)



73 (41.7%)

69 (39.9%)


21 (12.0%)

18 (10.4%)


17 (9.7%)

33 (19.1%)

  Unemployed or retired

53 (30.3%)

47 (27.2%)

 Originally from Merapi

158 (90.3%)

164 (94.8%)


 Median number of generations lived in Merapi (IQR)

5 (4–6)

5 (4–6)


 Mean length of stay in Merapi (SD) [years]

54.5 (19.2)

52.7 (18.1)


 With children staying in Merapi

160 (91.4%)

158 (91.3%)


 With grandchildren staying in Merapi

97 (55.4%)

93 (53.8%)


 House ownership

  Own house

138 (78.9%)

139 (80.3%)


  Stay with family

36 (20.6%)

31 (17.9%)

  Rental accommodation

1 (0.6%)

3 (1.7%)

Experience of volcano eruptions

 Median number of eruptions experienced before 2010 (IQR)

3 (1–3)

3 (3–5)


 Had to relocate

48 (27.4%)

6 (3.5%)


 Family members died in eruptions before 2010

0 (0%)

11 (6.4%)


 Family members died in 2010 eruption

46 (26.3%)

2 (1.2%)


 Family members injured in 2010 eruption

22 (12.6%)

1 (0.6%)


 Median number of evacuations experienced in 2010 eruption (IQR)

1 (1–3)

2 (2–3)


Loss of property


50 (28.6%)

103 (59.5%)


 Loss 1–2 items

99 (56.6%)

70 (40.5%)

 Loss ≥3 items

26 (14.3%)

0 (0%)

SD Standard deviation, IQR inter-quartile range.

The median number of volcanic eruptions experienced by residents of the two districts was 3, respectively, however, statistically Pakem residents had experienced significantly more previous eruptions than residents of Cangkringan (P < 0.001) (Table 1). The experience of the eruption in 2010 was more serious for the Cangkringan district with 46 (26.3%) respondents reporting death of at least one family member compared to 2 (1.2%) from the Pakem subdistrict (P < 0.001). Similarly, family members in Cangkringan were more frequently injured and reported greater loss of property during the 2010 eruption compared to respondents from the Pakem subdistrict (P < 0.001, respectively; see Table 1).

Five of the seven subscales of the I-EDS showed higher values for Cangkringan compared to Pakem (P < 0.001, respectively; Table 2). The mean score for the subscale “place attachment” was significantly lower in Cangkringan (39.7 vs 40.5; P = 0.014). Overall the total I-EDS scores were significantly higher in Cangkringan (mean 15.8) compared to Pakem (mean 14.6; P < 0.001) (Table 2). It can be concluded that people in Cangkringan experienced more environmental damage, feel more threatened and experience more of an impact from the changes to the environment than the respondents from Pakem. As a consequence of the eruption, people from Cangkringan also felt less attached to the land and experienced more sorrow than Pakem people. There was, however, no significant difference in terms of the action element between the two locations.
Table 2

Average Results of I-EDS Total Scale and Seven Subscales Stratified by Residence (Cankringan and Pakem Subdistricts) as Reported by 348 Survivors of the Mt. Merapi Eruption, Indonesia, October 2010.

I-EDS scores of subscales and total

Cangkringan (n = 175)

Pakem (n = 173)

P value

Mean “place attachment” score (SD)

39.7 (3.1)

40.5 (3.0)


Mean “observation” score (SD)

27.1 (2.5)

24.4 (2.9)


Median “frequency” score (IQR)

22 (18–26)

13 (10–18)


Mean “threat” score (SD)

28.2 (7.3)

25.2 (6.5)


Mean “impact” score (SD)

82.6 (10.1)

76.4 (8.8)


Mean “solastalgia” score (SD)

33 (4.9)

30.9 (4.0)


Median “action” score (IQR)

3 (1–4)

3 (2–4)


Mean of total I-EDS score (SD)

15.8 (1.6)

14.6 (1.3)


SD Standard deviation, IQR inter-quartile range.

Table 3 also demonstrates that Cangkringan respondents declared that they were exposed to the environmental destruction more frequently and felt more threatened by the environmental destruction compared to respondents from Pakem. However, respondents from Pakem felt the exposure to volcanic activity more significantly; for example visual pollution from smog (64.2 and 11.6% compare to 53.7 and .6%). Table 3 outlines the impact of environmental degradation 2 years post the eruption for Cangkringan respondents; for example, damage to infra-structure and the river by the volcano and by post-volcanic restoration and rebuilding work. The majority of respondents claimed the three biggest threats to the environment related to the erosion of the river mouth, damaged by the cold lava flows, pollution from trucks, and vibration from the mountain.
Table 3

Experienced Frequency and Threat of Environmental Hazard as Described by Survivors of the 2010 Merapi Eruption (N = 348).

Environmental hazards

Frequency (often and nearly always)

Strong threat





(n = 175)

(n = 173)

(n = 175)

(n = 173)

Hazard from truck


143 (81.7%)

60 (34.7%)

83 (47.4%)

21 (12.1%)


149 (85.1%)

65 (37.5%)

49 (28%)

7 (4%)


140 (80%)

42 (26.6%)

39 (22.3%)

5 (2.9%)

Hazard from mountain

 Visual pollution

94 (53.7%)

111 (64.2%)

34 (19.4%)

26 (15%)


15 (8.6%)

20 (11.6%)

71 (40.6%)

67 (38.7%)

Infrastructure damage

 Houses, building

92 (52.6%)

22 (12.7%)

63 (36%)

42 (24.3%)


98 (56%)

61 (35.3%)

19 (10.9)

17 (19.8%)

 Soil erosion in river-side

63 (36%)

47 (27.2%)

51 (29.1%)

26 (15%)


 Dust off the roads

112 (64%)

34 (19.7%)

47 (26.9%)

36 (20.8%)


32 (18.3%)

7 (4%)

26 (14.9%)

25 (14.5%)

Table 4 shows how respondents felt about the impact of the damage caused by the eruption including the negative impact on their daily life (long travelling times because of road damage, hot temperature), in terms of livelihood (trouble in finding feed for livestock, dry soil for the farmers), and the declining value of their properties. Cangkringan respondents were concerned by these problems more than respondents from Pakem.
Table 4

Impact of Environmental Degradation and Solastalgia Experienced by Survivors of the 2010 Merapi Eruption (N = 348).


Cangkringan (n = 175)

Pakem (n = 173)

Impacts of environmental change

 Health impact

  A lot of asthma or other respiratory disease because of air pollution

79 (45.1%)

68 (39.3%)

  Worried about the increasing mental health problems in neighbourhood

75 (42.9%)

59 (34.1%)

 Losses and negative impact on daily living

  A lot of people loss their family

152 (86.8%)

68 (39.3%)

  Took more time to go anywhere

141 (80%)

67 (38.7%)

  Temperature is getting hotter

123 (70.3%)

66 (38.1%)

  Farm land become drought

117 (66.8%)

53 (30.6%)

  The decreasing of home/property’s value

113 (64.5%)

92 (53.2%)

 Dissatisfaction and desperation

  The overall impact of disaster in this area is depressing

134 (76.6%)

113 (65.3%)

  Feel disappointed about degradation in local environment

114 (65.1%)

95 (55%)

 Positive impact

  Merapi eruption provides economic benefits

75 (42.9%)

62 (35.9%)

  Living more closely with neighbour

153 (87.4%)

145 (83.8%)

  People are accepting of whatever the situation is in Merapi

165 (94.3%)

167 (96.5%)


 Loss sense of peace and quite

135 (77.2%)

101 (58.4%)

 Threat of farming lifestyle

124 (70.8%)

74 (42.8%)

 Feel saddened to look at degraded landscapes and everything is buried

120 (68.5%)

111 (64.1%)

 The situation challenged my sense of belonging

27 (14.4%)

20 (16.6%)

 The thought of government forced to leave this place upsets me

32 (18.3%)

39 (22.5%)

 Feel good about restoration

162 (92.5%)

162 (93.7%)

Cangkringan respondents were also more worried about the physical and psychological health of family and community members compared to Pakem respondents (45.1 and 42.9% compare to 39.3 and 34.1%). The majority of respondents (more than 65% in Cangkringan and 55% in Pakem) felt profound sadness and disappointment on realising the great changes in their home environment. Regardless of their negative view of the damage, however, most survivors in both locations expressed acceptance of what happened and affirmed the need to continue to live in the damaged environment.

In term of solastalgia, Table 4 also indicates that although eruption survivors in both subdistricts were sad (over 64%) about the damage to the environment and missed their sense of place (77% in Cangkringan and 58% in Pakem), they did not lose their sense of belonging to Merapi (over 83%). They did, however, express willingness to relocate if the government wished them to do so (over 77% of respondents in both subdistricts).

To determine the dominant factors correlating with the respondents’ feeling of environmental distress, a multivariate analysis was performed. Multivariable analysis showed that older people scored lower on the I-EDS scale (P < 0.001) compared to younger people (regression coefficient: −0.031; P < 0.001), as did unemployed or retired people when compared to employees, farmers, merchants and labourers (regression coefficient: −0.428; P = 0.007), and those people with no formal education when compared to educated people (regression coefficient: −0.367; P = 0.037) (Table 5). Respondents from Cangkringan subdistrict scored higher on the I-EDS scale when compared with respondents from the Pakem subdistrict (P < 0.001) when adjusted for all other significant correlates and confounders, indicating that Cangkringan respondents were more affected by environmental damage. A significant two-way interaction was identified between occupational status and location. Unemployed or retired people from Cangkringan scored significantly lower on the I-EDS scale (regression coefficient: −0.619; P = 0.048) compared with respondents from Pakem or employees, farmers, merchants and labourers (Table 5).
Table 5

Characteristics Correlated with the Overall Weighted I-EDS score. Results of Multivariable Linear Regression Analysis Based on 348 Mt Merapi Volcano Eruption Survivors.


Regression coefficient

95% confidence interval

P value

Final stable model of significant characteristics (r2 = 35.1%)

 Age (continuous)


−0.040 to −0.022


 Cangkringan (baseline: Pakem)


1.049 to 1.585


 Unemployed or retired (baseline: employee, farmer, merchant, and labourer)


−0.689 to −0.097


 No formal education (baseline: with education)


−0.680 to −0.003


Model with confounders (r2 = 35.2%)

 Age (continuous)


−0.040 to −0.022


 Cangkringan (baseline: Pakem)




 Unemployed or retired (baseline: employee, farmer, merchant, and labourer)


−0.737 to −0.119


 No formal education (baseline: with education)


−0.712 to −0.022


 Male gender (baseline: female)


−0.395 to 0.172


Model with confounders and interactionsa (r= 36.0%)

 Age (continuous)


−0.040 to −0.022


 Cangkringan (baseline: Pakem)




 Unemployed or retired (baseline: employee, farmer, merchant, and labourer)


−0.540 to 0.341


 No formal education (baseline: with education)


−0.669 to 0.023


 Male gender (baseline: female)


−0.377 to 0.189


 Interaction: Unemployed or retired and residence in Cangkringan (baseline: residence in Pakem or employee, farmer, merchant and labourer)


−1.213 to −0.025


Collinearity diagnostics: maximum condition index 10.8; lowest eigenvalue 0.035 (involving age and level of education). The model r2 was 35.1, 35.2, and 36.0%, respectively; Collinearity diagnostics: maximum condition index 10.8; lowest eigenvalue 0.035 (involving age and level of education).

a The statistically significant interaction involved occupation and residence; as a consequence the effects of these two original variables cannot be any longer assessed in this hierarchical model.


The study was undertaken to determine if people living in an area damaged by a natural disaster experience environmental distress known, including feelings of solastalgia. The results confirm that 2 years after the disaster the survivors living in the area most damaged by the latest eruption of Mt Merapi (Cangkringan respondents) experienced the highest level of distress related to the environmental damage caused by the eruption. Interestingly, survivors reported most distress about the effects of volcanic dust sticking to the road and about the activities of trucks driving back and forth hauling and materials for mining and construction of the relocation area. The survivors reported little distress related to volcanic activities such as vibration, rumbling sounds, and air pollution caused by smoke or dust coming out of Mount Merapi. It can be concluded that the actual problems faced by the community 2 years after the eruption were not associated directly with environmental damage resulting from the eruption, but rather were occurring as a secondary problem that has emerged out of the process of recovery and rehabilitation.

This is a similar phenomenon to that experienced by people living in the mining area in the Upper Hunter Valley. Nearby residents of the mine also felt the impact of dust produced from the mining activities and by trucks or other heavy vehicles (Higginbotham et al. 2007). Results from both studies similarly illustrate how the communities were concerned about damage to homes, public buildings, and rivers. The difference is that Higginbotham’s (2007) study suggested that respondents suffered more from the changes occurring in the natural environment where they lived, soil erosion, and damage to buildings, while in this study, the community is more distressed by the impact of truck-related activities and the damaged river. In the study by Cunsolo Willox et al. (2012) in Canada, respondents were more concerned about the environmental changes associated with the weather, such as the reduction and quality of snow and ice, and the high incidence of storms.

In terms of environmental threats, the studies revealed different results. Higginbotham’s (2007) study suggested that the majority of respondents (53–65%) considered the greatest threat to be dust and its effects on health, but in this study, the greatest threat felt by respondents was from river erosion and pollution caused by trucks. This is not surprising considering that river erosion may increase the risk of cold lava flooding (volcanic debris washed down the river during or after periods of heavy rain), which creates hazards for residents living along the river during the wet season. The risk of cold lava flooding is greater if the river bed is shallow and the water level is high.

Perceived health impacts of both studies also show similarities. Respondents in both studies expressed their concern over the health effects resulting from the environmental destruction, especially in terms of respiratory problems (asthma). Higginbotham’s (2007) study also showed that respondents expressed concern about health problems such as cancer and birth defects as did the study by Cunsolo Willox et al. (2012) who reported a higher incidence of life style-related diseases such as diabetes and high blood pressure due to changes in eating habit resulting from environmental changes. In this study, however, the major concern was about the level of mental health disorders.

The emotional responses due to environmental change impact reported in current study are always higher than that reported in Cunsolo Willox study in term of sadness feeling (42 and 83.4%, respectively); anger (26 and 65.1%); anxiety (35 and 42–52%) and distress (22 and 76.6%). However, the level of helplessness of respondents in Cunsolo Willox study is slightly higher than that in this study (32 and 25.2%). After an eruption, many factors—loss of family, loss of jobs and property, as well as fears of a repeat disaster—may make the survivors vulnerable to psychiatric disorders.

Interestingly, this study shows that no matter how great the negative impacts of the eruption, respondents still express acceptance of the incident. In fact, many respondents indicate that positive lessons and experiences had come from the eruption, such as feeling closer to their neighbours. A number of Merapi survivors believe the environmental damage has a positive impact such as the creation of jobs for some people and the fact that the soil will be more fertile in the future.

In regard to the notion of solastalgia, Cangkringan residents showed significant solastalgia in contrast to that of Pakem residents. This implies that Cangkringan residents have lost their feeling of safety and comfort in their home environment as a consequence of the disaster and the devastation of the surrounding area. In term of farming pattern changing this study produced similar results with Higginbotham et al.’s (2007) and Cunsolo Willox et al.’s (2012). However, in term of changes to the sense of belonging as a result of environmental degradation this research has different results with previous researches. Changes to the sense of place, similar to the notion of the sense of belonging, were expressed by participants of the Cunsolo Willox et al. (2012) study, resulting from the changes to their enviornment resulting in disruption to hunting, trapping, fishing, and foraging which has been linked to physical and psychosocial outcomes. In the case of the participants in this study, this difference may result from positive outcomes that participants believed they gained as a result of the eruption and their acceptance that natural disasters are an expected phenomenon in their area. Interestingly, the respondents in this study indicated that despite their great sense of belonging to the area, they would relocate if instructed to do so. While the processes involved in deciding to relocate are complex, it is important to remember it is also an adaptive strategy (King et al. 2014).

Meanwhile the level of sadness reported in Higginbotham’s study (2007) is slighty higher than that reported in current study (74 and 68.5%) but respondents in this study missing the peace and quiet more than respondents in Higginbotham study. However, as there is no defined solastalgia cut-off point determined by Higginbotham et al. (2007) who developed the original EDS, it cannot be determined whether the eruption survivors in Cangkringan experience the same level of solastalgia as the participants in Higginbotham’s study. Additional studies are needed to further explore this concept.

The stress response of a person is strongly influenced by his/her perception to an event (appraisal) (Antai-Otong 2008). Lazarus in his theory of cognitive appraisal affirms the importance of primary appraisal in the stress response (Lazarus 1998). In this phase, a stressor or the resembling event may be perceived by someone as a variety of things-irrelevant (loss), benign (challenge), or hazardous (threat). If one assumes the damage to the environment as a threat then the appraisal process will continue into secondary appraisal to overcome it, while if one does not regard this as a problem then there is no alarm “stress” and there should be no attempt to overcome the event. This study indicates that younger people (aged range 18–59), those who have jobs, and who are formally educated suffer the most environmental distress feeling. Conversely, older-aged residents, people without a job (retired), and people who have not had any formal education, seemed less affected. The way people perceive a stressor (primary appraisal), may explain the higher level of distress in the elderly, people who are not employed (retired), and people who have a lower level of education (uneducated people). In addition, according to Evans and Cohen (1987), one of the factors that influence the primary appraisal of stress is the duration of the stressor. This is consistent with the theory of Stress and Adaptation by Stuart (2013) which suggests that the longer a person is exposed to a stressor, the more vulnerable that person experiences a stress (time precipitating factor).

The younger age group, the employed and educated people may experience environmental damage as a threat that interferes with them in carrying out their role as breadwinner for their families. Damaged roads and bridges, scattered dust resulting from the eruption and truck activities can be interpreted as a bothering stressor because of the long duration of the exposure. Meanwhile, for the unemployed and uneducated groups who are dominated by the elderly and housewives, environmental damage is considered less of a threat. This is because the elderly, housewives, and the unemployed spend most of their time in the local neighbourhood close to their houses so that they are less exposed to environmental damage and its impact.

People in the younger age group may be more affected also because of their family responsibilities. Lazarus argues that demographic factors such as level of education, income, and employment affect someone in perceiving something as a threat or a nuisance (Lazarus 1998). For the middle-aged group, the major attention is on the economic issues including family life, occupation, property, and investment. Environmental damage caused by the eruption has led to a decline in the selling price of land and houses, loss of property, investment and livelihoods thus being a potential to make them more vulnerable to distress.


This is the first research conducted into environmental distress resulting from a natural disaster. The study found that those people who live in the most devastated environment were the most distressed supporting the work of Albrecht (2005). Those most affected by the psychosocial distress in this study were adults between 18 and 59 years, employed adults, and those adults who were educated.


As this is only the third time the EDS has been used, many items require further revision and development. It is possible that the data collection procedure used in this study may have affected the answers provided to some degree. This study was conducted in Indonesia. To determine the psychosocial distress and solastalgia resulting from environmental damage caused by a volcanic eruption will require further similar studies in other countries.



All participants are gratefully acknowledged for their time and participation. We also thank the research assistants who assisted with data collection and the Chief of each hamlet who supported this research.

Conflict of interest



Funding was received from the School of Nursing, Midwifery & Nutrition and the Graduate Research School at James Cook University, Australia.


  1. Albrecht G (2005). ‘Solastalgia’: A New Concept in Health and Identity. Philosophy Activism Nature 3:41-55.Google Scholar
  2. Antai-Otong D (2008). Psychiatric Nursing Biological & Behavioral Concepts, 2 edition. Thomson Delmar Learning, New York.Google Scholar
  3. Chester DK (1988). Volcanoes and climate: recent volcanological perspectives. Progress in Physical Geography12:1-35.CrossRefGoogle Scholar
  4. Connor L, Albrecht G, Higginbotham N, Freeman S, and Smith W (2004). Environmental Change and Human Health in Upper Hunter Communities of New South Wales, Australia. EcoHealth 1:47–58.CrossRefGoogle Scholar
  5. Crighton EJ, Elliott SJ, Meer Jvd, Small I, and Upshur R (2003). Impacts of an environmental disaster on psychosocial health and well-being in Karakalpakstan. Social science & medicine 56:551-567.CrossRefGoogle Scholar
  6. Cunsolo Willox A, Harper SL, Ford JD, Landman K, Houle K, and Edge VL (2012). “From this place and of this place:” Climate change, sense of place, and health in Nunatsiavut, Canada. Social science & medicine 75:538-547.CrossRefGoogle Scholar
  7. Evans GW, and Cohen S, editors. (1987). Environmental Stress. John Wiley & Sons, New York.Google Scholar
  8. Feldman JN, and Tilling RI, editors. (2011). Chapter 15: Volcanic Eruptions, Hazards, and Mitigation. Volcano Hazards, 6 edition. Mosby Elsevier, Philadelphia.Google Scholar
  9. Higginbotham N, Connor L, Albrecht G, Freeman S, and Agho K (2007). Validation of an Environmental Distress Scale. EcoHealth 3:245-254.CrossRefGoogle Scholar
  10. Jones PS, Lee JW, Phillips LR, Zhang XE, and Jaceldo KB (2001). An adaptation of Brislin’s Translation Model for Cross Cultural Research. Nursing Research50:300-304.CrossRefPubMedGoogle Scholar
  11. King D, Bird D, Haynes K, Boon H, Cottrell A, Millar J, et al. (2014). Voluntary relocation as an adaptation strategy to extreme weather events. International Journal of Disaster Risk Reduction 8:83-90.CrossRefGoogle Scholar
  12. Lazarus RS (1998) Fifty years of the Research and Theory of R.S. Lazarus: An Analysis of Historical and Perennial Issues. Lawrence Erlbaum Associates, Inc., New JerseyGoogle Scholar
  13. Nugroho SP (2013) Banjir lahar dingin Merapi seret Penambang Pasir, 1 tewas. Accessed on 13 February 2013.
  14. Oppenheimer C (2003). Climatic, environmental and human consequences of the largest known historic eruption: Tambora volcano (Indonesia) 1815. Progress in Physical Geography27:230-259.CrossRefGoogle Scholar
  15. Rampino MR (2002). Supereruptions as aThreat to Civilizations on Earth-like Planets. Icarus156:562-569.CrossRefGoogle Scholar
  16. Sadler JP, and Grattan JP (1999). Volcanoes as agents of past environmental change. Global and Planetary Change21:181-196.CrossRefGoogle Scholar
  17. Sartore G-M, Kelly B, Stain H, Albrecht G, and Higginbotham N (2008). Control, uncertainty, and expectations for the future: a qualitative study of the impact of drought on arural Australian community. Rural and Remote Health8:1-14.Google Scholar
  18. Self S (2006). The effects and consequences of very large explosive volcanic eruptions. Philosophical Transactions of The Royal Society A364:2073-2097.CrossRefGoogle Scholar
  19. Speldewinde PC, Cook A, Davies P, and Weinstein P (2009). A relationship between environmental degradation and mental health in rural Western Australia. Health & Place15:880-887.CrossRefGoogle Scholar
  20. Stuart GW (2013). Principles and Practise of Psychiatric Nursing, 10 edition. Elsevier Mosby, St Louis Missouri.Google Scholar
  21. Van Haaften EH, and Van De Vijver FJ (1996). Psychological consequences of environmental degradation. Journal of Health Psychology1:411-429.CrossRefPubMedGoogle Scholar
  22. Warsini S, Mills J, and Usher K (2014). Solastalgia: Living With the Environmental Damage Caused By Natural Disasters. Prehospital and Disaster Medicine 29:1-4.CrossRefGoogle Scholar
  23. Witham CS (2005). Volcanic disasters and incidents: A new database. Journal of Volcanology and Geothermal Research148:191- 233.CrossRefGoogle Scholar

Copyright information

© International Association for Ecology and Health 2014

Authors and Affiliations

  • Sri Warsini
    • 1
  • Petra Buettner
    • 2
  • Jane Mills
    • 1
  • Caryn West
    • 1
  • Kim Usher
    • 1
  1. 1.School of Nursing, Midwifery and NutritionJames Cook UniversityCairnsAustralia
  2. 2.School of Public Health, Tropical Medicine and Rehabilitation SciencesJames Cook UniversityTownsvilleAustralia

Personalised recommendations