Introducing Japanese Landslide Warning and Evacuation System to Sri Lanka: Field Survey of Social Aspect in the Arayanake Area

Abstract

The central highland in Sri Lanka has a landslide-prone geo-hydrological condition. In addition, Sri Lanka has extreme rainfalls in two monsoon seasons, and most of the landslides occur during two monsoon seasons. Landslides in the central highland had been isolated events. However, the area has been developed for agriculture and human settlements, a series of landslides occurred in the mid-1980s. Since the topography, such as mountains with steep slopes and weather, such as high precipitation, are similar in Japan and Sri Lanka, similar landslide phenomena have been seen. One of the common landslide phenomena is rain-induced rapid and long-travelling landslides (RRLL). Landslide disaster risk reduction technologies have recently been developed in Sri Lanka, and foreign technologies have been introduced. Early warning and evacuation using a hazard map is a major system for landslide disaster risk reduction both in Japan and Sri Lanka. Japan has already developed and used the early warning and evacuation system using hazard maps. Since the system has been developed based on Japan’s socio-economic background, it is used successfully in Japan. Thus, when this Japanese technology is used in Sri Lanka, in addition to the engineers’ availability to master technologies, local people’s availability to accept the technology is also examined. In this research, the social background for introducing Japan’s early warning system in Sri Lanka is analyzed based on the field survey at Arayanake. The questionnaire survey and interview were implemented in January 2023, and there were 50 respondents. The survey showed some essential conditions to be improved for the successful warning and evacuation.

1 Introduction

Though similar landslide phenomena are seen in different countries, the landslide disaster risk reduction technology in one country may not be useful for other countries because of the differences in social background. Early warning and evacuation systems using hazard maps have been developed and used in Japan for landslide disaster risk reduction. It is reviewed since there are some essential conditions for effective use.

1.1 Similarity of the Natural Conditions Which Are the Causes of Landslides

The topography of the mountain area is similar in Japan and Sri Lanka. Japan has archipelagos and high steep mountains in the center of the main island. Most precipitation is brought during rainy seasons, and heavy rain-triggered landslides frequently occur in mountain areas. In Sri Lanka, the central part of the island’s southern half is a mountainous area with steep slopes, and a large part of the landslide-prone area is covered with a metamorphosed layer of thick weathered gneiss (Konagai et al., 2023). Most of the landslides occur during extreme rainfalls in two monsoon seasons. The rainfall and geological conditions are the major triggers of landslides. Because of the similarity, similar landslide phenomena are seen in Japan and Sri Lanka.

1.2 Different Social Backgrounds

For successful technology transfer, the social differences between both countries need to be recognized. For example, the mountain surrounding environment is different. The mountain in Sri Lanka is covered with farmland. However, the mountain in Japan is covered with forest. In addition, the population and age category in the mountain area is also different. Young people live in mountain areas in Sri Lanka, and the population is increasing. On the other hand, old people live in mountain areas in Japan, and the population is decreasing (Table 1; Fujita 2022). Since landslides in Japan have triggered frequent disasters, the old people in mountainous areas have experienced landslides and know the timing and safe route to evacuate, and the research for disaster risk reduction has been developed for a long time. The information system for evacuation is well developed, hazard maps are available, and disaster education is provided to the local people.

Table 1 Socio-economic differences in mountain areas in Japan and Sri Lanka

1.3 Necessary Social Background for Effective Early Warning and Evacuation in Japan

Based on accurate regional rainfall forecasting, early warning is effectively used under integrated management in landslide-prone areas in Japan. Following are essential conditions for local people to evacuate (Fujita and Shaw 2019).

• 1. Ability to interpret hazard maps by local people.

  Japan’s map education starts from elementary school. Therefore, the hazard map is well understood. In addition, designated evacuation places are also on the map. However, map education in Sri Lanka is not compulsory since geography is an elective subject for senior secondary level from 14 to 15 years old (Kumara and Ananda 2006). Therefore, it is necessary to know whether the local people in the selected area can interpret the map.

• 2. Effective information system

  Since landslides occurred frequently in Japan, the information system in mountain areas has been developed. There are several information devices, wireless devices, fixed and mobile phones, etc. The network has also been developed between national and local governments, fire stations, firefighters, and local people.

• 3. External help for evacuation and after evacuation

  External help is also expected when they evacuate. In Japan, if evacuees can reach the designated evacuation places, they can receive necessities such as water, food, and blankets provided by various organizations and individuals such as central or local government, private companies, volunteers, etc.

2 Basic Information of Selected Area: Arayanake

Arayanake, Kegalle District (Fig. 1) was selected as the survey area. Rain-induced Rapid and Long-travelling Landslide (RRLL) occurred in Arayanake on 17 May 2016, due to the heavy rainfall from 15 May 2016. It caused 31 deaths and 96 missing persons (Handa et al. 2018). Small-scale family farming, especially tea farming, is common in this area.

Fig. 1

Arayanake, Kegalle District

2.1 Questionnaire Survey

The social background for introducing Japan’s early warning system in Sri Lanka is analyzed based on the literature review and field survey. The questionnaire survey was conducted near the Arayanake landslide area to know the local people’s perception of landslide disaster risk reduction. The National Building Research Organisation (NBRO) provided the map of the river catchment in the Arayanake Landslide area (Fig. 2). The Red zone is the landslide area that occurred in 2016. Black spots are the buildings, and most of them are houses.

Fig. 2

Arayanake Landslide Area

Based on the NBRO-provided map, two areas for the survey were selected, near the main road and mountain area. The questionnaire survey was conducted at 50 houses. All of them are Sinhalese. There are five categories of questions as follows:

• 1. Demographic information: Q1-11

• 2. Livelihood: Q12–14

• 3. Experience of landslide: Q15

• 4. Hazard map and early warning: Q16-29

• 5. External help: Q30-36

There were 50 respondents in the selected area, 16 males and 34 females (Fig. 3). The questionnaire survey and interview were conducted during the daytime at each respondent’s house. The major age group is from 30 to 60. Most of them live in their own houses. 21 people out of 50 respondents work in their tea garden (Fig. 4) next to /near their houses. Since there was a landslide in May 2016, it was asked if they evacuated, and 49 respondents said yes.

Fig. 3

Basic information of respondents

Fig. 4

Occupation of the respondents

2.2 Ability to Interpret Hazard Map By Local People

The 1:50,000 landslide hazard maps are available for the central highland (Bandara and Jayasingha 2018). Several maps are seen in the survey area (Fig. 5a and b). Figure 5a was put on the wall of a community center after the landslide in 2016. A few villagers at the community center said they realized they lived in a landslide hazard area after they saw the map.

Fig. 5

(a) Left: Hazard map -Community center near Grama Niladhari Office (b) Right: Landslide early warning response plan -near Hathgampola Maha Vidyalaya (School)

Based on the literature review, it is said that map education is available. However, it cannot be said that they can interpret maps. As Fig. 6 shows, 21 respondents (42%) said they had never seen a hazard map in their living area. Figure 7 shows where the questionnaire survey was conducted, colored blue if respondents have seen a hazard map and colored red if respondents have never seen a hazard map. There is no particular area which shows a high rate of map availability. Therefore, it is necessary to clarify whether there are hazard maps in each community.

Fig. 6

Hazard map availability

Fig. 7

Location of the respondents’ houses: respondents who have seen the hazard map (blue points) or not (red points)

2.3 Reliable/Available Information System for Warning and Evacuation

Fourty nine persons out of 50 respondents experienced a landslide in May 2016, six persons (12%) out of 49 respondents received a warning, and 43 persons (88%) did not receive the warning (Fig. 8). However, half of the respondents could receive early warning until now (Fig. 9). Figure 10 shows the locations which respondents have received a warning (blue points) or not (red points). There is no particular area which shows a high rate of warning reception. Even near the main road, some of them cannot receive a warning, and even in the hilly area, some of them can receive a warning. Since receiving early warning is important for evacuation, it is necessary to know more about the information system.

Fig. 8

Warning received / not received in case of landslide in May 2016

Fig. 9

Experience of receiving landslide warning

Fig. 10

Location of the respondents’ houses: respondents who have received the warning (blue points) or not (red points)

Getting information through many devices is preferred since one may not be enough in a disaster. For example, mobile phones may not be used during disasters, and other devices or communication systems may be effective. Therefore, available information devices are asked. As Fig. 11 shows, most respondents receive information through TV. Though the number who get daily information through mobile phones is 34, all of them have mobile phones. Mobile phones are useful not only for getting information but also for sharing information. Therefore, the use of a mobile phone is expected to be used during a disaster.

Fig. 11

Source of daily information

Since the use of mobile phones during a disaster is expected, the usable function of mobile phones is asked. As Fig. 12 shows, all of them use the phone function, and 41 use SMS, such as WhatsApp, Instagram, and Facebook. It seems sharing information during landslide disasters is possible through mobile phones. However, there is the issue of signal. Respondents, especially those who live on mountain slopes, said the signal is weak, and they cannot use mobile phones at home. They need to go out and find a place to use mobile phones. Though they cannot use mobile phones at home, there is a good community network, so if someone can receive emergency information in the community, they visit neighbors to convey information. In addition, the mobile phone is also expected to be used in the farm sector. However, poor internet literacy and accessibility are the issues (De Silva Lasanthi et al. 2012).

Fig. 12

Using the function of a mobile phone

Then, the necessary information for evacuation is asked to know their needs for evacuation. 44 said they need information on where to evacuate (Fig. 13). Even if they can receive a warning, they don’t know where to evacuate. However, they evacuated in case of a landslide in 2016. They were not sure whether their evacuated places were safe or not, so it can be said that they were luckily evacuated to safe areas.

Fig. 13

Necessary information for evacuation

2.4 External Help for Evacuation and after Evacuation

In the case of the landslide in May 2016, 49 respondents evacuated, and one person did not evacuate since she needed to take care of an old person in her family. Based on the interview, there were several reasons that they hesitated to evacuate and did not know the safe places for evacuation. One reason is because of their family members who are old and/or sick. Another reason is because of bad road conditions. Figure 14 shows their evacuated places and the number of respondents who evacuated. Some of the respondents evacuated a few times in different places, therefore, the number of evacuated places is more than 50. Most of them evacuated temples or schools. Two respondents answered that they just evacuated to the main road since they did not know where to evacuate. Figure 15 shows where the questionnaire survey was conducted (50 red spotted houses), and the major places where they evacuated (yellow mark). Twenty-six respondents evacuated to the Rathnajoothia Ramaya temple, and 12 evacuated Hathgampala school. They are located along the main road. Dippitiya is also the area for them to evacuate, though it is about 2 km from Hathgampala. seven respondents evacuated to Dippitiya school.

Fig. 14

Evacuated Places for Arayanake Landslide in May 2016

Fig. 15

Location of the houses of the respondents (red) and evacuated places (yellow)

When local people evacuate, external help is expected. Therefore, it was asked if there were external help during the evacuation. As Fig. 16 shows, safe water, food, etc. were provided by various organizations such as international NGOs, local government, etc. 21 respondents answered they didn’t know.

Fig. 16

External help of materials

Education for disaster risk reduction is also an external help, and the availability of disaster education was asked. The local government is the major organization that provides landslide disaster education (Fig. 17). Twelve respondents answered that they don’t know.

Fig. 17

Education of landslide disaster risk reduction

3 Interview at a Temple and Schools

Based on the result of the evacuation in the 2016 landslide, most people evacuated to temples, especially Rathnajoothia Ramaya (also called Rathnajothi /Rathnajothia /Viyaneliya temple) and schools, especially Hathgampala Maha Vidyalaya (also called Hathgampala school) (Fig. 12). Since they are not located inside the red and yellow zones of the map in Fig. 2, it is considered to be safe and suitable places to evacuate. Therefore, the interview was conducted with a monk at Rathnajoothia Ramaya temple and the principals at Hathgampala primary and secondary schools to know what external support is expected as expected evacuation places.

3.1 Rathnajoothis Ramaya Temple

Now, 5 monks live in the temple, and it was possible to interview one of them. When the landslide occurred in 2016, 600 people evacuated from the temple, and the longest period of evacuees was 3 months. There was no financial support from anybody. However, Muslim and local people provided food and water. Though he could not receive a warning from the government, warning information was shared by many villagers.

3.2 Hathgampala Schools

The interview was conducted with the principals of Hathgampala primary and secondary schools. The schools are located within walking distance. Disaster training is provided to students twice a year in Hathgampala Primary School, with about 200 students of 5–10 years old. The students can study map education in environment class. Disaster education and geography class is provided at Hathbampala Secondary School, with about 800 students.

The schools were expected to be evacuation places when the landslide occurred in 2016. They could have temporary external help during and after the landslide, such as tents and sanitary toilets. The principals have been involved in a WhatsApp group called “Disaster Management Mawanell.” They have received disaster information through the WhatsApp group, such as warnings, maps, and pictures from NBRO and other government organisations. In addition, they could share information through the WhatsApp group. WhatsApp is also used for communication among teachers and parents. However, some parents cannot receive information through WhatsApp because of the weak or no signal of mobile phones.

The training program in landslide disaster risk reduction is appreciated by both Hathgampala primary and secondary schools if materials are provided. The principal of the primary school said about 7 years have passed since the landslide occurred in 2016, and some students are still afraid of the landslides. For some students, it seems the landslide disaster is a past event, so it will be a good timing to have the disaster education with the help of others now.

4 Conclusion

As a result of the questionnaire survey and interview, three major essential conditions need improvement for successful early warning and evacuation. The following are the difficult conditions.

• 1. More than half of the respondents do not know if there is a hazard map or not.

  Since they live in a landslide-prone area, knowing the safe place to evacuate is important. If all of them can see the hazard map of their living area, they will know the available place to evacuate.

• 2. They have several information devices to receive a warning, however, six persons received a warning, and 43 persons did not receive a warning in case of the 2016 landslide disaster. In addition, half of them have never received a warning so far.

  Mobile phones are very useful not only for receiving information but for sharing information. However, because of no signal or weak signal, many respondents cannot receive information through mobile phones. Though mobile phones are not used by all, the community network is effective. The survey shows they already have a good communication network and know who cannot receive information through mobile phones. In this case, neighbors with good signals can knock on their doors and share information. However, it is appreciated that they can have a few ways to receive the information since only one way to receive the information is not working sometimes. A further survey for improving the information system is needed.

• 3. Even if they can receive a warning, 44 out of 50 respondents do not know the safe places to evacuate.

  The most important thing to be considered is that they do not know the safe area to evacuate. 49 respondents evacuated in the case of a landslide in 2016, however, 44 persons out of 50 respondents answered the questionnaire survey that “They need the information for a safe place to evacuate”, so they were not sure whether their evacuated places were safe or not. They might evacuate based on their experience or intuition and lucky. Experience and intuition are important, however, if all of them in the selected area know the safe area to evacuate, the number of casualties will decrease. The temples and schools they evacuated are the right places to evacuate since they are out of the hazard area. If all local people recognize the safe area, it will be easier for them to evacuate. In addition, if some temples and schools are designated or recognized as evacuation places by both local people and external organizations such as official and private supporters, the local people will think they can receive any support, and supporters can provide necessary to the designated or recognized evacuation places. It will encourage the local people to evacuate.

Providing the training program is suggested to improve the above-mentioned difficult conditions for better early warning and evacuation. During the training program, it is necessary to clarify whether hazard maps exist in each community. If there is no hazard map, it can provide the hazard maps since NBRO already prepared the hazard maps. Map education is also possible be provided, then participants can know where they live, evacuation routes and the safe places to evacuate. In addition, during the training program, it is also possible to discuss the available information system and how to utilize it for warning and evacuation. Since the local government is the major organization to provide landslide disaster education, as the survey result shows, their collaboration is expected for providing the training program.

When the interview was conducted at both Hathgampala primary and secondary schools, a disaster training program is appreciated if materials are provided. Since the schools are well known, and many people evacuated there in case of the 2016 landslide, the Hathgampala schools are considered suitable organisations to start the training program. Then, providing training programs at some community centers may also be possible.