Advertisement

Waste Disposal & Sustainable Energy

, Volume 1, Issue 4, pp 301–317 | Cite as

Lessons learned from the Ezgeleh–Sarpol Zahab earthquake of November 2017: status of damage and disposal of disaster waste

  • Amirhomayoun SaffarzadehEmail author
  • Takayuki Shimaoka
  • Hirofumi Nakayama
  • Saeed Afsari Fard
Article
  • 213 Downloads

Abstract

A massive earthquake of magnitude Mw 7.3 shook Kermanshah Province in Western Iran along the Iraqi border on November 12, 2017. The epicenter of the earthquake was approximately 10 km southwest of Ezgeleh Town in Kermanshah Province. Field observations almost 4 months after the disaster indicated that the earthquake had caused tremendous damage to most structures in both urban and rural areas, and left an enormous amount of disaster waste. To investigate the status of the damage and disposal of the disaster waste, remote sensing was conducted using an unmanned aerial vehicle (drone). Through the capture of low-altitude images by drone and the generation of 3D models, the quantity of debris accumulated in a waste disposal facility near Sarpol Zahab was estimated at approximately 480,000 m3. A compositional analysis of the disaster waste was performed using an imaging technique. This revealed that the disaster waste was largely composed of concrete (39.6%), hollow brick (35.4%), and gypsum (21.2%) in the urban area, whereas soil was the dominant component (77.4%) in the rural area. The damage caused to most buildings was essentially due to their non-standard construction. The management of debris from the damaged buildings was a critical issue for the authorities, and the lack of preparedness was a serious drawback that consumed an enormous amount of time, budget, and workforce. A practical post-disaster preparedness plan would help the decision-makers and the public to manage the otherwise overwhelming nature of the post-disaster conditions in a more reasonable manner.

Keywords

2017 Iran Earthquake Ezgeleh–Sarpol Zahab Disaster waste disposal Waste compositional analysis Remote sensing 

Introduction

Over the past few decades, the occurrence and intensity of climatic and geological hazards have increased significantly. Earthquakes are one of the most devastating hazards, typically hitting an extensive geographic area and leaving an enormous amount of debris from the partial or complete destruction of buildings, roads, bridges, and infrastructure in a very short period of time. Earthquakes may also trigger secondary hazards, such as tsunamis, landslides, land subsidence, liquefaction, site amplification, lateral spreading, and other geohazards.

Earthquakes of different magnitudes often strike the Iranian Plateau, which is entirely located at the heart of an active seismic zone. On Sunday, November 12, 2017, an earthquake of Mw 7.3 struck the western part of Kermanshah Province in Western Iran (Fig. 1) along the northern border of Iraq at 18:18 UTC (21:48 Iran Standard Time) [1]. The epicenter of the earthquake was at 34.77° N and 45.76° E with a focal depth of approximately 19 km [1]; the shallow depth caused tremendous damage to the western part of the province1,2. The epicenter was approximately 10 km southwest of Ezgeleh, 33 km northeast of Qasr-e-Shirin, and 39 km northwest of Sarpol Zahab [2]. Sarpol Zahab was the largest settlement near the epicenter in which the majority of deaths and damage occurred. According to a seismic hazard assessment of the Kermanshah–Sanandaj region of western Iran [3, 4], the recent earthquake could have occurred in the most active segments of the Zagros Main Recent Fault.
Fig. 1

Location of Kermanshah Province in Western Iran (right), and the detailed administrative division map of Kermanshah Province in which the earthquake epicenter is marked (left)

The earthquake was the deadliest of 2017 and was felt as far away as Turkey and Pakistan3; it was worse than the powerful September 19, 2017 earthquake in Mexico, which killed 369.4 According to the latest reports, the death toll rose to over 660, of which 518 were from the city of Sarpol Zahab.5 The earthquake was preceded by a foreshock of Mw 4.3, which alerted the community to leave their buildings, thereby potentially reducing the number of fatalities. The Iranian Seismological Center reported other smaller foreshocks that occurred in the epicentral area.6

In a preliminary report, the Kermanshah Governorate announced an estimated damage cost of 26 trillion Iranian Rials (IRR) (ca. 84 million US$) to residential, administrative, and infrastructure facilities in the province. This was proportioned as some 8 trillion IRR (ca. 26 million US$) of damage to residential, service, and commercial units of the province, and approximately 18 trillion IRR (ca. 58 million US$) of damage to government buildings and infrastructure. It needs to be emphasized that these figures were only preliminary estimates in the early days after the disaster.

Regarding the damage to the 31,000 urban and rural residential units, 100% damage to at least 12,000 rural residential units and 5000 urban residential units was reported across the province. The number of damaged governmental units, including schools, offices, hospitals, and infrastructure, including water and electricity, was notably high7,8. In addition, critical water and power shortages were reported across Kermanshah Province. Piped water to the areas closest to the epicenter was cut off, and the water sources in Sarpol Zahab were contaminated.9 The earthquake also affected regions of Eastern Iraq; however, its impacts on Western Iran were more detrimental than those on the Iraqi side.

Direct observation plays a substantial role in understanding the features of a disaster and the dominant phenomena affecting the people, infrastructure, and natural environment. Fan et al. [5] suggest that the disaster damage assessment process is crucial for rescue and relief operations, providing a basis for developing a recovery and reconstruction plan in the disaster region and helping the authorities to prevent and mitigate future disasters. One of the most significant consequences of a disaster is the generation of an enormous amount of debris. In the past few decades, there have been efforts to assess the status of damage [6, 7] and the impact of natural disasters on waste generation [8], through direct observation and/or remote sensing.

Scientists from various geographic regions including Turkey [9, 10], Iran [11, 12, 13], and New Zealand [14, 15, 16] have suggested strategies for the sound disposal and recycling of wastes from natural disasters, particularly earthquakes. They have emphasized the identification, classification, management, recycling, and disposal of disaster debris as a key factor for rapid recovery and reconstruction. Remote sensing has contributed significantly to damage assessment and has proved to be an efficient technology for this purpose [5]. Several research groups from across the globe have systematically used this technology for precise damage and recovery assessment [17], risk management [18], disaster management and monitoring [5, 7], structural damage assessment [8], and building damage mapping [19, 20].

In order to evaluate the status of damage, disaster waste disposal, and its direct impact on the surrounding environment, a team composed of three members from Kyushu University departed for the region to conduct a field investigation on March 15 and 16, 2018, almost 4 months after the disaster. Time constraints, budget management, and administrative procedures, as well as necessary preparations for an overseas trip, were the main reasons for the delayed visit to the disaster area. The present investigation only focuses on the consequences of the earthquake in Kermanshah Province (Fig. 1).

The prime objective of this investigation was to observe the earthquake-induced damage and to evaluate the status of waste and debris disposal in the disaster area. The next objective was to use the following techniques in the field: (1) evaluate the capability of unmanned aerial vehicle (UAV) technology to capture images from the damaged zones and estimate the amount of accumulated disaster debris based on the generated 3D models, and (2) employ an innovative photography technique for on-site quantification of debris left from the disaster.

Earthquake mechanism and seismic features

In Western Asia, no fewer than four major tectonic plates (Arabian, Eurasian, Indian, and African) and one smaller tectonic block (Anatolian) are responsible for seismicity and tectonics in the Middle East and the surrounding region. The geologic development of the region is a consequence of a number of first-order plate tectonic processes, which include subduction, large-scale transform faulting, compressional mountain building, and crustal extension [3].

The 2017 earthquake in Western Iran occurred at the collision boundary between the Arabian and Eurasian plates, in the Zagros Mountains (Fig. 2). The thrust mechanism of this event is compatible with a fault, which accommodates the convergence of these plates. The shallow depth of its hypocenter explains the intensity of the shaking, as well as the possibility of other ground failures, such as landslides and ground subsidence, which increased the damage caused.
Fig. 2

Approximate location of the 2017 earthquake that occurred near the collision boundaries between the Arabian and Eurasian plates

(source: https://www.emsc-csem.org/Earthquake/260/lang=en)

To elucidate the focal mechanism of this earthquake, Zare et al. [21] proposed that the earthquake occurred owing to the activity of a fault dipping shallowly to the east-northeast or a fault dipping steeply to the southwest. Based on the active fault map of Iran, they suggested that this earthquake was presumably triggered by the movement of Zagros Mountain Front Fault (MFF) in the Pol-e-Zahab Region. Farzanegan et al. [2] also suggested that the earthquake had a thrust focal mechanism. Such earthquakes generate stronger shaking compared with earthquakes that occur because of normal or strike-slip faults.

From the epicenter and the focal mechanism of the earthquake, it was also reported that the fault responsible for the earthquake was probably one of the segments of the MFF with a north–northwest direction, dipping shallowly (10°–15°) to the east [1]. The very low angle faulting might have been responsible for the broad width of the aftershock zone. In addition, the existence of scattered clusters of aftershocks indicates that they might have been associated with the movement of a group of pre-existing minor faults that were activated by the earthquake. Despite the generally scattered distribution of the aftershocks, it is possible to define a NW–SE trend in them, which is consistent with the trends of the MFF and the High Zagros Fault (HZF) [1].

The preliminary results of geodetic source modeling suggest that the earthquake was generated by blind ENE oblique-thrust faulting, with an average slip of approximately 4 m at a depth of between 17 and 22 km [22]. Based on interferometric SAR (InSAR) analysis, Kobayashi et al. [23] proposed a deformation pattern in which a reverse slip occurred on a low dip angle fault plane, with a NNW–SSE strike direction. They also suggested that the seismic motion might have triggered many localized displacements at the mountain slopes. The IIEES [1] reported that the earthquake had at least three foreshocks, with magnitudes between 1.9–4.5, and 526 aftershocks with magnitudes of 1.8‒4.7 until November 28, 2017, 17 days after the occurrence of the main shock.

In the most recent analysis, Tavani et al. [24] indicated the coexistence of thin- and thick-skinned thrusting, the reactivation of inherited structures, and the occurrence of weak units promoting heterogeneous deformation within the Paleo-Cenozoic sedimentary cover and partial decoupling from the underlying basement. According to their study, the main shock of the November 2017 seismic sequence was located within the basement along the low-angle MFF system of the Zagros Mountains.

Scope of the study

Study area and field investigation

This study focuses on exploring the circumstances in the aftermath of the November 2017 earthquake in Western Iran. The study area was selected based on the major zones that were greatly affected by the earthquake. Accordingly, Sarpol Zahab and its neighborhood, as well as several communities to the north of it (Fig. 3), were the focus of the field investigation, which was carried out on two consecutive days in March 2018. During the site visits, the team was accompanied by an expert who was quite familiar with the geography and geology of the region, as well as the disaster itself, since its occurrence in November 2017. Based on the available reports, online sources, verbal discussions, and recommendations from the local guides, the team visited several sites as shown in Fig. 3.
Fig. 3

Study area and the locations visited (in sequence) including the human settlements severely affected by the earthquake: 1-Shiroudi Housing Complex in Sarpol Zahab, 5-Emam Abbas, 6-Tapani, 7-Koeeke, 8-Mela Kabod, and 9-Gorchi Bashi Villages, and the disaster waste disposal sites (2, 3, and 4).

Source: Google Map

The entirety of the evidence indicated that the major damage had occurred in a north–south direction, extending from Ezgeleh (near the epicenter) in the north to Sarpol Zahab and its suburbs in the south, very close to and parallel with the Iran–Iraq border (Fig. 3). The team attempted to cover the most heavily affected areas during this field survey. On day 1 (March 15), the team primarily focused on Sarpol Zahab and its surrounding districts. Sarpol Zahab is the largest human settlement in this zone and the one that received the most severe impacts from the earthquake, with entire buildings undergoing partial or complete collapse [6, 25]. The team also visited several dumping sites, including the main disposal site in which the disaster waste from Sarpol Zahab and the surrounding communities was accumulated.

On day 2 (March 16), the team visited several communities and locations that had received severe damage from specific earthquake-induced phenomena such as landslides. This survey mission started from one of the nearest villages to the epicenter, Emam Abbas in the north, and extended to Tapani, Koeeke, Mela Kabud, and Ghorchi Bashi in the southern regions. All of the visited areas are located north of Sarpol Zahab City (Fig. 3).

Methodologies

The principal goals of this investigation were to observe the status of damage and the ongoing operations for handling and disposing of disaster debris, and to conduct field measurements. In order to observe the damage and calculate the amount of disaster waste at a disposal site near Sarpol Zahab, an aerial survey was conducted using a UAV or drone (DJI Phantom 4 Pro). The drone was controlled by DJI GS Pro Software, which was installed on an iPad 4 mini device. For still image shooting, the flight altitude was set to 30 m, and the overlap ratio between the images was set to 80–90% front and 60% side. The images captured were later imported to Agisoft Photoscan v.1.2.5 for 3D modeling and estimation of the disaster waste. In order to estimate waste volume at the disposal site, the Structure-from-Motion (SfM) technique was used. SfM is a photogrammetric imaging technique for generating three-dimensional structures from two-dimensional image sequences [26, 27].

In order to quantify the components of disaster debris categories, a photographic compositional analysis method was adopted from the technique that was initially employed in Japan following the Great East Japan Earthquake and Tsunami of 2011 [28]. Based on this technique (with slight modifications), the compositional ratio of the disaster waste was estimated at several waste piles in Shiroudi Housing Complex and Emam Abbas Village.

Accordingly, the following steps were carried out in the field and in the laboratory. In the field: (1) several debris piles were spotted at each site; (2) a relatively flat surface was selected on each pile; (3) a wide measuring tape carried to the field was placed over the flat surfaces; and (4) several images of the piles were captured by a digital camera, from a distance of 1–1.5 m perpendicular to the designated region of interest (ROI). In the laboratory: (1) images from at least three piles were selected as representatives; (2) the selected images were divided into a 6 × 16 grid with 5 cm grid spacing, resulting in 96 cells; (3) each cell was marked with a circular symbol in the center; (4) the debris immediately below each symbol was deemed to be the dominant waste category in that particular cell; and (5) the total number of each waste category was counted and used in calculating the compositional ratios. Equation (1) was used to calculate the average compositional ratio (ACR) of each waste category:
$${\text{ACR}}\,(\% ) = \mathop \sum \nolimits {\text{count}}/\mathop \sum \nolimits {\text{total}} \times 100.$$
(1)

Impact of the earthquake on the built environment

Sarpol Zahab City

Because Sarpol Zahab suffered the majority of deaths and damage, the city was investigated in detail. The population of Sarpol Zahab was estimated at 45,481 in 2016.10 The city extends in an approximately NW–SE direction, and is divided into two segments by the Kermanshah–Qasr-e Shirin main road (Figs. 3, 4). It was observed that the city is bordered by low-height rocky outcrops on the NE edge, and extends towards the agricultural flatlands in the south, southwest, and west. A narrow perennial river runs approximately E–W through the northern part of the city (Fig. 4).
Fig. 4

Satellite image of Sarpol Zahab City. The location of the Shiroudi Housing Complex is marked on the image (see text for details).

Source of image: Google Earth

The buildings and city infrastructure in the southern part received more serious damage from the ground motions than those in the northern areas (nearest to the rocky outcrops). According to Ozcelik [29], the earthquake forces during the main shock are amplified by the poor dynamic characteristics of alluvial soils. This phenomenon was presumably a major factor causing additional damage to the buildings in the southern segment of the city.

In Sarpol Zahab, nearly the entire population was evacuated from their dwellings and resided in isolated, temporary refugee tents and prefabricated portable containers inside and outside the city. Figure 5 provides an oblique aerial view of the Shiroudi Housing Complex in the NW of Sarpol Zahab (see Fig. 4) and the temporary shelters (left) that were captured on site using the drone at an altitude of 30 m. With the exception of a sports complex, there were no other public facilities in the city or its suburbs to provide shelter for the evacuees.
Fig. 5

An oblique-angled view of the damaged buildings at the Shiroudi Housing Complex in NW of Sarpol Zahab and the isolated temporary shelters (left). The image was taken by a drone from an altitude of 30 m (view to the NE)

The structures in the city can be categorized into steel, concrete, and masonry (or non-engineered) types. The masonry structures were essentially made of adobe (mixed with hay and straw), brick clay, hollow cement blocks, or stones. The masonry buildings were the most frequent types, both in the city (Fig. 6a) and the neighboring rural areas. A significant percentage of the masonry buildings suffered total structural failure, which was the main cause of the casualties [30]. Even newly built and multistory buildings (Fig. 6b), and those which were under construction (either steel or concrete types) suffered severe damage, possibly due to improper design, structural vulnerabilities, weak foundations, poor engineering, lack of bracelets, poor quality concrete, inappropriate reinforcement, or ground failure.
Fig. 6

a Severe collapse of a masonry building, b partial collapse and settlement of a multistory steel frame building, c a house with intact façade and interior damage, and d progressive collapse of a concrete frame construction (left) next to a building (right) with concrete frame and hollow brick walls that survived in the earthquake

As a result, the majority of the buildings in the city became uninhabitable owing to a series of major to minor failures, although some looked undamaged in their exteriors and façades. Figure 6c shows a two-story private housing building that displayed almost no defects on the front face but became unlivable owing to internal damage, including fractured ceilings and tilted columns. In rare cases, however, it was possible to find structures that exhibited relatively successful performance. Figure 6d is an outstanding image that displays a building with a concrete frame structure and walls made of hollow blocks, which survived the earthquake (right), beside a large concrete structure (left) that suffered from a presumably progressive collapse. When a primary structural element fails, a building undergoes progressive collapse, resulting in the failure of adjoining structural elements, which in turn, causes further structural failure [31].

The city suffered from the partial or total collapse of public and governmental buildings, including the city governor’s office and the main hospital of the city. Such facilities must be able to maintain their full functionality for post-disaster relief operations. The city hospital, which had been established in 2008, was seriously damaged, owing to several structural and non-structural failures, and became fully inoperative after the earthquake (Fig. 7). The collapse of its structure led to the immediate death of at least 15 patients at the hospital.11 Presumably, if the hospital structure had been built to the national standard code [32], it would most likely have saved the lives of its inpatients at the time of the earthquake and many others who subsequently required hospital treatment as a consequence of the earthquake.
Fig. 7

a A view of the general hospital of Sarpol Zahab taken during the field survey, and b the interior damage of the hospital after the earthquake (photo courtesy of https://www.alef.ir)

Case study: The Shiroudi Housing Complex

The Shiroudi Housing Complex, which is located in the northwest corner of Sarpol Zahab (see Fig. 4), is the largest housing complex in the city, and is composed of several public buildings (seven stories) and a series of private houses (two to four stories). They are essentially concrete structures and were mostly occupied by their tenants at the time of the disaster. The government had completed the public housing for lower income families approximately 5 years before the earthquake. The earthquake caused serious casualties in this housing complex, leading to the deaths of at least 100 and serious injury of hundreds more.

Although these buildings had been constructed in recent years, they failed to withstand the devastating tremors of the earthquake. Many instances of total or partial collapse of the constructions were observed on site. Figure 8a displays the status of the public buildings 1 day after the disaster, showing that most of the exterior walls (particularly in the lower to middle floors) had collapsed, while the concrete frames of the buildings remained intact. Alavi et al. [30] linked these failures to poor construction quality, inappropriate connections between wall and structure, and the absence of wall posts. As can be seen in Fig. 8a, many other non-structural components, such as doors, windows, handrails, verandas, and cooling systems, had fallen as a result of being poorly affixed. According to local reports, such non-structural failures were responsible for most of the fatalities in this housing complex.
Fig. 8

a Status of multistory public housing at the Shiroudi Housing Complex one day after the disaster on Nov. 13, 2017, and b status of the buildings at the time of field survey (March 15, 2018)

Consequently, the entire area was uninhabitable, and only demolition and debris removal activities were in progress at the time of the survey. It was noted that the removal of the exterior walls and other non-structural components of the multistory buildings had been completed (Fig. 8b). There are plans to reinforce the existing frames of the public buildings and reconstruct them for future use. However, the privately owned houses of the Shiroudi Complex that were severely damaged had not been demolished at the time of the survey, possibly owing to legal issues and lack of budget.

In general, Alavi et al. [30] suggest several reasons for the instability of the concrete structures at this location and in the other zones across the disaster area. These include the following: inappropriate reinforcement of the joints and the critical ends of columns and beams; lack of transverse reinforcement; shear failure of short columns; poor quality of construction; inadequate length or location of splices; poor compressive strength of concrete; soft-story effect; and weak-column/strong-beam mechanism. Researchers from the countries neighboring Iran, including Pakistan [33] and Turkey [34, 35], reported similar structural failures caused by heavy earthquakes in their regions.

Apart from the direct impact of the earthquake on the structures, it is presumed that the insufficient strength of the loose agricultural soil amplified the tremors in this area as a site effect.12 Liquefaction was reported in the vicinity of this housing complex [1], though the evidence had disappeared by the time of the field investigation owing to rainfall. However, remnants of long fractures were observed in the ground just a few meters from the buildings (Fig. 9). Such phenomena emphasize the critical role of geotechnical evaluation of the soil before any construction project in minimizing the adverse effects of earthquakes.
Fig. 9

Ground failures (tip of arrow) in the loose agricultural soil, which were triggered by the earthquake in the neighborhood of the Shiroudi residential buildings

Emam Abbas Village

Emam Abbas was one of the nearest communities to the earthquake epicenter (Fig. 3) with four inhabitants dead. Almost no buildings could withstand the ground tremors in the village; only two buildings were undamaged, one of which was the village health center (Fig. 10), which had been built almost 25 years before the earthquake happened. This building was built with reinforced concrete (RC) on a thick concrete foundation. This indicates that the structures that were built to safety standards were robust against the earthquake, regardless of the age of the structure. Approximately 250 inhabitants lived in the village at the time of the visit, most of whom were accommodated in temporary shelters (tents and prefabricated portable containers). However, reconstruction had already begun in some places.
Fig. 10

Successful performance of a concrete structure at Emam Abbas Village, which had been built about 25 years before the 2017 earthquake

Tapani Village

Although this village was farther from the epicenter (Fig. 3), it had also suffered from very severe damage, based on field observation. There was no intact structure in the village, and almost everything was destroyed. The original population of the village was 535, of which approximately 16 were killed. The village was in a state of abandonment, and reconstruction was rare. Most of the residents had been evacuated to the temporary shelters in the neighboring villages.

Koeeke Villages

This is the largest rural community near the earthquake epicenter (Fig. 3) and consists of four neighboring villages: Hassan, Azizi Amin, Majid, and Mahmoud. The villages are located in the foothills of Mt. Shahneshin (Fig. 11a), which forms an anticline and is the main source of water for the agriculture in the region. The entire community is a subsidiary of Sarpol Zahab, with a population of over 1500 inhabitants, of which approximately 100 were reported dead after the earthquake. The villages had been vastly damaged, and most of the people were living in temporary shelters. Reconstruction projects were more noticeable in this area than in the other villages.
Fig. 11

a An aerial view of Mt. Shahneshin and the surrounding communities. The rectangle on the image approximately demarcates the outline of the image (b), and b a massive landslide that occurred shortly after the earthquake on the flanks of Mt. Shahneshin. Both Mela Kabod and Ghorchi Bashi Villages are located immediately before the toe of the landslide.

Base map: Google Earth

Mela Kabod and Ghorchi Bashi Villages

These small villages were not directly affected by the earthquake; however, a huge mass movement (landslide) triggered by the earthquake partially damaged some dwellings and endangered the entire settlements. The landslide occurred shortly after the earthquake on the flanks of Mt. Shahneshin (Fig. 11b). Rock and soil were displaced from their original position by a huge mass movement to the west, along an approximately 4 km rupture that displays a NW–SE direction (Fig. 11b). Both Mela Kabod and Ghorchi Bashi Villages (Fig. 11a) are located immediately before the “toe” of the landslide.

Figure 12a, b shows the direct collision of the landslide with the houses in Mela Kabod and Ghorchi Bashi Villages, respectively. The “main scarp” of the landslide can be clearly seen from the villages and even from Sarpol Zahab. The movement of soil and rock occurred over several hundreds of meters, down from the “crown” of the landslide to the west (Fig. 11b). It not only caused damage to the houses but also led to devastation of the farmlands and apiculture in both villages. Apart from the direct impact of the landslide on the buildings, its secondary risk was incredibly high, as it had the potential to move forward because of other natural phenomena, such as minor tremors, torrential rains, or simply gravity. Transverse and radial cracks/fissures, ranging from the micro- to macro-scale, were abundantly observed on the dislocated ground. Figure 12c displays a dense network of radial cracks in the landslide that overlooks Ghorchi Bashi Village. These lands were formerly used for agricultural purposes.
Fig. 12

Direct collision of the landslide with the buildings at a Mela Kabod, and b Ghorchi Bashi Villages. Long and deep radial cracks are observed in the body of the landslide overlooking Ghorchi Bashi Village (c)

Such significant ground failures and the magnitude of the landslide should be taken as warnings that the landslide might have immense potential for further movement at any time in the future, wiping out the surrounding settlements. In such circumstances, the authorities must prioritize the instant evacuation of the inhabitants in order to prevent them from any potential threat. The populations of Mela Kabod and Ghorchi Bashi are 100 and 240, respectively, which are not large for an urgent evacuation. Further geotechnical, geological, geophysical, hydrological, and remote sensing investigations should be carried out to evaluate the stability of the landslide and its future impacts on the human settlements, environment, agriculture, and hydrological regime of the region.

Disaster waste analysis

Identification of the quality and quantity of disaster waste can play a crucial role in post-disaster damage and recovery assessment [14, 15, 16]. As well as direct observation, remote-sensing-based investigations and associated technologies (using satellite data and UAV images) are vital assessment tools that have significantly contributed to the rapid identification of structural damage and post-disaster operations [5, 7]. Remote sensing technology can play a particularly significant role in the collection of rubble from demolished buildings, as well as the management of disaster debris. Based on these facts, we were determined to observe the status of waste disposal in the disaster area and to assess the capability of UAV technology and an on-site imagery technique as a tool for direct measurement of disaster waste.

Status of disaster waste disposal in the earthquake-affected regions

This earthquake affected a vast geographic area in Kermanshah Province (Western Iran) and consequently damaged several urban and rural areas across the region, including Sarpol Zahab (the largest settlement) and many other smaller communities. The direct observations revealed that the disaster had left an enormous amount of debris that originated from the completely or partially collapsed structures. Under such critical circumstances, the local government authorities generally lose control over public affairs, and exhaust workforce, organizational infrastructure, administrative integrity, field equipment, and resources. Therefore, handling the huge amount of waste generated from the disaster and subsequent relief operations would be completely beyond their capacity. In particular, when there are no national or local guidelines for waste management in emergency conditions, the situation becomes even more complicated, as was the case in this area. The status of disaster waste handling and disposal in a number of locations is described below.

Sarpol Zahab City and its suburbs

Because the number of damaged buildings in this area was high, both demolition and debris removal processes were in progress slowly. In Sarpol Zahab, the debris generated from the buildings was temporarily left on the streets next to the demolished buildings (Fig. 13a) before being carried to the final disposal site. Another issue related to waste handling was illegal dumping sites. While visiting the outskirts of the city, we noted several spots along major roads where debris had been dumped. Figure 13b displays piles of disaster waste that had accumulated adjacent to roads in Rikhak Olia, near Sarpol Zahab (location 2 on Fig. 3).
Fig. 13

Temporary accumulation of disaster debris at different locations: a Sarpol Zahab, b along a road in Rikhak Olia near Sarpol Zahab, c Emam Abbas Village, and d Koeeke Village

Disposal of waste and debris in unauthorized areas, including roadsides and farmlands, was encountered not only in this area, but also across the entire region, causing serious problems for the local authorities. Another location visited was an open dumping site SW of Sarpol Zahab (location 3 on Fig. 3), which regularly receives solid waste from household sources. In the aftermath of the earthquake, however, it became a destination for illegal (often nocturnal) dumping of disaster debris.

Apart from the aforementioned locations, the local authorities had designated a disposal facility in the mountains almost 10 km west of Sarpol Zahab (location 4 on Fig. 3) as a final destination for the majority of the disaster debris that originated from the city and the surrounding communities. At the disposal facility, mixed wastes were accumulated. The composition of the debris was principally soil and sand, stone, ceramics and tiles, glass shards, concrete, brick, hollow brick, and plaster (gypsum). Steel bars and other metal items were separated at the source (demolition sites) by private sector operators as valuable and recyclable construction materials; hence, they were not (or seldom) brought into this disposal facility.

The Provincial Support Headquarters (PSH) of various provinces (locally called Setad Moien Ostan), affiliated with the Iranian Housing Foundation, were involved in debris removal operations. Personnel and equipment from the PSH of Isfahan, Mazandaran, and Khorasan Razavi Provinces were responsible for transferring debris from Sarpol Zahab to the disposal facility. In parallel, the PSH of Hamedan, Markazi, and Ardebil Provinces were involved in transferring debris from the rural areas to this facility. The on-site personnel stated that apart from the disaster debris, combustible waste and household appliances were previously brought to the facility. It was intended that the facility would accept disaster debris until July 2018.

Emam Abbas and Tapani Villages

The debris left from the disaster was mostly scattered inside the village next to the damaged buildings, without separation or treatment (Fig. 13c). A significant amount of waste was also accumulated in the agricultural lands outside of the village or along the main road that connects the village to Ezgeleh. It was not clear whether this place was the final disposal site or whether the debris would be subsequently moved to a different location. The debris was essentially composed of soil and sediments, chunks of bricks and hollow blocks, stone, and concrete. Steel and other metal components were generally separated, with some sold to local customers for reuse. In Tapani Village, the situation was similar. The disaster debris had been left across the village or next to the destroyed buildings, and no disposal site was observed in this village.

Koeeke Villages

Considerable amounts of debris were distributed across these villages. In addition, a large volume of debris was laid on the foothills of Mt. Shahneshin as a final disposal site (Fig. 13d). This site is topographically located above the village, and it is believed that this location could enhance the risk of additional disasters, such as flooding during the rainy season, or airborne dust caused by the strong wind. Flash flooding could wash the loose disaster debris downslope and cause a secondary catastrophic event for the residents. The debris was compositionally similar to the other villages, and remained untreated. It largely consisted of soil and sediments, bricks, hollow blocks, stone, and concrete.

Aerial survey and 3D modeling of the final disposal facility

Scientists have attempted to use remote sensing technologies in order to identify and estimate quantities of post-disaster debris/waste for the purpose of designing a rapid recovery process [5, 7, 17, 18, 36]. To this end, data from various sources (satellites or UAVs) have been efficiently processed to evaluate the status of the damage and debris generated [8, 19, 20].

A DJI Phantom 4 Pro drone was used on site to capture video and still images in order to detect the shape and periphery of the final disposal facility. These data were used to create 3D models using the Structure-from-Motion (SfM) method, which was recently employed for estimating the amount of disaster waste in Japan [36]. Figure 14a displays an oblique aerial view of the disposal facility, working face, access road, and the active machinery at the time of the survey. Figure 14b presents the 3D model of the facility from almost the same angle as Fig. 14a. The facility presents a fan-shaped structure consisting of an upper and a lower slope. The former was the active zone of the facility, and the latter was presumably the passive zone, which also demarcates the border of the disposal site.
Fig. 14

a An oblique-angled view of the disposal facility, working face, access road, and active machinery captured by drone from an altitude of approximately 30 m, and b the 3D model of the disposal facility in which an upper (1) and a lower (2) slope can be obviously identified

The original topography of the land beneath the disposal facility consisted of several small V-shaped valleys surrounded by ephemeral watercourses (Fig. 14b). It occupied a large piece of land where a considerable amount of disaster waste was accumulated. By processing the 3D model, the quantity of waste accumulated on site was estimated at approximately 480,000 m3. However, as the accurate topography of the original ground was unknown at the time of the aerial survey, the disaster waste volume might have been underestimated. The personnel on site proposed that the height of the active zone (top surface) of the disposal site from the base (original topography) was approximately 70 m; however, it was estimated at 42 m by analyzing the 3D model. In such circumstances, a combination of ground mapping and aerial surveying would suggest more accurate and reliable results for the volume of waste deposited.

Compositional analysis of the disaster waste

An innovative and practical imagery technique [28] was tested in the field in order to carry out the compositional analysis and to quantify waste categories at several waste piles. It was realized that the composition of the debris from the demolished buildings varied from place to place. However, it was possible to divide it generally into the two categories of adobe (mud) and concrete-brick materials. The former was the dominant composition of the buildings in the rural areas, while the latter was largely derived from buildings in the urban areas. Figures 15 and 16 represent the types of debris derived from the demolished buildings of an urban area (Shiroudi Housing Complex in Sarpol Zahab) and a rural area (Emam Abbas Village), respectively. The images present the overall status of the disaster debris at each location.
Fig. 15

Basic gridding and counting method for measuring the compositional ratio of various components of disaster debris in an urban area, Shiroudi Housing Complex, Sarpol Zahab City (analytical data: pile 3 in Table 1)

Fig. 16

Basic gridding and counting method for measuring the compositional ratio of various components of disaster debris in a rural area, Emam Abbas Village (analytical data: pile 3 in Table 2)

Tables 1 and 2 present the ACR of disaster waste at Shiroudi Housing Complex and Emam Abbas Village, respectively. Data from three piles were used to calculate the related ACR at each site, according to Eq. (1). The majority of the waste categories at Shiroudi Housing Complex (Fig. 15) were concrete (39.6%), hollow brick (35.4%), and gypsum (21.2%) on average (Table 1). This procedure was also conducted for disaster waste piles at Emam Abbas Village (Fig. 16). The statistical results were significantly different on this site, because houses in this village (and most rural areas) were traditionally built with a combination of muddy soil, straw, and reed. As a result, soil was the dominant component (77.4%), as shown in Table 2. Tables 1 and 2 present the average compositional analyses of three waste piles from Shiroudi Housing Complex and Emam Abbas Village, respectively.
Table 1

Average compositional ratios of disaster debris components in three waste piles, Shiroudi Housing Complex, Sarpol Zahab City

The photograph of pile 3 and the gridding method is shown in Fig. 15 as a representative

Table 2

Average compositional ratios of disaster debris components in three waste piles, Emam Abbas Village

The photograph of pile 3 and the gridding method is shown in Fig. 16 as a representative

The aforementioned technique provides a cornerstone for rapidly estimating the total volume of the disaster waste, its components, and the compositional ratio of each waste category. However, weight conversion factors (e.g., density of each category of waste, compaction status, and homogeneity of the waste piles) would be required for precise estimation of the total amount (weight) of the disaster waste. Such simple and practical techniques provide a valuable opportunity to local authorities, policy-makers, and engineers, helping them adopt appropriate post-disaster waste management policies, including waste hauling and transfer, reuse, recycling, and disposal.

Lessons to be learned

On November 12, 2017, a massive earthquake (Mw 7.3) rocked Kermanshah Province in Western Iran, very close to the Iraqi border, leaving hundreds of causalities, thousands injured, and enormous damage in the epicentral area. Following a visit to the earthquake-hit area, a number of lessons were learned, which are summarized below:
  1. 1.

    Field observations indicated that tremendous damage had occurred to both the residential and administrative buildings across the city of Sarpol Zahab, the largest community in the region. For instance, the governor’s office, which operated as the command center for most activities across the region, was significantly damaged. However, because of the emergency conditions and despite severe aftershocks, the building was in use by the governor’s office employees. As another instance, the city hospital largely collapsed when the earthquake occurred, and its medical services were completely suspended in the early stages of the disaster. The collapse of the hospital caused significant injuries and deaths to the personnel and inpatients. Achour et al. [37] reported the satisfactory performance of a hospital following the 2011 Van Earthquake in Turkey, which effectively withstood the tremors. Such successful cases indicate that designing and building a structure according to updated building codes can save the structures and their residents. Therefore, in a seismic hot spot like Iran, developing and enforcing seismic codes would be the most reliable policy to minimize the losses caused by an earthquake.

     
  2. 2.

    Almost 4 months after the earthquake, management of debris from the damaged buildings as well as demolition and cleanup processes were among the most critical issues for authorities across the region. The disaster waste was a barrier to clean-up activities and reconstruction, as it blocked the roads. In addition, municipal waste and waste arising from relief operations were either mixed with or accumulated on top of the disaster waste. It was observed that the unmanaged disaster waste was a local source of airborne dust, which sometimes intensified during the dry and during windy periods, leading to an adverse impact on the environment, citizens, and agriculture. The entire region is a hub for several major agricultural products, particularly wheat, barley, maize and oilseeds, which are the main sources of income for most of the inhabitants.

     
  3. 3.

    The composition of disaster debris was significantly different depending on its source. Compositional analysis of the disaster waste in a typical urban area (Shiroudi Housing Complex) and a typical rural area (Emam Abbas Village) indicated that debris derived from recently developed urban areas was essentially composed of concrete-brick materials, while in the rural areas, soil and mud were the dominant components. In either situation, only a few buildings were observed that had withstood the earthquake shocks, possibly owing to their standard design and construction (Fig. 10).

     
  4. 4.

    Construction and retrofitting of buildings according to safety guidelines, such as Standard 2800 [32], must be prioritized for public, multi-purpose facilities, including healthcare centers, fire and police stations, armed/defense forces buildings, governmental offices, educational facilities, and sports complexes. Such facilities play substantial roles in the emergency response and evacuation operations in the aftermath of a disaster. It was observed that such facilities in Sarpol Zahab received almost the same level of damage as the other weak structures across the city, most probably owing to substandard constructions. Therefore, taking special precautions to protect educational facilities and other public buildings would be an important measure for saving the lives of the inhabitants and offering safe and secure shelters to the victims in the post-disaster stage. As an example, public buildings, particularly sports facilities, effectively provided temporary shelter to evacuees after an earthquake that hit the Kumamoto area of Western Japan in 2016 [36].

     
  5. 5.

    Although the Western Iran earthquake was the deadliest in 2017,13 the number of injured residents and the death toll were not significantly high, despite heavy structural damage. The main factor that saved the lives of the residents was a series of foreshocks (the most powerful measuring Mw 4.3) that occurred shortly prior to the main shock, causing people to be on alert. In addition, the earthquake and its aftershocks occurred in a region where the communities were geographically scattered, and there was no large city in the immediate vicinity of the epicenter. Special care must be taken when interpreting statistics on such disasters, as it was the above factors that led to the reduced number of casualties, rather than the efficiency of the infrastructure or reinforcement of buildings. Therefore, in planning for future scenarios, the magnitude of this earthquake and its potential impacts should not be overlooked owing to the relatively small death toll.

     
  6. 6.

    A critical issue that was very evident was the lack of sufficient financial resources for demolition of collapsed buildings, debris removal, and retrofitting/reconstruction projects. This was a serious problem for the local authorities and for most of the residents who suffered from the earthquake. An effective measure to cover the aforementioned costs would be to offer an early national insurance program (either compulsory or optional) to the citizens in order to support their housing against natural disasters including earthquakes. Earthquake insurance can significantly reduce the vulnerability of a society and speed up post-disaster recovery operations.

     
  7. 7.

    Lack of preparedness among the local authorities had clearly led them to taking hasty decisions. It was evident that the difficult post-earthquake conditions caused an enormous waste of time, resources, and manpower. If the local authorities and citizens had been prepared in advance, public safety could have been guaranteed. Instead of waiting until a disaster occurs, any society needs to have a concrete and operational disaster-preparedness plan, drafted by lawmakers, engineers, scientists, and government executives. In addition, both central and local governments must give priority to planning for a systematic recovery from any disaster instead of diverting attentions from them.

     

Footnotes

Notes

Acknowledgements

The authors are grateful to Prof. Mehdi Zare of IIEES for providing impressive comments and suggestions before and during the field survey. Special thanks are due to the directorate of Kermanshah Provincial Government and Sarpol Zahab Governor’s Office for facilitating and supporting the site visits across the region.

References

  1. 1.
    IIEES. Report on November 12, 2017 Sarpol-e-Zahab earthquake, Kermanshah Province, 5th edn, vol 1: Seismological features, December 2017, International Institute of Earthquake Engineering and Seismology (IIEES), Tehran, Iran (in Persian). 2017.Google Scholar
  2. 2.
    Farzanegan E, Pourmohammad Shahvar M, Eshraghi A, et al. Report of the November 12, 2017 Sarpol-e Zahab, Kermanshah Province Earthquake. Road, Housing and Urban Development Research Center (BHRC), Iran Strong Motion Network. 2017. https://www.bhrc.ac.ir/Portals/18/Sare%20Pole%20Zahab%20Earthquake.pdf.
  3. 3.
    Jenkins J, Turner B, Turner R, et al. Seismicity of the Earth 1900–2010 middle east and vicinity (ver 1.1, Jan. 28, 2014): U.S. Geological Survey Open-File Report 2010-1083-K, scale 1:7,000,000. 2013. https://pubs.usgs.gov/of/2010/1083/k/.
  4. 4.
    Shabani E, Mirzaei N. Probabilistic seismic hazard assessment of the Kermanshah–Sanandaj region of Western Iran. Earthq Spectra. 2007;23(1):175–97.CrossRefGoogle Scholar
  5. 5.
    Fan Y, Wen Q, Wang W, et al. Quantifying disaster physical damage using remote sensing data—A technical work flow and case study of the 2014 Ludian earthquake in China. Int J Disaster Risk Sci. 2017;8:471–88.CrossRefGoogle Scholar
  6. 6.
    Günce HB, San BT. Measuring earthquake induced deformation in south of Halabjah (Sarpol-e-Zahab) using Sentinel 1 data on November 12, 2017. The 2nd International electronic conference on remote sensing (ECRS 2018), 22 March–5 April 2018, Sciforum Electronic Conference Series, vol 2. 2018.Google Scholar
  7. 7.
    Chou TY, Yeh ML, Chen YC, et al. Disaster monitoring and management by the unmanned aerial vehicle technology. In: Wagner W., Székely, B. (eds.): ISPRS TC VII Symposium – 100 Years ISPRS, Vienna, Austria, July 5–7, 2010, IAPRS, Vol. XXXVIII, Part 7B. 2010.Google Scholar
  8. 8.
    Cusicanqui J. 3D scene reconstruction and structural damage assessment with aerial video frames and drone still imagery. Degree of Master of Sciences in Geo-Information Science and Earth Observation, University of Twente, The Netherlands; 2016. p 69.Google Scholar
  9. 9.
    Baycan F, Petersen M (2002) Disaster waste management—C&D waste. In: Annual conference of the International Solid Waste Association, Istanbul, Turkey.Google Scholar
  10. 10.
    Baycan F. Emergency planning for disaster waste: a proposal based on the experience of the Marmara Earthquake in Turkey. In: International conference and student competition on post-disaster reconstruction “planning for reconstruction”, Coventry, UK; 2004.Google Scholar
  11. 11.
    Rafee N, Karbassi AR, Nouri J, et al. Strategic management of municipal debris aftermath of an earthquake. Int J Environ Res. 2008;2(2):205–14.Google Scholar
  12. 12.
    Askarizadeh L, Karbassi AR, Ghalibaf MB, et al. Management of post-earthquake construction debris in Tehran Metropolitan. Int J Environ Sci Technol. 2016;13:639–48.CrossRefGoogle Scholar
  13. 13.
    Askarizadeh L, Karbassi AR, Ghalibaf MB, et al. Debris management after earthquake incidence in ancient City of Ray. Glob J Environ Sci Manag. 2017;3(4):447–56.Google Scholar
  14. 14.
    Brown C, Milke M, Seville E, et al. Disaster waste management on the road to recovery: L’Aquila earthquake case study. In: 14 European conference on earthquake engineering, Ohrid, Macedonia; 2010.Google Scholar
  15. 15.
    Brown C, Milke M, Seville E. Disaster waste management: a review article. Waste Manag. 2011;31:1085–98.CrossRefGoogle Scholar
  16. 16.
    Brown C, Milke M. Recycling disaster waste: feasibility, method and effectiveness. Resour Conserv Recycl. 2016;106:21–32.CrossRefGoogle Scholar
  17. 17.
    Ghaffarian S, Kerle N, Filatova T. Remote sensing-based proxies for urban disaster risk management and resilience: a review. Remote Sens. 2018;10(11):1760.CrossRefGoogle Scholar
  18. 18.
    Ghaffarian S, Kerle N. Towards post-disaster debris identification for precise damage and recovery assessments from UAV and satellite images. In: The international archives of the photogrammetry, remote sensing and spatial information sciences, volume XLII-2/W13, 2019 ISPRS geospatial week 2019, 10–14 June 2019, Enschede, The Netherlands; 2019.CrossRefGoogle Scholar
  19. 19.
    Matsuoka M, Yamazaki F. Building damage mapping of the 2003 Bam, Iran, earthquake using Envisat/ASAR intensity imagery. Earthq Spectra. 2005;21(S1):S285–94.CrossRefGoogle Scholar
  20. 20.
    Maruyama Y, Yamazaki F, Matsuzaki S, et al. Evaluation of building damage and tsunami inundation based on satellite images and GIS data following the 2010 Chile earthquake. Earthq Spectra. 2012;28(S1):S165–78.CrossRefGoogle Scholar
  21. 21.
    Zare M, Kamranzad F, Parcharidi SI, et al. Preliminary report of M w 7.3 Sarpol-e Zahab, Iran earthquake on November 12, 2017; 2017.Google Scholar
  22. 22.
    Motagh M, Vajedian S, Behling R, et al. 12 November 2017 M w 7.3 Sarpol-e Zahab, Iran, earthquake: results from combining radar and optical remote sensing measurements with geophysical modeling and field mapping. Geophysical Research Abstracts, vol 20, EGU2018-10528-4, EGU General Assembly 2018; 2018.Google Scholar
  23. 23.
    Kobayashi T, Morishita Y, Yarai H, et al. InSAR-derived crustal deformation and reverse fault motion of the 2017 Iran–Iraq earthquake in the Northwestern Part of the Zagros Orogenic Belt. 2018. http://www.gsi.go.jp/common/000197807.pdf.
  24. 24.
    Tavani S, Parente M, Puzone F, et al. The seismogenic fault system of the 2017 M w 7.3 Iran–Iraq earthquake: constraints from surface and subsurface data, cross-section balancing, and restoration. Solid Earth. 2018;9:821–31.CrossRefGoogle Scholar
  25. 25.
    DRES Earthquake Report. A preliminary report on school buildings performance during M 7.3 Ezgeleh, Iran earthquake of November 12, 2017. Organization for Development, Renovation and Equipping Schools of I. R. Iran, Tehran, Iran; 2017.Google Scholar
  26. 26.
    Westoby MJ, Brasington J, Glasser NF, Hambrey MJ, Reynolds JM. ‘Structure-from-Motion’ photogrammetry: a low-cost, effective tool for geoscience applications. Geomorphology. 2012;179:300–14.CrossRefGoogle Scholar
  27. 27.
    Fonstad MA, Dietrich JT, Courville BC, et al. Topographic structure from motion: a new development in photogrammetric measurement. Earth Surf Process Landf. 2013;38:421–30.CrossRefGoogle Scholar
  28. 28.
    Iwate Prefecture. Detailed plan of disaster waste disposal processing in Iwate Prefecture. 2013. http://iwate-archive.pref.iwate.jp/wp/wp-content/uploads/2017/02/c8d9cda44d86faa7ee9c3f6e5ffcf008.pdf. (in Japanese).
  29. 29.
    Ozcelik M. Back analysis of ground vibrations which cause cracks in buildings in residential areas Karakuyu (Dinar, Afyonkarahisar, Turkey). Nat Hazards. 2018;92(1):497–509.CrossRefGoogle Scholar
  30. 30.
    Alavi E, Mahootchian A, Yadegari S, et al. Report of: M 7.3 Ezgeleh, Kermanshah, Iran Earthquake on November 12, 2017. 2018. https://www.eeri.org/wp-content/uploads/Kermanshah-Earhquake-Report-E.A.pdf.
  31. 31.
    Ellingwood BR, Leyendecker EV. Approaches for design against progressive collapse. J Struct Div. 1978;104(3):413–23.Google Scholar
  32. 32.
    Standard 2800. Iranian code of practice for seismic resistant design of buildings (Standard No. 2800). 3rd edn, Building and Housing Research Center (BHRC), Tehran, Iran; 2007.Google Scholar
  33. 33.
    Naseer A, Khan AN, Hussain Z, et al. Observed seismic behavior of buildings in northern Pakistan during the 2005 Kashmir earthquake. Earthq Spectra. 2010;26(2):425–49.CrossRefGoogle Scholar
  34. 34.
    Yön B, Sayin E, Calayir Y, et al. Lessons learned from recent destructive Van, Turkey earthquakes. Earthq Struct. 2015;9(2):431–53.CrossRefGoogle Scholar
  35. 35.
    Yön B, Sayın E, Onat O. Earthquakes and structural damages. In Earthquakes, Tectonics, Hazard and Risk Mitigation, Edited by Taher Zouaghi, IntechOpen; 2017.  https://doi.org/10.5772/65425. https://www.intechopen.com/books/earthquakes-tectonics-hazard-and-risk-mitigation/earthquakes-and-structural-damages.Google Scholar
  36. 36.
    Saffarzadeh A, Shimaoka T, Nakayama H, et al. Tasks and problems involved in the handling of disaster waste upon April 2016 Kumamoto Earthquake, Japan. Nat Hazards. 2017;89:1273–90.CrossRefGoogle Scholar
  37. 37.
    Achour N, Pascale F, Price A, et al. Learning lessons from the 2011 Van Earthquake to enhance healthcare surge capacity in Turkey. Environ Hazards. 2016;15(1):74–94.CrossRefGoogle Scholar

Copyright information

© Zhejiang University Press 2019

Authors and Affiliations

  1. 1.Faculty of Engineering, Department of Urban and Environmental EngineeringKyushu UniversityFukuokaJapan
  2. 2.Mahab Ghodss Consulting Engineering Co.TehranIran

Personalised recommendations