Land subsidence is identified as a global problem and intensive studies are being conducted worldwide to detect and monitor risk of this problem. Risk assessment of land subsidence is simply an evaluation of the probability and frequency of occurrence of land subsidence, exposure of people and property to the subsidence and consequence of that exposure. Remote sensing technology was used to extract information of land subsidence in Kathmandu Valley, Nepal. Also, Disaster Risk Index method and Analytic Hierarchy Process (AHP) along with Geographic Information System (GIS) tools were used to assess risk of land subsidence in Kathmandu Valley, Nepal. Subsidence volume for locations Central Kathmandu, Chauni, Lalitpur, Imadol, Thimi, Madhyaour Thimi, New Baneshwor, Koteshwor and Gothatar was calculated using a simple mathematical formula. The subsidence depth for these locations was found to be in a range of 1 cm to 17 cm and the maximum subsidence velocity was found to be 4.8 cm/yr. This study revealed that the location where maximum subsidence was observed ( i.e. Central Kathmandu and Lalitpur) was found to be at high risk of experiencing land subsidence induced damage. Other location where subsidence was observed was found to be at medium risk and the rest of the Kathmandu valley was found to be at low risk with current data situation. This study can be considered as the first step towards other comprehensive study relating to land subsidence risk assessment. The outcome of this research provides a basic understanding of the current situation that can further assist in developing prevention and risk management techniques.
Land subsidence is just a geological phenomenon either triggered by natural or anthropogenic activities but when this phenomenon has the probability of resulting harmful consequences or the expected loss (of lives, property, livelihoods, economic activities or environment) then it is considered as risk [
The main damage on manmade infrastructure reported worldwide is mostly related to water transport structures [
The damage to manmade infrastructures is more emphasized and noticeable unlike to the damage to natural systems, which is invisible and generally more threatening. The main reason is that artificial infrastructure damages can generally be repaired opposed to natural system damage which is generally permanent. Some of the examples of natural structure damage are permanent compaction of aquifer system, change in topography which ultimately affects the river patterns and low lying areas.
The other main factor affected by land subsidence is damage to the social environment which includes the human society and the economic development level. The physical damage caused by land subsidence will eventually affect the social environment directly or indirectly but the intensity is determined by the recoverable capability of life, property and various economic activities in the disaster affected areas. Remarkable economic losses have been caused by land subsidence throughout the world [
Kathmandu valley, the capital and the urban core of a developing country Nepal is lagging in terms of data documentation and research work regarding land subsidence and its risk assessment. The factors that make a location prone to land subsidence risk (i.e. geology and groundwater extraction characteristics) are in favor of the valley, yet research is not being conducted. Also, the valley is experiencing rapid increase in population and economic development in the past few decades that will ultimately contribute to increase in risk of damage induced by land subsidence if no counter measures are considered.
Therefore, it is necessary to assess land subsidence risk for decision and policy makers to prevent a huge potential disaster. Risk assessment is simply an application of a methodology for evaluating risk, where risk is defined as the probability and frequency of occurrence of a hazardous event, exposure of people and property to the hazard and consequences of that exposure [
Most frequently deployed approach for land subsidence risk assessment are by the means of Geographic Information System (GIS) techniques and Disaster Risk Index Method and Analytic Hierarchy Process (AHP) [
The main objective of this study is to assess land subsidence risk in Kathmandu valley, Nepal, by using Geographic Information System (GIS) techniques, Disaster Risk Index Method and Analytic Hierarchy Process (AHP). Land subsidence map of Kathmandu valley for 2007 to 2010 was generated by applying the Differential Synthetic Aperture Radar Interferometry (D-InSAR) technique before conducting this study. Peer-reviewed paper related to this study can be found at http://www.mdpi.com/2073-445X/6/2/39.
Kathmandu valley is the capital city of Nepal and hence has been a center of ever growing economic activities from a very long time. Kathmandu valley is an urban agglomerate with a core urban center surrounded by extended urban economic zones. The lack of decentralization of developmental activities has propagated Kathmandu valley to be one of the most desired city to live in the country consequently, increasing the internal migration rates. The population density of Kathmandu valley is 2793 people per square kilometer as per the 2011 census [
Land subsidence volume and velocity obtained from D-InSAR processing (a remote sensing technique) of ALOS PALSAR data from 2007 to 2010 was used in this research. These results were obtained by the authors in a previous study considered as the first part of this study (refer http://www.mdpi.com/2073-445X/6/2/39).
Groundwater exploitation intensity data was used for hazard mapping of land subsidence and was provided by the Kathmandu Valley Water Supply Management Board (KVWSMB), Ministry of Water Supply and Sanitation, Government of Nepal.
Population density, Gross Domestic Product (GDP) and Construction Land Proportion data was used for Vulnerability mapping. Population density data was obtained from the Kathmandu Valley Development Authority (KVDA), Government of Nepal. GDP data for the study area was obtained from National Accounts Section, Central Bureau of Statistics, Government of Nepal. Construction land proportion data was obtained from the Ministry of Land Reform and Management, Nepal.
Through various literature review it was found that in most of the cases the shape of subsidence is very much like a cone shape. Therefore, the subsidence volume of each subsidence zone can be estimated by an assumption that the border extremities of a subsidence zone are linearly moving at a constant rate. A simple cone model designed by [
Subsidence volume represented by the shaded portion in
estimated by applying the formula of a volume of a cone which is shown in Equation (1) [
V = 1 3 [ A 1 + A 1 A 2 + A 2 ] h (1)
where, V is the subsidence volume to be estimated, A1 is the upper base area, A2 is the lower base area and h is the height or the perpendicular distance between the surface A1 and A2. Area A1 and A2 was calculated in ArcGIS (Version 10.4.1) by converting each subsidence zone obtained by DInSAR processing into a shapefile. The subsidence depth (h) and the velocity of land subsidence was also obtained from the DInSAR processing result. The methodology for D-InSAR processing to map land subsidence in Kathmandu valley from 2007 to 2010 has been explained in detail in the first part of this study which can be found at http://www.mdpi.com/2073-445X/6/2/39.
The main objective for assessing risk of land subsidence is to link the subsidence phenomenon with the damage it causes to the physical as well as social environment. The main factors that help to determine risk are hazard and vulnerability. Therefore, a hazard map and vulnerability map were generated and then combined to obtain the final risk map. The detailed methodology is explained as follows.
Disaster Risk Index Method
Risk assessment is an approach for evaluating risk, where risk is defined as the probability and frequency of occurrence of subsidence, exposure of people and property to the subsidence and consequence of that exposure [
D R = f ( H , V , C ) (2)
where, DR is the disaster risk, H is the hazard, V is the vulnerability and C is the capability of disaster prevention and reduction.
Hazard, in general can be defined as a phenomenon that has the potential to disrupt and damage people, property and their immediate environment. The hazard of land subsidence refers to the intensity and the probability that land subsidence will occur in a certain area in a certain period. As defined by the Asian Disaster Preparedness Center, Land subsidence hazard evaluation simply is the process of determining the degree of severity and the extent of the impact area. In this research, accumulative subsidence volume, land subsidence velocity and groundwater exploitation intensity was used as the indicators to evaluate hazard in the study area [
Vulnerability, in general can be defined as a concept that describes the factors (including economic, social and physical) aiding to reduce the ability to cope with the potential hazards impacts. The vulnerability of land subsidence refers to the measure of susceptibility to physical harm or damage caused due to land subsidence. Vulnerability includes the ability of the human society and the economic development level of the society to cope with the disaster caused by land subsidence. Land subsidence vulnerability evaluation is the process of assessing the sensitivity of the economy, population and physical infrastructure to the land subsidence phenomenon. In this research, population density, gross domestic product (GDP) and construction land proportion data was used as the indicators to evaluate vulnerability of the study area [
The capability of disaster prevention and reduction refers to ability of the country to prevent or reduce the effect of potential land subsidence on life, property and economy. However, the land subsidence monitoring of the study area being very poor it was assumed that the country has no ability to control land subsidence at present.
Analytic Hierarchy Process (AHP)
Analytic hierarchy process is a multi-criteria mathematical decision making process developed by Professor Thomas Saaty in (1977). This process uses hierarchical structures to derive relative priorities for criteria (indicators) employing pair wise comparisons. In this research, this process was used to give weights to the indicators identified for evaluating hazard and vulnerability in MSExcel. The basic procedure includes the following steps:
Pair-wise comparison: Pair-wise comparison matrix for hazard and vulnerability was developed separately to establish priorities among the indicators. The result of comparison is derived in terms of integer.
Normalization: The integers obtained from the above step is normalized to compute the priority vector which gives the relative weights among the indicators and ultimately helps to decide which indicator is relatively more important in determining land subsidence risk. Normalization generally means to average the values in each row to compute the corresponding weight.
Consistency Analysis: The main objective of this step is to check if the preference ratings made in the pair-wise comparison are consistent. This is measured in terms of Consistency Ratio (CR), which can be calculated using Equation (3) [
C R = C I R I (3)
where, CI is the consistency index and RI is the random inconsistency indices. RI is provided for each order of matrix by [
C I = λ max − n n − 1 (4)
where, λ max is the value obtained from the summation of product of each normalized weight and sum of columns of the reciprocal matrix, n is the number of indicators used. Saaty, (1980) suggests that the CR value equal to 0.1 or below shows that the comparison is consistent and hence acceptable. (For detail description of the methodology refer “The analytical hierarchy process” Saaty, 1980).
After obtaining the weights for each indicators Hazard map and Vulnerability map was generated in ArcGIS (Version 10.4.1). These two maps were then utilized to obtain the final Risk map by using Equation (2). The methodology flowchart for risk assessment is shown in
The subsidence volume for each location was estimated using Equation (1). The subsidence volume estimation for each location is shown in
Location | Area 1 (m2) | Area 2 (m2) | Height (m) | Volume (m3) |
---|---|---|---|---|
Cone 1 | 124,000 | 41,000 | 0.04 | 3151 |
Cone 2 | 162,000 | 44,000 | 0.04 | 3872 |
Upper part | 5,179,000 | 3,800,00 | 0.12 | 278,474 |
Location L1 | Total | 285,498 | ||
Location L2 | 1,369,000 | 12,000 | 0.11 | 55,336 |
Location L3 | 4,290,000 | 900,000 | 0.11 | 262,348 |
Location L4 | 3,696,000 | 399,000 | 0.11 | 194,677 |
Location L5 | 536,000 | 424,000 | 0.11 | 52,680 |
Location L6 | 1,092,000 | 194,000 | 0.11 | 64,030 |
Cone 1 | 1,079,000 | 211,000 | 0.05 | 29,452 |
Cone 2 | 784,000 | 52,000 | 0.05 | 17,299 |
Location L7 | Total | 46,751 | ||
Location L8 | 1,000,000 | 1,000,000 | 0.02 | 20,000 |
An indicator framework based on the Disaster Risk Index method was prepared with two significant factors for risk assessment along with three indicators for each factor. The indicators were weighted by its significance to govern land subsidence risk using the AHP multi-criteria decision making process. The values of each indicator were classified into three classes Low, Medium and High where low indicates the lowest hazard/vulnerability level and high indicates the highest hazard/vulnerability level.
Kathmandu valley was divided into three zones corresponding to the major cities Kathmandu, Bhaktapur and Lalitpur along with the subsidence locations obtained from previous processing for hazard evaluation. The data values of indicator for each zone and location is shown in
Index | Grade and value | ||||
---|---|---|---|---|---|
Low | Medium | High | |||
Factor | Indicator | Weighted value | [ | [ | [ |
Hazard | Accumulated Subsidence volume (mm) | 0.63 | 0 - 10000 | 10000 - 100000 | >100000 |
Subsidence velocity (mm/yr.) | 0.26 | 0 - 30 | 30 - 50 | >50 | |
Groundwater exploitation intensity (104 m3/yr.) | 0.11 | 0-8 | 8-16 | >16 | |
Vulnerability | Population density (person/km2) | 0.67 | 0 - 1000 | 1000 - 10000 | >10000 |
GDP per km2 (104 $) | 0.24 | 0 - 99 | 100 - 499 | >500 | |
Construction land proportion (%) | 0.09 | 0 - 49 | 50 - 79 | >80 |
Location | Volume (m3) | Maximum subsidence velocity (mm/yr.) | Average Discharge (104 m3/yr.) |
---|---|---|---|
Location L1 | 285,498 | 48 | 32 |
Location L2 | 55,336 | 26 | 4 |
Location L3 | 262,348 | 33 | 3 |
Location L4 | 194,677 | 30 | 2 |
Location L5 | 52,680 | 35 | 4 |
Location L6 | 64,030 | 29 | 1 |
Location L7 | 46,751 | 18 | 3 |
Location L8 | 20,000 | 11 | 2 |
Kathmandu zone | - | - | 4 |
Lalitpur zone | - | - | 4 |
Bhaktapur zone | - | - | 3 |
The land subsidence hazard map of Kathmandu valley is shown in
This indicates that there is a high probability that land subsidence occurrence is bound to intensify in locations L1, L3 and L4 considering there is no reduction in the average groundwater discharge.
In the same manner to hazard evaluation, Kathmandu valley was also divided into three zones corresponding to the major cities Kathmandu, Bhaktapur and Lalitpur along with the subsidence locations obtained from previous processing. The data values of indicator for each zone and location is shown in
The land subsidence vulnerability map for Kathmandu valley is shown in
Location | Municipality | Population density | GDP ($) | Construction land Proportion (%) |
---|---|---|---|---|
Location L1 | Kathmandu | 19,726 | 749,582 | 80 |
Location L2 | Kathmandu | 19,726 | 749,582 | 95 |
Location L3 | Lalitpur | 14,574 | 1,014,539 | 80 |
Location L4 | Lalitpur District | 3241 | 4,562,750 | 30 |
Location L5 | MadhyapurThimi | 7717 | 1,916,042 | 60 |
Location L6 | MadhyapurThimi | 7717 | 1,916,042 | 20 |
Location L7 | Kathmandu | 19,726 | 749,582 | 70 |
Location L8 | Kathmandu | 19,726 | 749,582 | 50 |
Kathmandu zone | 2382 | 6 | 60 | |
Lalitpur zone | 1197 | 12 | 40 | |
Bhaktapur zone | 680 | 22 | 30 |
vulnerability is seen in Bhaktapur zone as the population density and economic activity is lowest in this area. The result indicates that Location L1, L2, L3 L7 and L8 are most sensitive to damage caused by land subsidence.
Land subsidence risk map of Kathmandu valley was generated based on the hazard and vulnerability evaluation in ArcGIS (Version 10.4.1) using Equation (2) (
Shrestha, et al., (2017) [
Due to lack of proper scientific data and research, evidence of damage caused by land subsidence has not been reported in Kathmandu valley till date. Case studies from around the world can be referred to utilize the knowledge and experience in planning and policy making to reduce if not prevent the disastrous effect of land subsidence. The location of Kathmandu valley has a very close resemblance with the location of Mexico City where land subsidence has been documented very well. Hence, case study of land subsidence in Mexico City is discussed here.
Mexico City is the capital city of Mexico located in the valley of Mexico Basin. It is surrounded by volcanic chain and mountains with elevations reaching up to 5000 m [
The city has been suffering from groundwater extraction related subsidence since decades [
The main problems reported relating to land subsidence in Mexico City are
・ Decrease in runoff and wastewater drainage ability, that ultimately results in flooding during rainy season [
・ Disruption to the water transportation structures (i.e. pipelines, canals) resulting in interference with water supply [
・ Damage to the stability of manmade infrastructure (Buildings, transportation facilities like highways, roads and bridges) due to changes in surface gradients [
Since, there is a close resemblance in Kathmandu Valley and Mexico City, it can be expected that these problems may be encountered in Kathmandu Valley in near future if current situations prevail. The high-risk areas indicated by red color in
Land subsidence risk assessment based on current data revealed that the identified subsidence areas are at high and medium risk of suffering from subsidence induced damages like decrease in runoff and wastewater drainage ability, disruption to water transportation structures and damage artificial infrastructure stability whereas the rest of the Kathmandu valley is at low risk under the circumstance that similar conditions prevail. The outcomes of this research even though it is not directly validated can be used as a base for further detailed study. The results could also serve beneficial for developing disaster prevention policies. In addition, a more comprehensive risk assessment of land subsidence can be done by considering other indicators like the geological characteristics and land use type of the location. This would give a more detailed outlook on factors that can be controlled to reduce if not prevent a huge disaster. Also, in this study, it was assumed that the study area had no capacity for disaster prevention and reduction for risk assessment. However, different case scenarios like with government action to reduce groundwater exploitation and with government action to reduce or prevent construction land proportion can be employed to judge the situation in a different perspective.
The authors would like to thank Kathmandu Valley Water Supply Management Board (KVWSMB), Ministry of Water Supply and Sanitation, Kathmandu Valley Development Authority (KVDA) and National Accounts Section, Central Bureau of Statistics, Government of Nepal for providing necessary data to conduct this study. The authors would also like to express a hearty gratitude to Mr. Suyesh Bhattarai for his help in obtaining the required data.
Bhattarai, R. and Kondoh, A. (2017) Risk Assessment of Land Subsidence in Kathmandu Valley, Nepal, Using Remote Sensing and GIS. Advances in Remote Sensing, 6, 132-146. https://doi.org/10.4236/ars.2017.62010