Landslide Risks in a Changing Climate – Säveån River Valley
Part 1: Map Report and Summary of Results
SGI Publication 38–1E
Linköping 2017
Publication 38-1E Document reference:
SGI 2017, Landslide Risks in a Changing Climate – Säveån River Valley. Part 1: Map Report and Result Summary, Swedish Geotechnical Institute, SGI Publication 38–1E, Linköping.
Reference number: 1.1-1602-0091 Project number: 16029A
Orders:
Swedish Geotechnical Institute Information Service
SE-581 93 Linköping
Telephone: + 46 13-20 18 04 E-mail: info@swedgeo.se
Download this publication as a PDF document at www.swedgeo.se
Cover page photos: SGI (left), SGI/Swedish Mapping, Cadastral and Land Registration Authority (centre), SGI (right).
Landslide Risks in a Changing Climate – Säveån River Valley
Part 1: Map Report and Result Summary
SGI Publication 38–1E Linköping 2017
SGI Publication 38-1
4 (34)
SGI Publication 38–1E
Introduction
Society needs to adapt to a changing climate and consider the consequences when planning
buildings and infrastructure. Effective climate adaptation not only requires knowledge and decision support tools that are flexible and multidisciplinary, and which take into account local variations, but also that the measures that are implemented are coordinated at regional level.
Since 2009, SGI has allocated funds from government appropriation 1:10 Climate Change
Adaptation for a series of initiatives, including landslide risk mapping, methodology development, and the utilisation of materials from the mapping process.
Using material from the Göta River Valley Investigation (Göta älvutredning or GÄU) (SGI 2012), SGI has identified and prioritised additional watercourses for landslide risk mapping. The
Norsälven river valley was previously surveyed as a pilot area for the development of a simplified methodology for landslide risk mapping, using the Göta River Valley Investigation as a starting point. The investigation of the Säveån river valley builds on and develops the methodology used for the Norsälven river valley. The aim of the mapping of the Säveån valley is to provide wide- ranging documentation of the river, which that can then be used in municipal comprehensive planning to facilitate prioritisation for future municipal and regional work aimed at adapting society to a changing climate.
The investigation results and conclusions are presented in this report ‘Landside Risks in a Changing Climate – Säveån River Valley’, which consists of three parts:
• Part 1 – Map report and result summary, contains a summary of the government mission and sets out proposals for how the results can be used in climate adaptation work in different municipalities and counties. It also presents the landslide risks in map format, as well as maps of probability and consequences. The landslide risks maps contain risks both under current conditions, as well as the assessed sensitivities to future climate impact along the river valley.
• Part 2 – Mapping method, contains a description of the methodology, inventories, surveys, calculations and analyses.
• Part 3 – Detailed annex, contains a more detailed description of the methodology developed and used in the consequence analysis.
This investigation is available in map format via SGI’s map-viewing service. This service provides various map layers (GIS layers), which can be turned on or off, and which may prove to be
extremely valuable when used in planning activities.
The work on this report has been performed mainly by employees at SGI, and is organised as a core project, together with several sub-projects focusing on methodology development, analysis and investigation. The work was led by a senior management team comprising Karin Bergdahl, Karin Odén, Gunnel Göransson and Hjördis Löfroth. Åsa Jönsson, Ramona Kiilsgaard, Eva Narbrink and Lisa Van Well acted as sub-project leaders. Rebecca Bertilsson, Anette Björlin, Jim Hedfors, Ulrika Isacsson, Godefroid Ndayikengurukiye, Wilhelm Rankka, Gasper Sechu and Mats Öberg made significant contributions. In total, almost 30 SGI employees have contributed to this work. Bo Lind acted as internal report reviewer.
The work has been carried out in collaboration with other governmental authorities and research institutes, including the Geological Survey of Sweden (SGU), the Swedish Meteorological and
SGI Publication 38–1E
6 (33)
Hydrological Institute (SMHI), the Västra Götaland County Administrative Board, the City of Gothenburg, and the municipalities of Lerum and Partille.
Approval for publication was granted in Linköping in August 2017 by Charlotte Cederbom, the Head of SGI’s Land Use Planning and Climate Adaptation Department.
This is a translation from the Swedish original. SGI cannot be held responsible for any misunderstanding or error that may arise from the translation.
SGI Publication 38–1E
Contents
Introduction ... 5
Landslide risk analysis of the Säveån river valley – summary ... 8
1. Background ... 10
1.1 Scope and delimitation ... 11
2. Landslide risks in the Säveån river valley ... 12
2.1 Mapping – result summary ... 12
2.2 Presentation of the results in map form ... 18
2.3 Probability, consequence and risk ... 21
3. Recommendations for future work ... 28
3.1 Societal planning and construction ... 28
3.2 Climate adaptation measures ... 29
References ... 33
Appendices
Division of map sheets (Bladindelning) Landslide risk maps (Skredriskkartor) Probability maps (Sannolikhetskartor) Consequence maps (Konsekvenskartor)
Note: All maps are presented with the original Swedish legends.
SGI Publication 38–1E
8 (33)
Landslide risk analysis of the Säveån river valley – summary
The landslide risk analysis of the Säveån river valley has resulted in a general overview of the risk of landslides in both built and undeveloped areas, as well as in areas containing key societal
infrastructure, under both present and future climate conditions.
Landslide risk
A high landslide risk, i.e. where the probability of a landslide is substantial, and where the consequences of a landslide are serious for existing buildings and
infrastructure, has been found for around 10 per cent of the land surface within the investigated area. This includes industrial facilities in Gothenburg and Partille, as well as many buildings and larger road and rail facilities (the E20 motorway and the Western Main Line railway) along the Säveån river valley, mainly in Partille and Lerum.
A moderate landslide risk is common in areas closest to the river bank along large sections of the river, equivalent to around 15 per cent of the land surface within the investigated area. The probability of a landslide in these areas is substantial, even though they
lack key societal infrastructure. A major landslide in these areas could have secondary
consequences, such as flooding of buildings and infrastructure upstream and downstream. These have not been considered in this analysis.
The largest part of the investigated area (around 50 per cent of the land surface) falls into the low landslide risk category.
Sensitivity to climate change in the future is deemed to be high for parts of Partille (Kåhög- Jonsered) and within the central parts of Lerum. Other areas are deemed to have a moderate or low level of sensitivity to climate impact.
Measures to reduce landslide risk
Identified areas with a high landslide risk should be investigated in more detail before decisions are taken regarding possible action. Extensive work on more detailed investigations and measures is being carried out by the municipalities in the identified risk areas. The risk of a landslide along watercourses can be alleviated by introducing measures to either mitigate the probability of a landslide and/or the consequences of a landslide. Measures to mitigate the probability are usually physical in the form of excavations, stabilization berms and river bank protection. Measures to reduce the consequences could include relocation of buildings or facilities.
Foto: SGI
Map viewing service
A map viewing service can be found on our website:
swedgeo.se. The GIS layers with probability, consequence and landslide risk maps along the river are shown, as well as further location-specific information not presented in this report.
SGI Publication 38–1E
Further development of a simplified method
The method used for landslide risk mapping of the Säveån river valley is a simplified approach based on the method employed in the Göta River Valley Investigation. This method has been adjusted for each of the watercourses investigated. In this investigation, for example, methodology development related to the classification of consequences has been extended to include polluted areas. The method for risk mapping that has been developed can be applied to mapping landslide risks along other watercourses. The methods developed for probability classification and
consequence classification can be used independently of each other. The method for consequence classification can, for instance, be adapted for use when mapping other natural hazards.
SGI Publication 38–1E
10 (33)
1. Background
Expanding on the Göta River Valley Investigation (GÄU) (SGI 2012), SGI has developed methods for mapping landslide risks that can be applied to other watercourses.
SGI’s work with landslide risk mapping began with the Norsälven river valley as a pilot area and continued with the Säveån river valley, described in SGI’s budget plan as “An investigation of a more comprehensive nature that can be used as input for planning and decision-making by county administrative boards and municipal authorities in their climate adaptation work at regional or local level”.
With a more wide-ranging picture of landside risk areas along the prioritised watercourses (SGI 2013), it is possible to make a more substantiated assessment of areas that demand more detailed geotechnical investigations, and at the same time identify where adaptation measures (geotechnical reinforcement measures and climate adaptation measures) generate the greatest public benefit and are most cost-effective. The municipalities concerned also acquire a more complete basis with regard to the probability and societal consequences of landslides within built areas and in areas where new construction is planned.
The purpose of landslide risk mapping is to produce an overall map of landslide risks along the investigated watercourse. The map shows the distribution of risk levels and the probability and consequences (in pairs), as well as the impact of climate change from a 100-year perspective. The landslide risk map can be used as a basis for planning decisions regarding measures deriving from the municipal comprehensive planning process.
The methods employed in the Göta River Valley Investigation (GÄU) (SGI 2012) have been applied as far as possible to the Norsälven river valley (SGI 2015) and have been further developed for the Säveån river valley. Method development has been necessary to reduce the investigation costs and to simplify interpretation of the maps without it resulting in a critical impairment of the usability. Account has been taken of the evaluation and comments made after the GÄU. SGI has endeavoured to make the results more understandable.
Landslide risk mapping generates significant societal benefits by providing material that helps to:
• Avoid or mitigate the consequences of landslides
• Reduce the likelihood of landslides
• Support the fulfilment of environmental quality goals – a good built environment and a good non-toxic environment
• Provide input for planning for a changing climate in the future
There are many areas in Sweden that are susceptible to landslides. Areas along
watercourses flowing through layers of loose soil are often more vulnerable than other areas.
In these vulnerable areas, the effects of climate change can also be very tangible, e.g. a higher water flow that causes increased erosion and a deterioration in soil layer stability. The aim of landslide risk mapping is to provide an overall picture of the landslide risks along each specific watercourse, not only under current conditions but also from a long-term perspective in 2100.
SGI Publication 38–1E
1.1 Scope and delimitation
The investigated area along the Säveån river valley extends from Sävelången in the municipality of Lerum to Gamlestaden in Gothenburg, close to the confluence with the Göta Älv river, see Figure 1.1. The total distance is around 30 km, equivalent to 60 km of river banks. The width of the investigation area is limited to around 200 m from the banks, and in some cases a shorter distance when delimitation against solid ground was possible.
Figure 1.1 Investigation area – Säveån river valley.
SGI Publication 38–1E
12 (33)
2. Landslide risks in the Säveån river valley
This part of the publication provides a summary of the survey and mapping of the landslide risks along the Säveån river valley. Part 2 (SGI 2017b) and Part 3 (SGI 2017c) present the methods used for analysis and mapping in more detail.
2.1 Mapping – result summary
Areas that have a high landslide risk are relatively limited along the Säveån river valley (around 10% of the investigated area). Areas with a moderate landslide risk make up around 15% of the investigated area. The main part of the investigated area, i.e. around 50%, has a low landslide risk.
The remaining land surface is made up of solid ground or non-categorised surfaces.
The sensitivity to climate impact in the light of future changes in climate is assessed to be high for part of Partille (Kåhög-Jonsered) and within the central parts of Lerum. In the other areas the impact is deemed to be moderate or low.
The results for the different sections of the Säveån river valley are described in more detail below.
The measurement distance used in the description is based on the distance from the confluence of the Säveån river and the Göta Älv river. The results are also shown in the map annex: Landslide risk maps (pages 1-11), Probability maps (pages 1-11) and Consequence maps (pages 1-11).
2.1.1 Gothenburg
Probability
Around 30% of the investigated area within Gothenburg has a high or substantial probability of a landslide, particularly in the areas closest to the Säveån river. This area extends 50-150 m from the river. In some areas it extends slightly further, e.g. in areas with quick clay, where the extent of a landslide could be greater. In many of these areas, more detailed stability investigations have already been carried out or are ongoing. In several cases, measures have been carried out, although it has not been possible to take these measures into account in any detail in this general
investigation. Around 60% of the investigated area in Gothenburg has a negligible probability of a landslide, mainly in the areas farthest away from the river (more than 150 m).
Landslide risk is defined in this report as a combination of the probability of a landslide and the consequences of a landslide. The probability of a landslide and associated consequences are combined in a risk matrix. The risk is categorised on three levels – high, moderate and low – and with three corresponding colours – red, orange and yellow – on the maps showing the landslide risk.
The survey represents a general level of probability and consequences, where a qualitative assessment of the consequences has been carried out, and thus also a qualitative evaluation of the landslide risk.
SGI Publication 38–1E
Consequences
Surfaces with extremely serious and catastrophic consequences make up around 36% of the investigated area, varying depending on the distance from the river. However, in most of the investigated area the consequences are minor. The area in question consists mainly of natural areas and undeveloped land. These areas are large, predominantly on the north side of the river, while on the south side of the river they are smaller and relatively evenly distributed.
Landslide risk
Within the whole of the investigated area in Gothenburg, stretches with either a high or moderate landslide risk are found along both the south side and north side of the river. Further away from the river bank, the landslide risk level is generally low. In principle, there are geotechnical stability investigations at both a detailed and in-depth level throughout the entire stretch in Gothenburg.
These are marked on the map as faded areas and should be regarded as areas where the municipal authority has already stipulated the landslide risk. Stability-enhancement measures have also been implemented at several sites.
Figure 2.1 shows an overview of the landslide risk map in Gothenburg, produced in the light of the assessed climate impact, and indicating areas where detailed and in-depth stability surveys have already been carried out.
Climate impact
The climate impact has been assessed as moderate for a large part of the area, up to around distance point 4600, where the climate impact shifts to low.
Figure 2.1 shows an overview map of the landside risk levels in Gothenburg in the light of the assessed climate impact, and indicating the areas in which detailed and in-depth stability surveys have been carried out. Map sheets 1 and 2 present the landslide risk results for Gothenburg, including probability categories and consequence categories.
Figure 2.1 General landslide risk map of the Gothenburg sub-area. (© SGI, Swedish Mapping, Cadastral and Land Registration Authority, Geodata Collaboration).
SGI Publication 38–1E
14 (33)
2.1.2 Partille
Probability
In Partille, the area where there is a high or substantial probability of a landslide accounts for approximately 35% of the area covered by the investigation and in particular the areas closest to the Säveån river. The area extends 30-100 m from the river, in some cases slightly farther away, e.g. at Kåbäcken, Kåhög and Jonsered. There is quick clay along this stretch, which means the extent of the landslide would be greater. This is thus also the area in which there is a high or substantial probability that the landslide risk will be greater. In a large number of the areas where there is a high or substantial probability of a landslide, detailed investigations have been carried out or are in progress. In many cases, action has been taken. Around 50% of the land area has a negligible probability of a landslide, and this applies primarily to the areas some distance away from the river (more than 60-100 m).
Consequences
The area with extremely serious and catastrophic consequences represents approximately 33% of the investigation area and is at a varying distance from the river (although generally more than 60 m). The largest part of the area, however, has minor consequences, approximately 50%, and this is reflected in areas comprising natural land and undeveloped land. These areas take the form of large unbroken blocks, primarily on the north side of the river, in the eastern part to the east of
Brodalsbäcken, and in the area east of Uddaredsbäcken on the south side of the river.
Landslide risk
A moderate risk of a landslide exists along the river bank throughout the whole of the investigation area, and is disrupted in a number of high-risk areas, including the area where the Western Main Line railway crosses the river at the municipal border with Gothenburg, and in Kåhög and Jonsered, where the areas extend a couple of hundred metres up from the river bank. At many locations in Partille, geotechnical stability investigations have been conducted on both the detailed and in-depth level. These are shown on the map as faded areas, and should be regarded as areas where the municipal authority has already focused its attention on the landslide risk. Stability- enhancement measures have also been carried out at several locations.
Climate impact
In the light of the increased probability of a landslide, climate impact is considered to be low to moderate for a large part of Partille, with the exception of parts of Kåhög and Jonsered, where climate impact shifts from moderate to high.
Figure 2.2 contains an overview of the landslide risk map covering Partille, including the assessed climate impact, and with areas marked in which detailed and in-depth stability investigations have been conducted. Maps 2-4 show the results for Partille in terms of landslide risk, probability category and consequence category.
SGI Publication 38–1E
Figure 2.2 Overview, landslide risk map, Partille sub-area. (© SGI, Swedish Mapping, Cadastral and Land Registration Authority, Geodata Collaboration).
2.1.3 Lerum
Probability
The geological conditions along the Säveån river valley in Lerum are more fragmented and varied than further downstream. This is particularly the case upstream of the centre of Lerum. The interpolation between the representative sections are more difficult here, resulting in greater uncertainty regarding the probability classification. See Part 2, Annex 1 (SGI 2017b) for a description of the probability classification between the computation sections.
In Lerum, the area with a high or substantial probability comprises approximately 13% of the investigation area, concentrated along both sides of the Säveån river in the centre of Lerum, and on the north side of the river between Stenkullen and Floda. The area extends 0-100 m from the river, in certain cases slightly further, e.g. in the centre of Lerum. On this stretch, there is quick clay in the area, which extends the landslide risk area to more than 150 m. In large parts of these areas, detailed examinations have been carried out or are in progress. In many cases, action has been taken, although it has not been possible to take this into account in detail in this general
investigation. Around 43% of the investigation area has a negligible probability of a landslide, and this applies in particular to the areas some way from the river, i.e. more than 60-100 m from, for example, the centre of Lerum and from Hedefors eastwards on the north side of the river. On the south side of the river, east of the centre of Lerum, there is a negligible probability for long stretches, even in the sections closest to the river.
Consequences
The area with extremely serious and catastrophic consequences represents around 21% of the investigation area, and at varying distances from the river. In the centre of Lerum, the area with the greatest potential consequences is also very close to the river, although for other parts it is more than 50 m from the river. The majority of the area, however, has negligible consequences, approximately 55%, and this takes the form of natural land and undeveloped land. These areas
SGI Publication 38–1E
16 (33)
comprise large unbroken blocks east of the centre of Lerum, as far as Hedefors, between Stenkullen and Knavra bro, and between Stenkullen and Floda.
Landslide risk
Approximately 7% of the land area within Lerum is categorised as high risk, 8% as moderate risk, and 49% as low risk. Around Lake Aspen, the landslide risk level is mostly low, apart from along the Western Main Line railway to the south. In the area around the centre of Lerum, the risk level is moderate to high closest to the river, and low further away from the river. In this densely developed area, several geotechnical stability investigations have been carried out previously on a detailed and in-depth level. Shown on the map as faded areas, they should be regarded as areas where the municipal authority has already drawn attention to the landslide risk. Stability-enhancement measures have been carried out at several locations. There are also small high-risk areas around Hedefors, and further upstream on the north side of the river.
Climate impact
Climate impact, with account taken of an increased probability of a landslide, varies on the long section. Around Lake Aspen, it is low as the level of erosion is low, which can be attributed to the fact that it is a lake and not running water. Along the centre of Lerum, climate impact switches between low, moderate and high, the reasons for which are many, including the variation in the flow rate on the outer curves and inner curves, and the geological conditions on the river bed. East of the town centre, climate impact is largely low, although there is a transition to moderate further eastwards. It should also be added that a lack of river bed data between distance points 21800 and 22800, and between points 23400 and 28300, has meant that erosion of the river bed could not be measured. Instead, it has been assessed using the soil type map, erosion index, and modelled river bed shear stress.
Figure 2.3 shows an overview of the landslide risk map for Lerum, with assessed climate impact.
The areas that have been the subject of detailed and in-depth stability investigations are
highlighted. Maps 5-11 show the results for Lerum for landslide risks, probability categories and consequence categories.
SGI Publication 38–1E
Figure 2.3 Overview, landslide risk map, Lerum sub-area. (© SGI, Swedish Mapping, Cadastral and Land Registration Authority, Geodata Collaboration).
SGI Publication 38–1E
18 (33)
2.2 Presentation of the results in map form
The results are presented in the Map Annex in three different map series on a scale of 1:10 000 (A4). The maps are arranged into landslide risk maps, probability maps and consequence maps, divided into sheets 1-11. See sheet overview in the Map Annex.
• Landslide risk maps: The landslide risks and sensitivity to climate impact within different areas along the Säveån river valley are shown on the landslide maps (Sheets 1-11).
• Probability maps: The probability of a landslide and the sensitivity to climate impact along the Säveån river valley are shown on the probability maps (Sheets 1-11).
• Consequence maps: The consequences of landslides along the Säveån river valley are shown on the consequence maps (Sheets 1-11).
Figure 2.4 explains the different symbols and designations that appear in the risk map legends. The weighed risk (probability and consequence) is reported on three levels: low, moderate and high.
Current risk levels
Area with a LOW risk of a landslide. No specific investigations required for existing buildings and facilities. Stability investigations required for new development.
Area with a MODERATE risk of a landslide. Existing buildings and facilities are subject to a detailed stability investigation. In the case of new development, a detailed stability investigation is required along with possible measures.
Area with a HIGH risk of a landslide. The need for measures for existing buildings and facilities is determined by means of a detailed stability investigation. In the case of new development, a detailed stability investigation is required along with probable stability-enhancement measures.
Climate change impact
*)LOW sensitivity to climate impact. Climate change does not entail any general change in the probability classification. ‘No general change’ means that the probability is changed by up to half a step for areas with negligible to low probability (categories S1-S2) in today’s climate, whilst for areas with limited to substantial probability (categories S3-S5) it means no change.
MODERATE sensitivity to climate impact. Climate change means that the probability classification is changed by half to one step in areas with a low probability classification (categories S1-S2) in today’s climate. For areas with a limited to substantial probability (categories S3-S5) it is changed by up to half a step.
HIGH sensitivity to climate impact. Climate change means that the probability classification is changed by more than one step in areas with negligible to low probability (categories S1-S2) in today’s climate, and by more than half a step in areas with limited to substantial probability (categories S3-S5).
*) For areas that fall into the highest probability category (S5), the probability category cannot increase. In these areas, however, even a low level of impact caused by climate change could cause a landslide to occur.
Figure 2.4 Explanation of the legends for landslide maps. The line marking for climate impact is shown on the map along both sides of the river.
SGI Publication 38–1E
Use of the maps
The overall consequences of a landslide depend on the magnitude of the landslide, which is illustrated in Figure 2.5. The most common situation is that a landslide begins at the river bank and then progresses backwards to a varying extent, depending on the soil properties and the local topography. If there is quick clay in the area, the spread could be quite extensive. The
consequences as a whole are not shown on the consequence maps and must instead be assessed in each individual case.
Figure 2.5 Illustration of how the consequences of a landslide can depend on the magnitude and spread of the landslide.
Figure 2.6 shows an inventory of ‘actual’ landslide zones for part of the investigation area, presented in relation to the probability classification and consequence classification. This is an example of how the potential extent of a landslide can be assessed within the area, and gives an approximate idea of the extent of the consequences.
SGI Publication 38–1E
20 (33)
Figure 2.6 Map extracts showing examples of the spread of earlier landslides (marked with a light-blue line) in relation to the probability classification and consequence classification. The size of the landslide varies from around 400 m in length and 50-80 m in depth, to 75-100 m in length and 40-70 m in depth. © SGI, SGU, LM, Geodata Collaboration. The topography in the images is taken from the National Height Database. The image in the middle is the probability map and the image at the bottom is the consequence map.
SGI Publication 38–1E
2.3 Probability, consequence and risk
2.3.1 Probability of a landslide
The degree to which a landslide can be prevented, also termed stability, is normally expressed as the ratio between the driving and resisting forces on a slope. This ratio is known as the factor of safety. The resisting force mainly comprises the strength of the soil, but also stabilising forces, e.g.
from the water mass in a watercourse or additional resisting forces in the form of soil and stone infill in the lower part of the slope. The driving forces arise from the weight of the soil, as well the load exerted on the land, e.g. by buildings and material dumps.
To ensure the description is as close to reality as possible, the traditional computation of the factors of safety is complemented with an assessment of the probability of a landslide, where consideration is given to the uncertainty that is inherent in the parameters that have been incorporated. The stability is analysed with the aid of parameters that have been assigned a variation that reflects their uncertainty. The variation is determined in each individual case using experience from similar areas, and with statistical data gathered from investigations and measurements. A number of parameters change over time as a result of climate change, which means that computations need to be made both for the current situation and the situation in the future.
The probability of a landslide has been divided into five categories, S1-S5, see Table 2.1. The borders between the different probability categories have been set based on European and Swedish building standards, which are generally used for the construction of buildings. The categories have been selected in a way that probability category S5 means poorer conditions than the worst
category that is acceptable for temporary constructions, whilst probability category S1 means better stability than the requirements set for normal buildings (Berggren et al., 2011, GÄU Sub-Report 28). The computed probability has been linked together between the sections by assessing the probability for the areas between the sections based on their geotechnical and geometrical conditions in relation to the assessed conditions and results for the sections.
Table 2.1 Classification of landslide probability.
Probability category Probability of a landslide Relative probability of failure
S1 Negligible Pf < 310-6
S2 Low 310-6 ≤ Pf < 110-4
S3 Moderate 110-4 ≤ Pf < 310-3
S4 High 310-3 ≤ Pf < 110-1
S5 Substantial Pf ≥ 110-1
Consequences of a landslide
Alongside the computation of the probability of a landslide, the consequences of a landslide along the Säveån river valley have also been assessed. The consequences for buildings and transport routes in the area, as well as contaminated areas (MIFO-classified areas), have been evaluated qualitatively based on four aspects: life, environment, economic impact and societal function.
Landslides in the Säveån river valley could affect a large number of people as well as key societal functions. There is no exhaustive description of all the consequences, and in particular no economic
SGI Publication 38–1E
22 (33)
impact evaluation. Landslides also result in suffering, sadness or discomfort for many people.
These aspects have not been evaluated, and instead they are considered to be indirect consequences. The collective consequence of the landslide could embody several different consequences.
Consequences have not been assessed for a changing climate as this information is not available. It is assumed that societal planning takes into account both existing conditions as well as conditions resulting from a changing climate. In the investigation, the probability map could be used as a guide to determine where the geotechnical conditions ought to be taken into account specifically as part of the societal planning process (change of consequences).
The consequences have been divided into five categories, where the division is linked to the classification from earlier landslide risk investigations and represents a gradual increase in the consequences in the categories, see Table 2.2. Each data layer, i.e. buildings, roads and railways, as well as MIFO-classified areas 1 and 2, which include the different consequences, has subsequently been divided into squares, 10 x 10 m in size, and they have been combined in such a way that the greatest consequence within each square determines the consequence classification.
Table 2.2 The objects in the three data layers have been value-categorised based on four aspects: life, environment, economic impact and societal function, and have been placed into a value category ranging from 1-5.
Value category
Building Transport infrastructure
*
Contaminated areas **
5
Catastrophic consequences
LIFE
A number of people would probably be injured or killed if they were present in a multi-arena.
ENVIRONMENT
Catastrophic consequences for the environment, equivalent to a nuclear power station or Seveso facility.
ECONOMIC IMPACT
Catastrophically high financial losses that differ from the majority of economic losses, equivalent to a nuclear power station or Seveso facility.
SOCIETAL FUNCTION Loss of a major societal function, equivalent to a hospital.
LIFE
ENVIRONMENT
ECONOMIC IMPACT
Catastrophically serious economic losses that differ from the majority of economic losses, equivalent to a motorway or railway that is classified as being of national importance.
SOCIETAL FUNCTION Loss of a major societal function, equivalent to a motorway or railway that is classified as being of national interest.
LIFE
ENVIRONMENT
Catastrophic consequences for the environment, equivalent to the ecosystem of an entire watercourse being destroyed, along with its functions and structure, and the disappearance of key species.
Environmental quality norms for water are exceeded over a long period (generation perspective).
ECONOMIC IMPACT SOCIETAL FUNCTION
4
Extremely serious consequences
LIFE
A number of people would probably be injured or killed if they were present in a large school, multi-family dwelling or large railway station.
ENVIRONMENT
Extremely serious consequences for the environment, equivalent to a metal processing plant.
ECONOMIC IMPACT
Extremely serious economic losses, equivalent to a metal processing plant.
SOCIETAL FUNCTION Loss of a major societal function, such as a health centre.
LIFE
ENVIRONMENT
ECONOMIC IMPACT
Extremely serious economic losses, equivalent to a motorway or railway.
SOCIETAL FUNCTION Loss of a major societal function, equivalent to a motorway or railway.
LIFE
ENVIRONMENT
Extremely serious consequences for the environment, equivalent to the ecosystem of a whole watercourse being destroyed, together with its functions and structure, but which recovers gradually. Key species disappear temporarily. Environmental quality norms for water are exceeded over a long period (>6 years).
ECONOMIC IMPACT SOCIETAL FUNCTION
SGI Publication 38–1E
* For the transport infrastructure, the life aspect is not a significant factor, as it is probably the case that few people would be affected by a landslide on a road (compared with a building). Nor is the environment considered to be a significant factor within the transport infrastructure (compared with a building). The other aspects are of major significance.
** In the case of contaminated areas, only the environmental aspect has been used as a basis for value categorisation. The assessment has been made for the aquatic environment, not the land environment.
3
Very serious consequences
LIFE
A number of people would probably be injured or killed if they were present in a small multi-family dwelling.
ENVIRONMENT Major consequences for the environment, equivalent to an indoor swimming pool.
ECONOMIC IMPACT
Major economic losses, equivalent to an indoor swimming pool.
SOCIETAL FUNCTION Loss of a medium-sized societal function, equivalent to an animal hospital.
LIFE
ENVIRONMENT
ECONOMIC IMPACT
Major economic losses, equivalent to a public road, category I (width >7m) and a tram route.
SOCIETAL FUNCTION Loss of a medium-sized societal function, equivalent to a public road, category I (width >7m) as well as a tram route.
LIFE
ENVIRONMENT
Extremely serious consequences for the environment, equivalent to the ecosystem of a watercourse being destroyed locally together with its functions and structure.
Environmental quality norms for water are exceeded over a long period (>6 years).
ECONOMIC IMPACT SOCIETAL FUNCTION
2
Serious consequences
LIFE
A few people would probably be injured or killed.
ENVIRONMENT
Significant consequences for the environment, equivalent to an unspecified agricultural building.
ECONOMIC IMPACT Significant economic losses, equivalent to an unspecified agricultural building.
SOCIETAL FUNCTION Loss of a small societal function, equivalent to a distribution building.
LIFE
ENVIRONMENT
ECONOMIC IMPACT Significant economic losses, equivalent to a public road, category II (width 5-7m),
SOCIETAL FUNCTION Loss of a small societal function, equivalent to a public road, category II (width 5-7m).
LIFE
ENVIRONMENT
Serious consequences for the environment, equivalent to the ecosystem of a watercourse being destroyed locally together with its functions and structure.
Environmental quality standards for water are exceeded for a short period (<6 years) or only locally.
ECONOMIC IMPACT SOCIETAL FUNCTION
1
Minor
consequences LIFE
Probably no-one would be injured or killed.
ENVIRONMENT
Moderate consequences for the environment, equivalent to an auxiliary building.
ECONOMIC IMPACT
Slight economic losses, equivalent to an auxiliary building.
SOCIETAL FUNCTION Loss of a very minor societal function, equivalent to an auxiliary building.
LIFE
ENVIRONMENT
ECONOMIC IMPACT
Minor economic losses, equivalent to a public road, category III (width
<5m) and other roads, SOCIETAL FUNCTION Loss of a minor societal function, equivalent to a public road, category III (width <5m) and other roads.
LIFE
ENVIRONMENT Minor consequences for the environment, equivalent to a low level of local impact on the ecosystem and there is no risk of environmental quality standards for water being exceeded.
ECONOMIC IMPACT SOCIETAL FUNCTION
SGI Publication 38–1E
24 (33)
2.3.3 Landslide risk categories
The above-mentioned assessments mean that all parts of the investigation area have been allocated a probability categorisation and a consequence categorisation. The combination of these two categories forms a pair that determines the risk category. The classification can also be seen in a matrix, showing the ratio between different risk categories, see Figure 2.7.
Figure 2.7 Matrix for risk categories with examples of a combination of probability category and consequence category.
P ro b a b il it y c a te g o ry
S5
Substantial
S4
High
S3
Moderate
S2
Low
S1
Negligible
K1
Minor
K2
Serious
K3
Very serious
K4
Extremely serious
K5
Catastro- phic
Consequence category
S5/K5
Substantial probability of a
landslide with catastrophic consequences
S2/K2
Low probability of a landslide with
serious consequences
SGI Publication 38–1E
2.3.4 Landslide risk levels
To simplify the risk classification, the risk categories are grouped on three risk levels, comprising a number of categories that are equivalent to similar landslide risks, see Figure 2.8.
Probability category
S5 S5/K1 S5/K2 S5/K3 S5/K4 S5/K5
S4 S4/K1 S4/K2 S4/K3 S4/K4 S4/K5
S3 S3/K1 S3/K2 S3/K3 S3/K4 S3/K5
S2 S2/K1 S2/K2 S2/K3 S2/K4 S2/K5
S1 S1/K1 S1/K2 S1/K3 S1/K4 S1/K5
K1 K2 K3 K4 K5
Consequence category
Figure 2.8 Matrix showing landslide risk levels.
The risk levels are expressed as low, moderate and high, see Figure 2.4 for a more detailed description of the levels.
Areas with a low landslide risk Areas with a moderate landslide risk Areas with a high landslide risk
2.3.5 Risk of uphill progressive landslides and secondary effects
Within areas with quick clay, uphill progressive landslides can occur, with large areas affected as the landslide progresses. An event of this nature could lead to secondary consequences in addition to land loss and damage. Examples of potential secondary effects are the damming of a watercourse (or its subsidiaries) and flood waves, which could vary in magnitude depending on the volume of the landslide mass. These secondary consequences can, if the area affected is sufficiently large, supersede the primary consequences. The secondary consequences cannot be predicted with any certainty and have therefore not been addressed in this investigation.
SGI Publication 38–1E
26 (33)
2.3.6 Climate impact
Expected climate changes will entail higher river flow levels in the future, resulting in increased erosion of slopes and river beds as well as higher groundwater levels and porewater pressure. This will have an impact on the probability of a landslide occurring mainly due to the fact that the geometry of the slopes will be changed within parts of the investigation area. The effects of the probability are designated ‘sensitivity to climate impact’ and have been placed into three
categories: low, moderate and high. Sensitivity to climate impact is reported on the probability and landslide risk maps using different line types along the banks. Figure 2.4 contains a further
description of the sensitivity to climate impact.
Depending on which category an area falls into, climate change could lead to the probability category increasing within affected areas along the Säveån river valley. In areas that fall within the highest probability category (S5), even a low level of impact caused by climate change could result in a landslide.
2.3.7 Digital documentation
In the investigation, a large number of external documentary datasets have been gathered and used.
A good deal of data has been collected via the involvement by SGI in Geodata Collaboration, such as the Swedish Mapping, Cadastral and Land Registration Authority maps and the National Height Database. During the investigation, a whole range of new data and results have been produced.
Collection of documentation included entering the digital material into databases. One of the aims has been to gather data in GIS format. The data has been processed mainly in a GIS environment, using the coordinate systems SWEREF99TM (level) and RH2000 (height). During the course of the work, use has been made of software from ESRI ArcGIS and QGIS.
The end-results, in the form of landslide risks, probability categories and consequence categories, have been made available in various forms, including a map-viewing service that is open to external users. In the map window, selected documentary data is shown that could be of benefit when using the results from the investigation. The documentary data that has been produced by bodies other than SGI is shown in what is known as a WMS version from the data producer.
Figure 2.9 shows a segment from the map presentation service that has been produced as part of the investigation of the Norsälven river valley, and which has been supplemented with results for the Säveån river valley.
SGI Publication 38–1E
Figure 2.9 Section from the map presentation service, taking the Säveån river valley as an example.
https://gis.swedgeo.se/skredriskkarteringar/
SGI Publication 38–1E
28 (33)
3. Recommendations for future work
Proposals for measures for individual areas have not been produced within the framework of this assignment. In order to produce proposals for measures, more detailed investigations are required along with more advanced stability investigations. Below are a number of the more general proposals for how to proceed with this task. Further information about results and interpretation of the maps is provided in Part 2, Chapter 12 (SGI 2017b).
General recommendations for the investigation area along the Säveån river valley:
• Identified areas with a high risk of a landslide, and where sufficiently detailed studies have not been carried out, ought to be investigated further.
• When investigating measures, it is important to take account of the existence of quick clay, which would affect the spread of a landslide. Measures should also take climate change into account.
• An overview of existing erosion protection ought to be carried out, particularly within areas with moderate to high sensitivity to climate impact, and with account taken of maintenance and possible supplementary measures.
• Sustainable urban development can benefit by the municipal authority gathering information about the stability measures that have been undertaken. Information is needed about when, where and how measures have been carried out, as well as maintenance requirements.
• Climate adaptation measures should be coordinated
3.1 Societal planning and construction
The landslide risk plan has a resolution that is suitable for use on the comprehensive planning level.
It is also possible to weigh it together with other natural disaster risks, such as the flood risk, which is recommended.
In the case of local plans and building permits, more detailed geotechnical investigations need to be carried out, and with due account taken of the construction and the facilities that will be permitted
Measures to prevent landslides are an important part of the urban planning process, both with regard to the present-day climate and climate change in the long term. The landslide risk maps produced provide an illustration of the location of sensitive areas to proceed with and investigate in more detail. This is achieved by mapping the geotechnical conditions and the consequences that could arise in conjunction with a landslide. As part of this process,
account must be taken of a ‘probable’ landslide spread, as the consequences of the landslide are the sum of the consequences within the probable landslide area. The maps also indicate which areas could be affected most in conjunction with a change in climate as a result of a deterioration in stability conditions.
The results of this investigation can also be used to influence the location and implementation of new development and other activities in a way that the risk of a landslide can be prevented.
To mitigate the risk of a landslide along the Säveån river valley, measures can be taken to reduce the probability and/or the consequences. Probability-mitigation measures are often physical in the form of excavation, pressure berms and erosion protection. Consequence- mitigation measures could also include moving buildings and facilities.
SGI Publication 38–1E
within the framework of the plans/building permit. This is done to avoid increasing the probability of a landslide in the area in question, and to ensure the stability of new construction and existing development within the area covered by the local plan. According to European geotechnical standards, different levels of detail are required depending on the purpose. For this reason, geotechnical stability investigations are ideally carried out in stages, with an increasing level of detail. See also IEG 2010 for planning and IEG 2008 for planning and construction.
It is also important to bear in mind the fact that the risk level may increase if the consequences increase, i.e. if land is developed with the addition of buildings or facilities. Other land use changes could also be of significance to the landslide risk level. Examples include neglected maintenance of land drainage or dewatering as a result of a change in land use. This could result in a build-up of water pressure in the ground, which could increase the probability of a landslide.
3.2 Climate adaptation measures
Prior to making a decision about risk-mitigation measures, a quantitative cost-benefit analysis ought to be made by weighing up the cost of a particular measure against the benefit. In that case, the probability of a landslide and the consequences can be investigated in greater detail for a specified area prior to deciding about whether measures should be directed at mitigating the probability and/or the consequences. Physical probability-mitigation measures in the form of excavation, stabilizing berms, river bank protection, soil reinforcement and other similar measures, are often costly. Alternatives that reduce the consequences could in some instances be of greater benefit. These alternatives could include compulsory purchase/demolition of a building on a property, or the imposition of land use restrictions.
Preventive measures to counteract natural disasters could sometimes work against each other. A protective wall designed to prevent flooding could generate a load on the soil layers, thus reducing stability. It is therefore advisable to coordinate different climate adaptation measures.
Erosion protection and other preventive measures in watercourses could in many cases have a negative impact on the natural values that exist within and along watercourses. Traditionally, hard river bank protection has been employed, such as rip-rap using blasted rock or concrete blocks, although in later years, this and other methods have been called into question. Hard protection measures offer effective protection against erosion at the specific location where they are placed, but often result in the problem being shifted to another location. What is being sought nowadays in terms of erosion prevention is taking greater account of the natural values that exist within and along watercourses. There is every reason to investigate the conditions for naturally-adapted erosion protection that is beneficial to the environment and at the same time resistant to erosion (SGI 2016). Another important aspect is to analyse a long stretch of river to determine exactly which sections would be affected if erosion were to be prevented somewhere else and how this would take place. An example of combined erosion protection, where hard sections are combined with soft sections in order to blend in well with the surrounding natural environment and yet still remain more resistant when water levels and flow rates are higher, is shown in Figure 3.1.
SGI Publication 38–1E
30 (33)
Figure 3.1 Combined river bank protection (bushes and trees mixed with rip-rap).
Illustration: K. Gellerstam. (SGI 2016)
Figures 3.2 – 3.4 show examples of implemented and potential stability-enhancement measures.
Figure 3.2 Current (April 7, 2017) stability-enhancement work at Kvibergs Ängar in Gothenburg. River bank protection. Photo: SGI.
SGI Publication 38–1E
Figure 3.3 Current (April 7, 2017) stability-enhancement work at Kvibergs Ängar in Gothenburg. Excavation of soil and replacement with lightweight filling material. Photo: SGI.
SGI Publication 38–1E
32 (33)
UTFLACKNING AV SLÄNT
Figure 3.4 Examples of different measures to improve slope stability. Illustrations from Landslide Commission Report 5:95.
Support fill filltödfyllning
Soil removal
Slope flattening
Lime-cement column stabilisation
Soil nailing Erosion protection
Replacement of eroding material
SGI Publication 38–1E
References
Berggren, B, Alén, C, Bengtsson P-E & Falemo, S 2011, Metodbeskrivning sannolikhet för skred:
kvantitativ beräkningsmodell. Statens geotekniska institut, Göta älvutredningen, GÄU Delrapport 28, Linköping.
IEG 2008, Tillämpningsdokument, EN 1997-1 Kapitel 11 och 12, Slänter och bankar,
Implementeringskommissionen för Europastandarder inom Geoteknik, Rapport 6:2008, Rev 1, Stockholm.
IEG 2010, Tillståndsbedömning/klassificering av naturliga slänter och slänter med befintlig bebyggelse och anläggningar, Vägledning för tillämpning av Skredkommissionens rapporter 3:95 och 2:96 (delar av), Implementeringskommissionen för Europastandarder inom
Geoteknik, Rapport 4:2010, Stockholm.
SGI 2012, Skredrisker i Göta älvdalen i ett förändrat klimat, Statens geotekniska institut, Linköping.
SGI 2013, Prioritering av områden för skredriskanalys, Klimatanpassningsanslag 2013, Statens geotekniska institut, SGI Publikation 6, Linköping.
SGI 2015, Skredrisker i ett förändrat klimat – Norsälven, Statens geotekniska institut, SGI Publikation 18-1--4, Linköping.
SGI 2016, Naturanpassade erosionsskydd i vattendrag - En förstudie, Statens geotekniska institut, SGI Publikation 28, Linköping.
SGI 2017b, Skredrisker i ett förändrat klimat – Säveån, Del 2: Metodik för kartläggning, Statens geotekniska institut, SGI Publikation 38-2, Linköping.
SGI 2017c, Skredrisker i ett förändrat klimat – Säveån, Del 3: Fördjupningsbilaga konsekvensanalys, Statens geotekniska institut, SGI Publikation 38-3, Linköping.
Skredkommissionen 1995, Anvisningar för stabilitetsutredningar, Information, Skredkommissionen, Rapport 5:95, Linköping.
SGI Publication 38–1
Appendices
Division of map sheets (Bladindelning) Landslide risk maps (Skredriskkartor) Probability maps (Sannolikhetskartor) Consequence maps (Konsekvenskartor)
Blad 9
Blad 5 Blad 4
Blad 1 Blad 2 Blad 3
Blad 8 Blad 6 Blad 7
Blad 11 Blad 10
Göteborg
Partille
Lerum
1 0,5 0 1 2 3 4
km
±
Skredrisker i ett förändrat klimat - Säveån
Bladindelning av kartor
Landslide risk maps
E
E E
E E E E E
E E
E E
E
E E
E E
E E E E E E E
E E E E
E
E E E E E
S1/ K4 S1/ K1
S1/ K4
S1/ K1
S1/ K4
S1/ K4
S1/ K3
S5/ K1
S4/ K1
S4/ K1
S5/ K4 S1/ K3
S1/ K1
S1/ K1 S1/ K5
S1/ K3
S1/ K5 S4/ K1
S4/ K1
S1/ K1 S1/ K3
S1/ K1
S1/ K1 S1/ K4
S4/ K1
S5/ K3 S1/ K4
S5/ K1
S4/ K4
S1/ K2
S1/ K3
S4/ K4
S1/ K3
S1/ K4 S5/ K3
S1/ K1
S4/ K3
S1/ K1
S5/ K4
S1/ K1
S4/ K5
S5/ K1
S1/ K4
S1/ K4
S5/ K2
S4/ K4
S4/ K4
S4/ K1 S4/ K5
S5/ K4
S5/ K1
S5/ K4
S5/ K1
S1/ K5
S1/ K3
S1/ K3
S5/ K3
S5/ K3
S4/ K1
S1/ K1
S1/ K4
S5/ K4
S4/ K4
S4/ K1 S2/ K1
S5/ K1
S4/ K4
S4/ K4
S4/ K4
S4/ K4 S3/ K1
S4/ K4 S1/ K2
S4/ K4
S4/ K4
S3/ K4
S5/ K2
S3/ K4
S1/ K2
S3/ K1
S2/ K4 S1/ K2
S3/ K3
S1/ K2
S5/ K3
S5/ K3 S3/ K4
S2/ K1
S2/ K1
S4/ K4
S4/ K4
S3/ K1 S4/ K3
S5/ K2 S5/ K2
S3/ K4
S5/ K2
S2/ K4
S2/ K4 S3/ K4
S3/ K1
S2/ K4
S3/ K4
S2/ K1
S3/ K1
S2/ K4
S2/ K3
S3/ K3
S3/ K4 S2/ K1
S2/ K3
S3/ K1 S3/ K3
S3/ K3 S2/ K3
S3/ K1
S3/ K4
S3/ K3 4000
3000
2000
Skredrisker i ett förändrat klimat - Säveån.
Översiktligt planeringsunderlag för klimatanpassning.
Skredrisknivå
HÖG MEDEL LÅG
Ej klassat område
Klimatpåverkan
Liten Måttlig Stor
Underlag
E Längdmätning SGI Utredningsområde SGI
Fast mark Kommungräns
Ungefärligt område med detaljerad/fördjupad stabilitetsutredning
0 50100 200 300 400
±
Skala 1:10 000 (A4)
Blad 1
S1/K1 Sannolikhetsklass/Konsekvensklass
