<p>Large areas prone to clay slides are present in Canada, Sweden and Norway. Traditionally, mainly geotechnical approaches have been used to solve clay-slide issues. However, there is great potential to utilise interdisciplinary studies in order to prevent or reduce possible slide damage. The present study combines geology, morphology and geophysical investigations, with geotechnical data. The purpose is to better understand landscape development, formation of quick clay and sliding in clay.</p><p>Thick, marine clay deposits in valleys along the Norwegian coast are occasionally subjected to large landslides involving quick clay. The development of quick clay is closely related to the deglaciation history of the coastal parts of Norway. During and after the last ice age, up to several hundred metres of glaciomarine and marine deposits accumulated in Norwegian fjords. These sediments were subsequently exposed on land due to glacioisostatic rebound. Leaching of salt from the marine clay by groundwater resulted in the development of quick-clay layers or pockets, which completely liquefy when remoulded.</p><p>The study area of this thesis is the small valley Buvika, located in the Trondheimsfjorden area, Mid Norway. Buvika is characterised by its undulating terrain with numerous slide scars and ravines. There are thick occurrences of quick clay in the subsurface.</p><p>The existing geotechnical and geophysical data, combined with sedimentology, structural geology, morphology, geophysical results, and 14C-datings, have given input to the understanding of the landscape development of the study area. From this, a deglaciation history has been deduced, indicating at least one, and possibly two, minor glacier re-advances in Late Allerød/Early Younger Dryas time. This implies that there have been more and larger ice-front oscillations in the study area than earlier documented.</p><p>The lowered relative sea level led to incision by rivers accompanied by numerous slides involving quick clay. The erosion pattern of a valley filled with glaciomarine and marine deposits can be quite complex, but careful analyses have helped outlining the interplay between river and ravine incision, groundwater erosion and sliding. The study of sediments and structures in large excavated sections have resulted in the detection of slide material from old flake-type slides, where only a thin layer of quick clay acted as a slide plane. Younger slide scars cutting into theses older slide deposits show further quick-clay development. Mapping of the morphology in Buvika has identified numerous slide scars and ravines. A relative chronology of slide events has been established based upon the slide scars’ position in the terrain and/or results from 14C-datings of terrestrial organic material. Most of the historical slide scars are located in the northern part of the valley.</p><p>Detailed mapping of the quick-clay extent is of great interest for planning and protection purposes, as the position of quick clay within slopes has a major impact on the landslide risk. In this study, the resistivity method is found to be potentially well suited for outlining quick-clay occurrences since quick clay has a slightly higher electrical resistivity (10-80 Ωm) than intact unleached clay (1-10 Ωm). This is due to a higher salt content in the latter. These relationships are supported by pore water salt content measurements. The resistivity profiles that were acquired show good correlation with other geophysical data and geotechnical drillings. However, the resistivity method must be combined with other investigations, since both leached, non-quick clay and silty, non-sensitive material may give resistivity values of the same range as quick clay.</p><p>The stratigraphy of an area strongly influences the landscape development. It determines the morphology, such as ravine development, and size, shape and distribution of slides. To the east of Buvika, thick and frequent layers of sand and gravel in the dominant clay deposits drain the slopes, leading to development of deeply incising ravines. To the south and north, thinner layers of coarse material in the clay lead to pore-pressure build-ups and quick-clay development, resulting in numerous slide scars.</p>

Large areas prone to clay slides are present in Canada, Sweden and Norway. Traditionally, mainly geotechnical approaches have been used to solve clay-slide issues. However, there is great potential to utilise interdisciplinary studies in order to prevent or reduce possible slide damage. The present study combines geology, morphology and geophysical investigations, with geotechnical data. The purpose is to better understand landscape development, formation of quick clay and sliding in clay. Thick, marine clay deposits in valleys along the Norwegian coast are occasionally subjected to large landslides involving quick clay. The development of quick clay is closely related to the deglaciation history of the coastal parts of Norway. During and after the last ice age, up to several hundred metres of glaciomarine and marine deposits accumulated in Norwegian fjords. These sediments were subsequently exposed on land due to glacioisostatic rebound. Leaching of salt from the marine clay by groundwater resulted in the development of quick-clay layers or pockets, which completely liquefy when remoulded. The study area of this thesis is the small valley Buvika, located in the Trondheimsfjorden area, Mid Norway. Buvika is characterised by its undulating terrain with numerous slide scars and ravines. There are thick occurrences of quick clay in the subsurface. The existing geotechnical and geophysical data, combined with sedimentology, structural geology, morphology, geophysical results, and 14C-datings, have given input to the understanding of the landscape development of the study area. From this, a deglaciation history has been deduced, indicating at least one, and possibly two, minor glacier re-advances in Late Allerød/Early Younger Dryas time. This implies that there have been more and larger ice-front oscillations in the study area than earlier documented. The lowered relative sea level led to incision by rivers accompanied by numerous slides involving quick clay. The erosion pattern of a valley filled with glaciomarine and marine deposits can be quite complex, but careful analyses have helped outlining the interplay between river and ravine incision, groundwater erosion and sliding. The study of sediments and structures in large excavated sections have resulted in the detection of slide material from old flake-type slides, where only a thin layer of quick clay acted as a slide plane. Younger slide scars cutting into theses older slide deposits show further quick-clay development. Mapping of the morphology in Buvika has identified numerous slide scars and ravines. A relative chronology of slide events has been established based upon the slide scars’ position in the terrain and/or results from 14C-datings of terrestrial organic material. Most of the historical slide scars are located in the northern part of the valley. Detailed mapping of the quick-clay extent is of great interest for planning and protection purposes, as the position of quick clay within slopes has a major impact on the landslide risk. In this study, the resistivity method is found to be potentially well suited for outlining quick-clay occurrences since quick clay has a slightly higher electrical resistivity (10-80 Ωm) than intact unleached clay (1-10 Ωm). This is due to a higher salt content in the latter. These relationships are supported by pore water salt content measurements. The resistivity profiles that were acquired show good correlation with other geophysical data and geotechnical drillings. However, the resistivity method must be combined with other investigations, since both leached, non-quick clay and silty, non-sensitive material may give resistivity values of the same range as quick clay. The stratigraphy of an area strongly influences the landscape development. It determines the morphology, such as ravine development, and size, shape and distribution of slides. To the east of Buvika, thick and frequent layers of sand and gravel in the dominant clay deposits drain the slopes, leading to development of deeply incising ravines. To the south and north, thinner layers of coarse material in the clay lead to pore-pressure build-ups and quick-clay development, resulting in numerous slide scars.

<p>The use of improved equipment and methodology can result in considerable reductions in the drilling costs for medium- to large sized ground source heat pump system in crystalline bedrock. The main point has been to use special techniques within hydraulic fracturing to create a larger heat exchange area in the bedrock, and thus a greater energy extraction per borehole. The energy extraction is based on circulating groundwater.</p><p>Stimulation with hydraulic fracturing is a well known technique in order to improve borehole yields for drinking water-, oil-, and geothermal purposes. A procedure for injection of propping agents in selected borehole sections, and custom-made equipment for hydraulic fracturing in crystalline bedrock, a double packer, have been developed in this study. The propping agents are likely to ensure a permanent improvement of the hydraulic conductivity in a long-run perspective. </p><p>In addition to a pre-test, a comprehensive test programme has been performed at each of the two pilot plants at Bryn and at the former property of Energiselskapet Asker og Bærum (EAB) in Bærum municipality outside Oslo, Norway. A total of 125 stimulations with hydraulic fracturing using water-only and hydraulic fracturing with injection of sand have been performed in 9 boreholes. Test pumping and geophysical logging (temperature, electrical conductivity, gamma radiation, optical televiewer and flow measurements) have been carried out in order to document the effect of the hydraulic fracturing.</p><p>The pilot plants at Bryn and EAB, where the ground source heat pump systems are based on circulating groundwater, have demonstrated the short-period energy extraction, limitations and opportunities of the concept for hydraulic fracturing and increased energy extraction in different geological and hydrogeological areas. The bedrock at Bryn and EAB is characterized as a low-metamorphic sandstone and a nodular limestone, respectively. At Bryn, the five boreholes were organised with a central borehole encircled by four satellite boreholes 13 metres away from the central borehole. The central borehole at EAB was flanked with two boreholes 16 and 20 metres away. In operation mode, groundwater was pumped from the satellite boreholes, heat exchanged, and re-injected into the groundwater magazine via the central borehole. Routine operation of the plants has not yet been initiated.</p><p>The main findings from this study can be summarized as follows:</p><p>•Hydraulic fracturing with water-only results in an overall increase in water yield for the hard rock borehole.</p><p>•Hydraulic fracturing with injection of sand as propping agent also leads to an increased water yield.</p><p>•The use of sand as propping agent seems to be more required in fractures with high counter pressure, in this study higher than approximately 40 bars, compared with fractures with lower counter pressure. The particle size of the sand should also be adjusted to the appearing counter pressure, and injection of coarser sand is recommended in fractures with lower counter pressures.</p><p>•Comparing the results from the hydraulic fracturing performed at Bryn and EAB shows that the pressure levels, required to create new fractures, varied considerably. The maximum pressures present at Bryn were higher than the corresponding pressures at EAB. At Bryn 70% (44 out of 63) of the pressure-time curves from the hydraulic fracturing with water-only were interpreted as initiation or reopening of fractures, while the number for EAB was 97% (36 out of 37). The lower degree of fracturing at Bryn is likely to be a result of high rock stresses and high tensile strength of the bedrock, also confirmed by the results from the rock stress measurements performed at Bryn. Considering the bedrock at EAB, characterized as nodular limestone, the tensile strength is assumed to be less than the values for the low-metamorphic sandstone present at Bryn.</p><p>•The infiltration rate in the central boreholes is a critical factor for the energy extraction and a successful operation of ground source heat pump systems based on circulating groundwater. Results from the short-period circulation tests accomplished at Bryn and EAB show that the infiltration rate in the central borehole at Bryn (approximately 2500 litres/hour) was too low to obtain a satisfactory operation of the plant, while the infiltration rate at EAB (14000 litres/hour) was sufficient to achieve profitability. Under the actual conditions, a reduction in the construction costs, i.e. the drilling costs, for a conventional ground source heat pump system with single U-collectors in vertical boreholes, of more than 50% were achieved for the pilot plant at EAB when the energy extraction from water is more than 105 MWh. The large difference in the infiltration rate between Bryn and EAB was probably related to: (1) Large initial differences in the borehole yield prior to hydraulic fracturing (<560 litres/hour at Bryn and >6300 litres/hour at EAB). Nodular limestone generally has high permeability, while compact sandstone rocks are expected to have low permeability. (2) Hydraulic fracturing was most successful at EAB. (3) The higher rock stress level present at Bryn compared to EAB will increase the tendency to tighten the opened fractures, even the fractures with injected sand •The FEFLOW-modelling of the pilot plant at Bryn and EAB emphasized the important relation between the available heat exchange area in the bedrock, the thermal conductivity of the bedrock, and the energy potential. </p><p>•The increased borehole yields achieved by hydraulic fracturing in this study, and the improved, reliable and cost-effective hydraulic fracturing techniques in crystalline bedrock, will probably increase the interest for groundwater as a domestic water supply for small- to medium sized water works.</p>

The use of improved equipment and methodology can result in considerable reductions in the drilling costs for medium- to large sized ground source heat pump system in crystalline bedrock. The main point has been to use special techniques within hydraulic fracturing to create a larger heat exchange area in the bedrock, and thus a greater energy extraction per borehole. The energy extraction is based on circulating groundwater. Stimulation with hydraulic fracturing is a well known technique in order to improve borehole yields for drinking water-, oil-, and geothermal purposes. A procedure for injection of propping agents in selected borehole sections, and custom-made equipment for hydraulic fracturing in crystalline bedrock, a double packer, have been developed in this study. The propping agents are likely to ensure a permanent improvement of the hydraulic conductivity in a long-run perspective. In addition to a pre-test, a comprehensive test programme has been performed at each of the two pilot plants at Bryn and at the former property of Energiselskapet Asker og Bærum (EAB) in Bærum municipality outside Oslo, Norway. A total of 125 stimulations with hydraulic fracturing using water-only and hydraulic fracturing with injection of sand have been performed in 9 boreholes. Test pumping and geophysical logging (temperature, electrical conductivity, gamma radiation, optical televiewer and flow measurements) have been carried out in order to document the effect of the hydraulic fracturing. The pilot plants at Bryn and EAB, where the ground source heat pump systems are based on circulating groundwater, have demonstrated the short-period energy extraction, limitations and opportunities of the concept for hydraulic fracturing and increased energy extraction in different geological and hydrogeological areas. The bedrock at Bryn and EAB is characterized as a low-metamorphic sandstone and a nodular limestone, respectively. At Bryn, the five boreholes were organised with a central borehole encircled by four satellite boreholes 13 metres away from the central borehole. The central borehole at EAB was flanked with two boreholes 16 and 20 metres away. In operation mode, groundwater was pumped from the satellite boreholes, heat exchanged, and re-injected into the groundwater magazine via the central borehole. Routine operation of the plants has not yet been initiated. The main findings from this study can be summarized as follows: •Hydraulic fracturing with water-only results in an overall increase in water yield for the hard rock borehole. •Hydraulic fracturing with injection of sand as propping agent also leads to an increased water yield. •The use of sand as propping agent seems to be more required in fractures with high counter pressure, in this study higher than approximately 40 bars, compared with fractures with lower counter pressure. The particle size of the sand should also be adjusted to the appearing counter pressure, and injection of coarser sand is recommended in fractures with lower counter pressures. •Comparing the results from the hydraulic fracturing performed at Bryn and EAB shows that the pressure levels, required to create new fractures, varied considerably. The maximum pressures present at Bryn were higher than the corresponding pressures at EAB. At Bryn 70% (44 out of 63) of the pressure-time curves from the hydraulic fracturing with water-only were interpreted as initiation or reopening of fractures, while the number for EAB was 97% (36 out of 37). The lower degree of fracturing at Bryn is likely to be a result of high rock stresses and high tensile strength of the bedrock, also confirmed by the results from the rock stress measurements performed at Bryn. Considering the bedrock at EAB, characterized as nodular limestone, the tensile strength is assumed to be less than the values for the low-metamorphic sandstone present at Bryn. •The infiltration rate in the central boreholes is a critical factor for the energy extraction and a successful operation of ground source heat pump systems based on circulating groundwater. Results from the short-period circulation tests accomplished at Bryn and EAB show that the infiltration rate in the central borehole at Bryn (approximately 2500 litres/hour) was too low to obtain a satisfactory operation of the plant, while the infiltration rate at EAB (14000 litres/hour) was sufficient to achieve profitability. Under the actual conditions, a reduction in the construction costs, i.e. the drilling costs, for a conventional ground source heat pump system with single U-collectors in vertical boreholes, of more than 50% were achieved for the pilot plant at EAB when the energy extraction from water is more than 105 MWh. The large difference in the infiltration rate between Bryn and EAB was probably related to: (1) Large initial differences in the borehole yield prior to hydraulic fracturing (&lt;560 litres/hour at Bryn and &gt;6300 litres/hour at EAB). Nodular limestone generally has high permeability, while compact sandstone rocks are expected to have low permeability. (2) Hydraulic fracturing was most successful at EAB. (3) The higher rock stress level present at Bryn compared to EAB will increase the tendency to tighten the opened fractures, even the fractures with injected sand •The FEFLOW-modelling of the pilot plant at Bryn and EAB emphasized the important relation between the available heat exchange area in the bedrock, the thermal conductivity of the bedrock, and the energy potential. •The increased borehole yields achieved by hydraulic fracturing in this study, and the improved, reliable and cost-effective hydraulic fracturing techniques in crystalline bedrock, will probably increase the interest for groundwater as a domestic water supply for small- to medium sized water works.

<p>In co-operation with the Geological Survey of Norway, 395 samples of topsoil, 44 bedrock samples and 24 samples of sand from sandboxes at kindergartens taken in Oslo, the capital city of Norway, were analysed for a wide range of elements. To get a measure on the bio-availability of arsenic, a five stage sequential extraction procedure was accomplished on a subset of 98 topsoil samples and on all 24 samples of sand from sandboxes. These fractions (i.e. “Exchangeable”, “Carbonate”, “Reducible substrate”, “Organic” and “Residual”) were analysed for arsenic. Bulk concentrations in a representative subset of 10 bedrock samples were determined by XRF analysis.</p><p>For the concentrations of the elements As, B, Ba, Cd, Cr, Cu, Hg, Mn, Mo, Ni, P, Pb and Zn in soil, geologic sources are important, but human activity has had a large effect on concentrations in topsoil, especially in the central districts. At least 81% of the samples contain more arsenic than the Norm value for polluted soil used by the Norwegian authorities. For chromium, zinc and lead, the percentages are 65, 74 and 25 respectively. In topsoil, up to 50% of the arsenic is extractable in the exchangeable, the carbonate and the reducible substrate fractions, but only 2% of the samples exceeded the Norm value after the three extraction stages. After four stages of extraction, 69% of the samples exceeded the Norm value.</p><p>At 24 kindergartens in the central districts of Oslo, concentrations of As, Cd, Cr, Cu, Ni, Pb and Zn in topsoil samples are higher than in samples of sand from sandpits, but differences in arsenic are low. All topsoil samples and 63% of the sandbox samples exceed the arsenic Norm value for polluted soil used by the Norwegian authorities. Concentrations of at least one of the elements Cr, Cu, Ni, Pb and Zn exceed the Norm value in the topsoil. At two kindergartens all elements, except Cd, exceed the Norm value in the topsoil.</p><p>In the samples from the sandboxes, arsenic concentrations in the exchangeable, the carbonate and the reducible substrate fractions, lie at a higher level than in topsoil samples. Leakage of arsenic compounds from impregnated and painted wood have probably caused elevated concentrations of this element in the sand. At two kindergartens the cumulative arsenic content in samples from the sandboxes exceeds the Norm value of 2 mg/kg As for polluted soil already after two fractions (i.e. the “Exchangeable” and the “Carbonate” fraction). After three fractions, the arsenic content in sand at 5 kindergartens exceeds this value. After four fractions this number was still 5. Regarding soil, no kindergartens had a cumulative arsenic content after three stages of the extraction procedure. After four stages 16 kindergartens contained more arsenic than the Norwegian Norm value of 2 mg/kg As.</p><p>Bulk concentrations of metals, in the bedrock samples, are generally higher than the acid soluble concentrations. Differences are small for copper and zinc, somewhat higher for lead and nickel and considerably different for chromium. Acid soluble chromium concentrations in soil differ little when compared with acid soluble concentrations in bedrock, but are in general much lower than bulk concentrations in rock. Chromium is probably strongly incorporated in the mineral structure of rock samples and concentrated HNO3 cannot solubilize all chromium molecules. For copper, lead and zinc, the acid soluble concentrations in soil are higher than the bulk concentrations in bedrock. The parent rock material cannot be the only source for these elements in soils and various anthropogenic sources dispersed by air or water probably have major influence.</p><p>The enrichment of nickel in topsoil near highways in Oslo might be due to the bitumen content of road dust. The enrichment of chromium and copper might be due to wear and tear of cars and engine parts. There is no enrichment of cadmium and zinc in topsoil near highways. Concentrations of zinc in topsoil in Oslo are generally high, most probably because of industry. The possible addition of zinc from traffic to the topsoil is not observed, in this study.</p>

In co-operation with the Geological Survey of Norway, 395 samples of topsoil, 44 bedrock samples and 24 samples of sand from sandboxes at kindergartens taken in Oslo, the capital city of Norway, were analysed for a wide range of elements. To get a measure on the bio-availability of arsenic, a five stage sequential extraction procedure was accomplished on a subset of 98 topsoil samples and on all 24 samples of sand from sandboxes. These fractions (i.e. “Exchangeable”, “Carbonate”, “Reducible substrate”, “Organic” and “Residual”) were analysed for arsenic. Bulk concentrations in a representative subset of 10 bedrock samples were determined by XRF analysis. For the concentrations of the elements As, B, Ba, Cd, Cr, Cu, Hg, Mn, Mo, Ni, P, Pb and Zn in soil, geologic sources are important, but human activity has had a large effect on concentrations in topsoil, especially in the central districts. At least 81% of the samples contain more arsenic than the Norm value for polluted soil used by the Norwegian authorities. For chromium, zinc and lead, the percentages are 65, 74 and 25 respectively. In topsoil, up to 50% of the arsenic is extractable in the exchangeable, the carbonate and the reducible substrate fractions, but only 2% of the samples exceeded the Norm value after the three extraction stages. After four stages of extraction, 69% of the samples exceeded the Norm value. At 24 kindergartens in the central districts of Oslo, concentrations of As, Cd, Cr, Cu, Ni, Pb and Zn in topsoil samples are higher than in samples of sand from sandpits, but differences in arsenic are low. All topsoil samples and 63% of the sandbox samples exceed the arsenic Norm value for polluted soil used by the Norwegian authorities. Concentrations of at least one of the elements Cr, Cu, Ni, Pb and Zn exceed the Norm value in the topsoil. At two kindergartens all elements, except Cd, exceed the Norm value in the topsoil. In the samples from the sandboxes, arsenic concentrations in the exchangeable, the carbonate and the reducible substrate fractions, lie at a higher level than in topsoil samples. Leakage of arsenic compounds from impregnated and painted wood have probably caused elevated concentrations of this element in the sand. At two kindergartens the cumulative arsenic content in samples from the sandboxes exceeds the Norm value of 2 mg/kg As for polluted soil already after two fractions (i.e. the “Exchangeable” and the “Carbonate” fraction). After three fractions, the arsenic content in sand at 5 kindergartens exceeds this value. After four fractions this number was still 5. Regarding soil, no kindergartens had a cumulative arsenic content after three stages of the extraction procedure. After four stages 16 kindergartens contained more arsenic than the Norwegian Norm value of 2 mg/kg As. Bulk concentrations of metals, in the bedrock samples, are generally higher than the acid soluble concentrations. Differences are small for copper and zinc, somewhat higher for lead and nickel and considerably different for chromium. Acid soluble chromium concentrations in soil differ little when compared with acid soluble concentrations in bedrock, but are in general much lower than bulk concentrations in rock. Chromium is probably strongly incorporated in the mineral structure of rock samples and concentrated HNO3 cannot solubilize all chromium molecules. For copper, lead and zinc, the acid soluble concentrations in soil are higher than the bulk concentrations in bedrock. The parent rock material cannot be the only source for these elements in soils and various anthropogenic sources dispersed by air or water probably have major influence. The enrichment of nickel in topsoil near highways in Oslo might be due to the bitumen content of road dust. The enrichment of chromium and copper might be due to wear and tear of cars and engine parts. There is no enrichment of cadmium and zinc in topsoil near highways. Concentrations of zinc in topsoil in Oslo are generally high, most probably because of industry. The possible addition of zinc from traffic to the topsoil is not observed, in this study.

No description available.

No description available.

För att förnya och förbättra den spårbundna trafiken inom Sverige är stambanor för höghastighetståg ett alternativ som utreds då det skulle innebära förkortade transporttider både för person- och godstrafik. Innan höghastighetsbanor kan etableras i Sverige finns det många faktorer som måste utredas. Eftersom tågen går med betydligt högre hastigheter än de tåg som redan används blir känsligheten för problem som kan uppstå i undergrunden betydligt högre och toleransen för sättningar och tjällyft är i princip lika med noll. På grund av klimatet som råder i Sverige finns det mycket att reda ut angående tjälproblematiken. Förutom låga temperaturer och vatten i jorden kan en trumma påverka hur tjälen utvecklas i järnvägsbanken. I den här studien analyseras påverkan som en trumma i järnvägsbanken har på tjälutvecklingen. Inverkan har analyserats utifrån ett tvärsnitt inuti järnvägsbanken för två olika köldmängder, -500 graddagar och -1300 graddagar. Innertemperaturen i trumman har antagits variera längs med trummans längd. Temperaturen har varierats inne i trumman att vara från 10 % upp till 75 % av köldmängden i uteluften. Förutom att analysera hur olika temperaturer påverkar tjälutvecklingen har den mängd material som behöver schaktas bort och ersättas med ett isoleringsmaterial för att förhindra att tjälproblem uppstår beräknats. Oberoende av trummans storlek och temperatur kommer trumman inte att kunna placeras direkt på terrassnivån utan att tjälfronten når det tjälfarliga materialet och risk för att tjälproblem ska uppstå. När temperaturen inne i trumman är 75% av köldmängden utomhus, och därmed nära temperaturen utomhus, blir tjälfronten även brantare än för lägre innertemperaturer. En brant tjälfront kan inducera större och mer ojämna tjällyft. För att undvika att problem uppstår måste ett isoleringslager av ett tjälpassivt material anläggas. Mängden material som måste ersättas ökar med trumdiameter och ökar betydligt beroende på om temperaturen inne i trumman är 10% eller 75% av utomhusköldmängden. Exempelvis ökar mängden under en trumma med 30 centimeters diameter från 1 m3 till 35,5 m3 respektive från 10,1 m3 till 58,4 m3 för en 150 centimeters diameter. Ökade mängder medför högre kostnader i både material och transporter, vilket är något som tillsammans med övriga faktorer senare måste analyseras vidare för att få mer kunskap och underlag inför en möjlig etablering av höghastighetsbanor.

Suffusion, also called internal instability, is an internal erosion mechanism that occurs in embankment dams when fine-grained particles are washed out of the core soil matrix by seepage. Initiation of internal erosion depends mainly on three major factors: grain size distribution of the soil, stress conditions and hydraulic load; whilst its continuation depends of the filter properties. Broadly graded moraines, as glacial tills, are more susceptible to internal erosion by suffusion than other types of soils used in dams. Most embankment dams in Sweden consist of a central core of glacial till built more than 50 years ago. At the time of its construction, the available guidelines did not include specific grain size boundaries for the core and the filter related to internal erosion susceptibility. Today, several Swedish embankment dams have experienced incident of internal erosion such as leakages and sinkholes, making internal erosion an important safety issue. This circumstance allows raising the questions: How safe are the Swedish embankment dams and what conditions are needed for internal erosion by suffusion to initiate? This research aims to contribute to the assessment of dam safety by giving inputs regarding the characterization of internal erosion by suffusion and the relation among the main factors involved on its occurrence (geotechnical characteristics of soil material, degree of compaction and hydraulic load). This in order to increase the knowledge regarding the critical hydraulic gradient needed to develop suffusion in a given till material with a known degree of compaction. The research includes a laboratory program consistent on suffusion tests, which is an extension of the standard permeability test, and considers post-test examination and diagnosis of the samples. Two main groups of tests were performed: small and large suffusion tests. The small tests serve as a reference of the expected behaviour of soil samples under different boundary and test conditions, which allows optimizing the number of test to be performed in large tests. Results show that suffusion mechanism can be classified as internal suffusion (or filtration) and external suffusion (loss of soil particles from the soil matrix). The influence of compaction degree on the initiation of suffusion is limited in internally stable soils. However, poorly compacted specimens exposed to high hydraulic gradients could develop both internal and external suffusion if the filter is not capable to retain the eroded particles. The hydraulic conductivity of specimens with internal suffusion tends to decrease with a step wise increase of the hydraulic gradient. Such tendency is the result of the matrix of soil reaching equilibrium with the new seepage stresses. The hydraulic conductivity of specimens with external suffusion tends to increase with the increase of the hydraulic gradient.