Case Study Experiment - Preventing floods and landslides
The researchers in Norway tested flood retention and the impact of vegetation on river bank stability in the Morsa Catchment. They modelled the effect of different sized retention dams in forest areas (upstream agricultural fields) and the root strength of river bank vegetation.
|Preventing floods and landslides|
Water retention ponds
The flood peak is most likely lowered by the inclusion of a retention area in the catchment. The retention area has a large impact on decreasing soil erosion on the down-slope agricultural fields. It captures and diverts surface runoff, resulting in less sheet and especially gully erosion on the agricultural land.
Although the number of floods was not quantified during the experiment, according to the statements of the farmers, the flooding over the agricultural land is reduced from 2-3 times a year to once every second year (reduction in local flooding).
|Example of the modelling results for the area without and with implemented retention pond|
- The grass and shrub plots showed similar dynamics both in groundwater level and soil water content fluctuation, while clear differences were observed in hydrological responses within the plot containing trees. The trends observed in soil water content variation corresponded to the groundwater level fluctuations. The differences can be explained by different root depths and subsequent root water uptake.
- Soil-root strength changes with time. In all three plots, higher values of shear strength were observed during late spring and summer. This trend stems from two factors: intensity of the vegetation growth (higher root density) and lower soil moisture content. Variation in modelled slope stability corresponds to variation in groundwater levels and stream water levels. The probability of riverbank failure was higher during spring and early autumn, and lower during the summer. The tree plot was the most stable and shows the highest capacity to accommodate potential shear stress.
|Changes in average shear strength with time and with
depth, for the three plots
|Frequency of Fs for all simulated scenarios, for min and max slope angles. Stability classes: red – unstable slope; yellow – conditional stability; green – stable slope|
The Vansjø-Hobøl catchment covers about 690 km2 (see below). The catchment is located in southeast Norway and drains into the Oslo Fjord, south of the capital Oslo. The main lake system is Lake Vansjø (25 m asl; with a surface area of 36 km2), and its downstream river, the Hobølelva. The lake consists of several sub-basins; the two main basins are called Storefjorden (eastern basin) and Vanemfjorden (western basin). The outlet river is the Mosseelva River. Due to a dam at Moss, there is at times no water discharge in the Mosseelva. The main river in this catchment starts in the northern part of the drainage area and runs through forested areas through several lakes. At Kure, there is a water quality monitoring station.
The highest point of the catchment area is about 346 m a.s.l., whereas the lowest is at 0 m a.s.l., since the river system drains into the Oslo Fjord - see below. Due to its relatively high proportion of agricultural land (103 km2 in total) (Blankenberg et al., 2008), the Morsa catchment has been a pilot in the implementation of the EU Water Framework Directive (WFD). The catchment area of the Hobølelva River is 333 km2, and thus covers about half of the total catchment area of Morsa (690 km2).
Topography and points of interest of the Vansjø-Hobøl Catchment with sub-catchments and lakes
Main soil threat
Extreme storm events cause damages amounting to several million Norwegian kroner each year. In addition to agricultural runoff, the erosion of river banks due to floods, results in a marked increase in both suspended sediment and nutrient loads (the marine clay is rich in the phosphorus-rich mineral apatite). In the last few years, clay avalanches have contributed significantly to increased suspended sediments and phosphorus in the river system (Fig. 1). Municipalities, the road authorities (Vegvesenet) and insurance companies in Norway are active in finding solutions to this problem in urban and peri-urban areas. Buffer zones have been established amongst others in the eastern part of the basin. Along the Hobølelva and Kråkstadelva rivers, for instance, there is only commercial grass production occurring in these zones, and only nitrogen is applied as fertiliser. The buffer zones are intended to stop surface runoff from the fields to the rivers and also protect against river bank erosion.
The pressures on the Vansjø-Hobøl catchment include;
- Water quality pressures: High nutrient and sediment loads to the lake from the tributaries, deriving mainly from wastewater and agricultural runoff.
- Hydrological pressures: Water level fluctuations in Lake Vansjø that are least partly due to the regulation of the lake. Flooding during spring/autumn (results in risk of overflowing sewage treatment plants, flooded fields with increased nutrient runoff and damage to infrastructure).
- Climate change pressures: May generate an increased frequency of flooding and therefore increased erosion of river banks which are more unstable in winter with increased nutrient runoff.
The main socio-economic and socio-cultural forces that drive human activities include a) Food production, including the results of Norwegian policies regarding agriculture and remote settlements, b) Scattered dwellings (without satisfactory sewage systems), c) Economic drivers linked to the requirements for sufficient water flow at the outlet of the lake for purposes such as industry and hydropower, d) Requirements for drinking water extraction from the eastern basin, and e) Requirements for water clean enough for swimming and other recreational use.
The bedrock of the northern forested areas is mainly Pre-Cambrian bedrock with predominantly gneiss (Dons, 1977). Relatively thin moraine layers give poor soil quality for agriculture, but further downstream in the catchment the soils become rich in silt and clay minerals, as this area was submerged under the sea during the Quaternary period.
When the great inland glacier melted, the land rose from the sea and the marine clay deposits became the most fertile soils in Norway. In the very south of the catchment, a huge end moraine is effectively damming Lake Vansjø, causing the catchment to drain to the west instead of to the south. The Lake Vansjø basin, therefore, has a very characteristic form, with several basins and bays, which is rather unusual for Norwegian lakes. The soils in the catchment are closely linked to the geological history described above (see below). Thus, in the forested areas, the soil is predominantly coarse moraine whereas in the southern area's soils are dominated by marine deposits with silt loam and silty clay loam texture, of which a substantial part has been artificially levelled. Fluvial deposits with silt and silt loam texture are found along the river. Marine shore deposits are also represented, with textures loamy sand and sandy silt. The most widespread soils are classified as Epistagnic and Endostagnic Albeluvisols (Siltic), Luvic Stagnosol (Siltic) and levelled soil (Hauken and Kværnø, 2013).
Map of the catchment with combined land cover and soil map.
The soil map covers arable land only, and is here presented as topsoil texture classes.
About 16% of the River Hobølelva catchment is covered by agricultural land, about 5% is covered by waterbodies, and the remaining 79% is covered by forest (see below). Whereas Moss is the largest township in the entire catchment, the township of Ski is the most significant settlement within this tributary’s catchment; located in the very upstream part. For many years, Norwegian policy has been to enhance cereal cultivation in this part of Norway, due to its productive soils (see below). Hence, apart from a few cattle and chicken farms, livestock-production here is low. Timber has also been a major source of income (Martinsen, 2007). Before 1500 the Catholic Church owned large parts of the land and forest in this area. Around 1600-1700, the ownership was transferred to the richest farms (so-called “herregårder”). After 1814, their power diminished and the ownership of the forest was distributed to several farmers, including smaller livings. There is presently little information on forestry in the catchment. Finally, fishing and hunting for recreational use is common in the catchment. Two nature reserves exist in the recipient Lake Vansjø ; one located in the western basin (about 330 ha) in the municipalities of Moss and Rygge, the other at Moaskjæra/Danskebukta (about 100 ha) in the municipality of Råde. The objective of both reserves is to preserve important wetlands.
|Water-bodies||Forest||Grass-production||Cereal fields||Potato fields||Vegetable fields||Total area|
|1,565 (4.7%)||263 (79%)||403 (1.21%)||4995 (15%)||1 (%)||57 (0.17%)||333 (100%)|
Land use distribution in River Hobølelva catchment in km2 (%) (adopted from Skarbøvik et al., 2007)
The catchment has a continental climate in the north and more coastal in the south, and with a growing season of about 200 days. The mean annual air temperature is 5.6oC and the annual precipitation is about 829 mm (see below), as measured at Rygge meteorological station located in the very south of the catchment area for the period 1961 – 1990.
Annual (left) and mean monthly (right) precipitation at the Rygge meteorological station for the period 2000-2014 and water discharge in Hobølelva River for the period 1977-2010. The meteorological station is situated in the lower parts of the catchment, in an area which often has more precipitation than upstream in the catchment, therefore figure may be underestimating events in the entire catchment.
The Hobølelva River has its sources in the northern parts of the drainage area and runs through forests and lakes (main upstream lakes being Lakes Sætertjern, Bindingsvann, Langen, Våg and Mjær). Downstream of Lake Mjær the river changes character as it enters into marine clay deposits. This lower part of the catchment covers about 186 km2, and the river is here characterised by meanders through agricultural fields. Included in the estimated area of 186 km2 is the catchment area of the tributary Kråkstadelva, which also drains agricultural fields and joins the main river some 12 km downstream of Lake Mjær. Downstream of the confluence, the Hobølelva River runs through two waterfalls; the first is called Høgfoss where there is a mill and a stage gauge, and the second is called Kurefoss, where there is a monitoring station for water quality.
The river drains into Lake Vansjø (25 m a.s.l.) covering an area of 36 km2 with a mean depth of 7 m. In terms of floods, the Hobølelva River and Lake Vansjø are considered as parts of an interlinked system: when the water discharge in the Hobølelva River increases, the shores of Lake Vansjø often get inundated by flood water. This is mainly due to the narrow outlet of the lake. The outlet is controlled by a dam, but even when all gates of the dam are open, the narrow straits leading down to the dam slows down the water flow and cause inundation of agricultural lands upstream. The main hydrological stations include Høgfoss in River Hobølelva and Moss Dam in River Mosseelva for measuring discharge, and Rødsund Bridge in Vansjø measuring stage.
According to the monitoring station at Høgfoss, the annual runoff in the Hobølelva River at Høgfoss is 4.5 m3/s (1977-2002), thus comprising about 40% of the water discharge into Lake Vansjø. The highest water discharges are usually in April and October/November (see above). River Hobølelva has several old dams and barriers; the largest are located at Høgfoss (location of stage gauge) and at Kure (location of water quality station). These dams have no upstream reservoirs, although the river just upstream flows more slowly due to the dams. Despite the dams that were originally contrasted to facilitate grain mills, downstream of their location the river drops to waterfalls. A mill is still in operation at Høgfoss, whereas the mill at Kure has long been removed. River Mosseelva is dammed before its outlet to the Oslo Fjord. The waterfall from the dam to the sea is presently utilised for hydropower. As of 1973, the catchment is protected against any further hydropower developments (St.prp., 1973; Fylkesmannen and Østfold, 1999).
Drivers and pressures
In the Morsa catchment, as in most of Norway, floods typically occur during spring (snowmelt) and autumn (heavy rains). Floods can, however, also occur during summer (heavy rains) and winter (melting-freezing episodes, as well as rainstorms). The geography and hydrology of this catchment also contribute to the flood risk pressure. When the water discharge in the Hobølelva River increases, the shores of Lake Vansjø often get inundated by flood water, mainly due to the narrow outlet of the lake.
Two types of landslides occur: One is mainly confined to steeper slopes and induced by the combination of gravity and low shear stress in the soil due to heavy rains and/or snowmelt, and is not very common in this catchment. When it occurs, it is mainly a threat to buildings and less to soils. The other type is induced by quick-clay. This phenomenon is caused by the fact that the clay minerals in the soils are kept stable by high salt contents, since the soils were deposited in brackish or salt water. Gradually, salt is washed out of the soil, producing in an unstable clay structure that can collapse if an external pressure is imposed upon it. The result is a high viscosity liquid that flows at critical velocities much higher than that of water thus posing a great risk to life and property (Figure 9.4). Erosion in river banks is a typical triggering factor for such landslides. Climate change may pose additional pressure by increasing flood frequency or flood magnitude. Climate change is expected to lead to higher precipitation and more intense rainfall in this region of Norway (Hansen-Bauer 2009; Engen-Skaugen m.fl. 2010), thus increasing the risk of both floods and landslides.
Status of soil threat
In 2011, a workshop was held for farmers with fields bordering Lake Vansjø. Based on workshop results and several assumptions (see Skarbøvik et al., 2011), the area of inundated land, as well as the maximum depth of the inundated areas, were calculated (see below). While the workshop was not focused on the River Hobøl catchment but rather on the fields bordering the lake, results include farms within the lower reaches of the Hobølelva River. Several landslides have recently been observed in the catchment (Figure 9.4), but relevant data on frequency and size is not currently available.
|Flood return period||2-years||10-years||50-years||100-years*|
|Inundated area [ha]||42||135||219||269|
|Maximum depth of inundated area [cm]||25||80||130||160|
Inundated area of fields along Lake Vansjø for different flood return periods (data based on interviews with farmers as outlined in Skarbøvik et al. 2011).
Quick clay landslide along the Hobølelva River, 2008. Photo: Eva Skarbøvik
Maps on the current state of land use, soil degradation and soil conservation in the case study area have been produced using the WOCAT (World Overview of Conservation Approaches and Technologies) methodology
The steps of this process are as follows:
1) The area to be mapped is divided into distinctive land use systems (LUS).
2) The team gathers the necessary data on soil degradation and conservation for each LUS using a standardised questionnaire, in close consultation with local land users, and supported where possible by remote sensing or field data.
3) For each LUS, the soil degradation type, extent, degree, impact on ecosystem services, direct and indirect causes of degradation, as well as all soil conservation practices, are determined.
4) Once collected, the data is entered in the on-line WOCAT-QM Mapping Database from which various maps can be generated.
Following the principles of all WOCAT questionnaires, the collected data are largely qualitative, based on expert opinion and consultation of land users. This allows a rapid and broad spatial assessment of soil degradation and conservation/SLM, including information on the causes and impacts of degradation and soil conservation on ecosystem services.
More details about the methodology used to produce these maps and their interpretation can be found here.
Land Use (click on maps to expand)
The "effectiveness" of conservation is defined in terms of how much it reduces the degree of degradation, or how well it is preventing degradation. The Effectiveness trend indicates whether over time a technology has increased in effectiveness.
Effects of floods and landslides on ecosystem services
The following table summarises and ranks the effects of floods/landslides on soil functions in this Case Study. The table should be regarded as a temporary assessment and may need adjustments as the project continues.
|Biomass production||Reduced biomass production due to anaerobic conditions (floods), or soils loss (erosion due to floods, landslides).||M|
|Environmental interactions||Floods lead to erosion and the soils in this Case Study contain high levels of phosphorus thus increasing eutrophication risk in the downstream lake (Vansjø).||M|
|Gene reservoir/ Biodiversity pool||May be affected if soils are inundated and anaerobic over longer periods and/or by soil loss through landslides, but it is assumed that the relative effect of this as compared to the total biodiversity of soils in the catchment is less important.||N|
|Physical medium||Base for built development is affected by both floods and landslides. In the River Hobølelva, floods and landslides have so far mainly affected soil loss and less damage is observed on houses and infrastructure, hence only less impacts.||L|
|Source of raw materials||No known impacts in the Case Study area.||U|
|Carbon pool||No information available on the effects in this Case Study.||N|
|Cultural heritage||Assumed to be of little consequence in this Case Study area.||N|
Effects of floods and landslides on soil functions in the Hobølelva River Case Study. (N: None; L: Low; M: Medium; H: High; U: Unknown).
Administrative and socio-economic setting
The EU Water Framework Directive (WFD) is implemented in Norway, and the Morsa river basin was, together with 29 other Norwegian river basins, a pilot for the implementation. Flood protection responsibility in Norway falls under the Norwegian Water Resources and Energy Directorate, but is usually confined to the protection of infrastructure. Landslides are also the responsibility of this directorate. Again, protection measures are designed to protect houses, infrastructure and lives, and not soil. Soil protection, in general, is the responsibility of the Ministry of Food and Agriculture, and the Directorate of Agriculture.
The larger catchment of Morsa has about 40,000 inhabitants, and is situated in the counties of Akershus and Østfold. The population development in these counties is shown in the figure below. However, it is assumed that most of the population increase has occurred in the cities, and in this catchment the only townships are Ski in the upper part and Moss at the outlet of Lake Vansjø. Since the remaining settlements mainly comprise farms, some smaller communities and cottages, it is likely that the population increase in this catchment is less than what is indicated below (left). Industry is concentrated in the area around Moss. In the very south, the airport at Rygge is located at the catchment border. The standard of living in this area is believed to be close to the average for Norway as a whole (below right).
Population in Østfold and Akershus counties (NUTS3) (left) and GDP per capita trends for Norway and the Euro Area (right).
Soil loss due to erosion, floods and landslides affects the water quality in this catchment since the soil has high levels of phosphorus, both because of the use of fertilisers and because the clay is rich in apatite, which contains phosphorus. This means that control of soil loss is closely linked to water quality issues. Maps of quick clay pockets have been produced by the Norwegian Geotechnical Institute (NGI), to improve preparedness against landslides. Soft flood management measures, i.e. avoiding building on high-risk areas are often the most effective approach against damages to infrastructure and buildings. Since more than 100,000 people in Norway live in areas with a potential risk of quick clay landslides, preparedness against this type of landslides is, in general, more advanced than for other types of landslides. Typical measures include protecting river banks against erosion (see below) and drilling calcium/concrete columns into the soil in risk areas. Not all areas with a risk of quick clay landslides can be protected, and the mapping is used to find the areas where there is risk of fatalities, health or technical installations/material values. Agricultural soil is not listed specifically as a value that should be protected. Both long term and short term management options have been outlined in river basin management plans, and implemented through stakeholder interaction exercises.
(Left:) Protection works against river bank erosion, with the purpose to reduce the risk of quick clay landslides, in a river near the city of Fredikstad (south of the Morsa Catchment). Photo by NGI. (Right:) Armouring of river banks in the Hobølelva River. Photo: Eva Skarbøvik.
Implemented abatement measures
Some erosion protection of river banks has been implemented in Hobølelva River downstream of Lake Mjær. More such measures are under planning, amongst others in the Kråkstad River, but it is uncertain if they will be implemented. These measures can both protect the land from erosion due to floods, and can prevent that pockets of quick-clay are disturbed. A new operation scheme for Lake Vansjø was developed in 2011 (Skarbøvik et al. 2011) and has now been set in operation in a temporary arrangement (the final ratification by governmental directorates and ministries for a change of dam operation schemes usually takes several years). To protect the lake from excess nutrient overloads, the river banks have been used for grass production (buffer strips) for the last 5-10 years. However, the grass in the buffer strips is cut 1-2 times a year and the soil is ploughed every 5-7 years. Therefore, this measure can potentially enhance river bank erosion. Reduced tilling or tilling in spring has been done in almost all fields with high erosion classes due to a local regulation. This measure contributes to reduced overland flow and protects the soil against sheet and rill erosion. Similarly, local regulations have encouraged the establishment of grass-covered waterways.
Other abatement measures that have been/are being considered
In order to reduce the floods in Lake Vansjø, a flood tunnel from the lake directly to the sea/fjord, and/or widening of the outlet of the lake, are measures that have been discussed and partly outlined by consultants. These measures are expensive and a thorough analysis of consequences and costs has not been carried out yet. Measures to increase infiltration and retention of water in the catchment as a whole will delay the flood water and can potentially reduce river bank erosion, with subsequent decreased risks of quick clay landslides. These types of measures will have to be implemented not only in the agricultural areas but also in urban zones and forests, and have not yet been evaluated on a larger scale for the catchment. For RECARE, the interesting scientific challenge (both on a societal and natural sciences point of view) will be to approach this challenge from the perspective of soil care, which can present a new dimension to river basin management plans.
Relevant end-users and local stakeholder groups include;
- Glomma River Basin Authority and the Morsa River Basin District Organisation
- Inter-municipal water system company, MOVAR
- Østfold Nature Conservation
- Vansjø landowner organisation
- The Glommens and Laagens Brukseierforening (GLB). GLB is an organisation for all hydropower producers in the wider catchment area of the Glomma River System
- NVE – the Norwegian Water Resources and Energy Directorate – is the national regulation authority for hydropower. Any changes in the operation of the plant must be evaluated by NVE, and accepted by the King in Council.
- Involvement in Case Study
The stakeholders mentioned above will be actively involved in the RECARE project. NVE and Bioforsk have ongoing cooperation in the ExFlood project in developing mitigation measures, and together with the NGO's GLB, Østfold Nature Conservation and the Vansjø landowner organization, demonstration sites for these measures will be identified and established (see point 3 below). Apart from this activity, all stakeholders will be incorporated according to the dissemination plan of the RECARE project.
Gender and stakeholder workshops
The first study site workshop had a gender equal balanced group of 13 women and 11 men participating. Their roles were also equally divided, both men and women are water board- members, farmers association, local and regional authorities. The 2nd Stakeholder Workshop was hosted on Monday (23 November 2015). It took place in Ås, which is quite close to the Norwegian case study area; Vansjø. Eleven people attended altogether (a small, but good group!) The group consisted of 4 women and 7 men. Here are a few pictures showing the stakeholders in action evaluating measures to prevent floods and landslides.
Blankenberg, A.-G. B., Turtumøygard, S., Pengerud, A., Borch, H., Skarbøvik, E., Øygarden, L., Bechmann, M., Syversen, N., og Vagstad, N.H. 2008. Tiltaksanalyse for Morsa: Effekter av fosforreduserende tiltak i Morsa 2000-2006. (Mitigation analysis for the Morsa Catchment: Effects of measures to reduce phosphorus levels in the period 2000-2006; in Norwegian). Bioforsk Rapport 86, Vol.3, 2008, 54 pp.
Dons, J.A. 1977. Geologisk fører for Oslo-trakten. (Geology of the Oslo area; in Norwegian). Universitetsforlaget. 173 pp.
Engen-Skaugen, T., Førland, E. J., Hygen H.J. og Benestad, R. 2010. Klimaprojeksjoner frem til 2050; Grunnlag for sårbarhetsanalyse i utvalgte kommuner. met.no klima-rapport nr. 4, 2010, 58 s.
Fylkesmannen and Østfold 1999. Verdier i Vansjø-Hobølvassdraget, Hobøl, Våler, Råde, Rygge og Moss. Utgitt av Direktoratet for naturforvaltning i samarbeid med Norges vassdrags- og energidirektorat. (Values in the Vansjø-Hobøl Catchment; in Norwegian). VVV-Rapport 2000-15. Trondheim. 24 pp.
Hauken, M. and Kværnø, S., 2013. Chapter 2: Agricultural management in the JOVA catchments. In: Bechmann, M., Deelstra, J. (eds.). Agriculture and Environment. Long Term Monitoring in Norway. Akademika Publishing, Trondheim. ISBN 978-82-321-0014-9. Pp. 19-42.
Martinsen, Ø. 2007. Vansjø – en unik naturperle. (Lake Vansjø – a unique pearl of nature). Orion Forlag AS. 183 pp.
Skarbøvik, E., Barkved, L.J. og Stålnacke, P.G. 2007. Tilførsler av partikler og fosfor til Storefjorden - Utredninger Vansjø 2006. (Inputs of sediments and phosphorus to Storefjorden – Vansjø; in Norwegian). NIVA-Rapp 5389. 38 pp.
Skarbøvik, E., Udnes, H., Øgaard, A., Eggestad, H., Rohrlack, T., Tingvold, J. & Drageseth, T. 2011. Helhetlig utredning av manøvrering og flombegrensning i Vansjø. (Integrated assessment of the operation of the water levels in Lake Vansjø and flood reduction levels; in Norwegian). Bioforsk Report 6(63):60.
St.prp. 1973. Stortingsproposisjon nr. 4 (1972-73). Om verneplan for vassdrag. (Norwegian White Paper No. 4 (1972-73); Plans for watersheds protected against hydropower development; in Norwegian).