Ekskursjonsguide og bakgrunnsinformasjon GEOFAREDAGEN 2017 Ekskursjon til Romerike fredag 20.oktober Kvartærgeologi, skred og flom Tidsplan Ekskursjon Geofaredag – 20.10.2017
9.00: Avreise NGI
10 – 11: Minnesundtrinnet, flomsikring Tømte/Eidsvoll (jernbane)
11 – 12: Gardermoen (kvartærgeologi etc)
12 – 13: Lunch (kan også inntas ved første skredlokalitet, gjenstand for improvisasjon …)
13 – 15: Kvikkleireskred Sørum og andre
15 – 16.30: Flomsikring Lillestrøm (her blir det drop-off ved jernbanen for personer som skal til Gardermoen for å rekke et fly hjem)
17.00 +: Ankomst Oslo/NGI
Skred Kvartær
Flom
1
ROMERIKES GEOLOGI
Rolf Sørensen, Institutt for jord- og vannfag, Norges landbrukshøgskole, Ås. Særtrykk av: Det Norske Vivenskaps-akademi. Årbok 1996, side 127-133.
På høstekskursjonen lørdag 14. september, reiste vi gjennom et landskap som har en lang og komplisert geologisk historie. Dette landskapet har fasinert både berggrunns- og løsmasse- geologer i snart 200 år, og det er derfor et stort tilfang av nedskrevet kunnskap å øse fra når Romerikes geologi skal beskrives. I det følgende er bare noen få hovedpunkter nevnt, og med henvisning til noen av de viktigste, og nyeste publikasjonene fra området.
Berggrunnen Fjellknauser og mindre åser som stikker opp gjennom de tykke løsmasselagene på Romerike består av grunnfjell -, forskjellige typer gneiser som er mellom 1200 - 1800 mill. år gamle. Fra Fetsund (blant annet under Lense-museet), og nordvestover mot Skedsmokorset går det en sone med sterkt oppknuste bergarter som kalles 'Ørje mylonittsone'. Denne sonen skiller 'Romerike-komplekset' i nordøst fra 'Østfold-komplekset' sør-vestenfor (Berthelsen et al. 1996). Grunnfjellet grenser i vest, i Nittedal - Hakadal, mot Oslofeltes 'nordmarkitter' og andre vulkanske bergarter av permisk alder, det vil si at de er 'bare' ca. 270 mill. år gamle.
Løsmassene Allerede i 1838 skrev Keilhau at mektigheten av sand- og rullestensmassene på Romerike "kan på sine Steder ansættes til allerminst 100 Fod". Nyere undersøkelser viser at dette ikke var noen overdrivelse -, 100 meter med grus, sand eller leire er målt flere steder. En systematisk undersøkelse av de store isrand-terrassene ble utført av Olaf Holtedal (1924). I de siste 20 årene har en rekke viktige kvartærgeologiske undersøkelser blitt utført på Romerike. Noen av de viktigste vil bli omtalt etterhvert. Av disse kan først nevnes de moderne løsmasse-kartene i målestokk 1 : 50 000 som er utarbeidet av Østmo & Olsen (1978), Longva (1987, 1991 og 1997).
Isavsmeltingen Like nord for Lillestrøm, fra Berger til Asak og videre til Lystad, ligger det en rad med store sand- og grusavsetninger som danner det søndre og eldste av 'Romerike-trinnene', et vitnespyrd om den store innlands-isens avsmelting, se fig. 1. Det neste stopp i breens tilbakesmelting finnes over Jessheim og sør-østover mot Hvam. 2
Det største isrand-delta i Sør-Norge ligger ved Gardermoen - Hauerseter. Sand- og grusmassene er beregnet til å ha et volum på ca. 4.4 x 109 m3, og de dekker ca. 80 km2 (et område på størrelse med Nil-deltaet). Det er beregnet at brefronten sto stille ved Hauerseter-deltaet i ca. 50 år (Tuttle et al. 1996), engang mellom 9 500 til 9 700 radiokarbon-år før nåtid (Sørensen 1983, Longva & Thorensen 1989). Videre nordover på Romerike (utenfor kartet på fig. 1) ligger Dal- og Minnesund-trinnene. Det sist nevnte er datert til omtrent 9 400 radiokarbon-år før nåtid (Sørensen 1983). Hele Romerike (fra Lillestrøm til Eidsvoll) ble altså isfritt på ca. 400 år. Da jordskorpen i dette område ble fri for den tyngende bremassen, begynte landet å stige raskt, med ca. 8 cm pr år i det første millennium.
Nåværende vannskille R¿ros Vorma 62û Bredemte sj¿er Hersj¿
yy y yy y yy yyInnlands isen
60û Oslo ca. 9100 Hauerseter før nåtid Nes
yy yy y yyy yy y yy7û 58û 11û Jessheim
Kulmoen
yy y yy y yy yyy y N y
yy yy y yy yyy y
Berger Asak Glomma Lille- yy y yy yy y yyy y str¿m Lystad Tegnforklaring: Grus, Stein og blokker
yy yy yy yFet y yyyMorene, Fjell Morenerygg
Isfront ¯yeren 0 5 10 km yy yy yAntatt isfront
Fig. 1 Fordelingen av hovedtypene av løsmasser på Romerike (hvitt er silt og leire), og de viktigste oppholds-linjene i breens tilbaketrekking, omtrent slik som Holtedahl (1924) kartla området. Noe forenklet fra Longva (1987). 3
Havet som sto helt inn til brefronten ved Jessheim-trinnet, begynte å synke slik at Romeriksfjorden ble grunnere, og store sletter med finsand og silt ble etterhvert tørt land, hvor et 'pioner-plantesamfunn' med gras, urter og bjørkeskog etablerte seg i løpet av kort tid.
Romeriksfjorden Da brefronten sto ved Jessheim- og Hauerseter-trinnene for 9 700 - 9 500 år siden, var solinnstrålingen til den nordlige halvkule nær et maksimum (Mangerud 1989), og avsmeltingen av innlandsisen var enorm. Maksimum vannføring i breelvene som munnet ut på Hauerseter-deltaet er beregnet til å ha vært mellom 6,6 til 9,2 x 103 m3/s (Tuttle et al. 1996). En vannføring som er 2 - 3 ganger større enn ved Elverum under flommen i Glomma i 1995 (Bøe 1996). Sammen med sand- og grus ble det fraktet enorme mengder slam i breelvene, og dette ble bunnfelt i Romeriksfjorden utenfor brefronten. På de ca. 50 år brefronten sto ved Hauerseter ble det meste av fjorden utenfor fylt opp med breslam, eller det som vi idag kaller leirjord. Da brefronten hadde smeltet tilbake til sørenden av Mjøsa, hadde det meste av Romeriksfjorden blitt fylt opp med slam slik at det var dannet en jevn havbunns slette, med slakt fall mot sør og vest. Ved Nes var det en slik sammenhengende slette, hvor restene av sletten idag ligger 180 - 160 m o.h., og det vil si at fjorden var bare 30 - 50 m dyp da breen trakk seg vekk fra Romerike. Lenger sør og vest på Romerike var Romeriksfjorden noen dypere, og restene av den opprinnelige havbunn finnes idag på 155 - 160 m o.h. Denne fjorden hadde åpen forbindelse til den tids Oslofjord gjennom Øyern-bassenget, og for en kortere tid over 'Grorud-passet' mot indre Oslofjord. Det ferske vannet strømmet sørover på overflaten i fjorden og skapte en sterk inngående strøm av saltvann. Når breslammet (silt- og leir-partiklene) sank ned i det salte vannet, fnokket de seg sammen og dannet en meget løs struktur (korthus-struktur) med salt porevann mellom.
Den store flommen For ca. 9 100 radiokarbon-år siden, sto brefronten ved Elverum og over mot Moelv ved Mjøsa (Sørensen 1983, Longva et al.1994). Dalførene nordenfor var fylt med breis som hadde en maksimumhøyde ved nordenden av Storsjøen i Østerdalen, og like nord for Otta i Gudbrandsdalen. I Lesja, på Dovre og ved Tynset ble det dannet store bredemte sjøer mellom det nåværende vannskillet og restene av innlandsisen. Den største av de bredemte sjøene ved Tynset har blitt kalt 'Nedre Glomsjø', se fig. 2. Da isen ble tilstrekkelig tynn, tok smeltevannet veien under isen (vist på fig. 2), og flomvannet kom fram ved Elverum og dannet et "hjøkullaup" som det kalles på Island. Havet sto fortsatt langt innover i Glommas dalføre og nivået ved Elverum var ca. 200 m. Tappingen førte til at vann-nivået steg ca. 40 m 4
og vannføringen var ca. 350 000 m3/s under flomtoppen (omtrent tre ganger Amazonas vannføring !). Longva et al. (1994). Lenger sør -, på Romerike, var havnivået den gang ca. 150 m høyere enn i dag, men flodbølgen førte til at vann-nivået steg opptil 35 m, i løpet av en 12 - 17 dagers flomperiode (modell-beregning av Longva et al.1994). Sporene etter flommen finnes fortsatt i dag som et ca 1 m tykt lag med silt oppå den opprinnelige havbunn, der hvor den fortsatt er bevart (Longva 1987, og Longva et al. 1994). I tillegg til den såkalte Romeriksmjela, som dette siltlaget kalles, finnes det også spor etter isfjell som ble fraktet med flommen og 'løftet' inn over de oversvømte områdene (Longva & Bakkejord 1990). Da flomvannet trakk seg tilbake lå store isfjell (opptil 70 m 'brede' og 30 m tykke) strødd utover slettene på Romerike (horisontalt skravert område, fig. 2).
Nedre Glåmsjø Fig. 2. Situasjonen på Østlandet 0 50 km for ca. 9 100 år siden, med restene av innlands-isen, Sverige bresjøen ved Tynset, havbuktene Jutulhogget inn mot Hamar og Elverum, og
yyyy området som ble oversvømte da bresjøen ble tappet, gjennom Bre- "demning" 'Jutulhogget' under breen fram til yyyy Elverum. Elverum Noe modifisert, Tegnforklaring
yyyy fra Longva et al. (1994). Mj¿sa Omtrentlig isutbredelse Drenring i, Romerike og under breen yyyy Kongsvinger Havets utbredelse for 9100 år siden Oslo Oversvømt område M M M¿rkfoss
Leirskredene Da den gamle havbunn til Romeriksfjorden ble tørt land begynte overflatevannet å erodere i leirjorda, slik at det ble dannet et mønster av små V-formede daler over hele den enorme sletta. Størst graving ble likevel utført av Vorma og Glomma, som raskt laget sine dalfører etter som havnivået utenfor sank jevnt og trutt. 5
Det første historisk dokumenterte leirskred er fra ca. 1320 da omtrent 630 daa land ved middelaldergården By (nå Rotnes) ble ødelagt (Løken et al. 1970). Dette er et av de største leirskred på Romerike. Ca. 65 andre skred er omtalt etter den tid.
Nannestad
Vorma
yy yyy
yy y yy yy y
Hauerseter
yy yyy yy Jessheim y yy yy y y
N
a yyy y m yy yy y yy y yy m G l o
Nittedal y yy yy y y yy y y yy
Fjell i dagen
yy y y yy
yy y yy yy y yy Lillestr¿m Grus og sand
Elver og store bekker
Daterte kvikkleireskred
yy yy yyy yyFet Udaterte " 0 510 ¯yern km
Fig. 3. Skredkart over Romerike. De daterte skredene er omtalt i mange slags skrifter. De y y yy y 'udaterte' er kartlagt ved hjelp av flybildetolking og befaring i terrenget. Fra Løken, Jørstad & Heiberg (1970).
6
Særlig under 'Lille istid' gikk det mange store leirskred. Dette skyldes sannsynligvis at det kom mer nedbør, slik at leirjoren ble oppbløtt og kunne rase ut. Utførlige beskrivelser av kvikkleire og leirskred er gitt av Rosenqvist (1960) og Bjerrum (1971). Mellom 1725 og 1800 gikk det 6 veldig skred, og den største katastrofen på Romerike skjedde da Skjea-fallet i 1768 krevde 16 menneskeliv. Tilsammen har 27 mennesker mistet livet i slike skred på Romerike, sammen med et stort antall bufe. Av andre dramatiske hendelser kan nevnes Tesen-fallet (1795) da Vorma ble demmet opp i 111 dager, slik at Mjøsa steg 8 m. Løren-fallet som gikk året før, demmet opp Rømua med 19 meter, etter at omtrent 6 (seks) millioner m3 leire hadde rast ut. Det siste skredet på Romerike gikk ved Hekseberget i 1967, men bare 25 daa land gikk tapt.
Litteratur
Berthelsen, A., Olerud, S. & Sigmond, E.M.O. 1996. Geologiske kart over Norge - berggrunskart OSLO. 1 : 250 000. Norges geologiske undersøkelse.
Bjerrum, L. 1971. Kvikkleireskred. Et studium av årsaksforhold og forbygningsmuligheter. Norges Geotekniske Institutt, Oslo. Publikasjon Nr. 89. 14 s.
Bøe, P.Ch.(red.) 1996. Flommen 1995 i Glomma og Lågen. Glommen og Lågens Brukseierforening, Oslo. 19 s.
Keilhau, B.M. 1838. Undersøgelser om hvorvidt i Norge, saaledes som i Sverrig, findes Tegn til en Fremstiging af Landjorden in den nyere og nyeste geologiske Tid. Nyt Magasin for Naturvidenskaberne Bd. 1.
Mangerud, J. 1989. Hva er drivkreftene bak de store klimavariasjonene ? Norsk Polarinstitutt, Oslo. Rapportserie Nr. 53. 10 s.
Holtedahl, O. 1924. Studier over isrand-terrasene syd for de store Østlandske sjøer. Videnskabs Selskabets Skrifter, 1. Matematisk-Naturvidenskaplige Rekke Nr. 14: 110 s.
Longva, O. 1987. ULLENSAKER 1914 II Beskrivelse til kvartærgeologisk kart - M 1 : 50 000 Norges geologiske undersøkelse, Skrifter 76: 1 - 39.
Longva, O. 1991. FET 1914 I Kvartærgeologisk kart - M 1 : 50 000 Norges geologiske undersøkelse.
Longva, O. (i trykk 1997). STRØM 2015 III Kvartærgeologisk kart - M 1 : 50 000 Norges geologiske undersøkelse.
Longva, O. & Thoresen, M.K. 1989. The age of the Hauerseter delta. Norsk geologisk tidsskrift 69: 131 - 134.
Longva, O. & Bakkejord, K.J. 1990. Iceberg deformation and erosion in soft sediments, Souteast Norway. Marine Geology 92: 87 - 104.
7
Longva, O., Hafsten, U., Haugane, E. & Solli, A. (1994). The Romerike Silt Bed; Flood deposits from the catastrophic drainage of Preboreal glacial lake Nedere Glomsjø, SE Norway. I: Longva, O. Flood deposits and erosional features from the catastrophic drainage of Preboreal glacial lake Nedre Glåmsjø, SE Norway. 17 s + 24 fig. Unpublished Dr. Scient. thesis, Department of Geology, University of Bergen, 130 p.
Løken, T., Jørstad, F.A. & Heiberg,S. 1970. Gamle leirskred på Romerike. Romerike Historielags årbok, Bind VII: 51 - 69.
Rosenqvist, I.Th. 1960. Marine clays and quick clay slides in South and Central Norway. XXI International Geological Congress, Guide to excursion C 13. Norges geologiske undersøkelse. 212q: 1 - 26.
Sørensen, R. (red.) 1982. NORDQUA-Ekskursjon 1982. Preboreal - Boreal isavsmelting i Sørøst-Norge. Institutt for geologi, NLH, Ås. Rapport nr 17: 1 - 68.
Sørensen, R. 1983. Glacial deposits in the Oslofjord area. In: Ehlers, J. (ed.) Glacial deposits in North-west Europe, 19 - 28. Balkema, Rotterdam. 470 pp.
Tuttle, K.J., Østmo, S.R. & Andersen, B.G. 1996. Quantitative study of the distributary braidplain of the Preboreal ice-contact Gardermoen delta, Southeast Norway. Proceedings, J.-O. Englund Seminar, Gardermoen: 156 - 180.
Østmo, S.R. & Olsen K.S. 1978. NANNESTAD - 1915 III, kvartærgeologisk kart M - 1 : 50 000 Norges geologiskeundersøkelse.
WB- Excursion to Romerike
Figure 1 Excursion map
1
Background
The Romerike district is a lowland north east of Oslo situated mainly to the west of the river Glomma and its largest tributary, Vorma. It comprises approximately 1000 km2, a great part of which consist of Late Pleistocene marine sediments and glaciomarine deposits. Most of this area is located from 150 to slightly above 200 m above sealevel. Some places rock (PreEocambrian gneiss) outcropping are visible as hills above the sediments. The area is mainly drained by the two rivers, the Rømua and the Leira (the name means ”the clayey”) with their tributaties which enter the Glomma (or lake Øyern to the south). Glomma is the largest catchment in Norway (hydrological figure 2).
Figure 2. Marine deposits in Norway and mapped scars at Romerike.
The claciomarine deposits of Romerike, are the last and most northern of the glacial substages and provide a record of retreat of ice during the Late Pleistocene in Oslofjorden district, approximately 11000- 9000 years ago. During this period the Romerike area was a wide and shallow marine bay with the sea level between 190 and 210 m above present day sea level. In this bay several outwash deltas were formed in front of the retreating glacier. These deltas consist mainly of sand and gravel. Deltas are located at Berger (oldest), Jessheim and Hauerseter (youngest)
Later during the regression of the sea, courser material was deposited usually as a rather thin cover above the clay sediments. This is locally called ”mjele” and has a grain size distribution corresponding to a fine sand or a course silt. It is generally less than 1-2 m thick, and is believed to be deposited during dam breaks of glacial dammed lakes in upstream valleys.
2
The last 9000 years sculptured the clay areas in Romerike, i.e. river erosion and slides have modified the earlier sea bottom. The rivers Rømua and Leira show a characteristic dendritic drainage pattern which in Norway is a characteristic by river erosion in clay areas. Old slide scares can be seen and in some cases it seems like river erosion has erased former slide topography. During the 18th century several slides occurred and many of these slides have been described or mentioned in papers written at that time. The Skjea slide in 1768 and the Løren slide in 1794 are two of these and we will visit site of the the Holum slide from 1883.
The so far last contribution to landscaping, is manmade agriculture landscaping. In a period when the ravinated grassland was turned into agricultural fields, a number of large quick slides occurred. NGI developed guidelines for safe land development the slide problems diminished. Despite improved practices manmade construction work is still the most important factor for triggering this type of landslide. As described below, hazard and risk mapping of quick clay and land use planning in combination with mitigation, is important to mitigate this type of hazard.
Figere 3. Snow covered landscape south of Gardermoen, (about 1950?). Erosion and slides has formed the earlier sea bottom.
3
QUICK CLAY slides
Quick clay is deposited in marine environment during the last glaciations, probably brackish water, and later uplifted due to deglaciation (isostasi). In undisturbed state, quick clay has strength similar to normal clay. Quick clay shows however high sensitivity, meaning that the shear strength is reduced dramatically when remoulded (by more than a factor of 30). Geochemically quick clay has a low ionic content (salt), the salt content in the pore water is normally about 1g or less , which is 1/35 of the salt content in normal sea water. Marine clay forms from flocculated particles that settle down to seabed. The deposited clay has an open structure, with high water content. During collapse, the clay particles become suspended in their own pore water.
Figure 4
4
Quick clay
Initial slide
Figure 5
During failure the pore pressure increase in the soil and a local failure will propagat along concentrated shear bands. These landslides are retrogressive, they often start at a river, and progress upwards. They have been known to penetrate kilometres inland, and consume everything in their path.
STOP 1: Gjerdrum muncipality, Ask
10 years ago NGI made a Risk and Vulnerability study at the town Ask. The purpose was to identify areas for establishment of residences. The work resulted in a mitigation plan in an area which now is turned into a living area with acceptable safety. Counterfills and closure of brooks and ravines are techniques to increase the safety of an area.
A mapping method has now been standardized and hazard and risk maps have been produced in all areas where large occurrences of quick clay are known. Enclosed are two standard maps showing the centre of the towns Ask and Kløfta.
5
Figure 6 Risk map quick clay
The hazard map is classed based on factors as thickness of quick clay, sensitivity, pore pressure, slope height, river erosion. The hazard map is classified into 3 classes; High, medium and low.
The risk map is based on occurrence of buildings (reflects number of people), roads and other infrastructure inside the hazard zone plus possible downstream effects of the slide movements and deposits.
Figure 7 Hazard map quick clay
6
Kløfta
Kløfta is the centre in the Ullensaker municipality. During review of the municipal land use plan in this area, different suggestion of change in land use was suggested (blue polygons). NGI inspected these areas, and specified the necessary level for further investigations and required safety level if these areas was turned into living areas. Safety level for geotechnical issues differs from other type of hazards where frequency or return periods are used as. The safety level is the fraction between the strength of the material (soil) and the applied forces. This is similar to all type of safety assumptions used for construction works.
For quick clay areas it is required
1) No critical erosion 2) Safety factor larger than 1,4 or Substantial improvement of stability (~15% or 10 % dependent on risk level)
The safety factor or material factor 1,4 is also a general limit implemented in Eurocode 7
Figure 8 Map of areas with suggested changes in the municipality land use plan (blue polygons)
7
STOP 2 - Holum, Ullensaker 26. november 1883
Figure 9 Hand drawing from 1983, after the large slide at Holum 26. november 1883. Two farms disappeared (Northern and Soutern Holum), 6 people killed.
Holum is located north of Kløfta in Ullensaker municipality. The autumn 1983 had been unusually wet and small slides had occurred earlier this autumn. On November 25 a large slide occurred and during the night of 26. November a rapid slide occurred, destroying the houses on two farms (Northern and Southern Holum) 6 people died, all from the southern farm. 28 animals were killed.
It is estimated that 1, 3 mill m3 clay was released into the river Leira. The river was dammed up, and created a 6,5 km long lake. The water level rose 12,5 m and when the dambreak occured, all downstream bridges were destroyed.
The slide scar at Holum is approximately 450 m long and with 20 m elevation diffference. This means that the average slope angle along the failure plan was less than 2.5 °. The slide area is approximately 100 000 m2. The river Leira is ~900 m west of the scar.
8
STOP 3: Dikes and flood at Leirsund
The River Leira has a catchment of about 660 km2, and the main part of the catchment is forest and hilly areas. At Leirsund, the downstream influence from the lake Øyern i limited. The lower part of the River Leira runs through a meandering river plain. There are two bridges at Leirsund, one new and one old. The old has acceptable flood capacity. The new bridge is shown in the figure below was closed last week due to flooding.
Figure 10 Leirsund bridge (new), autumn 2000
9
Figure 11 June 6 1995,. Lillestrøm from south. Photo : Fotonor
STOP 4: Dikes around Lillestrøm
The flood level at Lillestrøm is mainly dependent on the water level in the lake Øyern. More than 100 years back, the natural yearly fluctuation of water in Øyern was more than 8 m. Today it is in average about 2 m which has allowed the development of the city. The reduction in the spring flood is caused by enlargement of the outlet from the lake and water regulation in the catchment itself. Upstream hydropower development has also reduced the flood size. One of the basic rules in hydropower development is that the dam or its gates shall be constructed or maneuvered so that natural floods do not increase in magnitude due to the regulation. Øyern is a part of the Glomma catchment (see below), the largest catchment in Norway.
Lillestrøm experienced extreme floods in 1966 and 1967 and then again in 1995.The floods in 1967 culminated at level 106.67 m.a.s.l. and 1200 houses were damaged. Measures in the outlet of
10
Øyeren after this event led to a reduction of flood top estimated at approximately 2.3 m. The floods in 1995 culminated at level +104.45 m.a.s.l.
After the flood in 1995 Skedsmo municipality decided that Lillestrøm should be protected from flood by building an embankment. The dike is now laid out over a distance of 6.4 km from Nebbursvollen in the northwest to Dyneaområdet in the south and up to Fetveien (see enclosed Flood Inundation Map, Lillestrøm). It is mainly built at the level +106.5 m.a.s.l with opportunities for extension up to the contour 107.5 using sand bags or other temporary measures. To not destroy the view from low- lying homes, it was on certain sections given dispensation to lower the top of the embankment to level+105.5. These sections account for nearly 1.7 km.
The municipality is today prepared to handle a flood level of 105.0 including a 0.5 m safety zone. For higher flood levels the municipality must have external help, to put sand bags on the low sections between level +105.5 and +106.5
The municipality has prepared a contingency plan that specifies various actions dependent of water level in Øyern/Nitelva. This include closing of outlet pipes to the river to prevent the river from entering the gray water system and starting flood pumping stations to keep local water and waste water out of the city.
Until 1995 safety level of dikes was selected to maximum observed water level plus a security level. This is common all over the world where people has expired an extreme flood. The table below shows the present statistically estimated flood water level at Lillestrøm. The outlet capacity for the lake Øyern is again improved and taken into account.
Flood water levels at Lillestrøm:
Normal water level: approx. +101.50 m.a.s.l
10 years flood: level 103.10 m.a.s.l
20-years flood: level 103.40 m.a.s.
50 years flood: level 104.10 m.a.s.l
100-years flood: level 104.70 m.a.s.
200-years flood: level 105.30 m.a.s.l
500-yeasr flood: level 106.15 m.a.s.l
NVE has a system where 80% financial coverage from NVE, the rest has to be covered by the muncipality or land owner . However during extreme events the local financial contribution is often set to zero.
11
Glomma catchment
Glomma river basin stretches 600 km from north to south, from Røros in the northeast and Grotli in the northwest down to Fredrikstad in the south. The water system has two main branches, Glomma in Øster dalen and river Laagen in Gudbrandsdalen. The total area of the catchment area is the entire 41,200 km2, which accounts for 13% of Norway's total land area.
Norway's largest lake Mjøsa (362 km2) is located in the part of the catchment. Glomma catchment covering heights from sea level up to 2469 m on top of the mountain Galdhøpiggen. Approx. 30% of the catchment area is over 1000 meters above sea level, ie above the tree line, and approx. 30% less than 500 meters above sea level
Glomma river hydrology reflects the great variations in climate within the river system, from mountain areas in north and west to the lowlands in the south. River's average annual inflow measured by Solbergfoss at the end of Øyern is about 22,000 million m3.
Average annual water flow at Solbergfoss is 700 m3 / s. Here in the lower reaches of the river water flow can vary from 150to 3,500 m3 / s during the year. Annual precipitation varies across the field from 260 mm in the arid areas of northern Gudbrandsdalen and North Eastern Valleys to 1,050 mm in the wettest parts of the catchment area in the northwest. Of Norway's approximately 4.8 million people, live about 15% of the catchment to Glomma River.
The water control in the catchment is run by Glommens and Laagen Brukseierforening, GLB - an organization for hydropower producers in the Glomma River Basin. The catchment has 26 regulated basins and about 50 hydropower stations. The total energy production is about 11 Terawatt- hours.
What is a water management organization
A water management organization is an organization for all hydropower producers in a river system, founded on the Watercourse Regulation Act of 1917. The main task of such a water management organization is to coordinate the use of water resources in the river, to the mutual benefit of all power plants. This is achieved in that the responsibility for all key hydropower reservoirs in the catchment is run by the association, which then also will be responsible for the licensing provisions and flow regulations are complied with. The water management organization is thus a natural link between power companies and the government / public. In Glomma river handled this task Glommens and Laagen Brukseierforening, which was created in 1918.
12
Figure 12 Glomma catchment
13
Hazard mapping and Land Use planning
The flood inundation project in Norway was started after a White paper that followed an extreme flood in 1995. More about this historical process are give in the enclosed paper by Hallvard Berg. Good practices in land use planning is the most cost effective and environmentally sound way of reducing risk of damage from flooding and other hazards. There is a continued and increasing pressure for urban development along rivers.
The Norwegian Planning and Building Act states that development is not allowed, unless safety is at an “acceptable level”, with local municipalities responsible for ensuring that this is the case. NVE offers guidance to local municipalities in the form of flood inundation maps, maps showing areas at risk of quick clay landslides and gives expert advice to municipal land use plans. NGI carry out work identifying hazards zones according to Norwegian national standard (TEK).
NVE has also developed a national guideline defining acceptable safety levels with respect to floods and other hazards related to rivers. The safety levels are differentiated related to hazard type and type of asset. A stepwise procedure for assessing the hazards has been designed to fit the planning process and levels typical for a local municipality. The following procedure is recommended:
• Municipal plan: potential hazards should be identified • Zoning plan: the actual hazard should be described and risk quantified • Building case: a satisfactory level of safety must be documented
The objective of this procedure is to ensure that areas with a potential hazard are identified at an early stage in the planning process giving municipalities more reliable and predictable land use plans.
NVE can raise an objection to a land use plan if flood or landslide hazard has not been considered.
Flood inundation maps
Flood inundation maps present the area prone to flooding at one or more floods with given return periods.
After a major flood in south eastern Norway in 1995, a governmental commission gave several recommendations in order to reduce flood damage in the future. The damage in 1995 amounted to about 1.8 billion NOK (200 mill USD).
The average annual cost of flood damage in Norway is about 200 mill NOK. One of the recommendations from the commission on flood protection measures was to produce flood inundation maps for the areas with the largest damage potential.
Why produce flood inundation maps?
The overall objective of the mapping is to reduce flood damages, through improved land use planning and emergency preparedness. The main target groups are municipalities and county officials, who are responsible for land use planning and emergency planning at local, respectively county level. The flood inundation map represents a tool to achieve:
• Improved land use planning with respect to flood hazards. A sensible use of flood prone areas is in Norway regarded as the best way of keeping the damage potential at a reasonable 14
level. Improvement in land use planning with respect to risk of flooding is among the most important measures to achieve this goal. • Improved flood warning and emergency preparedness. The maps will be useful in emergency planning and action connected to flood situations. The basis data and model results from the mapping will make quantitative flood forecasting possible, i.e. forecasting of water levels locally. Flood inundation maps can be generated related to the forecasted flood levels, allowing quick assessment of the potential impacts of a given flood. The maps will simplify rescue operations such as evacuation, and give background information when setting priorities to other actions. • Improved flood protection plans.
National standards
NVE has defined guidelines for land use planning and flood protection in flood prone areas. The acceptable flood risk is differentiated in relation to hazard type (risk of life, risk of economic loss) and type of assets to be protected. Domestic buildings for instance, should be safe against flooding up to 100 years floods, while industry and important infrastructure should be safe against at least 200 years floods.
The flood inundation map project includes the areas in Norway with the largest damage potential. The original plan was to map 188 river streches, covering 1750 km river length in 168 municipalities. Total cost is estimated to 8 mill USD. Project period: 1998-2007
The flood inundation map project was started by NVE (Norwegian Water Resources and Energy Administration) in 1998. NVE is manager of the project and NVE professionals also do most analyses. Other organisations such as the Norwegian mapping authority and private consultants participate with basis data. The municipalities are active partners in the mapping process and contribute with local information on water levels in previous flood events as well as measurements during floods within the project period.
Production method
The maps are produced digitally, to make the users able to make their own presentations in combination with other information, using their own tools. High accuracy mapping was chosen, in order to make the users able to use the results in land use planning without further analyses.
Land surface is represented by a DEM (Digital Elevation Model) based on detailed elevation data and the riverbed is represented by surveyed cross sections. Expected accuracy of the DEM is +/- 30 cm.
Through flood frequency analyses and hydraulic simulations water levels for 10, 20, 50, 100, 200 and 500 years floods are calculated. Expected accuracy of the computed water levels is +/- 30 cm.
Inundated areas are determined using GIS (Geographical Information System).
Historic events related to other known hazards in the river system, such as ice jams, ice run, erosion, debris flows etc, are identified based on information from local informers and archives, without trying to relate the events to statistical probability.
15
The final results from each river reach are delivered to the users both as a report with paper maps and as digital data. The presentation is standardized at scale 1:15000 with cross sections, levees etc marked. Water levels for all computed floods are presented both in a table and in a graph (longitudinal profile).
Flood estimation
Basis data are long term water stage observation series from hydrometric stations and catchment characteristics. Discharges are estimated for all projects within a catchment, based on flood frequency analyses. The result is the best coherence between discharge and return period
Flood elevation profiles
Basis data are cross sections and discharges from the flood estimation
Flood profiles are calculated in hydraulic simulation programmes (MIKE 11 or HEC-RAS)
The result from the hydraulic simulations is water level in each cross section
Land surface model
A digital elevation model (DEM) with high resolution (5-10 meters) and vertical accuracy (+/- 30 cm) is made by generating TIN models for all elevation models.
Base map data: detailed elevation data of terrain as well as other available elevation data (roads, levees, water contour etc).
Flood surface
Flood surface derived from TIN based on flood elevation in cross sections from the hydraulic simulation. The inundated areas are identified by subtracting the DEM from the flood surface, resulting in positive values in inundated areas.
16
Vi takker alle våre sponsorer. Uten generøse bidrag fra disse organisasjonene, ville det ikke vært mulig å avholde Geofaredagen 2017 med ekskursjonen som et gratis arrangement. Vi er sikre på at dette har sikret en rekordstor deltakelse.
TUSEN TAKK!