ICSU Dark Nature


Aims, scope and workplan

IGCP 490

Megafloods meeting

November 1 - 5, 2004

Bobole, Mozambique


Modern Mega Flooding in Kenya especially Budalangi Division and Tana River Districts.

Pamela Abuodha1 and John Omenge2
1: Kenya Marine and Fisheries Research Institute, P. O. Box 81651 - 80100 Mombasa, Kenya
2: Mines and Geological Department, P. O. Box 85420 - 80100 Mombasa, Kenya

Floods occasionally cause disaster in Kenya. Areas of Budalangi in Western Province and the lower parts of the Tana River are susceptible to floods. In 1997/98 the El Nino phenomenon affected many parts of Kenya causing damage worth millions of shillings, loss of lives, famine and waterborne disease epidemics. With inadequate preparation for the El Nino floods, national resources were over-stretched in the response phase. In 2002 floods affected Budalangi and Tana River resulting in the death of people and livestocks. The situation is aggravated by deforestation, rapid urbanization and weak enforcement of the physical planning Act, resulting in the location of human settlements in flood plains, steep hillsides, without observing building codes, design and standards.

Budalangi division lies to the north of Lake Victoria near the Kenya-Uganda border. Rainfall pattern in Budalangi is mainly bi-modal (two rainfall seasons in a year). The major season occurs in March to May (the long-rains season) while the other season (short-rains) occurs in October to December. Tana River district is in Coast Province of Kenya. The area may be considered a generally dry area. It also observes a bi-modal rainfall pattern (March-May and October-December) but the average rainfall peaks in both seasons fall below 100 mm. January-February and June-September are normally very dry.

Budalangi and Tana River areas have been identified with floods for decades. As a matter of fact, the flooding does not occur due to heavy rainfall in the areas. Annual rainfall analyses indicate that the amounts of rainfall in the areas alone may not be enough to cause such floods. Massive water in-flows emanating from the bursting of River Nzoia banks happens to be the main cause of the Budalangi floods. The River originates from two high-ground areas of Mt. Elgon and Cherengany Hills and drains into Lake Victoria. These two areas are known to have high rainfall amounts almost throughout the year. They receive average annual rainfall amounts of over 1250 mm while Budalangi area receives an average of about 1100 mm. The Nzoia River gathers strength as it flows downstream to an extent of bursting as it reaches the Budalangi areas. It all depends on the intensities of rainfall in the upstream regions Elgon, Cherengany and the surrounding areas. The displaced people usually damage some of the dykes, when they move with their belongings to higher areas that include the dykes, while others, whose homes have been submerged, use the dykes to bury their dead.

A lasting solution to the flooding problem has not yet been found. Every year the area floods, lives are lost, others are evacuated, relief assistance is sought, during dry seasons survivors go back to their land and rebuild their homes and with the next rains, they go through the same cycle. The study of the psychological and sociological attitude to floods reveals that there is remarkably little conscious human adjustment to this danger and that, even in places where action is taken to reduce the flood hazard, it may be 'casual, improvised, ineffective and far from optimum'.

Meanwhile, some technical measures can help to control extreme floods;

In this paper, Budalangi and Tana River floods are described, causes and effects of flooding are mentioned and ways to eliminate the floods permanently are suggested.

Are Kampala's Floods driven by Geo-Physical Processes? An Assessment of the causes and Impacts to the Urban Population

Bamutaze Yazidhi and Lwasa Shuaib

Department of Geography, Makerere University, PO Box 7062, Kampala, Uganda

Flooding is one of the major environmental and planning problems in Kampala city. This is because the frequency and occurrence of floods in many parts of the city especially the suburbs has increased tremendously in the last 10 to 15 years. Kampala is largely experiencing flashfloods that occur mainly within the rainy seasons. Flooding in the city affects the population in different ways but is more severe to the urban poor who have mainly settled the flood prone areas that are converted wetlands. Thus even a slight down pour during the rainy seasons may have far reaching impacts to the people in Kampala. Land use changes associated with the geo-physical processes as well as the topographic setup of Kampala have played a significant role in the increase of floods. This paper derives from a compendium of studies (that have been undertaken from both the socio-economic and geomorpholigcal perspectives) to analyze interacting processes of human activity with the natural processes in increasing flooding in Kampala. The paper also analyses the social economic impacts of flooding especially on the urban poor who reside in informal settlements that occupy wetlands mostly. The paper presents an introduction of flooding problem in Kampala before briefly expounding on the drivers for flooding and an assessment of the impacts to the urban population. The paper concludes with a note on the way forward in management of floods in the city that comprehensively needs to consider the anthropogenic processes and geo-physical processes underlying flooding.

The 2000 flood in the lower Limpopo River basin

Sebasito I. Famba Faculty of Agronomy and Forestry Engineering, Eduardo M. University, Maputo, Mozambique

The Limpopo catchment forms part of four countries: Botswana, Zimbabwe, South Africa (upper part of catchment) and Mozambique (17%: the whole lower part of the catchment). 13 mill people live in the catchment, with the densest population close to its mouth. In recent years the river has experienced floods in 1955, 58, 67, 72, 75, 81 and 2000. Discharge measurements in Chókwe in Mozambique showed three times higher flood peak than during any of these earlier reported floods. The main negative impacts of the 2000 flood in the Limpopo Valley are the following:

The total cost of the floods damage and relief works is equivalent with about 20% of Mozambique's GDP. In addition to this, there became a shortage of drinking and sanitation water and insecurity in circulation of people, good and services. Environmental impacts include: Erosion and deposition, reduced vegetation cover (floods afected especially mangrove in the delta area), change in ecosystems and bio-diversity (e.g. occurrence of exotic commercial fish in the river after the flood). Floods are not necessarily disasters, and the possible advantages of floods includes:

Flood mitigation strategies for the future include:

Many positive ecological impacts of the river system are related to flooding cycle. Floods are recurrent events, will come again; they are integral part of the hydrologic cycle - but each new food will differ in magnitude and characteristics to the previous. There is a need to record the floods (flood history, flood height, inundated area, satellite images, video and photos) and to consolidate the data in a report.

Flood Hazard Mapping and Monitoring Using GIS and Remote Sensing

Marek Graniczny Polish Geological Institute, Tel/fax: (22) 8494926; marek.graniczny@pgi.gov.pl>

Severe rains affects the Earth almost every year and cause floods, bringing serious damages to towns, roads, agriculture and to the environment in general, sometimes with loss of human lives. Polish Geological Institute and BGR (Geological Survey of Germany) started common studies in the Odra valley on the beginning of 90th. The multitemporal Landsat TM images, satellite radar data and aerial photos were widely used for geological, geomorphological and landuse mapping. In the first half of July 1997, heavy rains falling on the border areas between Poland, the Czech Republic, Austria and Slovakia, swelled the water courses and caused floods in the southern part of this region. Within a 10-day period, over 100 people died in Poland and in Czech Republic. On 15 July ERS-2 SAR data revealed consistent floods near Wrocław on the Odra River and westwards, along the river course. The extent of the flooding along the Odra River was revealed by ERS-2 SAR multi-temporal images on 18 July. Additional SAR data, collected from ERS-2 on 21 July, provided, up-to-date information on the event. The flooding along the Odra reached the border between Germany and Poland with a high water pressure that seriously threatened the resistance of a 160-km dike along the Odra near Frankfurt. Threatened zones of the dike could be identified at the Landsat TM data, registered 22nd July. On 23 July a 160-km dike collapsed. Two days later, the residents in the Frankfurt neighbourhood had to be evacuated. In the Czech Republic, thousand of homes were destroyed and thousand of acres of farmlands badly affected. In Poland over 149 villages were submerged and almost as many were threatened by new floods. On 26 July about 15 000 citizens had to leave the Town of Słubice on the Odra. The Polish side of the Odra was more in danger than the German side, because of the height difference between the two river banks (1-3 m lower in Poland). In the night 27-28 July, the water level in Frankfurt reached a record height 6.75 m. The situation improved during the second half of August. Waters retreated, thus reducing the risk of further dike cracks, with the exception of the Oderbruck region. Here, the high water level was still threating villages and farmlands. By using satellite information (optical and microwave), local authorities, civil protection entities and insurance and re-insurance companies are offered one more tool to monitor flood events and to assess damages. Furthermore, by combining the satellite information with topographic data (DTM), geological and hydrological data, even more end-user-oriented products can be obtained for direct utilization by entities in charge of risk management and hazard prevention.

Dark Nature: Environmental Catastrophes and recoveries in the Holocene

Haldorsen, Sylvi Department of Plants and Environmental Sciences, Agricultural University of Norway, P.O.Box 5003, N-1432 Aas, Norway e-mail: sylvi.haldorsen@ipm.nlh.no

Environmental catastrophes become a task of increasing interest during the last years. In 2003 a group of scientists, lead by Prof. Suzanne Leroy at the Brunel University in London took the initiative to establish a project on natural catastrophes. The project was launched in 1994, with funding from ICSU, IGCP, IUGS and INQUA. The purpose of the project is to focus on historical and pre-historical catastrophes and to see which effect these have had on societies. The project will mainly be run as workshops, each of them dealing with one particular kind of natural hazards. The following themes will be addressed in the workshops:

1. Desertification and sever droughts (Mauritania 2004)
2. Mega-floods (Mozambique 2004)
3. Landslides (Argentina 2005)
4. Earth quakes (Black Sea 2005)
5. Melting of permafrost (Canada 2005)
6. Final meeting, with field trips to areas of neotectonics (Italy 2005)

Most of the catastrophic events listed above are driven by climate. In global change prognoses one will very often meet the prediction that climate extremes will become more frequent during a global warming. One of the main aims of the Dark Nature project is to look at past events, in order to see how they are linked to wet and dry, warm and cold climate episodes. Another aim has been to analyse historical data to understand how societies and ecosystems in an area exposed to natural catastrophes responded during the different events.

When we analyse data for example for mega-droughts, we often see that the relation between climate and the catastrophic events is not necessary linked in the way predicted by climate modellers. For example, in several areas climate prognoses link “warm and dry” while data from the past indicate that “warm” has been associated with “wet”. In other areas a climate prognosis may say that large floods will become more frequent, while data from the past may tell us that the most catastrophic mega-flood events may have occurred during dry climate periods. It is necessary to understand what causes a catastrophic event, how it develops over time. This requires an integrated knowledge about climate, landscape, vegetation and land use in the area where it happens.

This meeting will in particular focus on recent mega-floods. The main focus will be on the interplay between nature and people.

Geochemical flux from a mountain catchment in southeastern Norway during the 200 year flood in 1995

Haldorsen, Sylvi; Jakobsen, Leif; Jorgensen, Per and Stromme, Gaute Department of Plants and Environmental Sciences, Agricultural University of Norway, P.O.Box 5003, N-1432 Aas, Norway e-mail: sylvi.haldorsen@ipm.nlh.no

In 1995 eastern Norway experienced the largest flood ever recorded. The flood was caused by a snow-rich winter, along with a late and sudden melting period. The flood was in particular severe along the two large rivers of Glma and Lgen with their tributaries. The discharges are normally controlled via a series of hydroelectrical dams, but the flood was too large to be managed by opening and closing of the dams. Our study was done in a sub-catchment of the largest tributary of Glma, the Godlidalen, situated on the mountain plateau between Glma and Lgen. The sub-catchment is natural and not manipulated by any dams. Discharge and geochemical flux were measured during the whole flood event, and was compared with the flux during the rest of the year 2005 and with the snow melt flood in 2004. In 1995 about 50 % of the annual ion flux out of the valley occurred during the first week of July. The discharge was the twice the discharge of the snow-melt flood in the same catchment in 2004, and the cation flux was 20 % higher. During the first month after the flood, the electrical conductivity of the river water was the lowest we have ever measured during the period 1985 - 1996. The lowest electrical conductivity was measured after the flood peak. This shows that ions are removed from the groundwater zone during the whole flood event from the start of the flood until its end. The geochemical flux during the 1995 flood is equivalent to the ion reservoir in the upper 2 m of the groundwater zone in the whole catchment.

Mega-floods in southeast Norway were considerably more frequent during the sold climate period from 1800 to 1880 than they have been during the warm period 1920 - 2000. Predicted climate changes in the region indicate that the amounts of lsnow in the studied catchment will decrease. The risk of mega-floods will most likely be reduced if the climate prognoses prove to be correct.

Record of Abrupt climate change during Early Holocene on the western continental margin of India

Pratima M. Kessarkar National Institute of Oceanography, Dona Paula - 403 004, Goa, India

Flood is one of the natural hazards that affect most of the countries in the world. The developing countries, particularly those in the tropics, with high and seasonal precipitation are subjected to widespread flooding. These floods are often related to deforestation, Industrialization etc. For the purpose of detecting the effects of human activities on climate change, it is important to document natural change in past climate.

India is monsoon-influenced country; the Western Ghats provides a principal geographical barrier in the path of the Arabian Sea. Southwest monsoon season (june to september) is the principal rainy season, over 90 % of annual rainfall is realised in this period. The southwest coast of India is directly influenced by SW monsoons with high annual rainfall of 3,107 mm (national average is 1,197mm) in Kerala.

To check the influence of monsoon induced flooding in the marine environment, three sediment cores collected from the southwestern margin of India were studied for various sedimentological parameters (organic carbon, CaCO3 content, grain size, coarse fraction constituents) and the results compared from the northwestern margin of India which is less monsoon affected. Our cores are located off Thiruvananthapuram, Mangalore and Goa and were collected from water depths of 1420 (core 1), 1940 (core 2) and 2650 (core 3) m respectively. Silty clays are the predominant sediment type in all the cores. The organic carbon (OC) content in the core-tops decreases from 4.4% (core 1) through 2.4% (core 2) to 0.6% (core 3). The CaCO3 content is 38%, 26% and 43% in cores 1, 2 and 3 respectively. Down-core variations reveal two intervals of high OC associated with sediments of late Holocene and the Last Glacial Maximum (LGM), and an interval of low OC during early Holocene. Increased terrigenous clay content coincides with low OC and CaCO3. Although high OC corresponds with high CaCO3 in late Holocene sediments, there is no such relation in some early Holocene and LGM sediments. Down-core variations in OC are similar in southwestern margin cores. In contrast, a sediment core from the northwestern margin of India (at 1900 m water depth) exhibits low OC associated with LGM sediments. A gentle increase in organic carbon in late Holocene sediments is documented in all the cores.

An increase or decrease of OC in the LGM sediments of the southwestern / northwestern margin of India may be related to the presence / absence of convective mixing associated with the NE monsoon in these regions. The decreased OC and CaCO3 and increased clay content evidence an intensified SW monsoon during early Holocene. Dilution by terrigenous matter and a stronger near-surface stratification during periods of intensified monsoons may have led to lower productivity. These features observed in the most of the cores on western margin of India suggest possibility of megaflood during this period. Increased OC in late Holocene sediments again suggests increased productivity and commencement of present day condition.

Paleoflood and hydroclimatic analysis of the Snake River, Idaho/Oregon

Rhodes, G.B., Geology Department, University of Maryland, College Park, MD USA 20742 Ely, L.L., Geology Department, Central Washington University, Ellensburg, WA USA 98926

Combined paleoflood and hydroclimatic analysis can be used to suggest the timing of modern and ancient extreme floods and causal mechanisms associated with their occurrence. Flood terrace stratigraphy at the Tin Shed and China Rapid sites along the Hells Canyon reach of the Snake River, Idaho/Oregon were examined to chronicle flood-related sedimentary deposition. Further, to associate climatic and meteorological conditions to the extreme floods (>10 yr recurrence interval) in the central and lower Snake River basin composite maps of geopotential height anomalies over the North Pacific Ocean were constructed based on historical stream gauge and precipitation records.

Flood terrace stratigraphy at Tin Shed and China Rapids suggests 25 extreme floods have occurred over approximately 5,000 years. A set of younger flood layers, which are <1 m below the surface yield samples with radiocarbon ages between 0-2 cal yr B.P. and 573-462 cal yr B.P. Some of these layers were probably deposited during extreme floods that occurred between 1894 and 1964. An older set of layers >1m below the surface yield samples with ages between 1269-1045 cal yr B.P. and 8661-8451 cal yr B.P. Younger and older sets of layers are separated by 0-2 units, which suggests a temporal gap.

Seventy percent of the daily composite maps constructed for dates prior to winter (DJF) extreme floods in the historical record show a high-pressure anomaly near the Aleutian Islands and Bering Strait. In contrast, only 35% of daily composite maps based on extreme precipitation (90th percentile) show high-pressure anomalies over the Aleutians Islands and Bering Strait. High-pressure anomalies in this area are associated with snowmelt, rain-on-snow events or extended periods of extreme precipitation. The difference in mapped anomalies and extreme stream flow and precipitation is probably due to stream flow response lag time and the complexity of flood-generating meteorological conditions.

Geochemistry in the interpretation of floodplain sediments

John Ridgway, British Geological Survey, Keyworth, Nottingham NG12 5GG, UK

River floodplains provide a history of the their development through the preservation in their deposits of sedimentary features, fossil plant and animal life, and both physical and geochemical records. Together these features provide information on the fluvial environment in which the sediments were deposited, their age, the prevailing climate and land use at the time, and the origin of the materials. The interpretation of these features, however, is complex and multi-disciplinary studies are required in order to achieve a proper understanding of floodplain evolution. The geochemistry of floodplain sediments is one of the tools that can be used, in conjunction with archaeological and industrial records and knowledge of the regional geology, geomorphology and geochemistry, to help decipher the age and origin of deposits and the processes of floodplain development.

Floodplain deposits can be sampled in natural riverbank and terrace exposures, by hand digging, the use of industrial excavators, or with a variety of drilling and coring techniques. The design of sampling programmes, wherever possible, should be based on an understanding of the geomorphology of the floodplain. Sampling intervals and the analytical programme depend to some extent on the aims of the study, but the possibility of post-depositional chemical movement under the influence of, for instance, changing water tables, must not be overlooked.

In drainage basins subject to mining or industrial activity, the geochemistry of floodplain sediments can be linked to records of contaminant output to help date flood events and cast light on how sediment moves through the system.

In general, the use of geochemistry in the interpretation of floodplain deposits depends on the close link between the geology of an area and the geochemistry of the sediments derived through erosion of that area. Regional geochemistry can be used to model inputs to a fluvial system from different parts of the drainage basin and to remove the effects of mining and industrial contamination. Even where regional geochemical data are not available, if the geology is known, limited-sampling programmes can provide sufficient information to develop useful models.

Catastrophic floods and abrupt global change: how glacial Lake Agassiz changed the world.

James T. Teller, Department of Geological Sciences, University of Manitoba, Winnipeg Manitoba, Canada

During the last glacial period, nortward-draining rivers in North America were dammed by Pleistocene ice, forming an extensive although discontinuous fringe of lakes across Canada that extended from the Mackenzie River basin across the Prairies to the Great Lakes and into the St. Lawrence Valley. This vast network of lakes was episodically connected as glaciers retreated, and overflow from them influenced most major hydrological systems east of the Rocky Mountains, as well as the oceans into which the lakes drained. Glacial Lake Agassiz was the largest of these lakes, covering large areas in Manitoba, Saskatchewan, Ontario, and western Quebec, as well as the largest lake in the world, and its geological record extends over more than 1.5 million km2. Lake Agassiz influenced the climate, vegetation, and people of this vast region, and its overflow played a role in the evolution of the Great Lakes, St. Lawrence Valley, Mackenzie River Valley, and Mississippi River Valley systems. Overflow from Lake Agassiz entered 3 different oceans during its 5000-year life span, and periodic catastrophic outbursts of thousands of cubic kilometres of water altered global ocean circulation and brought about at least 3 episodes of global cooling during a period when the earth was warming. The record of overflow from this lake is being used to model potential impact of large fluxes of freshwater on global ocean circulation and climate.

Although none of the outbursts from Lake Agassiz resulted in sea level rises of more than 0.5 m, some caused rapid transgressions of the ocean across shallow continental shelves and marine basins. For example, the final outburst from Lake Agassiz 8400 years ago (163,000 km3) would have caused an abrupt transgression of 0.7 km in one year across the floor of a gentle continental shelf with a slope of 1 in 1500, like that of the Mississippi River delta. On the nearly flat floor of the Persian Gulf (1:25,000 slope), which was dry during the last glacial maximum, the final outburst from Lake Agassiz resulted in a transgression of 12 km in only one year. Thus, in the Persian Gulf basin, not only did melting ice sheets produce a continuing ocean transgression of 140 m every year for 7000 years, but there was also an abrupt transgressive flood of 12 km about 8400 years ago. Generation after generation of humans living on the floor of the Persian Gulf would have been forced to move to higher and higher land: on at least one occasion, they would have had to move quickly. This must have prompted stories about a “flood”, such as that recorded in Babylonian history as the Epic of Gilmamesh (in the first cuneiform script on clay tablets, which were buried in the NW end of the Persian Gulf) and, perhaps, in the Bible as Noah's Flood.

The basis for the contingency plan in Mozambique

César Tembe

Instituto Nacional de Gesto de Calamidades, Maputo, Mozambique

The master document for actions to be undertaken before and after a catastrophic event is an annually based document produced since 1996, and includes:

This is an official documentation produced by the National Institute for Calamity Management, Instituto Nacional de Gesto de Calamidades, and aimed to:

There are seven working groups for execution, monitoring and evaluating of specific activities when a calamity strikes:

The impact of mega-floods. How to identify mega-floods in palaeorecords - Introduction to the workshop

Lopo Vasconcelos, Department of Geology, University of Eduardo Mondlane, Maputo Mozambique

This lecture will give an overview of the geology and geography of Mozambique, as a background to understanding of the flooding in the country.

Northern Mozambique is dominated by Precambrian rocks, with a belt of sedimentary rocks along the coast in the east and Karoo age rocks in the northwestern corner, along the border to Tanzania and Malawi. The Zambezi River forms the boundary to Central Mozambique.

In central Mozambique there are two areas of Precambrian rocks to the west, surrounded by a belt of sedimentary rocks, with the main occurrence in the central part of this region.

Southern Mozambique is dominated by sedimentary rocks, with volcanic Karoo age rocks along the border to South Africa. These form the Lebombo Mountains.

Quaternary unconsolidated sediments cover most of Southern Mozambique and the coastal zone of Central Mozambique.

A great part of land is a lowlying coastal plain, especially in southern Mozambique, with altitudes below 200 m a.s.l. In these areas the typical floodplains occur. Areas influenced only by precipitation falling in Mozambique seldom experience floods. For areas influenced by precipitation in neighbouring countries, there is a high risk of flooding. There are very few dams in Mozambique.

Topics of this workshop will be:

Floods in Mozambique Lopo Vasconcelos, Department of Geology, University of Eduardo Mondlane, Maputo Mozambique

As it was worldwide broadcasted, Mozambique faced the most severe floods of the last 50 years, due to heavy rains and 2 cyclones. The floods affected the southern provinces of Mozambique: Maputo, Gaza, Inhambane, Manica and Sofala. The cyclones affected Inhambane, Sofala, Manica Zambézia, Nampula and Cabo Delgado Provinces

In February 2000, the precipitation system produced rainfall at a rate exceeding 500 mm a month in the damaged area. Heavy rain fell intermittently for three weeks from Feb. 26. This rate is several times the normal yearly rates.

As a result, rainwater flowed into the valley of the Limpopo River and caused a flood. The hypsometric map of Mozambique shows that in the Limpopo area, altitudes do not reach 200m The Limpopo River is usually 10 km wide, but reached a maximum width of 125km.

Risks of flood:

In areas influenced by precipitation only in Mozambique, floods rarely occur. Areas influenced by precipitation in neighbouring countries are high risk flooding areas. The high risk flooding areas are the low-lying areas found along the coast

Because of the large areas with low-lying coasts along river mouths, Mozambique is a highly vulnerable country in terms of water control.

The impact of mega-floods: Excursion - field guide

Mussa Achimo, Joo Mugabe and Fortunato Cuamba Department of Geology, University of Eduardo Mondlane, Maputo, Mozambique

One major concern in the recent discussion on global climate change is the prospective frequency and severity in the very near future of extreme events such as heavy rainfall, flooding, drought etc. But estimating the return periods for these recent floods is difficult due to short instrumental records (rarely >100 years in most cases). In Maputo, the rainfall is recorded since the 1910s, but we only could get data from the 1950s. Floods are the most common type of natural disasters in Mozambique because of the climate irregularities and its unique drainage. In 2000 southern Mozambique experienced the worst floods in 50 years.

Two training areas were selected for the course of this workshop: Area 1 is within the Mkomati River meander bend north of Marracuene village, 25 km north of Maputo. The area was selected due to the fact that the Mkomati River floodplain areas are inundated periodically by the over bank flow, and the floodplain deposit here reflects the mechanism by which the sediments are transported and deposited. These include transfer from the main channel during over bank flow, and by slope wash from the red sand dunes of the Inland Dune Formation. The 2000 flood sediments are easily identified in cores close the Dune Formation. Area 2 is located in the Maputo City and this was selected in order to show some of the impacts of the 2000 floods on the geomorphology. Large gullies were formed with sedimentation on the lowland surface flats. This can be estimated by the heights of the buried houses. In the elevated areas the gullies have cut and/or removed houses and moved entire houses down to the lowland areas. Water supply systems have been cut off in many places.

The field excursion will includes a tour to destroyed areas of Maputo City, as well as a trip to Xai-Xai City, 200 km north of Maputo, one of the most affected city by 2000 floods of Limpopo River.

Applications of SAR remote sensing in landslide and subsidence studies Krakow , 29 October 2004. Ground - motion monitoring from the distance of 800 km; TerraFirma project

Marek Graniczny, Janusz Jureczka and Zbigniew Kowalski Geological Survey of Poland

The SAR is an active, coherent acquisition system: both amplitude and phase information are recorded. Phase values contain information about the distance between the sensor and the target on ground. Data are available since 1992. The method has been used in Poland.

Today 3 SAR platforms are available for civil applications, and can be applied to measure small ground displacements. In Poland the method has been used to estimate ground subsidence. How

ever, the method can also be used to look estimate flood water levels. Preliminary conclusions:

1. The measuring data shows subsidence of the area limited to few millimetres per year 1 - 2 mm. Such small values suggest that these subsidences have not relation to underground exploitation. Only, about 1% of measurements show values extending from 1 to 2 cm, which could be indication of the hard coal exploitation. From other sources (geodetic measurements, precise levelling etc.) the recorded subsidences reach 10 - 20 cm per year and even more.

2. The strong correlation between the recorded negative values and structural pattern of Carboniferous strata is out of question. It is difficult to determine character of this correlation and its genesis, at this moment. The measurement's results indicate that subsidence is present at the areas of synclinal structures and the areas of the dropped wings of the two big regional faults (Będzin and Kłodnica). Undoubtedly, the two above-mentioned faults of the Variscan origin were rejuvenated during the Alpine Orogeny (the Triassic deposits were found in the dropped wings of faults). It could be indication of the active tectonic movements, too. The further studies are needed to prove that hypothesis.

GEOINDICATORS: An introduction and their relevance to flood studies

John Ridgway, British Geological Survey, Keyworth, Nottingham NG12 5GG, UK

In 1992, the International Union of Geological Sciences (IUGS) established, through its Commission on Geological Sciences for Environmental Planning (COGEOENVIRONMENT), a working group to develop geological indicators of rapid environmental change. With the assistance of specialists in a variety of geoscience disciplines, the working group compiled a Geoindicator Checklist. This was designed to provide decision-makers, managers and planners with a series of Earth science related tools to aid in the assessment and monitoring of Earth system processes and changes that are important to the way man lives in, uses and manages urban, agricultural and wilderness environments along with their mineral and biological resources. The Geoindicators Initiative (GEOIN) has developed from this to encourage the application of geoscience to environmental concerns through monitoring and assessing rapid geological change. GEOIN operates via the Internet (www.geoindicator.org) as a medium for the exchange of data, information and experiences and is involved in the organisation of conferences, workshops, training courses, publications and other activities. The website provides an up-to-date list of activities and publications on geoindicators.

Geoindicators are measures (magnitudes, frequencies, rates and trends) of geological processes and phenomena occurring at or near the Earth's surface and subject to changes that are significant for understanding environmental change over periods of 100 years or less. They are applicable to both catastrophic events and those that take place more gradually, but which are evident within a human lifespan. Typically, they are high-resolution measures of dominantly abiotic, short-term changes in the geological environment, many of which can result in irreversible ecosystem disturbances at a variety of scales. Standard techniques in geology, geochemistry, geophysics, geomorphology, hydrology, physical geography and other Earth sciences form the basis for the development of geoindicators. Some measure simple observable changes that are easily represented on maps (e.g. shoreline position or the areal extent of a lake), whilst others (e.g. soil or groundwater quality) require measurements of chemical, physical, or biological parameters.

Twenty-seven geoindicators have been identified that are applicable to monitoring and assessing geological changes in fluvial, coastal, desert, mountain and other terrestrial areas. Examples of typical geoindicators are given and the twelve indicators that provide tools of relevance to the study of floods, their causes and effects, are briefly described.

This website is hosted by Memorial University with support from the Canadian Quaternary Association