Sediment yield estimation from a Sediment yield estimation from a hydrographic survey: A case study hydrographic survey: A case study for the for the KremastaKremasta reservoir, Western reservoir, Western GreeceGreece

D. D. ZarrisZarris(1)(1), E. , E. LykoudiLykoudi(2)(2) & D. & D. KoutsoyiannisKoutsoyiannis(1)(1)

(1) (1) Department of Water Resources, Faculty of Civil Engineering, NatDepartment of Water Resources, Faculty of Civil Engineering, National Technical ional Technical University of Athens, 5 University of Athens, 5 HeroonHeroon PolytechneiouPolytechneiou St., Athens 15780, GreeceSt., Athens 15780, Greece(2) Department of Geological Sciences, Faculty of Mining Enginee(2) Department of Geological Sciences, Faculty of Mining Engineering and ring and Metallurgy, National Technical University of Athens, 5 Metallurgy, National Technical University of Athens, 5 Heroon PolytechneiouHeroon Polytechneiou St., St., Athens 15780, GreeceAthens 15780, Greece

55thth International Conference International Conference “Water Resources “Water Resources Management in the Era of Transition”Management in the Era of Transition”, Athens, 4, Athens, 4--8 8 September 2002September 2002

Brief Outline of the PresentationBrief Outline of the Presentation

Introduction (Sediment yield and delivery processes, sediment accumulation in reservoirs, source erosion)

Research Project

Appraisal of river sediment deposits in reservoirs of hydropower dams, Funded by Public Power Corporation (PPC) and the General Secretariat of Research and Technology (GSRT), 1998-2001

Research method (Hydrographic survey of the Kremastareservoir, Western Greece)

Computation of total mass of the deposited sediments for the total period of the reservoir’s operation

Catchment’s sediment yield and delivery ratio estimation

Comparison with other published data from the international literature

Sediment delivery processesSediment delivery processesSediment source (wash load versus river bed material)

Magnitude and proximity to the outlet of the source erosion areasCharacteristics of the drainage network (density and frequency, slope gradients, watershed area)Frequency, intensity and duration of the erosion producing storms (wash load)Geological formations and soil characteristics (erodibility)Geomorphologic characteristics (faults, orographic uplifting, etc.)Depositional potential of the catchment (surface roughness, depressions, man-made sediment storages)

Sediment yield processesSediment yield processes

Temporal variability both in annual yields but also in inter-storm amounts

Sediment yield estimates...

1. Simple statistical regression models (e.g. sediment rating curves, sediment yield with catchment area)

Sediment yield processes as functions of spatial scale (e.g. vegetation cover for hillslope scales and partial rainfall coverage and drainage density to watershed scales)

Strongly influenced, but not completely determined, by watershed area

2. Conceptual or physically-based mathematical models (e.g. LISEM, WEPP, EUROSEM)

3. Reservoir deposits’ measurements by (repeated) hydrographic surveys

Sediment yield characteristics...

Precise processes still unknown, lack of mathematical expression with universal applicability

Temporal variability of sediment dischargesTemporal variability of sediment dischargesSediment discarge vs discharge measurments in Aliakmonas R.

at Ilarionas, Northern Greece

0,1

1

10

100

1000

10000

1 10 100 1000Discharge (m 3 /s)

Sedi

men

t dis

char

ge (K

gr/s

)

Dry period

W et period

Spatial variability of sediment yield Spatial variability of sediment yield (lack of universal expression)(lack of universal expression)

1

10

100

1000

10000

100 1000 10000 100000

Area (km2)

Sedi

men

t yie

ld (t

/km

2Data from 42 Data from 42 international watershedsinternational watersheds

KremastaKremasta reservoir watershedreservoir watershed

LANDSAT 7 (ETM+) ImageCatchment drainage network

KremastaKremasta reservoir watershed reservoir watershed ((contcont.).)

Digital Terrain Model (DTM) Mean annual rainfall (mm)

KremastaKremasta reservoir watershed reservoir watershed ((contcont.).)

Land usesGeologic formations

KremastaKremasta reservoir hydrographic surveyreservoir hydrographic survey

Positioning: Differential Global Positioning System (DGPS) (reference station and moving receiver) with accuracy in horizontal plane 2-5 m(1)

Additional information...

Dionysos Satellites Centre, Department of Topography, Faculty of Surveying and Rural Engineering, National Technical University of Athens

Distance between echo-sounding routes ranging from 50 to150 m, additionally to check routesDepth measurement: Hydrographic echo-sounder Raytheon DE 719B operating at the frequency of 200 kHz(2)

2. Depth measurement error: 0.5%±1 in of the total depthValuable contribution by...

1. In level of significance 95% with selected availability

Depth measurement illustrationDepth measurement illustration

Hydrographic Hydrographic routes for routes for reservoir reservoir scanningscanning

DTM Generation and Calculation of DTM Generation and Calculation of Deposits’ VolumeDeposits’ Volume

Digitization of initial topographic maps

Scale 1:5000

Extraction of spatially irregularmesh of points and creation

of the Triangulated Irregular Network (ΤΙΝ)

Grid formation

size 6 m*6 m

Algorithm:Triangulation with linear interpolation

(SURFER 7.0)Union of DGPS files and echo-sounder entries

in one uniqueASCII file

Creation of the TIN surface and grid formation

size 6 m*6 m

INITIAL TOPOGRAPHIC MAPSINITIAL TOPOGRAPHIC MAPS

HYDROGRAPHIC SURVEYINGHYDROGRAPHIC SURVEYING

Typical errors of hydrographic surveyingTypical errors of hydrographic surveying

DGPS accuracy on horizontal plane (due to selected availability)Obscure definition of the water-mud interfaceVariation of the hydrographic boat speed

Digitization errors both of the initial topographic maps but also from the echo-sounder charts

Errors in x-y-z plane from the construction of the initial topographic maps prior to the dam construction

Significant non-typical error (uncertainty): Areas as earth material banks for dam construction not known

Indicative profile of fluvial sediment Indicative profile of fluvial sediment deposits (a)deposits (a)

245

255

265

275

-5000 -4980 -4960 -4940 -4920 -4900 -4880 -4860 -4840Longitude (m)

Lev

el (m

)

Prior to the dam construction

Year of the hydrographic survey

Section near the reservoir delta at Acheloos R. branch

Indicative profile of fluvial sediment Indicative profile of fluvial sediment deposits (b)deposits (b)

160

180

200

220

240

260

280

300

4900 5100 5300 5500 5700 5900 6100 6300 6500

Longitude (m)

Lev

el (m

)

Prior to the dam construction

Year of the hydrographic survey

Section at the inner part of the reservoir

Identification Identification of reservoir of reservoir segments with segments with sediment sediment deposits (a)deposits (a)

Identification Identification of reservoir of reservoir segments segments with with sediment sediment deposits (b)deposits (b)

Deposits

Deposits

Deposits

Non-Deposits

ResultsResults –– Deposits’ VolumeDeposits’ Volume

10.466.6TOTAL

2.212.2MEGDOVAS R.

2.513.1AGRAFIOTIS R.

5.741.3ACHELOOS R.

Deposits’ area(km2)

Deposits’ Volume(hm3)

Reservoir Sections

INITIAL DESIGN STUDY INITIAL DESIGN STUDY ESTIMATE FOR DESIGN ESTIMATE FOR DESIGN

PERIOD 50 YEARSPERIOD 50 YEARS394394 hmhm33

Results Results –– Deposits’ MassDeposits’ Mass

Undisturbed samples were impossible to be collected due to the physical properties of the deposited sediments

Collection of two core samples from the reservoir’s invert• Evaluation of deposits’ thickness• Analysis of sediments physical characteristics and mineral composition• Estimation of deposits’ density• Estimation of real deposition rate• Measurement of sediment organic content

Additional Information...

(1) Lane and Kolzer formula from percentage quantities of sand, silt and clay(correspondingly 71.9% sand, 23.3% silt and 4.8% clay)

(2) Density estimation after 34 years of reservoir operation 1692 kg/m3

TOTAL DEPOSITS’ MASSTOTAL DEPOSITS’ MASS112.5 112.5 MtMt

ACHELOOSACHELOOS AGRAFIOTISAGRAFIOTIS MEGDOVASMEGDOVAS69.869.8 Mt 22.1 Mt 20.6 MtMt 22.1 Mt 20.6 Mt

Sediment yield of Sediment yield of KremastaKremasta reservoir reservoir watershedwatershed

TOTAL

MEGDOVAS R.

AGRAFIOTIS R.

ACHELOOS R.

Subcatchment

3292106.41005.6

123919.5489.4

32020.92034.8

173366.01184.6

Subcatchment area

A (km2)

Mean annual sediment dischargeQs (kg/s)

Mean annual sediment yield

Sy (t/km2)

Comparison with internationally published dataComparison with internationally published dataa) Sediment yielda) Sediment yield

���������������������������������������������������������������������

����������������������������������������������������������������������������������������������������������������������������������������������������������������������

��������������������������������������������������������������������������������������������������������������������������������

�����������������������������������������������������������������������

����������������������������������������

1

10

100

1000

10000

100 1000 10000 100000Area (km2)

Sedi

men

t yie

ld (t

/km

2)

Agrafiotis R.Acheloos R.

Megdovas R.

Data from Dendy and Bolton (1976)

Comparison with internationally published data Comparison with internationally published data b) Sediment delivery ratiob) Sediment delivery ratio

0 .0 0

0 .1 0

0 .2 0

0 .3 0

0 .4 0

0 .5 0

1 0 0 1 0 0 0 1 0 0 0 0 1 0 0 0 0 0

A re a (k m2)

Del

iver

y ra

tio

A grafio tis R .

A che loo s R .

M e gdovas R .

Laure nce , 1996

R e nfro , 1972

Soil erosion and sediment delivery ratio Soil erosion and sediment delivery ratio estimationestimation

0.2050401005.6TOTAL

0.222251489.4MEGDOVAS

0.4248472034.8AGRAFIOTIS

0.1770771184.6ACHELOOS

Sediment delivery ratio

Soil erosionΑ (t/km2/y)

Mean annual sediment yield

Sy (t/km2)Subcatchment

Soil erosion computed from a GIS based modelof the Universal Soil Loss Equation (USLE)

ConclusionsConclusions

Measurements of deposited sediments within a reservoir could be an effective method for reconstructing long term catchment sediment yieldsThe reservoir under study should be large enough so that trap efficiency could be assumed as unity This method is unable to estimate sediment yield of finer time scales (e.g. annually) unless more frequent hydrographic surveys are accomplished This method combined with sediment discharge measurements in an upstream site and/or alternative measurement techniques (e.g. turbidity) can be an effective tool on integrated catchment management

ConclusionsConclusions ((cont.cont.))

Dead volume principle, at least for large reservoirs, should be reconsidered in terms of the spatial accumulation of deposited sediment as describedCatchment sediment yields under study exhibit considerably higher values than other published data from throughout the globeGeomorphologic controls such as tectonic activity, orographic uplifting, hydrological parameters (e.g. intense storms) and also the dominant geological formation (e.g. highly erodible flysch) are responsible for this considerable difference

THANK YOU VERY MUCHTHANK YOU VERY MUCH