# JSSBIO1Huttula Lecture Set 41 5. Sediment transport models.

## Presentation on theme: "JSSBIO1Huttula Lecture Set 41 5. Sediment transport models."— Presentation transcript:

JSSBIO1Huttula Lecture Set 41 5. Sediment transport models

JSSBIO1Huttula Lecture Set 42 Settling velocity (v f ): Stoke’s equation Assumption: spherical particles Gravity force = drag force Particle reaches a constant settling velocity This velocituy is dependent on: fluid viscosity (  ), density difference between the particle and water (  s -  ) and particle diameter (d), g=gravity constant= 9,81 m/s 2 Velocity range: From 0.07 m/d (clay, d=1.2  m) to 710 m/d (sand, d=200  m), density=2.5 gcm -3

JSSBIO1Huttula Lecture Set 43 Settling speed in nature Particles are seldom spherical: clay particles are like plates Aggregation of particles happens due to the electromagnetic forces  cohesive soils (clays, mud…) Organic compounds like humic substances have a very fragile structure  changes even in water column  velocity from Stoke’s equation has to be corrected with empirical relations Baba& Komar, 1981: v real =0.761v f In sediment transport models v f is calculated using the median particle size from a surface sediment sample

JSSBIO1Huttula Lecture Set 44 Erosion or resuspension from bottom Acting forces on a particle laying on the bottom 1.difference between gravity on buoyancy 2.drag force by the current 3.lifting force due to the pressure differences as caused by water flowing between particles 4.electromagnetic forces causing aggregation Term 1.  density difference and particle (diameter) 3 Terms 2 and 3.  shear force caused by current and particle (diameter) 2 Shields’s empirical curve for erosion  in design of structures A simplified erosion curve by Hjulström (erosion vs. current velocity) In models we use most often the critical shear concept

JSSBIO1Huttula Lecture Set 45 Hjulström’s curve for erosion

JSSBIO1Huttula Lecture Set 46 Critical shear Total shear (  ) on the lake bottom = shear by orbital movements of waves = f(wind fetch over lake, lake mean depth, wind velocity and duration)Materials\Lake Säkylän Pyhäjärvi.pdfMaterials\Lake Säkylän Pyhäjärvi.pdf shear by currents  > critical shear (  cr ), erosion happens with a rate  a*(excess shear) b  cr, a and b are experimental values, which we calibrate during model application values for  cr : 0.008…1 Nm -2, b=1..3, a = depends on sediment In this formulation there is no consolidation effects and bottom morphology included

JSSBIO1Huttula Lecture Set 47 Calculation of sediment transport Simple screening tools:http://el.erdc.usace.army.mil/dots/doer/tools.htmlhttp://el.erdc.usace.army.mil/dots/doer/tools.html Using numerical flow models for predicting the horizontal current field Suspended solids concentration is calculated with concentration equation Following terms in concentration equation: –advection with settling speed in vertical dimension –turbulence –mass flow from tributaries and to out flowing river –settling and deposition to bottom –erosion or resuspension from bottom

JSSBIO1Huttula Lecture Set 48 Example from Mänttä 2DH flow model with BOD7 water quality compartment Sediment was light organic fibre Short term regulation at hydropower plant

JSSBIO1Huttula Lecture Set 49 Example from Karhijärvi Three different models were tested 2DH, 2DV and 3D model Models were tested in an runoff case in Oct 1992, when heavy rains caused erosion from watershed and a heavy suspended solids load to lake Data: winds on the lake, water current observations, turbidity observations  3D model gave best results

JSSBIO1Huttula Lecture Set 410 Mänttä: Transport model result Sediment: fibrous material

JSSBIO1Huttula Lecture Set 411 Sediment transport in Tanganyika Model simulation –lake wide circulation model  boundary values (current velocity) for high resolution model at river mouths –flow model and suspended sediment transport models –SS input was estimated from historical data –real winds from atmospheric model HIRLAM (this model was used first time in tropics)

JSSBIO1Huttula Lecture Set 412 Calculated depth-averaged flow on 24.08.97 04:00. Calculated depth-averaged flow on 24.08.97 20:00 3D FLOW MODEL

JSSBIO1Huttula Lecture Set 413 (A), 12:00 24.08.97 (B) 0:00 28.08.97 (C) after 22, 82 and 166 hours after the simulation start respectively. 3D SEDIMENT TRANSPORT MODEL

JSSBIO1Huttula Lecture Set 414 6. Water quality models

JSSBIO1Huttula Lecture Set 415

JSSBIO1Huttula Lecture Set 416 Concentration equation where, –c = concentration, qL= amount of loading release, n= length measure against release, u,v,w = x-, y- ja z- advective velocities in x-, y- ja z- directions, D x, D y, D z = dispersion coefficients, R(T,c) = biogeochemical changes in concentration

JSSBIO1Huttula Lecture Set 417 Application of WQ-models We include : –Advection –Dispersion –Settling on the bottom Bio- chemical processes –Decomposition, respiration, aeration, anaerobic release of P from the bottom –Select the most important variables concerning the problem –Oxygen, nutrients (like P,N), chlorophyll-a and some conservative substance (like Na) –Limiting factors (light, nutrients, …) must be included. Check! –Temperature corrections must be included. Check!

JSSBIO1Huttula Lecture Set 418 Lake Lappajärvi WQ-model PROBE temperature model –Materials\Effects of Climate Change....pdfMaterials\Effects of Climate Change....pdf PROBE-WQ model –Materials\Lappajarvi_WQ.pdfMaterials\Lappajarvi_WQ.pdf

JSSBIO1Huttula Lecture Set 419 Oxygen model

JSSBIO1Huttula Lecture Set 420 Phytoplankton biomass and ToTP

JSSBIO1Huttula Lecture Set 421

JSSBIO1Huttula Lecture Set 422

JSSBIO1Huttula Lecture Set 423

JSSBIO1Huttula Lecture Set 424

JSSBIO1Huttula Lecture Set 425

JSSBIO1Huttula Lecture Set 426 Interactions in EIA-SYKE-model

JSSBIO1Huttula Lecture Set 427

JSSBIO1Huttula Lecture Set 428

JSSBIO1Huttula Lecture Set 429

JSSBIO1Huttula Lecture Set 430

JSSBIO1Huttula Lecture Set 431 Other WQ-model applications Several case studies Materials\Flow_and_WQ_Models_Sarkkula.pdf Materials\Flow_and_WQ_Models_Sarkkula.pdf Lake Pyhäselkä –Materials\Pyhaselka.pdfMaterials\Pyhaselka.pdf What happened to WQ after real reduction of loads? –Materials\IAWQ99.pdfMaterials\IAWQ99.pdf

JSSBIO1Huttula Lecture Set 432 Summary of WQ-model calculations Check that you have data to describe WQ in variable discharge and loading conditions Select those properties (variables), which describe best the effects of loading and concentrate calibration on them Use most simple parameterization of the variables First coefficient values from literature and by experience Compare the calculated and observed values Select the conditions (weather, discharge and loading) during which the effects are described ….and run the model!!