Controls on particle settling velocity and bed - PowerPoint Presentation

Slide1

Controls on particle settling velocity and bed erodibility in the presence of muddy flocs and pellets as inferred by ADVs, York River estuary, Virginia, USA

Kelsey Fall*, Carl Friedrichs, and Grace CartwrightVirginia Institute of Marine Science Slide2

Motivation: Determine fundamental controls on sediment settling velocity and bed erodibility in muddy estuaries

Physical

-biological

g

radient

found along the York estuary

:

-- In the middle to upper York River estuary, disturbance by sediment transport reduces macrobenthic activity, and sediment layering is often preserved. (e.g., Clay Bank – “Intermediate Site”)-- In the lower York and neighboring Chesapeake Bay, layering is often destroyed by bioturbation. (e.g., Gloucester Point – “Biological Site”)-- NSF MUDBED project ADV tripods provide long-term observations within a strong physical-biological gradient.

Study site: York River Estuary, VA

(X-rays

courtesy ofL. Schaffner)

Schaffner et al., 2001

1/11Slide3

ADV at deployment

-- ADVs often provide quality long-term data sets despite extensive

biofouling

.

-

-

ADVs provide continual long-term estimates of:

Suspended mass concentration (c) from acoustic backscatter

Bed

Stress (τ

b): ρ

*<u’w’>

Bulk Settling

Velocity (

w

sBULK

): <w’c’>/csetErodibility (ε) given by ε = τb/M, where M is depth-integrated C

ADVafter retrieval

Observations provided by an Acoustic Doppler VelocimeterSensing volume ~ 35 cmab

(Photos by C. Cartwright)

Fugate and Friedrichs ,2002; Friedrichs et al., 2009; Cartwright, et al. 2009 and Dickhudt et al., 2010

2

/11Slide4

Biological

site

Generally < 1

kg

/m

2

/Pa

Intermediate site

ε

varies from

~ 3 kg/m

2

/

Pa (Regime 1) to

~ 1 kg/m

2

/

Pa (Regime 2)

1

2

3

4

5

6

ε (kg/m

2

/Pa)

Seasonal Variability in bulk settling

velocity (

W

sBULK

)

and bed

erodibility

(ε) is observed

at the Intermediate Site.

3-day mean of

ε

from fits

to M = ε

τ

b

using ADVs

Biological

site

W

sBULK

~

1 mm/

s

Intermediate

site

W

sBULK

varies from ~ 0.5 mm/s (Regime 1) to ~ 1 mm/s (Regime 2)

3- day Mean

WsBULK from fits to <w’c'> = WsBULK<C> using ADVs

2

1.5

1.0

0.5

0

WsBULK (mm/s)

Cartwright et al., 2009

3/11Slide5

Biological

site

Generally < 1

kg

/m

2

/Pa

Intermediate site

ε

varies from

~ 3 kg/m

2

/

Pa (Regime 1) to

~ 1 kg/m

2

/

Pa (Regime 2)

1

2

3

4

5

6

ε (kg/m

2

/Pa)

3-day mean of

ε

from fits

to M = ε

τ

b

using ADVs

Biological

site

W

sBULK

~

1 mm/

s

Intermediate

site

W

sBULK

varies

from ~ 0.5 mm/

s (Regime 1) to

~ 1 mm/s

(Regime 2)

3- day Mean

W

sBULK

from fits to <

w’c'> = WsBULK<C> using ADVs

21.5

1.0

0.5

0

W

sBULK

(mm/s)Cartwright et al., 2009

What is happening at Intermediate Site

when

Regime 1



Regime 2?

3/11Slide6

W

sBULK = <w’c’>/<c> (mm/s)(a) Sediment Bulk Settling Velocity, W

sBULKPhase-Averaged Settling Velocity for Two Regimes Regime 1

Regime 2

Increasing |u| and τ

b

Tidal Velocity Phase (q/p)0.1 0.2 0.3 0.4 0.5Similar WsBULK at the beginning of tidal phase suggest presence of flocs during

both regimesRegime 1:

Flocs-Lower observed W

sBULK at peak |u| and τb (<0.8 mm/s)

Regime 2: Pellets+Flocs

-Higher

observed

W

sBULK

at peak |u| and τ

b (~1.2 mm/s)-Influence of pellets on WsBULK7/11

(Note that Bulk Settling Velocity, wsBULK

= <w’c’>/cset is considered reliable for mud only during accelerating half of tidal cycle.)Slide7

Tidal Analysis highlights differences

in Regime 1 and Regime 2.Tidal Velocity Phase(θ/π)

Increasing IuI

Decreasing IuI

(b) Bed Stress (Pa)

(d) Concentration (mg/L)

0

0.5150100150200

0.050.1

0.150.2

0.25

(c) Drag Coefficient

0

0.5

1

0.00004

0.00008

0.0012

0.0016

CWASHCWASH(a) Tidal Current Speed (cm/s)

153045

Tidal Velocity Phase(θ/π)Increasing IuIDecreasing IuI5/11Slide8

(a) Tidal Current Speed (cm/s)

153045

Tidal Analysis highlights differences

in Regime 1 and Regime 2.

Tidal Velocity Phase

(θ/π)

Increasing IuI

Decreasing IuI(b) Bed Stress (Pa)(d) Concentration (mg/L)00.5

1

50

100150

200

0.05

0.1

0.15

0.2

0.25

(c) Drag Coefficient

0

0.510.000040.000080.00120.0016CWASHC

WASH

Regime 1: Flocs -High C at relatively low τb -Lower ADV derived Cd (more stratified water column)-Lower τb despite higher similar current speeds

Regime 1Regime 1

Regime 1

Regime 1

Tidal Velocity Phase

(θ/π)

Increasing IuI

Decreasing IuI

5/11Slide9

(a) Tidal Current Speed (cm/s)

153045

Tidal Analysis highlights differences

in Regime 1 and Regime 2.

Tidal Velocity Phase

(θ/π)

Increasing IuI

Decreasing IuI(b) Bed Stress (Pa)(d) Concentration (mg/L)00.5

1

50

100150

200

0.05

0.1

0.15

0.2

0.25

(c) Drag Coefficient

0

0.510.000040.000080.00120.0016CWASHC

WASH

Regime 1: Flocs -High C at relatively low τb -Lower ADV derived Cd (more stratified water column)-Lower τb despite higher similar current speeds

Regime 2: Pellets+Flocs-Lower C at

high

τ

b

-Increase in C

d

(Water column less stratified)

Regime 2

Regime 2

Regime 2

Regime 2

Tidal Velocity Phase

(θ/π)

Increasing IuI

Decreasing IuI

5/11Slide10

Concentration (mg/L)

(a)(b)

Hysteresis plots of C vs.

t

b

for the top 20 % of tidal cycles with the strongest

t

b for (a) Regime 1 and (b) Regime 2 . τcDEP flocs = ~ 0.08 PaWashload (~20%)Flocs (~80%)

Washload (~20%)

Flocs (~50%)

Pellets (~30%)

Bed Stress (Pa)

Bed Stress (Pa)

Concentration (mg/L)

τ

cDEP

flocs = ~ 0.08 Pa

τ

cINT

= ~ 0.05 PaτcINT

= ~ 0.02 PaRegime 1 Regime

2 -- Once tb increases past a critical stress for initiation (tcINIT), C continually increases for both Regime 1 and for Regime 2Erosion

-- As t

b

decreases

for

Regime 1

, C does not fall off quickly until

t

b

≤ 0.08 Pa, suggests that over individual tidal cycles, cohesion of settling

flocs to the surface of the seabed is inhibited for τb

larger than ~ 0.08 Pa. -- As tb decreases for

Regime 2, C decreases more continually, suggesting pellets without as clear a t

cDEP. But the decline in C accelerates for tb ≤ ~

0.08 Pa, suggesting (i) a transition to floc deposition and (ii) that settling C component is ~ 3/8 pellets, ~ 5/8 flocs.

Deposition

6

/11Slide11

W

sBULK = <w’c’>/<c> (mm/s)(a) Sediment Bulk Settling Velocity, W

sBULKPhase-Averaged Settling Velocity for Two Regimes Regime 1

Regime 2

Increasing |u| and τ

b

Tidal Velocity Phase (q/p)0.1 0.2 0.3 0.4 0.5Similar WsBULK at the beginning of tidal phase suggest presence of flocs during

both regimesRegime 1:

Flocs-Lower observed W

sBULK at peak |u| and τb (<0.8 mm/s)

Regime 2: Pellets+Flocs

-Lower

observed

W

sBULK

at peak |u| and τ

b (~1.2 mm/s)-Influence of pellets on WsBULK7/11

(Note that Bulk Settling Velocity, wsBULK

= <w’c’>/cset is considered reliable for mud only during accelerating half of tidal cycle.)Slide12

W

sBULK = <w’c’>/<c> (mm/s)

WsDEP = (c/(c-cwash))*WsBULK

(mm/s)Analysis of WsBULK by removing CWASH and solving for settling velocity of the depositing component (

W

sDEP

) during increasing

tb allows separate estimates for settling velocities of flocs (WsFLOCS) and pellets (WsPELLETS).(a) Sediment Bulk Settling Velocity, WsBULK(b)Phase-Averaged Settling Velocity for Two Regimes Remove cwashRegime 1 Regime

2 Tidal

Velocity Phase (q/p

)

0.1 0.2 0.3 0.4 0.5

Regime 1

Regime

2

0.1 0.2 0.3 0.4 0.5

(b) Depositing component of Settling Velocity,

W

sDEPIncreasing |u| and τb Increasing |u| and τb 8/11Recall: peak τb

~ 0.15 Pa for Regime 1, and peak τb ~ 0.22 Pa for Regime

2 Slide13

W

sBULK = <w’c’>/<c> (mm/s)

WsDEP = (c/(c-cwash))*WsBULK

(mm/s)Analysis of WsBULK by removing CWASH and solving for settling velocity of the depositing component (

W

sDEP

) during increasing

tb allows separate estimates for settling velocities of flocs (WsFLOCS) and pellets (WsPELLETS).(a) Sediment Bulk Settling Velocity, WsBULK(b)Phase-Averaged Settling Velocity for Two Regimes Remove cwashRegime 1 Regime

2 Tidal

Velocity Phase (q/p

)

0.1 0.2 0.3 0.4 0.5

Regime 1

Regime

2

0.1 0.2 0.3 0.4 0.5

(b) Depositing component of Settling Velocity,

W

sDEPIncreasing |u| and τb Increasing |u| and τb W

sFLOC = ~ 0.85 mm/sImplies floc size is limited by settling-induced shear rather than tb .

WsDEP = WsFLOCS8/11Recall: peak τb ~ 0.15 Pa for Regime 1, and peak τb ~ 0.22 Pa for Regime 2

Slide14

W

sBULK = <w’c’>/<c> (mm/s)

WsDEP = (c/(c-cwash))*WsBULK

(mm/s)Analysis of WsBULK by removing CWASH and solving for settling velocity of the depositing component (

W

sDEP

) during increasing

tb allows separate estimates for settling velocities of flocs (WsFLOCS) and pellets (WsPELLETS).(a) Sediment Bulk Settling Velocity, WsBULK(b)Phase-Averaged Settling Velocity for Two Regimes Remove cwashRegime 1 Regime

2 Tidal

Velocity Phase (q/p

)

0.1 0.2 0.3 0.4 0.5

Regime 1

Regime

2

0.1 0.2 0.3 0.4 0.5

(b) Depositing component of Settling Velocity,

W

sDEPIncreasing |u| and τb Increasing |u| and τb WsDEP = WsFLOCS

WsDEP = f

FWsFLOCS + fFWsPELLETS= ~ 1.5 mm/s at peak tb Assume: fF = 5/8, fP = 3/8 This gives:

WsPELLETS = ~ 2 mm/s

8

/11

W

sFLOC

= ~ 0.85 mm/s

Implies floc size is limited by settling-induced shear rather than

t

b

.

Recall: peak

τ

b

~ 0.15 Pa for

Regime 1

, and peak τ

b ~ 0.22 Pa for Regime

2 Slide15

25

or 120 Hour Averaged Bed Stress (Pa)25 Hour Averaged Erodibility, (kg/m2/Pa) Daily-averaged erodibility is correlated either to 5-Day-averaged

tb (Regime 1) or to daily-averaged tb (

Regime 2), revealing two distinct relationships between ε and tb.

Regime 1

:

Erodibility (

ε) increases proportional to the average stress over the last 5 days, consistent with cohesive bed evolution dominated by the consolidation state of flocs.Regime 2: Erodibility (ε) decreases with greater stress, possibly associated with the effects of bed armoring by the pellet component. Influence of Stress History on Bed Erodibility for Two Regimes

Regime 1 Regime

2 9/11Slide16

Summary

and Future Work:

York River sediment settling velocity (

W

s

) and erodibility (

ε

) are described by two contrasting regimes: (i) Regime 1: a period dominated by muddy flocs [lower Ws, higher ε].(ii) Regime 2: a period characterized by pellets mixed with flocs [higher Ws, lower ε]. Tidal phase-averaging of ADV records for the strongest 20% of tides for June to August 2007 reveals:A non-settling wash load (CWASH) is always present during both

Regimes.Once stress (τb

) exceeds an initial critical value (τcINIT) of ~ 0.02 to 0.05 Pa, sediment concentration (C) continually

increases with τb for both Regimes.As

τb decreases, cohesion of settling flocs to the surface of the seabed is inhibited for τ

b

larger than ~ 0.08

Pa for

both

Regimes

. Subtraction of CWASH from WSBULK for Regime 1 results in a stable floc settling velocity of WsFLOC ≈ 0.85 mm/s. The constant floc settling velocity implies that floc size is limited by settling-induced shear rather than turbulence associated with bed stress.Separation of WsFLOC and CWASH from WSBULK

for Regime 2 finally yields WSPELLET ≈

2 mm/s.During Regime 1, ε increases with tb averaged over the previous 5 days, consistent with cohesive bed evolution; while for Regime 2, ε decreases with daily tb, perhaps consistent with bed armoring.Future work will include (i) vertically stacked ADVs and (ii) deployment of a high-definition particle settling video camera.

10/11Slide17

Acknowledgements

Marjy

FriedrichsTim GassWayne Reisner Funding:

Julia

Moriarity

Carissa Wilkerson

Questions?11/11