Bridging the Gap between idealized and realistic numerical river plume simulations

Bridging the Gap between idealized and realistic numerical river plume

simulations

Robert HetlandTexas A&M University

Feedback between idealized and realistic modeling

Idealiz

ed

modelin

g

Realis

tic

mod

elin

g

Numerical consistency

Physical consistency

Metrics(Depends on user

requirements)

Transition to operational models

• What are the requirements of the model end-user?– A metric that embodies this requirement.

• How do you quantify good model performance?– Skill assessment measures model’s ability to

reproduce the desired metric.

Metric - Fresh water flux

• Broad, integrative metric for buoyancy flow as a whole.

• Depends on definition of reference salinity.

• Also interested in the nature of the fresh water flux (e.g., what kind of water carries the fresh water?)

Model skill - ‘Synoptic skill’

Anatomy of a river plume

• Assume that the plume may be divided into dynamically distinct regions:– Near field where advection of

momentum is important.– Far field where wind forcing and the

earth’s rotation are the dominant physics.

Dynamically definedplume regions - cross section

Dynamically definedplume regions - cross section

Plume structure - Day 8.3 (mean flow, no wind)

Surface salinity Fresh water thickness

Mixing and the Froude # Max. vertical salt flux Froude number

Dynamically definedplume regions - cross section

Plume structure

Upwelling Downwelling DownwellingUpwelling

Surface salinity Fresh water thickness

Vertical salt flux - with windupwelling downwelling No wind

Integrated vertical salt fluxWind/no wind comparison

Wind

No wind

Surface salinity - wind vs. none

QuickTime™ and aGIF decompressorare needed to see this picture.

Gulf o

f M

ain

eSea s

urf

ace

salin

ity

Hetl

and &

Sig

nell

Fresh water flux EMCC

Fresh water flux WMCC

Model skill of buoyancy driven coastal current transport

EMCC WMCC

33 hr 7 day 33 hr 7 day

Along-shore currents -0.41 0.01 -0.87 -0.74

Cross-shore currents -1.06 -0.46 -0.51 0.57

Surface salinity -0.47 -0.36 0.83 0.98

Fresh water flux 0.52 0.86 0.03 0.66

Realistic model conclusions

• While the model does not have high skill at point-by-point comparisons, the model is good at simulating the large ‘whole plume’ scale.

• The model resolution is not high enough to resolve the estuary/near field regime, but still simulates the fresh water transport correctly.

Wind forcing must reduce the sensitivity to the source

Fresh water (salinity class)Q

f = 1

000

m3 s

-1

max= 0.1 m2 s-2Q

f = 3

000

m3 s

-1

max= 0.2 m2 s-2

Future direction

• Idealized modeling:– How does wind control the plume salinity

and spatial dimensions?– How do other types forcing (background

flow, pulsing discharge) influence water mass modification by the wind.

• Realistic modeling: – What are reasonable ranges in parameter

space?– What is required to resolve the small-scale

features of the plume?