Upload
others
View
1
Download
0
Embed Size (px)
Citation preview
1/28/2019
1
Modeling Cyanobacteria Movement in Milford Lake,
Part IIAuthors:
C. Bryan Young1, Ph.D., P.E.
Z. Charlie Zheng2, Ph.D.
Haidong Liu2, Ph.D.
Ted D. Harris3, Ph.D.
1University of Kansas, Department of Civil, Environmental and Architectural Engineering2University of Kansas, Department of Aerospace Engineering3University of Kansas, Kansas Biological Survey
1
Background
• Research Funded by KDHE• Use computer modeling to develop a better understanding of cyanobacteria transport
• Develop animations for different scenarios, varying:• Lake Level
• Wind Conditions
• Flow Rate
2
1/28/2019
2
Model Description
• Research Model Developed at KU• 2D hydrodynamics
• Finite difference
• Rectangular, regular mesh
• Fully implicit
• Parallelized
• Constituent transport• Sediment
• Contaminant
3
Background
4
1/28/2019
3
Background
• Two Projects• Phase 1: 2D model of algae transport in Milford Zone C
• 26 eight‐day simulations
• Project completed in 2017
• Phase 2: 2D model of algae transport in entire lake• 31 twenty‐day simulations
• Project completed in 2018
5
Background
• Two Projects• Phase 1: 2D model of algae transport in Milford Zone C
• 26 eight‐day simulations
• Project completed in 2017
• Phase 2: 2D model of algae transport in entire lake• 31 twenty‐day simulations
• Project completed in 2018
6
1/28/2019
4
Phase 2 Simulations
7
Days 1 ‐ 10 Days 11 ‐ 20 Beginning Days 1 ‐ 10 Days 11 ‐ 20
Wind Dir. Wind Speed Wind Dir. Wind Speed Lake Level Inflow Outflow Inflow Outflow
Case Bloom Location (from __) (mph) (from __) (mph) (ft) (cfs) (cfs) (cfs) (cfs)
1 1, 4, 5, 6 N 12 N 12 1138.4 25 25 2000 2000
2 1, 4, 5, 6 N 12 N 12 1141.4 25 25 2000 2000
3 1, 4, 5, 6 N 12 N 12 1144.4 25 25 2000 2000
4 1, 4, 5, 6 N 12 N 12 1146.4 25 25 2000 2000
5 1, 4, 5, 6 N 12 N 12 1141.4 25 25 25 25
6 1, 4, 5, 6 S 12 S 12 1138.4 25 25 2000 2000
7 1, 4, 5, 6 S 12 S 12 1141.4 25 25 2000 2000
8 1, 4, 5, 6 S 12 S 12 1144.4 25 25 2000 2000
9 1, 4, 5, 6 S 12 S 12 1146.4 25 25 2000 2000
10 1, 4, 5, 6 S 12 S 12 1141.4 25 25 25 25
11 1, 4, 5, 6 W 12 N 12 1138.4 25 25 2000 2000
12 1, 4, 5, 6 W 12 N 12 1141.4 25 25 2000 2000
13 1, 4, 5, 6 W 12 N 12 1144.4 25 25 2000 2000
14 1, 4, 5, 6 W 12 N 12 1146.4 25 25 2000 2000
15 1, 4, 5, 6 S 12 S 12 1138.4 25 25 25 2000
16 1, 4, 5, 6 S 12 S 12 1141.4 25 25 25 2000
17 1, 4, 5, 6 S 12 S 12 1144.4 25 25 25 2000
18 1, 4, 5, 6 S 12 S 12 1146.4 25 25 25 2000
19 1, 4, 5, 6 S 12 S 12 1141.4 100 100 100 100
20 1, 4, 5, 6 S 12 S 12 1141.4 500 500 500 500
21 1, 4, 5, 6 S 12 S 12 1141.4 1000 1000 1000 1000
22 1, 4, 5, 6 S 20 S 20 1141.4 1000 1000 1000 1000
23 1, 4, 5, 6 S 12 N 12 1444.4 25 25 25 25
24 1, 4, 5, 6 N 12 S 12 1444.4 25 25 25 25
25 1, 4, 5, 6 S 12 N 12 1444.4 25 25 2000 25
26 1, 4, 5, 6 S 12 N 12 1444.4 25 25 1000 25
27 1, 4, 5, 6 S 12 S 12 1444.4 25 25 2000 25
28 7, 8 N 12 N 12 1444.4 25 25 2000 2000
29 7, 8 E 12 E 12 1444.4 25 25 2000 2000
30 7, 8 S 12 S 12 1444.4 25 25 2000 2000
31 7, 8 W 12 W 12 1444.4 25 25 2000 2000
8
Phase 2 Results
1/28/2019
5
9
Phase 2 Results
10
Phase 2 Results
1/28/2019
6
Summary of Phase 2
• What it does do• Builds a qualitative understanding of algal transport in entire lake
• Evaluates a wide variety of wind and flow conditions
• Limitations• Simulation in 2D, not 3D
• Treats algae as a conservative tracer
11
2D vs 3D
12
1/28/2019
7
2D vs 3D
13
3D Model
• 3D hydrodynamics• Non‐hydrostatic simulation
• Vertical movements of algae
• Automated wet/dry cell detection
• Parallelized high performance cluster computing
• Algae growth and decay due to environment factors
• Algae transport (convection and diffusion
• Varying buoyancy
• Irradiation, nutrients, temperature, and salinity
14
1/28/2019
8
Comparison to the Environmental Fluid Dynamics Code (EFDC)
Pros:• the Virginia Institute of Marine Science (VIMS) and School of Marine
Science of the College of William and Mary, by Dr. John M. Hamrick.
• Multifunctional surface water modeling system
• Hydrodynamics, sediment, algae, and other contaminants
• Over 100 water bodies including rivers, lakes, reservoirs, wetlands, estuaries, and coastal ocean regions
• One of the most widely used and technically defensible hydrodynamic models in the world
Cons:• Hydrostatic assumption
• Constant vertical algae velocity
3D Hydrodynamic Validation: Oscillating Basin
Initial free surface elevation:0.1 cos
Casulli, V. (1999). “A semi-implicit finite difference method for non-hydrostatic, free-surface flows.” Int. J. Numer. Methods Fluids, 30(4), 425–440.
Initial Free Surface Elevation along X-axis
1/28/2019
9
3D Simulation of Milford Lake
17
(Foster, et al. 2015)
18
1/28/2019
10
Comparison with Surface Measurement at F1(Liu, Zheng, Young, and Harris, Ecological Modelling, 2018)
19
• Wind effect (with varying speed and direction) may be dominant to cause Microcystis transport
• Wind induced mixing may cause Microcystis colonies to sink
• Wind diminished in the last 10 hours of the simulation, and a better agreement shows.
Summary
• Phase 1 – Complete • Algal transport in Milford Lake, Zone C
• 26 scenarios, 82 total simulations
• Phase 2 – Complete • Algal transport in entire lake
• 31 scenarios, 116 total simulations
• 3D Model with Growth Dynamics Developed
20
1/28/2019
11
Questions?
21
For more information, contact:
C. Bryan Young, Ph.D., P.E. ([email protected]) or
Z. Charlie Zheng, Ph.D. ([email protected])