Louisiana State University Louisiana State University

LSU Digital Commons LSU Digital Commons

LSU Master's Theses Graduate School

July 2020

HYDRAULIC EFFECTS OF THE LOWER MISSISSIPPI RIVER HYDRAULIC EFFECTS OF THE LOWER MISSISSIPPI RIVER

BATTURE BATTURE

David P. May Louisiana State University and Agricultural and Mechanical College

Follow this and additional works at: https://digitalcommons.lsu.edu/gradschool_theses

Part of the Civil Engineering Commons, and the Hydraulic Engineering Commons

Recommended Citation Recommended Citation May, David P., "HYDRAULIC EFFECTS OF THE LOWER MISSISSIPPI RIVER BATTURE" (2020). LSU Master's Theses. 5199. https://digitalcommons.lsu.edu/gradschool_theses/5199

This Thesis is brought to you for free and open access by the Graduate School at LSU Digital Commons. It has been accepted for inclusion in LSU Master's Theses by an authorized graduate school editor of LSU Digital Commons. For more information, please contact gradetd@lsu.edu.

HYDRAULIC EFFECTS OF THE LOWER MISSISSIPPI RIVER BATTURE

A Thesis

Submitted to the Graduate Faculty of the Louisiana State University and

Agricultural and Mechanical College in partial fulfillment of

the requirements for the degree of Master of Science

in

The Department of Civil and Environmental Engineering

by David Patrick May

B.S., University of Mississippi, 2014 August 2020

ii

Table of Contents

Table of Contents ............................................................................................................... ii

Tables ................................................................................................................................. iii

Figures ................................................................................................................................ v

Abstract ............................................................................................................................ vii

Introduction ........................................................................................................................ 1

1. Literature Review ........................................................................................................ 6

2. Model Characteristics: Adaptive Hydraulics (AdH)............................................... 10 2.1 Roughness in AdH ..................................................................................... 11

3. Model Development ................................................................................................. 13

4. Methodology and Analysis ....................................................................................... 15 4.1 Steady State Hydrograph ........................................................................15 4.2 Existing Condition ..................................................................................16 4.3 Alternative Scenarios ..............................................................................17

5. Results and Analysis ................................................................................................. 25 5.1 Observation Points .....................................................................................25 5.2 River Observation Arc ...............................................................................30 5.3 Upper and Lower Model Domain Flow Split ............................................36

6. Discussion .................................................................................................................. 46

7. Summary.................................................................................................................... 49

Appendix. Supplemental Analysis ................................................................................. 50

References ........................................................................................................................ 62

Curriculum Vitae ............................................................................................................. 64

iii

Tables

Table 1. Alternative Scenario Overview ........................................................................ 20

Table 2. Upstream Percentage of Total Flow by Scenario ............................................ 44

Table 3. Downstream Percentage of Total Flow by Scenario ....................................... 45

iv

v

Figures

Figure 1. Location Map of the LMR and Model Domain Location .............................. 2

Figure 2. Model Domain and Mesh ................................................................................ 4

Figure 3. Helena, Arkansas Specific Gage Analysis ....................................................... 5

Figure 4. Model Validation ............................................................................................ 14

Figure 5. Mesh Element Resolution .............................................................................. 14

Figure 6. Idealized Steady State Hydrograph ............................................................... 16

Figure 7. Model Material Types; Channel (red), Island (orange), and batture (green)17

Figure 8. All URV Scenarios; Depth vs Equivalent Manning’s n ............................... 20

Figure 9. URV2: Existing Condition; Depth vs Equivalent Manning’s n .................. 21

Figure 10. URV3: 3 inch tall grass; Depth vs Equivalent Manning’s n ...................... 21

Figure 11. URV4:1 inch tall grass; Depth vs Equivalent Manning’s n ....................... 22

Figure 12. URV5:12 inch tall grass; Depth vs Equivalent Manning’s n ..................... 22

Figure 13. URV6: Select Cut; Depth vs Equivalent Manning’s n ............................... 23

Figure 14. URV7: Select Cut, 12 inch diameter; Depth vs Equivalent Manning’s n 23

Figure 15. URV8: 6 inch diameter; Depth vs Equivalent Manning’s n ...................... 24

Figure 16. URV9: Future Growth; Depth vs Equivalent Manning’s n ....................... 24

Figure 17. Main Channel and Overbank Observation Points ...................................... 26

Figure 18. MC4 WSE at 650,000 cfs of Discharge ...................................................... 27

Figure 19. MC4 at 1 Million cfs of Discharge .............................................................. 27

vi

Figure 20. MC4 at 1.5 Million cfs of Discharge .......................................................... 29

Figure 21. MC4 at 2.2 Million cfs of Discharge .......................................................... 29

Figure 22. River Observation Arc ................................................................................. 31

Figure 23. River Observation Arc at a Discharge of 650,000 cfs ................................ 32

Figure 24. River Observation Arc at a Discharge of 1 Million cfs .............................. 32

Figure 25. River Observation Arc at a Discharge of 1.5 Million cfs .......................... 33

Figure 26. River Observation Arc at a Discharge of 2.2 Million cfs .......................... 34

Figure 27. Alternative Scenario WSE versus Existing Condition at 1.5 Million cfs . 35

Figure 28. Alternative Scenario WSE versus Existing Condition at 2.2 Million cfs . 35

Figure 29. Upper Model Domain Flow Split Arcs ....................................................... 37

Figure 30. Upper Model Domain Flow in Main Channel ............................................ 38

Figure 31. Upper Model Domain Flow in Left Descending Overbank ...................... 38

Figure 32. Upper Model Domain Flow in Right Descending Overbank .................... 39

Figure 33. Lower Model Domain Flow Split Arcs ...................................................... 40

Figure 34. Lower Model Domain Flow in Main Channel ........................................... 40

Figure 35. Lower Model Domain Flow in Right Descending Overbank ................... 41

Figure 36. Lower Model Domain Flow in Left Descending Overbank ..................... 42

Figure 37. Upstream Main Channel Discharge as a Percent of the Total Discharge . 43

Figure 38. Downstream Main Channel Discharge as a Percent of the Total Discharge43

vii

Abstract

The Lower Mississippi River (LMR), since the creation of the Mississippi River and

Tributaries (MR&T) project, has been extensively modified to support navigation and flood risk

reduction. The components of the MR&T system, implemented to support the above missions,

have performed successfully for many decades, but not without contention. Rising stages in

recent years have led to the questioning of the hydraulic impact of navigation dikes and river

training methodologies. Multiple studies have been performed (Biedenharn, 2000, May 2017,

Mayne, 2018, and Simon, 2019) that indicate these river training structures have minimal impact

to stage during flood conditions. It has been hypothesized that the large batture area of the LMR

may play a significant role in overbank stages, particularly at flood conditions.

The LMR batture, defined as the portion of the floodplain confined by levees and/or

valley walls, is one of the largest of such riverine areas in the world and stretches continually for

nearly 700 miles. Upstream of Baton Rouge, Louisiana, the batture has an average width of

almost 6.5 miles and can be generally classified as a heavily forested area. At high flows, the

LMR batture becomes inundated, thus activating a substantial area of amplified roughness as

compared to the main channel of the Mississippi River. Through the use of a large-scale two-

dimensional hydraulic numerical model, an attempt to isolate and quantify the hydraulic effects

of the LMR batture area roughness on stage trends was conducted. Analysis of the model results

within this effort show that the batture of the LMR represents a substantial area of flow for the

Mississippi River at flood stages and the hydraulic roughness of said area has a measurable effect

on water surface elevation. Common forestry management methods, such as select cutting to

reduce tree density, resulted in a decrease of water surface elevation at flood stage in the range of

feet.

1

Introduction

The isolation and quantification of batture effects on stage in the Lower Mississippi River

(LMR) (Figure 1) requires that certain criteria must be met for accurate analysis. First, a large

enough area must be modeled/simulated to capture the cumulative effects of the batture

roughness on stage. Second, a numerical model capable of capturing the floodplain

hydrodynamics over such a large area and that also facilitates the manipulation of variables

necessary to quantify and isolate stage effects from hydraulic roughness is pertinent. Last, an

area that exhibits divergent stage trends, decreasing specific gage trends at low discharges and

increasing specific gage trends at flood discharges, is favorable as it indicates batture influence

on stage at flood flows.

DISCLAIMER: The contents of this report are not to be used for advertising, publication, or promotional purposes. Citation of trade names does not constitute an official endorsement or approval of the use of such commercial products. All product names and trademarks cited are the property of their respective owners. The findings of this report are not to be construed as an official Department of the Army position unless so designated by other authorized documents. DESTROY THIS DOCUMENT WHEN NO LONGER NEEDED. DO NOT RETURN IT TO THE ORIGINATOR.

2

Figure 1. Location Map of the LMR and Model Domain Location

3

The area selected for analysis includes nearly thirty-four river miles of the LMR south of

Helena, Arkansas (River Miles 621-655). In this area the batture ranges from seven to eleven

miles wide and is heavily forested. A United States Army Corps of Engineers (USACE) study

was completed, publication in draft as of this date, in this area with focus on dike notching

activities near Island 63 in the LMR. Additionally, a Mississippi River Geomorphology and

Potamology Program (MRG&P) study (Biedenharn, et al. 2015) identified the gage at Helena,

Arkansas, as exhibiting divergent stage trends (Figure 3). This divergence was identified through

the use of specific gage analysis. In this method of analysis certain flows, ranging from low flow

to flood flows, and their respective water surface elevations (stage) are plotted over the entire

gage time period of record. The resulting plot provides visual representation of the vertical

elevation trends at that location in the river channel. A declining trend characterizes a

degradational or lowering river bed whereas an increasing trend insinuates aggradation in the

river bed. Generally, these trends are consistent throughout the range of flows. However, at the

Helena gage these trends diverge drastically once flow activates the floodplain or batture around

1 Million cfs. This divergent phenomenon indicates substantial batture influence that could be

attributed to its vegetative roughness.

4

Figure 2. Model Domain and Mesh

A two-dimensional (2-D) Adaptive Hydraulics (AdH) numerical model was created as

part of the Island 63 effort. The simulation capabilities available in the AdH software, in regards

to the manipulation of hydraulic roughness variation, are highly suited for capturing the complex

interactions of flow over vegetation in a floodplain (see Section 2.1). Due to the overall size,

batture area captured, established 2-D model framework, and available bathymetric data

available within this model domain (Figure 2) it served as the ideal platform for the analysis

outlined within this thesis. The methodology and analysis conducted herein, with all of the above

criterion met, consists of a 2-D numerical hydraulic model which simulates scenarios of bankfull

and flood flows based on the nearest LMR gage data. Each flow scenario will be simulated on

alternative geometric meshes that represent different floodplain land management conditions

through the use of the FR URV card.

5

Land management alternative conditions from exaggerated clear-cut scenarios to large

twenty-plus-year forest growth will be analyzed based on water surface elevation and other

measurable hydraulic effects. All alternatives will have a determined Manning’s roughness

coefficient assigned based on available literature (Arcement, 1989). The objective of these

alternatives and their subsequent results is intended to isolate and quantify the water surface

elevation impacts of various roughness conditions in the batture area. By taking advantage of the

simulation capabilities of roughness in AdH, the parameters of each scenario will allow for the

isolation of roughness characteristics, ranging from dense forest to realistic land management

conditions. The resultant water surface elevations of each scenario will likely yield water surface

elevation differences reflective of the vegetation induced hydraulic resistance. Through the

above, a feasible and actionable floodplain management condition capable of lowering water

surface elevations at flood stages, could be identified.

Figure 3. Helena, Arkansas Specific Gage Analysis

6

1. Literature Review

The relationship between vegetation and hydraulic roughness, usually assigned in

the form of Manning’s roughness coefficient, has been well established in the world of

hydraulics and hydraulic modeling. It is widely understood and accepted that the

presence of vegetation in a riverine environment results in hydraulic resistance that leads

to a rise in water level. However, the cumulative hydraulic impact of vegetation on a

riverine system has rarely been evaluated for a significant reach of a riverine

environment. Equally as important, few studies assess the hydraulic impacts of the

continuing development of vegetation through time for a large reach of river. Analysis

and understanding of the balance of flood risk management, ecosystem restoration, and

land use management is needed to effectively manage river systems. The following

studies, the majority completed in Europe, separately encompass efforts to understand

riverine flood risk management in relation to floodplain management.

Kiss at al. (2019) focused on the rapid increase of invasive vegetation on the

Tisza River’s floodplain in Hungary. The study was initiated when the Tisza River

experienced, based on historical data, an increase in flood stages without an increase in

discharge. Significant land use change, from ploughed fields to widespread forestation,

and the emergence of the invasive species throughout the Tisza’s floodplain yielded an

increase in hydraulic roughness and sediment deposition that decreased flow conveyance

and increased flood levels. Modeling was used to simulate the long-term land cover

changes at three sections. These models simulated the difference between unmanaged and

managed vegetation on the floodplains. Site measurements concluded that the removal of

7

the invasive species would result in an 86% reduction in vegetative roughness in the

floodplains. When numerically modeled, the removal of the invasive species yielded a

13–34 centimeter reduction in peak flood stages. The authors stated that one of the most

important elements of floodplain management is providing rivers with rapid and

unobstructed flow conveyance in both the main channel and floodplains.

Makaske et al. (2011) focused on the government-mandated ecosystem

rehabilitation taking place in Dutch leveed floodplains and raised the question of its

compatibility with flood safety standards. The leveed floodplain of these areas strongly

resembles the LMR in terms of the land use change moving from largely agricultural to

heavily vegetated. Focused on one branch of the Dutch Rhine River, the authors

hydraulically modeled future vegetation growth succession of a large-scale riverine

ecosystem rehabilitation and determined the vegetation could lead to 0.6 meter higher

flood stages. The authors strongly suggested that future river ecosystem rehabilitation

efforts take into account vegetation succession growth, climate change, and river

engineering. While the scale here is much smaller, this study strongly resembles the LMR

batture and its land use changes, as well as its increased stages in recent history.

Klimas, (1987) focuses on the effectiveness of forested buffer zones in protection

of the revetted banks of the LMR in areas where tree clearing had occurred. Post tree

clearing in some areas resulted in excessive erosion in the overbank area. Velocities and

scour were measured at several sites exhibiting varying standing tree configurations. It

was determined that select-cut thinning and perpendicular-to-flow strip cuts were equally

effective in maintaining the flow resistance necessary to curtail damaging erosion.

8

Regardless of the forestry method used to remove trees, the author also states that a

buffer strip of at 600 feet is necessary to protect the overbank areas from excessive

erosion. Should the LMR batture move back towards a culture and practice of select

harvesting of timber the guidance in this document will be critical in protecting existing

river training while increasing flow conveyance in the floodplain.

Galema, 2009 provides a thorough evaluation of vegetation resistance as it

pertains to modeling the behavior of water levels in rivers for flood management

purposes. While noting that the presence of vegetation has a major effect on flow

resistance and stage, the author’s main focus is that of evaluating the suitability of

different vegetation resistance descriptions. The author accomplishes this through the

compilation of numerous flow data sets to evaluate the ranges of applicability of three

emergent rigid vegetation methods and seven submerged vegetation descriptions.

Through this research, and extensive literature reviews, the author concludes that

constant roughness parameters, often used in hydraulic modeling, are not suitable for

capturing vegetation resistance through a range of depths. In section 2.1, the numerical

representation of roughness utilized in this effort is discussed. Its applicability to

analyzing the hydraulic effects of the LMR batture align well with Galema, 2009 in that

constant roughness parameters, such as Manning’s roughness coefficient are likely not

appropriate for capturing the hydraulic effects of vegetative induced roughness.

The studies summarized above exhibit awareness of the relationship between

floodplain management, flood stages as a result of floodplain roughness, and the

suitability of the practices used to assess these conditions. However, the technology and

9

methodologies applied fall short in capturing, isolating, and quantifying the vegetative

roughness of the floodplain at flood stages. This being largely due to the limits of

vegetative roughness simulation through the use of Manning’s n and the scale of the

models used for analysis. The simulation of vegetative roughness contained in this effort

provides the capability to not only represent vegetative roughness as hydraulic resistance,

but to also simulate vegetation stem density and diameter. Though numerical, the

representation of vegetative roughness in terms of diameter and density allows for easier

relation to physical floodplain condition. Furthermore, the scale of the model domain

analyzed will better capture the cumulative impacts on water surface elevation onset by

hydraulic resistance in the floodplain. Lastly, through this effort a feasible floodplain

management condition can be determined that is both applicable and capable of reducing

water surface elevation at flood flows.

10

2. Model Characteristics: Adaptive Hydraulics (AdH)

The Adaptive Hydraulics model, AdH, is a finite element model that is capable of

simulating three-dimensional Navier Stokes equations, two and three-dimensional shallow water

equations, and groundwater equations (AdH website). It can be used in a serial or multiprocessor

mode on personal computers, UNIX, Silicon Graphics, and CRAY operating systems. For this

study, AdH was applied in 2-D depth-averaged mode. The AdH-2D model utilizes the depth-

averaged, Reynolds Averaged Navier- Stokes (RANS) equations under the assumption that (1)

the horizontal length scale is much greater than the vertical length scale and (2) the pressure is

hydrostatic. The assumption of a hydrostatic water column reduces the RANS equations to the

well-known 2D shallow water (SW) equations. In these equations, the conservation of mass and

momentum for a continuum of incompressible fluid is mathematically described by the

continuity and momentum equations (Brown, 2018).

The adaptive aspect of AdH is its ability to dynamically refine the domain mesh in areas

where more resolution is needed at certain times during the simulation due to changes in the flow

and/or transport conditions. However, this feature was not used in the analysis due to the high

mesh resolution of the model used. AdH also has an adaptive time-stepping capability where the

model can reduce the time step during a simulation to improve the convergence values. AdH can

simulate the transport of conservative constituents, such as dye clouds, as well as sediment

transport that is coupled to bed and hydrodynamic changes. The ability of AdH to allow the

domain to wet and dry as the tide and/or river stage changes is important for simulating the

Mississippi River and associated flood plain over the wide range of flows common to the system

(AdH website). This tool has been developed at USACE Engineer Research and Development

11

Center, Coastal and Hydraulics Laboratory (ERDC-CHL), and has been used to model sediment

transport in such varied environments as the Mississippi River (Brown, et al. 2019).

2.1 Roughness in AdH

In AdH models, the roughness value for a material type (or region) can be easily adjusted

via the boundary condition file. Note that AdH does not employ the classic form of Manning’s

equation. The classic form of Manning’s equation was originally developed empirically and is

only approximately valid over a limited range of roughness-to-depth ratios. The log profile

roughness utilized in AdH, however, is theoretically based and hence is valid over the full range

of roughness-to-depth ratios, as long as the flow is in the turbulent, rough range (Brown, 2018).

While the variation between the two methods is nominal in-channel, it can be significant in the

floodplains. For this reason, the assigned roughness values in AdH differ numerically from

classic Manning’s n values commonly discussed in the literature.

The model used in the analysis outlined in this research employed the AdH unsubmerged

rigid vegetation (FR URV) card in the batture areas. For all alternatives simulated, the FR URV

parameters are plotted as depth and equivalent Manning’s n in section 4.3. The unsubmerged

rigid vegetation method is used to compute a shear stress drag based co-efficient for use in

computing the bottom shear stress resulting from a steady (or quasi-steady) current through rigid,

unsubmerged vegetation (FR URV). FR URV input parameters include bed roughness height,

average stem diameter, and average stem density.

12

Some examples of this condition might include flow through mangrove stands, through

phragmites in coastal wetlands, or through trees and other obstructions in coastal storm surge

flooding or riverine flooding. The formulation is taken from Walton and Christensen (1980) and

it includes both the form drag induced by flow through the obstructions and the skin drag

induced by flow over the bed. The equation is given as follows:

Equation 1: 𝐶𝐶𝐷𝐷 =0.32(1−𝑚𝑚𝜋𝜋

4 𝑑𝑑2)

[ln�10.94ℎ𝑘𝑘 +1�]−2

+𝐶𝐶𝐷𝐷.𝑆𝑆.ℎ𝑚𝑚𝑚𝑚

Where: 𝐶𝐶𝐷𝐷 = 𝑏𝑏𝑏𝑏𝑚𝑚 𝑠𝑠ℎ𝑏𝑏𝑒𝑒𝑒𝑒 𝑠𝑠𝑠𝑠𝑒𝑒𝑏𝑏𝑠𝑠𝑠𝑠 𝑚𝑚𝑒𝑒𝑒𝑒𝑑𝑑 𝑐𝑐𝑐𝑐𝑏𝑏𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑏𝑏𝑐𝑐𝑠𝑠,ℎ = 𝑤𝑤𝑒𝑒𝑠𝑠𝑏𝑏𝑒𝑒 𝑚𝑚𝑏𝑏𝑑𝑑𝑠𝑠ℎ,𝑘𝑘 =

𝑏𝑏𝑒𝑒𝑒𝑒𝑐𝑐𝑒𝑒𝑒𝑒𝑒𝑒𝑏𝑏𝑐𝑐𝑠𝑠 𝑒𝑒𝑐𝑐𝑒𝑒𝑑𝑑ℎ𝑐𝑐𝑏𝑏𝑠𝑠𝑠𝑠 ℎ𝑏𝑏𝑐𝑐𝑑𝑑ℎ𝑠𝑠,𝐶𝐶𝐷𝐷.𝑆𝑆. = 𝑚𝑚𝑒𝑒𝑒𝑒𝑑𝑑 𝑐𝑐𝑐𝑐𝑏𝑏𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑏𝑏𝑐𝑐𝑠𝑠 𝑐𝑐𝑐𝑐 𝑠𝑠ℎ𝑏𝑏 𝑠𝑠𝑠𝑠𝑏𝑏𝑚𝑚𝑠𝑠,𝑚𝑚 =

𝑒𝑒𝑒𝑒𝑏𝑏𝑒𝑒𝑒𝑒𝑑𝑑𝑏𝑏 𝑠𝑠𝑠𝑠𝑏𝑏𝑚𝑚 𝑚𝑚𝑐𝑐𝑒𝑒𝑚𝑚𝑏𝑏𝑠𝑠𝑏𝑏𝑒𝑒, 𝑒𝑒𝑐𝑐𝑚𝑚 𝑚𝑚 = 𝑒𝑒𝑒𝑒𝑏𝑏𝑒𝑒𝑒𝑒𝑑𝑑𝑏𝑏 𝑠𝑠𝑠𝑠𝑏𝑏𝑚𝑚 𝑚𝑚𝑏𝑏𝑐𝑐𝑠𝑠𝑐𝑐𝑠𝑠𝑑𝑑

More details about AdH and its computational philosophy and equations can be found in CHL 2018.

13

3. Model Development

The model mesh extends from just South of Helena, Arkansas, approximately 32 river

miles downstream (Figure 2). The mesh extends east and west to Mississippi River and

Tributaries project (MR&T) levees. Multibeam bathymetry from 2015 and LiDAR from the

same time period were used for mesh development. Due to the prior purpose of this model, all

river training structures are represented in the model mesh. The horizontal datum for the model is

North American Datum of 1983 (NAD83) Zone 15, feet. The vertical datum is North American

Vertical Datum of 1988 (NAVD88), feet. The mesh domain includes over 98,000 acres (153

square miles), over 210,000 elements, and approximately 105,000 nodes. The mesh resolution is

set such that the river channel has 100 foot spacing on average, the element size increases toward

the mesh boundaries and decreases over in channel features (Figure 5). Gages within the model

domain and used in the hydrograph development and calibration include Helena, Friar Point, and

Fair Landing. The Friar Point gage, located near the middle of the model domain, was used to

verify model calibration to existing conditions utilizing steady state conditions (Figure 4). All

simulations reported in this document were run on the ERDC High Performance Computing

(HPC) system named JIM. Average simulation time was 23–26 hours for all alternatives run.

14

Figure 4. Model Validation

Figure 5. Mesh Element Resolution

15

4. Methodology and Analysis

In this section, the modeling methodology, numerical mesh general description, and alternative

roughness scenario development are discussed.

4.1 Steady State Hydrograph

The AdH model used in this effort was previously calibrated for a standard hydrograph

simulation. In order to accurately capture the hydraulic effects of the LMR batture, an idealized

steady state hydrograph, matching historic flows and stages recorded within the reach of river

captured in the model domain, was developed to ensure a steady hydraulic state was achieved for

every flow simulated. Based on existing hydraulic data on the LMR, discharges of up to 2.2

Million cubic feet per second (cfs) have occurred along this reach of river. Discharges from

290,000 cfs to 2.2 million cfs were simulated over a period of 250 days. Intermediate flows were

held constant for a minimum of ten days, and flows above 1 Million cfs were held constant for a

minimum of 20 days. A graphical representation of the steady state hydrograph used for the

analysis can be seen in Figure 6.

16

Figure 6. Idealized Steady State Hydrograph

4.2 Existing Condition

The existing condition, in the context of this analysis, refers to the calibrated

hydraulically modeled parameters of the AdH model. Calibration was achieved using the

AdH log profile roughness version of Manning’s roughness in all areas of the model

other than the overbank (batture) area, which utilized the FR URV card. Model material

types were separated into areas of main channel, islands, and overbank. The main channel

and islands were assigned AdH Manning’s roughness coefficient values of .03 and .035,

respectively. In all alternatives simulated, the FR URV parameters of the overbank

material type was the only variable adjusted. Table 1 below summarizes the existing

condition overbank area roughness FR URV parameters as well as all alternatives. An

example of the model material types can be seen in Figure 7.

17

Figure 7. Model Material Types; Channel (red), Island (orange), and batture (green)

4.3 Alternative Scenarios

In order to isolate and quantify the hydraulic effects of the LMR batture on stage,

a wide range of overbank roughness scenarios was simulated. Each of these scenarios is

intended to represent varying forms and stages of land management. In addition to the

existing condition, seven additional scenarios were modeled. These seven scenarios range

from unrealistic conditions, such as the overbank consisting of only 1 inch tall grass, to

more realistic conditions representative of a standard select cut in which timber of

specific dimensions is removed (Table 1).

18

Table 1. Alternative Scenario Overview

Scenario Bed Roughness Height

Average Stem Diameter (in)

Average Stem

Density (10ft2)

URV2: Existing Condition; 14-18 inch dimeter trees based on field observations.

.02 1.2 .04

URV3: 3 inch tall grass

.2 0 0

URV4: 1 inch tall grass

.08 0 0

URV5: 12 inch tall grass/dense undergrowth

1.0 0 0

URV6: Select cut. Half the density as existing condition

.02 1.2 .02

URV7: Half the density of existing condition. Trees greater than 1 foot in diameter removed.

.02 1.0 .02

Table cont’d.

19

Scenario Bed Roughness Height

Average Stem Diameter (in)

Average Stem

Density (10ft2)

URV8: Half the density of existing condition. Trees greater than .5 foot in diameter removed.

.02 .5 .02

URV9: Large growth trees. Future condition in which growth continues and trees are allowed to increase in density and diameter up to 18-24 inches.

.02 2 .05

For each alternative scenario, the FR URV parameters were used to calculate the

equivalent Manning’s roughness coefficient (Equation 1) based on depths encountered in the

batture area of the model to ensure agreement with data from (Arcement, 1989). Figures 8–16

also provide visualization of the FR URV card’s adaptation of hydraulic resistance as related to

depth. Figure 8 provides a summary plot of all overbank roughness scenarios, while Figures 9–

16 are each alternative scenario plotted individually. Note that in scenarios URV3, URV4, and

URV5, which simulate varying heights of grass, the greater the depth the lower the resultant

20

friction is a result of vegetative roughness. This is due to the water level reaching a depth that is

greater than the bed roughness height of the vegetation being simulated. The opposite is true for

the remaining scenarios that simulate forested areas where the friction increases as the depth

increases.

Figure 8. All URV Scenarios; Depth vs Equivalent Manning’s n

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0 5 10 15 20 25 30

EQU

IVAL

ENT

MAN

NIN

G'S

N

WATER DEPTH (FT)

Rigid, Unsubmerged Vegetative Roughness:All URV Scenarios

URV2; ExistingConditionURV3

URV4

URV5

URV6

URV7

URV8

URV9

21

Figure 9. URV2: Existing Condition; Depth vs Equivalent Manning’s n

Figure 10. URV3: 3 inch tall grass; Depth vs Equivalent Manning’s n

22

Figure 11. URV4:1 inch tall grass; Depth vs Equivalent Manning’s n

Figure 12. URV5:12 inch tall grass; Depth vs Equivalent Manning’s n

23

Figure 13. URV6: Select Cut; Depth vs Equivalent Manning’s n

Figure 14. URV7: Select Cut, no tree greater than 12 inch diameter; Depth vs Equivalent Manning’s n

24

Figure 15. URV8: no tree greater than 6 inch diameter; Depth vs Equivalent Manning’s n

Figure 16. URV9: Future Growth; Depth vs Equivalent Manning’s n

0

0.1

0.2

0.3

0.4

0.5

0.6

0 1 2 3 4 5

EQU

IVAL

ENT

MAN

NIN

G'S

N

WATER DEPTH (FT)

Rigid, Unsubmerged Vegetative Roughness:URV9

25

5. Results and Analysis

The following sections outline the results of the analysis completed within this effort. For

every method of analysis, interpretation and relation to floodplain management are given.

5.1 Observation Points

A total of six observation points were created within the model domain to record

hydraulic parameters for each of the alternative scenarios (Figure 17). The four main

channel observation points are located in the thalweg of the channel while the overbank

points are located in the center and lower portion of the model domain. Water surface

elevation at discharges of 650,000 cfs, 1 Million cfs, 1.5 Million cfs, and 2.2 Million cfs

was recorded for all six observation points. The water surface elevation trends for each of

the observation points is very similar for each respective flow. For this reason only the

results of MC4 (Figures 18–21) will be shown in the body of the report. The remaining

plots can be found in the Appendix of this document. In the following plots, please note

the variations of scale in the y-axis.

26

Figure 17. Main Channel and Overbank Observation Points

For flows of approximately 1,000,000 cfs and less, the batture of the LMR in this reach

of river is not yet activated by flows from the main channel, and the overbank roughness effect

upon the main channel is negligible. Due to this, the difference in water surface elevation of each

alternative scenario is nearly indiscernible. Differences between the conditions averages less than

a tenth of a foot. Also notable in Figures 18 and 19 is that URV9, the highest vegetative

roughness condition, reports the lowest water surface elevation value. This can be explained by

certain lower areas of overbank that are activated at these discharges holding water longer than

the other scenarios due to the increased hydraulic resistance of this scenario. Although the

difference is minuscule, it illustrates the accuracy of the model in capturing the effects of

increased vegetative roughness.

27

Figure 18. MC4 WSE at 650,000 cfs of Discharge

Figure 19. MC4 at 1 Million cfs of Discharge

28

Figures 20 and 21 show the effect of batture vegetative roughness once the floodplain is

activated. At 1.5 Million cfs, a condition in which approximately sixty percent of the entire

batture in the model domain is inundated, water surface elevation differences between each

alternative scenario range from a tenth of a foot to a half of a foot. Once the model reaches a

steady state discharge of 2.2 Million cfs (Figure 21), the batture hydraulic roughness has a

significant impact on water surface elevation between all alternatives. For this condition the

entire batture is inundated by seven to eleven feet of water. This condition yields a difference of

two feet in water surface elevation in the more extreme alternative scenarios, such as the existing

condition versus URV3. While URV3 is an unrealistic level of roughness in the LMR batture, the

select cut conditions, such as URV6, result in a one-half to one foot reduction in stage and

represent feasible floodplain management conditions.

29

Figure 20. MC4 at 1.5 Million cfs of Discharge

Figure 21. MC4 at 2.2 Million cfs of Discharge

30

5.2 River Observation Arc

In addition to the observation points used to record water surface elevation at various

locations within the model domain, an observation arc was also created along the entire length of

the main channel to record water surface elevation. (Figure 22). The observation arc runs the

entire length of the main channel thalweg in the model domain and records water surface

elevation at the specified discharges. The observation arc strongly reflects the trends of the

observation points in that the batture roughness influence is negligible until the floodplain is

activated. At major flood discharges, 1.5 Million cfs to 2.2 Million cfs, the water surface

elevation impact is in the range of feet. Figures 23–26 show all alternative scenarios plotted for

discharges of 650,000 cfs, 1 Million cfs, 1.5 Million cfs, and 2.2 Million cfs. The tailwater

controlled aspect of the AdH model is evident towards the downstream 130,000 to 200,000 feet

of the observation arc. For this reason, the diminishing effects of water surface elevation

differences near the downstream boundary are simply an artifact of the modeling technique.

31

Figure 22. River Observation Arc

32

Figure 23. River Observation Arc at a Discharge of 650,000 cfs

Figure 24. River Observation Arc at a Discharge of 1 Million cfs

142

144

146

148

150

152

154

156

158

160

162

0 50000 100000 150000 200000

WSE

(NAV

D 88

FT)

DISTANCE (FT)

River Arc WSE 650kcfs

URV2

URV3

URV4

URV5

URV6

URV7

URV8

URV9

150

155

160

165

170

175

0 50000 100000 150000 200000

WSE

(NAV

D 88

FT)

DISTANCE (FT)

River Arc WSE 1Mcfs

URV2

URV3

URV4

URV5

URV6

URV7

URV8

URV9

33

Figures 23 and 24 show that during lower discharges, there is essentially no discernible

water surface elevation difference among all alternative scenarios. However, as shown in Figures

25 and 26, the cumulative and compounding impacts of batture vegetative roughness at

discharges above approximately 1 Million cfs can induce water surface elevation change in the

range of feet.

Figure 25. River Observation Arc at a Discharge of 1.5 Million cfs

160

165

170

175

180

185

0 50000 100000 150000 200000

WSE

(NAV

D 88

FT)

DISTANCE (FT)

River Arc WSE 1.5Mcfs

URV2

URV3

URV4

URV5

URV6

URV7

URV8

URV9

34

Figure 26. River Observation Arc at a Discharge of 2.2 Million cfs

Given the negligible difference in water surface elevations between the alternative

scenarios at flows of 650,000 cfs and 1 Million cfs and the substantial difference of the same

alternative scenarios at 1.5 Million cfs and 2.2 Million cfs, additional analysis was completed on

the river observation arc at the two higher flows. The arc difference plots in Figures 27 and 28

are the difference in water surface elevations between each alternative as compared to the

existing condition. Again, the more realistic floodplain management conditions, such as the

select cutting of trees represented by URV6, URV7, and URV8, results in reduction in water

surface elevations of approximately one foot.

170

175

180

185

190

195

200

0 50000 100000 150000 200000

WSE

(NAV

D 88

FT)

DISTANCE (FT)

River Arc WSE 2.2Mcfs

URV2

URV3

URV4

URV5

URV6

URV7

URV8

URV9

35

Figure 27. Alternative Scenarios Water Surface Elevations versus Existing Condition at 1.5 Million cfs

Figure 28. Alternative Scenarios Water Surface Elevations versus Existing Condition at 2.2 Million cfs

-1.40

-1.20

-1.00

-0.80

-0.60

-0.40

-0.20

0.00

0.20

0.40

0.60

0 20000 40000 60000 80000 100000 120000 140000 160000

WSE

(NAV

D 88

FT)

DISTANCE (FT)

River Arc WSE 1.5Mcfs_WSE_DIFF

URV2 vsURV9URV2 vsURV6URV2 vsURV3URV2 vsURV4URV2 vsURV5URV2 vsURV7URV2 vsURV8

-7.00

-6.00

-5.00

-4.00

-3.00

-2.00

-1.00

0.00

1.00

2.00

0 20000 40000 60000 80000 100000 120000 140000 160000

WSE

(NAV

D 88

FT)

DISTANCE (FT)

River Arc WSE 2.2Mcfs_WSE_DIFF

URV2 vs URV9

URV2 vs URV6

URV2 vs URV3

URV2 vs URV4

URV2 vs URV5

URV2 vs URV7

URV2 vs URV8

36

5.3 Upper and Lower Model Domain Flow Split

A final form of analysis was utilized to further isolate and quantify the effects of the

LMR batture and its associated vegetative roughness on water surface elevation. In both the

upper and lower portions of the model domain, observation arcs were placed laterally across the

entire cross section. These arcs were separated into right overbank area, main channel, and left

overbank area (Figure 29). At each of the six arcs, hydraulic flux (Q) was calculated to determine

the flow distribution impacts of each alternative’s roughness variation during flows of 1.5

Million cfs to 2.2 Million cfs. (Figures 30-32.) The results provide both a quantification of the

impacts on flow and a visual representation of the divergence of stage trends.

37

Figure 29. Upper Model Domain Flow Split Arcs

38

Figure 30. Upper Model Domain Flow in Main Channel

Figure 31. Upper Model Domain Flow in Left Descending Overbank

1,000,000

1,100,000

1,200,000

1,300,000

1,400,000

1,500,000

1,600,000

1,700,000

1,800,000

1,900,000

2,000,000

5,200 5,300 5,400 5,500 5,600 5,700 5,800 5,900 6,000 6,100

Q (C

FS)

MODEL TIME (HRS)

Upstream Main Channel Q(cfs);1.5Mcfs-2.2McfsExistingConditionURV6_MC

URV9_MC

URV3_MC

URV4_MC

URV5_MC

URV7_MC

URV8_MC

0

100,000

200,000

300,000

400,000

500,000

600,000

700,000

5,200 5,300 5,400 5,500 5,600 5,700 5,800 5,900 6,000 6,100

Q (C

FS)

MODEL TIME (HRS)

Upstream Left Overbank Q(cfs);1.5Mcfs-2.2Mcfs Existing

ConditionURV6_LOVB

URV9_LOVB

URV3_LOVB

URV4_LOVB

URV5_LOVB

URV7_LOVB

URV8_LOVB

39

Figure 32. Upper Model Domain Flow in Right Descending Overbank

While the cross sections differ in size between the upper and lower portions of the model

domain, distribution of flow as related to the alternative scenarios is similar. The sizeable

difference in flow through their respective main channel and overbank areas is quantifiable

evidence of the hydraulic effects of batture roughness on not only water surface elevation but

also the distribution of flow itself. Such a change in flow distribution, and its accompanying

velocities, must be considered when determining the suitability of a land management condition.

A dramatic increase in overbank velocities could lead to excessive scour. Conversely, a

substantial decrease in overbank velocities could induce deposition and an overall decrease in

conveyance. Additional research regarding the associated velocities of varying land management

conditions would serve to determine a condition in which conveyance is restored but erosion

does not become damaging. The flow distribution analysis for the lower model domain is shown

below in Figures 33–36.

0

5,000

10,000

15,000

20,000

25,000

30,000

35,000

40,000

5,200 5,300 5,400 5,500 5,600 5,700 5,800 5,900 6,000 6,100

Q (C

FS)

MODEL TIME (HRS)

Upstream Right Overbank Q(cfs);1.5Mcfs-2.2Mcfs

ExistingConditionURV6_ROVB

URV9_ROVB

URV3_ROVB

URV4_ROVB

URV5_ROVB

URV7_ROVB

URV8_ROVB

40

Figure 33. Lower Model Domain Flow Split Arcs

Figure 34. Lower Model Domain Flow in Main Channel

1,000,000

1,200,000

1,400,000

1,600,000

1,800,000

2,000,000

2,200,000

5,200 5,300 5,400 5,500 5,600 5,700 5,800 5,900 6,000 6,100

Q (C

FS)

MODEL TIME (HRS)

Downstream Main Channel Q(cfs);1.5Mcfs-2.2Mcfs

ExistingConditionURV6_MC

URV9_MC

URV3_MC

URV4_MC

URV5_MC

URV7_MC

41

Figure 35. Lower Model Domain Flow in Right Descending Overbank

0

10,000

20,000

30,000

40,000

50,000

60,000

70,000

80,000

90,000

100,000

5,200 5,300 5,400 5,500 5,600 5,700 5,800 5,900 6,000 6,100

Q (C

FS)

MODEL TIME (HRS)

Downstream Right Overbank Q(cfs);1.5Mcfs-2.2Mcfs

ExistingConditionURV6_ROVB

URV9_ROVB

URV3_ROVB

URV4_ROVB

URV5_ROVB

URV7_ROVB

42

Figure 36. Lower Model Domain Flow in Left Descending Overbank

In the series of plots above it is evident, during flood flows, the batture roughness

seemingly dictates how much flow is conveyed through the main channel area of the cross

section. In order to grasp a better understanding of this, the main channel sections of the upper

and lower flow split arcs were used to calculate the percentage of the total cross sectional

discharge passing through the main channel for each alternative. The results of this analysis are

shown in Figures 37 and 38. For interpretive purposes, Figure 37 shows a total cross-sectional

discharge of 2.2 Million cfs in alternative scenario URV9, of the 2.2 Million cfs, 1.95 Million

cfs or 88% is flowing in the main channel. Additionally, based on the interpretation of the

figures below, the higher the overbank roughness the greater the discharge passing through the

main channel.

0

100,000

200,000

300,000

400,000

500,000

600,000

700,000

5,200 5,300 5,400 5,500 5,600 5,700 5,800 5,900 6,000 6,100

Q (C

FS)

MODEL TIME (HRS)

Downstream Left Overbank Q(cfs);1.5Mcfs-2.2Mcfs

ExistingConditionURV6_LOVB

URV9_LOVB

URV3_LOVB

URV4_LOVB

URV7_LOVB

URV8_LOVB

43

Figure 37. Upstream Main Channel Discharge as a Percent of the Total Discharge

Figure 38. Downstream Main Channel Discharge as a Percent of the Total Discharge

1,783,652, 81%

1,682,813, 76%

1,950,633, 88%

1,522,517, 69%

1,512,59…

1,549,499, 70%

1,655,631, 75%1,556,714, 70%

60%

65%

70%

75%

80%

85%

90%

95%

100%

1,500,000 1,700,000 1,900,000 2,100,000

% O

F TO

TAL

DISC

HARG

E (C

FS)

MAIN CHANNEL DISCHARGE (CFS)

Upstream Main Channel Discharge; 1.5-2.2Mcfs

US_FLOW_SPLIT_MC_Existing ConditionUS_FLOW_SPLIT_MC_URV6US_FLOW_SPLIT_MC_URV9US_FLOW_SPLIT_MC_URV3US_FLOW_SPLIT_MC_URV4US_FLOW_SPLIT_MC_URV5US_FLOW_SPLIT_MC_URV7US_FLOW_SPLIT_MC_URV8

1,873,773, 85%1,795,141, 81%

1,991,532, 90%

1,504,143, 68%1,490,848, 68%

1,545,899, 70%

1,772,113, 80%1,680,458, 76%

60%

65%

70%

75%

80%

85%

90%

95%

100%

1,400,000 1,600,000 1,800,000 2,000,000 2,200,000

% O

F TO

TAL

DISC

HARG

E (C

FS)

MAIN CHANNEL DISCHARGE (CFS)

Downstream Main Channel Discharge; 1.5-2.2Mcfs

DS_FLOW_SPLIT_MC_Existing ConditionDS_FLOW_SPLIT_MC_URV6DS_FLOW_SPLIT_MC_URV9DS_FLOW_SPLIT_MC_URV3DS_FLOW_SPLIT_MC_URV4DS_FLOW_SPLIT_MC_URV5DS_FLOW_SPLIT_MC_URV7DS_FLOW_SPLIT_MC_URV8

44

The relationship of overbank roughness and the percent of total flow being passed in the

main channel, for all alternative scenarios simulated in this effort, is summarized below in Tables

2 and 3. For both tables, the alternative scenarios are listed in descending order, from smoothest

to roughest, based on their associated physical parameters. These data suggest that a simple

select cutting to reduce the density of trees in the existing condition could reduce the main

channel flow at 2.2 Million cfs by approximately 5% (see Tables 2 and 3 below). While

seemingly small, such a reduction of flow in the main channel and consequent increase of flow

in the batture could greatly improve this section of the LMR’s ability to convey flood flows.

Table 2. Upstream Percentage of Total Flow by Scenario

Alternative 1Mcfs 1.5Mcfs 2Mcfs 2.2McfsURV4 94% 85% 78% 68%URV3 94% 86% 78% 69%URV5 95% 86% 79% 70%URV8 94% 87% 81% 70%URV7 94% 87% 81% 75%URV6 94% 88% 82% 76%

Existing Condition

(URV2)95% 90% 85% 81%

URV9 96% 93% 90% 88%

U/S Main Channel % of Total Flow

45

Table 3. Downstream Percentage of Total Flow by Scenario

Alternative 1Mcfs 1.5Mcfs 2Mcfs 2.2McfsURV4 97% 90% 79% 68%URV3 97% 90% 80% 68%URV5 97% 91% 81% 70%URV8 95% 88% 82% 76%URV7 95% 89% 84% 80%URV6 95% 88% 85% 81%

Existing Condition

(URV2) 95% 90% 87% 85%URV9 95% 92% 91% 90%

D/S Main Channel % of Total Flow

46

6. Discussion

The results of the analysis above exhibit the measurable effect of overbank

roughness on water surface elevation in the model domain utilized. Through the

manipulation of the roughness simulation capabilities in AdH, the parameters of each

scenario were represented by varying stem density and diameters of the batture

vegetation in order to isolate roughness characteristics. The resultant water surface

elevations of each scenario yielded quantifiable water surface elevation differences that

effectively captured the vegetation induced hydraulic resistance. From these results

feasible and actionable floodplain management conditions capable of lowering water

surface elevations at flood stages were identified. Due to the compounding nature of

hydraulic resistance on water surface elevation, a larger model domain would likely

produce larger variations in stage due to the removal of backwater effects. Furthermore,

the change in flow distribution between the main channel and overbank areas shown for

each alternative scenario must be taken into consideration when determining the

suitability of a land management condition. A dramatic increase in overbank velocities

could lead to excessive scour. Conversely, a substantial decrease in overbank velocities

could induce deposition and an overall decrease in conveyance. Additional research

regarding the associated velocities of varying land management conditions would serve

to determine a condition in which conveyance is restored but erosion does not become

damaging.

47

As previously stated, the alternative scenarios ranged from unrealistic conditions,

such as one inch tall grass, to realistic states such as a select cut and continued tree

growth. URV9, representative of the existing condition allowed to mature as a forest and

resulting in water surface elevation increase, is equally as important as the alternative

scenarios that resulted in water surface elevation decreases. The frequency of the LMR

reaching flood stage has greatly increased in recent times, despite some reaches being

degradational, and could be indicative of the ever increasing roughness throughout the

batture. This is a situation that will not improve without intervention in the form of

floodplain land management. URV6, which represents a feasible and operable land

management practice, resulted in water surface elevation decreases that are favorable for

flood risk management without being harmful environmentally. Water surface elevation

decreases of approximately one foot were evident in all forms of analysis undergone in

this study for URV6. Again, these decreases would likely be much more significant if

simulated in a larger model domain.

Building upon the fact that hydraulic resistance is a compounding phenomenon

and that rivers should be treated as a system, it should be noted that while an overbank

roughness reduction in an area will likely reduce water surface elevation in that area it

will not have an effect on the system as a whole. Even in the area of reduced roughness,

the backwater effect of the next downstream area of increased roughness would diminish

the decrease in water surface elevation of the land managed area. For this reason, should

an overbank roughness reduction effort be initiated on the LMR, the entire batture would

have to be treated as a system for a measurable decrease in water surface elevation to

occur. With a larger modeling effort, possibly encompassing the entire LMR and its

48

batture, a similar series of alternative scenarios could be simulated to quantify water

surface impacts on the system level. Once quantified, the resultant water surface

elevation decrease could likely justify the enactment of a floodplain management plan

that would include forestry management techniques such as select cutting to reduce tree

density.

49

7. Summary Since the establishment of the MR&T project, the LMR has been extensively engineered

to successfully meet the navigation and flood risk needs of the nation. Within the boundaries of

the MR&T project on the LMR lies the batture. Despite this, the batture has not been managed

or engineered to meet the same needs as the rest of the MR&T project area due, likely, to private

ownership of much of the batture. Regardless of ownership, the batture and its hydraulic effects

on the LMR need to be understood in order to be better managed for flood risk. The contents of

this document indicate that the batture of the LMR represents a substantial area of flow for the

Mississippi River at flood stages and the hydraulic roughness of said area has a measurable effect

on water surface elevation. Despite the small area encompassed within the model domain of this

analysis relative to the entire LMR, land management practices that result in overbank roughness

reduction provide a feasible strategy to reduce flood stages.

50

Appendix. Supplemental Analysis

A.1 MC1 at 650 kcfs

51

A.2 MC2 at 650 kcfs

A.3 MC3 at 650 kcfs

52

A.4 OVB1 at 650 kcfs

A.5 OVB2 at 650 kcfs

53

A.6 MC1 at 1Mcfs

A.7 MC2 at 1Mcfs

54

A.8 MC3 at 1Mcfs

A.9 OVB1 at 1Mcfs

55

A.10 OVB2 at 1Mcfs

A.11 MC1 at 1.5Mcfs

56

A.12 MC2 at 1.5Mcfs

A.13 MC3 at 1.5Mcfs

57

A.14 OVB1 at 1.5Mcfs

A.15 OVB2 at 1.5Mcfs

58

A.16 MC1 at 2.2Mcfs

A.17 MC2 at 2.2Mcfs

59

A.18 MC3 at 2.2Mcfs

A.19 OVB1 at 2.2Mcfs

60

A.20 OVB2 at 2.2Mcfs

A.21 River Arc at 650kcfs

61

A.22 River Arc at 650kcfs

62

References

Adaptive Hydraulics Information: https://www.erdc.usace.army.mil/Media/Fact-Sheets/Fact-Sheet-Article-View/Article/476708/adaptive-hydraulics-model-system/

Arcement, G.J (1989) Guide for Selecting Manning’s Roughness Coefficients for Natural Channels and Flood Plains, U.S. Geological Survey Water-Supply Paper 2339.

Bernard, B., (1992) Depth-Average Numerical Modeling for Curved Channels Technical Report HL-92-9, U.S. Army Corps of Engineers, Engineering Research and Development Center, Vicksburg, MS.

Biedenharn, D. S., and Watson, C. C. (1997). “Stage adjustment in the Lower Mississippi River, USA.” Regulated Rivers: Research & Management, 13(6), 517–536.

Biedenharn, D. S., L. Hubbard, and P. H. Hofman. 2000. Historical analysis of dike systems on the Lower Mississippi River. Draft Report to the U. S. Army Engineer Research and Development Center, Coastal and Hydraulics Laboratory, 172 p.; Vicksburg, MS: U. S. Army Corps of Engineers

Biedenharn, D. S., M. A. Allison, C. D. Little, Jr., C. R. Thorne, and C. C. Watson. 2015. Largescale geomorphic change in the Mississippi River from St. Louis, MO, to Donaldsonville, LA, as revealed by specific gage records. MRG&P Technical Report (in press), 50 p.; Vicksburg, MS: U. S. Army Corps of Engineers, Mississippi Valley Division.

Brown, Gary L., McAlpin, Jennifer N., Pevey, Kimberly C., Luong, Phu V., Price, Cherie R., and Barbara A. Kleiss (2019). Mississippi River Hydrodynamic and Delta Management Study: Delta Management Modeling: AdH/SEDLIB Multi-Dimensional Model Validation and Scenario Analysis Report” Coastal and Hydraulics Laboratory Technical Report. ERDC/CHL TR-19-2.

Brown, Gary, 2018. AdH Friction Library; https://www.erdc.usace.army.mil/Media/Fact-Sheets/Fact-Sheet-Article-View/Article/476708/adaptive-hydraulics-model-system/

Brown, Gary L, 2012, “A Quasi-3D Suspended Sediment Model Using a Set of Correction Factors Applied to a Depth Averaged Advection Diffusion Equation”,Proceedings, IIHR 3rd International Shallow Flows Symposium, University of Iowa, 2012.

Brown, Gary L, 2012, “Modification of the bed sediment equations of Spasojevic and Holly (1993) to account for variable porosity, variable grain specific gravity, and nonerodable boundaries” Proceedings, IIHR 3rd International Shallow Flows Symposium, University of Iowa, 2012.

63

Brown, Gary L., McAlpin, Jennifer N., Pevey, Kimberly C., Luong, Phu V., Price, Cherie R., and Barbara A. Kleiss (2019). Mississippi River Hydrodynamic and Delta Management Study: Delta Management Modeling: AdH/SEDLIB Multi-Dimensional Model Validation and Scenario Analysis Report. Coastal and Hydraulics Laboratory Technical Report. ERDC/CHL TR-19-2.

CHL (2018). Adaptive Hydraulics 2D Shallow Water (AdH-SW2D) User Manual (Version 4.6).

Fischenich J.C. 1997. Hydraulic Impacts of Riparian Vegetation, Summary of the Literature. Vicksburg (MS) US Army Engineer Waterways Experiment Station Technical Report EL-97-9.

Galema, Alida. (2009). Vegetation Resistance: Evaluation of vegetation resistance descriptors for flood management. Master’s Thesis; University of Twente.

Kiss, T., Nagy, J. (2019). “(Mis) management of floodplain vegetation: The effect of invasive species on vegetation roughness and flood levels”. Science of the Total Environment 686 (2019), pages 931-945.

Klimas, Charles V. (1987). Influence of Forest Vegetation in Mediating Overbank Flow Velocities in the Vicinity of Lower Mississippi River Revetted Banks. ERDC-MP-EL-87

Leech, James R., May, David P., and Tate O. McAlpin (2018). Mississippi River: Natchez to Baton Rouge Hydraulic and Sediment Transport Model. Mississippi River and Tributaries Report. MRG&P Report No. 19.

Makaske, B. Maas, G.J. (2011). The Influence of Floodplain Vegetation Succession on Hydraulic Roughness: Is Ecosystem Rehabilitation in Dutch Embanked Floodplains Compatible with Flood Safety Standards? Royal Swedish Academy of Sciences.

May, D.P., Biedenharn, D.S., McAlpin, T.O. 2017. Hydraulic Dike Effects Investigation on the Mississippi River: Natchez to Baton Rouge. MRG&P Technical Report (in press), Vicksburg, MS: U. S. Army Corps of Engineers, Mississippi Valley Division.

Mayne, C.M., May, D.P., Biedenharn, D.S. 2019. Empirical Analysis of Effects of Dike Systems on Channel Morphology and Flowlines. MRG&P Technical Report (in press), Vicksburg, MS: U. S. Army Corps of Engineers, Mississippi Valley Division.

Walton R., and Christensen, B. A. (1980). “Friction factors in storm surges over inland areas” J. Waterw. Port, Coastal Ocean Div., Am. Soc. Civ. Eng., 106(2), 261-271.

Winkley, B.R., 1977. Man-made Cutoffs on the Lower Mississippi River, conception, Construction, and River Response, US Army Corps of Engineer Potamology Investigations Report 300-2, Vicksburg, MS, 209 pp.

64

Curriculum Vitae

DAVID P. MAY RESEARCH HYDRAULIC ENGINEER

INLAND DEPUTY PROGRAM MANAGER RSM Develop, implement, and manage projects throughout USACE to solve complex River

Engineering problems

Principal Investigator able to complete multiphase projects, from plan development to oversight of costs, schedules, data collection, methodology, and team development/management.

Relationship builder; excel at collaborating with coworkers, customers, and other agencies to meet or exceed deliverables, project, timeline, and funding expectations.

Inland Deputy Program Manager for the National Regional Sediment Management Program Graduate of both levels of ERDC’s Leadership Development Program.

EXPERTISE

Project Management:

Relationship development, need based solution development, project level scoping, project funding management, team development, multi-disciplinary team supervision, and performance assessment

Hydraulic Engineering Disciplines:

Geomorphology, Hydraulics, Hydrodynamics, Field Data Collection, Data Management, Streambank Stabilization, Sediment Transport, Numerical Modeling,

Technical Proficiencies: 2D AdH, HEC-RAS, ArcGIS, SIAM, ISSDOTv2, Microsoft Office Suite, USACE Research Scientific Scuba Diver

PROFESSIONAL PUBLICATIONS

May, David P. “West Fork of San Jacinto River Preliminary Geomorphic Assessment”. ERDC Technical Report; June 2018

May, David P., Haring, Christopher P. “Eagle Creek Canyon Watershed Assessment”. ERDC DOTS Report; June 2018

65

May, David P., Biedenharn, David S, Mayne, Casey, “Coffee Point Dike System Pre-Notch Analysis”. MRG&P Technical Report No. X. (In preparation);June 2018 May, David P., Biedenharn, David S., McAlpin, Tate O. “Hydraulic Dike Effects Investigation Mississippi River Merriweather-Cherokee AdH Model”. MRG&P Technical Report No. X; July 2018

May, David P., Biedenharn, David S., Johnson, Brian S., and Howe, Edmund, “Mississippi River 2016 Winter Stage Trends”. MRG&P Technical Report No. 15. February 2017

May, David P., Biedenharn, David S., McAlpin, Tate, “Hydraulic Dike Effects Investigation Mississippi River Natchez to Baton Rouge”. MRG&P Technical Report No. X December 2018

May, David P., Leech, James R., McAlpin, Tate O. “Overbank Structure Rehabilitation AdH Modeling Effort”. In Draft. ERDC-CHL Technical Report; June 2018

May, David P., Leech, James R., Heath, Ronald E. , “Morganza Control Structure Forebay Numerical Hydraulic Model Investigation”. MRG&P Technical Report No. X July 2017

May, David P., Leech, James R. “Brazos and Colorado River Erosion DOTS Request”. ERDC DOTS Report; August 2016

May, David P., Haring, Christopher P. “Santa Clara Canyon Watershed Assessment”. ERDC DOTS Report; August 2017

Hamilton, Paul, May, David P., Haring, Christopher P. “Brazos River Erosion Management Study Richmond, TX”. Planning Assistance to States (PAS) Technical Memorandum August 2017. 60% Contribution.; November 2017

McAlpin, Tate O., Abraham, David D., May, David P. “The Integrated Section, Surface Difference Over Time (ISSDOTv2) Method and Numerical Code” ERDC-CHL Technical Report. 70% contribution.: 2016

Abraham, David D., Pratt, Thad, May, David P. “Missouri River Sediment Bedload Computation and Analysis’”. CHL Letter Report. 60% contribution. 2016

Shelley, John, McAlpin, Tate O., Abraham, David D, May, David P., Adenihun, J. “Verification of Bedload Transport Formula at the Sub-Cross-Section Scale” ERDC-CHL Technical Report. 30% Contribution.; 2015

Sharp, Jeremy A., Heath, Ronald E., Park, Howard E., May, David P. “Scour Model Study of Proposed Pier Extensions” CHL Letter Report. 50% contribution.;2014

66

Abraham, David D., May, David P. “Boyer Bend AdH Numerical Model Investigation” CHL Letter Report. 90% contribution.;2013

Abraham, David D., Ramos-Villanueva, Mareilys, Pratt, Thad, Ganesh, Naveeen, May, David P. “Sediment and Hydraulic Measurements with Computed Bed Load on the Missouri River, Sioux City to Hermann,2014”. ERDC-CHL Technical Report. 40% Contribution.2015

EDUCATION & PROFESSIONAL AFFILIATION/ACCOLADES

BS in Civil Engineering, UNIVERSITY OF MISSISSIPPI (Oxford, MS) GPA 3.7 MS in Civil Engineering, LOUISIANA STATE UNIVERSITY (Baton Rouge, LA) GPA

4.4 Degree Completion: August 2020

Affiliations Member, American Society of Civil Engineering Member, Engineers Without Borders, Past Vice President, Engineers Without Borders, Ole Miss Chapter Member, Society of American Military Engineers

Accolades

Director of U.S. Department of the Interior, International Technical Assistance Program letter of Thanks/Recognition for hosting a group of Cambodian Foreign National Dignitaries on a Tech Transfer Mississippi River Tour/Planning Trip in September 2016

Letter of Recognition from Colonel Richard P. Pannell, Commander of the Galveston District for my efforts in the 2016 Texas Spring Flooding

USACE Dam Safety Team Award of Excellence for 2017; 2016 “Tax Day” Flood in Houston Texas

Nominee of the 2017 Army Innovation of the Year Award as member of the Overbank Floodway Modeling Team for New Orleans District

2016 Excellence in Customer Care CHL Award Recipient Certified U.S. Army Corps of Engineers Research Scientific Scuba Diver ERDC LDP 1 Graduate 2017 ERDC LDP 2 graduate 2018