<%BANNER%>

Channel Planform Analysis of the Leaf River and Tributaries in Mississippi

Permanent Link: http://ufdc.ufl.edu/UFE0024106/00001

Material Information

Title: Channel Planform Analysis of the Leaf River and Tributaries in Mississippi A Decade after an In-Stream Mining Moratorium
Physical Description: 1 online resource (101 p.)
Language: english
Creator: Garfield, Ursula
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: bogue, change, creek, disturbed, gravel, homo, in, lateral, leaf, migration, mining, planform, power, stream, thompson
Geography -- Dissertations, Academic -- UF
Genre: Geography thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: In-stream mining of alluvial rivers is a readily available source of construction material. River managers of the years did not consider the ecological and physical effects that removal of alluvial material has caused. Our study examines the impact of in-stream and floodplain mining on the Leaf River, Mississippi and tributaries and determines if a decade after a mining moratorium is a sufficient amount of time to stop degradation. The Leaf River is a large tributary of the Pascagoula River with no in-stream mining but some floodplain mining. The Leaf River has five major tributaries of these three rivers were analyzed. The Bowie River has large in-stream mining pits and adjacent floodplain mining, the Bogue Homo River has no history of in-stream or floodplain mining and Thompson Creek has in-stream and flood plain mining. Lateral migration rates and point bar areas have decreased in the intensively mined rivers suggesting some recovery in the channel system. An increase in vegetative cover on point bars are contributing to a level of stability in the mined reaches. Stream power is a driving force of channel change in streams and rivers. The higher the stream power the more change likely to be found in the system. The Leaf River and two tributaries were analyzed. It was found that sinuosity, point bar area, erosion change indices and average lateral migration rates of the stream channels had a correlation to specific stream power, but not all change characteristic variables were consistent for all rivers. The more disturbed streams showed a correlation to lateral migration and point bar area but not erosion change indices. The planform change response in disturbed river systems is difficult to predict. Channel response and recovery is dependent upon the extent of human disturbance such as mining and resistance factors such as geology and vegetation.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Ursula Garfield.
Thesis: Thesis (M.S.)--University of Florida, 2008.
Local: Adviser: Mossa, Joann.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0024106:00001

Permanent Link: http://ufdc.ufl.edu/UFE0024106/00001

Material Information

Title: Channel Planform Analysis of the Leaf River and Tributaries in Mississippi A Decade after an In-Stream Mining Moratorium
Physical Description: 1 online resource (101 p.)
Language: english
Creator: Garfield, Ursula
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: bogue, change, creek, disturbed, gravel, homo, in, lateral, leaf, migration, mining, planform, power, stream, thompson
Geography -- Dissertations, Academic -- UF
Genre: Geography thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: In-stream mining of alluvial rivers is a readily available source of construction material. River managers of the years did not consider the ecological and physical effects that removal of alluvial material has caused. Our study examines the impact of in-stream and floodplain mining on the Leaf River, Mississippi and tributaries and determines if a decade after a mining moratorium is a sufficient amount of time to stop degradation. The Leaf River is a large tributary of the Pascagoula River with no in-stream mining but some floodplain mining. The Leaf River has five major tributaries of these three rivers were analyzed. The Bowie River has large in-stream mining pits and adjacent floodplain mining, the Bogue Homo River has no history of in-stream or floodplain mining and Thompson Creek has in-stream and flood plain mining. Lateral migration rates and point bar areas have decreased in the intensively mined rivers suggesting some recovery in the channel system. An increase in vegetative cover on point bars are contributing to a level of stability in the mined reaches. Stream power is a driving force of channel change in streams and rivers. The higher the stream power the more change likely to be found in the system. The Leaf River and two tributaries were analyzed. It was found that sinuosity, point bar area, erosion change indices and average lateral migration rates of the stream channels had a correlation to specific stream power, but not all change characteristic variables were consistent for all rivers. The more disturbed streams showed a correlation to lateral migration and point bar area but not erosion change indices. The planform change response in disturbed river systems is difficult to predict. Channel response and recovery is dependent upon the extent of human disturbance such as mining and resistance factors such as geology and vegetation.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Ursula Garfield.
Thesis: Thesis (M.S.)--University of Florida, 2008.
Local: Adviser: Mossa, Joann.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0024106:00001


This item has the following downloads:


Full Text

PAGE 1

CHANNEL PLANFORM ANALYSIS OF THE LEAF RIVER AND TRIBUTARIES IN MISSISSIPPI: A DECADE AFTER AN IN-STREAM MINING MORATORIUM By URSULA A.B. GARFIELD A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFULLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2008 1

PAGE 2

2008 Ursula A.B. Garfield 2

PAGE 3

To my family for their unquestioning support on this journey. 3

PAGE 4

ACKNOWLEDGMENTS I want to thank my committee members Dr. Jo ann Mossa, Dr. Peter Waylen and Dr. Ilir Bejleri for their wisdom and support in finishi ng this work. Dr. Fik and Jim Rasmussen were invaluable in the interpretation of statistical results and David Coley for his work on the original dataset. The help of undergraduate students was i ndispensible for this work. It would not have been possible to complete without the help of Mary Santello and Steven Marks who spent hours and months digitizing the entire Pasca goula River basin from aerial photographs. Without the funding by the US Geological Surv ey from the pooled funds provided by the US Army Corps of Engineers, the Pat Harrison Waterway Di strict, the Mississippi Nature Conservancy for the initial project entitled Geomor phic Assessment of the Pascagoula River, this project would not have gotten off of the ground. Special thanks go to Dr. Nick Funicelli (USGS retired) for inspiring me to finish school. His encouragement allowed me to complete my undergraduate degree in Geography. I want to thank my husband Albert for putting up with me during this time; and my two sons, Andrew and Stephen, for filling in for me when the chores had to be done. 4

PAGE 5

TABLE OF CONTENTS page ACKNOWLEDGMENTS ............................................................................................................... 4 LIST OF TABLES ...........................................................................................................................7 LIST OF FIGURES .........................................................................................................................9 ABSTRACT ...................................................................................................................... .............12 CHAPTER 1 INTRODUCTION ................................................................................................................ ..14 2 ANALYSIS OF CHANNEL PLANFO RM CHANGE on THE LEAF RIVER, MISSISSIPPI AND THREE TRIBUTARIES ........................................................................17 Introduction .................................................................................................................. ...........17 Study Area ..............................................................................................................................18 Literature Review ...................................................................................................................18 Methods ..................................................................................................................................20 Sinuosity ..................................................................................................................... .....21 Point Bar Area ................................................................................................................ .22 Change Indices ................................................................................................................22 Average Lateral Migration ..............................................................................................23 Average Channel Width ..................................................................................................23 Statistical Analysis .......................................................................................................... 23 Results .....................................................................................................................................24 Sinuosity ..................................................................................................................... .....24 Point Bar Area ................................................................................................................ .25 Change Ratios ..................................................................................................................26 Average Lateral Migration ..............................................................................................27 Average Channel Width ..................................................................................................28 Statistical Analysis .......................................................................................................... 28 Discussion .................................................................................................................... ...........30 Leaf River ........................................................................................................................30 Bowie River .....................................................................................................................31 Bogue Homo River ..........................................................................................................31 Thompson Creek .............................................................................................................32 Summary and Conclusions .....................................................................................................33 3 STREAM POWER IN COMPARISON TO CHANNEL PLANFORM OCCURING ON THE LEAF RIVER, MISSISSI PPI AND TWO TRIBUTARIES ..........................................65 Introduction .................................................................................................................. ...........65 Literature Review ...................................................................................................................66 5

PAGE 6

Methods ..................................................................................................................................67 Discharge ..................................................................................................................... ....67 Stream power .................................................................................................................. .69 Statistical Analysis .......................................................................................................... 70 Results .....................................................................................................................................70 Statistical Analysis .......................................................................................................... ........72 Discussion .................................................................................................................... ...........74 Conclusions .............................................................................................................................75 4 CONCLUSIONS ................................................................................................................. ...93 APPENDIX A METHODOLOGY FOR CALCULATING AVERAGE YEARLY LATERAL MIGRATION ..................................................................................................................... ....94 B CALCULATING INSTENTANIOUS PE AK FLOW DATA USING THE LOGPEARSON TYPE III DISTRIBUTION .................................................................................95 LIST OF REFERENCES ...............................................................................................................97 BIOGRAPHICAL SKETCH .......................................................................................................101 6

PAGE 7

LIST OF TABLES Table page 2-1 Shapefile Names ........................................................................................................... .....352-2 Shapefile Attributes ...........................................................................................................352-3 Topology Rules ..................................................................................................................352-4 Pascagoula project geod atabase data utilized ....................................................................362-5 Attributes for main channel unions ....................................................................................362-6 Proportional area change ratios (Mossa and McLean 1997) .............................................362-7 Average channel width test of normality ...........................................................................372-8 Average channel width Wilcoxon Signed Rank Test statistics .........................................372-9 Average lateral migra tion tests of normality .....................................................................372-10 Average lateral migration W ilcoxon Signed Rank Test Statistics .....................................382-11 Point bar area tests of normality ........................................................................................ 382-12 Point bar area Wilcoxon Si gned Rank Test statistics ........................................................382-13 Sinuosity tests of normality ............................................................................................. ..392-14 Sinuosity Wilcoxon Signed Rank Test statistics ...............................................................392-15 Change indices tests of normality ......................................................................................402-16 Change indices Wilcoxon Signed Rank Test statistics ......................................................412-17 Leaf River change indices W ilcoxon Signed Rank Test statistics .....................................412-18 Leaf River, MS discharge data for imagery fly dates ........................................................413-1 Peak discharge data at USGS gauging stations ..................................................................763-2 Stream change characteristics ............................................................................................763-3 Shapiro-Wilk Test for normality ........................................................................................763-4 Leaf River reaches 7-40 Spearman Rank Order Correlation in comparison to stream power..................................................................................................................................77 7

PAGE 8

3-5 Leaf River reaches 41-57 Spearman Rank Order Correlation in comparison to stream power..................................................................................................................................773-6 Leaf River reaches 58-79 Spearman Rank Order Correlation in comparison to stream power..................................................................................................................................773-7 Bogue Homo River Spearman Rank Orde r Correlation in comparison to stream power..................................................................................................................................773-8 Thompson Creek Spearman Rank Order Correl ation in comparison to stream power .....773-9 Confluence and mining pond list .......................................................................................783-10 Summary of published rela tionships between lateral migration rates and other parameters.. .................................................................................................................. ......78 8

PAGE 9

LIST OF FIGURES Figure page 2-1 Pascagoula River basin a nd tributaries under study ..........................................................422-2 2005 imagery fly dates .......................................................................................................432-3 Imagery fly dates in comparison to discharge at Hattiesburg, MS ....................................442-4 Imagery fly dates in comparis on to discharge at McLain, MS ..........................................442-5 Change indices methods (Mossa and McLean 1997) ........................................................452-6 Leaf River sinuosity index 1955-2005 ...............................................................................462-7 Bowie River sinuosity index 1955 2005 ..........................................................................462-8 Thompson Creek sinuosity index 1955-2005 ....................................................................472-9 Bogue Homo River sinuosity index 1955-2005.................................................................482-10 Bowie River point bar areas 1958 2005 ..........................................................................482-11 Bogue Homo River point bar areas 1958-2005 .................................................................492-12 Thompson Creek point bar areas 1958-2005 .....................................................................492-13 Leaf River point bar areas 1955-2005 ...............................................................................502-14 Leaf River cumulative point bar area .................................................................................502-15 Bogue Homo River cumulative point bar area ..................................................................512-16 Bowie River cumulative point bar area .............................................................................512-17 Thompson Creek cumulative point bar area ......................................................................522-18 Leaf River change indices 1996-2005 ...............................................................................522-19 Bowie River in-stream mining pits ....................................................................................532-20 Leaf River instability at kilometer 98 ................................................................................542-21 Bogue Homo River ch ange indices 1996-2005 .................................................................552-22 Confluences of the Bogue Homo River and Tiger Creek ..................................................562-23 Bowie River change indices 1996-2005 ............................................................................57 9

PAGE 10

2-24 Thompson Creek change iIndices 1996-2005....................................................................572-25 Bowie River average late ral migration per year ................................................................582-26 Bogue Homo River average lateral migration per year .....................................................582-27 Thompson Creek average lateral migration Per year .........................................................592-28 Leaf River average late ral migration per year ...................................................................592-29 Bowie River average width ................................................................................................602-30 Bogue Homo River average width .....................................................................................602-31 Thompson Creek average width ........................................................................................612-32 Leaf River average width ...................................................................................................612-33 Bowie River floodplain pits in relation to river channel ....................................................622-34 Thompson Creek floodplain pits ........................................................................................632-35 Thompson Creek mining areas ..........................................................................................643-1 Two year flood return for USGS gauge st ations calculated using Log-Pearson Type III distribution vs drainage area measured at gauge stations .............................................793-2 Comparison of discharg e calculations methods .................................................................793-3 Point bar area vs stream power for Leaf River ..................................................................803-4 Sinuosity vs stream power for Leaf River .........................................................................803-5 Lateral migration vs stre am power for Leaf River .............................................................813-6 Between change indices vs stream power for Leaf River ..................................................813-7 Deposition change indices L eaf River vs stream power ....................................................823-8 Erosion change indices L eaf River vs stream power .........................................................823-9 Unchanged change indices vs stream power for Leaf River ..............................................833-10 Point bar area vs stream power for Thompson Creek ........................................................833-11 Sinuosity vs stream power for Thompson Creek ...............................................................843-12 Average lateral migration vs stream power for Thompson Creek .....................................843-13 Between change indices vs stream power for Thompson Creek .......................................85 10

PAGE 11

3-14 Deposition change indices vs stream power for Thompson Creek ....................................853-15 Erosion change indices vs st ream power for Thompson Creek .........................................863-16 Unchanged change indices vs stream power for Thompson Creek ...................................863-17 Point bar area vs stream power for Bogue Homo River ....................................................873-18 Sinuosity vs stream power for Bogue Homo River ...........................................................873-19 Average lateral migration vs st ream power for Bogue Homo River .................................883-20 Between change indices vs stream power for Bogue Homo River ....................................883-21 Deposition change indices vs str eam power for Bogue Homo River ................................893-22 Erosion change indices vs stre am power for Bogue Homo River .....................................893-23 Unchanged change indices vs str eam power for Bogue Homo River. ..............................903-24 Stream slope vs stream power............................................................................................903-25 Leaf River: Slope and stream power by reach ...................................................................913-26 Bogue Homo River: Slope and stream power by reach .....................................................913-27 Thompson Creek: Slope a nd stream power by reach.........................................................92 11

PAGE 12

Abstract of Thesis Presen ted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science CHANNEL PLANFORM ANALYSIS OF THE LEAF RIVER AND TRIBUTARIES IN MISSISSIPPI: A DECADE AFTER AN IN-STREAM MINING MORATORIUM By Ursula A.B. Garfield December 2008 Chair: Joann Mossa Major: Geography In-stream mining of alluvial rivers is a readily available so urce of construction material. River managers of the years did not consider the ecological and phys ical effects that removal of alluvial material has caused. Our study examin es the impact of in-stream and floodplain mining on the Leaf River, Mississippi and tributarie s and determines if a decade after a mining moratorium is a sufficient amount of time to stop de gradation. The Leaf Rive r is a large tributary of the Pascagoula River with no in-stream mining but some floodplain mining. The Leaf River has five major tributaries of these three rivers were analyzed. The Bowie River has large instream mining pits and adjacent floodplain mining, the Bogue Homo River has no history of instream or floodplain mining and Thompson Cr eek has in-stream and flood plain mining. Lateral migration rates and point bar areas have decreased in the intensively mined rivers suggesting some recovery in the channel system An increase in vegetative cover on point bars are contributing to a level of stability in the mined reaches. Stream power is a driving force of channel change in streams and rivers. The higher the stream power the more change likely to be found in the system. The Leaf River and two tributaries were analyzed. It was found that sinuos ity, point bar area, erosion change indices and 12

PAGE 13

13 average lateral migration rates of the stream channels had a correlation to specific stream power, but not all change characteristic variables were consistent for all rivers. The more disturbed streams showed a correlation to lateral migrati on and point bar area but not erosion change indices. The planform change response in disturbed ri ver systems is difficult to predict. Channel response and recovery is dependent upon the exte nt of human disturbance such as mining and resistance factors such as geology and vegetation.

PAGE 14

CHAPTER 1 INTRODUCTION River managers have been neglected the effects of in-stream mining due to poor documentation (Rinaldi et al. 2005). In-str eam mining has a profound morphological impact on river systems. Sand and gravel extracted are us ed in construction and road building trades. Avulsions into floodplain mining pits during large floods may ca use increased incision upstream and downstream of the mined areas. Lateral ch annel instability may cause undermining of structures such as bridges (Rinaldi et al. 2005) and are a concern to landowners and road managers. Channel migration can compromise the in tegrity of structures built in floodplains. In some instances large riprap structures must be built to prevent channel migration (Figure 2-20). The impacts of actively changing rivers increase susceptibility due to flooding do to a lack of natural levee forma tion (Charlton 2008). Stream power is a geomorphic driving force of channel change in alluvial rivers. Streams are classified into low, medium and high ener gy by their characteristics (Nanson and Croke 1992). Stream power is dependent upon the slop e and width of the reach being analyzed (Reinfelds et al. 2003). According to Brookes ( 1988), a river system main tains stability with a stream power at a particular threshold theref ore minimizing lateral migration and downstream migration (Knighton 1998) and seeking equilibrium or a state of balance within a system (Charlton 2008). Stream characteristics analyzed include channel sinuosity, point bar area, channel change over two time periods, average la teral migration and average wi dth. Sinuosity ratio gives an indication of how much bend is in a channel plan form. Channels with a sinuosity ration between of less than 1.1 are defined as straight, 1.1 and 1.5 are sinuous and channe ls with a ratio over 1.5 are meandering (Charlton 2008). Point bars are sand ba rs usually located at the inside curve of a 14

PAGE 15

channel bendway. Change indices indicate the per cent a channel has changed or stayed the same when comparing two time periods. Average late ral migration is defined as the amount the channel centerline moved per year. Average ch annel width describes the mean width of a channel. These characteristics are important because they are in indication of the health of a stream system physically and ecologically. According to Simon and Downs (1995) increasing stresses being placed on alluvial channel systems through continued encroachment of urban areas and transportation systems, identifie s a need for assessment of ch annel conditions and the relative sensitivity of channels to disturbance or altere d environmental conditions. Stream with excessive sediment can produce large sandbars in low stream power areas causing the sediment load to settle filling channels and resu lting in an increase in stream bed elevation. The change in bed elevation makes the floodplain more susceptibl e to flooding and catastrop hic land cover change. If the centerline of a channel cha nges abruptly, this is an indica tion of lateral migration of the channel may cause loss of adjacent land to landowners. Excessive sediment is also a cause of impe rilment of spawning areas in streams. Large increases in sediment loads in streams can cr eate substrate composition detrimental to fish spawning and rearing habitats (Platts et al. 1989 ). Mitigation of suspended sediments is an important management action for the conservation of fishes that spawn in benthic habitats (Burkhead and Jelks 2001). Gravel mining is a major industry in Mississipp i. The Pascagoula River, MS and tributaries is one of a few basins in southeast United Stat es with no major flow control devices such as dams and diversions (Dynesius and Nilsson 199 4), but anthropogenic acti vity such as mining, has had a diverse impact on the basin. The Leaf River combines with the Chickasawhay River to 15

PAGE 16

16 form the Pascagoula River at Merrill, MS (Figure 2-1). The Leaf River has five major tributaries that contribute to the drainage area. Mining has occurred in the Leaf River basin since the early 1900s but was expanded greatly in the 1980s (Mossa et al. 2007). In 1995, the State of Mississippi changed th e law concerning in-stream mining. The law prevented further in-stream mining, but sti ll allowed floodplain mini ng (Mississippi State Legislature 1995). Previous studies in the basin examined channel changes from the mid-1950s to the mid1990s (Mossa et al. 2007). Our study examined th e impact of mining on the planform on a major river system within the Pascagoula River basin after a decade of mining moratorium, analyzing channel characteristics and specific stream power to determine if the streams have become stabilized into equilibrium after ten years of no in-stream mining in the ba sin or are the rivers continuing to change. Of the five tributaries of the Leaf River, our study examined changes that occurred from the mid-1990s to 2005 within the Leaf River, Bowie River, Bogue Ho mo River and Thompson Creek. All rivers were compared to historic ch annel characteristics and the Leaf River and two tributaries were examined to determine if stre am power played a role in observed channel characteristics. Chapter 2 analyzes stream char acteristics such as channel sinuosity, point bar area, channel change indices, average lateral mi gration rate and averag e channel width between the mid-1990s and 2005. Chapter 3 examines if ther e is a correlation be tween specific stream power and the defined channel characteristics. The goal was to determine if sufficient time has elapsed to facilitate stream recovery after a ten-year mo ratorium on in-stream mining.

PAGE 17

CHAPTER 2 ANALYSIS OF CHANNEL PLANFORM CHANGE ON THE LEAF RIVER, MISSISSIPPI AND THREE TRIBUTARIES Introduction In a previous study concerning channel changes in the Pasca goula River basin, one of the findings was that major channel planform changes occurred in reaches affected by floodplain and in-stream mining (Mossa et al. 2007) from th e mid 1950s to mid 1990s. In 1995, a Mississippi law (Mississippi State Legisl ature) prohibited in-stream mi ning, but continued to allow floodplain mining. The rationale for studying human induced imp acts on river system include concern over future environmental changes, institutional needs and a desire to restore natural functionality to stream along with defining the destructive role of humans on nature (James and Marcus 2006). Planform changes in the river system examined are sinuosity change, point bar area change, change indices, average lateral migra tion rate and average ch annel width change. Sinuosity of a river is defined as the ratio of channel to valley lengt h (Schumm 1977). Point bar area is the size of the dry point ba rs and is variable by the water le vel in the river. Change indices are scale-independent ratios that compare channel planform changes along and between rivers (Mossa 2006). Average lateral migr ation is the migration rate of the stream centerline per year. Average channel width was determined by centerline length of the stream and the wetted area of the stream. Each of the planform changes was pl otted by an average reac h block length of 2km for the Leaf River and a 1km reach block fo r the Bowie River, Bogue Homo River and Thompson Creek. Reach blocks were created every 1km or 2km extending across the floodplain encapsulating the probable width that the channe l may migrate over the period being evaluated. The reach blocks used in our study were the same as the previous study by Mossa et al. (2007). 17

PAGE 18

Geographic Information Systems (GIS) allows for re mote analysis of the stream system which in the past would have required much field work to evaluate the study area and a large amount of time and money to complete the task of analyzin g an entire drainage basin. The use of aerial imagery that has been remotely sensed over mu ltiple time periods is an invaluable tool in assessing channel planform. This chapter will examine the stream characteristics for the Leaf River, a major tributary of the Pascagoula River, and three of the Leaf Rivers tributaries from 2005 imagery and compare them to data from the previous study by Mossa et al. (2007). The purpose of our study is to determine if rivers examined have become le ss unstable seeking equilibrium where the amount of sediment eroded and deposited are equal afte r a ten year moratorium on in-stream mining. Study Area The Pascagoula River basin is located in south east United States in the state of Mississippi (Figure 2-1). The Leaf River combines with the Chickasawhay River to form the Pascagoula River. The Leaf River has a drainage area of 9280 km2 (3580 mi2) and is free of in-stream mining with some floodplain mining occurring adjacen t to the river channel south of the City of Petal, MS. Three tributaries analyzed are as follows: the Bogue Homo River, a natural flowing river with no record of mining; the Bowie River, an extensively mined river with large in-stream and floodplain pits; and Thompson Creek, a small tributary with a history of in-stream and floodplain mining. Literature Review Geographic Information Systems (GIS) in rive r channel change analysis has been used since the early advent of the so ftware, allowing for the creation of temporal and spatial data in river floodplain studies (Oetter et al. 2004). The current technology allows for the use of varied methodologies in detecting change in river systems utilizing oblique digital imagery (Chander et 18

PAGE 19

al. 2002; Gilvear et al. 2000), multi-spectral imager y (Winterbottom and Gilvear 1997; Gilvear et al. 2004), aerial photography (Gurnell 1997) digital color aerial photogra phy (Gilvear et al. 2004) and scanned historic maps (Winterbotto m 2000; Marston et al. 2003; Oetter et al. 2004, Mossa et al., 2007) over multiple years. Some techniques utilized classified and unclassified remotely sensed raster data (G ilvear et al. 2004), and vector data (Mossa and McLean 1997; Wellmeyer et al. 2005) to detect planform ch ange (Gurnell 1997; Mossa 2006; Winterbottom 2000) and lateral migration rate s (Wellmeyer et al. 2005). In addition to GIS, channel change has also been detected using topographic surveys and channel profiles from the 19th century to present day (Rinal di and Simon 1998; VanLooy and Martin 2005). Other methods used to identify channel change over the last 100 years include sedimentation analysis, botanical, historical s ources, planimetric resurvey, cross-profiling and erosion pins (Lawer 1993). Channel change in rivers is defined by va rious authors as either caused by hydrology or sedimentation (Gregory 2006). Lateral migration of some channels is defined as expected channel change where the channel is transl ated downstream (Winterbottom 2000). Channel width is dependent upon if the river system is aggrading or degrading and incising (Winterbottom 2000). Sediment derived from incisi on is stored in point bars, terraces, deltas and floodplains playing a role in creating downstr eam features and physic al and aquatic impacts (Simon and Rinaldi 2006). Alluvial rivers are an attractive source of readily available sand and gravel for road building and construction in many areas of the United States. Mining in the rivers or on the floodplain can cause considerable disturbanc e in the channel morphology (Kondolf 1997) and biology (Meador and Layher 1998; Brown et al. 1998) caused by ex cess sedimentation upstream 19

PAGE 20

and downstream (Simon and Rinald i 2006). In-stream mining has also caused lowering of the stream bed resulting in ecological consequences to the alluvial plain downstream (Petit et al. 1996), movement of mining pits (Lee et al. 1993), movement in sand gr avel transition zones (Knighton 1999) and migration of nick points upst ream (Marston et al. 2003). With the regrowth of vegetation on the stream banks (Knox 2004) a nd sedimentation rates decreasing due to reforestation, channel stability may occur after several decades with no stream restrictions such as bridges therefore controlli ng force levels in disturbed streams (Graf 1979). Methods The imagery used in this research was the Unit ed States Department of Agriculture Farm Services Agency (USDA-FSA) NAIP 2005 county mo saic image files of Pascagoula River basin and its tributaries. The imagery was downloa ded from the Mississippi Automated Resource Information System (MARIS) website. Th e NAIP imagery product has a ground sampling distance of 2 meters and was or tho rectified to digital ortho quarter quads (DOQQ) with an accuracy of +/10 meters. The data and imagery were projected using the Mississippi Transverse Mercator (MSTM) Projection whic h is a customized Transverse Mercator projection created by MARIS Technical Center (MTC) and the Missis sippi Department of Transportation (MDOT) and based on the North American Da tum of 1983 (NAD83) (MARIS 2007). Fly dates for the imagery varied from July 7, 2005 to November 11, 2005 (Figure 2-2). Depending upon the basin, the imagery mosaic was digi tized with generally low flows or slightly above measured at Hattiesburg, MS (Figure 2-3) and McLain, MS (F igure 2-4) or if the imagery overlapped, the low flow images were used. The imagery was then put into a Geographic In formation System (GIS) using ESRI ArcGIS Editor Version 9.2 using the MSTM projecti on and NAD83 datum. The imagery was then digitized at 1:2000 scale. For each feature type, a separate shapefile was created (Table 2-1) and 20

PAGE 21

attributes for each shapefile were standardiz ed (Table 2-2). A new polygon or polyline was created at each new quadrangle or river junction, creating multiple features for each river in the shapefile. After the main channel, channel bars and vegetated channel ba rs were digitized; the channel bars and vegetated channel bars were cu t out from the main channels creating holes in the original polygon where the channel bar features were located. The whole basin was checked for topology using the topology f eature in ArcGIS and any erro rs were corrected. The topology rules utilized are detailed in Table 2-3. Our study builds on a previous geomorphic as sessment of the Pascagoula River basin. Imagery and GIS data from the previous project (Mossa et al. 2007) was utilized for the earlier years of the 1950s and 1990s. Imagery fly date s varied for 1950s data from May 2, 1955 to October 23, 1960 and the 1990s data from Februa ry 8, 1992 to January 11, 1997 for the rivers analyzed (Figures 2-2 & 2-3). The majority of the imagery in th e study area was from 1996 except for the lower 2 km of the Leaf River at the Chicksawhay River confluence which was from 1992. The same methodology was followed for this resear ch as was previous done (Mossa et al. 2007). Dr. Joann Mossa provided the geodatabase of the 1950s and 1990s data used for our analysis, University of Florida Department of Geography dated March 27, 2006. Table 2-4 details the data utilized from the previous proj ect. The shapefile database files (.dbf) were saved as Microsoft Excel (.xls) files before manipulation, otherwise th e shapefiles would have become corrupt and unusable. Sinuosity To determine the river sinuosity, the valley le ngth from the previous study dataset and 2005 center line shapef iles were overlaid with the rive r reach polygons using the intersect function in ArcGIS to create valley length by reach and 2005 center line by reach. The line 21

PAGE 22

lengths for both intersected files were then calc ulated using the software Data East LLC Xtools Pro 4.2.0 (Novosibirsk, Russia). The two shapefile databases were then opened in Microsoft Excel to create the sinuosity index by river reach. P = Lc / Lv Where P is the sinuosity index by reach, Lc is the river centerline length and Lv is the valley length by reach. The file was then sorted by reach number and the sinuosity indices were graphed in Microsoft Excel for past and present time periods. Point Bar Area To calculate the point bar area, Xtools Pro was used to calculate the area of the point bars in ArcGIS. The shapefile was then intersected with the river reach shapefile to create a point bar by reach shapefile. The database was then opene d in Excel and the areas of the polygons were summed by reach block and graphed. Change Indices The change indices were cr eated in ArcGIS using the union function by overlaying the main channel polygon shapefiles for 2005 (T2) and 1990s (T1); and for the second union 2005 (T2) and 1950s (T0) for each river analyzed (F igure 2-5) (Mossa and McLean 1997). When the union is completed topology is run again using the m ust not have gaps (area) rule to create the polygons between the other polygons. The resulting polygons were then attributed (Table 2-5). The data were opened in Microsoft Excel sorted by reach and the areas were summed by reach. Ratios were created for each reach with the init ial area being defined as I = D + U (Table 2-6). The data were sorted again by indices and graphed. 22

PAGE 23

Average Lateral Migration Average lateral migration was measured and graphed for each river by time periods using the centerline shapefile s for 2005 (T2), 1990s (T1) and 1950s (T0). To calculate lateral migration, ArcGIS Spatial Analyst tools were us ed. For specific instructions see Appendix A. Average Channel Width The mean width channel is calculated by using the area of the river polygon and dividing by the centerline length. Wa= Ar / Lr where Wa is average width of the channel by reach Ar is the area of the channel by reach and Lr is the centerline length of the channel by reach. In ArcGIS, the river of interest is selected in the main channel polygon shapefile and exported to a new shapefile. The intersect tool is used with the reach polygon shap efile to create a new shapefile (i.e.: BHmainchnl2005R ID) and Xtools Pro is run to calculate the polygon area. Using the centerline by river reach shap efile (i.e.: BHcntlineRID) created in the sinuosity measurement and Xtools Pro is used to calcula te the line length. Both files ar e opened in Excel, sorting each file by reach and summing the polygons area and centerline length. Then the centerline area field is divided by the centerline length field to ge t the average width by reach, repeating for each river and time step and graphing the results. Statistical Analysis Statistical analysis was performed on the st ream characteristic si nuosity, point bar area, channel change indices, averag e lateral migration and averag e width using SPSS software to determine if first the characteris tics are distributed normally and second to determine if the two temporal samples are the same. The non-parame tric Kolmogorov-Smirnov test for normality was used because it makes no assumption of normality. Shapiro-Wilk test of normality was also used to assess how well the observed frequency dist ribution fits the expected cumulative frequency 23

PAGE 24

curve and if the data is normally distribute d. The non-parametric Wilcoxon Signed Rank test was used to compare the stream planform charact eristics for the two time periods because the samples tested are related. Results Sinuosity Sinuosity for the four rivers was analyzed for the decade from mid-1990s to 2005. Data from the previous study was utilized for this an alysis for the mid-1950s to mid-1990s (Tables 2-6 to 2-9) (Mossa et al. 2007). Examination of the plotted graph showed river sinuosity for 2005 deferred very little from the 1990s data. The upper Leaf (Figure 2-6) at kilometer 22 did show reduced sinuosity from 1996 due to a recent meander cutoff identified in the aerial photographs. The middle Leaf River (Reach 41 to 57) and lower Leaf River (Reach 58 to 82) showed no notable changes from the mid-1990s. The Bowie River (Figure 2-7) at kilometer reac h 1 to 2 also shows a decrease in sinuosity from 2.5 to 1.8 and was caused by a change in th e confluence of Bouie Creek, Okatoma Creek which forms the Bowie River. The large pit areas of the Bowie River did not show notable changes when compared to the 1990s graph. The Bogue Homo River (Figure 2-9) sinuos ity analysis indicated no changes when compared to 1996 and very little when compared to 1950s. Thompson Creek (Figure 2-8) at reach 6 of within a mining pond sinuosity increased slightly from 2.0 to 2.1 and a slight decrease at reach 10 from 1.9 in 1990s to 1.75 in 2005. The lower reach of Thompson Creek (Reach 12 to16) showed a slight increase in the intensively mined areas when compared to 1996. 24

PAGE 25

Point Bar Area Since point bar area is affected by discharge level a similar stream gauge level would be necessary for the basin. The 2005 fly dates for the analyzed rivers were two days apart and the discharges (Table 2-18) were generally similar throughout the basin for the available imagery except for the upper reaches of the Bogue Homo River. Point bar area by reach and cumulative point bar area for all four rivers was plotted. Leaf River point bars area (Figure 2-13) appear to be stable upstream of the confluence of the Bowie River (km 94), but downstream of the confluence (km 98 to 112) point bar area has increased by 88% between the years 1996 and 2005. The Leaf River showed an increasing area trend downstream of the Bowie River confluence starting at km 106 in comparison to previous time periods (Figure 2-14). Point bar area was reduced in the mined reaches (km 11 to 15) of the Bowie River by 50% (Figure 2-10) due to infilling and vegetation of the mining pit banks, but increased 41% upstream of the mining reach (km 7 to 11). The Bowie River (Figure 2-16) showed a decreasing trend in point bar area at km 12 and an increase at km 9 to 11. The Bogue Homo River point bar area (Figure 2-11) increased by an average of 47% from km 10 to 29 and may be due to exposure of the 1996 submerged point bars. A large increase of point bar area (km 7 to 10) may be a result of very recent deforestation of land adjacent to the Snake Creek tributary. The Bogue Homo River poi nt bar (Figure 2-15) ar ea was higher for the entire river but had a similar tr end when compared to 1996 and 1958. Thompson Creek point bar area (Figure 2-12) ha s decreased by 67% in the mined area due to infilling and vegetation of th e mined pits (km 12 to 16). Thompson Creek (Figure 2-17) charts showed a decreasing trend in point bar area at km 13 although the values were considerably higher than in the 1950s. 25

PAGE 26

Change Ratios Scale independent change ratios have been us ed to determine stability of a river (Mossa and McLean 1997; Mossa et al. 2007). When short time periods are examined, the B/I ratio is typically zero, unless there are meander cutoffs, avulsions or very rapi d lateral migration. U/I shows the proportion of the river unchanged or in its initial boundary, D/ I shows what proportion is deposited from earlier time period, and E/I show the proportion of the initial channel that was eroded (Figure 2-5) (Mo ssa and McLean 1997). The Leaf River change indices graph comp aring 1996 to 2005 (Figure 2-18) indicates the area upstream of the Bowie River confluence has a steady and similar ratio of E/I and D/I with an average around 20% of the channel undergoing change with 80% in the same position. At kilometer 93 just upstream of th e confluence (Figure 2-19), the percentage of change is about 50% of the unchanged, erosion and deposition indices that are unseen in the rest of the river. Downstream of the Bowie River confluence U/I index rises to 0.85 and th en dips to 0.45 at kilometer 98 that is near the man-made sewage treatment pond (Figure 2-20) and then continues to rise to an average of 0.75. The sewage treat ment plant has a manmade rock rip rap revetment to prevent capture by the Leaf River and to redu ce lateral migration in that area. Concurrently D/I lowers to 0.35 to 0.20 and E/I lowers to an average of 0.10. At the confluence of the Chickasawhay and Pascagoula rivers the indices spike again. B/I for the entire river stayed between 0 and 0.10, with large values associat ed mostly with zones of rapid migration. Bowie River (Figure 2-23) indices compari ng 1996 to 2005 generally are level with U/I starting at 0.70 and rising to over 0.90, D/I re ducing from 0.30 to 0.15, E/I lowering from 0.30 to near 0 and B/I at 0 except for the mined reach at kilometer 13 rising to 0.10. This is much lower than the values of cited in Mossa et al. (2007) over the time period of act ive in-stream mining but indicate little change in the in-strea m pit areas from the mid-1990s to 2005. 26

PAGE 27

When comparing change indices between the years 1996 to 2005, the Bogue Homo River (Figure 2-21) B/I ratio is at or close to 0, U/I indices had an average of 0.60, D/I averages 0.40 and E/I goes from 0.0 to 0.40. The highest erosion ratios occur near th e confluence of Tiger Creek (Figure 2-22) and may be caused by land cover changes due to clearing of a forested area. Just upstream of the confluence of the Leaf Ri ver at reach 29 U/I rise s to 0.85, D/I dips to 0.15, B/I is 0.0 and E/I becomes 0.10. Thompson Creek change indices graph compar ing 1996 to 2005 (Figure 2-24) starting at kilometer 4, U/I averages only 0.45 with D/I averaging 0.55. This suggests only 45% of the channel is where it was before, with some reduc tion due to lower water le vels and migration and re-vegetation of point ba rs. E/I is near 0.20 and then rises to over 0.50 just upstream of second mining reach and lowers to below 0.10 and again ri ses near the confluence of the Leaf River. Average Lateral Migration Leaf River lateral migration rates (Figure 228) for the upper Leaf is on average 0.50 m/yr for 1996 to 2005, generally the same or slightly higher than previous through reach block 40. The area just upstream of the Bowie River conf luence changes from 0.5 to 2.25 m/yr which is lower than the previous time peri od (up to 3.1 m/yr), but downstr eam of the juncture migration rates increase to 2 to 3.5 m/yr and are higher than the previous time period, which has maxima of about 2 m/yr. After the Thompson Creek confluen ce, the migration rates are the same as the previous time period. Just upstream of the Pa scagoula River confluen ce, the rate rises dramatically to 7 m/yr for 1996 to 2005 from ju st less than 2 m/yr for 1982 to 1996, indicating a shift in the juncture with the Chickasawhay River. Average lateral migration rates per year for the Bowie River (Figure 2.25) are generally the same for 1996 to 2005 as the previous time peri od (1982 to 1996) averaging 1.5 meters per year. For the upper 11 km examined near km 12 which is before the Glendale Bridge the rate jumps to 27

PAGE 28

3.5 m/yr compared to the previous time period and then slowly returns to 1 meter at the confluence of the Leaf River. The Bogue Homo River lateral migration rate s (Figure 2-26) have increased from an average of 0.5 m/yr in 1982-1996 to an average of 1.25 m/yr for 1996-2005. Thompson Creek lateral migration (Figure 2-27) is only slightly higher in the mined area (around 1 m/yr) than the previous time period (0.5 m/yr) for the uppe r 13 km, it stabilizes to an average of 0.75 m/yr. This reach has become more stable, such that migration rates have dropped from 3 meters per year to 1.2 m/yr. Average Channel Width Leaf River average channel width (Figure 232) upstream of the Bowie River confluence stayed the same as in 1996. The lower Leaf Ri ver average width has decreased from 100 to approximately 60 meters due to incision, lower water levels or a combination. Average channel widths for the Bowie River (Figure 2-29) and Bogue Homo River (Figure 2-30) have not generally changed from 1996 to 2005. Thompson Creek average channel width (Figure 2-31) has decreased overall but especially in the mined reach of km 13 to 18. Statistical Analysis Characteristics of the Leaf River, Bowie Ri ver, Bogue Homo River, and Thompson Creek were analyzed to determine if there was statis tical evidence of change. Sinuosity, point bar area, channel change indices, average lateral migration and average channel width were first tested for normality using SPSS version 15 and then tested for significance at the 95th percentile. The tests performed were Kolmogorov-Smirnov and Shap iro-Wilk for normality and Wilcoxon Signed Rank for significance. The results of the normalit y test and significance can be found in Tables 2-7 to 2-16. Because the test for significance determined some of th e characteristics that were to 28

PAGE 29

be compared were not normal and the size of the population was small, a t-test could not be performed. A Wilcoxon Signed Rank test wa s chosen for all characteristics. The hypothesis is as follows: H0 : two samples are the same HA : two samples not the same Sinuosity for the Bowie River and Thompson Creek failed to reject the null hypothesis while the Bogue Homo and Leaf Rivers were st atistically different (T able 2-14). The Bowie River and Thompson Creek point bar areas were not statistically different and failed to reject the null hypothesis. The Bogue Homo River and all r eaches of the Leaf River show statistically different point bar areas. (Table 2-12). The change indices analysis (T able 2-16) for the Leaf River was broken up into three areas to see if they were significantly different from previous years; upper reac hes (7 to 40), middle reaches (41 to 57) and lower reaches (58 to 82) These reaches were chosen because initial analysis indicated all of reaches were significantly different (Table 2-17) when the point bar area was analyzed. When the reaches were analyz ed, the upper reaches failed to reject the null hypothesis for erosion, but was rejected for the between, unchanged and deposition change indices; middle reaches rejected the between and erosion change indices a nd failed to reject the unchanged and deposition change indices; lower r eaches failed to reject the between change indices and rejected for unchanged, erosion a nd deposition change indices. Bowie River and Bogue Homo River (Table 2-16) sh owed statistically significant di fferences for all of the change indices. The null hypothesis was rejected for betw een and erosive change indices and failed to reject for unchanged and deposition change indices on Thompson Creek. Average lateral migration only s howed that the Bowie River as not statistically different from the previous time period. In comparison the Leaf River, Bogue Homo River and Thompson Creek were statistically different and rejects the null hypothesis (Table 2-10). 29

PAGE 30

Average channel width and point bar area are al so stage dependent so water levels in the rivers could account for some of the changes det ected. Average channel widt h statistical analysis identified all of the rivers as being statisti cally different sizes between the years 1996 to 2005 (Table 2.8). Discussion The use of GIS technologies and aerial photogr aphy is invaluable in the research of channel change. The use of high resolution imagery is preferable when compared to maps and low resolution imagery. The detection of small cha nnel features such as point bars and in-stream bars is easily accomplished. Leaf River The Leaf River continues to indicate change upstream and downstream of the confluence of the Bowie River. The change indices indicat e straightening of the channel upstream of the confluence (Figure 2-19) and continued lateral migration downstream even though measures are in place to prevent migration through the use of rip-rap along the banks of the Hattiesburg sewage plant (Figure 2-20). The channel also appears to be narrowing downstream when compared to upstream of the confluence from the previous study (Mossa et al. 2007). Point bar area is also increasing downstream of the confluence compared to the previous study (Mossa et al. 2007). Wilcoxon signed-rank test is a non-parametric te st for two related samples compares the differences between measurements measured at an in terval level. Utilizing this test the statistical analysis of the Leaf River indicates the point ba r area and change indices indicate that the Leaf River for the lower reaches (58 to 82) show stat istically significant chan ge is still occurring, while the middle reaches (41 to 57) indicates eros ion is still occurring and the between area is also significantly different supporting the conclusion of straightening in the middle reaches. 30

PAGE 31

Bowie River The Bowie River is an extensively mined tributary (Figure 2-33) of the Leaf River. Sediment from upstream is filling in portions of the mined pits and vegetation is stabilizing the visible point bars as indicated by the decrease in point bar area in the mined reaches and increase in channel bars. During flood conditions, the floo dplain mining areas are very susceptible to capture (Kondolf 1997) causing diversion of the river. The change indices for the Bowie River also indicate stability in the mined areas but decreased stability directly upstream of the mined areas at 8 kilometers (Kondolf 1993). The lateral migration characteristic indicates continued stability upstream of the mined reaches and shifting of the main channel in the mined reaches as compared to the previous study (Mossa et al. 20 07). Average width is decreasing indicating some incision occurring upstream of the mined areas or decreased flows. The mined areas of the river could also be susceptible to pit migration (Lee et al. 1993). Wilcoxon signed-rank statistical an alysis of the Bowie River indicates lateral migration was not statistically significantly different for th e two change periods analyzed, but this may be due to geologic controls caused by the size of the mined area and exposed bedrock. Point bars area shows the area not significantly different from the 1990s but the change indices showed a significant difference for all change indices. Observations in the field show the in-stream mining pits may be acting as a reservoir upstream of the Glendale bridge area (Figure 2-33 ) allowing for the settlin g of sediment in the water column and resulting in infilling and vegetative encroachment on the spoil piles. Bogue Homo River The Bogue Homo River is a natural flowing trib utary of the Leaf River with no history of mining. According to Rosson (2001), 81-100% of th e land area in Perry County occupied by the Bogue Homo River is in timberl and, of which about 40% of land is owned by timber companies. 31

PAGE 32

However land use records of the early 1900s to 1930s indicated intensive timber harvesting activity (Mossa et al. 2007). The Bogue Homo River change i ndices with deposition highs of 0.45 and erosion low below 0.10 and lateral migrati on at a higher rate th an the previous study appears to be straightening in place when compared to the prev ious study (Mossa et al. 2007). But with the historical background of major landuse change, the rive r could still be considered a disturbed river due to ex cess sediment caused by clear cutting of forest for timber harvest (Simon and Downs, 1995) and continued use of the su rrounding landscape for fo rest products (Howe 2001). Point bar areas has generally stayed the same or decreased slightly except at the confluence of Tiger Creek, where poin t bar area has changed from 10,000 m2 to 25,000 m2 (Figure 2-11). This may indicate a disturbance on Tiger Creek due to land cover changes (Figure 2-22). Wilcoxon signed-rank statistical analysis of the Bogue Homo River showed statistically significant changes in point bar ar ea sinuosity and change indices. A further visual analysis of aerial photographs from 2005 indicates land cover ch anges from forest to cleared and replanted forest occurring in the area of Tiger Creek (F igure 2-22) which may be causing the indicated changes. Thompson Creek Thompson Creek is a small mined tributary of the Leaf River. As with the Bowie River, Thompson Creek shows increased stability in the mined reach areas with decreases in point bar area caused by increased vegetati on but also indicates shifting of the channe l due to sediment deposition (Kondolf 1993) which is confirmed by only 40% of th e channel being in the same location and 60% of the 1990s channel changed to land in 2005. Latera l migration has also decreased in the mined reaches but increased di rectly upstream (Figure 2-34) and downstream (Figure 2-35) of the mined areas as compared to previous studies (Mossa et al. 2007). Channel 32

PAGE 33

width has decreased overall in the basin, due to lower water levels and possible incision. The mining pits in the floodplain are highly suscepti ble to capture or avulsion in a flood event (Kondolf 1997) due to their proximity to the main channel and the inter-pit channels that exist on the floodplain (Figure 2-34). Statistical analysis of Thompson Creek show s point bar area and sinuosity as being not significantly different from the previous year analyzed indicating some stability, while the change indices show no significance in the de position and unchanged indi ces. Lateral migration is still occurring which may be due to avul sions into the mining ponds to some degree. Summary and Conclusions The use of GIS allowed for analysis of large areas of a stream basin that would not have been possible with the time money constraints allotted in our study. Th e use of the previous studies data was invaluable in determining how much change occurred in the studied streams from time period to time period. As identified in a previous study (Mossa et al. 2007) the Leaf River channel continues to change and may be affected by changes occurring on its tributaries. The addition of rip-rap is stabilizing migration of the Leaf River channel downstream of th e Bowie River confluence but is only a temporary measure to prevent migration. Areas of the Leaf River above the confluence of the Bowie appear to be straightening. Infilling of mining ponds on the Bowie River appears to be con tinuing and causing the flow of upstream sediment not to be passed into the Leaf River. Vegetation is encroaching on the spoil piles adjacent to the in-stream pits aiding with the stabilization of piles. As determined by the average lateral migration rate, the Bogue Homo River is migrating downstream but is within normal range of ch anges (Winterbottom 2000). Changes associated with disturbance are indicated in the area of Tiger Creek due to possible land cover changes. 33

PAGE 34

Thompson Creek in-stream mining pits are in filling causing shifts in channel location. Adjacent floodplain pits are susceptible to capture with large flood events. Stream recovery from human disturbance su ch as in-stream mining is unpredictable. Dependent upon the extent of disturbance the possibility of stability is not realistic for a large disturbed basin within a short time frame. Further studies in the future may be able to determine the length of recovery from human impacts of in-stream mining. Thes e changes identified indicate a river system that is continuing to adjust to disturbance during the last century, including in-stream and floodplain mining. 34

PAGE 35

Table 2-1. Shapefile names Shapefile name Type Channel_bars2005 polygon Channel_bars_veg2005 polygon Cntlines2005 polyline Main_channels05 polygon Meander_islands2005 polygon Point_bars2005 polygon Secondary_channels2005 polygon Table 2-2. Shapefile attributes Attribute Type/length RIVER Text/25 UPDATED date TECH Text/2 FEATURE Text/25 QUAD_NUM Text/25 QUAD Text/25 YEAR_ Short integer Table 2-3. Topology rules Feature Rule Feature Class Mainchannels_05 Must Not Overlap Secondary_channels2005 Must Not Overlap with Mainchannels_05 Point_bars2005 Must Not Over lap with Mainchannels_05 Point_bars2005 Must Not Overla p with Secondary_channels2005 Mainchannels_05 Must Not Over lap with Secondary_channels2005 Mainchannels_05 Must Not Overlap with Point_bars2005 Meander_islands2005 Must Not Ov erlap with Mainchannels_05 Meander_islands2005 Must Not Overlap with Secondary_channels2005 35

PAGE 36

Table 2-4. Pascagoula project geodatabase data utilized Feature Dataset Feature Class Year Type Notes ValleyLength ValleyLengths polyline Digitized from 1980s DRGs Strms90s Main_channels90s 1990s polygon Strms90s Secondary_channels90s 1990s polygon Strms90s Point_bars90s 1990s polygon Strms90s Channel_bars90s 1990s polygon Strms90s Channel_bars_veg90s 1990s polygon Strms90s Meander_islands90s 1990s polygon Strms90s Cntlines90s 1990s polyline ReachPolygons LeafRiverRIDs polygon Digitized from 1980s DRGs ReachPolygons BogueHomaRIDs polygon Digitized from 1980s DRGs ReachPolygons BowieRiverRIDs polygon Digitized from 1980s DRGs ReachPolygons ThompsonCrRIDs polygon Digitized from 1980s DRGs Strms50s Main_channels50s 1950s polygon Strms50s Secondary_channels50s 1950s polygon Strms50s Point_bars50s 1950s polygon Strms50s Channel_bars50s 1950s polygon Strms50s Channel_bars_veg50s 1950s polygon Strms50s Meander_islands50s 1950s polygon Strms50s Cntlines50s 1950s polyline Table 2-5. Attributes for main channel unions Index T0 T1 E 1996 0 D 0 2005 B 0 0 U 1996 2005 Table 2-6. Proportional area change ratios (Mossa and McLean 1997) Indices Description Comment U/I Ratio of U to I Shows proportion of initial channel area in same position D/I Ratio of D to I Shows proportion of initial channel area abandoned E/I Ratio of E to I Shows proportion of initial channel area eroded or created B/I Ratio of B to I Shows proportion of initial channel area between channels I D + U Initial area 36

PAGE 37

Table 2-7. Average channel width test of normality Kolmogorov-Smirnova Shapiro-Wilk Statistic df Sig. Statistic df Sig. Leaf Avg Width 2005 0.150 10 0.200 0.973 10 0.918 Leaf Avg Width1996 0.206 10 0.200 0.858 10 0.073 Bowie Avg Width 2005 0.256 10 0.061 0.827 10 0.030 Bowie Avg Width 1996 0.278 10 0.028 0.810 10 0.019 Thompson Avg Width 2005 0.235 10 0.126 0.877 10 0.120 Thompson Avg Width 1996 0.214 10 0.200 0.936 10 0.507 Bogue Homo Avg Width 2005 0.167 10 0.200 0.897 10 0.202 Bogue Homo Avg Width 1996 0.204 10 0.200 0.891 10 0.176 *This is a lower bound if the true significance aLilliefors Significance Correction Table 2-8. Average channel width W ilcoxon Signed Rank Test statistics Leaf Avg Width 1996 & 2005 Bowie Avg Width 1996 & 2005 Thompson Avg Width 1996 & 2005 Bogue Homo Avg Width 1996 & 2005 Z -6.146a -3.464a -3.823a -2.685a Asymp. Sig. (2-tailed) 0.000 0.001 0.000 0.007 aBased on negative ranks. Table 2-9. Average lateral migration tests of normality Kolmogorov-Smirnova Shapiro-Wilk Statistic df Sig. Statistic df Sig. Leaf Avg Lat Mig 96-05 0.246 15 0.015 0.777 15 0.002 Leaf Avg Lat Mig 82-96 0.193 15 0.137 0.837 15 0.011 Bowie Avg Lat Mig 96-05 0.343 15 0.000 0.669 15 0.000 Bowie Avg Lat Mig 82-96 0.254 15 0.010 0.672 15 0.000 Thompson Avg Lat Mig 96-05 0.253 15 0.011 0.742 15 0.001 Thompson Avg Lat Mig 82-96 0.265 15 0.006 0.630 15 0.000 Bogue Homo Avg Lat Mig 96-05 0.124 15 0.200 0.963 15 0.738 Bogue Homo Avg Lat Mig 82-96 0.158 15 0.200 0.943 15 0.416 *This is a lower bound if the true significance aLilliefors Significance Correction 37

PAGE 38

Table 2-10. Average lateral migration Wilcoxon Signed Rank Test statistics Leaf Avg Lat Mig 82-96 & Leaf Avg Lat Mig 96-05 Bowie Avg Lat Mig 82-96 & Bowie Avg Lat Mig 96-05 Thompson Avg Lat Mig 82-96 & Thompson Avg Lat Mig 96-05 Bogue Homo Avg Lat Mig 82-96 & Bogue Homo Avg Lat Mig 96-05 Z -4.634a -0.155a -2.495a -4.268a Asymp. Sig. (2-tailed) 0.000 0.877 0.013 0.000 a Based on negative ranks. Table 2-11. Point bar area Tests of normality Kolmogorov-Smirnova Shapiro-Wilk Statistic df Sig. Statistic df Sig. Leaf Pt Bars 96 Reach 1-40 0.246 16 0.011 0.845 16 0.011 Leaf Pt Bars 05 Reach 1-40 0.171 16 0.200 0.869 16 0.026 Leaf Pt Bars 96 Reach 41-57 0.190 16 0.127 0.906 16 0.100 Leaf Pt Bars 05 Reach 41-57 0.181 16 0.169 0.942 16 0.375 Leaf Pt Bars 96 Reach 58-82 0.122 16 0.200 0.925 16 0.203 Leaf Pt Bars 05 Reach 58-82 0.194 16 0.109 0.961 16 0.680 Bowie Point Bars 1996 0.219 16 0.038 0.776 16 0.001 Bowie Point Bars 2005 0.155 16 0.200 0.936 16 0.302 Thompson Point Bars 1996 0.299 16 0.000 0.636 16 0.000 Thompson Point Bars 2005 0.218 16 0.040 0.833 16 0.008 Bogue Homo Point Bars 1996 0.347 16 0.000 0.599 16 0.000 Bogue Homo Point Bars 2005 0.266 16 0.004 0.691 16 0.000 This is a lower bound if the true significance a Lilliefors Significance Correction Table 2-12. Point bar area Wilcoxon Signed Rank Test statistics Leaf Pt Bars Reach 1-40 2005 &1996 Leaf Pt Bars Reach 41-57 2005 &1996 Leaf Pt Bars Reach 58-82 2005 & 1996 Bowie Point Bars 2005 & 1996 Thompson Point Bars 2005 & 1996 Bogue Homo Point Bars 2005 & 1996 Z -4.100 b -2.249a -4.345a -0.227a -1.036a -4.114a Asymp. Sig. (2tailed) 0.000 0.025 0.000 0.820 0.300 0.000 aBased on negative ranks. bBased on positive ranks. 38

PAGE 39

Table 2-13. Sinuosity tests of normality Kolmogorov-Smirnova Shapiro-Wilk Statistic df Sig. Statistic df Sig. Sinuosity Leaf 2005 0.146 16 0.200 0.943 16 0.386 Sinuosity Leaf 1996 0.236 16 0.018 0.810 16 0.004 Sinuosity Bowie 2005 0.208 16 0.061 0.910 16 0.117 Sinuosity Bowie 1996 0.186 16 0.142 0.897 16 0.071 Sinuosity Thompson 2005 0.167 16 0.200 0.930 16 0.247 Sinuosity Thompson 1992 0.118 16 0.200 0.967 16 0.794 Sinuosity BogueHomo 2005 0.108 16 0.200 0.946 16 0.432 Sinuosity BogueHomo 1996 0.097 16 0.200 0.951 16 0.498 This is a lower bound of the true significance. aLilliefors Significance Correction Table 2-14. Sinuosity Wilcoxon Sign ed Rank Test statistics Sinuosity Leaf 1996 & 2005 Sinuosity Bowie 1996 & 2005 Sinuosity Thompson 1992 & 2005 Sinuosity BogueHomo 1996 & 2005 Sinuosity Leaf 1996 & 2005 Z -3.289a -1.758a -0.443a -2.371a -3.289a Asymp. Sig. (2-tailed) 0.001 0.079 0.658 0.018 0.001 aBased on positive ranks. 39

PAGE 40

Table 2-15. Change Indices tests of normality Kolmogorov-Smirnova Shapiro-Wilk Statistic df Sig. Statistic df Sig. LFciB8296R7-40 0.316 15 0.000 0.534 15 0.000 LFciB9605R7-40 0.461 15 0.000 0.367 15 0.000 LFciB8296R41-57 0.252 15 0.011 0.782 15 0.002 LFciB9605R41-57 0.462 15 0.000 0.311 15 0.000 LFciB8296R58-82 0.432 15 0.000 0.523 15 0.000 LFciB9605R58-82 0.489 15 0.000 0.297 15 0.000 LFciD8296R7-40 0.181 15 0.199 0.942 15 0.413 LFciD9605R7-40 0.217 15 0.055 0.842 15 0.013 LFciD8296R41-57 0.181 15 0.200* 0.929 15 0.262 LFciD9605R41-57 0.216 15 0.058 0.929 15 0.263 LFciD8296R58-82 0.169 15 0.200* 0.931 15 0.282 LFciD9605R58-82 0.126 15 0.200* 0.954 15 0.585 LFciE8296R7-40 0.210 15 0.075 0.844 15 0.014 LFciE9605R7-40 0.275 15 0.003 0.860 15 0.024 LFciE8296R41-57 0.158 15 0.200* 0.970 15 0.852 LFciE9605R41-57 0.167 15 0.200* 0.874 15 0.039 LFciE8296R58-82 0.144 15 0.200* 0.963 15 0.751 LFciE9605R58-82 0.196 15 0.125 0.892 15 0.072 LFciU8296R7-40 0.181 15 0.199 0.942 15 0.413 LFciU9605R7-40 0.217 15 0.055 0.842 15 0.013 LFciU8296R41-57 0.181 15 0.200* 0.929 15 0.262 LFciU9605R41-57 0.216 15 0.058 0.929 15 0.263 LFciU8296R58-82 0.169 15 0.200* 0.931 15 0.282 LFciU9605R58-82 0.126 15 0.200* 0.954 15 0.585 BWciB8296 0.303 15 0.001 0.694 15 0.000 BWciB9605 0.477 15 0.000 0.320 15 0.000 BWciD8296 0.137 15 0.200* 0.935 15 0.321 BWciD9605 0.135 15 0.200* 0.974 15 0.916 BWciE8296 0.253 15 0.011 0.652 15 0.000 BWciE9605 0.239 15 0.021 0.808 15 0.005 BWciU8296 0.137 15 0.200* 0.935 15 0.321 BWciU9605 0.135 15 0.200* 0.974 15 0.916 THciB8296 0.288 15 0.002 0.573 15 0.000 THciB9605 0.310 15 0.000 0.750 15 0.001 THciD8296 0.146 15 0.200* 0.961 15 0.707 THciD9605 0.184 15 0.181 0.894 15 0.076 THciE8296 0.143 15 0.200* 0.916 15 0.169 THciE9605 0.150 15 0.200* 0.882 15 0.050 THciU8296 0.146 15 0.200* 0.961 15 0.707 THciU9605 0.184 15 0.181 0.894 15 0.076 BHciB8296 0.278 15 0.003 0.802 15 0.004 BHciB9605 0.327 15 0.000 0.700 15 0.000 BHciD8296 0.182 15 0.194 0.949 15 0.503 BHciD9605 0.104 15 0.200* 0.977 15 0.944 BHciE8296 0.125 15 0.200* 0.951 15 0.544 BHciE9605 0.257 15 0.009 0.846 15 0.015 BHciU8296 0.182 15 0.194 0.949 15 0.503 BHciU9605 0.104 15 0.200* 0.977 15 0.944 This is the lower bound of the true significance a Lilliefors Significance Correction 40

PAGE 41

Table 2-16. Change indices Wilc oxon Signed Rank Test statistics LFciB960 5 & LFciB829 6 Reach 740 LFciD960 5 & LFciD829 6 Reach 740 LFciE960 5 & LFciE829 6 Reach 7-40 LFciU960 5 & LFciU829 6 Reach 740 LFciB960 5 & LFciB829 6 Reach 41-57 LFciD960 5 & LFciD829 6 Reach 41-57 LFciE960 5 & LFciE829 6 Reach 41-57 LFciU960 5 & LFciU829 6 Reach 41-57 Z -2.875a -4.796a -0.846a -4.796b -2.934a -0.639a -3.574a -0.639 b Asymp Sig. (2tailed) 0.004 0.000 0.397 0.000 0.003 0.523 0.000 0.523 Table 2-16. Change indices Wilcoxon Signed Rank Test statistics (continued) LFciB96 05 & LFciB82 96 Reach 58-82 LFciD96 05 & LFciD82 96 Reach 58-82 LFciE960 5 & LFciE829 6 Reach 58-82 LFciU960 5 & LFciU829 6 Reach 58-82 BWciB960 5 & BWciB829 6 BWciD960 5 &BWciD829 6 BWciE960 5 & BWciE829 6 BWciU960 5 & BWciU829 6 Z -1.423a -2.516b -4.372a -2.516a -2.073a -2.172a -3.516a -2.172b Asymp. Sig. (2tailed) 0.155 0.012 0.000 0.012 0.038 0.030 0.000 0.030 Table 2-16. Change indices Wilcoxon Signed Rank Test statistics (continued) THciB9 605& THciB8 296 THciD96 05& THciD82 96 THciE96 05 & THciE82 96 THciU9605 & THciU8296 BHciB96 05 & BHciB82 96 BHciD960 5 & BHciD829 6 BHciE96 05 & BHciE82 96 BHciU96 05 &BHciU82 96 Z -2.430a -1.913b -2.896a -1.913a -3.490a -2.685b -4.684a -2.685a Asymp. Sig. (2-tailed) 0.015 0.056 0.004 0.056 0.000 0.007 0.000 0.007 aBased on positive ranks bBased on negative ranks Table 2-17. Leaf River change indices Wilcoxon Signed Rank Test statistics D/I 1996-2005 D/I 1982-1996 E/I 1996-2005 E/I 1982-1996 U/I 1996-2005 U/I 1982-1996 1996-2005 B/I 1982-1996 B/I Z -2.402a -5.757a -2.402 b -4.383a Asymp. Sig. (2-tailed) 0.016 0.000 0.016 0.000 aBased on positive ranks bBased on negative ranks Table 2-18. Leaf River, MS discharge data for imagery fly Dates Date USGS Gauge Station Discharge (cfs) 7-7-2005 Hattiesburg, MS 1070 7-9-2005 Hattiesburg, MS 992 8-15-2005 Hattiesburg, MS 979 8-18-2005 Hattiesburg, MS 814 7-7-2005 McLain. MS 2380 9-8-2005 McLain. MS 5170 41

PAGE 42

Figure 2-1. Pascagoula River basi n and tributaries under study 42

PAGE 43

Figure 2-2. 2005 imagery fly dates 43

PAGE 44

100 1000 10000 100000 1000000 195019601970198019902000201 0 YearQ (Ft3/sec) Q photo dates Figure 2-3. Imagery fly dates in compar ison to discharge at Hattiesburg, MS 100 1000 10000 100000 1000000 1950196019701980199020002010 YearQ (Ft3/sec) Q photo dates Figure 2-4. Imagery fly dates in comp arison to discharge at McLain, MS 44

PAGE 45

Figure 2-5. Change indices met hods (Mossa and McLean 1997) 45

PAGE 46

Figure 2-6. Leaf River sinuosity index 1955-2005 Figure 2-7. Bowie River sinuosity index 1955 2005 46

PAGE 47

Figure 2-8. Thompson Creek sinuosity index 1955-2005 47

PAGE 48

Figure 2-9. Bogue Homo Ri ver sinuosity index 1955-2005 Figure 2-10. Bowie River point bar areas 1958 2005 48

PAGE 49

Figure 2-11. Bogue Homo Ri ver point bar areas 1958-2005 Figure 2-12. Thompson Creek point bar areas 1958-2005 49

PAGE 50

Figure 2-13. Leaf River point bar areas 1955-2005 Figure 2-14. Leaf River cumulative point bar area 50

PAGE 51

Figure 2-15. Bogue Homo River cumulative point bar area Figure 2-16. Bowie River cumulative point bar area 51

PAGE 52

Figure 2-17. Thompson Creek cu mulative point bar area 0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 1.00 143046627894110126142158 Distance (km)Change Indice B/I D/I E/I U/I Bowie River E Tallahala Cr. Bogue Homa River Thompson Cr. Figure 2-18. Leaf River change indices 1996-2005 52

PAGE 53

Figure 2-19. Bowie River in-stream mining pits 53

PAGE 54

Figure 2-20. Leaf River inst ability at kilometer 98 54

PAGE 55

0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 1.00 2345678910111213141516171819202122232425262728293031 Distance (km)Change Indice B/I D/I E/I U/I Tiger Creek Buck Creek Figure 2-21. Bogue Homo Ri ver change indices 1996-2005 55

PAGE 56

Figure 2-22. Confluences of the B ogue Homo River and Tiger Creek 56

PAGE 57

g 0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 1.00 12345678910111213141516 Distance (km)Change Indice B/I D/I E/I U/I Mining area Figure 2-23. Bowie River change indices 1996-2005 Figure 2-24. Thompson Creek change indices 1996-2005 57

PAGE 58

0 1 1 2 2 3 3 4 4 5 12345678910111213141516 D i s t a n c e ( m / Y e a r ) Distance (km) 1996-2005 1982-1996 1955-1982 6 -RR bridge, Peps Point 10 -I-59 bridge 12-15 -Reach of intensive Mining 16 -Mining pond, US11 bridge, USGS gauge 13 Glendale bridge Figure 2-25. Bowie River average lateral migration per year 0.0 0.5 1.0 1.5 2.0 2.5 24681012141618202224262830Distance (m/Year)Distance (km) 1996-2005 1982-1996 1955-1982 2 -Bridge Ovett Rd 12 -Bridge Whitfield Rd & USGS gauge Richton 17 Bridge SR 42 18 -Buck Creek juncture 8 -Tiger Creek juncture 28 -Bridge Old Augusta Rd Figure 2-26. Bogue Homo River aver age lateral migr ation per year 58

PAGE 59

Figure 2-27. Thompson Creek averag e lateral migration per year 0 1 2 3 4 5 6 7 81422303846546270788694102110118126134142150158D i s t a n c e ( m / y e a r ) Distance (km) 1996-2005 1982-1996 1955-1982 20 -W. Tallahala Creek 34 Fisher Creek 46 Oakaha y Creek 60 -Big Creek 66 Lowrey Creek 94 -USGS gauge Hattiesburg, US 11 bridge, 100 No Name 118 -SR 29 bridge, USGS gauge New Augusta, E Tallahala Creek 124 Bogue Homa River 136 Thompson Creek 144 Gaines Creek 148 USGS gauge, SR57 bridge McLain Figure 2-28. Leaf River averag e lateral migration per year 59

PAGE 60

0 25 50 75 100 125 150 175 200 225 250 275 12345678910111213141516Width (m)Distance (km) 2005 1996 1955 Figure 2-29. Bowie River average width 0 5 10 15 20 25 30 35 40 45 135791113151719212325272931Width (m)Distance (km) 2005 1996 1955 Figure 2-30. Bogue Homo River average width 60

PAGE 61

0 10 20 30 40 50 60 70 45678910111213141516171819 W i d t h ( m ) Distance (km) 2005 1996 1955 Figure 2-31. Thompson Creek average width 0 10 20 30 40 50 60 70 80 90 100 110 120 130 1432506886104122140158 W i d t h ( m ) Distance (km) 2005 1996 1955 Figure 2-32. Leaf River average width 61

PAGE 62

Figure 2-33. Bowie River floodplain p its in relation to river channel 62

PAGE 63

Figure 2-34. Thompson Creek floodplain pits 63

PAGE 64

Figure 2-35. Thompson Creek mining areas 64

PAGE 65

CHAPTER 3 STREAM POWER IN COMPARISON TO CHANNEL PLANFORM OCCURING ON THE LEAF RIVER, MISSISSIPPI AND TWO TRIBUTARIES Introduction Anthropogenic activities such as floodplain and in-stream mining leave rivers more vulnerable to various types of channel change. Th e purpose of this analysis was to determine if stream power had any effect on stream characteristics in river basins. Rhodes (1987) identifies many definitions of stre am power and identifies stream power as a factor in channel change. Stream power formulas generally utilize the weig ht of water, drainage area, slope and discharge in the calculation. As drainage area increases, discharge increases which directly affects stream power. Specific st ream power adds width of the channel to the analysis and was therefore utilized in this analysis (Knighton 1999). Brookes (1988: 172) identified streams that have specific stream power of greater than 35 m-2 as actively changing and seeking dynamic equilibrium while specific stream power in the range of 1 to 35 m-2 are stable (Brookes 1988: 98). Nans on and Croke (1992) also identified in their classification system, rivers with stream power ranges 10 to 60 m-2 as meandering streams. The Leaf River, Bogue Homo River and Thompson Creek are meandering rivers with lateral point bars. Accord ing to Nanson and Croke (1992) in th eir classification of rivers based on stream power and characteristics, categor izes these rivers as the medium-energy. Because of the few gauging station located though out the research area, this analysis also considers varying methods of determining stream discharge and determine how stream powers varies downstream. The calculated stream discharg e was used to calculate specific stream power for the Leaf River, Bogue Homo River and Thompson Creek, MS by 1km to 2km reaches. Changes analyzed from 1996 to 2005 on the Leaf River and two of its tributaries Bogue Homo River and Thompson Creek were used as indicators of stream change. Stream characteristics 65

PAGE 66

used in this analysis from the previous chapter were lateral mi gration, sinuosity, point bar area, and change indices. The purpose of the study is to determine if stream power is influencing any of the identified stream characteristics. Literature Review Studies of disturbed low stream power wa tersheds have identified a response to channelization as being induced by the distur bance (Urban and Rhoads 2003) even though the amount of available stream power was below the 100 threshold of reco very of sinuosity (Magilligan 1992) or the 35 level of persistent erosiv e adjustments (Brookes 1985). To evaluate potential instability in alluvial rivers methods includi ng site evaluations and GIS based input was used to evaluate the magnitude of and type of instabili ty in the evaluation of large basin areas (Simon and Downs 1995). This method evaluated the basin using available stream power and identified a threshold between erosion and deposition and the level of critical stream power as basin area increased. Mined streams are susceptible to incision caused by excess stream power for the amount of available sediment. The nature of channel inci sion in disturbed alluvial systems and some of the drivers include incision (Simon and Rinaldi 2 006). Incision was placed into a broader context of drivers and resisting forces governing channel adjustment by an alyzing channel changes as an imbalance between sediment load and stream power. Stream patterns have been examined to dete rmine with evaluated peak flows and stream power if the channel is sus ceptible to change over time (S chumm 1985). Morphological effects such as upstream and downstream incision of the mined areas in turn cause hydrological effects such as water table lowering and reduced flood frequency. Incision causes increased sediment downstream and reduced flow into estuaries in turn causing loss of riparian and aquatic habitat, and a variety of physical, ecological and envi ronmental effects (Rinaldi et al. 2005). 66

PAGE 67

Previous studies indicated a relationship betw een change such as channel migration and stream power in fluvial systems (Nanson and Hi ckin 1986). Richard et al (2005) summarized the relationship between lateral migration rates and channel-form variables from previous studies (Table 3.10) and indicated as flow energy increases lateral migration rates also increase (Nanson and Hickin 1986; Lawler et al 1999). Lateral migration rates were determined to be correlated with active channel width and total st ream power (Richard et al (2005). Methods To determine stream power the discharge of the river was calculated. A two year flood return was used as a baseline for this analysis. In most rivers this value is close to the bankfull flow (Knighton 1998) and the Q1.5 year flood (Leopold et al. 1964). Discharge The two methods of calculation stream disc harge were used: Landers Wilson (1991) and Log-Pearson using peak discharge. The first st ep in determining stream discharge was to calculate drainage area. To accomplish this ta sk, 10 meter resolution Digital Elevation Model (DEM) imagery were downloaded from Mississi ppi Automated Resource Information System (MARIS). The DEMs were combined to create one mosaic DEM for easier analysis. The DEMs were converted to raster in Ar cGIS toolbox using the DEM to raster tool. The rasterized DEMs were used for the remaining analysis. ArcGIS 9.2 Spatial Analysis Hydrology tools were used to, determine flow direction, fill sinks, calculate flow accumulation, delineate watersheds, and create stream networks. ArcHydro extension was loaded in ArcGIS to delineate sub-basin area according to the predetermined reaches. Discharge points were cr eated where the stream intersects the reach block using the Sub-Basin tool in ArcHydro. The sub-basin area wa s recalculated using Hawths Analysis Tools version 3.27 to insure accuracy in calculation. The slope of the reach was 67

PAGE 68

determined by calculating Rise/Run using the information tool in ArcGIS to determine the elevation value on the DEM for the stream networ k. The elevation value wa s converted from feet to meters for each reach intersection and the slope was determined by the following formula. S = (Eo Er) / Lcl Where S slope, Eo is the elevation at the top of the reach block, Er is the elevation at the end of the reach block, and Lcl is the centerline length from the beginning of the reach block to the end of the reach block. In most stream power calculations the bankfull discharge is used. The bankfull discharge has a reported recurrence interval of 1.5 years in the United States (Leopold et al. 1964). To determine discharge for each reach bloc k outlet, several methods were used to determine the best method. Landers and Wilson in Flood Characteristics of Mississippi Streams (1991) defines Q2 for a drainage area greater than 800 mi2 as Q2 = 131(A) 0.97 (S) 0.21 L -0.47 and for drainage areas less than 800 mi2 for Eastern Mississippi as Q2 = 296 (A) 0.81 (S) 0.03 (L) -0.36 where Q2 is the discharge in Ft3/ s for a two year flood return, S is slope, and L is centerline length. To make these calculations, all calculations were converted to English measurements to ensure compatibility to the Landers and Wilson document. The final calculations were converted to metric meas urements for comparison to other methods. According to the Interagency Advisory Co mmittee on Water Data (1982), the Log-Pearson Type III Distribution is the recommended technique for flood frequency analysis. The LogPearson Type III distribution from the website: http://water.oregonstate.edu/stream flow/analysis/floodfreq/index.htm The website instructions were used as a comparison method to Landers and Wilson (1992). The Log-Pearson Type III distribution is used to fit flood frequency data to predict the design 68

PAGE 69

flood for a river at some site. Historic data was used to calculate bank full discharge for a two year flood return by using instanta neous peak stream flow data fo r the available gauging stations from the USGS website http://water.usgs.gov/nwis/peak For details of how the Log_Pearson Type III distribution was used to calculate two year flood return discharge see Appendix B. Gauging station data used in the analysis (Tab le 3-1) shows the name, discharge area and the number of years of available data of the. Th e discharge area as specified by the USGS gauge information and historical peak discharge data was used to determine the two year flood frequency (Q2) discharge using the Log-Pearson Type III distribution. The flood frequency calculations for Q2 were plotted against drai nage area to determine a trend line and formula (Figures 3-1). Y = 0.1437x + 70.05 Where Y is the Q2 two year flood return discharge and x is drainage area. The Q2 discharges for all seven gauging stations were recalculated using the above formula to determine discharge. Met hods of calculating stream discharge were compared for LandersWilson, Log-Pearson Type III and the trend line of the Log-Pearson Type III analysis (Figure 32). In comparison to the Landers-Wilson method the trend line of Log-Pearson Type III utilized peak stream data captured after 1990 and the Log-Pearson Type III was only calculated at available gauging stations. Since discharge for e ach reach outlet was required the trend line of the Log-Pearson Type III distribution was used to calculate the two year flood return discharge Stream power Stream power was calculated using the trend line formula Q2 = 0.1437x + 70.05 for estimation of discharge with a two year flood re turn, slope, width and th e weight of water. = ( Q2 S) / W 69

PAGE 70

Where is stream power, is (9800 N m 3) the specific weight of clear water at 10 C, where N is force measured in Newton; Q2 is the discharge for a two year flood return at each reach block (m3/second); S is local slope ( Elevation/centerline length) and W width in meters. Stream power was then compared to stream change characteristics (Tab le 3-2) for the Leaf River, Bogue Homo River and Thompson Creek. Slope, width and discharge varies for each defined stream reach. Statistical Analysis The statistical analysis was performed usi ng SigmaPlot 11.1 software (Systat Software Inc., San Jose, CA). The Shapiro-Wilk test for normally was run for the specific stream power and slope data to determine the type of correlatio n to run on the stream ch aracteristic variables. Correlations were done using the Spearman Rank Or der tests because the data failed tests for normality. Results Except for average channel width, which was used to compute specific stream power a cursory review of the graphs did not show a st rong visual correlation be tween stream power and stream planform characteristics on the Leaf Rive r (Figures 3-3 to 3-9) and Thompson Creek (Figures 3-10 to 3-16). Bogue Homo River (Figures 3-17 to 3-23) appeared to act as expected for an undisturbed stream with low specific st ream power variables (Schumm 1985, Magilligan 1992). For the Leaf River (Figure 3-3) stream power was in the range of 75 to 120 indicating active channel change (Brookes 1985). The upper L eaf River from km 14 to km 60 fluctuated around an average of 90 At km 60 to 78 specific stream power increases to 120 and then decreases to 80 at km 80. The middle Leaf River from km 80 to 94 shows a peak in specific stream power of 110 upstream of the confluence of the Bowie River and after the confluence 70

PAGE 71

the stream power goes down to an average of just below 80 At the beginning of the lower reaches of the Leaf River, specifi c stream power again peaks to 110 at km 114 and settles to an average specific stream power of 100 to km 144. When comparing specific stream power to slope of the Leaf River channel stream, the va riation of stream power throughout the basin did not appear to have a visual correlation. The Bogue Homo River (Figure 3-17) had a stream power calculation range from 19 to 13 indicating little work occurring. In areas id entified with increasing point bar areas stream power did not fluctuate. The increase in point bar area possibly indica tes the additional of sediment possibly coming from Tiger Creek. Thompson Creek (Figure 3-10) stream power in creased at the beginning of the analyzed area from 145 to 165 at km 5 to 6. The mined areas of the stream at km 10 to 16 showed a slow increase in stream power from 85 to 105 Stream power increased notability to 140 near the confluence with the Leaf River. When plotting specific stream power against slope through the basin the mined areas with variable slope through the mined areas (km10 to 16) showed a flattening of stream power at 90 even when the slope incr eased slightly through the reach. Stream power was plotted against slope (Fi gures 3-24 to 3-27) for the basin. The upper Leaf River appeared to correspond to an increase in slope with increased specific stream power downstream until the lower section of the reach at km 66 where Lo wrey Creek joins the drainage area. At this confluence stream power appears to increase from 80 to 120 The middle Leaf River slope appeared to flatte n even though the stream power wa s increasing with no increase in drainage area just upstream of the Bowie River confluence. The middle Leaf River section ends 71

PAGE 72

with the addition of East Tallahala Creek at km 114. The lower Leaf River specific stream power assumes a downward trend of stream power as the river widens downstream. The middle Leaf River showed a variation from the upper Leaf River where the slope stayed consistent but stream po wer increased. This same pattern was seen in the lower Leaf River and in Thompson Creek (Figure 24). This ma y indicate slope as not being a factor in the change in stream power but is expected in the lower reaches of a stream due to hydraulic geometry downstream. Bogue Homo River showed a specific stream power level between 13 to 17 resulting in an expected trend of specific stream power lo wering going downstream (Figure 3-26) even with fluctuating widths, slope and discharge. When comparing the Bogue Homo River specific stream power to the Leaf River upper reaches, some of the differences seen are related to scale. The Bogue Homo River is 31 km long while the upper reaches of the Leaf River is over 80 km long. More variation will occur when you have a longer re ach with more small tributaries entering into the system causing the variati on seen in the Leaf River spec ific stream power analysis. When specific stream power was plotted agai nst slope for Thompson Creek varying levels of stream power 85 to 180 were identified for the same slope. In the mined reaches between km 10 to 16 the stream widens and specific stream power lowers dramatically just upstream of the mined reach of 105 to 85 in the mined reaches. Stream power increases to over 140 just downstream of the mined reaches. Statistical Analysis In order to quantify the results a statistical analysis was necessary to create any conclusions on the analyzed rivers. A test for normality was run for the three rivers for slope and specific stream power (Table 3-3) and failed for Leaf River slope and Thompson Creek specific stream power. Normality test indicated Spearman Rank Orde r correlation had to be used in the analysis. 72

PAGE 73

Statistical analysis was run for slope, speci fic stream power, sinuosity, point bar area, change indices and average lateral migration fo r the three rivers analyzed. Spearman Rank Order Correlation was used because the st rength of association needed to be determined and the all data being compared was not normally distributed. The Leaf River upper reaches (km 14 to 80) showed lateral migration, sinuosity and erosion change indices as significant to the 90th percentile when compared to specific stream power. Lateral migration and erosion change in dices showed a moderate negative correlation, while sinuosity showed a moderate positive corr elation. The middle reaches of the Leaf River (km 41 to 114) between and erosion change indi ces showed statistical significance to the 90th percentile when compared to specific stream power. Between change indices showed a relatively strong positive correlation to spec ific stream power and erosion change indices showed a strong positive correlation when compared to specific stream power. The lower Leaf River (km 116 to 158) showed no characteristics signif icant to specific stream power. The Bogue Homo River showed erosion change indices as being si gnificant to the 90 percentile when compared to specific stream power and a moderate positive correlation to specific stream power. Thompson Creek lateral migration and point bar area showed si gnificance to the 90th percentile when compared to spec ific stream power. Both had a strong negative correlation when compared to specific stream power. Comparing stream power to stream change ch aracteristics for mined rivers was visually inconclusive. However for stream s with stream power above 70 to 120 slope is not as evident as on the Leaf River that stream power was a determ ining factor. Thompson Creek has stream power between 80 to 180 but the slope was relatively the same for this range of 73

PAGE 74

stream power varying from 0.001 and 0.0012 mm-1. The same was determined for Thompson Creek (Figures 3-10 to 3-16) with a stream power calculation in the range of 90 to 170 The Leaf River and Thompson Creek have active ch ange occurring as determined by the stream change characteristics. Discussion In the basin analyzed, low power streams with an external source of disturbance did concur with Urban and Rhoads (2003) as being not cond ucive to rapid channel change. Stream with stream power above 100 showed significant changes occurring especially in areas that were undisturbed but showed little la teral migration and sinuosity. Str eams with the highest stream power did not indicate an increase in slope or channel width as the cause of increasing stream power. Mined steams showed a lower level of stream power in the mined reaches due to widening of the channel in the mined reaches. Upstream of the mined reaches did not show significant narrowing but downstream of the mine d reach did show a narrower channel. The stream power threshold between erosion and de position did show significance in areas of the stream that indicated erosion due to la rge mined areas in an adjacent stream. Stream pattern was necessarily not a predictor of stream power Areas with high levels of stream power (above 100 w) did not show signifi cance when compared to sinuosity either in undisturbed or disturbed landscapes The disturbed landscapes did i ndicate a pattern and level of moderate stream power that has a relativ ely low stability factor (Schumm 1985). Utilizing a basin model to evaluate stability may not show the individual characteristics of a human disturbed landscape. A large model may be a starting point but could not be the only method for predicting change that occurs w ithin individual str eams in the basin. 74

PAGE 75

Conclusions Stream power showed to be a useful tool in determining the amount of stability in a stream basin. This comparison is limited by the unpredicta ble characteristics of in-stream mining areas. Further study could include evaluation of width depth ratios and stream power in mined areas. Geology and vegetation, which provide resistance must also be evaluated to determine how stream power is restricted with constraine d areas of stream wi th in-stream mining. In mined rivers stream power in relation to hydraulic geometry downstream is changed because of the human induced changes that occur in the stream basin. Some of the in-stream mining pits can be shallow or very deep causing unpredictable results when calculation specific stream power. Perhaps a better method should have been using channel depth measurements to more accurately calculating the amount of stream power through the mined reaches. 75

PAGE 76

Table 3-1. Peak discharge data at USGS gauging stations Gauging Station River Years Entries Near Collins Leaf 1900-2006 70 Hattiesburg Leaf 1900-2006 103 Near Mclain Leaf 1900-2006 68 Near New Augusta Leaf 1900-2006 29 Near Raleigh Leaf 1900-2006 55 Near Taylorsville Leaf 1900-2006 42 Near Richton Thompson 1998-2006 9 Table 3-2. Stream change characteristics Characteristic Width Point Bar Area Change Indices Sinuosity Lateral Migration Table 3-3. Shapiro-Wilk Test for normality W-Statistic P Result Leaf River Slope reach 1-40 0.913 0.010 Failed Leaf River Slope reach 41-57 0.967 0.763 Passed Leaf River Slope reach 58-79 0.939 0.188 Passed Leaf River Specific Stream power reach 1-40 0.921 0.017 Failed Leaf River Specific Stream power reach 41-57 0.899 0.066 Passed Leaf River Specific Stream power reach 58-79 0.932 0.134 Passed Bogue Homo River Slope 0.966 0.493 Passed Bogue Homo River Specific Stream power 0.966 0.493 Passed Thompson Creek Slope 0.959 0.677 Passed Thompson Creek Specific Stream power 0.847 0.016 Failed 76

PAGE 77

Table 3-4. Leaf River reaches 7-40 Spearman Ra nk Order Correlation in comparison to stream power Lat Mig Sinuosity PB area CI 96-05 B CI 96-05 D CI 96-05 E CI 96-05 U Corr Coeff -0.423 0.418 -0.167 -0.0996 -0.269 -0.441 0.269 P Value 0.0129 0.0141 0.344 0.572 0.122 0.00922 0.122 # Samples 34 34 34 34 34 34 34 Table 3-5. Leaf River reaches 41-57 Spearman Ra nk Order Correlation in comparison to stream power Lat Mig Sinuosity PB area CI 96-05 B CI 96-05 D CI 96-05 E CI 96-05 U Corr Coeff -0.225 0.346 0.221 0.606 -0.152 0.809 0.152 P Value 0.377 0.169 0.387 0.00988 0.553 0.0000002 0.553 # Samples 17 17 17 17 17 17 17 Table 3-6. Leaf River reaches 58-79 Spearman Ra nk Order Correlation in comparison to stream power Lat Mig Sinuosity PB area CI 96-05 B CI 96-05 D CI 96-05 E CI 96-05 U Corr Coeff -0.0457 0.214 -0.351 0.198 0.0096 0.333 -0.0096 P Value 0.836 0.334 0.108 0.372 0.964 0.128 0.964 # Samples 22 22 22 22 22 22 22 Table 3-7. Bogue Homo River Spearman Rank Order Correlation in compar ison to stream power Lat Mig Sinuosity PB area CI 96-05 B CI 96-05 D CI 96-05 E CI 96-05 U Corr Coeff 0.311 -0.0372 -0.135 0.222 -0.0855 0.498 0.0855 P Value 0.113 0.851 0.498 0.262 0.668 0.00841 0.668 # Samples 27 27 27 27 27 27 27 Table 3-8. Thompson Creek Spearman Rank Order Correlation in comparison to stream power Lat mig Sinuosity PB Area CI 96-05 B CI 96-05 D CI 96-05 E CI 96-05 U Corr Coef -0.789 0.429 -0.836 -0.075 -0.411 0.075 0.411 P Value 0.0000002 0.107 0.0000002 0.783 0.124 0.783 0.124 #Samples 15 15 15 15 15 15 15 77

PAGE 78

Table 3-9. Confluence and mining pond list River Intersecting Stream Location (km) Leaf River West Tallahala Creek 20 Leaf River Fisher Creek 34 Leaf River Oakahay Creek 46 Leaf River Big Creek 60 Leaf River Lowrey Creek 66 Leaf River Bowie River 94 Leaf River No Name Creek 100 Leaf River East Tallahala Creek 118 Leaf River Bogue Homo River 124 Leaf River Thompson Creek 136 Leaf River Gaines Creek 144 Bogue Homo River Tiger Creek 8 Bogue Homo River Buck Creek 18 Thompson Creek Mining pond 6 Thompson Creek Mining ponds 10 to 16 Table 3-10. Summary of published relationships between latera l migration rates and other parameters. Adapted from Richard et al. 2005. Source Significant Relationship Additional notes Definitions Hooke (1979) M ~ Q peak, API M = migration rate API = antecedent precipitation index Hooke (1980) M ~ A A = drainage area Nanson and Hickin (1983) M ~ Q and S M ~ W and S Also identified bank texture, planform and sediment supply rate as important S = slope W = width MacDonald (1991) M ~ Q Q = discharge Lawler et al. (1999) M ~ L Also found stream power and bank material to be important L = distance downstream Richard et al. (2005) M ~ W M ~ QS QS = total stream power 78

PAGE 79

Figure 3-1. Two year flood return for USGS gauge stations calculated us ing Log-Pearson Type III distribution vs drainage area measured at gauge stations Figure 3-2. Comparison of disc harge calculations methods 79

PAGE 80

Figure 3-3. Point bar area vs stream power for L eaf River. For notes on arrows see Table 3-9. Middle reaches Lower reaches Upper reaches Middle reaches Lower reaches Upper reaches Figure 3-4. Sinuosity vs stream power for Leaf River. For notes on arrows see Table 3-9. 80

PAGE 81

Figure 3-5. Lateral migration vs st ream power for Leaf River. For notes on arrows see Table 3-9. Middle reaches Lower reaches Middle reaches Lower reaches Upper reaches Upper reaches Figure 3-6. Between change indices vs stream power for Leaf River. For notes on arrows see Table 3-9. 81

PAGE 82

Figure 3-7. Deposition change indi ces Leaf River vs stream po wer. For notes on arrows see Table 3-9. Lower reaches Middle reaches Middle reaches Lower reaches Upper reaches Upper reaches Figure 3-8. Erosion change indices Leaf River vs stream power. For notes on arrows see Table 39. 82

PAGE 83

Lower reaches Middle reaches Upper reaches Figure 3-9. Unchanged change indices vs stream power for Leaf River. For notes on arrows see Table 3-9. Figure 3-10. Point bar area vs stream power fo r Thompson Creek. For notes on arrows see Table 3-9. 83

PAGE 84

Figure 3-11. Sinuosity vs stream power for Thom pson Creek. For notes on arrows see Table 3-9. Figure 3-12. Average lateral migration vs stream power for Thompson Creek. For notes on arrows see Table 3-9. 84

PAGE 85

Figure 3-13. Between change indi ces vs stream power for Thompson Creek. For notes on arrows see Table 3-9. Figure 3-14. Deposition change indices vs st ream power for Thompson Creek. For notes on arrows see Table 3-9. 85

PAGE 86

Figure 3-15. Erosion change indi ces vs stream power for Thompson Creek. For notes on arrows see Table 3-9. Figure 3-16. Unchanged change indices vs stream power for Thompson Creek. For notes on arrows see Table 3-9. 86

PAGE 87

Figure 3-17. Point bar area vs stream power fo r Bogue Homo River. For notes on arrows see Table 3-9. Figure 3-18. Sinuosity vs stream power for Bogue Homo River. For notes on arrows see Table 39. 87

PAGE 88

Figure 3-19. Average lateral migration vs stre am power for Bogue Homo River. For notes on arrows see Table 3-9. Figure 3-20. Between change indices vs stream power for B ogue Homo River. For notes on arrows see Table 3-9. 88

PAGE 89

Figure 3-21. Deposition change indices vs stream power for Bogue Homo River. For notes on arrows see Table 3-9. Figure 3-22. Erosion change i ndices vs stream power for B ogue Homo River. For notes on arrows see Table 3-9. 89

PAGE 90

Figure 3-23. Unchanged change indices vs str eam power for Bogue Homo River. For notes on arrows see Table 3-9. Figure 3-24. Stream slope vs stream power. 90

PAGE 91

Figure 3-25. Leaf River: Slope and stream power by reach. For not es on arrows see Table 3-9. middle reach upper reach lower reach Figure 3-26. Bogue Homo River: Slope and stream power by reach. For notes on arrows see Table 3-9. 91

PAGE 92

Figure 3-27. Thompson Creek: Slop e and stream power by reach. Fo r notes on arrows see Table 3-9. 92

PAGE 93

CHAPTER 4 CONCLUSIONS Stream flow through disturbed rivers is severely impacted due the amount of change occurring in the stream network. Evaluating stabil ity or instability requi res an examination of many factors including a number of planform variables, patte rn and stream power. Stream characteristics such as stream power, sinuosity, lateral migration, point bar area, channel width and change indices are good indicators of the stability of a channel, but other factors such as adjacent disturbed rivers may cause variations of predicted outcomes. A deca de in the lifetime of a river is too small of a timeframe to determine if a river has and become more stable in an environment that is very di sturbed by human impacts. River managers are encouraged to investigate the physical and environmental impacts that in-stream and floodplain mining has in a stream ba sin before allowing direct human impact to occur. 93

PAGE 94

APPENDIX A METHODOLOGY FOR CALCULATING AVERAGE YEARLY LATERAL MIGRATION 1. To calculate lateral migration using ArcGIS 9.2 Spatial Analyst tools follow these steps. 2. Add a field to the T1 and T0 centerline shapef ile named Stream_ID and set the value to 1 for each feature. 3. In the centerline shapefile sele ct by attribute each separate river centerline and create a centerline shapefile for each river. 4. In the ArcGIS Spatial Analys t extension menu, go to options and set the extent under the extent tab, analysis extent set to the river reach polygon shapefile (i.e.: BogueHomoRIDs and press OK. 5. Again under Spatial Analyst select convert, se lect feature to raster. A menu will come up. Input feature is the T1 river centerline shapef ile (i.e.: BH1990sCntline), field is Stream_ID, an output cell size is 2, and then specify an output raster filename (i.e.: BH1990sRas). 6. Under Spatial Analyst select distance, then st raight line and a menu w ill come up. Input the T2 river centerline file (i.e.: BH2005Cntline) for the Distance to field, output cell size as 2 and output raster input Ri verT2Ras (i.e.: BH2005Ras) 7. Under Spatial Analyst select raster calculator an d multiply the two raster files just created for straight line distance and the rasteriz ed T1 or T2 file. (i.e.:BH1990sRas BH2005cntlineRas) 8. The resulting output will be added to the Tabl e of Contents in ArcG IS named Calculation. Right click the Calculation file in the table of contents, select data, make permanent, giving a relevant name the file (i.e.: BHLMcalc). Unde r Spatial Analyst select Zonal Statistics tool to summarize that distance raster using the origin al polygon layer as the zonal layer. Set zone dataset to the reach polygon shapefile (i.e.: BogueHomoRID), the zone field as Reach_ID and the value raster as the output from the cal culation (i.e.: BHLMcalc). This will create a table and graph in ArcGIS. 9. Repeat steps 3-7 for each river. 94

PAGE 95

APPENDIX B CALCULATING INSTENTANIOUS PEAK FLOW DATA USING THE LOG-PEARSON TYPE III DISTRIBUTION In Microsoft Excel perform the following steps: Step 1: Obtain stream flow data Step 2: Organize the information in a table. Step 3: Rank the data from largest discharge to smallest discharge using the "sort" command. Add a column for Rank and number each stream flow value from 1 to n (the total number of values in your dataset). Step 4: Create a column with the log of each max or peak stream flow using the Excel formula {log (Q)} and copy command. Step 5: Calculate the Average Max Q or Peak Q and the Average of the log (Q) Step 6: Create a column with the excel formula {(log Q avg(logQ))^2} Step 7: Create a column with the excel formula {(log Q avg(logQ))^3 Step 8: Create a column with the return period (Tr) for each discharge using the Excel formula {(n+1)/m}. Where n = the number of values in the dataset and m = the rank. Step 9: Complete the table with a final column showing the exceedence probability of each discharge using the excel formula {=1/Retu rn Period or 1/Tr} and the copy command. Step 10: Calculate the Sum fo r the {(logQ avg(logQ))^2} and the {(logQ avg(logQ))^3} columns. Step 11: Calculate the variance standard deviation and skew coefficient as follows: variance = standard deviation = skew coefficient = 95

PAGE 96

Excel functions can also be used to ca lculate the variance (=VAR( ) ), standard deviation (=STDEV( ) ), and skewness coe fficient (=SKEW( ) ). Note that you use these formulas with the data in the log(Q) column. Step 12: Calculat e weighted skewness Step 13: Calculate K values. Use the frequency factor table and the skew coefficient to find the K values for the 2,5,10,25,50,100, and 200 recurrence intervals Step 14: Using the general equation, list th e discharges associated with each recurrence interval general equation = Step 15: Create table of Discharge values found using the log Pearson analysis Source: http://water.oregonstate.edu/streamflow/analysis/floodfreq/tutorial.htm 96

PAGE 97

LIST OF REFERENCES Brookes, A. 1985. River channelization: Traditional engineering methods, physical consequences, and alternative practices. Progress in Physical Geography 9: 44-73. Brookes, A. 1988. Channelized Rivers: Perspe ctives for environmental management. Chicherster, UK: Wiley & Sons. Brown, A., Lyttle, M., and Brown, K. 1998. Impacts of gravel mining on gravel bed streams. Transactions of the American Fisheries Society 127: 979-994. Burkhead N. and Jelks, H. 2001. Effects of Suspended Sediment on the Reproductive Success of the Tricolor Shiner, a Crevic e-Spawning Minnow. Transactions of the American Fisheries Society 130: 959-968. Chandler, J., Ashmore, P., Paola, C., Gooch, M. and Varkaris, F. 2002. Monitoring river-channel change using terrestrial oblique digital im agery and automated digital photogrammetry. Annals of the Association of Am erican Geographers 92(4): 631-644. Dynesius, M, and C. Nilsson. 1994. Fragmentation and flow regulations of river systems in the northern third of the wo rld. Science 266: 753-762. Gilvear, D., Davids, C. and Tyler, A. 2004. The use of remotely sensed data to detect channel hydromorphology; River Tummel, Scotland. River Research and Applications 20: 795811. Gilvear, D., Winterbottom, S. and Sichingabula, H. 2000. Character of channel planform change and meander development: Luangwa River, Zambia. Earth Surface Processes and Landforms 24: 421-436. Graf, W. 1979. Mining and channel response. Annals of the Associ ation of American Geographer, 69(2): 262-275. Graf, W. 2001. Damage control: rest oring the physical integrity of Americas rivers. Annals of the Association of American Geographers 91(2): 1-27. Gregory, K. J. 2006. The human role in cha nging river channels. Ge omorphology 79: 172-191. Gurnell, A. 1997. Channel change on the River Dee meanders, 1946-1992, from the analysis of air photos. Regulated Rivers: Re search & Management 13: 13-26. Howe, T. 2001 Growth of the Lumber Industr y (1840-1930): Mississippi Hi story Now, an online publication of the Mississippi Hist orical Society, Jackson, MS. http://mshistory.k12.ms.us/index.php?id=171 Interagency Advisory Committee on Water Data 1982. Guidelines for Determining Flood Flow Frequency. Bulletin #17B of the Hydrology Subcommittee. Reston, VA: US Department of Interior Geological Survey O ffice of Water Data Coordination. 97

PAGE 98

James A. L., and Marcus, W. A. 2006. The human role in changing fluvial systems: Retrospect, inventory and prospect. Geomorphology 79: 152-171. Knighton, A. D. 1999. The gravel-sand transition in a disturbed catchment. Geomorphology 27: 325-341. Knighton, A. D. 1998. Fluvial Forms and Pro cesses: A New Perspe ctive. London: Arnold. Knighton, A. D. 1999. Downstream variation in stream power. Ge omorphology 29: 293-306. Knox, J. 1987. Historical valley floor sedimentation in the upper Mississippi Valley. Annals of the Association of American Geographers 77(2): 224-244. Kondolf, G.M. 1993. The reclamation concept in regulation of gravel mining in California, Journal of Environmental Pla nning and Management 36(3): 395-406. Kondolf, G.M. 1997. Hungry Water: Effects of dams and gravel mining on river channels. Environmental Management 21(4): 533-551. Landers, M.N. and Wilson, K.V., Jr. 1991. Flood char acteristics of Missi ssippi streams. U.S. Geological Survey Water-Resources Investigations Report 91-4037, 82 p. Lawler, D. M. 1993. The measurement of river bank erosion and lateral channel change: a review. Earth Surface Processes and Landforms 18: 777-821. Lee, H., Fu, D. and Song, M. 1993. Migration of rectangular mining pit composed of uniform sediments. Journal of Hydrau lic Engineering 119(1): 64-80. Leopold, L. B., Wolman, M. G., and Miller, J. P., 1964, Fluvial processes in geomorphology. San Francisco, CA: W. H. Freeman and Co. Magilligan, F. J. 1992. Thresholds and the spa tial variability of fl ood power during extreme events. Geomorphology 5: 373-390. United States Department of Agriculture. 2005. 2005 National Agriculture Imagery Program. MARIS Mississippi Automated Resource Inform ation System Technical Center. Jackson, MS. http://www.maris.state.ms.us. Accessed March 6, 2007. Mark, D. 1983. Relations between field-surveyed channel networks and map based geomorphic measures, Inez, Kentucky. Annals of the A ssociation of American Geographers 73(3): 358372. Marston, R., Bravard, J.-P. and Green T. 2003. Imp acts of reforestation and gravel mining on the Malnant River, Haute-Savoie, French Alps. Geomorphology 55: 65-74. Meador, M and Layher, A.1998. In-stream sand and gravel mining: environmental issues and regulatory process in the United States. Fisheries 23(11): 6-12. Mississippi State Legislature. 1995. Miss. Code Ann. 53-711 section 400.C. Jackson, MS. 98

PAGE 99

Mossa, J. 2006. Quantifying channel planform change. Papers of the Applied Geography Conference 29: 65-70. Mossa, J., Coley, D. and Rasmussen, J. 2007. Ge omorphic assessment of channel changes, Pascagoula River and tributaries, Mississippi. Report submitted to the U.S. Army Corps of Engineers, Pat Harrison Waterway Distri ct, and Mississippi Nature Conservancy. Gainesville, FL: Department of Ge ography, University of Florida. Mossa, J. and McLean, M.B. 1997. Channel planform and land cover changes on a mined river floodplain: Amite,River, Louisiana, USA. Applied Geography 17(1): 43-54. Nanson, G. and Hickin, E. 1986. A st atistical analysis of bank er osion and channel migration in western Canada. Geological Society of America Bulletin 97: 497-504. Nanson, G. and Croke J. 1992. A genetic classi fication of floodplains. Geomorphology 4: 459496. Oetter, D., Ashkenas, L., Gregory, S. and Mi near, P. 2004. GIS methodology for characterizing historical conditions of the Willamette Rive r flood plain, Oregon. Transactions in GIS 8(3): 367-383. Petit, F., Poinsart, D. and Bravard J.-P. 1996. Channel incision, gravel mining and bedload transport in the Rhone River upstream of L yon, France: canal de Mi ribel. Catena 26: 209226. Platts, W., Torquemada, R., McHenry, M. and Graham, C. 1989. Changes in Salmon Spawning and Rearing Habitat from Increased Delivery of Fine Sediment to the South Fork Salmon River, Idaho. Transactions of the American Fisheries Society 118: 274-283. Reinfelds, I., Cohen, T., Batten, P. and Brierl ey, G. 2009. Assessment of downstream trends in channel gradient, total and specific str eam power: a GIS approach. Geomorphology 60: 403-416. Rhodes, B.L. 1987. Stream Power Terminol ogy. Professional Geographer 39(1): 189-195. Richard, G. 2005. Statistical anal ysis of lateral migration of the Rio Grande, New Mexico. Geomorphology 71: 139-155. Rinaldi, M. and Simon, A. 1998. Bed-level adju stments in the Arno River, central Italy. Geomorphology 22: 57-71. Rinaldi, M., Wyzga, B. and Surian, N. 2005. Sedi ment mining in alluvi al channels: Physical effects and management perspectives. Rive r Research and Applications 21: 805-828. Rosson, J. 2001. Forest resources of Mississippi, 1994. United State Depart ment of Agriculture, Forest Service Resource Bulletin SRS-61. Ashvi lle, NC: Southern Research Station. Schumm, S. A. 1977. The Fl uvial System. Caldwell, NJ: The Blackburn Press. 99

PAGE 100

100 Schumm, S. A. 1985. Patterns of al luvial rivers. Annual Reviews Earth Planet Science 13: 5-27. Simon, A. and Downs, P. 1995. An interdiscip linary approach to evaluation of potential instability in alluvial ch annels. Geomorphology 12: 215-232. Simon, A. and Rinaldi, M. 2006. Disturbance, str eam incision, and channe l evolution: the roles of excess transport capacity and boundary materials in co ntrolling channel response. Geomorphology 79: 361-383. Urban, M. A. and Rhoads, B. L. 2003. Catastro phic human-indices change in stream-channel planform and geometry in an agricultural watershed, Illinois, USA. Annals of the Association of American Geographers 93(4): 783-796. Van Looy, J. and Martin C. 2005. Channel and vegetation change on the Cimarron River, southwestern Kansas, 1953-2001. Annals of th e Association of American Geographers 95(4): 727-739. Wellmeyer, J., Slattery, M. and Phillips, J. 2005. Quantifying dow nstream impacts of impoundment on flow regime and channel pl anform, lower Trinity River, Texas. Geomorphology 69: 1-13. Winterbottom, S. 2000. Medium and short-term ch annel planform changes on the Rivers Tay and Tummel, Scotland. Geomorphology 34: 195-208. Winterbottom, S. and Gilvear, D. 1997. Quantifi cation of channel bed morphology in gravel-bed rivers using airborne multispectral imagery and aerial photography. Regulated Rivers: Research & Management 13: 489-499.

PAGE 101

BIOGRAPHICAL SKETCH Ursula Garfield was born in Neptune, New Je rsey, to Martha and Udo Anckarstrom-Bohm, immigrants from Germany. She has two brothers and three sisters and lived in Spring Lake, New Jersey until 1976 when her family moved to Lake Worth, Florida. Ursula completed high school at Lake Worth High School and went on to college at Broward Community College. After working for 20 years in the information technology field, she returned to college to finish her Associate of Arts degree at Santa Fe Co mmunity College (Gainesville, Florida) in December 2002. Ursula entered the University of Fl orida as a transfer student in the Department of Geography where she finished her Bachelor of Arts in summer 2005. Ursula entered the masters degree program in the Department of Geography in fall 2005, and is planning to continue as a Ph.D. student. Ursula is married to Albert Garfield and has two children: Stephen Albert Garfield and Andrew Johann Garfield. She curre ntly lives in rural Marion County, Florida, just outside of Flemington on a 21-one acre farm. 101