Battle Creek River Watershed Sediment Budget Study › wp-content › uploads › ... · A number of stream and watershed conservation efforts have been undertaken to minimize the

U.S. Army Corps of Engineers Detroit District

Battle Creek River Watershed Sediment Budget Study

March 2008

Battle Creek River Watershed Sediment Budget Final Report

Prepared for U.S. Army Corps of Engineers Detroit District Prepared by NTH Consultants, Ltd. and

Baird W.F. Baird & Associates Ltd.

For further information please contact

Travis Dahl at (313) 226-3398

Cover photograph courtesy of Willard Library: Verona Dam near Battle Creek, MI, March 1908

TABLE OF CONTENTS UNIT CONVENTIONS ..................................................................................................................... 1 Executive Summary ........................................................................................................................... 2 1.0 Introduction ................................................................................................................................. 3

1.1 Objectives ................................................................................................................................. 3 1.2 Battle Creek River Watershed ............................................................................................... 4

2.0 Methods ........................................................................................................................................ 9 2.1 Literature Review .................................................................................................................. 10 2.2 Regional Sedimentation Study ............................................................................................ 11

2.2.1 Sediment Yield ............................................................................................................... 11 2.2.2 Soil Erosion ..................................................................................................................... 14 2.2.3 Fluvial Processes ............................................................................................................ 15 2.2.4 Mass Wasting .................................................................................................................. 15

2.3 Watershed Hydrology and Sediment Yield Modeling .................................................... 16 2.3.1 Hydrologic Data ............................................................................................................. 16 2.3.2 Spatial Data ..................................................................................................................... 18 2.3.3 SWAT Model Creation .................................................................................................. 28

3.0 Battle Creek River Sediment Budget Results ................................................................... 30 3.1 Literature Review .................................................................................................................. 30 3.2 Regional Sedimentation Study ............................................................................................ 31

3.2.1 Sediment Yield ............................................................................................................... 31 3.2.2 Watershed Soil Erosion ................................................................................................. 33 3.2.3 Fluvial Processes ............................................................................................................ 35

3.3 Watershed Hydrology and Sediment Yield Modeling .................................................... 38 3.3.1 Model Calibration Results ............................................................................................ 38 3.3.2 Model Results ................................................................................................................. 42

4.0 Conclusions and Recommendations ................................................................................. 47 Cited Literature ................................................................................................................................ 49 Tables ................................................................................................................................................. 53

ACRONYMS AVSWAT Arc View Soil and Water Assessment Took interface BCRW Battle Creek River Watershed BMP Best Management Practice CMI Clean Michigan Initiative FSA Farm Service Agency GIS Geographic Information System MDEQ Michigan Department of Environmental Quality MiGDL Michigan Geographic Data Library MRLC Multi-Resolution Land Characteristics Consortium NHD National Hydrography Database NID National Inventory of Dams NLCD National Land Cover Dataset NRCS Natural Resources Conservation Service RUSLE Revised Universal Soil Loss Equation SI Metric Units (Système international d'unités) SIAM Sediment Impact Assessment Model SSURGO Soil Survey Geographic Database STATSGO State Soil Geographic Database SWAT Soil and Water Assessment Tool TIGER Topologically Integrated Geographic Encoding and Referencing TMDL Total Maximum Daily Load USACE United States Army Corps of Engineers USGS United States Geological Survey USLE Universal Soil Loss Equation WRDA Water Resources Development Act

FIGURES Figure 1: Location map of Battle Creek Watershed ...................................................................... 6 Figure 2: 2001 NLCD Landuse classifications ................................................................................ 7 Figure 3: Previous sediment budget results for Great Lakes Watersheds from past 516(e)

studies. ....................................................................................................................................... 13 Figure 4: Weather stations and climatic regions for the Battle Creek River Watershed........ 20 Figure 5: Quaternary geology ........................................................................................................ 21 Figure 6: Elevation data ................................................................................................................. 22 Figure 7: SSURGO soils ................................................................................................................... 23 Figure 8: STATSGO soils ................................................................................................................. 24 Figure 9: Hydrography within Battle Creek Watershed ............................................................ 25 Figure 10: Locations of Dams in the Battle Creek River Watershed from the National

Inventory of Dams and MI Dept. of Natural Resources. ................................................... 26 Figure 11: 2006 FSA Compliance Imagery ................................................................................... 29 Figure 12: Sediment BMPs as listed in the "Battle Creek River Watershed Management

Plan" ........................................................................................................................................... 34 Figure 13: Engineering of channels (Boley-Morse, 2004) ........................................................... 36 Figure 14: Engineering of channels (2) (Boley-Morse, 2004) ...................................................... 36 Figure 15: CMI erosion site before stabilization (Source: www.kalamazooriver.net) ........... 37 Figure 16: CMI erosion site after bank stabilization (Source: www.kalamazooriver.net) .... 37 Figure 17: SWAT calibration comparison .................................................................................... 39 Figure 18: SWAT calibrated flow data versus observed data from the USGS discharge gage

at Battle Creek ........................................................................................................................... 40 Figure 19: SWAT output. Total water yield and sediment yield for entire basin ................ 41 Figure 20: Average annual channel deposition or erosion per mile of channel. .................... 44 Figure 21: Average annual sediment yield by sub-basin. ......................................................... 45 Figure 22: Sediment budget from SWAT model results by source .......................................... 46 TABLES Table 1: Land type distribution within the Battle Creek River Watershed .............................. 8 Table 2: Discharge Data for the Battle Creek River Watershed ................................................ 17 Table 3: Sediment Yield Estimates for the Battle Creek River Watershed ............................... 31 Table 4: Average annual channel erosion or deposition estimation by sub-basin ................. 42 Table 5: Average annual sediment load estimates from watershed sources ......................... 43 Table 6: Sediment BMPs listed for BC Watershed ...................................................................... 54

B a t t l e C r e e k R i v e r S e d i m e n t S t u d y P a g e 1 F i n a l R e p o r t 1 1 2 0 3 . 0 1 0

UNIT CONVENTIONS

This report relied upon a wide variety of data from numerous sources. Most published values of

scientific data are reported in metric units according to the International System of Units (SI -

Système international d'unités). However, many values from agency reports and engineering

documents use the U.S. Customary System of units. Throughout this report, the results of

computations are reported in SI units to adhere to standard convention. The following abbreviations

and conversions can be used to compare data:

• Ton (T) – U.S. customary short ton = 2,000 pounds = 907.2 kg.

• ton (t) – metric ton = 1.1 Ton.

• Acre (ac) = 43,560 ft2 = 4,046 m2 = 0.046 Hectare (ha).

• Cubic meters per second (cms) = 35.32 cubic feet per second (cfs).

• Acre*feet (Acft) = 1,233 m3.

• gram/cubic centimeter (g/cc) = 1 part per million (ppm) = 1 t/m3.

• 1 Ton per square mile per year = 0.00156 T/ac/yr = 0.350 t/km2/yr = 0.00350 t/ha/yr


EXECUTIVE SUMMARY

The Battle Creek River is a large tributary to the Kalamazoo River in southwestern lower Michigan. This primarily agricultural watershed has been implicated as the largest contributor of sediment to the Kalamazoo River. Potential sources of sediment include bank erosion, agricultural runoff, road crossings, construction runoff, ditch maintenance, dams, bed erosion, and urban runoff. A Clean Water Act Section (319) grant was recently awarded to local agencies to foster the development and utilization of soil conservation and water quality improvement best management practices for the Battle Creek River Watershed, 241 mi2 (BCRW – Figure 1). The potential impacts of these activities on soil erosion and sedimentation could be significant. Under Section 516(e) of the Water Resources Development Act, the U.S. Army Corps of Engineers Detroit District was authorized to conduct a sediment budget analysis for the BCRW. This report presents the results of this sediment budget and potential impacts of the Section 319 efforts on sediment reductions. A variety of regional sediment yield tools were applied to provide estimates of the expected range of total sediment yield from the BCRW, These ranged from 0.04 to 5.3 t/ha/yr with the most consistent estimates ranging from approximately 0.4 to 0.8 t/ha/yr. A watershed hydrology and sedimentation model was built using the Soil and Water Assessment Tool (SWAT) to provide a quantitative and physically based estimate of total sediment yield by sources for the BCRW. Sources included non-point source runoff from soil erosion and erosion of in channel sediments by fluvial scour. SWAT estimated total sediment yield of 0.84 t/ha/yr, 68,500 tons from soil erosion and 20 tons from channel processes. While sediment contributions from bank erosion may be quite high at a local level, the fate of these sediments was only estimated with SWAT. For example, SWAT predicted a number of reaches that would be likely sinks of sediment eroded from upstream channels. This would be consistent with many portions of the drainage network that require regular dredging to maintain conveyance capacity in the agricultural ditches. Unfortunately, the volume of material dredged from the channels is an unknown component of the sediment budget. The dredged channel sediments are often deposited as “berms” along the drainage ditches and may be reintroduced to the drainage system over time through bank erosion. Bank erosion has been suggested as being the largest portion of the sediment budget for the Battle Creek River, but there has not been enough data collected to properly estimate the contribution of sediment from bank erosion. These sources may represent large components of sediment budgets at the local and watershed scale. Future efforts to manage and reduce sedimentation in the BCRW should emphasize quantifying these contributions.


1.0 INTRODUCTION

The Battle Creek River Sediment Budget Study builds on previous studies performed under

authority of Section 516(e) of the Water Resources Development Act (WRDA) 1996. Section

516(e) directs the U.S. Army Corps of Engineers (USACE) to develop sediment transport models for

all major Great Lakes tributaries contributing sediment to Federal navigation projects or Areas of

Concern (AOCs). The goal of this project was to develop a sediment budget for the Battle Creek

River Watershed (BCRW) that may be used by local and regional stakeholders and regulatory

authorities to facilitate the identification, prioritization, and implementation of sedimentation

reduction strategies.

The Battle Creek River is a tributary of the Kalamazoo River in the agricultural lands of

southwestern lower Michigan (Figure 1). Consequently, the drainage system is made up of a mix of

natural rivers, tile drains, and ditches. Upstream of Bellevue in Eaton County, much of the river and

its tributaries have been channelized. Many of the typical issues that affect creeks and rivers in

agricultural lands have contributed to historic sedimentation issues in Battle Creek including; soil

erosion and runoff, stream bank and channel instability, and loss of riparian corridor (MDEQ, 2004).

Using available data for these processes from local, regional, state, and federal stakeholders, a

sediment budget was developed for the BCRW.

1.1 Objectives

The goal of this project was to develop a comprehensive sediment budget for the BCRW using

existing data. The following objectives were established to meet this goal:

1. Gather and review all potential sources of sedimentation information;

2. Process, quality check, and synthesize these to develop core sediment budget components;

• Soil erosion

• Stream bank erosion

• Channel erosion and deposition

• Total BCRW sediment yield

3. Develop total watershed sediment budget.


1.2 Battle Creek River Watershed

The Battle Creek River is a tributary to the Kalamazoo River in southwestern Michigan. The

drainage area is approximately 241 square miles where the Battle Creek River joins the Kalamazoo

in the city of Battle Creek. The majority of the land (52%) within the watershed is used for

agriculture while the rest is forested, wetlands, and developed (Figure 2). Table 1 shows a complete

breakdown of land uses within the study watershed. Populated areas within the watershed include

Charlotte, Olivet, Bellevue, and the city of Battle Creek.

Excessive sediment loads and sedimentation in the Battle Creek River and its tributaries have been

implicated with the degradation of water quality and aquatic habitat here as well as downstream in

the Kalamazoo River (Calhoun Conservation District, 2004). Issues of concern are typical for

agricultural watersheds in this region and include:

• Increased runoff from:

o Agricultural land use conversion

o Tile drainage

• Increased magnitude and frequency of discharge events from:

o Increased runoff

o Straightening natural channel

o Drainage ditch construction

• Increased sediment yield from:

o Soil erosion from agricultural lands

o Channel incision

o Stream bank instability

o Ditch disturbance from maintenance activities

A number of stream and watershed conservation efforts have been undertaken to minimize the

impacts of these disturbances on the hydrology and stability of the river system. Soil conservation

and agricultural best management practices have been implemented on a number of farms under

local, state, and federal conservation authorities. Riparian restoration and buffer projects have been

developed in conservation corridors to reduce sediment yield from fields and enhance stream bank

stability. Stream bank stabilization projects have been undertaken to cut back eroding stream banks,


remove stream bank berms and dredge material, reduce entrenchment of streams, and improve

stream bank stability and habitat. Perhaps the most visible restoration action was the removal of the

Elm Street Dam on the Battle Creek River in the city of Battle Creek in August of 2005. This dam

was built in the early 1900’s to supply cooling water for a coal fired power plant. Removal of the

dam included comprehensive analyses of river hydraulics, sediment transport capacity for alternative

scenarios, and natural channel design for the impacted sections of river (MDEQ, 2005).


Figure 1: Location map of Battle Creek Watershed


Figure 2: 2001 NLCD Landuse classifications


Table 1: Land type distribution within the Battle Creek River Watershed

NLCD 2001 Classification % of Total Area Open Water 1.2%Developed, Open Space 5.5%Developed, Low Intensity 3.3%Developed, Medium Intensity 0.7%Developed, High Intensity 0.4%Barren Land 0.4%Deciduous Forest 16.5%Evergreen Forest 0.5%Mixed Forest 0.4%Scrub/Shrub 0.2%Grasslands/Herbaceous 1.5%Pasture/Hay 16.7%Cultivated Crops 35.5%Woody Wetlands 17.0%Emergent Herbaceous Wetland 0.3%


2.0 METHODS

A combination of literature review, regional sedimentation analyses, and watershed hydrology

and sedimentation modeling were used to develop a detailed sediment budget for the BCRW.

The literature review is described in section 2.1 below.

Results of the literature review, including published soil erosion, sediment yield, and watershed

and basin denudation studies from watersheds in similar physiographic regimes, were used to

determine the expected range of values for sediment budget components for the BCRW.

Geographical scaling of these results and sediment budget analyses were used to develop

sediment budget approximations for the BCRW, section 2.2.

A typical mass-balance approach was originally planned to develop the final sediment budget for

BCRW. Following the comprehensive review of existing discharge, suspended sediment, river

survey, and fluvial data, it was concluded these data could not be used to develop a sediment

budget. Consequently, a more advanced sediment modeling approach requiring the development

and calibration of a watershed hydrology and sedimentation model was developed. This is

described in section 2.3.

B a t t l e C r e e k R i v e r S e d i m e n t S t u d y P a g e 1 0 F i n a l R e p o r t 1 1 2 0 3 . 0 1 0

2.1 Literature Review

An extensive literature review was conducted to gather all information related to sediment issues

within the BCRW. Potential sources of information included geospatial data, scientific literature

and publications, agency reports, and local media sources. The literature review also included

telephone and email interviews with relevant local, state, and federal agency officials to gather

unpublished reports and data including stream bank erosion assessments, conservation reserve

program (CRP) enrollment and performance statistics, and hydraulic models developed for

floodplain mapping and hydraulic structure analyses (e.g. bridge and dam removal studies). A

list of reviewed literature can be found at the end of this report.


2.2 Regional Sedimentation Study

The term sediment budget is widely used and often, with various meanings. One of the most

important steps in determining a sediment budget is defining what one actually is. While

numerous publications and authors have reported on this, Lane, et. al. (1997) provide one of the

more thorough reviews and discussions available and emphasized the importance of identifying

source, transport, and sink areas across spatial scales. All of these areas exist at essentially every

scale, from plot (10-6 km2) to regional and continental scale basins (107 km2). Quite simply, a

watershed sediment budget represents an accounting of all sediment sources and sinks. While

sediment transport processes redistribute sediments and may be represented in sediment budgets,

they need not be. Indeed, sediment transport processes and sediment sources and sinks may be

interchanged depending upon the time scale of concern. For example, flood plain sediments

represent transient storage in a long-term average annual sediment budget, transport in geologic

time, and may locally be a dynamic source and or sink on an event basis. For the BCRW

Sediment Budget, sources include any potential sediment sources upstream of the BCRW outlet

at the U.S. Geological Survey (USGS) discharge monitoring station in Battle Creek, Michigan

(Latitude 42°19'53", Longitude 85°09'13", Gauge id 04050003). Sediment yield and the

sediment budget represent total sediment discharge past this point.

2.2.1 Sediment Yield

The most important part of a sediment budget is the total sediment yield from a watershed

(sediment budget end member). This value represents the maximum amount of sediment

exported from the watershed and is the net sum of all sediment sources; it is typically represented

as a long-term average annual yield (mass/time) or sediment discharge value (mass). The four

primary methods to estimate total yield are denudation rate computations, regional sediment

yield relationships developed from gauge data, depositional sediment surveys in receiving water

bodies, and mechanistic sediment yield modeling.

2.2.1.1 Denudation Rate

The term denudation rate refers to the spatially averaged rate of erosion across a basin.

Denudation rates are typically generated for specific physiographic settings having similar

geologic, climatic, and land use conditions. While the process of basin denudation includes both


dissolved and particulate loads, denudation rates are often reported just for particulate loads.

Denudation represents the simple difference between terrain loss (erosion and solution) and

terrain building (uplift, windblown soil, etc.). While terrain building often exceeds erosion

(which is the case in much of the Great Lakes physiographic region), there is still denudation.

The broad scale estimate of denudation for Great Lakes lowlands with cold winters as provided

by Leopold, et. al. was 2.9 cm per 1,000 years (1995; Corbel, 1959).

2.2.1.2 Regional Sediment Yield Data

A wholly separate estimate for the same physiographic region listed minimum, average, and

maximum values as 10, 100, and 800-tons/square mile/yr, respectively (0.035, 0.35, and 2.8

t/ha/yr) (U.S. Water Resources Council, 1968).

Previous sediment budgets created for similar Great Lakes Watersheds show a strong correlation

between watershed area and sediment yield. These results represent sediment yields estimated

by a variety of physically based, calibrated watershed sedimentation models 516 (e) studies

conducted over the past ten years and can be seen in Figure 3. These results were used to

develop the regional relationship and estimate sediment yield for the BCRW.

A recent study on dam removal dynamics within the Plainwell-Otsego impoundment

downstream of Battle Creek was conducted to determine the sediment loading into the

impoundments in that reach of the river by Wells, Langendoen, and Simon (2007). This study

looked at sediment deposition within the impoundments between Otsego and Plainwell. Their

study concluded that sediment load rate into the Plainwell impoundment, the farthest upstream,

was approximately 71 T/day (Wells, Langendoen, & Simon, 2007).

A variety of other studies focusing on this region were used to estimate the sediment load from

the BCRW. These studies used many different methods for calculating sediment loads and were

applied to this watershed for comparison purposes.


Figure 3: Previous sediment budget results for Great Lakes Watersheds from past 516(e) studies.

2.2.1.3 Sediment Survey

Sediment contributions from upstream sources may be estimated using bathymetric surveys and

sediment depth estimations. Since there are a few impoundments within the BCRW, this was

initially considered as a method for determining the sediment budget for the watershed.

Following conversation with local agencies, it was determined that there had not been

comprehensive bathymetric and sediment surveys of the reservoirs within the BCRW for use of

this method.

2.2.1.3 Regional Sediment Yield Modeling

The watershed sediment yield model of Lane (2001) is based upon a transport limited

assumption, that is, sediment sources are readily available within the watershed and total

sediment yield is controlled by available precipitation and runoff to erode and transport

sediments to, through, and out of the watershed (Lane, 2001). This method was initially


considered for analysis of the watershed, but following data review, it was determined that there

was not enough sediment data within the Battle Creek River to successfully use this method.

2.2.2 Soil Erosion

The use of simple soil erosion models (e.g. USLE, RUSLE) often results in the over prediction

of sediment yield values because sediment delivery is either not considered or not conducted at

the proper scale to reflect erosion and sediment transport processes (Riedel, et. al., 2005a; Riedel

and Vose, 2002; Bolstad, et. al., 2005; Lane, 2001; Trimble and Pierre, 2000; Nearing, 1998).

These simple models were developed for plot level studies and do not represent landscape to

watershed scale sedimentation processes.

Soil erosion and sedimentation can have a large impact on the environment and economy within

the Great Lakes region. These include loss and degradation of farming land, loss of aquatic

habitats, reduction in water quality, and filling in of impoundments and navigation channels.

The Great Lakes Tributary Modeling Program through the U.S. Army Corps of Engineers (2005)

and the Great Lakes Basin program for soil erosion and sediment control (Great Lakes

Commission, 2007) are two programs being implemented throughout Great Lakes states to

reduce sediment loads from soil erosion in Great Lakes tributaries.

Within the BCRW Management Plan drafted under a Clean Water Act 319 grant, a list of critical

sediment source areas within the watershed was created. Areas where Best Management

Practices (BMPs) could be implemented were suggested along with their estimated pollutant

reduction potential. Many of these critical areas listed streambank erosion as contributing large

amounts of sediment to the river. A location map of all the sediment BMPs noted can be seen in

Figure 12. Table 6, adapted from the Watershed Management Plan, lists all BMPs where the

cause of water quality degradation was excessive sedimentation.


2.2.3 Fluvial Processes

2.2.2.1 Bank Erosion

It has been noted that bank erosion is the largest contributor of sediment from the BCRW

(Calhoun Conservation District, 2004). Many of the waterways in the upper portion of the

watershed are established county drains and have been straightened, widened, and are regularly

dredged to support agricultural practices (Figure 13, Figure 14). These channels often become

entrenched and disconnected from the floodplain, causing further erosion during the flooding

season. The practice of disposing of dredge materials along the channel banks creates berms of

sediment that exacerbate the issue of channel and bank instability by partially or wholly

confining high frequency, low magnitude flood events within the channels. This concentrates

flood flows within channels and prevents the spreading and dissipation of flood energy across

broad floodplains.

Bank erosion varies widely across the watershed with differences in watershed, river and valley

characteristics. Some areas of the river run through riparian forested or wetland corridors,

mostly downstream of Bellevue. Other areas have had channel alterations but are still within the

active riparian corridor. Some of the channels have been completely altered and have

agricultural fields right up to their edge. Several different sites within the Kalamazoo Watershed

have been monitored to estimate sediment contribution due to stream bank erosion through the

Clean Michigan Initiative (CMI) Grant Project (www.kalamazooriver.net).

2.2.4 Mass Wasting

Mass wasting is a general term that encapsulates a variety of forms of large bluff, hill slope, soil,

or other form of earth movement/displacement. Typical forms of mass wasting for the Great

Lakes region include slumping of hillslopes (small landslides), bluff failure, and cliff failure

(Riedel, et. al., 2005b). Mass wasting was not mentioned within the watershed management plan

(Calhoun Conservation District, 2004) and is not a significant sedimentation process in the

BCRW. Mass wasting is typical of high-relief areas and therefore is not an issue in this low-

relief agricultural area.


2.3 Watershed Hydrology and Sediment Yield Modeling

A distributed watershed hydrology and sedimentation modeling approach was developed to

estimate sources of sediment in the BCRW (eroding from land and entering the drains and river).

Following a model review process, the SWAT (Soil and Water Assessment Tool) model was

determined to be the most appropriate tool for this region because it implicitly accounts for all of

the agricultural practices in this region including agricultural cropping practices, soil

conservation practices, artificial watershed drainage by tile drains and ditches, and degradation

of channel beds. SWAT is a physically based model that was developed to aid in prediction of

the impacts of climate change, reservoir management, land use change, and watershed

management practices on water, sediment, and chemical dynamics in complex watershed

systems. Much of the data needed to create a SWAT model is available free to the public; a

model can easily be developed using available climate, terrain, soils, and land use data.

The Sediment Impact Assessment Model (SIAM) was also considered as a potential tool to

estimate sediment sources and fluxes from fluvial processes (Mooney; 2006; Mooney, et. al.,

2001). However, the specific data needs of SIAM requiring channel metrics, discharge and

hydraulic measurements, and sediment distributions for the fluvial network in the BCRW

precluded it as these data did not exist with sufficient geographic coverage.

2.3.1 Hydrologic Data

Hydrologic data were obtained to support the development of a hydrologic budget and SWAT

model development for the BCRW. Primary data sources were climatic and hydrologic

observations.

2.3.1.1 Climatic Data

Observed, daily climatic data for the period 1998 to 2007 were obtained from the National

Weather Service’s National Climatic Data Center (NCDC) at www.ncdc.noaa.gov. This period

of record was chosen because continuous precipitation and temperature data needed for model

calibration were available from the nearby weather stations at the Battle Creek and Lansing

airports. The spatial distribution of these gauges helps represent the spatial variability of rainfall

and temperature across the watershed. These stations were also chosen because they were within


the same climatic region as the BCRW, which is located in the far southern portion of the DFb

climate (snow, fully humid, warm summer), as shown in Figure 4.

Climatic data for hydrology and sedimentation forecasts were obtained from the GIS database

included within SWAT. This database includes locations and climatic statistics for over 11,000

National Weather Service (NWS) Cooperative Observer stations that can be used to generate the

SWAT model data requirements including rainfall, temperature, solar radiation, wind speed and

relative humidity.

2.3.1.2 Discharge Data

Discharge data for calibration were available for USGS gauging stations in the BCRW. These

records contain daily average discharge data for the corresponding period of record of observed

climatic data. Table 2 lists the stations used for development of the SWAT model and their

corresponding period of record.

Table 2: Discharge Data for the Battle Creek River Watershed

HUC Site Number Site Name Area

(mi2) Dec. Lat. Dec. Lon. Begin Date End Date Records

04050003 04105000 Battle Creek at Battle

Creek, MI 241 41.33139 85.15361 10/30/1930 11/5/2007 27533

04050003 04104945 Wanadoga Creek

Near Battle Creek, MI 48.3 42.39639 85.13167 10/1/1994 11/5/2007 4784

2.3.1.3. Base Flow Data

Base flow for every reach of the National Hydrography Dataset (NHD) (Figure 9) within the

state of Michigan has been calculated under the Groundwater Inventory and Mapping Project

(GWIM) (MDEQ et al., 2006). These data were used in conjunction with the discharge data at

the USGS gauges to compare observed base flows with the base flow from the SWAT model.

2.3.1.4 River Hydraulic Data

No regional curve data relating channel dimensions to watershed area were available for the

BCRW. However, regional data were available for the Upper Menominee River watershed,

which is in the same physiographic region, Central Lowland Eastern Lake section, as the BCRW


(Mistak & Stille, 2007). These data were compared to surveyed channel metrics at USGS stream

gage sites in the BCRW and high-resolution aerial imagery and found to be consistent for use in

developing the hydraulic routing and channel dynamics capabilities in SWAT.

2.3.2 Spatial Data

A large variety of public GIS data were gathered from public sources, agency clearinghouses,

and agency personnel to build the GIS sediment budget database for the BCRW. These data

were all projected into the common projection and datum, NAD 1983 – Michigan State Plane

South. These data are described below and were used to build a watershed hydrology and

sediment yield model using SWAT and the GIS based ArcView SWAT interface (AVSWAT).

2.3.2.1 Climatic Stations

Climatic station data were obtained from the National Weather Service, National Climatic Data

Center (NCDC) (http://www.ncdc.noaa.gov/oa/ncdc.html). Additional stations and climatic

region information were obtained from the Michigan Geographic Data Library (MiGDL)

(http://www.mcgi.state.mi.us/mgdl/) and the Weather Generator database in SWAT. These data

were used to develop a GIS point file showing the locations of climatic stations suitable for

hydrologic and sediment transport analyses in the BCRW. The spatial distribution of weather

gauges used for the SWAT model are illustrated in Figure 4.

2.3.2.2 Geologic Data

During the last glaciation, Battle Creek watershed was at the southern extent of advancing glacial

lobes. Consequently, the surficial geology includes a mix of inter-bedded glacial lake deposits

(lacustrine clay), sand ridges from old shorelines, unsorted till, and sorted outwash deposits.

Quaternary geology data for the Battle Creek watershed were obtained from the USGS,

1:100,000 scale series of geologic maps for the Great Lakes region (Figure 5).

http://pubs.usgs.gov/dds/dds38/shape.html

2.3.2.3 Terrain Data

Terrain in this region is generally subdued. Uplands include rolling hills and flat fields

interspersed with meandering stream valleys (Figure 6). Elevation within the watershed ranges


from 800 feet near the city of Battle Creek to 1,060 feet farther upstream. The highest quality

terrain data here are 1/3 arc second (10 meter) data from the USGS (http://ned.usgs.gov/).

2.3.2.4 Soils Data

There are two levels of soil data available, Soil Survey Geographic Database (SSURGO) and

State Soil Geographic Database (STATSGO). Field mapping methods used national standards to

construct the soil maps in the SSURGO database range from 1:12,000 to 1:63,360 and are the

most detailed level of soil mapping (Figure 7). STATSGO mapping provides general planning

level information for soils and was generated at larger scale (Figure 8).

http://www.ncgc.nrcs.usda.gov/products/datasets/

2.3.2.5 Hydrography Data

The USGS National Hydrography Database (NHD) represents the most current hydrography for

the region (Figure 9) and was developed from 1:24,000 scale topographic maps

(http://nhd.usgs.gov/). The 10-meter terrain model DEM was used to enhance digital

hydrography data for the BCRW. The digitally generated streams were compared with the NHD

dataset to insure accuracy with the model.

2.3.2.6 Dams Data

Data for dams within the watershed were retrieved from the U.S. Army Corps of Engineers

National Inventory of Dams (NID) (http://crunch.tec.army.mil/nidpublic/webpages/nid.cfm).

This website provides hydraulic data necessary for hydrologic modeling for the dams such as

volume, surface area, and maximum discharge. The dams within the BCRW listed within NID

are shown in Figure 10. It is important to note that recently removed dams (Elm Street Dam

within the city of Battle Creek and the Charlotte Dam) are not included with this figure. The

NID only includes dams that fall under the following categories:

1. High potential hazard class

2. Low potential hazard class exceeding 25 feet high and 15 acre-feet storage

3. Low potential hazard class exceeding 6 feet high and 50 acre-feet storage


Figure 4: Weather stations and climatic regions for the Battle Creek River Watershed


Figure 5: Quaternary geology


Figure 6: Elevation data


Figure 7: SSURGO soils


Figure 8: STATSGO soils


Figure 9: National Hydrography Dataset (NHD) streams within Battle Creek Watershed


Figure 10: Locations of Dams in the Battle Creek River Watershed from the National Inventory of Dams and MI Dept. of Natural Resources.


2.3.2.6 Land Use Data

There were two potential sources of publicly available land use data included in the watershed

geodatabase: the 1992 and 2001 versions of the National Land Cover Databases (NLCD) developed

by the federal level, multi-agency, Multi-Resolution Land Characteristics Consortium (MRLC).

Both of these data sources may be obtained from:

MRLC - http://landcover.usgs.gov/classes.php.

There are three primary differences between the NLCD1992 and NLCD2001 databases. First,

NLCD2001 data are of generally higher quality because the 2001 initiative took advantage of

“lessons learned” during the production of the 1992 data. This includes improved coordination and

timing of satellite flights, improved processing methods, and optimized land cover classifications.

Second, NLCD2001 data are more current as they were developed using satellite imagery from 2001

whereas NLCD is based upon 1992 satellite imagery. Third, the NLCD2001 database has included

additional low-density residential areas where TIGER Roads data were buffered. The differences in

production methods and database quality between the NLCD1992 and NLCD2001 are extensive.

Complete descriptions of each database may be found at;

NCLD1992 - http://landcover.usgs.gov/natllandcover.php

NLCD2001 - http://www.mrlc.gov/mrlc2k_nlcd.asp

The NCLD1992 includes 21 specific land uses that represent a variety of urban, agricultural and

natural landscapes. The NLCD2001 dataset contains a further refined land use scheme including 28

different land use classes. The physiology of the vegetation for these landscapes is represented in

SWAT and includes typical characteristics necessary to simulate hydrologic and erosion processes

including, but not limited to, growing season length, phyto-productivity, leaf area, plant water use,

organic matter accumulation, nutrient uptake, and soil protection.

The NLCD 2001 data was used for analysis purposes, as it was the most detailed and accurate data

available. These were shown previously in Figure 2.


2.3.2.7 Aerial Imagery

High-resolution real color aerial imagery was developed for this region by the U.S. Dept. of

Agriculture Farm Service Agency (FSA) and was obtained from the Natural Resources Conservation

Service (NRCS). This imagery is shown in Figure 11. The high resolution of this imagery allowed

for quality assurance checking and identification of land cover including agricultural lands,

forestland, presence of riparian areas, and large areas of bank erosion.

2.3.3 SWAT Model Creation

Data from the sources described above was compiled to create a complete SWAT model for the

watershed. Daily precipitation and temperature data from the NCDC was formatted for input into

SWAT. Land cover (2001), terrain, and STATSGO soils GIS data were compiled and brought into

the SWAT project through the SWAT’s Arcview interface AVSWAT.


Figure 11: 2006 FSA Compliance Imagery


3.0 BATTLE CREEK RIVER SEDIMENT BUDGET RESULTS

3.1 Literature Review

A list of reviewed sources can be found in the Cited Literature section of this report. Literature

came from a variety of sources that accurately represented the region and watershed. Many studies

on sediment had been conducted on the impoundments downstream of confluence of Battle Creek

with superfund sites and TMDLs being conducted. Several references had been made throughout

the literature that the Battle Creek River was the largest contributor of sediment to the Kalamazoo.

The results of the literature review and interviews with local authorities failed to identify the source

of this common notion; it could not be validated. The focus of research and water quality projects in

the area has been on phosphorous data collection and the impacts of phosphorus loadings on water

quality in the Kalamazoo River, not sediment.

The BCRW Management Plan, drafted by the Calhoun Conservation District (2004), addresses a

wide variety of environmental issues within the BCRW including areas that have erosion and

sedimentation issues. These areas are shown in Figure 12. The areas listed within the Watershed

Management Plan focused on locations where Best Management Practices (BMP) could be applied

and developed cost estimates for their implementation. It was estimated within the report that if

applied, these BMPs could reduce sediment loads to the watershed by 2,800-3,000 tons per year.

In addition to available literature, conversations with local stakeholders indicated that publications

on the impact of dam removal within the BCRW will be released soon following completion of field

studies.


3.2 Regional Sedimentation Study

The estimates for total BCRW sediment from the sediment budget methods are summarized in Table

3 and spanned nearly two orders of magnitude. Specific results for each method and sample

computations are presented in section 3.2.1.

Table 3: Sediment Yield Estimates for the Battle Creek River Watershed

Source Sediment Yield T/mi2/yr t/ha/yr U.S. Water Resources Council, 1968 10-800 0.35 Leopold et. al, 1995; Corbel, 1959 131 0.46 Brune, 1951 1514 5.30 Dendy and Bolton, 1976 685 2.40 Syed, Bennett, & Rachol, 2004 22 0.08 Ouyang, Bartholic, & Selegan, 2005 25-49 0.09–0.17 Past 516(e) studies 154 0.54 SWAT model 240 0.84

3.2.1 Sediment Yield

3.2.1.1 Denudation Rate

The broad scale estimate of denudation for lowlands with cold winters as provided by Leopold, et.

al. was 2.9 cm per 1,000 years (1995; Corbel, 1959). If this rate is reflective of BCRW, then the

total sediment yield can be estimated as:

Denudation rate * unit conversion * Bulk density1 * Basin area = Yield (2.9cm)/(1000yr) * (.01m/cm) * (1.6 t/m3) * (6.242e8 m2) = 28, 960 t/yr = 0.46 t/ha/yr = 131 T/mi2/yr

While computed from a broad, physiographic based estimate of denudation rate, this estimate of

sediment yield is certainly within a reasonable range for a Great Lakes watershed dominated by

agricultural land use (Inamdar, S., and Naumov; Ouyand, et. al., 2005).

1 Average bulk density according to N.R.C.S. Soil Survey data – www.websoilsurvey.nrcs.usda.gov.


3.2.1.2 Regional Sediment Yield Data

A wholly separate estimate for the same physiographic region listed minimum, average, and

maximum values as 10, 100, and 800-tons/square mile/yr, respectively (0.035, 0.35, and 2.8 t/ha/yr)

(U.S. Water Resources Council, 1968). The average is similar to that from section 3.2.1.1.

Brune (1951) conducted a comprehensive survey of hundreds on reservoir sediment survey data

from old Soil Conservation Service dams across the United States with watersheds covering nearly

the entire range of physiographic regions and land types. While there was no single relationship to

predict sediment yield across the database, sediment yield did show significant dependence on

watershed area when controlled for lithology and land cover type. Brune used these results to

develop nomographs to predict sediment yield for various regions based upon unit volume of annual

runoff (volume / watershed area), percentage of land under cultivation, and watershed size. Results

from this nomograph for Battle Creek Watershed indicate the expected sediment yield would be

approximately 1,500 T/mi2/yr, or 5.3 t/ha/yr.

Dendy and Bolton (1976) developed sediment yield relationships for the physiographic regions of

the United States using observed sediment discharge data from gauging stations. For the Great

Lakes region, their data included a number of agricultural basins similar to BCRW. Sediment yield

predicted according to their method is:

Sediment Yield = 674 (Watershed Area)-0.16

= 674 (624.2km2) -0.16

= 2.4 t/ha/yr = 685 T/mi2/yr

In another study, using a combination of modeling methods including the Revised Universal Soil

Loss Equation (RUSLE) for soil erosion and a GIS based sediment delivery model (SEDMOD),

Ouyang, et al (2005) estimated sediment loading rates for many Great Lakes watersheds considered

to be high contributors of sediment. These methods use information about soils, rainfall, slopes,

land roughness, and land management practices to estimate average annual sediment loads from each

watershed. The estimate of average annual sediment load for the Kalamazoo River basin from this

study was 50,000 to 100,000 tons/year or 25 to 49 T/mi2/yr (0.09 to 0.17 t/ha/yr).

Using the modeling results from the Wells, et al. studying the sediment dynamics within the

Plainwell-Otsego impoundment (2005), the assumed sediment load rate into the Plainwell


impoundment is 71 t/day (25,915 t/year). Assuming the upstream contribution of sediment is

proportional to watershed area, the estimated sediment load from the BCRW is estimated as:

Sediment Deposition Rate * Battle Creek Area / Plainwell Area = Yield (25,915 t/yr) * (241 mi^2) / (1,299 mi^2) = 4,808 t/yr = 0.077 t/ha/yr = 22 T/mi2/yr

In addition to the above listed studies, previous 516(e) watershed budget projects on tributaries to the

Great Lakes were examined. A strong correlation exists within these studies between watershed area

and sediment yield (Figure 3). The relationship was applied to the drainage area of the BCRW with

the following results:

Sediment Yield (t/yr) = 272.2 (Watershed Area (km2))0.75

= 272.2 (624 km2) 0.75

= 34,000 t/yr = 0.54 t/ha/yr = 154 T/mi2/yr

3.2.1.3 Sediment Survey

As mentioned in section 2.0, this study was inconclusive because of a lack of data for impoundments

within the watershed.

3.2.2 Watershed Soil Erosion

Within the BCRW Management Plan drafted under a Clean Water Act 319 grant, a list of critical

areas within the watershed was listed as mentioned previously in the Reviewed Literature section of

the report. BMPs were suggested along with estimated sediment reduction potential. Many of these

critical areas listed stated a problem of streambank erosion contributing large amounts of sediment to

the river. A location map of all the sediment BMPs noted can be seen in Figure 12. Table 6,

adapted from the Watershed Management Plan, lists all BMPs where the source of water quality

degradation was excessive sedimentation. If applied, these BMPs could theoretically reduce

sediment loads by 2,800-3,000 tons per year. These numbers represent potential reductions and

were not compared to existing total loadings because sediment budget data were not available at that

time.


Figure 12: Sediment BMPs as listed in the "Battle Creek River Watershed Management Plan"


3.2.3 Fluvial Processes

3.2.3.1 Bank Erosion

Bank erosion has been monitored at several sites through the Kalamazoo Watershed to estimate

sediment and phosphorous loading and implement erosion reduction measures. The site within the

BCRW that was studied is in Convis Township. This site was at a sharp bend within the river and

was experiencing significant erosion problems (Figure 15). This site was loading an estimated 61

tons per year into the Battle Creek River, which was the second highest sediment load of ten sites

surveyed within the Kalamazoo River Watershed for the Clean Michigan Initiative (CMI) Grant

Project. Additional vegetation, riprap, and crib walls were added to reduce further erosion of the

bank at that site (Figure 16).

Three sites were designated for geomorphic assessment within the BCRW in conjunction with the

Watershed Management Plan (Boley-Morse, 2004). These sites included an un-channelized

riparian-forested corridor, channelized riparian-forested corridor, and a channelized agricultural

corridor. In order to establish an estimate of annual sediment loading from streambank erosion for

the entire watershed, more sites should be surveyed to accurately represent the bank erosion within

the BCRW.


Figure 13: Engineering of channels (Boley-Morse, 2004)

Figure 14: Engineering of channels (2) (Boley-Morse, 2004)


Figure 15: CMI erosion site before stabilization (Source: www.kalamazooriver.net)

Figure 16: CMI erosion site after bank stabilization (Source: www.kalamazooriver.net)


3.3 Watershed Hydrology and Sediment Yield Modeling

3.3.1 Model Calibration Results

The SWAT model was calibrated to match the overall water budget for the entire watershed (Figure

17) and the total flow at the USGS gage at Battle Creek (Figure 18). The most important aspects of

calibrating distributed, GIS based watershed hydrology and sedimentation models is calibrating

hydrology and using highest resolution terrain data. This is because model performance and

accuracy is most limited by the resolution of computational grid size as set from the terrain data

(Jenks, et. al., 2006; Riedel, et. al., 2005; Riedel, et. al., 2002). Following calibration for hydrology

using terrain data, proper land use and soils data are the most influential factors in determining

accuracy in sediment yield estimates (Jenks, et. al., 2006; Cotter, A.S., 2003). With the surface

hydrology calibrated and best available land use and soils data, SWAT estimated sediment loading

based on input parameters. A summary figure of sediment yield (tons per square mile) for the entire

watershed as compared to water yield and precipitation can be seen in Figure 19. From this figure, it

is obvious that there is a strong correlation between water yield and sediment yield for a given year.

The water yield data was obtained from the USGS gage 04105000 (Battle Creek at Battle Creek,

MI).


Figure 17: SWAT calibration comparison


Figure 18: SWAT calibrated flow data versus observed data from the USGS discharge gage at Battle Creek


Figure 19: SWAT output. Total water yield and sediment yield for entire basin


3.3.2 Model Results

3.3.2.1 Channel Sources

Bedload sediment transport within the Battle Creek River and its tributaries was estimated with the

SWAT model that was developed and calibrated for the watershed. Model results of bedload include

erosion and deposition occurring within the channel of each subwatershed. Table 4 summarizes

estimated channel erosion and deposition within each subwatershed with positive values showing

erosion of bed sediments and negative values showing deposition of bed sediments. These results

should not be viewed as “absolute” estimates because no calibration was possible to test the validity

and absolute accuracy of these results. However, the computational methods are mechanistically

based and consistent across the reaches. Consequently, the relative differences between reaches

should provide reasonable approximation of differences in bed material flux. These results can also

be seen in Figure 20. The estimated sediment load at the confluence with the Kalamazoo from

bedload transport is approximately 20 tons per year.

Table 4: Average annual channel erosion or deposition estimation by sub-basin

Subwatershed Length Erosion or (-) Deposition miles Tons T/mi t/km

1 0.78 0 0.0 0.02 2.84 4 1.4 0.83 16.28 -1 -0.1 0.04 5.19 -4 -0.8 -0.45 0.50 -5 -10.0 -5.66 11.47 10 0.9 0.57 0.88 0 0.0 0.08 4.27 1 0.2 0.19 2.00 2 1.0 0.6

10 3.42 6 1.8 1.011 1.13 3 2.7 1.512 1.48 -9 -6.1 -3.413 0.83 0 0.0 0.014 2.47 -3 -1.2 -0.715 8.65 3 0.3 0.216 5.02 1 0.2 0.117 2.05 -2 -1.0 -0.618 2.94 5 1.7 1.019 5.15 2 0.4 0.220 6.46 13 2.0 1.121 6.26 -6 -1.0 -0.5


3.3.2.2 Watershed Sources

After running the SWAT model over a number of years, estimates of average annual sediment load

by subwatershed were made. Figure 21 illustrates the average annual sediment load in different

subwatersheds from the SWAT modeling results. Table 5 shows sediment loads in tons per year for

each subwatershed as calculated from SWAT. The total estimated annual sediment load from

watershed sources is approximately 68,500 tons per year or 0.84 t/ha/yr.

Table 5: Average annual sediment load estimates from watershed sources

Subwatershed Area Average Annual Sediment Load

Average Annual

Sediment Load

mi^2 T/mi^2 t/ha Tons 1 8.1 240 0.8 1,9002 1.8 160 0.6 3003 44.8 220 0.8 10,0004 17.1 400 1.4 6,8005 4.9 540 1.9 2,6006 22.4 230 0.8 5,3007 7.9 230 0.8 1,8008 21.3 210 0.7 4,4009 8.5 290 1.0 2,400

10 16.7 410 1.4 6,90011 8.3 280 1.0 2,30012 3.0 200 0.7 60013 0.4 100 0.4 014 9.8 330 1.2 3,30015 20.0 260 0.9 5,10016 17.5 200 0.7 3,50017 13.3 100 0.3 1,30018 16.3 110 0.4 1,70019 8.1 220 0.8 1,80020 21.7 240 0.8 5,10021 13.5 100 0.4 1,400

The overall contribution of bedload sediment (20 T/yr) as compared to the total watershed sediment

load (68,500 T/yr) is small. Figure 22 shows a breakdown of the sediment budget results from the

SWAT model by source. This figure clearly shows that the bedload contribution to the sediment

budget is very small (0.03%) as compared to the watershed sources.


Figure 20: Average annual channel deposition or erosion per mile of channel.


Figure 21: Average annual sediment yield by sub-basin.


Figure 22: Sediment budget from SWAT model results by source


4.0 CONCLUSIONS AND RECOMMENDATIONS

The results of the BCRW Sediment Budget Study revealed a wide variation in total estimated

sediment yield. However, results from comprehensive modeling and the most relevant regional

methods were consistent and estimated yields in the range of 0.5 to 0.8 t/ha/yr. While bed material

in channels was predicted to be somewhat dynamic, it was not a significant portion of the overall

sediment budget. Reviews of recent reports, literature, and field projects identified streambank

erosion, instability, and ditch dredging as a potentially large and relatively undocumented portion of

local sediment budgets at the reach. Unfortunately, quantitative estimates of these contributions to

the total sediment budget could not be made. Sediment yields from bank erosion processes,

especially in agricultural areas where floodplain and stream banks have undergone active accretion

from heavy sediment loads due to past agricultural activities, have been identified as potentially

large component of the total sediment budget in similar watersheds being studied under the

Conservation Effects Assessment Program (CEAP) (Simon, 2005). It is recommended that future

projects in the watershed focus on these processes and the acquisition of field data to support the

development and analyses of reach scale fluvial sediment budgets using SIAM.

The results of the SWAT modeling done for the BCRW fall well within the published values of

sediment yield for the region. The model was calibrated with daily flow data from the USGS gage at

Battle Creek. No sediment data was available for calibration, but with the hydrology correct, SWAT

gave a reasonable estimate for sediment loading. They were also similar to results of past 516(e)

studies developing sediment budgets for the region.

Soil erosion and sedimentation can have a significant impact not only on the environment but the

economy as well. A few of the economic effects include loss of topsoil and nutrients in croplands,

increased fertilizer needs due to nutrient losses, road and highway structural damage due to

streambank erosion, additional need for water supply treatment, loss of storage capacity in flood

protection impoundments, depredation of water-recreational areas, decreasing depth of navigation

channels, and increased dredging costs (U.S. Army Corps of Engineers, 2005). These sediments can

be sources of a wide array of environmental pollutants including nutrients, residual agricultural

chemicals, and, while not present in BCRW, sediments contaminated with toxic or hazardous


substances which threaten aquatic species. These issues can be very important as cumulative effects

of pollutant loadings to harbors and the Great Lakes can also have significant economic impacts;

market values of homes decrease in closer proximity to contaminated areas (Braden, et al., 2006).

Efforts to reduce sedimentation from Great Lakes tributaries since the early 90s have been

successful. The Great Lakes Tributary Modeling Program through the U.S. Army Corps of

Engineers has addressed this problem with application of computer-based models and development

of tools to evaluate the impact that forestry, farming, urban sprawl, storm water management,

wetland protection, and BMP applications have on sediment loading rates (U.S. Army Corps of

Engineers, 2005). Over the last 18 years, the Great Lakes Basin program for soil erosion and

sediment control has protected over 129,000 acres of land against soil loss and prevented the

discharge of over 1.6 million tons of sediment to surface waters (Great Lakes Commission, 2007).


CITED LITERATURE

Battle Creek GIS Department. (2007). NPDES Storm Sewer Outfall Map. Boley-Morse, K. (2004). “The Battle Creek River Watershed Project.” Presentation for the 319

grant. Braden, J.B., Taylor, L.O., Won, D., Mays, N., Cangelosi, A., and Patunru, A.A. (2006). “Economic

Benefits of Sediment Remediation.” Final report for project GL-96553601, 120 p. Brune, G.M. (1951). Sediment Records in Midwestern United States; International Association of

Science Hydrology, Tome III, Brussels Symposium, pp. 29-38. Calhoun Conservation District. (2004). The Battle Creek River Watershed Management Plan. Clean

Water Act 319 Planning Grant, 162 p. Corbel, J. (1959). Vitesse de l’erosion:Zeits, Geomorphologie, 3:1-28. Cotter, A.S., 2003. Water quality model output uncertainty as affected by spatial resolution of input

data. J. American Water Resources Association, August. Dendy, F.E. and Bolton, G.C. (1976). Sediment Yield-Runoff Drainage Area Relationships in the

United States, J. Soil and Water Conservation, 31:264-266. Deutsch, M., Vanlier, K.E., and Giroux, P.R. (1960). “Ground-Water Hydrology and Glacial

Geology of the Kalamazoo Area, Michigan.” State of Michigan Department of Conservation Geological Survey Division, Progress Report Number 23, 22 p.

Fongers, D., and Henderson, J. (2006). “Peak Flow Analysis of Michigan USGS Gages.” MDEQ

Land and Water Management Division, 139 p. Fongers, D., Manning, K. and Rathbun, J. (2007). “Application of Richards-Baker Flashiness Index

to Gaged Michigan Rivers and Streams.” Michigan Department of Environmental Quality, 102 p. Great Lakes Commission. (2007). “Great Lakes Basin Program for Soil Erosion and Sediment

Control 2007: Annual Report.” Keeping it on the Land, 9(3): 12 p. Hansen, E.A. (1971). “Sediment in a Michigan Trout Stream: Its Source, Movement and Some

Effects on Fish Habitat.” North Central Forest Experiment Station, U.S. Department of Agriculture.

Homer, C.H., Yang, L., Wylie, B., and Coan, M. (2004). Development of a 2001 National Land-

Cover Database for the United States, Photogrammetric Engineering & Remote Sensing, 70(7): 829-840.


Inamdar, S., and Naumov, A. 2006. Assessment of Sediment Yields for a Mixed-landuse Great Lakes Watershed: Lessons from Field Measurements and Modeling. Journal of Great Lakes Research, 32:471-488.

Jenks, A., Bolstad, P.V., Riedel, M.S., and Vose, J.M. 2006. Estimating Sediment Yield in the

Southern Appalachians using WCS-SED, Proc. Joint 8th Federal Interagency Sedimentation and 3rd Hydrologic Modeling Conferences, Reno, NV, April 2-6.

Kalamazoo River Study Group. (2007). “Supplemental Remedial Investigation/Feasibility Study

Work Plan: Morrow Dam to Plainwell.” 51 p. Kieser and Associates (2003). “Nonpoint Source Opportunities for Point Sources: Streambank

Restoration/Clean Michigan Initiative Sites.” www.kalamazooriver.net. Kieser, M.S. and Ott, N. (2001). “Post-BMP Implementation Monitoring Results for the Cooper

Township Agricultural Site #2 Area A: Potential for a Small-scale agricultural Site to Generate Phosphorus Credits in the Kalamazoo RiverWatershed.” Prepared for the Water Environment Research Foundation, Project 97-IRM-5C

Kirsch, K., Kirsch, A. and Arnold, J.G., 2002. Predicting Sediment and Phosphorus Loads in the

Rock River Basin Using SWAT. Transactions of the ASAE, 45(6): 1757-1769. Lane, L.J. (2001). A sediment transport and yield model for alluvial streams, Seventh Federal

Interagency Sedimentation Conference, Reno, Nevada, pp. I-8 - I-15. Lane, L.J., Hernandez, M. and Nichols, M.H. (1997). Processes Controlling Sediment Yield from

Watersheds as Functions of Spatial Scale. Journal of Environmental Modeling & Software, 12(4): 355-369.

Luukkonen, C.L., Blumer, S.P., Weaver, T.L., and Jean, J. (2004). “Simulation of the Ground-

Water-Flow System in the Kalamazoo County Area, Michigan.” USGS Scientific Investigations Report 2004-5154, 77 p.

MDEQ (1998). “Guidebook of Best Management Practices for Michigan Watersheds.” Michigan

Department of Environmental Quality Surface Water Quality Division, 640 p. MDEQ (2002). “Water Quality Monitoring of Lake Allegan and Tributaries, 2001.” Michigan

Department of Environmental Quality Surface Water Quality Division, Staff report, MI/DEQ/SWQ-02/064, 23 p.

MDEQ (2005). “Removing Dams Improves Water Quality, Fisheries Habitat and Recreational Use.”

August 11, 2005: http://www.michigan.gov/dnr/0,1607,7-153-10369-124681--,00.html. MDEQ, Calhoun Conservation District. (2003). “The Watershed Partner Project.” Michigan’s

Nonpoint Source Program, Tracking Number 2003-0025, 2 p.


MDEQ, Calhoun Conservation District. (2004). “Battle Creek Watershed Planning Project.” Clean Water Act 319 Nonpoint Source Grant, 1 p.

MDEQ, City of Battle Creek. (2005). “City of Battle Creek Illicit Discharge Elimination.” Clean

Michigan Initiative Clean Water Fund Grant, 2001-0027, 2 p. MDEQ, City of Battle Creek. (2006). “City of Battle Creek City Hall Runoff Project.” Clean

Michigan Initiative Clean Water Fund Grant, 2004-0152, 2 p. MDEQ, USGS, Michigan State University. (2006). “Groundwater Inventory and Map (GWIM)

Project Technical Report.” State of Michigan Public Act 148, 254 p. Michigan Department of Natural Resources. (1987). “Kalamazoo River Remedial Action Plan.”

Second Draft, 279 p. Mistak, J.L, and Stille, D.A. (2007). “Upper Menominee River Regional Curve.” Regional

Reference Curves, Michigan Stream Team, 19 p. Mooney, D.M., 2006. SIAM, Sediment Impact Analysis Methods for Evaluating Sedimentation

Causes and Effects. Joint Federal Interagecy Conference, Reno, NV. Mooney, D.M., Biedenharn, D.S., and Watson, C.C. 2001. Sediment Impact Assessment Model. Seventh Federal Interagency Sedimentation Conference, Reno, NV. Nearing, M.A., 1998. Why soil erosion models over-predict small soil losses and under-predict large

soil losses. Catena, 32(1): 15-22. Ouyang D, Bartholic J, Selegean J. 2005. Assessing Sediment Loading from Agricultural Croplands

in the Great Lakes Basin. Journal of American Science, 1(2): 14-21. Rachol, C.M., Fitzpatrick, F.A., and Rossi, T. (2005). “Historical and Simulated Changes in Channel

Characteristics of the Kalamazoo River, Plainwell to Otsego, Michigan.” USGS Scientific Investigations Report 2005-5044, 67p.

Renwick, W.H., Smith, S.V., Bartley, J.D. and Buddemeierc, R.W. (2005). The role of

impoundments in the sediment budget of the conterminous United States. Geomorphology, 71(2005): 99-111.

Rheaume, S.J., Hubbell, D.L., Rachol, C.M., Simard, A., and Fuller, L.M. (2004). “Sediment

Characteristics and Configuration within the Otsego City Dam Impoundment on the Kalamazoo River, Michigan, 2001-02.” USGS Water Resources Investigations Report 03-4218, 62 p.

Riedel, M.S., Jenks, A., Bolstad, P.V., and Vose, J.M. 2005a. Application and Validation of a

Sediment Yield Model in Developing Regions of the Southern Appalachians, Watershed Management to Meet Water Quality Standards and Emerging TMDL, Amer. Soc. Agricultural Engineers, Atlanta, GA. March 5-9. 655 p.


Riedel, M.S., Brooks, K.N., and Verry, E.S. 2005b. Impacts of land use conversion on bankfull discharge and mass wasting. Journal of Environmental Management, 76(2005)326–337.

Riedel, MS and JM Vose. (2002). Forest road erosion, sediment transport and model validation in

the Southern Appalachians. Proc. Second Interagency Hydrologic Modeling Conference, United States Advisory Committee on Water Information – Subcommittee on Hydrology, Las Vegas, NV, July 28th – August 1st.

Robertson, D.M., Saad, D.A., and Heisey, D.M. (2006). “Present and Reference Concentrations and

Yields of Suspended Sediment in Streams in the Great Lakes Region and Adjacent Areas.” USGS Scientific Investigations Report 2006-5066.

Simon, A., 2005. Evaluation of the Importance of Channel Processes in Ceap-Watershed Suspended-

Sediment Yields, Eighth Federal Interagency Sedimentation Conference. U.S. Subcommittee on Sedimentation, Reno, NV.

Soller, D.H. and Packard, P.H. (1998). Digital Representation of a Map Showing the Thickness and

Character of Quaternary Sediments in the Glaciated United States East of the Rocky Mountains, U.S. Geological Survey Digital Data Series DDS-38.

Syed, A.U., Bennett, J.P., and Rachol, C.M. (2004). “A Pre-Dam Removal Assessment of Sediment

Transport for Four Dams on the Kalamazoo River between Plainwell and Allegan, Michigan.” USGS Scientific Investigations Report 2004-5178, 43 p.

Trimble, S.W. and Pierre, C. (2000). US Soil Erosion Rates--Myth and Reality. Science, 289: 248-

250. U.S. Army Corps of Engineers (1989). “Sedimentation Investigations of Rivers and Reservoirs.”

Engineer Manual 1110-2-4000. U.S. Army Corps of Engineers (2005). “Great Lakes Tributary Modeling Program: Status Report.”

U.S. Army Corps of Engineers Great Lakes & Ohio River Division, 82 p. U.S. Army Corps of Engineers (2007). “Annual Report/Contract Dredging Report: Saugatuck

Harbor, MI.” U.S. Army Corps of Engineers Detroit District, Project Operations Section. U.S. Water Resource Council. (1968). The Nation’s Water Resources, Part 5, Chapter 5, Washington

D.C. Wells, R.R., Langendoen, E.J., and Simon, A. (2007). “Modeling Pre- and Post-Dam Removal

Sediment Dynamics: The Kalamazoo River, Michigan.” Journal of the American Water Resources Association, 43(3):773-785.

Wesley, J.K. (2005). “Kalamazoo River Assessment.” Michigan Department of Natural Resources

Fisheries Division, Special Report 35, 123 p.


TABLES


Table 6: Sediment BMPs listed for BC Watershed

Station # County Township Section Stream Problem/

Concern BMP Sediment

Reduction/ year

1b Calhoun COB 1 BC River Streambank Erosion

Stabilize 250' of streambank utilizing bio-engineering along Linear Path 69 tons

4 Calhoun Emmett 6 BC River

Manicured lawn with no

vegetative buffer Undercutting streambank

Stabilize 10' of streambank utilizing bio-engineering Install 50' vegetative bufferstrip 6 tons

5a Calhoun Emmett 6 BC River Streambank erosion at culvert Stabilize 10' of streambank around culvert utilizing bil-engineering 1 ton

10 Calhoun Pennfield 28 BC River


vegetative buffer Streambank

erosion

Stability 700' of streambank utilizing bio-engineering Install 700' vegetative bufferstrip 385 tons

11b Calhoun Pennfield 21 BC RiverStreambank erosion from

human access Install canoe launch and access stairs 2 tons

12 Calhoun Convis 18 BC River

Pasture gully erosion to

wetland adjacent to the BC River

Install a rotational grazing system with 1000' of fencing and a 200' vegetative bufferstrip adjacent to wetland 11 tons

14 Eaton Bellevue 28 BC River Road edge erosion off bridge Plant vegetation that can withstand roadside abuse 1 ton

15 Eaton Bellevue 28 BC River


vegetative buffer and streambank

erosion

Stability 50' of streambank utilizing bio-engineering Install 50' vegetative bufferstrip 1 ton

18 Eaton Walton 18 BC River Unlimited livestock access

Install 3000' of fencing, 200' vegetative bufferstrip, revegetate streambank with seeding, and 2 alternative watering systems 41 tons




Reduction/ year

20 Eaton

Walton, Carmel,Eaton, and Brookfie

ld

16 15 10 11 2 1 35 26 25 24 19 20 29 28 33 5 8 17 20 21 22

27

BC River

River disconnected

from the floodplain from bermed banks

due to historical dredging

Resotre 1 mile of river from a Type F5 stream to a Type C5 stream utilizing natural channel design 986 tons

21 Eaton Carmel 35 BC River Unpaved road eroding off bridge Pave 50'x30' bridge area Install curbing and filter strip 38 tons

22 Eaton Carmel 26 BC River Unpaved road eroding off bridge Pave 50'x30' bridge area Install curbing 38 tons

23a Eaton Carmel 25 BC River Limited livestock access Install 500' vegetative bufferstrip and 2 watering systems 14 tons

23b Eaton Carmel 25 BC River Unpaved road eroding off bridge Pave 50'x30' bridge area Install curbing and filter strip 38 tons

24 Eaton Carmel 24 BC River Possible run-off from pasture

Install 3000' vegetative bufferstrip on left bank and 1000' vegetative bufferstrip on right bank Install livestock stream crossing and 2

watering systems

75% reduction

26a Eaton Eaton 20 BC River Unpaved road eroding off bridge Pave 50'x30' bridge area Install curbing and filter strip 38 tons

26b Eaton Eaton 20 BC RiverEroding and undercutting stream bank

Construct in-stream geomorphic assessment to determine reason for streambank erosion, stabilize 50' of streambank utilizing bil-

engineering or an in-stream structure Install 50' vegetative bufferstrip1 ton

28 Eaton Eaton 20 BC River

Cropland planted adjacent to the road with no

buffer uphill from river

Install 500' vegetative bufferstrip 65% reduction

29a Eaton Eaton 29 BC River Limited livstock access

Install 1,000' of fencing, 500' vegetative bufferstrip and 1 watering system

65% reduction

29b Eaton Eaton 29 BC River Unpaved road eroding off bridge Pave 50'x30' bridge area Install curbing and filter strip 38 tons




Reduction/ year

30 Eaton Brookfield 20 BC River


vegetative buffer Streambank

erosion

Conductin in-stream assessment to determine reason for streambank erosion, stabilize 50' of streambank utilizing bio-engineering or in-

stream structure Install 50' vegetative bufferstrip 14 tons

31 Eaton Brookfield 21 BC River Unlimited

livestock accessInstall 1,000' of fencing, 500' vegetative bufferstrip and 1 watering

system 14 tons

37a Barry Assyria 34 Wanandoga Creek

3 perched culverts

positioned causing

streambank erosion

Reposition culverts utilizing the MESBOA technique 21 tons

38 Barry Assyria Wanandoga Creek

Entrenched channel with

severe streambank

erosion at bend and only one

culvert is receiving the active flow

Replace culverts utilizing the MESBOA technique, stabilize streambank by gathering in-stream data and install structure, and

establish vegetation 28 tons


Entrenched channel with

severe streambank

erosion

Replace culverts utilizing the MESBOA technique, stabilize streambank by gathering in-stream data and install structure, and

establish vegetation 83 tons


Unlimited livestock access Install 1000' of fencing, 1 livestock crossing, and 2 watering systems 10 tons

41 Eaton Bellevue 7 Wanandoga Creek

Bridge abutments are not as wide

as bankfull causing

streambank erosion

Conduct in-stream geomorphic assessment to determine bridge design, replace structure, install structure to stabilize bank 21 tons

43 Eaton Bellevue 7 Wanandoga Creek

Unpaved road eroding off bridge Pave 30'x30' bridge area Installing curbing and filter strip 38 tons




Reduction/ year

44b Eaton Bellevue 7 Wanandoga Creek

Unpaved road eroding off bridge Pave 30'x30' bridge area Installing curbing and filter strip 38 tons

45a Eaton Bellevue 5 Wanandoga Creek

Perched and inadequate sized

culvert that is creating erosion from backwater

Replace culverts utilizing the MESBOA technique, stabilize streambank by gathering in-stream data and installing an in-stream

structure, and re-establish vegetation 17 tons

45b Eaton Bellevue 4 Wanandoga Creek

Culvert positioning is

creating streambank erosion and

narrow vegetative bufferstrip adjacent to cropland

Reposition and size culvert utilizing the MESBOA technique, stabilize streambank by gathering in-stream data and installing and in-stream structure, re-establish vegetation, and expand vegetative bufferstrip

25 tons

47 Calhoun Convis 10 Ackley Creek

Perched and inadequate sized culvert, eroding stream channel,

erosion from road run-off, and

manicured lawns with no

vegetative bufferstrip

Replace culverts utilizing MESBOA technique and install a 400' vegetative bufferstrip along both sides of creek 1 ton

48b Eaton Walton 29 Indian Creek

Eroding streambank from human access

Install access stairs and fishing platform 2 tons

49 Eaton Walton 29 Indian Creek

Severely eroding stormwater discharge channel

Remove concrete slabs and reshape bank behind culvert outfall, extend and redirect culvert, reshape banks to 2:1 slope, use rock rip

rap, bio-engineering, seeding, and log deflectors Work with Olivet College to install some innovative stormwater BMPs such as rain

gardens, roof water control, and parking lot design

525 tons

51 Eaton Walton 28 Indian Creek

Eroding streambank from human access

Stabilize 75' of streambank utilizing bio-engineering 21 tons




Reduction/ year

52 Calhoun Lee 6 Indian Creek

Eroding and undercutting stream bank

Stabilize 50' of streambank utilizing bio-engineering and install 50' vegetative bufferstrip 6 tons

53 Calhoun Lee 5 Indian Creek

Barnyard run-off from roof water and livestock in

wetland

Install roof water control structures, rotational grazing system with 4000' of fencing, manure storage facility, 2500' vegetative bufferstrip, 4

watering systems, and restore 2 small wetlands

65% reduction

55 Calhoun Convis 13 Indian Creek

Narrow vegetative buffer

adjacent to cropland and 2 drain tile outlets

Widen vegetative buffer 45' Install 2 tile inlet filters 65% reduction

56a Calhoun Convis 18 State and

Indian Drain

Eroding and undercutting stream bank

Stabilize 60' streambank utilizing bio-engineering 8 tons

56b Calhoun Lee 18 State and

Indian Drain

Lack of vegetation

growing on fabric from reshaped

bank adjacent to a manicured lawn

Revegetate bank and establish a 75' bufferstrip 65% reduction

57 Calhoun Lee 18 State and

Indian Drain

Unlimited livestock access

Install 1000' of fencing, 500' vegetative bufferstrip, and 1 watering system 65%

60 Calhoun Clarence 21 State and

Indian Drain

Eroding stream bank and

manicured lawn with lack of vegetative bufferstrip

Stabilize 100' streambank utilizing bio-engineering and install a vegetative bufferstrip 25 tons

62 Eaton Brookfield 29

Hogle and

Miller Drain

Unlimited livestock access causing severe

streambank erosion

Install 5000' of fencing and vegetative bufferstrip, 1 livestock crossing, and 2 watering systems Seed eroding bank 275 tons

Battle Creek River Watershed Sediment Budget Study › wp-content › uploads › ... · A number of stream and watershed conservation efforts have been undertaken to minimize the

Documents