Remote Sensing as a Tool for Monitoring Soil Moisture Dynamics in

Tile-Drained, Agricultural Watersheds

ABSTRACT

Two tile-drained, agricultural watersheds in the Headwaters Area of East-Central Illinois and one in Minnesota will be evaluated, using remotely-sensed and field collected data, to ascertain the feasibility of utilizing tiled soils as a reservoir for flood damage mitigation. An algorithm for automatic detection of tile lines from remotely-sensed data will be built. An expansive model will be developed for one watershed to extrapolate seasonal/interannual to decadal/century variability in anticipation of 50 and 100-year floods. Guidelines for application of this technology will be published and distributed to community decision-makers and farmers located in the watersheds.

INTRODUCTION

Since the late 1860's, subsurface tile drains have been used to convert swamplands to productive farmlands in much of the glaciated Midwest. Drains, made of clay or concrete sections or perforated plastic are placed 2-5 feet below the soil surface to deliver excess water to drainage ditches that connect with natural surface channels so that crop roots remain aerated. The focus of this proposal is the use of structures connected with tile drain systems to manage the local water table for flood control, and agronomic and environmental benefits. Remote sensing techniques are proposed as a means for determining soil moisture conditions which should guide operation of drainage system control structures. The Embarras River watershed upstream of Villa Grove, Illinois is the selected study area for quantifying the effects of tile drain control on downstream flooding. Selected sites in both Illinois and Minnesota are proposed for field-scale testing.

Team members for this three-year study are listed below:

Illinois Center for Agricultural Geographic Excellence (ILCAGE), a consortium of the University of Illinois Department of Agricultural Engineering, the Champaign County Soil and Water Conservation District (CCSWCD), and Illinois Agribusinesses

Biosystems and Agricultural Engineering Department, University of Minnesota

The Environmental Research Institute of Michigan, International (ERIM)

Daedalus Enterprises, Inc.

Precision Aviation, Inc.

Upper Embarras River Basin Planning Committee

Principal investigators from the University of Illinois and the University of Minnesota provide the theoretical understanding and computer modeling expertise for investigating control of subsurface drainage systems for water table management. The Champaign County Soil and Water Conservation District provides the critical link between theory and implementation of the design strategy by individual landowners. ERIM International, with over fifty years experience in remote sensing technology, system design, data processing, and applications, provides expertise and software for process automation and modeling. The Airborne Multispectral Digital Camera (AMDC), designed and built by Daedalus Enterprises, Inc., for the National Aeronautics and Space Administration (NASA) under a Small Business Innovation Research (SBIR) venture, is the primary remote sensing data collector. Precision Aviation, Inc. will collect the aerial imagery and provide local knowledge of the region. The Upper Embarras River Basin Planning Committee will play a key role in informing and involving landowners within the study region. Included in this Step 1 proposal are brief descriptions of the Research to Be Conducted, Motivation and Expected Consequences, Technical Approach, and Estimate of Cost. Copies of the curriculum vitae of the principal investigators are provided as attachments.

RESEARCH TO BE CONDUCTED

This project addresses NRA-97-MTPE-12 priority topics 1, 2, and 4:

Priority topic 1: Flood hazards.

Priority topic 2: Field and numerical experiments on land-atmosphere interaction

Priority topic 4: Seasonal-to-interannual variability in terrestrial hydrologic systems.

The team will use a combination of remote sensing and "in situ" field sampling technology to evaluate the impact of controlled retention and release of water in subsurface tile drains for flood control, water quality improvement, and crop drought protection. The project has five phases:

Phase 1: Remote sensing.

Phase 2: Field system experiments.

Phase 3: Field-scale and watershed-scale computer modeling.

Phase 4: Guidelines for system portability and application.

Phase 5: Publication and distribution.

Phase 1: Remote Sensing

The goal of Phase 1 is to develop a soil moisture signature library from aerial infrared images of sufficient quality to allow identification of tile drainage systems, as well as predictive modeling of available water storage in the soil profile for a variety of soil types. Imagery will be collected for various soil moisture and cover conditions over the entire test watershed, as well as selected fields in the other two watersheds.

Phase 2: Field System Experiments

Selected tile drain systems in Illinois and Minnesota will be outfitted with structures for controlling water table height. The purpose will be to provide ground-truthing in conjunction with Phase 1, as well as to investigate agronomic risks and benefits associated with water table management. Analysis results will be inputs to a model for estimating tile system outlet hydrographs.

Phase 3: Computer Modeling

At the field scale, the goals of water table management are to be able to remove excess water quickly during wet periods and store water for use during drought. In addition, such control can provide a means for encouraging denitrification during fallow periods so that excess nitrogen is released to the atmosphere instead of to surface waters that may serve as drinking water supplies. At the watershed scale, control of flow through individual tile systems must be coordinated such that peak flows at the watershed outlet are reduced, thereby mitigating flood impact. Studies will be directed towards determining cause and effect relationships between ground water retention and release, flood damage, and crop drowning. Computer models will be developed to provide cost-benefit data to decision-makers.

Phase 4: Guidelines for System Portability and Application.

Based on the results from Phases 1, 2, and 3, specific flood control recommendations will be made for Villa Grove. Guidelines for applying the method in other communities will be proposed. In addition, recommendations for maximizing field-scale benefits of water table management will be developed.

Phase 5: Publication and Distribution

Recommendations will be compiled in reader-friendly handbooks geared towards

landowners and community decision-makers.

MOTIVATION AND EXPECTED CONSEQUENCES

Communities located downslope in mildly-sloping watersheds of the glaciated Midwest are naturally subject to flooding. Villa Grove, IL in the Embarras River watershed is the target community for this project. Flooding has been officially considered critical for Villa Grove since 1947. Many studies have proposed various flood control methods and plans with cost-benefit analyses. Structural solutions have been considered cost ineffective. Current measures employ flood plain acquisition, evacuation, flood proofing, zoning, and flood insurance. In addition to flood control, water table management in tile-drained fields offers a means for removal of excess nitrate through denitrification, and storage of groundwater against drought. Remote sensing can be used to estimate available storage in the soil profile which will aid in coordinating storage and release of water retained in field drainage tile. Some expected outcomes include:

Development of soil moisture signatures for various soil types.

Identification of tile lines throughout the watershed draining to Villa Grove, IL.

A plan for reducing peak flows at Villa Grove, IL.

Development of guidance materials for landowners and community leaders in tile-drained regions.

Performance evaluation of the NASA AMDC.

TECHNICAL APPROACH

The overall approach is to determine relationships between remotely-sensed soil/tile signatures (from reflected infrared wavelengths) and field-measured soil moisture, and to use those relationships to develop guidelines for water table management in tile-drained regions. Specific plans are outlined by phase below:

Phase 1: Remote Sensing

Multispectral imagery of 1- to sub-meter spatial resolution will be collected and analyzed during bare soil (fallow) conditions and at full crop cover each of the three years. Images will be collected after significant rainfall (during active tile function) for the Embarras watershed draining to Villa Grove, and for a minimum of three test fields (described under Phase 2). Images will be collected using the AMDC developed for NASA by Daedalus Enterprises, Inc. Combinations of nine bands in the reflected infrared will be tested. A contingency exists for using thermal bands during the second or third year of testing if the budget permits the additional expense. ERIM will develop an automated process for identifying tile lines from the images.

Phase 2: Field System Experiments

A minimum of three fields with mapped subsurface drainage systems will be outfitted with structures for controlling water table height. The selected fields will represent the three most common subsurface drainage designs:

Regularly-spaced lateral drain tubes connecting to a main

Irregularly-spaced drain tubes

Vertical surface inlet connecting to subsurface main to drain "prairie potholes"

Piezometers will be installed at each site to monitor the depth to the local water table. Rainfall will be measured at each site. Flow and nitrate exiting the drain system outlet will also be measured. Surface soil moisture will be measured on days aerial imagery is collected. The goal is to relate soil moisture, water table position, and image response such that available soil water storage volume can eventually be estimated solely from remote sensing techniques. Candidate field sites will be described in the Step 2 proposal.

Phase 3: Computer Modeling

The University of Illinois, University of Minnesota, and ERIM will develop computer models to address both field-scale and watershed-scale response to water table management and create the cost-benefit tables. The field-scale model will be designed to predict the outflow hydrograph and crop yield for a variety of weather conditions. The watershed-scale model will combine drainage system outlet hydrographs with surface flows to predict watershed outlet response. Portions of several existing models will be investigated, including the "Drainage Response of Pothole Landscapes and the Erosion and Transport of Sediment Model (DROPLETS) developed at the University of Minnesota.

Phase 4: Guidelines for System Portability and Application.

Field testing and modeling results will be used to develop guidelines for other Midwestern locations, taking into account soil differences and cropping practices.

Phase 5: Publication and Distribution

Handbooks containing recommendations and guidelines will be published in non-technical language built around explanatory visuals. Distribution will be to decision-makers, community leaders, farm managers, and any farmer, landowner, or operator expressing interest. Key to the dissemination of recommendations to landowners and community policy-makers will be close interaction with the Champaign County Soil and Water Conservation District and the Upper Embarras River Basin Planning Committee.

ESTIMATE OF COST

Total estimated cost of the project is $1.5 million, spread over three years. A detailed budget will be included in the Step 2 proposal. Main budget items are outlined below:

Salaries and benefits for professional and supporting personnel: $450,000

Hardware for image processing and analysis: $65,000

Field monitoring equipment and control structures: $40,000

Remotely-sensed data: $90,000

Travel: $25,000

Publication and printing: $20,000

Subcontractors' fees (ERIM, UMN, Precision Aviation): $500,000

Indirect costs: $310,000

Total: $1,500,000

The proposed budget allows for hiring of a half-time GIS/Remote Sensing professional and graduate assistants. Hardware purchases include a workstation data server and laptop units for field work. The total includes estimated costs of remotely-sensed imagery for the Embarras watershed and selected fields, performance evaluation of the AMDC, development of tile detection algorithms, and production of 1,000 copies of the resulting water table management handbook.

Curriculum Vita - Richard A. C. Cooke, Ph.D.

Assistant Professor, Agricultural Engineering Department,

University of Illinois

Education:

Degrees Field Institutions Date

BS Agricultural Engineering University of the West Indies 1981

MS Agricultural Engineering University of Guelph 1986

PhD Agricultural Engineering Virginia Polytechnic Institute 1993

Professional Experience:

Assistant Professor, Agricultural Engineering Department,

University of Illinois: 11/94-present

Research Associate, Agricultural Engineering Department,

Virginia Tech: 11/93-10/94

Lecturer, Faculty of Engineering, University of the West Indies 8/87-6/90

Assistant Lecturer, Faculty of Engineering,

University of the West Indies 8/96-6/87

Selected Publications

Mostaghimi, S., S.W. Park, R.A. Cooke and Y. Wang. 1997. Assessment of management alternatives on a small agricultural watershed. Water Research (In Press).

Kurien, V. M., R. A. Cooke, M. C. Hirschi, J. K. Mitchell. 1997. Estimating drain spacing of incomplete drainage systems. Transactions of the ASAE 40: 377-382.

Cooke, R.A., S.E. Walker, M.C. Hirschi and J.K. Mitchell. 1997. Modeling flow and transport in the Little Vermilion River watershed. Invited Paper at the Spring Meeting of the American Geophysical Union (Baltimore, Maryland, May, 1997).

Northcott, W.J., R.A. Cooke, A.K. Verma, S.E. Walker, M.C. Hirschi and J.K. Mitchell. 1997. Water quality modeling of an East Central Illinois watershed in a GIS environment. AWRA's Annual conference and symposium on conjunctive use of water resources: aquifer storage and recovery. October 19-23, Long Beach'97, California.

Verma, A.K., R.A. Cooke, M.C. Hirschi and S. E. Walker. 1997. Potential of Integrating Remote Sensing and Geographic Information Systems in Soil Erosion Modeling. Proc. of the Conference on Management of Landscapes Disturbed by Channel Incision, Oxford, Mississippi.

Walker, S.E., R.A. Cooke, A.K. Verma, S. Kim, W.J. Northcott, M.C. Hirschi and J.K. Mitchell. 1997. Estimating flow from irregularly spaced subsurface drainage systems. AWRA's Annual conference and symposium on conjunctive use of water resources: aquifer storage and recovery. October 19-23, Long Beach'97, California.

Curriculum Vita - Michael C. Hirschi, Ph.D.

Associate Professor, Agricultural Engineering Department,

University of Illinois

Education:

Degrees Field Institutions Date

Assoc. General/Engr. Transfer Anoka-Ramsey Comm. Coll. 1976

BSAgE Agricultural Engineering University of Minnesota 1978

MSAgE Agricultural Engineering University of Minnesota 1980

PhD Agricultural Engineering University of Kentucky 1985

Professional Experience:

Section Leader, Soil and Water, Agricultural Engineering Dept.

University of Illinois 1996-present

Water Quality Program Coordinator,

Cooperative Extension Service, University of Illinois: 1992-present

Associate Professor, Agricultural Engineering Department,

University of Illinois: 1991-present

Visiting Research Scientist, National Center for

Supercomputing Applications, University of Illinois: 1/94-8/94

Research Associate, Corps of Engineers

U. S. Army Construction Engineering Research Lab,

Champaign, Illinois: 1989-1994

Interim Associate Director, Cooperative Extension Service,

University of Illinois: 2/92-12/92

Acting Assistant Director for Agriculture & Natural Resources,

Cooperative Extension Service, University of Illinois: 7/91-12/91

Assistant Professor, Agricultural Engineering Department,

University of Illinois: 1985-1991

Selected Publications:

Mitchell, J. K., M. C. Hirschi and A. S. Felsot. 1994. Nitrate losses from cropping management systems. Proceedings of the National Symposium on Protecting Rural America's Water Resources: Partnerships for Pollution Solutions, October 23-26, Omni Shoreham Hotel, Washington, DC. Ground Water Protection Council, Oklahoma City, OK, pp. 25-30.

Kurien, V.M., R.A. Cooke, M.C. Hirschi, and J.K. Mitchell. 1997. Estimation of Effective Drain Spacing for Incomplete Drainage Systems. Transactions of the ASAE 40(2):377-382.

Lesikar, B.J., A.L. Kenimer, M.C. Hirschi, J.K. Mitchell, and A.S. Felsot. 1997. Modeling Alachlor Movement under Spill Conditions Using LEACHMP. Accepted for publication in Transactions of the ASAE.

Curriculum Vita - Jay D. Davis

GIS Coordinator, Champaign County Soil and Water Conservation District

Civilian and Military Education:

Degrees Field Institutions Date

BS Fish & Wildlife Mgt. University of Missouri 1976

MA Mgt. & Public Admin. Webster University, St. Louis, MO 1982

MILITARY COURSES

Basic Imagery Interpretation Course, Ft. Huachuca, AZ 1977

Defense Sensor Interpretation & Analysis Training Program (DSIATP),

Offutt AFB, NE 1980

Advanced Imagery Interpretation Course, Weisbaden-Scheirstein, FRG 1983

Advanced Non-Commissioned Officer Course, Ft. Huachuca, AZ 1984

Advanced Sensor Imagery Analysis Training, Washington, D.C. 1986

Professional Experience:

GIS Coordinator, CCSWCD, Champaign, IL 1996-present

Duties include creation of maps depicting the location of field drainage tile using scanned aerial CIR imagery, planning and incrementally implementing layers for countywide agricultural databases, acting as district representative on county-wide task force to enable farmers to use the INTERNET in daily operations.

Cartographic Aid, Illinois State Office, NRCS, Champaign, IL 1994-1996

Duties included analog mapping of wetlands from color aerial crop compliance slides onto 7.5 minute quad sheets

Chief Imagery Analyst, Master Sergeant (MSG),

US Army, Washington., DC 1986 - 1993

Duties included supervisory and non-supervisory positions, introduction of commercial multispectral imagery into military intelligence products, acting as Non-Commissioned Officer In Charge, Imagery Analysis Cell, Joint Intelligence Center (1990-1991) during Operations Desert Shield and Desert Storm, eighteen months as project manager for imagery-derived military intelligence products, representing US Army on the Tactical and Military Multispectral Requirements Evaluation Group, TaMMREG, a National-level multispectral task force, and producing military intelligence geographic products for strategic and tactical support.

Senior Imagery Analyst, Sergeant First Class (SFC), US Army,

Wiesbaden-Schierstein, FRG 1982 - 1985

Duties included developing and executing analog imagery analytic procedures, and the supervision of production and dissemination of first and second echelon imagery-derived intelligence products.

Imagery Analyst, Staff Sergeant (SSG), US Army, Ft. Bragg, NC 1979 - 1982

Duties included developing and executing analog analytic procedures, and producing third echelon imagery derived intelligence products.

Curriculum Vita - Larry E. Reed

Manager, Information Production Department, Earth Sciences Group, ERIM, International

Education:

Degrees Field Institutions Date

BBA Operations Research/ Eastern Michigan University 1982

Information Systems

Professional Experience:

Manager, Multispectral Program Office,

Environmental Research Institute of Michigan (ERIM) 1991-present

Currently responsible for all business activities associated with the IPF including system development, product development, multispectral training, and all other activities associated with the business area utilizing commercially available multispectral information.

Manager, Information Production Department, ERIM 1978-present

Managed the development of the Image Processing Facility (IPF) which represents the accumulation of one of the most robust collections of software available for manipulation of all types of multispectral data including MSS, TM, CZCS, AVHRR, SPOT, and scanned map and photographic data.

Supervisor, Computer Operations, ERDC,

Bendix Aerospace Systems Division 1972-1978

Contributed to development of computer processing systems, algorithms, and processing techniques to handle Landsat MSS data. Directed numerous projects with both aircraft and Landsat data including: monitoring strip mining, water depth analysis, analysis of land cover, and hydrologic and geologic projects. Participated in the design, development, training, and delivery of one of the first digital data processing systems (MDAS) to Japan, Argentina, Egypt, and India. Directed efforts to combine photographic processing techniques with digital techniques to produce high-quality output products to demonstrate the results of digital processing, to

introduce geometric control to multispectral processing to allow production of cartographically accurate resource inventory maps, and to refine resampling techniques to insure the best possible pixel representation during the resampling process.

Image Interpretation Specialist, U.S. Air Force 1965-1972

Duties included a variety of assignments (Southeast Asia, Japan, SAC, and TAC) with exposure to all types of reconnaissance systems such as synthetic aperture radar (SAR), infrared (IR), and electro-optical (EO) used by the Air Force for both tactical and strategic intelligence applications. Attended several advanced intelligence schools and applied manual interpretation skills to all types of intelligence applications.

Curriculum Vita - Bruce N. Wilson, PE, Ph.D.

Associate Professor, Biosystems and Agricultural Engineering Department,

University of Minnesota

Education:

Degrees Field Institutions Date

BS Agricultural Engineering University of Minnesota 1976

MS Agricultural Engineering University of Minnesota 1979

PhD Agricultural Engineering University of Kentucky 1984

Professional Experience

Associate Professor, Biosystems and Agricultural Engineering Department

University of Minnesota 1994-present

Assistant Professor, Agricultural Engineering Department,

University of Minnesota 1991-1994

Associate Professor, Agricultural Engineering Department

Oklahoma State University 1987-1991

Assistant Professor, Agricultural Engineering Department,

Oklahoma State University 1983-1987

Research Specialist, Agricultural Engineering Department,

University of Kentucky 1979-1983

Research and Teaching Assistant/Associate,

Agricultural Engineering Department, University of Minnesota 1976-1979

Selected Publications

Couger, G., B.N. Wilson and C.T. Rice. 1992. Determination of drainage networks from plot-size and basin-size areas. Applied Engineering in Agriculture, Vol. 8(2): 185-189.

Gruber, A.M., B.N. Wilson, J.L. Nieber, B. Hansen, J.W. Brown and E. Brooks. 1996. Observed results from a field study of depression focused recharge. ASAE Paper 962089, Presented at the 1996 ASAE Annual International Meeting, ASAE, 2950 Niles Road, St. Joseph, MI.

Rice, C.T., B.N. Wilson and M. Appleman. 1988. Soil topography measurements using image processing techniques. Computer and Electronics in Agriculture, Vol. 3: 97-107.

Wilson, B.N. and D. Storm. 1993. Fractal analysis of surface drainage networks for small upland areas. Transactions of the ASAE, Vol. 36(5): 1319-1326.

Wilson, B.N. and D. E. Storm. 1993. Surface flow paths for small drainage areas. ASAE Paper 932135, Presented at the 1993 ASAE Summer Meeting, Spokane, WA.