Ritter and Adel Shirmohammadi CONTENTS 8.1 Introduction 8.2 History of Drainage in the United States 8.3 Materials and Methods for Subsurface Drainge 8.4 Types of Drainage Systems 8.4.1
Trang 1Agricultural Drainage and Water Quality
William F Ritter and Adel Shirmohammadi
CONTENTS
8.1 Introduction
8.2 History of Drainage in the United States
8.3 Materials and Methods for Subsurface Drainge
8.4 Types of Drainage Systems
8.4.1 Surface Drainage
8.4.2 Conventional Subsurface Drainage
8.4.3 Water-Table Management
8.5 Water-Table Management Design
8.5.1 Preliminary Evaluation and Feasibility of Site
8.5.4 System Layout and Installation
8.5.5 Operations and System Management
8.6 Soil and Crop Management Aspects of Water-Table Management
8.7 Water Quality Impacts
Trang 28.8 Impact of Drainage of Surface Water Quality
8.9 Institutional and Social Constraints
8.10 Summary
References
8.1 INTRODUCTION
Water management for agricultural purposes can be traced to Mesopotamia about
9000 years ago.1Herodotus, a Greek historian of the fifth century B.C., wrote about
a drainage works near the city of Memphis in Egypt
Drainage has been part of American agriculture since colonial times Withoutdrainage, it is hard to imagine the U.S Midwest as we know it in the 20th century,the epitome of agricultural production Much of Ohio, Indiana, Illinois, and Iowaoriginally was swamp, or at least too wet to farm Without drainage, irrigation devel-opment in the western United States would have failed because of waterlogging andsalinity
In the 1960s and 1970s, drainage was considered an honorable and viable soiland water conservation practice Drainage technology developed rapidly during thisera In the 1990s, drainage is greeted with angry response in many quarters Because
of drainage, better than half the original wetlands in this country no longer excist Inaddition, drainage has reduced the habitat for birds and wildlife and has had detri-mental effects on water quality2 Today the design and operation of drainage systemsmust satisfy both agricultural and environmental objectives
8.2 HISTORY OF DRAINAGE IN THE UNITED STATES
Early settlers brought European drainage methods with them to North America.These methods included small open ditches to drain wet spots in fields and to cleanout small streams In New York and New England, early settlers used subsurfacedrainage in addition to open ditches Material used for buried drains prior to the use
of clay-fired tile pipes included poles, logs, brush, lumber of all sorts, stones laid invarious patterns, bricks, and straw
In 1754, the Colony of South Carolina passed an act for draining the CacawSwamp.3The Dismal Swamp area of Virginia and North Carolina was surveyed byGeorge Washington for reclamation in 1763, and in 1778 the Dismal Swamp CanalCompany was chartered A drainage outlet for the City of New Orleans was cons-tructed around 1794.4
The first known colony-wide drainage law was enacted in New Jersey onSeptember 26, 1772 Early drainage works were constructed in Delaware, Maryland,New Jersey, Massachusetts, South Carolina, and Georgia under the authority of colo-nial and state laws The first organized drainage project in Maryland was authorized
by the legislature for draining the Long Marsh in Queen Anne and Caroline Counties.5Similarly, legislation authorizing drainage projects in Delaware dates back to 1793.6Drainage in the midwestern U.S began after 1850, when the Swamp Land Act
of 1849 and 1850 released large amounts of swamp and wetland still owned by the
Trang 3Federal government These lands were released for private development, with thefunds from their sale used to build drains and levees The Reclamation Act of 1902established the Bureau of Agricultural Engineering within the U.S Department ofAgriculture, which was responsible for the design and construction of many of themajor drainage ditches that were installed to create surface water outlets Drainagedistricts began to be organized in the early 1900s In its natural state, much of the fer-tile land in northwestern Ohio, northern Indiana, northcentral Illinois, northcentralIowa, and southeastern Missouri was either swamp or frequently too wet to farmbefore drainage was installed Drainage also permitted large areas in westernMinnesota, the gulf plains of Texas, northeastern Arkansas, and the delta area ofMississippi and Louisiana to be cultivated.7
Drainage problems developed as a consequence of irrigation developed in thearid west In the San Joaquin Valley of California, the Modesto Irrigation Districtdrained more than 18,000 ha In the Imperial Valley of California, over 81,000 ha ofcropland had drainage problems by 1919 Today over 80% of the cropland in theImperial Valley is drained Bureau of Reclamation irrigation projects such as theColumbia Basin in Washington, the Grand Valley (Nebraska), Big Horn Basin(Montana and Wyoming), Oahe (South Dakota), Weber Basin (Utah), Garrison(North Dakota), and Big Thompson (Colorado) have required drainage as a conse-quence of irrigation.3
8.3 MATERIALS AND METHODS FOR SUBSURFACE
DRAINAGE
The first use of clay tile for farm drainage is attributed to John Johnston, who lived
in the Finger Lakes region of New York Johnston imported patterns for type drain tile from Scotland in December 1835 Tiles were made from these patterns
horseshoe-at the B.F Whartenby pottery horseshoe-at Whorseshoe-aterloo, N.Y in 1835 They were made entirely byhand A crude molding machine was installed in 1838 in the Whartenby factory thatmade the process cheaper and faster.8Sometime after 1851, John Dixon developed amuch improved machine for making horseshoe tile In the 1870s, another newmethod of tilemaking that used a rectangular slab of clay instead of a conventionalmold was introduced.8
The first tilemaking machine, the “Scraggs,” was brought to America in 1848from England The machine operated on the extrusion process.8Many locally manu-factured tilemaking machines were patterned after the Scraggs machine; most of theearly manufacturers were located in New York State
Weaver8also discussed the early use of concrete tile for subsurface drainage In
1862, David Ogden developed a machine for making drain tile from cement and sand.Until 1900, concrete drain tile was used primarily where good clay was not available
In the 1940s, bituminized fiber pipe was used in the eastern States and generation plastic tubes were also introduced By 1967, corrugated plastic tubing wasmanufactured commercially in the United States from polyvinyl and polyethyleneresins The agricultural market tubing was very light and flexible and greatly reducedhandling and shipping costs Tile alignment problems were avoided.3By 1983, 95%
Trang 4early-of all agricultural subsurface drains installed annually in the U.S and more than 80percent in Canada consisted of corrugated plastic tubing.9
Subsurface drains were first installed in hand-dug trenches, followed by a bination of plowing and hand digging The first trencher introduced in 1855 was thePratt Ditch Digger revolving-wheel type that was horsedrawn.3The Hickok and theRennie elevator ditchers were patented in 1869 Another early machine was theJohnston Tile Ditcher made in Ottawa, Illinois All of the early machines requiredmore than one pass over the trench to excavate it to the required depth Singlepassmachines powered by horses came next and included the Blickensderfer TileDitching Machine, the Heath’s Ditching Machine, the Paul’s Ditching Machine andthe Fowler Drain Plow In the early 1880s, steam-powered wheel trenches were intro-duced The Bucheye steampowered trencher was introduced in 1882 In 1908, steampower was replaced with a gasoline engine on the Buckeye, which was the forerun-ner of today’s high-speed trenchers and laser-controlled drain plows
com-8.4 TYPES OF DRAINAGE SYSTEMS
8.4.1 S URFACE D RAINAGE
Surface drainage is used to remove water that collects on the land surface Surfacedrainage is used primarily on flat or undulating land where slow infiltration, slow per-meability, restricting layers in the soil profile, or shallowness of soil over rock or deepclays A surface drainage system usually consists of an outlet channel, lateral ditches,and field ditches Lateral ditches carry the water received from field ditches or fromthe field surface to the outlet channel.10
Surface drainage systems include land smoothing or grading, and field ditches.Land grading is the shaping of the land surface with scrapers and land planes toplanned surface grades Land smoothing removes small depressions and irregulari-ties in the land surface
Field ditches may be either random or parallel The random ditch pattern is used
in fields having depressional areas that are too large to be eliminated by land ing Field ditches connect the low spots and remove excess water from them Whenthe topography is flat and regular, a parallel ditch pattern is used The row directionshould be perpendicular to the ditch Drains do not have to be equally spaced andwater may flow in only one direction The drain should have a minimum depth of0.23 m and have a minimum crosssectional area of 0.50 m2 The sideslopes of theditches should be 8:1 or flatter to allow machinery to cross.11
smooth-8.4.2 C ONVENTIONAL S UBSURFACE D RAINAGE
Subsurface drains consist of underground pipe systems to collect excess water fromthe root zone and lower the water-table Subsurface drainage falls into two classes:relief and interception drainage.10Relief drainage is used to lower a high water-tablethat is generally flat or of very low gradient Interception drainage is to intercept,reduce the flow, and lower the flowline of the water in the problem area Relief drainsnormally consist of a system of parallel collection drains connected to a main drainlocated on the low side of a field or along a low waterway in the field The main drain
Trang 5transports the collected water to the outlet An interceptor drain often consists of asingle drain which intercepts lateral flow of groundwater caused by canal seepage,reservoir seepage, or levee-protected areas.
8.4.3 W ATER -T ABLE M ANAGEMENT
The trend in the humid areas of the United States is to develop a total water ment system Water-table management strategies can be grouped into three types:subsurface drainage, controlled drainage, controlled drainage–subsurface irrigation.12Subsurface drainage alone lowers the water table during wet periods and is governed
manage-by drainage system depth Controlled drainage is achieved manage-by placing a control ture, such as a flashboard riser in the outlet ditch or a subsurface drain outlet, to con-trol the rate of subsurface drainage Controlled drainage-subirrigation is similar to thecontrolled drainage system, except that supplemental water is pumped into the system
struc-to maintain the water table at a current level during drought periods Drainage is vided during wet periods by allowing excess water to flow over the control structure,which may be adjusted in elevation depending upon the rainfall (Figure 8.1) Thepractice has been used for years in peat and muck soils with high permeability and animpervious layer below the drains or with a naturally high water table.13
pro-The system can be applied in both the field and watershed scale using variouswater control structures and operational procedures.12,14 Water-table managementoffers more possibilities for flood control, improved water conservation, andimproved water quality than conventional drainage systems15 The greatest potentialfor water-table management systems is on relatively large flat land areas where highwater tables persist for long periods during the year There have been a number ofpapers in recent years dealing with the design, economics, and environmentalimpacts of controlled drainage systems.16,17
8.5 WATER TABLE MANAGEMENT DESIGN
Shirmohammadi et al.12outlined five tasks that must be performed to design a cessful and efficient water-table management system These tasks include prelimi-nary evaluation and feasibility of the site, detailed field investigation, designcomputations, system layout and installation, and operation and management Each
suc-of these tasks is discussed by Evans and Skaggs18in detail ASAE19has also loped a design, installation, and operation standard for water table management systems
deve-8.5.1 P RELIMINARY E VALUATION AND F EASIBILITY OF S ITE
Six site characteristics should be considered for successful performance of table management systems:
water-8.5.1.1 Drainage Characteristics
The site must require improved subsurface drainage to remove excess water that otherwise would restrict farm operations and crop growth Soils classified as
Trang 6“somewhat poorly drained,” “poorly drained,” and “very poorly drained” are primecandidates for water-table management Natural Resources Conservation Service soilsurvey manuals provide soil maps and classifications for each state within theAtlantic Coastal Plain.
8.5.1.2 Topography
Surface slopes should not exceed 1% for the system to be economically feasible Asthe slope increases, more control structures are required to maintain a uniform watertable
Trang 7systems to perform satisfactorily The deeper the barrier, the larger the volume ofwater required to fill the soil profile and raise the water table during irrigation.
8.5.1.5 Drainage Outlet
A good gravity or pumped drainage outlet is needed to provide adequate flow city for expected peak discharges For gravity flow systems, the drainage outletshould be at least 1.2 m below the average land surface A sump equipped with anappropriate pump can be constructed to collect the surface and subsurface drainageflow where an adequate natural drainage outlet is not present
capa-8.5.1.6 Water Supply
An adequate water supply must be available for the subirrigation mode Location,quantity, and quality of the water must be taken into consideration during the plan-ning stage
8.5.2 D ETAILED F IELD I NVESTIGATIONS
For efficient design, soil type and arrangement of soil horizons, soil hydraulic perties, crops, water supply, and various climatological and topographical parametersmust be considered Soil type, arrangement of soil horizons, soil hydraulic properties,and hydraulic conductivity (lateral conductivity values and soil water characteristicdata) determine drain line depth and spacing The crop and its rooting depth may alsoinfluence system design
pro-An accurate topographic map is required to evaluate the slope of the land and itsadequacy for any type of water-table management system A general guideline is toinstall the drain lines perpendicular to the slope, but this guideline can be modified,depending upon site conditions
Climatological data, such as rainfall, temperature, and solar radiation, are tant parameters Knowledge of climatological data can provide a good understanding
impor-of crop water use and periods impor-of peak water requirement Crop water requirementinformation is required for a controlled drainage–subirrigation system to determinethe external water supply size, pumping plant size, and overall management strategy.Design criteria also should be evaluated for each site based on economic and envi-ronmental quality considerations
Trang 88.5.3 D ESIGN C OMPUTATIONS
Data collected from the field investigation enables the design engineer to computeproper drain depth, drain spacing, drain grades, number and size of control structuresneeded to maintain a uniform water table, and a proper pump capacity required forthe water supply and the drainage outlet if a sump is used at the outlet Soil horizonarrangement data, topography, and crop-rooting characteristics will help to determinethe proper drain depth, which generally ranges from 1 to 2 m, depending upon siteconditions.18Soil hydraulic conductivity values and depth to the impermeable layerwill enable the engineer to evaluate the drain spacings, using the Hooghoudt’s steadystate drainage rate method for drainage conditions However, other procedures must
be used to evaluate the drain spacings if subirrigation is a part of the overall plan.18DRAINMOD, a water table management model for shallow water table condi-tions, is probably the most comprehensive model available for design of subsurfacedrainage, controlled drainage, and controlled drainage–subirrigation systems, pro-vided the required input data are available.20
8.5.4 S YSTEM L AYOUT AND I NSTALLATION
Using the information obtained during the first three steps, the design engineer needs
to prepare a map showing the field, location of laterals and mains, and location andnumber of control structures Appropriate grades for drains must also be specifiedusing the design standards and site information The type of water table managementsystem should also be specified
A contour map prepared during the second phase of planning must be used toidentify the location and grade of the drain lines and the control structures Locations
of the control structures are selected so that they provide the most uniform water tableelevations possible Water table fluctuations of 0.30 to 0.45 m and 0.15 to 0.20 m may
be tolerated for grain crops and shallow-rooted vegetable crops.18
Once the system layout is completed on a well prepared map, the size, spacing,and grade of drain lines and the size and capacity of the control structures are speci-fied A contractor then can initiate the installation according to specifications.Autolevel, laser-controlled plows and trenchers that provide accurate and fast in-stallation of the system are currently available However, caution is necessary regard-ing the hand installation of laterals and main to the drain in a closed system to ensurethat none will be left unattached
8.5.5 O PERATIONS AND S YSTEM M ANAGEMENT
This task is one of the most important aspects of the overall effort; traditionally, it hasbeen performed by the producer and most usually on a trial-and-error basis Selectingthe proper weir elevation, maintenance of the system, and timing of the subirrigationand drainage phases are part of the operation and management of the system Onlarge-scale fields (40.5 ha), there may be high spots and depressions that were notconsidered in designing the depth and spacings of the drain lines because of the eco-nomics of the system During the operation mode, however, a producer may adjust
Trang 9the control structure setting so that neither drought in high spots nor excess water indepressions will harm the crop Similarly, knowing when to reverse from the drainagemode to the subirrigation mode in a controlled drainage-subirrigation system requiresexperience as well as soil moisture measurement, using such devices as tensiometers.Tensiometers indicate the soil-water potential from which one may judge the timing
of subirrigation Weather forecasts can be used to evaluate the time for lowering thewater table to provide proper storage for incoming rain
Manual adjustment of the control structure setting is laborious; consequently, it
is often not adjusted because of the farmer’s conflicting schedule Research ments have enabled linking weather forecast data to the control structures throughcomputers, modems, and telephone lines.21In the future this type of system will prob-ably be used in commercial systems
develop-8.6 SOIL AND CROP MANAGEMENT ASPECTS OF
WATER-TABLE MANAGEMENT
The Southeast and Mid-Atlantic Coastal Plain have variable rainfall during the ing season This, combined with sandy soils with low water holding capacity, cancause drought conditions.22 These conditions are worse in soils with shallow rootzones caused by subsurface hardpans that could be controlled by deep chisel plow-ing Water-table management by controlled drainage–subirrigation can amelioratevariability of water supply.22, 23
grow-Intense rains in some regions are possible during the growing season.22As aresult of such rainfall, the shallow water tables that result from controlleddrainage–subirrigation leave fields vulnerable to flooding To prevent this, systemshave been designed to link controlled drainage–subirrigation to weather predictions.Fouss and Cooper21stopped subirrigation when a 55% or greater rainfall probability
is predicted They also recommended free drainage of the soil in advance of a dicted storm If free drainage is used, precautions must be taken not to drop the watertable so much that reestablishment of the desired level would be difficult.23,24
pre-For controlled drainage–subirrigation systems to be successful, the depth of thewater table must be low enough to prevent aeration problems and high enough to per-mit capillary rise into the root zone for plant uptake The capillary water contribution
to root uptake is negligible for water table depths 76 cm below the bottom of the rootzone in sandy soils or 92 cm in clay soils.23Doty26found the best water-table depthfor corn on sands or sandy loam in the Coastal Plain was 76–89 cm The recommended depth of the water table is 92–153 cm for clay soils.27The crop typeand climate in addition to soils determine where, within these ranges, the water tableshould be set
If the ratio of deep percolation to infiltration is greater than 1:10, a water tablewill not perch adequately and the site is unsuitable for controlled drainage–subirri-gation.27 Other soil factors that affect water-table management are poor surfacedrainage, organic soils that subside, and soil strength Poor surface drainage mayaffect trafficable conditions and soil aeration.22Shih et al.28recommended different
Trang 10water table depths for different crops and different times of the year on organic soils
to provide irrigation and reduce subsidence Deep tillage combined with controlledwater table depth can eliminate hard-pan problems that limit root growth depth.29
8.7 WATER QUALITY IMPACTS
8.7.1.1 Conventional Drainage
Land development using conventional drainage generally increases total annual flows from fields and peak outflow rates Studies in North Carolina have shown thatannual outflows increased 5% for surface drainage and 20% for subsurfacedrainage30, 31 when compared with natural undrained conditions Peak flow rates typically increased up to four times with surface drainage compared with natural con-ditions Subsurface drainage peak flow rates doubled compared with natural systems.Peak outflow rates varied greatly depending upon storm intensity, antecedent mois-ture, and drainage intensity The natural areas used for comparison were unmanagedforested areas without drainage improvement, flat (0.01 slope or less), and broad(exceeding km2)
out-Bengston et al.32 measured surface runoff and outflow from four plots inLouisiana on Commerce clay loam soil from 1982 to 1991 Two of the plots had bothsurface and subsurface drainage and two of the plots had surface drainage only Theaverage annual surface drainage was 402 mm from the surface and subsurface-drained plots and 614 mm from plots only with surface drainage The annual runofffrom surface and subsurface-drained versus only surface drained plots ranged from ahigh of 775 and 1085 mm in 1989 to a low of 150 and 208 mm in 1984, respectively.Subsurface drainage reduced surface runoff by an average of 35%, but the totaldrainage flow from surface and subsurface drain plots (i.e., runoff plus subsurfacedrain outflow) was about 35% more than for the plots with only surface drainage
by less than 15% compared with conventional drainage
8.7.2 N UTRIENTS
8.7.2.1 Conventional Drainage
The earliest research on tile drainage water quality was reported by Willrich.33Willrich collected water samples twice a month from 10 subsurface drainage outlets
Trang 11draining 2.4–148 ha in Iowa The median values for chemical properties of thedrainage water ranged as follows: total N 12 to 27 mg/L, ortho P 0.1 to 0.3mg/L; K 0.2 to 0.8 mg/L; hardness 350 to 440 mg/L as CaCO3, alkalinity 260
to 330 mg/ L, and pH from 7.4 to 7.8 The N was mostly in the NO3form
Bolton et al.34were the first to study the effect of agricultural drainage on waterquality in Ontario They measured nutrient losses in tile drainage on a Brookston claysoil in continuous corn, continuous bluegrass, and a four-year rotation of corn, oats,alfalfa, and alfalfa No fertilization was compared with fertilizer application rates of
17 kg/ha of N and 67 kg/ha P for all crops except first- and second-year alfalfa in therotation The corn received an additional 112 kg/ha of N The average annual N and
P losses are presented in Table 8.1 Nitrogen losses increased with fertilizer tions in four of the six cropping seasons Nitrate concentrations in the tile outflowwere above 10 mg/L for fertilized rotation corn and second-year alfalfa Croppingsystems had little effect on P concentrations Fertilizer application caused a smallincrease in P losses
applica-Baker and Johnson,35in a summary paper of several studies, concluded that centrations of NO3-N were greater in subsurface drainage than in surface runoff; NH3concentrations in runoff were usually greater than in subsurface drainage and P con-centrations in subsurface drainage were usually less than in runoff Baker andJohnson based their conclusions on a number of studies in different locations and re-present general conditions that exist for runoff and subsurface drainage water quality.Other studies have also shown that N losses in tile drainage increase with fertilizerapplication Logan and Schwab36monitored subsurface drainage water quality fromthree field-sized areas on glacial till soils in Union County, Ohio They found sea-sonal N losses varied from 0.1 to 45.6 kg/ha The highest loss was on a site where 224kg/ha of N was applied preplant to corn In 1972, only 22 kg/ha of N fertilizer wasapplied, but the seasonal N loss was still 36.4 kg/ha On the site where continuousalfalfa was grown, the seasonal N losses were 0.1 and 0.9 kg/ha in 1972 and 1973
con-TABLE 8.1
Average Annual N & P Losses in Tile Drains34
Trang 12No fertilizer was applied to the alfalfa, and the tile discharge was much lower thanfrom the other two sites where corn was grown.
Baker and Johnson37compared differential nitrogen fertilization rates and tile
NO3-N discharge rates on a Webster slit loam soil in Iowa The 5-year average annual
NO3-N loss from an area receiving an average of 56 kg/ha of N fertilizer was 26kg/ha The high fertilization rate area had an average annual NO3-N loss of 48 kg/haand received an average of 116 kg/ha/yr of N fertilizer The average annual flow vol-ume from the tile lines was 132 mm, which represents a significant contribution tostream flows in central Iowa
In another study on a Webster clay loam soil in southern Minnesota, Gast et al.38measured NO3-N losses from tile lines for annual N applications of 20, 112, 224, and
448 kg/ha to continuous corn Each treatment was replicated three times on plots 13.7
by 15.3 m Nitrate losses and tile flow volumes are summarized in Table 8.2 Waterflow through the tile lines occurred annually for approximately 6 weeks in the periodfrom mid-April through early July and constituted an equivalent flow from 7 to 22%
of the annual precipitation during the 3-year study Nitrate losses from the tile linesafter fertilizer applications for 3 years (1975) were 19, 25, 59, and 120 kg/ha/yr forthe 20, 112, 224, and 448 kg/ha N application rates Application of the recommended
112 kg/yr resulted in only slight increases in NO3-N concentrations in the tile water
or total losses from the tile lines compared with the 20 kg/ha treatment
Tillage also has an effect on the amount and timing of NO3-N and total N in surface drainage waters Gold and Loudon39compared P and N losses from conser-vation tillage (chisel plow) and conventional tillage (moldboard plow) from two 4-hawatersheds in the Saginaw Bay area of Michigan Total P and soluble P concentra-tions were higher in tile flow from conservation tillage than conventional tillage Thegreater losses of P in surface runoff for conventional tillage more than offset thelarger losses in P in tile flow for conservation tillage Nitrate concentrations weresimilar in the tile flow from both tillage systems (11.7 and 10.5 mg/L) but were higherthan in the surface runoff Kjeldahl N concentrations were higher in surface runoffthan in tile flow