Agricultural Engineering · Land & Water Resources Engineering · also Irrigation and Drainage Engineering, Irrigation Water Management
How much water a crop needs, when to give it, how to measure and convey it, and how to design surface irrigation.
In the exams: GATE AG Section 5 (irrigation depth, interval and command area in 5 of 8 papers 2019–26); ICAR AIEEA-PG Unit VII; ICAR AICE subject 43; NCEES PE irrigation; PRC ABE; IA CAIS
Irrigation engineering is the part of agricultural engineering that decides how much water a crop needs, when it needs it, how that water is measured and carried from the source to the field, and how it is spread over the field so that it ends up in the root zone rather than in drains, ditches or deep groundwater. This manual covers the soil-water-plant foundation, evapotranspiration and crop water requirement, effective rainfall, irrigation scheduling, flow measurement, the design and lining of open channels, buried pipeline distribution, the three surface methods (border, basin and furrow), the efficiencies used to judge a system, and the management of large canal commands. Sprinkler and drip design, drainage and groundwater are treated in their own subjects.
The subject matters because irrigated land produces a large share of the world's food from a smaller share of its cropland, and because irrigation is by far the largest human use of fresh water. Most of that water is still applied by surface methods, and a large fraction of what leaves a reservoir or well never reaches a plant root. The practising engineer's job is to close that gap with numbers: a depth to apply, an interval, a stream size, a channel section, a flume reading, an efficiency figure that can be checked in the field.
Worldwide the subject is examined in the same way. Indian entrance and fellowship papers (GATE AG, ICAR AIEEA-PG and AICE, ASRB NET, state public service commissions) ask short calculations on irrigation depth, interval, command area, weirs, duty and delta, and efficiencies. Licensing examinations in the United States (NCEES PE), the Philippines (ABELE), Nepal and Bangladesh test design with standard formulas, and the Irrigation Association certifications test scheduling, uniformity and water management as field practice. Where practice or standards differ between countries this manual says so; where a value depends on soil, crop or climate it is given as a range or left for local data.
Soil is a porous medium of mineral particles, organic matter and pore space. Water in the pores is held by adsorption on particle surfaces and by capillary forces in small pores. The energy state of this water is described by soil water potential, the work needed to move a unit quantity of water from the soil to a reference pool of pure free water. Its components are matric potential (from capillarity and adsorption, always negative in unsaturated soil), osmotic potential (from dissolved salts, negative), gravitational potential (from elevation above a datum) and pressure potential (positive below a water table). Water moves from higher to lower total potential. Potential is expressed as energy per unit volume (kPa), per unit weight (metres of head) or in bars; 1 bar is 100 kPa and roughly 10.2 m of water. The relation between water content and matric potential is the soil water characteristic or retention curve. It is steep for sands, which release most of their water at small suctions, and flat for clays, which hold water at high suctions. The curve shows hysteresis: at the same suction a drying soil holds more water than a wetting soil.
Water content is stated on a mass basis (mass of water per mass of oven-dry soil) or a volume basis (volume of water per bulk volume of soil). They are linked by bulk density: volumetric content equals mass content multiplied by the ratio of bulk density to the density of water. Confusing the two bases is the most common error in irrigation calculations. Saturation is the state in which all pores are filled. Field capacity is the water content after free drainage has largely stopped, usually one to three days after a thorough wetting; it is conventionally associated with a matric potential of about −33 kPa (one third of a bar), though −10 kPa is often used for sandy soils, and it is really a field property rather than a fixed laboratory point. The permanent wilting point is the water content at which plants wilt and do not recover overnight in a humid atmosphere; it is conventionally taken at −1500 kPa (15 bar). Water held between saturation and field capacity is gravitational water, which drains away and is not counted as available. Water held more tightly than the wilting point, including the hygroscopic water on particle surfaces, is not available to crops.
Available water capacity is the difference between field capacity and wilting point. Multiplied by root zone depth it gives total available water, TAW, as a depth: TAW (mm) = 1000 (θfc − θwp) Zr, with θ volumetric and Zr in metres. Loams and silt loams hold the most available water per metre; sands the least. Plants cannot extract all of TAW without stress, because as the soil dries the matric potential falls and hydraulic conductivity drops steeply. The fraction that can be removed before yield or quality suffers is called management allowed depletion (MAD) in North American practice and the depletion fraction p in FAO-56. The readily available water is RAW = p × TAW. Typical values of p lie between about 0.3 for shallow-rooted, sensitive crops such as many vegetables and about 0.6 or more for tolerant field crops; FAO-56 tabulates p by crop and adjusts it for evaporative demand. MAD is also a management choice: a farmer with limited labour may accept deeper depletion and longer intervals, and a high-value crop may be managed at a smaller MAD. Roots do not draw evenly from the root zone. The classical extraction pattern for a uniform soil is roughly 40, 30, 20 and 10 percent of use from successive quarters of the root depth, which is why sensors are placed in the upper half of the root zone.
Infiltration is the entry of water through the soil surface. The rate is high at first and declines toward a steady value close to the saturated hydraulic conductivity. Empirical models include Kostiakov, cumulative infiltration I = a t^b with b between 0 and 1; the modified Kostiakov form, which adds a steady term; Horton's exponential decay; and Philip's two-term equation, I = S t^0.5 + A t, where S is sorptivity. In the field, infiltration for basins and borders is measured with double-ring infiltrometers, the outer ring reducing lateral spread under the inner ring; for furrows, inflow-outflow tests on a furrow reach are more representative. Soil water content is measured directly by the gravimetric method, drying a sample at 105 °C to constant mass, which is the reference for calibrating every other method. Indirect methods include tensiometers, which read matric potential and work only in the wet range (above about −80 kPa, because the water column breaks at lower suctions); electrical resistance blocks, which work in the drier range; the neutron probe, which counts slow neutrons scattered by hydrogen; and dielectric methods such as time-domain reflectometry and capacitance probes. The pressure plate apparatus is used in the laboratory to establish the retention curve and the −33 and −1500 kPa points.
Evapotranspiration (ET) is the combined loss of water by evaporation from soil and plant surfaces and by transpiration through stomata. The two are hard to separate in the field, so they are treated together. ET needs energy (mainly net solar radiation), a vapour pressure gradient between the evaporating surface and the air, and turbulent transfer to carry vapour away. About 2.45 MJ of energy is needed to evaporate one kilogram of water at ordinary temperatures, which is why 1 MJ m⁻² day⁻¹ of energy is equivalent to about 0.408 mm day⁻¹ of evaporation. Three ET concepts must be kept apart. Reference evapotranspiration, ET₀, is the ET of a defined reference surface and depends only on weather. Crop evapotranspiration under standard conditions, ETc, is the ET of a disease-free, well-fertilised crop in a large field with no water limitation. Actual or adjusted crop ET, ETc adj, allows for water stress, salinity, poor stand or management.
FAO Irrigation and Drainage Paper 56 (Allen, Pereira, Raes and Smith, 1998) defined the reference as a hypothetical grass crop 0.12 m tall with a fixed surface resistance of 70 s m⁻¹ and an albedo of 0.23, closely resembling an extensive, actively growing, well-watered green grass. The daily form of the equation is ET₀ = [0.408 Δ (Rn − G) + γ (900 / (T + 273)) u₂ (es − ea)] / [Δ + γ (1 + 0.34 u₂)], where ET₀ is in mm day⁻¹, Rn is net radiation and G soil heat flux (MJ m⁻² day⁻¹), T is mean daily air temperature at 2 m (°C), u₂ is wind speed at 2 m (m s⁻¹), es − ea is the vapour pressure deficit (kPa), Δ is the slope of the saturation vapour pressure curve and γ is the psychrometric constant (kPa °C⁻¹). The first term in the numerator is the radiation (energy) term; the second is the aerodynamic term. For daily steps G is small and is usually taken as zero. The psychrometric constant is γ = 0.665 × 10⁻³ P with P the atmospheric pressure in kPa, so it falls with altitude. Wind measured at another height is converted to 2 m with u₂ = uz × 4.87 / ln(67.8 z − 5.42). Saturation vapour pressure should be computed as the mean of es at Tmax and Tmin, not at mean temperature, because the curve is non-linear.
Older methods each captured one driver: the Blaney-Criddle method used temperature and daylength, the radiation method used solar radiation, and pan methods used a measured evaporation. Each needed local calibration and tended to overestimate or underestimate in climates unlike those where it was developed. The FAO-56 Penman-Monteith equation combines the energy balance with an aerodynamic resistance and a canopy (surface) resistance, so it responds physically to all four weather drivers. FAO recommends it as the sole standard method for computing ET₀, and ASCE adopted a standardised form (the ASCE standardized reference ET equation) with a short grass reference close to FAO-56 and a tall (alfalfa) reference. When the full data set is missing, FAO-56 gives procedures for estimating radiation from sunshine hours or temperature range, humidity from minimum temperature, and wind from regional values, and it accepts the Hargreaves equation, ET₀ = 0.0023 Ra (Tmean + 17.8) (Tmax − Tmin)^0.5 with Ra as extraterrestrial radiation in evaporation-equivalent mm day⁻¹, as a fallback once checked against Penman-Monteith for the region.
The Class A pan is a circular galvanised or stainless pan about 121 cm in diameter and 25 cm deep, set on a wooden frame above the ground and filled to within a few centimetres of the rim. Because a pan stores and exchanges heat differently from a crop, pan evaporation is converted to ET₀ with a pan coefficient: ET₀ = Kp × Epan. Kp depends on the pan's surroundings (green crop or dry fallow upwind and the fetch distance), on relative humidity and on wind, and FAO-56 tabulates values from about 0.35 to 0.85. In India the IW/CPE scheduling method uses pan evaporation directly. Lysimeters are the reference measurement for ET: a block of soil with a crop, isolated from the surrounding soil, whose water balance is measured either by weighing (weighing lysimeters, which can resolve hourly changes) or by measuring drainage (non-weighing or drainage lysimeters, which give ET over longer periods). Eddy covariance, Bowen ratio and scintillometer systems measure ET from turbulent fluxes, and satellite energy-balance methods estimate ET over large areas; all of these are research or regional tools rather than field scheduling tools.
Crop evapotranspiration under standard conditions is ETc = Kc × ET₀. The crop coefficient Kc captures the differences between the crop and the reference grass: crop height (which affects aerodynamic roughness), albedo, canopy resistance and the amount of exposed wet soil. Kc changes over the season. FAO-56 divides the season into four stages: initial (planting to about 10 % ground cover), crop development (to effective full cover, often near flowering), mid-season (full cover to the start of maturity) and late season (maturity to harvest). Three values define the curve: Kc ini, Kc mid and Kc end. The curve is flat at Kc ini through the initial stage, rises linearly through development, is flat at Kc mid, and falls linearly (or stays level for crops harvested fresh) to Kc end. For many full-cover field crops Kc mid lies roughly between 1.0 and 1.2, but the tabulated values in FAO-56 and ASCE manuals, not a single number, should be used, and stage lengths should come from local planting dates and varieties.
The FAO-56 tabulated values assume a sub-humid climate with minimum relative humidity of about 45 % and a 2 m wind speed of about 2 m s⁻¹. For Kc mid and Kc end above 0.45 in other climates, FAO-56 adjusts by adding [0.04 (u₂ − 2) − 0.004 (RHmin − 45)] (h/3)^0.3, where h is mean plant height in metres. Tall crops in dry, windy climates therefore get a higher Kc, because they are better coupled to the dry air. Kc ini is dominated by soil evaporation; it depends on how often the soil surface is wetted, how deeply, and on the evaporative demand, and FAO-56 gives charts for it. When wetting is frequent, Kc ini can approach or exceed 1.0; when the surface is dry most of the time it may be well below 0.5. The dual approach splits Kc into a basal crop coefficient Kcb, representing transpiration with a dry soil surface, and a soil evaporation coefficient Ke: ETc = (Kcb + Ke) ET₀. It needs a daily water balance of the surface soil layer and is used for high-frequency irrigation (drip, frequent sprinkling), for research, and where the single Kc would be misleading after each wetting. When soil water falls below the readily available level, transpiration is reduced by a water stress coefficient Ks: ETc adj = Ks Kc ET₀ (or Ks Kcb + Ke in the dual form). In FAO-56, Ks = (TAW − Dr) / ((1 − p) TAW) when root zone depletion Dr exceeds RAW, and Ks = 1 otherwise. Salinity reduces Ks further. Ks is the formal link between Chapter 1 and scheduling: keeping depletion within RAW keeps Ks at 1.
Crop water requirement is the depth of water needed to meet ETc of a disease-free crop over its season. The net irrigation requirement is what irrigation must supply after allowing for other sources: NIR = ETc − Pe − Ge − ΔSW, where Pe is effective rainfall, Ge is capillary contribution from a shallow water table and ΔSW is useful soil water stored at planting. Special needs are added: pre-sowing irrigation, leaching to control salts, frost protection, and, for transplanted rice, water for nursery, land soaking and puddling, standing water, and seepage and percolation losses. The gross irrigation requirement divides the net requirement by the application efficiency, and the requirement at the headworks further divides by conveyance and distribution efficiencies. Summing ETc day by day or stage by stage is standard; FAO's CROPWAT software performs the calculation using Penman-Monteith ET₀, single Kc curves and effective rainfall. For project design the peak-period ETc, not the seasonal total, sizes canals, pipes and pumps, and it should be taken from the design year (a dry year of chosen probability), not an average.
Effective rainfall is the part of rainfall that is stored in the root zone and used by the crop. The remainder is lost to surface runoff, to deep percolation below the root zone, or is intercepted and evaporated from the canopy without replacing transpiration. The definition depends on the purpose. To an agronomist, rain that leaches salts may be useful even though it passes the root zone; to a hydrologist, runoff is not lost at all but becomes someone else's supply. In irrigation design it means rain that reduces the irrigation needed. For lowland rice it includes rain retained in the bunded field, net of overflow and percolation. Effectiveness depends on rainfall intensity and amount (small showers are largely evaporated from the canopy and soil surface; large storms produce runoff and deep percolation); on the soil's infiltration rate and the water-holding capacity remaining when the rain falls (rain on a dry profile is more effective than rain just after irrigation); on slope, surface cover and bunding; on crop cover and root depth; and on how irrigation is managed. A scheduling rule that keeps the soil near field capacity leaves little room to store rain and therefore lowers effective rainfall. This is why good schedulers leave some storage unfilled when rain is likely.
Direct measurement uses a field water balance or lysimeters and is rarely practical. Empirical methods are used instead. The fixed-percentage method takes a set fraction, often 70–90 %, of monthly rainfall, with the fraction chosen locally. The dependable rainfall method uses rainfall at a stated probability of exceedance, commonly 80 %, obtained from a frequency analysis, and then takes a fraction of it; this is appropriate for design because it guards against dry years. The USDA Soil Conservation Service (now NRCS) method relates monthly effective rainfall to monthly rainfall and monthly ETc, with a correction for the net irrigation depth. A simplified form of it, used in FAO's CROPWAT, is: Pe = P (125 − 0.2 P) / 125 for monthly P up to 250 mm, and Pe = 125 + 0.1 P for P above 250 mm. The FAO/AGLW formula and the USBR tables are other empirical options. All empirical methods are calibrated in particular climates and should be checked locally. For project planning, effective rainfall is subtracted from ETc month by month or decade by decade to give net irrigation requirement, and dependable rather than mean rainfall is used for capacity. For real-time scheduling it is better to track the root zone water balance day by day, adding measured rain up to the storage available and treating any excess as runoff or deep percolation; this implicitly computes effective rainfall. In monsoon climates, most rain may fall in a few months while the dry-season crop receives almost none, so effective rainfall for the rabi or dry season crop may be near zero even where the annual total is large.
Scheduling answers two questions: when to irrigate and how much to apply. The engineering answer is a soil water balance. Starting from a known depletion, each day's depletion increases by ETc and decreases by effective rain and irrigation: Dr,i = Dr,i−1 − Pe − I + ETc + DP, with deep percolation DP arising only when the root zone is refilled beyond field capacity. Irrigation is triggered when depletion reaches the allowable limit (RAW, or MAD × TAW), and the net depth applied is normally the depletion at that time, returning the root zone to field capacity. The interval between irrigations at peak demand is then the net depth divided by ETc. If the soil water is known on a mass basis, the net depth is dn = (θfc − θi) × (ρb/ρw) × D, with θ as mass fractions and D the root depth; with volumetric contents the density ratio is dropped. The gross depth is dn / Ea, where Ea is the application efficiency.
Four relations recur. First, net depth from soil data, as above. Second, interval = RAW / ETc, rounded down to whole days. Third, the basic irrigation equation Q t = A d: a flow Q running for time t covers area A to depth d. In SI, a flow of 1 L s⁻¹ running one hour delivers 3.6 m³, and 1 ha-cm is 100 m³, so the hours needed to apply d cm on A hectares with Q in L s⁻¹ are t = 100 A d / (3.6 Q) ≈ 27.8 A d / Q. In US units, one cubic foot per second running one hour gives almost exactly one acre-inch (0.992; 1 acre-inch per hour is 1.008 cfs). Fourth, the area a source can command when it runs a fixed number of hours per day: A = Q × hours per day × interval / gross depth. A tubewell of 30 L s⁻¹ running 16 h a day for a 10-day interval with 80 mm gross depth delivers 30 × 3.6 × 16 × 10 = 17,280 m³ per cycle, enough for 17,280 / 0.08 = 216,000 m², or 21.6 ha.
Soil-based criteria use measured soil water content or tension: irrigate when tensiometers at a chosen depth read a set suction, or when a probe shows a set depletion. Suitable trigger tensions vary widely by crop and soil, from a few tens of kPa for sensitive vegetables to several hundred kPa for tolerant field crops on clays (the latter beyond tensiometer range). Climate-based criteria use the water balance computed from ET₀ and Kc, or, in India, the IW/CPE ratio: irrigation water depth divided by cumulative pan evaporation since the last irrigation. With a fixed irrigation depth of 60 mm and an IW/CPE ratio of 0.75, irrigation is given when cumulative pan evaporation reaches 60 / 0.75 = 80 mm. Plant-based criteria use visual symptoms, leaf water potential, stomatal conductance, sap flow or canopy temperature (the crop water stress index compares canopy-air temperature difference with non-stressed and fully stressed baselines). Critical growth stages are those where stress causes disproportionate yield loss: for wheat in the Indian recommendation, crown root initiation about three weeks after sowing is the most critical; flowering and grain filling are critical for most cereals; for maize, tasseling and silking; for rice, panicle initiation to flowering. When water is short, it is better to meet the crop's needs in its critical stages and allow controlled deficits elsewhere; deficit irrigation accepts some yield loss per hectare to spread water over more area or save it for other uses. FAO Irrigation and Drainage Paper 33 relates relative yield decrease to relative ET deficit through the yield response factor Ky: 1 − Ya/Ym = Ky (1 − ETa/ETm). Crops or stages with Ky above 1 are more sensitive than proportional. In canal systems farmers cannot irrigate on demand; supply arrives by rotation. Scheduling then becomes the reverse problem: given a fixed turn and depth, choose crops and areas whose peak needs fit the rotation interval. Modern practice combines weather-station ET, soil moisture sensors and simple software or mobile advisories, but the underlying balance is unchanged.
Water that is not measured cannot be managed or fairly shared. Measurement is needed to charge for water, to check that a canal delivers what the schedule says, to compute efficiencies and to set stream sizes on fields. Methods fall into four groups. Volumetric methods time the filling of a container of known volume and suit small flows. Velocity-area methods multiply cross-sectional area by mean velocity measured with a current meter, an acoustic Doppler instrument or, crudely, floats; surface float velocity is multiplied by a coefficient (commonly around 0.8–0.9) to estimate mean velocity. Constriction or head-discharge devices (weirs, flumes, orifices) create a known relation between upstream head and discharge. Pipe devices include propeller meters, differential-pressure meters (venturi, orifice plate), electromagnetic and ultrasonic meters, and the trajectory or coordinate method for a free jet from a pipe. Whatever the method, accuracy depends on calibration and upkeep: a device that silts up, leaks round its sides or is read at the wrong point gives numbers that look precise but are wrong.
A weir is a notch over which water flows; the discharge depends on the head H measured above the crest at a point upstream, at a distance of at least about four times the maximum head, where the water surface is not yet drawn down. The theoretical discharge over a rectangular weir of length L is Q = (2/3) Cd √(2g) L H^1.5. For a sharp-crested rectangular weir with end contractions the Francis formula in SI is Q = 1.84 (L − 0.1 n H) H^1.5, with Q in m³ s⁻¹, L and H in m, and n the number of end contractions (two for a fully contracted weir, zero for a suppressed weir spanning the channel). The Cipolletti weir is trapezoidal with side slopes of 1 horizontal to 4 vertical, chosen so that the extra area of the sloping sides compensates for end contraction; its SI formula is Q = 1.86 L H^1.5. The triangular or V-notch weir measures small flows accurately because the flow area grows rapidly with head: Q = (8/15) Cd √(2g) tan(θ/2) H^2.5, which for a 90° notch with Cd about 0.58 becomes approximately Q = 1.38 H^2.5 (SI). Weirs need free fall with air under the nappe, a crest height and approach channel large enough to keep approach velocity low, and a sharp upstream edge. Their main drawbacks are head loss and silt deposition behind the crest.
A Parshall flume is a critical-depth flume with a converging inlet, a throat with a downward-sloping floor and a diverging outlet; its size is named by throat width W. Flow passes through critical depth in the throat, so under free flow a single head reading Ha at a specified point in the converging section gives the discharge by Q = K Ha^n, with K and n depending on throat width; for flumes from 1 to 8 ft the US customary formula is Q = 4 W Ha^(1.522 W^0.026), with Q in ft³ s⁻¹ and W and Ha in ft. When downstream water rises, a second head Hb is read in the throat. The ratio Hb/Ha is the submergence. Free flow holds up to a transition submergence that rises with size: 50 % for the smallest flumes (1–3 inch), 60 % for 6 and 9 inch, 70 % for 1–8 ft and 80 % for larger flumes. Above it, a submergence correction from standard tables is subtracted. Advantages: small head loss (about a quarter of a weir's for the same flow), self-cleaning of sediment, and accuracy over a wide range. Disadvantages: exact construction and higher cost. Long-throated flumes and broad-crested weirs (ramp flumes) designed with software from the US Bureau of Reclamation and partners are now widely used because they can be fitted into existing canals and rated from theory. Cutthroat flumes omit the throat section and are simpler to build.
An orifice is an opening of known area below the water surface. Q = Cd A √(2 g h), where h is the head on the centre of a free orifice or the difference between upstream and downstream water levels for a submerged orifice. For a sharp-edged, fully contracted orifice Cd is about 0.6. Submerged orifices are used as turnouts and meter gates where little head is available. Siphon tubes over a field channel bank are small orifices, and their discharge is found from the head difference and the tube diameter. Pipe flow from a partly full horizontal pipe can be estimated by the California pipe method, and from a full pipe discharging horizontally by the coordinate method using the horizontal and vertical drop of the jet. Standard references are the US Bureau of Reclamation Water Measurement Manual and ISO standards for weirs and flumes; any device should be installed as the reference specifies, since installation errors are the largest source of measurement error in the field.
Irrigation channels are designed for steady uniform flow, in which depth, area and velocity do not change along the reach and the water surface is parallel to the bed. Manning's equation in SI units is V = (1/n) R^(2/3) S^(1/2), where V is mean velocity (m s⁻¹), R the hydraulic radius (flow area divided by wetted perimeter, in m), S the bed slope and n the roughness coefficient; Q = A V. In US customary units the constant 1.49 appears in the numerator. Typical n values are about 0.011–0.015 for smooth concrete, somewhat higher for rough concrete, brick or stone masonry, around 0.02–0.025 for clean, regular earth channels, and considerably higher for channels with weeds or irregular sections; Chow's Open-Channel Hydraulics tabulates ranges. Because depth appears implicitly, the normal depth for a given Q, n, S and section is found by trial, chart or a small program. The Chezy equation, V = C √(R S), is the older form; C and n are related by C = R^(1/6)/n.
For a given area, the section that minimises wetted perimeter carries the most flow; it is called the best hydraulic or most efficient section. For a rectangle it is a width equal to twice the depth, and for every best section the hydraulic radius equals half the depth. The best trapezoid is half a regular hexagon, with side slopes of 60° to the horizontal, but earth channels need flatter slopes for stability, typically from about 1:1 to 2:1 (horizontal:vertical) depending on soil, so the best trapezoid for a fixed side slope is used: its top width equals twice the length of a side, and R = y/2. In practice, excavation cost, available land, the need for a shallow channel to command fields, and stability all shift the chosen section away from the hydraulic optimum. Small farm channels in many countries are trapezoidal or, where lined with concrete or brick, rectangular or semicircular. Freeboard, the vertical distance from the design water surface to the top of the bank, protects against waves, surges and errors; it increases with discharge and is set by local standards.
An earth channel must not erode at design flow and must not silt up. The maximum permissible (non-erosive) velocity depends on soil texture and cohesion, on sediment in the water and on channel age; tables of permissible velocities and tractive stresses (Fortier and Scobey, Lane) give ranges, and the designer should use local values. The minimum velocity to avoid silting and weed growth is usually a few tenths of a metre per second. Two regime theories developed on the alluvial canals of the Indo-Gangetic plain are standard in South Asian syllabi. Kennedy's theory (1895) defined the critical velocity that neither silts nor scours: V₀ = 0.55 m D^0.64 in SI, where D is depth and m is the critical velocity ratio (1 for the Upper Bari Doab sediment). Lacey's regime theory relates all dimensions to discharge and a silt factor f = 1.76 √d, with d the mean particle size in mm: V = (Q f² / 140)^(1/6), wetted perimeter P = 4.75 √Q, hydraulic radius R = 5V² / (2f), and bed slope S = f^(5/3) / (3340 Q^(1/6)). These are empirical and apply to alluvial channels carrying sediment; for lined or non-alluvial channels, Manning with permissible velocity or tractive force is used.
A channel network needs structures to control levels and flows. Drops (falls) dissipate energy where the ground slope is steeper than the bed slope. Checks and cross-regulators raise the water level upstream so that turnouts can draw water. Turnouts and outlets deliver water from a distributary or watercourse to a farm or field; in the Indian subcontinent, outlets are classed as non-modular (discharge depends on both upstream and downstream levels, such as a pipe outlet), semi-modular (depends on upstream level only, such as the open flume outlet) and modular (independent of both within limits). Culverts and aqueducts carry water across roads and drainage lines. Division boxes split flows between field channels. Energy dissipation at drops is achieved with stilling basins or baffle arrangements, and any change in section needs transitions to avoid scour. Every structure should be sized for the full supply discharge with freeboard, and checked for the lowest operating flow, when levels may fall below turnout sills.
Unlined canals lose water by seepage through their bed and banks. Losses depend on soil permeability, wetted perimeter, water depth, the depth to groundwater, sediment that clogs pores, and the canal's age. In permeable soils a large fraction of the water entering a network can be lost before reaching the fields, although part of it recharges groundwater and may be pumped back elsewhere. Seepage is measured by the ponding method (isolating a reach with temporary dams and recording the fall of the level, corrected for evaporation), which is regarded as the most accurate; by the inflow-outflow method, which suits long reaches but is sensitive to measurement errors; and by seepage meters placed on the bed. Lining reduces seepage and, in addition, lowers roughness so that the same discharge needs a smaller section, allows higher velocities, controls weeds, reduces maintenance and breaches, and reduces waterlogging in the command. Its cost must be weighed against the value of the water saved and the other benefits; an economic analysis compares the annualised cost of lining with the annual value of water saved plus maintenance savings.
Hard-surface linings include cast-in-place concrete, precast concrete slabs, shotcrete, brick or tile masonry, stone masonry, and asphaltic concrete. Concrete is durable and smooth but must be protected against cracking by contraction and expansion joints filled with a sealant, and against uplift. Brick lining is common in South Asia where bricks are cheap and labour available. Earth linings use compacted earth, soil-cement, or clay and bentonite blankets to reduce the permeability of the bed. Buried membrane linings use a flexible plastic film such as low-density polyethylene or other geomembranes covered by a protective layer of earth or concrete; the cover protects the film from sunlight, animals and machinery. Exposed membranes are sometimes used for farm ponds. Choice depends on the availability and cost of materials, labour, canal size, soil conditions, climate (freezing damages rigid linings) and required life. Lined channels are usually designed as trapezoidal or, for small sizes, rectangular or curved-bottom sections with the lower roughness of the lining material. Because the lining cannot erode, velocity may be higher than in earth, limited mainly by the need to avoid supercritical flow and damage to joints. When the water table rises above the canal bed or the canal is drained quickly, water behind the lining exerts uplift pressure that can crack or lift panels; the remedies are under-drainage (porous filter layers with longitudinal drains), weep holes and pressure-relief valves. Expansive clays need a non-expansive bedding. Freeboard on a lined canal is often divided into lining freeboard (height of lining above water) and bank freeboard. Earthwork beneath the lining must be compacted to avoid settlement.
A low-pressure buried pipeline can replace open field channels and watercourses for surface irrigation. Water from a pump or canal outlet enters through an inlet stand (a vertical riser) that provides the head, allows air to escape and damps surges; the pipeline then runs below plough depth to risers with outlet valves (commonly called alfalfa valves or orchard valves) that discharge into borders, basins or furrows, or feed gated pipe. Common materials are rigid PVC, high-density polyethylene and unreinforced or reinforced concrete; asbestos-cement is no longer used in many countries because of health rules. Benefits: almost no conveyance loss, no land lost to channels, no weeds, easy crossing of uneven ground, and better control of flow. Design uses the irrigation stream size as the pipe flow and the hydraulic grade line from the inlet to each outlet. Friction loss is computed with the Hazen-Williams formula, hf = 10.67 L Q^1.852 / (C^1.852 D^4.87) in SI with C about 140–150 for plastic pipe, or with Darcy-Weisbach, hf = f (L/D) V² / (2g). The available head at every outlet must exceed the losses; stand heights are set so that the hydraulic grade line stays below the stand tops. Protection includes air release valves at high points, vacuum relief, pressure relief, drain valves at low points and thrust blocks at bends. Water hammer from sudden valve closure raises pressure by ΔP = ρ c ΔV (Joukowsky), where c is the wave speed; slow closure and design velocity limits (guidance in the USDA NRCS practice standard for plastic irrigation pipelines commonly uses about 1.5 m s⁻¹, 5 ft s⁻¹) control it.
In surface irrigation water flows over the field by gravity and infiltrates as it goes. Every event has four phases: advance, as the front moves down the field; storage or ponding, after the front reaches the end and inflow continues; depletion, after inflow is cut off, as the surface water recedes at the upstream end; and recession, as the water disappears along the field. The time a point is covered is its infiltration opportunity time, the difference between recession and advance times at that point. Uniformity requires the opportunity time to be similar along the field, which means advance must be fast compared with the time needed to infiltrate the required depth. Criddle's quarter-time rule of thumb says the advance time to the end should be about a quarter of the required opportunity time. Design variables are the field length and slope, the inflow rate (stream size) and the cut-off time; the soil's infiltration function and the surface roughness (Manning n for the surface) are the controlling properties. Volume balance models, zero-inertia and full hydrodynamic models are used for design; the USDA software WinSRFR is a standard tool. Laser-guided land levelling to a precise grade or to level is the single most effective way to improve all surface methods.
A border is a long, usually rectangular strip bounded by low dikes (levees), with a gentle slope along its length and no cross slope. Water enters at the upper end and flows down as a sheet. It suits close-growing crops such as wheat, pasture, alfalfa and small grains on medium soils. Border width is set to a multiple of implement widths and to the ability of the stream to spread across the strip; length depends on soil intake rate, slope and stream size, being shorter on sandy soils. The stream size per unit width must be large enough to advance quickly and spread across, but not so large that it erodes the surface or overtops the dikes; the maximum non-erosive stream falls as slope increases, and is larger for dense sod crops than for clean-tilled soil. In graded borders with free outflow, inflow is cut off before the front reaches the end, so that the water already on the surface completes the irrigation during recession; cutting too late produces runoff, too early leaves the lower end under-irrigated. Blocked-end borders avoid runoff but pond water at the lower end.
A basin is a level area surrounded by dikes and filled rapidly; water spreads in all directions and infiltrates with almost no runoff. Basins suit rice (which needs standing water), fodder and field crops on level land, and orchards, where individual trees may be ringed (ring basins) or rows bunded (check basins). They work on a wide range of soils if the stream is large enough; level basins are especially suited to fine-textured, low-intake soils and are ineffective where intake is very high unless basins are small. Basin size is governed by stream size, intake rate and the depth to be applied: for a fixed stream, sandier soils need smaller basins. With precise levelling and a large inflow, level basins can reach very high application efficiency and uniformity, and they tolerate simple management. Drawbacks: dikes interfere with machinery, crops sensitive to waterlogging suffer on heavy soils, and levelling cost can be significant. For rice, the basin also holds rain, which raises effective rainfall.
Furrows are small channels between crop rows; water flows along them and moves sideways and downward into the ridges. They suit row crops (maize, cotton, sugarcane, potatoes, vegetables) and avoid wetting the whole surface, which reduces crusting and evaporation. Furrow spacing is set by crop row spacing and the lateral spread of water, which is wide in clays and narrow in sands. Slopes are usually gentle, and on steeper land furrows may follow the contour. The maximum non-erosive stream is estimated by the empirical relation Qmax = 0.6 / S, with Q in litres per second and S the slope in percent (about 10 / S in US gallons per minute). Techniques that improve furrow performance include cutback irrigation (a large stream for fast advance, then reduced to match intake and limit runoff), surge irrigation (intermittent flow in on-off cycles, which reduces intake of the wetted reach and speeds advance on many soils), alternate-furrow irrigation, tailwater recovery and reuse systems, and gated pipe or siphons for even distribution. The irrigation time for a furrow follows from volume: t = d w L / q, where d is the gross depth, w the furrow spacing, L the length and q the furrow stream. Corrugations are small, closely spaced furrows used to guide water on close-growing crops.
Efficiency terms compare volumes of water; each must state its boundary. Water conveyance efficiency is the ratio of water delivered to the field (or farm) to water diverted at the source: ηc = Wf / Wr. Water application efficiency is the ratio of water stored in the root zone to water delivered to the field: Ea = Ws / Wf, the losses being surface runoff (tailwater) and deep percolation. Water storage efficiency (also called requirement efficiency) is the ratio of water stored to water needed in the root zone before irrigation: Es = Ws / Wn; it shows whether the irrigation filled the root zone. Water use efficiency is used in two senses: crop water use efficiency (yield per unit ET) and field water use efficiency (yield per unit water applied). Project or overall efficiency is the product of conveyance, distribution and application efficiencies. The International Commission on Irrigation and Drainage (ICID) and ASCE task committees have recommended clearer performance terms, notably irrigation efficiency (volume of irrigation water beneficially used divided by volume applied minus change in storage) and irrigation sagacity (including reasonable non-beneficial uses).
Efficiency says how much water went where it was wanted on average; uniformity says how evenly it was spread. Christiansen's coefficient of uniformity, developed for sprinklers but used generally, is CU = 100 (1 − Σ|x − x̄| / (n x̄)), where x are individual depths, x̄ their mean and n the number of observations. Distribution uniformity of the low quarter is DU = 100 × (mean of the lowest quarter of depths) / (mean of all depths); the Irrigation Association and USDA NRCS use it widely, and for surface irrigation it is estimated from infiltrated depths along the field. The older 'water distribution efficiency' in Indian textbooks, Ed = 100 (1 − y/d), where y is the average numerical deviation from the mean depth d, is mathematically the same as Christiansen's CU. For a normal distribution of depths, CU and DU are approximately related by CU ≈ 100 − 0.63 (100 − DU).
With non-uniform application, a manager must choose how much of the field to leave under-irrigated. To make sure the low quarter receives the required depth, the average applied depth must be the required depth divided by the low-quarter DU, and the extra becomes deep percolation. This is why application efficiency in the low quarter, AELQ, equals the average low-quarter depth infiltrated (up to the requirement) divided by the average depth applied; for a system with no runoff and adequate low-quarter application, it is approximately DU. High efficiency can be achieved with poor uniformity simply by under-irrigating, so efficiency figures must always be read together with adequacy or storage efficiency. In surface irrigation, low uniformity comes from the difference in opportunity time between head and tail, from uneven land and from soil variability; in pressurised systems from pressure variation, emitter or nozzle variation and wind. Water 'lost' from one field as runoff or deep percolation is often recovered downstream or by wells. Improving field efficiency therefore does not always save water at basin scale; it may reduce return flows others depend on while allowing more ET from an expanded area. The distinction between consumptive use (ET, water that leaves the system as vapour) and non-consumptive losses (recoverable return flows versus flows to saline sinks) is central to water accounting. For a practitioner, the lesson is to state which efficiency is meant, over what boundary, and whether saved water is really saved.
The gross command area (GCA) of a canal is all the land that could be irrigated from it by gravity; the culturable command area (CCA) excludes roads, settlements, ponds and barren land; the area actually irrigated in a season, as a percentage of CCA, is the intensity of irrigation. Delta (Δ) is the total depth of water supplied to a crop over its base period B (the time between the first and last watering). Duty (D) is the area that one unit of continuous flow can irrigate over the base period, usually in hectares per cumec. The relation is Δ = 8.64 B / D, with Δ in metres, B in days and D in ha per m³ s⁻¹ (because one cumec for one day is 86,400 m³). Duty rises (more area per cumec) as losses fall and methods improve, and it must be quoted at a stated point: duty at the field is higher than duty at the head of the canal, because conveyance losses lie between them. The capacity of a channel follows from the area it must serve at peak demand divided by the duty, plus allowances. Time factor is the ratio of days the canal actually runs to days in the period; capacity factor is the ratio of mean supply to full supply capacity. Kor watering is the first watering after sowing (for wheat or rice in South Asian usage), whose depth and the kor period over which it must be given often govern canal capacity.
Large canal projects have frequently irrigated less area than planned, with high losses, waterlogging and salinity in the head reaches and scarcity at the tail. Causes include incomplete field channels and drains, unlevelled land, rigid or unreliable deliveries, weak water rights, and farmers growing water-intensive crops near the head. India launched the Centrally Sponsored Command Area Development Programme in 1974–75 to close this gap. It was restructured as Command Area Development and Water Management (CADWM) in 2004 and has been implemented under the Pradhan Mantri Krishi Sinchayee Yojana (Har Khet Ko Pani component) since 2015–16. Its typical components are on-farm development works (field channels, field drains, land levelling and shaping, and realignment of field boundaries), warabandi, conjunctive use of surface and groundwater, adaptive trials and training, correction of system deficiencies, and support to water users' associations. Other countries have similar programmes under different names; the Philippines, for example, relies on irrigators' associations in national irrigation systems.
Canal water can be delivered continuously, on demand, or by rotation. Rotation is the rule on large gravity systems in South Asia. Warabandi, used in the Indus basin of north-west India and Pakistan, is a fixed rotation in which each farmer on an outlet gets the full outlet flow for a time proportional to his landholding, on a fixed day and hour each week; the roster is public, and the system requires no discretion by officials. It is equitable and simple but not responsive to crop needs, so farmers adjust by choosing crops and by using groundwater. Osrabandi and other rotations operate at distributary level. Modern systems with lined canals, automatic downstream control, pipelines and measurement structures can offer arranged or demand deliveries, and flexible schedules allow on-farm efficiency improvements to translate into real gains. Where rotation continues, reliability matters more than flexibility: a farmer who knows exactly when water will come can plan sowing, labour and groundwater pumping around it.
Participatory irrigation management (PIM) and irrigation management transfer give farmer organisations responsibility for operating and maintaining parts of a system, usually the minors and watercourses, and for collecting fees. Mexico, Turkey, the Philippines and many Indian states have enacted laws or policies for water users' associations; outcomes vary with the legal backing, the financial viability of the associations and the reliability of the main system above them. Conjunctive use of canal water and groundwater is the planned use of both to meet demand: canal recharge raises water tables that farmers tap in dry periods, and pumping lowers water tables in waterlogged areas. Planning a command therefore requires hydrological analysis of recharge, drainage and salt balance, not only of canal capacity. Monitoring with remote sensing (actual ET, cropping pattern, irrigated area) is now a routine tool for evaluating command performance, alongside indicators such as relative water supply, delivery performance ratio and equity between head and tail.
Learn to move between mass and volume water contents, between depth and volume (1 ha-cm = 100 m³), and between flow and time (Q t = A d) without error; most examination marks in this subject are lost on unit slips, not on concepts. Work one full FAO-56 ET₀ calculation by hand from a day's weather so the equation stops being a black box, then use CROPWAT or a spreadsheet. Build a Kc curve for a local crop and a season's water balance, and compute net irrigation requirement with and without effective rainfall. In the field laboratory, run a double-ring infiltrometer test, calibrate a V-notch against a bucket and stopwatch, and measure advance on a furrow. These few exercises cover most of what GATE, ICAR and licensing papers ask.
At five years an engineer designs and checks systems: sets border and furrow lengths, stream sizes and cut-off times for a farm; sizes a buried pipeline and its stands; specifies a Parshall or long-throated flume for a turnout and writes its rating; prepares an estimate for field channels and lining under a government scheme; and advises farmers on schedules using pan or weather-station ET and soil moisture sensors. The typical decisions are practical trade-offs: laser levelling versus shorter runs, lining a watercourse versus burying pipe, a fixed rotation versus sensor-based timing. Keep field records of inflow, advance and runoff; they are what later justify a design change.
A senior engineer with twenty-five years plans and manages commands: fixes canal capacities from crop plans and duty, sets rotation schedules that are equitable between head and tail, decides where lining pays and where seepage is recharging wells that farmers rely on, and organises water users' associations to take over minors and watercourses. The work is as much institutional as hydraulic: water rights, fee collection, maintenance budgets, and resolving disputes with measured data. Performance indicators such as relative water supply and delivery performance ratio, together with satellite estimates of ET and cropped area, are the tools for showing where a system is failing and for defending investment.
A researcher with long field experience works on the open questions: how to model surface irrigation with spatially variable infiltration and use the models for real-time control; how to estimate actual ET at field scale from satellites and sensor networks and assimilate it into scheduling; how to account for water at basin scale so that efficiency gains are real savings rather than transfers; how deficit irrigation and crop response functions behave under climate change; and why irrigation management transfer succeeds in some systems and fails in others. The standards of evidence are long-term, measured water balances and yields, not single-season trials.
Sprinkler and drip design, agricultural drainage, salinity management and groundwater pumping are covered in their own subjects and only touched on here. The manual does not tabulate crop-specific Kc values, stage lengths, depletion fractions, permissible velocities, freeboard or maximum border stream sizes, because these depend on crop, soil and national standard and must be taken from FAO-56, NRCS or the local code. Full surface irrigation hydraulic models (zero-inertia, hydrodynamic) are described but not derived. Country-specific water law and scheme details beyond India's CAD programme are only outlined, and scheme names and funding rules should be checked against current government guidelines.
Original practice questions written to the syllabus and pattern of the exams named in each question's examTags. They are not past papers and do not reproduce any board's questions.
These notes are free to read here, with no account.
What the Laisha app adds:
Free for ten days, then a subscription. We would rather write that here than have you find it later.