A. Agronomic factors

A.1 Growing cycle and growing period
A.2 Radiation
A.3 Temperature
A.4 Rooting
A.5 Aeration
A.6 Water quantity
A.7 Nutrients (NPK)
A.8 Water quality
A.9 Salinity
A.10 Sodicity
A.11 pH, micronutrients and toxicities
A.12 Pests, diseases and weeds
A.13 Flood, storm, wind and frost

Crop Requirements and Limitations

The Crop Environment

Factors that are principally agronomic are discussed under thirteen headings in this section; these are: growing cycle and growing period; radiation; temperature; rooting; aeration; water quantity; nutrients (NPK); water quality; salinity; sodicity, boron and chloride toxicities; pH, micronutrients and other toxicities; pests, diseases and weeds; flood, storm, wind and frost. Some aspects impinging on management are inevitably included; in such cases these management considerations are excluded from Section B (Management).

A.1 Growing cycle and growing period

A.1.1 Critical limits of growing period

The growing cycle is the period required for an annual crop to complete its annual cycle of establishment, growth and production of harvested part. Perennial crops have growing cycles of more than one year.

The growing period for annual crops is the duration of the year when temperature, soil. water supply and other factors permit crop growth and development.

Thus, a growing cycle is a property of the crop (i.e. a crop requirement) whereas a growing period is a condition of the land (i.e. a land quality or land characteristic).

The growing period is a major determinant of land suitability for crops and cultivars on a worldwide and continental scale as described in FAO, 1978a. Tables 32 and 33 illustrate the concept. Within a project development area, growing periods often vary due to variable temperature, water supply and rainfall characteristics.

In subtropical and temperate climates, there are winter and summer growing periods due to seasonal temperature changes. For example, in lower Egypt, temperate crops such as berseem clover, wheat, barley and beans are grown in the winter and crops with higher temperature requirements such as cotton, rice and maize are grown during the summer. Normally, these seasonal temperature variations will not be class-determining. Thus LUTs can be described for the rotational cropping system spanning the winter and summer growing periods. In hilly areas where temperatures vary with altitude, or where frost occurs in valley bottoms, the growing period may be considered 'class-determining', also where water supplies vary from place to place.

The period during which irrigation water supplies are available determines the growing period in many countries. Irrigation projects in south east Asia, or in Middle Eastern countries relying on run-of-river or spate storm flows can include land on which the growing period and the period of irrigation may vary from 12 months of the year to a few months a year. Such variations may be 'class-determining' where they exist within one project area.

Growing periods can be constrained by wet or humid conditions that limit opportunities for ripening and drying the crop, or which lead to problems of quality (e.g. reduced sugar content of sugarcane, staining of cotton, blemishes on fruits, etc.).

Table 32 MAJOR CLIMATES

Compare Table 33

Climate		Major climates during growing period	24 hr mean (daily) temperature (C) regime during the growing period	Suitable for consideration for crop group (Table 33)
	No.	Descriptive name
Tropics All months with monthly mean temperatures, corrected to sea level, above 18°C	1	Warm tropics	More than 20	II and III
	2	Moderately cool tropics	15-20	I and IV
	3	Cool tropics	5/10 - 15	I
	4	Cold tropics	Less than 5	Not suitable
Subtropics One or more months with monthly mean temperatures, corrected to sea level, below 18°C but all months above 5°C	5	Warm/moderately cool subtropics (summer rainfall)	More than 20	II and III
	6	Warm/moderately cool subtropics (summer rainfall)	15 - 20	I and IV
	7	Warm subtropics (summer rainfall)	More than 20	II and III
	8	Moderately cool subtropics (summer rainfall)	15 - 20	I and IV
	9	Cool subtropics (summer rainfall)	5/10 - 20	I
	10	Cold subtropics (summer rainfall)	Less than 5	Not suitable
	11	Cool subtropics (winter rainfall)	5/10 - 20	I
	12	Cold subtropics (winter rainfall)	Less than 5	Not suitable
Temperate One or more months with monthly mean temperatures, corrected to sea level, below 5 C	13	Cool temperate	5/10 - 20	I
	14	Cold temperate	Less than 5	Not suitable

Source: FAO 1980c, p. 355; Higgins and Kassam 1981.

Table 33 CROP ADAPTABILITY GROUPS, BASED ON PHOTOSYNTHETIC PATHWAY AND RESPONSE TO RADIATION AND TEMPERATURE

Compare Table 32

Crop adaptability group	I	II	III	IV	V
Photo-synthetic pathway	C3	C3	C4	C4	CAM
Optimum temperature for photosynthesis (°C)	15-20	25-30	30-35	20-30	25-35
	Sugarbeet Phaseolus Wheat Barley Oats Potato Bean (TE) Chickpea	Soybean (TR) Phaseolus Rice Cassava Sweet Potato Yams Bean (TR) Groundnut Cotton Tobacco Banana Coconut Rubber Oil palm	Sorghum (TR) Maize (TR) Pearl millet Panicum Millet (TR) Finger millet Setaria Sugarcane	Panicum Millet (TE, TH) Sorghum (TE, TH) Maize (TE, TH) Setaria	Sisal Pineapple

TE = Temperate cultivars; TR = Tropical (lowland) cultivars; TH = Tropical (highland) cultivars.

Source: Based on information extracted from FAO 1978a and FAO 1980c.

A.1.1 Critical limits of growing period

Method 1: General Method (FAO 1978a)

This approach is useful in reconnaissance studies or low intensity investigations especially where the need for irrigation is to be evaluated. It is based on mean daily temperature (T), precipitation (P), and the potential evapotranspiration (PET) to produce data such as is illustrated in Figure 6. Data for 10-day periods or months can be used, or monthly data can be converted into data for 10-day periods. The procedure is:

i. Temperature constraint: The growing period is confined to 10-day periods in which mean daily temperature equals or exceeds a minimum temperature (e.g. 5°C).

ii. Beginning of the growing period: Under rainfed conditions this is taken as the time at which precipitation equals or exceeds half the potential evapotranspiration.

iii. Humid period: Under rainfed conditions a normal growing period must include at least one 10-day humid period defined as a period in which rainfall exceeds potential evapotranspiration.

iv. End of the rains: This can be taken as the time at which precipitation falls below half the potential evapotranspiration.

v. End of growing period: The growing period ends when the reserve of water stored in the soil following the cessation of rainfall and irrigation is depleted.

Figure 6 Examples of four types of growing period (under rainfed conditions which might be modified by irrigation) - Normal

Figure 6 Examples of four types of growing period (under rainfed conditions which might be modified by irrigation) - Intermediate

Figure 6 Examples of four types of growing period (under rainfed conditions which might be modified by irrigation) - All year round humid

Figure 6 Examples of four types of growing period (under rainfed conditions which might be modified by irrigation) - All year round dry

a - Beginning of rains and growing period
b₁ and b₂ - Start and end of humid period respectively
c - End of rains and rainy season
d - End of growing period
P - Precipitation
PET - Potential evapotranspiration

(after FAO 1978a)

Isolines for growing periods on a continental scale are published in the results of the Agro-ecological Zones Project (FAO 1978/80/81). These are generalized to meet the constraints of the scale of publication. For more intensive studies, isolines of growing period may be drawn at 75-day, 90-day and then at 30-day intervals up to 365 days. These can be drawn for the existing agriculture and for the agriculture anticipated after supplemental irrigation. The growing periods may indicate the need for one or two crops in succession, or for different cultivars of the same crop.

Method 2: Rainfall Related Events (Stern and Coe 1982)

This approach is useful if the growing period depends on seasonal and year-to-year variations in rainfall.

This method relies on an analysis of daily rainfall for individual years of the rainfall record. The distinctive feature of the method is that each year provides one number for any event or characteristic of interest. The event is defined by the user, i.e. as a set of rainfall characteristics that could, for example, define a dry period, the start of the rains, the end of the rains, the length of the growing period between the start and end of the rains, or the distribution of rainfall amounts throughout the year. Each event in each year is listed (e.g. as the day or period of occurrence). An estimate of the probability of an event occurring can then be made directly from its relative frequency of occurrence or, alternatively, a distribution (such as the normal) can be fitted.

A further development of this method is to compute a daily soil water balance sheet on the best obtainable information. This balance sheet can show the soil water content between a defined field capacity water content (upper boundary) and permanent wilting point (lower boundary) according to the daily gains and losses of water. Figure 7 illustrates the output of such a method for a permeable soil showing that growing periods under rainfed conditions vary from season to season and year to year.

The use of this method is facilitated by computer technology but the data can be processed equally well by hand (Stern and Coe 1982) if computers are not available. One of the main advantages over more general methods is that conditions in individual extreme years can be identified from the historical rainfall record.

Figure 7 Computed available soil water for 1965-78, based on daily rainfall, field capacity and permanent wilting point of Kilinochchi chromic luvisol, with 304 mm (12 inches) of available water in 250 cm rooting depth, and an evapotranspiration of 5 mm (0.2 inches) per day

Source: Robertson and Eavis 1983

A.2 Radiation

Three relevant aspects of radiation are (i) daylength, (ii) its influence on photosynthesis and dry matter accumulation in crops, and (iii) its effects on evapotranspiration. Radiation levels may also be important in the drying and ripening of crops, but this is evaluated under heading B.17.

Daylength may be a relevant class-determining factor in evaluations carried out at low intensity across different latitudes as already discussed under 'Growing Period' (Tables 32 and 33). Daylength affects photoperiod-sensitive cultivars of crops such as rice, influencing floral initiation and the onset or length of vegetative and reproductive phases of growth and development. The interaction of daylength with water availability or temperature can sometimes prove 'class-determining' at project level (e.g. in influencing the flowering of sugarcane, flowering and fruiting of mangoes, and in the bulbing and ripening of onions, etc.).

The influence of radiation on photosynthesis and dry matter accumulation in crops has been reviewed by Monteith (1972). The relationships for C3 and C4 crops differ as indicated in Table 33.

The components of the radiation balance that may be used to define critical limits are shown in Figure 8, and as follows:

Figure 8 Illustration of the radiation balance

Source: FAO 1977b

i. Extraterrestrial radiation (R_a) received at the top of the atmosphere. This is dependent on latitude and time of year.

ii. Solar radiation (R_s) is that part of the extraterrestrial radiation which is not absorbed and scattered when passing through the atmosphere, together with some of the scattered radiation that also reaches the earth's surface. A proportion of this radiation (about 50%) is photosynthetically active radiation (PAR) (Szeicz 1974). Values of solar radiation can be obtained from direct measurements or approximated by using:

R_s = (0.25 + 0.05 n/N) R_a,

where n is the actual bright sunshine hours (e.g. measured with a Campbell Stokes solarimeter) and M is the maximum possible sunshine hours for a given month and latitude (given in standard tables, e.g. see FAO Irrigation and Drainage Paper No. 24, 1977b).

iii. Net shortwave solar radiation (R_ns): Part of the solar radiation (R_s) is reflected back directly by the soil and crop and is lost to the atmosphere. Reflection depends on the nature of the surface cover and is approximately 5% for water and 20-25% for many green-leafed crops. That which remains after losses by reflection is the net shortwave radiation, R_ns.

To obtain the net shortwave radiation, the solar radiation must be corrected for the reflectivity of the crop surface with:

R_ns = (1 - a) R_s

For example, for green crops covering the ground, the value of a is taken as 0.25 and R_ns is 75% of the solar radiation.

iv. Net longwave radiation (R_nl,) is the difference between outgoing and incoming longwave radiation. Outgoing radiation is usually greater than incoming longwave radiation so that additional losses at the earth's surface occur because of longwave radiation; thus net longwave radiation represents an energy loss. This can be determined from temperature, vapour pressure (ed), and the ratio n/N (actual to possible hours of bright sunshine). Values for the functions f(t), f(ed), and f(n/N) are given in FAO Irrigation and Drainage Paper No. 24, Tables 15, 16 and 17.

v. Total net radiation (R_n) is equal to the difference between net shortwave radiation and net longwave radiation so that:

R_n = R_ns - R_nl

Total net radiation is used in estimating losses of water by evaporation. The unit of Rn =1 cal/cm²/min is approximately equivalent to the energy required to evaporate 1 mm of water per hour. Radiation in S1 units is given as mW/m² and 1 mW/m² is required to evaporate 830 nun/day.

The vegetative growth of most plants increases linearly with solar radiation up to a limit beyond which no further increase occurs. In many tropical areas, water shortages rather than radiation limit growth and the radiation-limited potential is not attained. However, marked seasonal effects on yields may be evident (e.g. in the Philippines and other south east Asian countries, irrigated modern rice varieties plentifully supplied with water, yield more in the dry season than in the overcast wet season). In temperate countries, radiation is one of the most dominant growth-limiting factors in winter months and land characteristics such as aspect may be used to define critical limits, as appropriate.

Tables for relating radiation data and sunshine hours to dry matter production rate of a 'standard' crop are given by FAO (1978, 1980, 1981) and Doorenbos and Kassam (in FAO 1979a). These figures, together with temperature and crop phenological data, can be used as a basis for calculating biomass converted to crop yields with respect to radiation for given areas.

A.3 Temperature

Temperature has already been discussed under A.1 Growing Cycle and Growing Period. The growth of most crops ceases below a critical low temperature and very high temperatures (usually above 30 - 35 C) have adverse effects. Crops are divided into five adaptability groups on the basis of their photosynthetic carbon assimilation pathways (C3, C4 or CAM) and according to the effects of radiation and temperature on photosynthesis (see Table 33). Between the minimum temperature for growth and the optimum temperature for photosynthesis, the rate of growth increases more or less linearly with temperature; the growth rate then reaches a plateau within the optimum temperature range before falling off at higher temperatures, Temperature interacts with radiation; the highest potential for growth is achieved with both radiation and temperatures in the optimal range.

In many temperate climates and at high altitudes in tropical countries, the temperature for growth is below optimum during part of the growing season.

Critical limits to define s1, s2, s3, n1 and n2 levels of suitability can be specified in terms of ranges of temperature in C, or other units. For example, heat units are sometimes used to indicate seasonal conditions for certain crops, such as cotton, in units of day-degrees accumulated over a growing season.

A.4 Rooting

Plants require water and nutrients, which are conveyed from the soil to the productive parts of the plant through roots. If root growth, or the development or function of a root system is impaired by adverse land characteristics, the growth and yield of the crop may likewise be impaired from lack of water or nutrients.

The growth and distribution of roots in the soil may be affected by;

i. the supply of assimilates (sugars, etc.) from the shoots and leaves to the roots;

ii. soil temperatures;

iii. soil water;

iv. soil nutrients and the chemical environment including salinity, sodicity, pH, micronutrient deficiencies and toxicities;

v. the supply of oxygen to meet oxygen requirements for respiring root tissues:

vi. mechanical impedance to root penetration;

vii. pests and diseases of the root system.

Because of their separate importance in land evaluation, it is convenient to evaluate some of these factors under different headings. Therefore, soil aeration is discussed under A.5, water under A.6, salinity under A.9, sodicity under A.10, pH, micronutrient deficiencies and toxicities under A.11, pests and diseases under A.12, etc. All these factors may affect root growth and root system development or function, as well as the growth of the whole crop.

Under the present heading A.4, 'rooting' will signify root room and mechanical impedance.

Root room is the space for root development. It can be represented by critical limits of (i) effective soil depth, (ii) volume percent occupied (or not occupied) by stones, or (iii) the impenetrable (or penetrable) soil volume, as appropriate. Root-occupied soil volume varies with time in the case of annual crops developing root systems from seedling establishment to plant maturity and this process can be slowed by mechanical impedance.

Mechanical impedance to root penetration is the force that roots must exert or resistance they must overcome to penetrate the soil. This depends on the soil strength.

Root room and mechanical impedance may be 'class-determining' where they vary sufficiently from one land unit to another to produce differences in water and nutrient uptake by crops that affect final yields, production or quality.

Effective soil depth and the volume percent of stones are estimated by standard soil survey techniques. The effective soil depth for rooting may be limited by (i) induced hardpans arising as a result of management practices (e.g. heavy traffic or continued soil submergence), (ii) by genetic hardpans such as claypans, siltpans, fragipans, cemented and indurated hardpans, gypsiferous and calcareous hardpans, etc., (iii) by restrictive C or D horizons, (iv) by horizons of low or high pH, with toxic accumulations of aluminium, iron, manganese or sulphidic compounds (low pH), or sodium with carbonates and bicarbonates (high pH).

Soil strength and mechanical impedance to root penetration vary with (i) soil texture or grain size distribution, (ii) soil structure or consistency, and (iii) soil water content. Soil strength and resistance to penetration increase as the soil bulk density is increased by compaction, and as the soil becomes drier. For examples of these effects in sands of different grain size and a sandy loam see Eavis (1972a) and Warnaars and Eavis (1972). Mechanical impedance can be estimated with a penetrometer or may be inferred from soil strength measurements or by observations of root systems in soil profile pits (swollen root tips, limited distribution etc.). Triaxial apparatus used in soil mechanics can measure soil strength. A simpler and satisfactory approach is to apply a 'normal' load via a plate to the top of an unconfined soil core placed on the pan of a top-loading balance. The reading of the balance when the soil core fractures is the unconfined soil strength. Alternatively, a small penetrometer made from an inverted pin can be inserted into the soil core, registering the forces resisting penetration as observed on the scale of the top-loading balance.

Soil strength and penetrometer measurements must be made at appropriate soil water contents as they are much affected by the wetness of soil. The soil water contents selected should reflect the conditions that plants experience in the field.

Soil strength and penetrometer measurements should not be used uncritically. Roots may proliferate through cracks, termite or worm passages and old root channels etc. However, the size of pores that retain water in unsaturated soils are always too small to accommodate roots. Therefore, roots must generally deform the soil in order to penetrate.

Critical limits for 'Rooting' using characteristics such as effective soil depth, volume percent of stones, and soil strength, can be established in terms of the requirements of particular crops or cropping systems, using the s1, s2, s3, n1 and n2 levels of suitability with the help of the above suggestions.

A.5 Aeration

Respiring plant roots consume large quantities of oxygen and, on average, at 25°C they consume about nine times their volume of oxygen gas each day. The roots of plants that grow under submerged conditions can get little oxygen from the surrounding soil and water. Flooded rice and some bog plants can obtain their supplies of oxygen mostly by transport through air passages from the aerial parts to the roots (Greenwood 1968).

Plants, other than rice and bog plants, must obtain their oxygen mainly through the soil. Thus an adequate supply of oxygen through the soil throughout the growing season is a requirement for many crops. Micro-organisms in the soil also consume large quantities of oxygen, and under anaerobic conditions may produce stimulatory or inhibitory levels of ethylene gas, a plant hormone.

In order to reach roots, oxygen diffuses via gas-filled pores and from thence through water films and through the respiring root tissues themselves. Oxygen diffuses 10 000 times more quickly through the gas than through the liquid phase, so that the oxygen concentration at different points throughout the interconnecting network of gas-filled pores is generally fairly uniform. In contrast, concentration gradients of oxygen across the water films and root tissues are large. For this reason, the water content of soils and the thickness of water barriers to oxygen movement around the roots greatly influence oxygen availability.

Although deficient aeration can be readily detected in soils from standard soil survey observations such as gley colours, there is no easily measurable property of soil, or reliable instrument, for determining soil aeration status. Gas-filled pore space in field soils changes inversely with soil water content, therefore, computer simulations similar to that described in Figure 7 might be modified to indicate the variable risks of deficient aeration. The duration of periods of saturation may be deduced by computing a daily soil water balance and the periods of heavy rainfall when the soil exceeds field capacity, due to insufficiently rapid drainage. Thus an important characteristic may be soil permeability or the readiness with which the soil transmits water to drainage. This may be affected by either a water table or by a barrier layer with a relatively low permeability or hydraulic conductivity compared with the overlying soil. The resistance to vertical flow (C) through a barrier equals the thickness of the layer divided by its vertical hydraulic conductivity. Critical limits of C can be set between C = 50 or less representing s1 or virtually no barrier, to C = 250 or over which constitutes a real barrier to flow (FAO Irrigation and Drainage Paper No. 38, 1980a).

The importance of drainage in removing excess water and salts is further discussed in C.21.

Sometimes it is important to recognize that adverse effects due to poor aeration in a wet period may be offset by additional growth in a subsequent dry period. This additional growth may result from additional water stored in the soil prior to the drought. In other words, benefits due to additional residual soil water for growth, following the end of a period of waterlogging or soil saturation, compensate for the effects of poor aeration (Eavis 1971).

If it is found necessary to investigate possible soil aeration problems for a particular soil, a bio-assay technique suitable for soil cores maintained in the laboratory at a range of the field water contents can be used with pea or other seedlings as a test plant (Eavis 1972). An 'aeration deficiency index' can be established by deviations from the relationship between root extension rate and penetrometer resistance. This can sometimes establish whether deficient aeration or some other factor is responsible for poor growth or suboptimal yields on a particular soil.

Although lack of oxygen is often the principal reason for the adverse effects of poor aeration on crop growth, there may be other equally important influences, which include root and foot rots caused by fungal or bacterial pathogens. These can be 'class-determining' for particular crops; for example, citrus and other fruit trees may be prone to gummosis and other diseases on soils subject to periods of poor aeration.

Poor aeration may lead to inefficient use of nitrogen applied in manures and fertilizers. Losses of nitrogen may occur from denitrification and leaching.

In evaluating the factor 'Aeration', the management practices and land development requirements for minimizing adverse effects should be considered. Costs of permanent field drains should be evaluated under heading C.21, and the costs of temporary field drains under B.16.

A.6 Water quantity

A.6.1 The significance of water quantity as a class-determining factor
A.6.2 Water requirement in relation to water supplies
A.6.3 Crop yields and water stress
A.6.4 Estimating irrigation and crop water requirements
A.6.5 The contribution of rainfall to water requirements in farmers fields (effective rainfall)
A.6.6 Seepage and percolation in wetland rice

In quantifying how much water is required for irrigation, it is necessary to distinguish between crop water requirement, net irrigation water requirement, gross irrigation water requirement, and their components as listed below.

Water requirements can be expressed in terms of depth of water (mm) or volume (m³), One mm of water depth on one hectare of land equals a volume of 10 m³/ha (i.e. to convert data in mm to m³/ha, multiply by 10). Water quality is discussed under heading A8. Key factors in determining the supply of water to an irrigated crop are shown in Figure 9.

Figure 9 Flow diagram of key factors determining the supply of water to an irrigated crop

Source: CSIRO Annual Report 1978/80

i. Crop water requirement is defined as the water necessary to meet the maximum evapotranspiration rate of the crop when soil water is not limiting.

ii. Net irrigation water requirement is defined as the water required to meet the crop water requirement, minus contributions in the field by precipitation, run-on, groundwater and stored soil water, plus field losses due to run-off, seepage and percolation.

iii. Gross irrigation water requirement is defined as the net irrigation water requirement, plus conveyance losses between the source of the water and the field, plus any additional water for leaching over and above percolation.

iv. Evapotranspiration is the rate of water loss through transpiration from vegetation, plus evaporation from the soil surface or from standing water on the soil surface. The terms reference crop evapotranspiration (ETo), maximum evapotranspiration rate of the crop (ETm or ETcrop), and actual crop evapotranspiration rate of the crop (ETa), are defined in the Glossary.

v. Effective precipitation (or effective rainfall) is the part of the precipitation that contributes to crop water requirement, net irrigation water requirement or both.

vi. Run-on refers to the contribution of surface water from adjacent land, and run-off the losses to adjacent land.

vii. Groundwater refers to the contribution of water from depth.

viii. Soil water is water stored in the soil. ix- Seepage (following IRRI) refers to losses of water from the field by lateral, surface flow through the bunds (earthen banks) of rice fields. x. Percolation refers to the losses of water from the field by vertical flow through the soil profile.

xi. S & P denotes seepage and percolation (in wetland rice).

xii. Conveyance losses are losses due to evaporation, percolation or breaches in the network of irrigation canals between the source of water and field.

xiii. Leaching requirement is the water required to drain through the root zone to control soil salinity (sometimes expressed as a fraction of the net irrigation water requirement or leaching fraction).

A.6.1 The significance of water quantity as a class-determining factor

In some countries and project areas, the supply of water does not limit crop production and is tailored to meet the full requirement. In others, the abundance of water varies markedly through the year and from year to year. Water supplies can modify the length of the growing period as already discussed under heading A.1.

A.6.2 Water requirement in relation to water supplies

The degree of regulation of water supply in a river valley can vary enormously from little control to full control. Spate irrigation and run-of-river gravity irrigation regulated by diversion structures without storage may result in a very variable water supply. In the absence of sufficient storage within the irrigation network to transfer water from the wet to the dry season, situations arise during the year which are best described as 'land-limited' and 'water-limited'. In 'land-limited' months of the year, there are abundant supplies of water and insufficient land on which to apply it. In 'water-limited' months, the irrigable area is constrained by water supplies, not land area; certain areas of land may then receive more water than other areas.

The area which can be irrigated is, in fact, constrained by the amount of water available in the month or period of limited water (Eavis, Socratous and Makin 1979; Hazlewood and Livingstone 1978: Livingstone and Hazlewood 1979). In the months of abundant water surplus, water that cannot be stored can be spilled, or can be used by the farmer to assist in land preparation (i.e. softening the land) or weed control (i.e. deep standing water in rice fields). It follows that for many gravity-fed run-of-river networks (and also for many schemes with limited storage), the critical limits for defining levels of suitability (s1, s2, s3, n1, n2) must be concerned with the 'water-limited' period and not necessarily with the overall water quantity available for use throughout the year.

An important part of any evaluation of water supply and water requirements, where water is a scarce commodity and seasonally variable, is to match the water supply and water requirement (demand) profiles as closely as possible. For example, cropping systems, and areas occupied by various crops, can be manipulated to accommodate a diminishing water supply towards the end of a rainy season. Also, land preparation and dates of planting can be staggered to smooth away peak water demand where it exceeds the water supply in certain months. Figure 10 illustrates how irrigation scheduling affects the matching of water supply and requirements. If the irrigation project is planned to supply water on a constant amount-constant frequency schedule, the matching of supply and requirements may be poor. Nevertheless, many irrigation projects have to be planned on such a basis to simplify administrative arrangements for issuing water- The matching of supply-demand profiles can be improved by varying the quantity or frequency of water application, or both, as illustrated in Figure 10.

Year to year variations in water supplies are often as important in land evaluation as seasonal variations. Decisions on the size of the 'irrigable' land area are based on matching water requirements and supply; in situations where there are unreliable and erratic supplies, the 'irrigable' area should be calculated at an acceptable level of risk. One must choose between providing water reliably to a small area of land and providing it less reliably to a large area of land; the latter may utilize the available water supply better than the former in wetter years. The compromise should aim to maximize net benefits per project or per unit of water, rather than be certain to achieve high yields on a smaller land area (Eavis, Socratous and Makin 1978). If water supplies are limited by annual variations, high yields per unit of land can only be achieved regularly by foregoing the opportunity to irrigate a larger area in wetter years, unless adequate storage is possible.

Broadly, there are two ways in which water requirements can be reduced to match a scarce water supply in dry periods. Either the land area irrigated can be reduced temporarily during the period of shortage by completely cutting off supplies to certain areas, or the water supplies can be reduced to less than the optimum requirement with the consequence that crop yields are reduced.

Clearly, areas of land that will receive water in some years or seasons, but not during periods of shortage must be given a lower level of suitability than those which have more reliable water supplies. Crops that tolerate drought, yet respond when water is plentiful, can sometimes be grown on such land, or short duration crops can be dropped from the cropping system during the dry period. Critical limits than define s1, s2, s3, n1 and n2 levels of suitability for a LUT can be set from the above considerations.

Figure 10a Effect of irrigation scheduling on the matching of water supply and water requirements - Constant amount-constant frequency (rotational) schedule

Figure 10b Effect of irrigation scheduling on the matching of water supply and water requirements - Constant amount-variable frequency schedule (variable frequency rotation)

Figure 10c Effect of irrigation scheduling on the matching of water supply and water requirements - Varied amount-constant frequency schedule (varied amount rotation)

Source: Replogle and Merriam 1980

A.6.3 Crop yields and water stress

The effects of water deficit on the yields of many crops have been described in FAO Irrigation and Drainage Paper No. 33 (FAO 1979c). Experimentally, crop yield and transpiration are often reported to be directly proportional and for many practical purposes a linear relationship between crop yield and the actual crop evapotranspiration (ETa) is a good approximation (at least over the marketable yield range)- Some crops suffer from lack of water more at some stages of development than others and this can be taken into account.

The type of production function often found experimentally is illustrated in Figure 11. Yield and evapotranspiration are plotted as ratios of the maximum yield (Ym) and the maximum evapotranspiration (ETm or ETcrop). Subtracting these ratios from unity (1), transforms the scales so that they indicate relative yield decrease (1 - Ya/Ym) and relative evapotranspiration deficit (1 - ETa/ETm). Critical limits may be set as illustrated in Figure 11 for appropriate ranges of yield. If the water supply varies from year to year, the percentage of years in which an 'n' rating occurs can be the basis for selecting critical limits (see Eavis, Socratous and Makin 1979).

A.6.4 Estimating irrigation and crop water requirements

Methods for estimating water requirements are described in FAO Irrigation and Drainage Paper No. 24. The main steps generally necessary are listed below. A more sophisticated procedure based on water balance models, that takes into account variations of daily rainfall in farmers' fields, the soil as a reservoir, and water shortages, is described by Eavis, Socratous and Makin (1979).

Figure 11 Relationship between crop evapotranspiration and yield for potatoes, based on irrigation experiments in Cyprus, showing how critical limits were defined. (The proportion of years in which water supply and relative yields fell to n1 [N1 on the figure] was the basis for sizing the irrigable area)

The main steps in estimating crop water and irrigation requirements are as follows:

i. set out a cropping calendar of 10-day or weekly periods for land preparation, planting, (draining the wetland rice field), harvesting, etc.;

ii. calculate 'reference crop evapotranspiration' (ETo) for each 10-day or weekly period. Use climatic data or records of pan evaporation by the methods described in FAO Irrigation and Drainage Paper No. 24;

iii. select crop coefficients (kc) according to instructions in FAO Irrigation and Drainage Paper No. 24 (kc = ETm/ETo for different stages of crop development);

iv. obtain the maximum crop evapotranspiration (ETm) by multiplying ii. and iii- (ETm = ETo.kc) for different stages of crop development. This assumes no water shortages occur;

v. add in water requirements for wetting the soil initially if it is dry, and for land preparation; also that for draining rice fields for weeding, etc.;

vi. subtract water requirements supplied by residual soil moisture towards the end of the growing season (if appropriate);

vii. add in estimates of losses from run-off, seepage and percolation, or gains from run-on or groundwater;

viii. calculate the leaching requirement (see Figure 16). If the expected percolation is insufficient to keep soil salinity within the required range, add the appropriate amount to the water requirement;

ix. deduct the contribution from precipitation or rainfall in farmers' fields (effective rainfall) from the irrigation water requirement;

x. convert the above requirements in mm, into volumes of water per irrigated area (i.e. mm x 10 x ha = m³;

xi. add on the conveyance losses between source of water and the field;

xii. on the basis of the irrigation application technique, decide on irrigation schedules (e.g. see Figure 10) in terms of frequency, rate and duration of water application;

xiii. determine peak water requirements in terms of flow rates (litres per sec per ha, l/s/ha);

xiv. match supply and requirement profiles by review and iteration.

A detailed description of the above procedures is beyond the scope of this publication. Figure 12 with its accompanying Table 34 illustrates by example some estimates of irrigation water requirements for double cropped rice in Bali, Indonesia (Eavis and Walker 1976). The range of irrigation water requirements for different land units in the project area of this example differed mainly because of the variation in field losses from seepage and percolation (see Section A.6.6).

Table 34 GROSS ANNUAL IRRIGATION WATER REQUIREMENTS FOR LOCAL 140-DAY RICE FOLLOWED BY HYV 120-DAY RICE (to accompany Figure 12)

Seepage and percolation losses mm/day		Rice 1 crop (140 day)	Rice 2 crop (120 day)	Gross annual irrigation water requirements mm/year 1/	Peak demand rate l/sec/ha
		mm/crop
1	5.0	1 352	1 901	3 253	2.15
2	10.0	2 146	2 744	4 890	3.26
3	15.0	3 096	3 701	6 797	3.59

1/ mm/year x 10 = the gross irrigation water required in m³/ha/yr.

Figure 12 Irrigation water requirements of local rice followed by HYV rice, north coast of Bali, Indonesia

Assumptions:

a) Seepage and percolation losses were taken as 5 mm/day, 10 mm/day and 15 mm/day to meet the range of conditions in the different land units 1, 2 and 3 (shaded areas).

b) Planting spread over 60 days.

c) Local variety (traditional) 140 days' duration; HYV 120 days.

d) Pre-saturation 250 mm for first crop and 100 mm for second crop.

e) Losses in conveyance 15% gross supply plus 5% due to administrative wastage: total loss up to field gate = 20% gross supply.

f) 80% probability rainfall was the figure used to approximate the effective rainfall.

A.6.5 The contribution of rainfall to water requirements in farmers fields (effective rainfall)

In humid areas, crop water requirement may be partially provided by rain falling directly on the farmers' fields. The respective proportions of the irrigation water requirement provided (i) from rainfall or precipitation, and (ii) from irrigation, may differ from year to year and season to season.

Not all the rain received in the field directly is effective. Part is lost by run-off, deep percolation, or by evaporation of rain intercepted by the plant foliage. Where run-off is not important, the best method for estimating the direct contribution of rainfall to water requirements, is to construct a daily soil water balance using historical daily rainfall (Stern and Coe 1982). Simpler approximations have often been used (e.g. the 80% probability rainfall), sometimes with misleading results. Older methods for computing rainfall probabilities and effective rainfall are given in FAO Irrigation and Drainage Paper No. 24.

Land characteristics such as slope, relief, infiltration rate, cracking, permeability and soil management may all 'influence the utilization of rainfall and critical limits of the important ones can be used in land evaluation.

In wetland rice, particularly, the contribution of effective rainfall depends on water conservation practices. Continuous flow of water by gravity in many rice areas keeps the paddy fields full of water. Rainfall on a full paddy overspills and may not be effectively utilized further downstream. To improve the use of direct rainfall, farmers should block off the through flow and allow the standing water levels in the fields to decline. This partially empties the paddy field and presents an opportunity for the temporary storage of rainfall: it also reduces seepage and percolation losses (see below).

Generally, the proportion of the total rainfall which is effective is greater in dry periods than in wet periods. Snow melt, and run-on from adjacent areas are of great importance in some areas.

A.6.6 Seepage and percolation in wetland rice

In many rice areas, water losses by seepage and percolation (S & P) greatly exceed evapotranspiration. If water quantity is limiting, S & P and the land characteristics affecting them, are important candidates for evaluation.

Seepage is the lateral movement of water through the soil and through levees, embankments or bunds around fields. Percolation is the vertical movement downward towards a water table. Seepage and percolation cannot always be clearly differentiated under field conditions and therefore are often considered together.

In relatively flat areas with few drains and a high water table, seepage and percolation depend on the total outflow in response to differential water heads and the resistance to flow through bunds and the soil. In a series of paddies on a slope, seepage from one paddy to the next is offset by seepage from higher paddies and net losses can be measured from the last paddy of the system, which is usually located along a drain or unplanted area acting as a sink for the entire system.

Percolation rates are governed by the water head (pressure) and the resistance to water movement through the soil profile. The depth of standing water in paddy fields, due to increased head, has a marked influence on percolation rates through the floor of the field and through the bunds, on permeable soils. Soil permeability is affected by soil structure, texture and the interfaces between horizons, including the presence of claypans or hardpans which may give rise to a perched water table distinct from and above the true water table. In many rice areas the water table itself controls percolation losses of water.

On sloping land where the water table is below the 25-30 cm rooting depth of rice, S & P are related to how watertight the terraced paddies can be made by puddling the floor of the field, and by blocking off leakage through the bunds. Puddling breaks down the soil structure and helps sealing against water losses. The effectiveness of puddling depends on characteristics such as texture, the clay minerals predominating, swelling and shrinking, salinity, organic matter content and water control. The effectiveness of puddling is often a prime consideration in evaluating water losses by S & P. Newly created riceland may not puddle effectively and it is common for the effectiveness of puddling to improve over the first five to ten years.

Losses of water to a water table not only occur through the floor of the field, but also through the bunds. In many rice areas, permanent bunds have well-structured permeable soils permitting leakages which increase with the depth of standing water in the field (Walker and Rushton 1984); these losses may be the principal concern. If the bunds are permanent terrace structures the difficulty of preventing such losses is likely to be greater than where bunds are destroyed after each rice crop, and then reformed and compacted during land preparation.

Soil texture, the percentage of clay and the clay mineralogy may all contribute to an evaluation of S & P water losses. Net irrigation water requirements for rice on sandy soils are likely to be greater than on clay soils. Chin and Lee (1961) reported 7, 9, 10 and 12 mm/day for Taiwan soils having 25-30%, 15-25%, 10-15%, 5-10% clay content respectively. Gupta and Bhattacharya (1969) reported decreasing percolation losses successively for sandy soils, sandloams, fine sandy loams and heavy clays. But Achar and Dastane (1971) reported percolation losses as high as 19 mm/day from heavy black vertisols with as much as 50% clay. The type of clay mineral and salt content of the soil and water are important factors to consider. Salts lead to the aggregation of montmorillonitic and other clays and aggravate S & P losses. Kaolinitic clays swell relatively little and puddle less effectively than montmorillonitic clays.

S & P will often dictate the limit of wetland rice irrigation on catenary topographic sequences- In valley bottoms, S & P are less than further up-slope, partly due to soil texture and structure changes. Hence S & P will often be important in deciding the size of the irrigable area. A range in S & P between 2 and 15 mm/day in the example given in Figure 12 corresponded to a threefold increase in net irrigation water requirement per year, and a similar increase in peak demand rate of supply.

Guidelines for critical limits of seepage and percolation for wetland rice are suggested as follows but these must be modified to reflect the significance of water supplies under local conditions as already explained in A.6.1:

s1	0-2 mm/day
s2	2-5 mm/day
s3	5-15 mm/day
n1	more than 15 mm/day

Approximate values for S & P may be obtained from observations of the recession of standing water in existing rice paddies on soils similar to those of interest. In the sloping gauge technique, an inclined metre stick (slope 1:5) is installed in the representative paddies. The decrease in water level is measured on the gauge and represents the combined losses from evapotranspiration, inflow (irrigation + rainfall), and S & P. If no water is added or drained from the paddy, then the total water used (indicated by change in depth) is simply evapotranspiration plus S & P. Over small areas the evapotranspiration is approximately the same and differences between locations can be assumed to be differences in seepage and percolation.

A.7 Nutrients (NPK)

A.7.1 Nitrogen
A.7.2 Phosphorus
A.7.3 Potassium
A.7.4 Factor rating 'NPK nutrition'

The discussion of crop nutrition in this section is mainly restricted to nitrogen (N), phosphorus (P) and potassium (K), the three major nutrients commonly supplied as fertilizers. Other major and micro-nutrients and the effects of pH and toxicities are discussed under heading A.11.

A distinction must first be drawn between:

- crop nutrient requirements based on the nutrient content of the crop,
- requirements for fertilizers and manures.

The mineral composition of plant dry matter as a measure of crop nutrient requirements necessitates serial sampling during the life of the crop for accurate results; however crop nutrient uptake is usually taken as the nutrient content of the harvested crops. This gives a guide as to the nutrients required to maintain soil fertility at about the existing level. Supplies of plant nutrients to replace those removed at harvest may come from:

i. soil mineralization (i.e. the transformation of soil minerals or organic matter from non-available into available nutrients);

ii. manures and fertilizers, or

iii. fixation from the air, in the case of some of the nitrogen.

Losses also occur from leaching, fixation in unavailable forms and demineralization, etc. Fertilizer and manurial requirements depend on all these considerations and are usually determined empirically by experiment.

A.7.1 Nitrogen

Nitrogen is second only to water in importance as a factor affecting the yields of most irrigated crops. The nitrogen cycle is illustrated in Figure 13. Nitrogen fertilizers give fairly predictable yields where lack of nitrogen is the principal factor limiting yields. The main considerations in deciding how much nitrogen should be applied to obtain a given yield are:

i. the amounts of nitrogen removed by the crop;
ii. the initial nitrogen content of the soil;
iii. the contribution from nitrogen fixation;
iv. the losses of nitrogen from leaching, denitrification, etc.

Figure 13 Transformations of nitrogen in the rhizosphere

Source: FAO 1980d

A further consideration, where high yields of short duration leafy crops are required, is the need for temporarily high concentrations of nitrate or ammonium nitrogen in the soil to stimulate uptake at critical stages of growth; such applications often lead to additional losses by leaching, and hence the amount of nitrogen applied in intensive cultivation may be considerably greater than the nitrogen removed by the crop. On the other hand, contributions may come from the air or from mineralization of organic matter. Traditional rice varieties have been grown for centuries in many parts of Asia without the use of nitrogenous manures or fertilizers, owing to the atmospheric fixation of nitrogen by micro-organisms and algae. Both flooded rice and leguminous crops can contribute in the order of 30-75 kg N/ha during favourable conditions over a three to four month period, though usually not enough for optimum or maximum yields with modern varieties.

If factors other than water and nutrients are not limiting, then the interaction between nitrogen fertilizer and water application is frequently highly significant. Recently reclaimed soils in arid and semi-arid areas are often inherently very low in organic matter and nitrogen content, yet very suitable for irrigation. On such soils the type of response and interaction to water and nitrogen fertilizers illustrated in Figure 14 is obtained; a bigger response to nitrogen is obtained if water is not scarce. The variable performance of land units can be evaluated in terms of such a response surface. Alternatively, critical limits may be defined as the cost of fertilizer inputs on different land units.

Figure 14 An example of the yield response to fertilizer and water for a soil low in available nitrogen

The cost of applying fertilizer nitrogen may vary from land unit to land unit. Soils requiring high nitrogen inputs may be initially low in nitrogen, or may utilize nitrogen applications inefficiently due to leaching or other losses. In practice, however, farmers often use the same amounts of fertilizer on all their land and yields may then vary on account of different efficiencies of utilization.

Nitrogen deficiency is especially common on sandy and well-weathered soils in areas of high rainfall and on soils low in organic matter. Total nitrogen and nitrate nitrogen contents of soils give some indication of severe deficiency but, in intermediate ranges and because of seasonal changes, are of little help in determining immediate fertilizer requirements.

Arable soils have a variable nitrate content ranging from less than 2 to 60 mg/l of nitrogen as nitrate. High levels of nitrate nitrogen may indicate that little or no nitrogen need be applied. Total soil nitrogen is low if it is less than 0.1% and high if it is more than 0.3% of the oven dry soil.

Tissue analyses of plant leaves give a range in total nitrogen from about 1.5% (low) to 3.5% (high), depending on the crop and age of the leaf, etc. (Chapman 1973).

A.7.2 Phosphorus

Much attention has been given to the development of chemical methods for determining the available phosphorus in a soil, where availability is defined as the amount a crop will take up from the soil or, alternatively, where it is used as a measure of the ability of soil to supply the amounts needed for maximum crop yield under the system of agriculture being practised (Russell 1973), The soil analyses can generally detect gross deficiency but do not have much general predictive value in deciding phosphorus fertilizer requirements to achieve various yields unless first proved suited to a particular area. The uptake of available soil phosphorus by crops depends on many factors including:

i. how fast the unavailable forms of phosphorus are transformed into exchangeable forms and vice versa;

ii. the rate at which the available and exchangeable forms are released into the soil solution;

iii. the soil water content and solution concentrations during the period of growth;

iv. the crop requirement for phosphorus;

v. how effectively the root system explores the soil volume and absorbs and utilizes the phosphorus present.

Phosphorus deficiency most commonly occurs on highly weathered tropical soils, calcareous soils, and peat and muck soils but there is a response to fertilizer phosphorus on a very wide range of soils.

Highly weathered tropical acid soils include some that absorb phosphate so strongly that its concentration in the soil solution remains too low for the crop to benefit from it without massive applications. Yet some crops (e.g. cassava) can utilize the phosphate on such soils. In land evaluation it is particularly important to identify soils that strongly absorb phosphorus - In the tropics these soils are often oxisols or ultisols. Russell (1973) suggests that they could be identified by a laboratory determination of the amount of phosphorus a soil sample must absorb to come into equilibrium with a solution of phosphorus of the same strength as is required for active phosphorus uptake by most crops (e.g. 10^-5 M, but less or more depending on the crop).

Acid soils usually require more added phosphorus than neutral and calcareous soils, and rock phosphate is effective on acid soils. Excess phosphorus in species such as citrus that are sensitive to phosphorus excess may induce both copper and zinc deficiency symptoms on calcareous soils.

Phosphorus in tropical soils is commonly mineralized from organic matter at the start of the rains or irrigation, following a dry period. The availability of phosphorus, whether judged by chemical methods or by plant uptake, increases on submerging a soil. The increase in the availability of phosphorus and other elements is often cited as one of the benefits of flooding rice soils (Ponnamperuma 1976). Nonetheless, the increase in the solubility of phosphorus on flooding is low in ultisols and oxisols.

Table 35 EXTRACTANTS USED TO ESTIMATE 'PLANT AVAILABLE' PHOSPHORUS IN SOILS

EXTRACTANT	SOILS	mg P/l OF DRY SOIL
		RESPONSE EXPECTED	RESPONSE PROBABLE	RESPONSE UNLIKELY
1. Acetate-acetic acid (Morgan) 1/ 0.0125 with respect to sodium 0.16N with respect to acetate pH 4.8 to 5.0	Acid soils	<5	6-15	>15
2. Hydrochloric-sulphuric acid-soluble phosphorus method (Mehlich) 5 g of soil shaken for 5 minutes with 20 ml of 0.05N HCl-0.25N H2SO4 solution. Filter.	Non calcareous acid soils that have not recently been treated with rock phosphate	<10	11-31	>31
3. Hydrochloric acid-ammonium fluoride-soluble phosphorus test (Bray No.1) Shake 1 part of soil for 1 minute with 7 parts of 0.025N HCl-0.03N ammonium fluoride. Filter.	Non calcareous soils	<7	7-20	>20
4. Sodium bicarbonate-soluble phosphorus test (Olsen) Add 100 ml of 0.5M sodium bicarbonate adjusted to pH 8.5 to 5 g of soil. Add one teaspoon of phosphorus-free carbon black, shake for 30 minutes. Filter and determine P on 5 ml per 25 ml final volume.	Neutral, calcareous and acid soils. Not for sodic soils high in organic matter	<5	6-10	>11
5. Carbonic acid (McGeorge) 50 g of soil placed in a cylinder with 250 ml of distilled water. CO2 passed through the suspension for 15 minutes before filtering.	Calcareous soils. Rapid routine standardized testing not always possible	<3	6-10	>10
6. Water soluble phosphorus method (Bingham) 10 g of soil shaken 15 minutes with 100 ml distilled water and filtered.	All soils	<1 or <0.13 in water extract	1-2	>2 or >0.13 in water extract

1/ In the Morgan test, ratios and extractant times vary widely, e.g. 1:2 soil: extract shaking time 1 minute, compared with 1:5 and shaking time 30 minutes.

In general, the phosphorus fertilizer or manurial requirement for rice is less than for other cereals. Grasses and cereals usually have a lower requirement than crops such as potato, sugarbeet, and leafy vegetable crops which may respond to two or three times more than the application rate for the former (Bingham 1973). Wherever possible, such generalizations should be confirmed by experiments which also examine the optimal sources and inputs of phosphorus fertilizers in their various forms as rock phosphate, superphosphate, triple superphosphate, etc.

Many chemical techniques for measuring 'plant available' soil phosphorus are in current use and these are summarized (Table 35) in terms of the responses that are anticipated in the area where they are used (Bingham 1973).

Total phosphorus content in plant tissues ranges from 0.05% to 0.5%, depending upon the state of nutrition, plant species, season and tissue sampled. Leaf values from tree crops are usually lower than those for most annuals, ranging from 0.05% to 0.10% total phosphorus for deficiency, and 0.2% to 0.4% in the satisfactory range. Excess is not readily detected by tissue analysis, but if values greater than 0.5% are encountered in trees, further examination may be advisable (Bingham 1973).

A.7.3 Potassium

Potassium deficiency, as indicated by low exchangeable potassium (Ulrich and Ohki 1973), commonly occurs on:

i. sandy soils that leach excessively;
ii. acid sandy soils;
iii. organic soils;
iv. soils that have been heavily cropped, leached or eroded;
v. highly leached ferralsols.

In contrast, soils in many arid and semi-arid areas have more than sufficient plant available potassium to meet the nutritional requirements of irrigated crops. This is because of the relatively small impact of weathering and leaching on dry-region soils. Exceptions on such soils can readily be detected from exchangeable potassium determinations, extracting the potassium with normal ammonium acetate or 0.5 normal sodium bicarbonate.

Soils with less than 100 kg/ha of exchangeable potassium in the root zone are often responsive to potassium fertilizer. If the soil contains more than 300 kg/ha of exchangeable potassium, very few crops are likely to respond (Ulrich and Ohki 1973).

The potassium content of leaves used for tissue analysis ranges from about 0.7 to 1.5% on a dry weight basis, for most plants. A few plant species (e.g. potato) require much higher concentrations in their tissues for normal growth: up to 5% potassium may be beneficial in intensive, early season temperate crops.

A.7.4 Factor rating 'NPK nutrition'

The selection of appropriate factor ratings for 'NPK Nutrition' requires an appreciation of likely yield response curves for crops on the land units being evaluated. For example. Figure 14 illustrates a family of yield response curves to nitrogen fertilizer (corresponding to different evapotranspiration levels). The physical responses to various levels of fertilizer can be translated into financial benefit/cost ratios. Factor ratings for 'NPK Nutrition' can be used to approximate benefit/costs; for example, if a land unit produced excellent yields without any fertilizer it would receive a factor rating of s1. A benefit/cost of less than one, which implies that costs of fertilizers exceed the financial value of the yield increment, would lead to a factor rating of n1 or n2. N, P and K can each be evaluated separately, and then be combined to give an overall factor rating for NPK Nutrition, for example as in Table 36. The 'significance' of this factor rating can then be judged relative to others listed on Format 3.

Table 36 FACTOR RATING OF 'NPK NUTRITION' IN TERMS OF FERTILIZER REQUIREMENTS FOR A SPECIFIC LAND UNIT AND LAND UTILIZATION TYPE

LUT: Crop or crops: Land Unit No(s):
	Factor Ratings					Selected rating
	s1	s2	s3	n1	n2
	Benefit/cost index 1/
	> 3	2-3	1-2	0.5-1	< 0.5
Nitrogen			Ö			s3
Phosphorus	Ö					s1
Potassium	Ö					s1
	Factor Rating for 'NPK Nutrition' (i.e. relatively high costs of nitrogen fertilizer are implied to obtain acceptable yields)					s3
Conclusion: Enter s3 on Format 3 for 'NPK Nutrition'

1/ Benefit/costs given are examples only.

A.8 Water quality

Guidelines for evaluating the suitability of water for irrigation are described in FAO 1976c, Irrigation and Drainage Paper No. 29 (Revision 1 is in press and will be issued in 1985). Values are suggested (Table 37) which relate to the general irrigation problems of salinity, infiltration and specific ion toxicity. The suitability of water depends on i) how it is managed, ii) the nature of the soil, and iii) crop tolerance to salinity of various types of irrigation water. Water quality criteria must be interpreted in the context of overall salt balances and toxicities and the effects on soil. The problems that result from using poor quality water vary both in kind and degree but the most common are:

Table 37 GUIDELINES FOR INTERPRETATION OF WATER QUALITY FOR IRRIGATION 1/

Potential Irrigation Problem			Restriction on Use
			Units	None	Slight to Moderate	Severe
Salinity (affects crop water availability) 2/
	ECw (or)		dS/m	<0.7	0.7 - 3.0	>3.0
	TDS		mg/l	<450	450 - 2 000	>2 000
Infiltration (affects infiltration rate of water into the soil. Evaluate using ECw and SAR together) 3/
SAR =0-3 and ECw =				>0.7	0.7 - 0.2	<0.2
	= 3 - 6 =			>1.2	1.2 - 0.3	<0.3
	= 6 - 12 =			>1.9	1.9 - 0.5	<0.5
	= 12 - 20 =			>2.9	2.9 - 1.3	<1.3
	= 20 - 40 =			>5.0	5.0 - 2.9	<2.9
Specific Ion Toxicity (affects sensitive crops)
	Sodium (Na) 4/
		surface irrigation	SAR	<3	3 - 9	>9
		sprinkler irrigation	me/l	<3	>3
	Chloride (Cl) V
		surface irrigation	me/l	<4	4 - 10	>10
		sprinkler irrigation	me/l	<3	>3
	Boron (B) 5/		mg/l	<0.7	0.7 - 3.0	>3.0
	Trace Elements (see FAO 1985)*
Miscellaneous Effects (affects susceptible crops)
	Nitrogen (NO3-N) 6/		mg/l	<5	5 - 30	>30
	Bicarbonate (HCO3) (overhead sprinkling only)		me/l	<1.5	1.5 - 8.5	>8.5
	pH			(Normal Range 6.5 - 8.4)

1/ Adapted from University of California Committee of Consultants, 1974.

2/ EC_w means electrical conductivity, a measure of the water salinity, reported in deciSemens per metre at 25°C (dS/m) or as previously reported in millimhos per centimetre at 25°C (mmho/cm); they are numerically equivalent. TDS means total dissolved solids, reported in milligrams per litre (mg/l).

3/ SAR means sodium adsorption ratio. SAR is sometimes reported by the symbol Tte. See Glossary for the SAR calculation procedure. At a given SAR, infiltration rate increases as water salinity increases. Evaluate the potential infiltration problem by SAR as modified by EC. Adapted from Rhoades 1977; and Oster and Schroer 1979.

4/ Values for sodium and chloride applicable to sensitive tree and woody plants with surface irrigation; many annual crops are less sensitive to these specifications. With overhead irrigation and low humidity (<30 percent), sodium and chloride absorbed through the leaves of sensitive crops can cause damage.

5/ For boron tolerances, see Table 42.

6/ NO₃-N means nitrate nitrogen reported in terms of elemental nitrogen. (NH₄-N and Organic-N should be included when wastewater is being tested).

* See revised FAO Irrigation and Drainage Paper No. 29 (1985) for trace elements and for new method to calculate adjusted SAR (adj. R_Na).

i. Salinity: A salinity problem can occur if the total quantity of salts in the irrigation water is high enough for the salts to accumulate in the crop root zone to the extent that yields are affected. Excessive soluble salts in the root zone inhibit water uptake by plants. The plants suffer from salt-induced drought. Plants respond more critically to salinity in the upper part of the soil profile than to the salinity levels at depth. Thus, managing this critical upper root zone may be as important as providing adequate leaching to prevent salt accumulation in the total root zone.

ii. Infiltration: An infiltration problem related to water quality occurs when the rate of water infiltration into and through the soil is reduced (because of this water quality) to such an extent that the crop is not adequately supplied with water and yield is reduced. The poor soil infiltration makes it more difficult to supply the crop with adequate water and may greatly add to cropping difficulties through crusting of seed beds, waterlogging of surface soil and accompanying disease, salinity, weed, oxygen and nutritional problems. It is evaluated, first, for total salts in the water because low salt water can result in poor soil infiltration due to the tremendous capacity of pure water to dissolve and remove calcium and other solubles in the soil; secondly, from a comparison of the relative content of sodium to calcium and magnesium in the water or SAR (sodium adsorption ratio). Thirdly, carbonates and bicarbonates can also affect infiltration and must be evaluated. These three factors interact in determining the long-term influence of a water on the soil infiltration rate.

iii. Toxicity: A toxicity problem occurs when certain constituents in the water are taken up by the crop and accumulate in amounts that result in a reduced yield. In arid and semi-arid areas this is usually related to one or more specific ions in the water, namely, boron, chloride and sodium.

iv. Miscellaneous: Various other problems may occur, e.g. from excessive nitrogen in the water supply, white deposits on fruit due to high bicarbonate in sprinkler applied water, and suspected abnormalities indicated by water with an unusual pH.

The guidelines presented in Table 37 should allow a determination that water of a given salinity, SAR and specific ion composition does or does not have a potential to limit crop production. Where limitations are indicated, the water may still be usable providing certain management steps are taken to alleviate the problem. The guidelines in Table 37 were drawn up on the assumption that the area under consideration is semi-arid or arid, with low rainfall, good drainage, no uncontrolled shallow water table and that surface or sprinkler applications of water were used with 15% of the water percolating through the root zone. The guidelines are possibly too strict for drip irrigation on highly permeable soils, but elsewhere the user must constantly guard against drawing unwarranted conclusions based strictly on laboratory results alone. Further discussion of salinity limitations continues in heading A.9.

The quality of water for localized irrigation techniques is an important management criteria that can be conveniently introduced here. Critical limits have been suggested by Bucks and Nakayama (see Howell, Bucks and Chesness 1980) as set out in Table 38.

Table 38 CRITICAL LIMITS FOR CLASSIFYING POTENTIAL CLOGGING HAZARD OF IRRIGATION WATER TO BE USED IN TRICKLE SYSTEMS

Factor		Clogging Hazard
		Slight	Moderate	Severe
Physical Suspended solids (Max. ppm) 1/		< 50	50-100	> 10
	Chemical
	pH	< 0.7	7.0-8.0	> 8.0
	Total Dissolved Solids (Max. ppm) 1/	< 500	500-2 000	> 2 000
	Manganese (Max. ppm) 1/	< 0.1	0.1-1.5	> 1.5
	Iron (Max. ppm) 1/	< 0.1	0.1-1.5	> 1.5
	Hydrogen sulphide (Max. ppm) 1/	< 0.5	0.5-2.0	> 2.0
Biological
	Bacteria populations (Max. no/ml) 2/	< 10 000	10 000-50 000	> 50 000

1/ Maximum measured concentration from a representative number of water samples using standard procedures for analysis.

2/ Maximum number of bacteria per millilitre can be obtained from portable field samplers and laboratory analysis.

Source: Bucks and Nakayama from Howell et al. 1980.

A.9 Salinity

The adverse effects of soil salinity on plant growth vary with the crop being grown. The presence of salinity in the soil solution resulting from either indigenous salt in the soil, or from salt added by irrigation water can affect growth (i) by reducing water available to the crop (the osmotic effect) and (ii) by increasing the concentration of certain ions that have a toxic effect on plant metabolism (the specific ion effect). There is an approximate tenfold range in salt tolerance of agricultural crops. This wide choice of crops greatly expands the usable range of water salinity for irrigation and emphasizes the fact that water quality and soil salinity are specific for the intended use. Many plants, for example, barley, wheat and maize, are sensitive to the osmotic effect during germination and the early seedling stages, but have greater tolerances at later stages (USDA 1954). Salt damage is aggravated by hot, dry conditions and may be less severe in cool humid conditions. Salt tolerance data for any given crop cannot be considered as fixed values, but should be used as guidelines.

The evaluation of plant salt-tolerance data by Maas and Hoffman (1977) suggests that for each crop a certain threshold value exists beyond which crop yields decrease linearly with increasing salinity. When the soil saturation extract ECe value is less than some prescribed threshold value, crop yields are unaffected and represent 100% relative yield.

Salinity tolerances for various crops are given in Figure 15 which also indicates approximate yield reductions in relationship to increasing salinity of the soil saturation extract ECe.

Figure 15 Salt tolerances of various crops to salinity as measured in the saturation extract ECe. Vegetable crops.

Figure 15 Salt tolerances of various crops to salinity as measured in the saturation extract ECe. Field crops.

Figure 15 Salt tolerances of various crops to salinity as measured in the saturation extract ECe. Fruit crops.

Figure 15 Salt tolerances of various crops to salinity as measured in the saturation extract ECe. Forage crops.

Source: Maas and Hoffmann 1977; James et al 1982.

The relation between the conductivity of saturated extract ECe and that of the irrigation water, ECw, can be approximated by assuming that the irrigation water concentrates three times as it becomes soil water and that the salinity of the saturation extract is half that of the soil water (i.e. ECe = 3/2 ECw). For example, the yield of sugarbeets is reduced 10% when the ECe reaches 8.6 dS/m (at 25°C). This corresponds to ECw = 5.7 which is the ECw of the irrigation water that would result in a 10% yield decrement if the assumptions are valid. This calculation also assumes that all the salinity is derived from the irrigation water and none from the soil; it may frequently be necessary to take the latter into account.

The leaching requirement using irrigation water of different salinities and for crops of different salt tolerances can be readily obtained from the graphical solution given in Figure 16. Thus if the conductivity of the irrigation water is 2 dS/m and the crop salt tolerance threshold is 4, the leaching requirement is about 0.10, or 10% of the applied water should leach through the soil.

Figure 16 Graphical solution for the leaching requirement (L), the minimum leaching fraction that prevents yield reduction, as a function of the salinity of the applied water and the salt tolerance threshold value for the crop

Source: Hoffmann and van Genuchten 1980

Additional water will be required to reduce an initially high salt content to an acceptable value (see C.24 Reclamation Leaching).

A.10 Sodicity

The detrimental effect associated with sodium accumulation in soil can be divided into two categories: i) deterioration of the physical condition of the soil; and ii) sodium toxicity.

i. physical effects of sodicity

The presence of excessive amounts of exchangeable sodium in soil promotes the dispersion and swelling of clay minerals. The soil becomes impermeable to both air and water. The infiltration and hydraulic conductivity decrease to the extent that little or no water movement occurs. The soil is plastic when wet and becomes hard (brick-like) when dry. Tillage becomes difficult and soil crusting occurs. Recent research (Frenkel et al. 1978) has indicated that dispersion blocks soil pores, whereas swelling reduces pore sizes. The effect is most pronounced on soils containing clays which swell and shrink. Soils containing non-expanding clays such as kaolinite and sesquioxides are relatively insensitive to the physical effects of exchangeable sodium. However, heavy cracking clays may be so impermeable when wet that the decreased permeability associated with a high sodium content may not matter.

Sodicity is determined as the exchangeable sodium percentage (ESP). In rating sodicity one should take into account the changes in ESP which will take place after the land is irrigated.

The actual sodicity which can be maintained in the soil with the anticipated soil and water management practices should be rated according to the overall effect on crop production. As the root systems of most crops develop best in the upper 30 cm of the soil, more attention may be given to the surface soil, except in the case of tree crops, or where sodicity in the subsoil is an important factor in drainage.

Both laboratory and field methods can be used to determine salinity and sodicity. Laboratory studies can be used to determine critical limits for the influence of exchangeable sodium on the physical characteristics (e.g. permeability) of individual soils.

In past surveys, sodicity phases of soils were often distinguished on the basis of a high pH measured in the field. However, pH is not only a function of sodicity but of salinity as well, and should not be used alone for the rating of sodicity. In case ESP is not determined, or only for a limited number of profiles, it may be estimated from pH-ESP-EC relationships that have been worked out for a number of areas. These relationships may not always hold, especially when gypsum is present. The SAR of the irrigation water will influence the ESP of the soil, but the relation between the two may not be straightforward, because the ESP of the soil is conditioned by the SAR of the soil solution, and this is constantly changing. After irrigation, the soil solution slowly becomes more concentrated as the crop transpires water, so its SAR rises, and if the effect of concentration is to cause some calcium or magnesium to precipitate out as the carbonate, this will also cause it to rise still further. The higher the total salt content of the irrigation water, the lower must be its SAR if the ESP of soil is to remain below a given level (Table 37).

Before leaching, saline-alkali soils often have a high ESP associated with the high salinity. The ESP values of soils before leaching may give a misleading impression of the potential sodicity hazard. The ESP values should be determined after leaching tests have been carried out with water of similar quality to that to be used for irrigation.

On some cracking clays (i.e. black clays high in montmorillonite in the Gezira, Sudan), which crack on drying, but which are otherwise impermeable, good crop yields can be obtained even though they have a high ESP (up to 40%) and an unstable structure. These clays may contain aluminium hydroxide films which probably help to stabilize the cracks to some extent.

In Table 39, based on work in the Sudan (Purnell, pers. comm.), critical limits are given for sodicity, using values for the ESP and

SAR of the soil solution after leaching. The ratings s1, s2, s3 and n reflect nonsodic, slightly sodic, moderately sodic and strongly sodic soils respectively. This data may not be satisfactory for a proper evaluation on its own. Other factors, including the internal drainage of the soil, the properties of the clay minerals present, the calcium and gypsum content, the particle size of carbonate fractions, and the salt and SAR of the irrigation water are all important.

Table 39 CRITICAL LIMITS FOR SODICITY TOLERANCE

Factor Ratings 1/	ESP %	ESP %	SAR 3/	SAR
	(0 - 30) 2/	(30 - 90)	(0 - 30)	(30 - 90)
s1	<10	<20	<8	<18
s2	10 - 20	20 - 35	8-18	18 - 38
s3	20 - 35	35 - 50	18 - 38	38 - 68
n	>35	>50	>38	>68

1/ Ratings may be raised one level if permeability is more than 2 cm/hr (e.g. as in loamy and sandy soils).

2/ Soil depth ranges in cm.

3/SAR may be used if ESP figures seem unreliable.

ii. Sodium toxicity

Plants vary considerably in their ability to tolerate sodium ions. Most tree crops and other woody-type perennials are particularly sensitive to low concentrations of sodium. Most annual crops are less sensitive, but may be affected by higher concentrations. Sodium toxicity is often modified and reduced if calcium is also present, therefore a reasonable evaluation of the potential toxicity is possible using the SAR for the soil water extract and the SAR of the irrigation water. Symptoms of sodium toxicity may appear only after a period of time during which toxic concentrations accumulate in the plant: the symptoms appear as a burn or drying of tissues first appearing at the outer edges of leaves. Table |40 can be used to evaluate the sodium hazard for representative crops.

iii. Chloride and boron toxicities

Critical values for tolerance for chloride and boron for various crops are given in Tables 41 and 42.

A.11 pH, micronutrients and toxicities

A.11.1 pH, micronutrient deficiencies and toxicities on non-rice cropland
A.11.2 Chemical characteristics of submerged rice soils (based on Ponnamperuma 1976)
A.11.3 Acid sulphate soils

Apart from NPK, discussed in A.7 and the toxicities caused by excess sodium, boron and chloride discussed in A.10, there remain limitations of pH, micronutrient deficiencies, and other toxicities. These are discussed for non-rice cropland, for submerged riceland, and for acid sulphate conditions in the three following sections A.11.1, A.11.2 and A.11.3.

Table 40 TOLERANCE OF VARIOUS CROPS TO EXCHANGEABLE SODIUM (ESP) UNDER NONSALINE CONDITIONS

Tolerance to ESP and range at which affected	Crop	Growth response under field conditions
Extremely sensitive (ESP = 2-10)	Deciduous fruits Nuts Citrus (Citrus spp.) Avocado (Persea americana Mill.)	Sodium toxicity symptoms even at low ESP values
Sensitive (ESP = 10-20)	Beans (Phaseolus vulgaris. L)	Stunted growth at these ESP values even though the physical condition of the soil may be good
Moderately tolerant (ESP = 20-40)	Clover (Trifolium spp.) Oats (Avena sativa L.) Tall fescue (Festuca arundinacea Schreb.) Rice (Oryza sativa L.) Dallis grass (Paspalum dilatum Poir.)	Stunted growth due to both nutritional factors and adverse soil conditions
Tolerant (ESP = 40-60)	Wheat (Triticum aestivum L.) Cotton (Gossypium hirsutum L.) Alfalfa (Medicago sativa L.) Barley (Hordeum vulgare L.) Tomatoes (Lycopersicon esculentum Mill.) Beets (Beta vulgaris L.)	Stunted growth usually due to adverse physical conditions of soil
Most tolerant (ESP = more than 60)	Crested and Fairway wheatgrass (Agropyron spp.) Tall wheatgrass (Agropyron elongatum (Host) Beau.) Rhodes grass (Chloris gayana Kunth)	Stunted growth usually due to adverse physical conditions of soil

Note- Estimates of the equilibrium ESP can be made from the irrigation water or more preferably from the SAR of the soil saturation extract using the nomogram in Appendix B of FAO Irrigation and Drainage Paper No. 29. This estimation method is not applicable where soil gypsum is present. Effectiveness of any planned corrective action should be field tested before being applied on a large scale. Soils at ESP = 20-40 and above will usually have too poor physical structure for good crop production. The research results given above were obtained with soils whose structure was stabilized with Krilium.

Source: Pearson 1960. For updated information see revised FAO Irrigation and Drainage Paper No. 29.

Table 41 CHLORIDE TOLERANCES IN THE SATURATION EXTRACT OF SOIL FOR FRUIT CROP ROOTSTOCKS AND VARIETIES IF LEAF INJURY IS TO BE AVOIDED

Crop	Rootstock or variety	Maximum permissible C1 in saturation extract me/l
	Rootstocks
Citrus (Citrus spp.)	Rangpur lime, Cleopatra mandarin	25
	Rough lemon, tangelo, sour orange	15
	Sweet orange, citrange	10
Stone fruit (Prunus spp.)	Marianna	25
	Lovell, Shalil	10
	Yunnan	7
Avocado (Persea americana Mill.)	West Indian Mexican	8
Grape (Vitis spp.)	Salt Creek, 1613-3	40
	Dog Ridge	30
	Varieties
Grape (Vitis spp.)	Thompson Seedless, Perlette	25
	Cardinal, Black rose	10
Berries 1/ (Rubus spp.)	Boysenberry	10
	Olallie blackberry	10
	Indian Summer raspberry	5
Strawberry (Fragaria spp.)	Lassen	8
	Shasta	5

1/ Data available for single variety of each crop only.

Source: Bernstein 1965. For updated information see revised FAO Irrigation and Drainage Paper No. 29.

Table 42 RELATIVE TOLERANCE OF CROPS AND ORNAMENTALS TO BORON 1/

Tolerance decreases in descending order in each column

Tolerant	Semi-tolerant	Sensitive
4.0 mg/l of boron	2.0 mg/l of boron	1.0 mg/l of boron
Athel (Tamarix aphylla)	Sunflower, native (Helianthus annuus L.)	Pecan (Carya illinoensis (Wang.) K. Koch)
Asparagus (Asparagus officinalis L.)	Potato (Solanum tuberosum L.)	Walnut, black and Persian or English (Juglans spp.)
Palm (Phoenix canariensis)	Cotton, Acala and Pima (Gossypium sp.)	Jerusalem artichoke (Helianthus tuberosus L.)
Date palm (P. dactylifera L.)	Tomato (Lycopersicon esculentum Mill.)	Navy bean (Phaseolus vulgaris L.)
Sugarbeet (Beta vulgaris L.)	Sweetpea (Lathyrus odoratus L.)	American elm (Ulmus americana L.)
Mangel (Beta vulgaris L.)	Radish (Raphanus sativus L.)	Plum (Prunus domestica L.)
Garden beet (Beta vulgaris L.)	Field pea (Pisum sativum L.)	Pear (Pyrus communis L.)
Alfalfa (Medicago sativa L.)	Rose (Rosa sp.)	Apple (Malus sylvestris Mill.)
Gladiolus (Gladiolus sp.)	Olive (Olea europaea L.)	Grape (Sultanina and Malaga) (Vitis sp.)
Broadbean (Vicia faba L.)	Barley (Hordeum vulgare L.)	Kadota fig (Ficus carica L.)
Onion (Allium cepa L.)	Wheat (Triticum aestivum L.)	Persimmon (Diospyros virginiana L.)
Turnip (Brassica rapa L.)	Corn (Maize) (Zea mays L.)	Cherry (Prunus sp.)
Cabbage (Brassica oleracea var. capitata L.)	Milo (Sorghum bicolor (L.) Moench)	Apricot (Prunus armeniaca L.)
Lettuce (Lactuca sativa L.)	Oat (Avena sativa L.)	Thornless blackberry (Rubus sp.)
Carrot (Daucus carota L.)	Zinnia (Zinnia elegans Jacq.)	Orange (Citrus sinensis (L.) Osbeck)
	Pumpkin (Cucurbita spp.)	Avocado (Persea americana Mill.)
	Bell pepper (Capsicum annuum L.)	Grapefruit (Citrus paradisi Macfad.)
	Sweet potato (Ipomoea batatas (L.) Lam.)	Lemon (Citrus limon (L.) Burm. f.)
	Lima bean (Phaseolus lunatus L.)
2.0 mg/l of boron	1.0 mg/l of boron	0.3 mg/l of boron

1/ Relative tolerance is based on boron in irrigation water at which boron toxicity symptoms were observed when plants were grown in sand culture. Does not necessarily indicate a reduction in yield.

Source: Wilcox 1960

Figure 17a General trend of the influence of reaction (pH) on the availability of plant nutrients (widest part of the bar indicates maximum availability) - Relative availability of common elements in mineral soils with pH (after Truog 1948)

Figure 17b General trend of the influence of reaction (pH) on the availability of plant nutrients (widest part of the bar indicates maximum availability) - Organic soils (after Lucas and Davis 1961)

A.11.1 pH, micronutrient deficiencies and toxicities on non-rice cropland

i. pH (General)

Crops vary in their response to pH; calcifuge plants dislike lime while calciphilous plants are lime-loving. There are very few crops that grow well in calcareous soils that do not grow equally well at a pH above 6 under lime-free conditions. Several crops, such as tea, require acid conditions. Many crops are affected by micro-nutrient deficiencies or toxicities at certain pH levels. The availability of various macro and micronutrients over the pH scale is illustrated in Figure 17; however, this availability varies from crop to crop.

The pH of soil suspensions varies according to whether the soil is shaken with water, or with an electrolyte such as normal potassium chloride. The pH of the latter may be a whole pH unit lower than that measured by shaking with water, but closer to the real pH on the soil particles themselves. It is best to use an electrolyte where the pH of saline soils is to be measured, especially when comparisons are necessary with nonsaline pHs.

In soil/water suspensions the pH may vary with the soil to water ratio. In the field, as the soil gets drier, the concentration of salts in the soil solution may rise, causing a fall in pH. If the soils contain substances susceptible to oxidation and reduction, the pH will fall or rise accordingly. Thus the pH of waterlogged soils containing sulphides will fall from pH 7 to below pH 4 if drained and aerated (see A.11.3). The pH of a soil is also influenced by the carbon dioxide concentration of the soil air; the higher this concentration the lower the pH, the effect being greater the higher the soil's pH. Because the pH of soil samples collected from the field depends on the conditions of measurement, collection and measurement should be standardized in any study. i

ii. Calcium and lime requirement

The amount of liming material required to neutralize soil acidity, and the final pH it is desirable to achieve, must generally be worked out under local conditions in field trials. In the laboratory, the amount of lime needed to bring the soil to a selected pH can be determined either by titrating the soil with lime to this pH or, more conveniently, by shaking the soil with a calcium solution buffered at this pH-

In many acid tropical soils, the object of applying lime is to neutralize exchangeable aluminium, not to achieve a particular pH and this seems to overcome the problems associated with over-liming. On some acid soils, over-liming may induce micronutrient deficiencies and the lime must be given in small, frequent dressings, keeping a check on pH and crop performance whenever additional dressings of lime are given.

Liming materials include calcium oxide (burnt or quick lime), calcium hydroxide (hydrated or slaked lime), finely-ground limestone, and chalk. In wet climates calcium bicarbonate is continually leached out of the soil; in the United Kingdom, for example, this is equivalent to about 200-400 kg of calcium carbonate per hectare per year (Russell 1973).

Calcium deficiency, particularly in fruit and vegetable crops, can occur on some acid soils although the harmful effects of acid soils are more usually caused by aluminium, iron, manganese or sulphur toxicities.

iii. Magnesium

Magnesium deficiency commonly occurs on acid, sandy soils in areas of moderate to high rainfall. Magnesium deficiency may be induced by applying too much potassium fertilizer, and occasionally even by mulching with grasses rich in potassium. The application of nitrogen tends to promote the uptake of magnesium. In sandy soils subject to leaching, soils with equal amounts of available magnesium may be more subject to magnesium deficiency at a low pH than at a higher pH (Chapman 1973).

Soils with high exchangeable magnesium and exhibiting the morphology and problems of sodic soils occur in western Canada, USA and Middle Eastern arid and semi-arid areas. Low permeability and intractable working conditions are more important than excess magnesium on these soils.

iv. Zinc

Zinc deficiency is very widespread in neutral to alkaline soils. Excessive soil phosphorus aggravates zinc deficiency. Some crops are affected more than others. For example, phaseolus beans, maize, potato, onion, citrus, cherry and peach are susceptible, whereas alfalfa, wheat, barley and grasses are rarely affected. A good prediction of zinc deficiency can be obtained using DTPA-extractable zinc which, if below 0.8 ppm on a dry soil basis, indicates the need for zinc applications to susceptible non-rice crops (see A.11.2 for rice). For field crops in arid and semi-arid areas, 10 kg zinc per hectare broadcast and incorporated into the soil will control zinc deficiency for three or four years. Foliar applications of 1-2 kg Zn/ha/yr to tree crops are very effective.

v. Iron

Iron deficiency, or lime-induced chlorosis, on calcareous soils is a very complex problem influenced by many physical, chemical and biological factors. Iron is absorbed by plants as Fe (II) which is relatively soluble, but in the pH range of neutral to alkaline soils this is rapidly converted to Fe (III) which is very insoluble. Foliar deficiency symptoms are quite specific for iron. Diagnostic soil tests for available iron have been generally unsuccessful and often are of little use as the condition is difficult to correct. However, Lindsay and Norvell (1978) in the USA and Stewart-Jones (1979) in Saudi Arabia have used DTPA extraction with some success. Additions of ferrous sulphate to soils which are grossly deficient in iron can sometimes produce substantial increases in yields; the response is dependent on microbial activity and the presence of organic matter. Lime-induced chlorosis within land units is generally very variable and unpredictable, but many soils will produce good overall production and yields, despite lower yields in affected patches of fields. Some crops have 'iron efficient' and 'iron inefficient' cultivars and with tree crops, the chlorosis often varies markedly from tree to tree. Tissue analysis does not identify this problem satisfactorily, but it is readily evident from leaf chlorosis.

vi. Sulphur

Sulphur deficiency occurs on old deeply weathered land surfaces where the soils have been strongly leached for a long period of time, and is only rarely found in arid areas. The sulphur supplying power of a soil can be estimated from the amount of water-soluble and absorbed sulphate in the root zone, for crops can use this absorbed sulphate quite readily. Less than 3 ppm SO₄-S in soil extracted with lithium chloride solution was well correlated with the yield of alfalfa in S-deficient soils of southern Idaho. Sulphur deficiency is readily corrected by sulphate containing fertilizers (e.g. superphosphate or sulphate of ammonia) or gypsum.

Sulphur toxicity occurs on acid-sulphate soils as discussed in A.11.3.

vii. Boron

Boron deficiency does not usually occur on arid and semi-arid land where boron toxicity is a much more probable occurrence as already discussed under A.10 and Table 42. Boron deficiency frequently occurs on calcareous soils, or on acid soils that have been limed, particularly when plants such as sugarbeet are under water stress in dry periods. Deficiency often occurs when hot water soluble boron of the soil is less than 0.3 to 1.0 ppm but the predictive value of this test is not always satisfactory. Plant tissue analysis is very reliable for confirming boron deficiency in plants. Note that, whereas boron deficiency is usually determined using the hot water extraction process, boron toxicity is generally identified from boron in the saturation extract of soils.

viii. Copper

Copper deficiency occurs on many ancient strongly-weathered soils (e.g. in Australia) which are low in copper, and on some sandy soils, especially calcareous sands and peats.

ix. Manganese

Manganese deficiency in neutral to alkaline soils is often associated with deficiencies of iron and zinc but rarely, if ever, both. Manganese deficiency, either alone or in combination with other elements occurs much less often than zinc and iron deficiencies. It is rarely found in field or vegetable crops in irrigated regions but is commonly a limitation in citrus and deciduous tree crops.

x. Molybdenum

Molybdenum deficiency is usually found only on acid soils and it can often be cured by liming or, more cheaply, by applications of sodium or ammonium molybdate to the soil, crop or seed. Poor nitrogen fixation by legume crops is associated with molybdenum deficiency.

xi. Aluminium

Aluminium toxicity accompanied by manganese and iron toxicity occurs on acid soils over large areas of oxisols and ultisols subject to seasonal wetting and drying in the humid tropics. Aluminium toxicity at pH values of less than 5 is one of the main causes of limited root penetration of annual crops, such as cotton, below certain soil depths (Pearson 1974). (For acid sulphate soils see A.11.3.)

In Table 43 international ranges of soil and plant analyses are tabulated, using data obtained by Sillanpää and as reported in Soils

Table 43 INTERNATIONAL AVERAGES AND RANGES OF SOME SOIL AND PLANT (MAIZE) ANALYSES 1/
(Number of samples = 1 976 from 25 countries, depth of sampling 0-20 cm)

		Units	Mean	± Standard deviation	Minimum	Maximum
General Soil Properties
Particle size distribution
	<.002	%	28	18	1	87
	.002 -.06	%	36	20	1	87
	.06-2	%	35	26	0	97
	> 2 mm	%	1	5	0	56
Texture index			42	17	9	91
Cation exchange capacity (unbuffered CEC)		me/100 g	27.0	15.4	2	99.7
pH (H2O)			6.92	1.05	4.00	8.95
pH (CaCl2)			6.40	1.11	3.62	8.55
Electrical conductivity (1:2.5 soil:water)		dS/m	2.1	3.2	.1	73.0
CaCO3 equivalent		%	2.8	7.1	0	67.1
Organic C		%	1.3	1.3	.1	39.1
N (total)		%	.135	.088	.008	1.657
Bulk density (disturbed, air-dried)		g/cm3	1.2	.14	.47	1.77
Soil Macronutrients 2/
N (total)		mg/l	1 547	793	97	14 729
P (0.5M Na bicarbonate, pH 8.5, Olsen)		mg/l	22.5	33.0	.1	656.0
Extractable K (1M ammonium acetate, pH 7) 3/		mg/l	330	356	18	5 598
Extractable Ca (ditto)		mg/l	3 450	2 815	10	17 995
Extractable Mg (ditto)		mg/l	446	462	2	6 490
Extractable Na (ditto)		mg/l	86	260	0	4 058
Soil Micronutrients 2/
B (hot water extraction) 4/		mg/l	.65	.71	.05	10.02
Cu (AAAc-EDTA extraction) 5/		mg/l	6.0	7.9	.1	99.7
Fe (AAAc-EDTA)		mg/l	160	139	14	2 275
Mn (DTPA extraction 6/		mg/l	43.2	38.4	.9	378.4
Mo (AO-OA extraction) 6/		mg/l	.212	.273	.010	3.560
Zn (DTPA extraction)		mg/l	2.14	6.49	.09	185.20
Maize Plant Nutrient Contents 7/ (grown on the above soils)
N		%	3.14	.87	.88	6.51
P		%	.330	.104	.050	1.038
K		%	3.13	.96	.58	6.71
Ca		%	.470	.205	.091	1.880
Mg		%	.251	.119	.036	1.125
B		ppm	9.24	8.00	1.88	100.04
Cu		ppm	11.6	4.2	2.0	99.6
Fe 8/
Mn		ppm	77.6	47.8	8.3	517.1
Mo		ppm	.86	1.35	.01	21.03
Zn		ppm	35.7	47.2	6.2	915.6

Footnotes for Table 43

1/ Source: Sillanpää, FAO Soils Bulletin 48, p. 433. This publication provides details of analytical methods and results from the study in which soil and plant samples were obtained from the 25 countries. Approximately two-thirds of the samples fall within the range represented by the mean plus or minus the standard deviation. See Sillanpää, p. 441 for a breakdown of results by FAO-Unesco soil units.

2/ Results reported here as mg/l are on a volume basis. They may be converted to a weight basis, i.e. to obtain extractable cations in traditional units of milli-equivalents per 100 g dry soil, divide the value in mg/l by the product (equivalent weight x bulk density x 10).

3/ 'Extractable' may be similar but not identical with 'exchangeable'; the importance of (i) the intensity of leaching with the extractant, (ii) presence of soluble cations not adsorbed on the exchange complex, and (iii) pH, should be appreciated if results are to be reported as 'exchangeable'.

4/ As an index of plant availability, soil B values may be corrected according to CEC (see Sillanpää p.49).

5/ Acid ammonium acetate-EDTA extractant (Sillanpää p.10). As an index of plant availability, soil Cu values may be corrected according to the organic C content (Sillanpää p. 56).

6/ The Mn content of plants decreases with increasing pH. As an index of plant availability, soil Mn values may be corrected according to pH (Sillanpää p.67). A pH correction factor may also be applied to soil Mo extracted with ammonium oxalate-oxalic acid solution.

7/ Plant Nutrient Contents expressed on a dry matter (105°C) basis. The results given are for field grown maize samples submitted with the soil samples. Pot-grown wheat plant analyses are also presented in Sillanpää, Appendix 4.

8/ Results not reliable, possibly because of contamination with soil.

Bulletin 48. About 2 000 soil and indicator plant samples were obtained by Sillanpää from representative soils in 25 countries. In Table 43 approximately two-thirds of the samples fall within the range represented by the mean plus or minus the standard deviation. Minimum and maximum values in this population of samples are also tabulated. The values given in this table can be used as guidelines, and the reader is referred to Soils Bulletin 48 for more detailed explanation of the analytical methods used.

A.11.2 Chemical characteristics of submerged rice soils (based on Ponnamperuma 1976)

Flooded rice soils undergo chemical changes that differ from those that are relevant under dryland conditions. In many parts of the world, the newer rice varieties have given disappointing results because soil problems have affected these newer varieties; this had earlier gone undetected owing to the use of traditional rice varieties tolerant to the adverse soil conditions. Iron toxicity on acid soils, phosphorus deficiency on ultisols, oxisols, vertisols and andepts, zinc deficiency on sodic, calcareous and peat soils, and iron deficiency on soils of high pH are the main problems to be identified in evaluating the effects of submergence in the cultivation of rice. Acid sulphate conditions are also potentially important for rice if the soils dry out periodically.

Important chemical changes (Ponnamperuma 1976) that have implications in land evaluation for wetland rice are given in the following eight paragraphs.

i. Change in pH: Within a few weeks of submergence the pH of acid soils increases and the pH of sodic and calcareous soils decreases. Thus the submergence causes the pH values of most acid and alkaline soils to converge between 6 and 7. The rate and degree of the pH changes depend on soil properties and temperature. Soils that have an adequate amount of organic matter (>2%) and active iron (>1%) and that are low in acid reserves attain a pH of about 6.5 within a few weeks of submergence. However, if acid soils are low in organic matter or in active iron, or are high in acid reserves, they may not attain a pH of more than 5 even after months of submergence. Thus, in evaluating land for rice production, the pH of the dry soil may not be as important as the factors that influence pH kinetics on soil submergence.

ii. Changes in salinity: The electrical conductivity of the soil solution after submergence increases with time, reaches a peak, and then decreases. Most submerged soils, regardless of their initial conductivities, have values exceeding 2 dS/m during a good part of the growing season. Conductivities are highest in saline soils and lowest in leached ultisols and oxisols and the course of conductivity changes varies markedly with the soil. Changes are highly correlated with the concentration of iron and manganese in the soil solutions of acid soils and with the calcium and magnesium bicarbonate concentration in alkaline soils.

The salinity hazard in flooded soils may be greater than the ECe values of the soil immediately after submergence may indicate. Soil reduction and the solvent action of carbon dioxide release large amounts of ions into the soil solution, but due to dilution it may be less than the ECe values may suggest.

iii. Reduction of Fe (III) to Fe (II): The most dramatic change that occurs when a soil is submerged and undergoes reduction is that Fe (III) oxide hydrates are reduced to Fe (II) compounds. Consequently, the soil colour changes from brown to grey, and large amounts of Fe (II) enter the solution phase. The concentration of water-soluble iron, which at submergence rarely exceeds 0.1 mg/l, may rise to 600 mg/l within a few weeks after flooding, it then declines or reaches a plateau. In acid sulphate soils the peak values may be as high as 5 000 ppm.

Iron toxicity may be a hazard for wetland rice on soils for which the main drawbacks for dryland crops are manganese and aluminium toxicities and a deficiency of macro-elements. Thus iron toxicity is common in submerged ultisols, oxisols, and acid sulphate soils in the tropics. It may also occur in acid sandy soils and in peat soils low in active iron, as in Akiochi soils. Low temperatures (<20°C), by bringing about late but high and persistent concentrations of water-soluble iron, may cause iron toxicity in soils in which, at 25-30°C, high concentrations are short-lived. Characteristics for predicting iron toxicity hazard are:

a. pH of the dry soil;
b. amount of reserve acidity?
c. reactivity and content of Fe (III) oxide hydrates;
d. soil temperature;
e. salt content;
f. percolation rate;
g. interflow from adjacent areas.

iv. Increase in supply and availability of nitrogen: (see heading A.7.1).

v. Increase in availability of phosphorus, silicon, molybdenum: The availability of phosphorus and silicon, whether judged by chemical methods or by plant uptake, increases on submerging a soil.

The concentration of water-soluble molybdenum increases on flooding, presumably as a result of desorption following reduction of ferric oxides. This may benefit nitrogen-fixing algae at the surface, anaerobic bacteria in the reduced soil, and aerobic bacteria on the roots.

vi. Decrease in concentrations of water-soluble zinc and copper: A decrease in concentrations of water-soluble zinc and copper is one of the few disadvantages of flooding soils for rice. Since 1966, zinc deficiency has been recognized as a widespread nutritional disorder of rice on sodic and calcareous soils. Recent work suggests that zinc deficiency (and perhaps copper deficiency) is a serious obstacle to the growth of rice on continuously wet soils and peat soil (Ponnamperuma 1965 and 1972). These deficiencies may not be as acute for dryland crops grown on those soils after they are drained. Therefore, the possibility of zinc and copper deficiencies should always be considered in evaluating land for wetland rice.

vii. Production of toxins: These include organic reduction products, organic acids, ethylene and hydrogen sulphide. Hydrogen sulphide is produced in submerged soils as a result of sulphate reduction and anaerobic decomposition of organic matter. In normal soils it is rendered harmless by precipitation as ferrous sulphide, but in soils high in sulphate and organic matter and low in iron, it may harm rice plants (see A.11.3).

viii. Implications for land evaluation for wetland rice: The chemical changes brought about by soil submergence may drastically alter the category in which a soil is placed on the basis of characteristics for dryland soil. Some soils may shift from suitable to unsuitable and vice versa. The same chemical changes, along with inherent soil properties, complicate enormously the evaluation of problem soils. Table 44 lists some of the growth limiting factors likely to be important on various soils.

A.11.3 Acid sulphate soils

Acid sulphate conditions may be anticipated where it is intended to drain submerged soils high in sulphate and organic matter, e.g. in mangrove swamps. Aeration of these soils when drained can lead to the oxidation of sulphur compounds and acidification to a very low pH. This acidification is potentially a problem in many coastal areas subject to tidal influence from saline sea water, especially mangrove swamps that are to be drained and reclaimed. Many tens of thousands of hectares in the humid tropics could be brought under rice or oil palm cultivation provided the soils are prevented from drying by careful water control all the year round.

To distinguish soils that are potentially hazardous from those that are not, the severity of acidification on drying can be measured in the field or laboratory from changes in pH. In the field, pH measurements can be made with 1:5 soil-water suspensions, as soon as possible after the time of sampling. This establishes the normal field values for the unreclaimed soil. To determine the effects of oxidation on these soils when they are drained, duplicate samples may be exposed to air by loosening the necks of polythene storage bags. Measurements are then made at intervals to monitor changes in pH. If the pH drops to less than 4 within 30 days, a level of soluble aluminium approaching 2 mg/l is possible which is harmful to rice. At a pH of 3.6 the soluble aluminium could be as high as 43 mg/l. Ferrous iron levels above 500 mg/l are also harmful to rice and other crops and in some acid sulphate soils can rise to 5 000 mg/l.

Table 44 GROWTH LIMITING CONDITIONS FOR RICE ON SUBMERGED SOILS OF VARIOUS TYPES

KIND OF SOIL AND MAIN LIMITATIONS	OTHER GROWTH LIMITING CONDITIONS
Saline soils
Arid saline soils	Alkalinity, Zn deficiency, N & P deficiencies
Acid coastal saline soils	Iron toxicity, P deficiency, deep water
Neutral and alkaline coastal and saline soils	Zn deficiency, deep water
Deltaic and estuarine acid sulphate soils	Iron toxicity, P deficiency, deep water
Coastal histosols	Nutrient deficiencies, H2S toxicity, toxicity of organic substances, deep water, Fe toxicity
Acid sulphate soils
Coastal soils	Salinity, Fe toxicity, N & P deficiencies, deep water
Old inland soils	N & P deficiencies
Histosols	Fe toxicity, H2S toxicity, nutrient deficiencies, deep water, salinity
Iron-toxic soils
Acid sulphate soils	Salinity, N & P deficiencies, deep water
Acid oxisols and ultisols	P deficiency, low base status, low Si content
Histosols	H2S toxicity, toxicity of organic substances, macro-nutrient deficiencies, Zn and Cu deficiencies, deep water
Phosphorus deficiency in wetland rice
Acid sulphate soils	Strong acidity, iron toxicity, low nutrient status, base deficiency, salinity
Acid oxisols and ultisols	Iron toxicity, base deficiency
Vertisols	Zinc deficiency, iron deficiency, salinity, alkalinity
Zinc deficient soils
Saline-sodic and sodic soils	Salinity, N & P and Fe deficiencies
Vertisols	P and Fe deficiencies, salinity, alkalinity
Calcareous soils	K deficiency
Wet soils	Cu deficiency
Histosols	N, P, K, Si, Cu, deficiencies; H2S toxicity, deep water

Source: after Ponnamperuma 1976.

Where laboratory facilities exist, tests can be carried out on duplicate soil samples that have been maintained free of air. One set is oxidized with hydrogen peroxide in the laboratory before analysing and the other is not. Total pyrite (FeS) is determined from the difference between the analyses, and its concentration indicates if the soil is potentially acid sulphate.

Potentially acid sulphate soils often have horizons with a soft buttery consistency, making the material difficult to extract with an auger. These soils are often conformable with Rhizophora mangrove and Nipah palm vegetation; there is often a smell of hydrogen disulphide in soil pits or auger holes, a low pH on drying, and a high sulphate content.

A.12 Pests, diseases and weeds

Insufficient regard for potential pest, disease and weed problems in land evaluation quite commonly results in the choice of unsuitable cropping systems and rotations leading to poor performance of irrigation projects.

The categories of problem may be listed as due to (i) wild animals, (ii) arthropods including insects and mites, (iii) parasitic nematodes, (iv) fungal pathogens, (v) bacterial pathogens, and (vi) virus diseases. In reconnaissance studies these should be considered in selecting alternative LUTs.

Pests, diseases and weeds may be 'class-determining' because of the variability from one land unit to another in exposure to wild animals, in microclimate or soils, or in other land characteristics. Insect problems, particularly in cotton, have led to the failure of large irrigation schemes (e.g. in Australia).

Certain crops need protection by fencing against wild animals and theft. This could be considered as an investment and land development cost under Section C. Site and aspect affecting microclimate may cause increased incidence of many fungal and bacterial leaf diseases. Cool temperatures at the base of slopes may downgrade the land because of proneness to diseases- Humid sites may be more disease-prone since the number of hours during which the leaf surface is wet often encourages fungal and bacterial pathogens, and reduces the effectiveness of control measures.

Poorly drained soils predispose certain crops to root and foot rots (see A.5). Nematode problems may be more severe on sandy soils than on clay soils.

Weed problems are often under-evaluated. Tens of thousands of hectares of land have been abandoned due to the difficulty of controlling weeds on certain soils. The impracticability of weed control during periods of wet weather on heavy soils restricts the range of crops that can be grown on non-rice cropland. Weeds that are not a problem early in the life of a project may become so with time or vice versa. For a list and details of the world's worst 100 weeds see Holm et al. (1977) and for rice land weeds see Moody (1981).

The cost of pesticides, herbicides and labour, etc. to control pests, diseases and weeds, including activities such as bird scaring, is an input to farm budgets, but initially can be scored using factor ratings.

Long-term hazards of pest, disease and weed build-up for given rotations and management should, where possible, be based on comparable experience in the locality.

A.13 Flood, storm, wind and frost

A.13.1 Flooding in rice cultivation
A.13.2 Flood hazard
A.13.3 Storm, hail and wind hazard
A.13.4 Frost hazard

The evaluation of these can be separated into sections concerned with floods and deep water in the cultivation of rice, flood hazard in general, storm and wind hazard and frost hazard as discussed in A.13.1, A.13.2, A.13.3 and A.13.4 respectively.

A.13.1 Flooding in rice cultivation

Twenty-five to thirty percent of the world's rice areas are subject to deep water flooding and grow traditional tall and floating rice varieties, and 25-30% are under shallow water (IRRI 1975b). Deep water rice land is mainly situated in densely populated valleys and the deltas of major rivers (e.g. Ganges, Brahmaputra, Godavari, Irrawaddy, Chao Phraya and Mekong), where low yielding indica varieties are grown. In the deepest areas (1-6 meters), floating rice varieties are grown. The stems of these varieties elongate as the water rises and the leaves float on the water. About 10% of the rice land in Asia and Africa is planted with floating rice; almost half of the total is planted with tall non-floating varieties adapted to medium-deep water. Land suitability depends on the varieties available and on the reliability of the flood. The speed of the rise or fall in water levels with respect to the varieties grown is very important. If the water recedes too quickly the varieties may lodge. Fast-moving water can flatten or uproot the crop or cover it with silt. Flooding by sea water causes additional damage by salts. Many rice varieties are tolerant to some submergence, or are able to grow through the deep water. Young seedlings are often more susceptible to flooding and submergence than older seedlings.

Floating rice is generally planted at the beginning of the first monsoon rains, the seeds being broadcast on dry or moist soil. During the early period, the rice grows as an upland crop and may even need to be drought tolerant at the seedling stage, photoperiodicity (see Radiation) causes flowers to form after the water starts to recede but before the end of the rains. Harvesting is sometimes from boats or it may take place after the water completely recedes and the soil is dry. | Land and water suitability depend on the reliability and depth of flooding, the duration and depth of flooding required by particular varieties arid the speed of recession of the water. The velocity of water flow is also an important factor.

It is possible that the variety of rice may be dictated by land characteristics, and if these varieties differ in yield potential, they affect land productivity and therefore the varietal requirement may be land class-determining- Five categories of rice varieties are recognized: irrigated wetland (lowland), shallow rainfed (lowland), intermediate-deep rainfed (lowland), deep water, and dry land. The varieties adapted to these various conditions have different yield potential, the greatest being for irrigated wetland and the lowest (usually) for dryland.

An example from the south coast of Java can be cited, where areas were distinguished as 'not subject to flooding', 'subject to moderate flooding' and 'subject to prolonged flooding'. Two crops of high yielding modern rice varieties could be grown on the land not subject to flooding, and the benefit of this could be offset against the cost of protecting additional land from flood. Tall rice varieties (some improved) were suitable for land subject to temporary flash flooding. The important aspects to be evaluated were the risk associated with different dates of planting, and whether the use of improved shorter duration rices would allow two crops instead of one per year. Benefits would then be set off against flood control and irrigation costs. In the areas subject to prolonged seasonal flooding, only one dry season rice crop could be grown. The rice production, input-output and land development investment costs on the different land units were expressed in economic terms to define land suitability classes. Alternatively, different LUTs could have been created but this would have led to too many combinations of LUTs and land units for evaluation.

A.13.2 Flood hazard

In shallow water rice areas and in areas producing other crops, spasmodic floods not only affect the crop, but also damage the soil and the infrastructure, e.g. rice-field bunds, pathways, temporary and permanent houses, roads and bridges etc. Flood damage is most likely to occur on river flood plains, alluvial and coastal plains, regions with large seasonal variations in rainfall and liable to intensive rain over hours or days. The detailed pattern of incidence is thus related to landforms.

In setting critical limits for flood hazard, two criteria may be used: period of inundation, and flood frequency. The period of inundation is the average number of days during the cropping season or year when the land is covered by water. This may be obtained from records or estimated. The flood frequency is the probability of occurrence of damaging floods during the year. A damaging flood is one that destroys or causes severe damage to the crop, land or infrastructure. Where required, a damaging flood may be defined quantitatively in terms of period of inundation and/or speed of flow or volume of discharge of moving water. The following scale can be applied quantitatively where data are available, but will usually form the basis for subjective estimation.

Frequency of damaging floods:

Very rare or never	Less than 1 year in 20 or never known to occur
Rare	Less than 1 year in 5
Infrequent	Between 1 year in 5 and one per year
Very frequent	More than 5 times per year

Particularly where rice is grown, it may be necessary to distinguish between floods with a low current, which may be beneficial, and floods with a strong current, which may damage field structures. An example of critical limits for floods of these two types in the Sudan is given in Table 45.

Table 45 EXAMPLE OF CRITICAL RANGES FOR ASSESSMENT OF FLOODING HAZARD

Factor rating	Frequency of Flooding
	Flooding with strong current during growing season	Flooding with low current
s1	Once every 10 years	None - 2 days
s2	Once every 6-10 years	2 days - 3 weeks
s3	Once every 3-5 years	3-20 years
n	Every 1-2 years	More than 20 weeks

A.13.3 Storm, hail and wind hazard

The exposure of land to storm and wind and the susceptibility or tolerance to these for different crops often needs assessment in land evaluation. A judgement needs to be made of the economic impact which is probable for respective land units and crops. Two aspects are the general prevalence of the hazard (e.g. wind) and the occurrence of special events such as high intensity rainfall, cyclones and hurricanes. The latter are considerations in the selection of LUTs, but the extent of the damage and the ability of the crop to survive and sustain production after the event may be aggravated at specific sites, which could be differentiated into factor ratings. Amongst crops there is a clear distinction between short-term crops and perennial crops. The survival of short-term crops in the event of an infrequent storm hazard is of less consequence than for tree crops and orchards which might be completely destroyed. Bananas have the capability of regrowth from underground shoots if the above ground parts of the plant are destroyed; most tree crops do not have this capability.

Hail can severely damage or destroy crops in many parts of the world and may have a bearing on the crops chosen. Hail damage is often very localized. The possibility of insurance against hail damage may also affect the choice of crops.

A.13.4 Frost hazard

Where it occurs, frost can be an important land class-determining factor. Frost pockets occur in valley floors owing to katabatic air movements. Frost can destroy the flowers of temperate fruit crops and consequently affect yields. Rare frosts are particularly important in the case of orchards (e.g. citrus) where trees of all ages may be destroyed. Damaging frosts can be defined in terms of temperatures, duration, and periods of the year during which damage may occur using data from climatic records. Local experience is often helpful in indicating the effect of landforms (i.e. the greater incidence in valley floors and the increase in incidence with altitude).