Chapter 12

Soil Fertility Principles

Fig. 12.1 Maize deficiency symptoms of the four main limiting macronutrients in the tropics.
Nitrogen (N) deficiency in southern Malawi; phosphorus (P) deficiency in western Kenya;
potassium (K) deficiency in Nigeria; and sulfur (S) deficiency in northern Malawi. Potassium and
sulfur photos courtesy of Ray Weil, University of Maryland

https://doi.org/10.1017/9781316809785.014 Published online by Cambridge University Press

 310 SOIL FERTILITY PRINCIPLES

Soil fertility is the capacity of soils to supply essential nutri- chewing insects, but high contents can damage milling
ents to plants. Currently, we recognize 17 elements as essen- machinery. Some grasses (rice, sugar cane) respond to sil-
tial plant nutrients, three of which, carbon (C), oxygen (O) icon in Oxisols and Histosols that have very low levels of
and hydrogen (H), are taken up from the air and soil pores, layer-silicate clays or weatherable minerals. Iodine is an
and the remaining 14 from the soil. Essential means that essential element for humans to develop intelligence early
without them plants, microbes or animals would not be able in life, and is treated as one of the five “micronutrients” in
to complete their life cycles (Barker 2010). Essential elem- human nutrition (Chapter 2). Selenium is very important in
ents are not fungible; it is not possible to substitute one for animals and is commonly a component of salt licks that are
another, all are equally essential. Table 12.1 lists them in used in livestock production. It is also important in wheat
order of their typical concentration in plants, indicating the growth (Ivan Ortiz-Monasterio, personal communication,
main ionic or molecular species that plants take up, and 2016). Chromium, vanadium and arsenic are also considered
their main functions in plant physiology. essential elements for animals (Muñiz 2008). Sodium is
In addition, there are other elements that are considered everywhere in fluids of living beings, but its essentiality
beneficial but not essential – silicon (Si), iodine (I), selenium has not been proven. However, sodium is believed to be
(Se), chromium (Cr), vanadium (V), arsenic (As), sodium (Na), required for nitrogen fixation.
and cobalt (Co) being the main ones. Silicon imparts While this list seems overwhelming, nature takes care of
strength to plants to stand up straight and deters some most of these plant requirements with little help from

Table 12.1 The essential nutrient elements, in order of typical concentration in plants. Adapted from Barker (2010) and
Muñiz (2008).

Nutrient Ion or molecule Typical content Element’s main functions or a constituent of specific plant
taken up in plant leaves compounds
Carbon CO2 50% Cell growth, energy
Oxygen CO2, H2O, some 40% Oxidant
other ions below
Hydrogen H2O 5% Reductant
+
Nitrogen NO3 , NH4 3% Protein, amino acids, DNA, RNA, chlorophyll
+
Potassium K 2% Electrolytic balance, protein synthesis, turgidity
Calcium Ca2+ 1% Pectin acts as “cement” between cells, mitosis, nodulation in legumes,
bones in animals
Magnesium Mg2+ 0.5% Chlorophyll, photosynthesis, respiration, adenosine triphosphate
(ATP; key to energy transfer during photosynthesis and respiration)
Sulfur SO42– 0.3% The amino acids cysteine and methionine in protein synthesis, structural
integrity of enzymes, characteristic flavors of garlic, onion and brassicas
Phosphorus H2PO4–, HPO42– 0.2% DNA, RNA, bones (in animals), energy (ATP), mitosis
Chlorine Cl 100 ppm Evolution of oxygen during photosynthesis
Iron Fe2+, Fe3+ 100 ppm Chlorophyll synthesis, proteins, respiration, ATP synthesis, antioxidant,
nitrogen-fixation. Increases in concentration of indole acetic acid, lignins,
flavonoids and aromatic amino acids
Manganese MnO42– 50 ppm Required for photosynthesis and activation of some enzymes
Boron BO33–, H3BO3 30 ppm Presumed to be needed for movement of sugar, lignification, pollination
and seed development; no known enzymes contain boron, it is the least
understood essential element
Zinc Zn2+ 20 ppm Activator of several enzymes involved in nitrogen and carbohydrate
metabolism, photosynthesis, DNA and RNA synthesis; main ones are
alcohol dehydrogenase for respiration and carbonic anhydrase for
photosynthesis
Copper Cu2+ 5 ppm Photosynthesis, respiration, reactions with molecular oxygen, antioxidant
Nickel Ni 2+
1 ppm Activity of urease, possibly nitrogen fixation
Molybdenum MoO42– 0.1 ppm Nitrate reduction, nitrogen fixation

https://doi.org/10.1017/9781316809785.014 Published online by Cambridge University Press

 12.3 NUTRIENT CYCLING 311

Genetic
yield limit

Yield limited by zinc

Threshold
Plateau yield for phosphorus
yield for
Yield limited by sulfur
phosphorus
Plant yield

Plateau yield for nitrogen

Yield limited by phosphorus

Threshold
yield for
nitrogen

N NP NPS NPSZn
Nutrients added

Fig. 12.2 Illustration of the law of the minimum.

Fig. 12.3 A typical nutrient-uptake curve.
humans, who focus on managing nutrients that are deficient,
or in some cases those that can reach toxic levels (iron, man-
ganese, boron, sodium), as well as occasional interactions up their root system, then a “grand” period of rapid growth,
between calcium, magnesium and potassium. Note that alu- then a slowing down as plants switch to their reproductive
minum (Al) is not a nutrient, but often reaches toxic levels in stage, which is followed by a plateau and, in certain cases, a
acid soils. Environmentally, carbon (CO2, CH4), nitrogen decrease, as physiological maturity approaches (Fig. 12.3).
(NO3–, N2O, NO) and phosphorus (H2PO4–, HPO42–) can harm The ideal way is to synchronize nutrient additions with
the environment, causing global warming or eutrophication. the plant’s nutrient requirements as it grows. This is the
There are three main principles of soil fertility: the law “synchrony principle,” developed under the leadership of
of the minimum, synchrony and nutrient cycling. ecologist Mike Swift (Swift 1985), and is the second principle
of soil fertility; one which represents the agronomist’s
dream, and is seldom, if ever, realized.
12.1 The Law of the Minimum The difficulties in synchronization are caused by two
main reasons. First, there is the total asynchrony between
Justus von Liebig (1840) established the law of the minimum, the germinating seedling and basal fertilizer applications
which states that plant growth will be limited by the essential before or at planting. The seedling, as mentioned before,
element that it is most deficient in. So, if it is nitrogen – as is draws nutrients from its seed reserves and is incapable of
mostly the case – plants will respond to nitrogen applications using any basal fertilizer, which is vulnerable to leaching
until a plateau is reached. Then if phosphorus is the second until the root system develops. In the case of nitrogen, basal
most limiting nutrient, additional growth will take place with applications are also vulnerable to denitrification and
phosphorus fertilization until a yield plateau is reached, and nitrous oxide (N2O) emissions to the atmosphere, and to
so on. If the limiting factor is physical, such as soil compac- ammonia (NH3) volatilization in soils with high pH. The
tion, plants will not respond adequately until that constraint second reason is the pulse effect of soluble fertilizer applica-
is largely overcome. Even if levels of all 14 nutrients are tions, as soluble mineral fertilizers, when exposed to soil
adequate, growth may be limited by water, the genetic poten- moisture, drastically increase the ionic content in the soil
tial of the crop’s cultivar, management and, ultimately, by solution of the nutrient in question. This lack of synchrony
temperature and solar radiation. The law of the minimum is is shown in Fig. 12.4. This figure shows the crux of the
shown in Fig. 12.2, and is the first principle of soil fertility. nutrient efficiency issues, a central challenge of soil man-
agement, and one now recognized to be a major player in
global warming due to N2O emissions to the atmosphere.
12.2 Synchrony
Unlike the law of the minimum, which shows a linear 12.3 Nutrient Cycling
response with plateaus, the time course of plant nutrient
uptake (nutrient accumulation) is a parabolic curve. It is very Nutrient cycling is the third fundamental principle of soil
flat at the beginning, mostly because the seedling is using fertility. Vitousek et al. (2002) noted the following salient
seed reserves, then shows a slow growth while plants build point: cycles of soil nutrient elements are different, varying

https://doi.org/10.1017/9781316809785.014 Published online by Cambridge University Press

 312 SOIL FERTILITY PRINCIPLES

Also, nutrient cycles vary in how closed or “tight” they

are because of different loss pathways. The nitrogen cycle is
very “leaky” because of gaseous and leaching losses. The
phosphorus cycle is “tight,” but the potassium cycle is also
leaky. The question of how to loosen or tighten nutrient
cycles in agricultural systems is a priority for soil scientists.

12.4 Nutrient Uptake by Crops

and Cycling
The demand side of soil fertility is best represented by the
nutrient uptake at harvest. Tables 12.2 and 12.3 give useful
information on crops at different yield levels and their
nutrient uptake. For industrial and fruit crops, only the
nutrients actually removed from the field are indicated
Fig. 12.4 Synchrony between a crop nutrient uptake pattern (Table 12.4).
and fertilizer application is very difficult to achieve, especially at In nutrient cycling, there are major differences in how
the early stages of growth. Adapted from Brady and Weil (2008), much of the nutrients tropical ecosystems take up and
with permission by Ray Weil return to the soil. Table 12.5, complied by a team of
ecologists and agronomists (Sanchez et al. 1989), illus-
trates several key points. Tropical rainforests, both on
in speed depending on the types of bonds the elements have acid and fertile soils, produce more than twice (~4 t C/ha
with carbon, and on their stoichiometry (element ratios). per year) the above-ground litter than tropical savannas.
Organic nitrogen (N) forms covalent bonds with organic Only high-input wetland rice and a maize–maize–soybean
carbon (C) (–C–N–), while organic phosphorus (P) and system produced similar annual amounts. But the cham-
organic sulfur (S) form ester bonds (–C–O–P or –C–O–S), pion of them all was the white potato in the Andes, which
and most other nutrient elements (potassium [K], calcium produced almost 7 t C/ha of tops in only 4 months. The
[Ca], magnesium [Mg], iron [Fe], manganese [Mn], zinc [Zn], lowest value was ~0.4 t C/ha of sorghum stover in
etc.) are either bonded ionically or are in loose associations the Sahel.
with soil organic carbon (SOC). High-input agroecosystems generally add similar nutri-
The covalent bonding of N with C results in a series of ent inputs to the soils as the natural systems. For example,
soil organic nitrogen (SON) compounds that are chemically residues returned to the soil by an annual maize–maize–
recalcitrant or physically protected in the slow and passive soybean rotation in the Peruvian Amazon provided more
SON pools. Phosphorus and sulfur compounds have a high nitrogen, phosphorus, potassium and similar levels of cal-
negative charge, which is believed to prevent them from cium and magnesium than litter fall from a tropical rainfor-
being entrapped in the slow and passive SOC pools, confin- est in similar soils. Residue return in intensive wetland rice
ing them to the “active” SOC and the structural and meta- production exceeded nutrient litter fall, particularly for
bolic carbon organic input pools. The ester links between potassium, compared to a tropical rainforest on fertile soils.
organic phosphorus and organic sulfur are readily split by Fertilizer and liming applications in these high-input
extracellular enzymes, such as phosphatase and sulfatase, systems is a principal reason for this.
which are produced by roots, mycorrhizae and soil micro- As expected, this did not happen in low-input systems, but
organisms. Organic phosphorus and sulfur mineralization when aluminum-tolerant cultivars were used in acid soils,
proceeds rapidly, once the extracellular enzymes break the some systems came surprisingly close to the natural ones.
ester bonds. The resulting phosphorus and sulfur anions The upland rice–cowpea rotation, after burning a secondary
react with soil minerals in ways that nitrogen does not. forest, recycles less carbon and nitrogen, but more phosphorus
This makes nitrogen mineralization slower and more and much more potassium than the rainforest in similarly
costly in terms of energy (supplied by carbon) than the acid soils. The aluminum-tolerant Brachiaria humidicola/Desmo-
mineralization of phosphorus and sulfur. In terms of dium ovalifolium grazed pasture in the Colombian Llanos
stoichiometry, plants have C:N ratios of 100 and above, returned more carbon, nitrogen and calcium, similar levels
while soil bacteria have low C:N ratios, about 6, which of phosphorus and magnesium, but less potassium, than the
means that bacteria require more nitrogen relative to the natural tropical savanna systems (Sanchez et al. 1989). This
energy available, and in fact they are usually nitrogen- indicates that well-managed, fertilized agricultural systems
starved. Phosphorus and sulfur cycles are more flexible than can recycle nutrients back to the soil similarly to the appropri-
the nitrogen cycle (Vitousek et al. 2002). ate natural system, even though the economic product,

https://doi.org/10.1017/9781316809785.014 Published online by Cambridge University Press

 12.5 DETERMINING FERTILIZER NEEDS 313

Table 12.2 Nutrients accumulated by major cereals, in yields of cereals on a dry weight basis (12–14 percent water). Typical
values.

Crop Plant part Yields Nutrient uptake at harvest (kg/ha)

(t/ha) N P K Ca Mg
Maize Grain 1.0 25 6 15 3.0 2.0
Stover 1.5 15 3 18 4.5 3.0
Total 2.5 40 9 33 7.5 5.0

Grain 4.0 63 12 30 8.0 6.0

Stover 4.0 37 6 38 10.0 8.0
Total 8.0 100 18 68 18.0 14.0

Grain 7.0 128 20 37 14.0 11.0

Stover 7.0 72 14 93 17.0 13.0
Total 14.0 200 34 130 31.0 24.0

Rice Grain 1.5 35 7 10 1.4 0.3

Straw 1.5 7 1 18 2.6 2.2
Total 3.0 42 8 28 4.0 2.5

Grain 8.0 106 32 20 4.0 1.0

Straw 8.0 35 5 70 24.0 13.0
Total 16.0 141 37 90 28.0 14.0

Wheat Grain 0.6 12 2.4 3 0.3 1.0

Straw 1.0 3 0.8 14 2.0 2.0
Total 1.6 15 3.2 17 2.3 3.0

principally grains, have been taken away. The trick is to ensure 12.5.1 Soil Sampling
that crop residues are returned to the soil. Taking a representative soil sample is both the first step and
the largest source of error in soil fertility evaluation, because
of the magnitude of extrapolation of the analytical results.
12.5 Determining Fertilizer Needs When a 5-g subsample taken from a 500-g composite sample
of a 2.5-ha field at 15-cm depth is used for phosphorus deter-
The three soil fertility principles are used as the basis for deter- mination, it represents one-billionth (10–9) of the total soil
mining mineral and organic fertilizer needs – one of the major volume for which the analysis is made (Perur et al. 1974).
agronomic challenges, not only for increasing food production, Soil scientists assume that the distribution of printed
but also to reduce leaching and greenhouse gas emissions. instructions and diagrams on how to take soil samples is
The previous section dealt with the demand side, or the sufficient to be assured of representative specimens. Unfortu-
amounts of nutrients needed by plants, as well as the nutrients nately, this is not always the case. A representative soil sample
returned to the soil by cycling. Now we turn to the supply side, is a composite of 10–20 samples of the root zone of a field with
determining how much mineral fertilizer is needed (organic no major variation in slope, drainage, color or past fertilizer
fertilizers are discussed in Chapters 11 and 13). The main steps history. For deep-rooted plants, I recommend taking a 0–20-cm
to determine fertilizer needs are soil sampling, tackling vari- and 20–50-cm sample, and for the deepest rooted ones, a
ability, laboratory analyses, soil test correlations, modeling 50–100-cm and 100–300-cm sample. Non-representative areas,
and developing fertilizer recommendations. such as fence rows, termite mounds, those where straw has

https://doi.org/10.1017/9781316809785.014 Published online by Cambridge University Press

 314 SOIL FERTILITY PRINCIPLES

Table 12.3 Nutrients accumulated by root and grain legume crops and forage grasses. Root crops are in fresh weight, grain
legume in grain yields, as for cereals in Table 12.2, grasses in above-ground dry mass. Typical values.

Crop Plant part Yields Nutrient uptake at harvest (kg/ha)

(t/ha) N P K Ca Mg
Roots:
Cassava Roots 8.0 30 10 50 20 10
Roots 16.0 64 21 100 41 21
Roots 30.0 120 40 187 77 40
Roots 59.0 42 28 291 43 19
Whole plant 59.0 64 19 176 102 26

Potato Tubers 12.0 52 10 80 22 14

Tubers 22.0 120 20 166 40 26
Tubers 40.0 172 34 232 70 48
Sweet potato Roots 16.5 72 8 88 – –

Grain legumes:
Common bean Grain 1.0 31 3.5 6.6 – –
Soybean Grain 1.0 49 7.2 21 – –
Peanut Unshelled nuts 1.0 49 5.2 27 – –

Grasses, annual dry mass production, cut every 2 months:

Guinea grass (Panicum maximum) Above-ground 10.0 107 27 180 78 49
23.0 288 44 363 149 99
35.0 560 77 600 230 133
Pangola grass (Digitaria decumbens) Above-ground 10.0 120 22 180 36 28
23.0 299 47 358 109 67
31.0 400 53 558 130 87
Napier or elephant grass Above-ground 10.0 144 24 180 35 30
(Pennisetum purpureum) 25.0 302 64 504 96 63
46.0 800 92 900 129 87

been burned and manure piles must be avoided. Appropriate important in oxidic subsoils for estimates of anion exchange
information is also needed, including the name and address of capacity, gypsum requirements and deep SOC.
the farmer, georeference, previous crop, and fertilizer prac- Another important question is where to sample when
tices. This is best provided with a web-based questionnaire in phosphorus has been applied in bands. The answer is
the local language that can be sent electronically to a central between the bands, if their locations are known. The same
laboratory. An example for maize-growing in Mindanao, Phil- is true for conservation tillage. A third consideration is the
ippines, has been successfully developed by the International time of the year. Soil samples should be taken a long time
Rice Research Institute (IRRI 2011). before planting so that the results will be available when the
For established pastures or permanent crops where fertil- decision is made as to how much fertilizer is to be applied.
izers are not likely to be incorporated, sampling the top 20 cm In ustic soil moisture regimes, this means sampling during
is usually sufficient. For nitrate analysis, the 20–50-cm subsoil the dry season, when upward ion movement takes place.
is usually sampled, and nitrate sampling tools and procedures This may change some of the soil test results, principally in
exist. Subsoil samples, to at least 50 cm and beyond, are regard to potassium.

https://doi.org/10.1017/9781316809785.014 Published online by Cambridge University Press

 12.5 DETERMINING FERTILIZER NEEDS 315

Table 12.4 Nutrients removed by major industrial and fruit crops. Typical values. From first edition.

Crop Plant part Yields Nutrient removed at harvest (kg/ha)

(t/ha) N P K Ca Mg
Sugar cane (2-year crop) Above-ground 100 75 20 125 28 10
200 149 29 316 55 58
300 254 35 499 96 80
Cotton Seed 0.8 30 4.4 7 – –
Coffee Dry beans 1.0 25 1.7 16 1 2
Tea Dry leaves 0.6 31 2.3 15 2 –
Tobacco Cured leaves 1.0 116 14 202 – –3.0
Rubber Dry latex 3.0 7 1.2 4 4 –
Cocoa Dry beans 0.5 10 2.2 5 1 1
Oil palm Kernels 15.0 90 8.8 112 28 –
Banana Bunches 10.0 19 2.0 54 23 30
Pseudo stem and leaves – 20 1.3 22 1 3
Total – 39 3.3 76 24 33
Bunches 30.0 56 6.0 161 70 82
Pseudo stem and leaves – 29 4.0 65 2 8
Total – 85 10.0 226 72 90
Pineapple Fruit 12.5 9 2.3 29 3 –
Coconut Dry copra 1.2 60 7.2 40 – –

How often a field should be sampled depends on the ammonium (NH4+) content when both were extracted by
intensity of fertilizer use and the economic value of the crop. 1 M KCl. No changes in pH, available phosphorus, calcium
For average management intensity, once every 3 years has or magnesium were observed.
been recommended by the International Fertilizer Evalu-
ation and Improvement Program (ISFEIP; 1967–1974), while 12.5.2 Tackling Soil Variability
for very intensively managed areas, annual sampling is neces- Spatial variability in soil properties is always an issue
sary. In my opinion, deep soil sampling in oxidic soils should because soil is far from a homogeneous system and bears
be carried out once every 5 years, or at a changing point in a the footprints of generations of farmers who add temporal
rotation. Deep sampling and related shallower sampling can variability. The question is how wide-ranging is the variabil-
estimate changes in soil carbon at that time. ity. I have seen thousands of hectares of very uniform soils
As soil testing is a service program, the time between soil that follow a predictable pattern due to landscape position
sampling and its analysis should be minimized. As empha- in Oxisols of South America, where large-scale farming is
sized in the first edition, delays in shipping, power failures practiced. The uniformity of these soils results from the
and the absence of some reagents in many tropical labora- deep, weathered sediments that they are derived from; vari-
tories are unfortunately very common. Several studies have ability of these soils is minimal, and is found mainly along
been made on the effect of time in storage and drying on slopes. Local extension workers recommend sampling every
analytical results. This is still valid, four decades later, in tens or hundreds of hectares, every few years.
much of the tropics. The mixing and leveling of flooded and puddled rice soils
After arrival at the laboratory, the soil sample is dried, for decades or centuries in Asia acts essentially like the
then ground, either by pounding or by electric-powered blender we use in our kitchens, making a pretty uniform
grinders, passed through a 2-mm sieve, and assigned a slurry. Variability is thus minimized. At the other extreme,
laboratory number or stored until ready for analysis. Boul- in smallholder farms in Africa, the human footprint is enor-
din et al. (1971) studied the effect of drying versus keeping mous. Lots of natural spatial variability exists in a mosaic of
several Oxisols and Ultisols moist for a period of months. soils due to the topography and because the soils are derived
Drying decreased the exchangeable aluminum ion (Al3+) from different kinds of rock, and also from management, as
content by 20 percent and increased the exchangeable illustrated in Fig. 12.5. When people farm 1 or 2 hectares,

https://doi.org/10.1017/9781316809785.014 Published online by Cambridge University Press

 316 SOIL FERTILITY PRINCIPLES

Table 12.5 Biomass and nutrient uptake of above-ground organic resources that are recycled back to the soil in tropical
natural systems and agroecosystems, assuming total residue return, but the economic parts were harvested.
Representative values. Adapted from Sanchez et al. (1989).

System description and duration Economic Yield: Input Dry Nutrients returned to soil
(in parenthesis) yield (t/ha) residue biomass (kg/ha)
ratio (t/ha)
Ca N P K Ca
Natural systems (1 year):
Tropical rainforest, acid soilsb – – Litter fall 8.8 3960 108 3 22 53
Tropical rainforest, fertile soilsc – – Litter fall 10.5 4725 162 9 41 171
Tropical savanna d
– – Litter fall 3.8 1719 25 5 31 11

Short-cycle tropical food cropsc:

Rice (4 months)e 2.2 0.8 Straw 2.8 1260 15 2 37 11
f
Maize (4 months) 1.3 0.7 Stover 2 900 18 3 19 7
Sorghum (5 months)f 0.7 0.8 Stover 0.9 405 8 1 10 4
f
Potato (4 months) 10.7 0.7 Tops 15.1 6975 17 1 43 14
Soybean (3 months)g 1.7 0.7 Stover 2.4 1080 27 2 24 22

High-input tropical agroecosystems:

Wetland rice, two crops per year (1 year)e 11 1 Straw 11 4950 59 9 151 –
Maize–maize–soybean, three crops per year 8.7 – Stover 9.3 4185 139 15 98 –
(1 year)h
Soybeans, Cerrado (4 months)g 2.3 0.7 Stover 3.5 1575 86 8 43 –
Cocoa/Erythrina agroforestry, Brazil (1 year) i
1 0.2 Leaf litter 6 2700 81 14 71 –

Low-input tropical agroecosystems:

Upland rice–rice–cowpea, Peru (1 year)i 4.7 – Straw/ 6 2700 77 12 188 –
stover
Brachiaria humidicola/Desmodium ovalifolium – – Leaf litter 7 3153 60 5 12 –
grazed pasture, Colombia (1 year)j
Alley cropping, Inga edulis (1 year)k – – Tree 6 2700 137 10 52 –
prunings
Cocoa/Erythrina agroforestry, Brazil (1 year)i 1 0.2 Leaf litter 6 2700 81 14 71 –
a
Vitousek and Sanford (1986).
b
Sarmiento (1984).
c
Nutrient contents from DeDatta (1981).
d
Yields for the tropics from the FAOSTAT statistics database (fao.org/faostat), data for 1985, accessed in 1989.
e
Nutrient content from first edition.
f
Yields from Goedert (1986), nutrient content from Henderson and Kamprath (1970).
g
First edition; Sanchez et al. 1983; TropSoils (1987).
h
CEPLAC (1989).
i
Sanchez and Benites (1987).
j
CIAT (1985).
k
Szott (1987).

https://doi.org/10.1017/9781316809785.014 Published online by Cambridge University Press

 12.5 DETERMINING FERTILIZER NEEDS 317

they get to know their fields very well, and such knowledge 12.5.3 Laboratory Analysis
helps in determining where to sample. The most widely used way to assess the status of a nutrient in
Fields close to the household (infields) are usually more a soil is by an extraction in a wet chemistry laboratory. Soils
fertile than those farther away (outfields) because small- are shaken with an extractant, which is supposed to indicate
holder households dump garbage and kitchens ashes, and in a few minutes the amount of a nutrient a typical crop
their domestic animals often urinate and defecate in these would take up from the soil over its lifetime. These values
nearby fields. This results in a fertility gradient, illustrated are variously known as available phosphorus, potassium,
in Fig. 12.6, using a portable pH meter. etc., usually including the name of the procedure’s author
(Olsen, for example). Water is also used as an extractant, and
resin, to capture ions in a soil slurry. All these extracted
products are artifacts (like humic or fulvic acids), which
makes them difficult to relate to reflectance spectrometry
on the whole soil, but when they correlate with plant growth,
they become useful metrics. Because of its mobility in soils,
there are no similar soil tests for nitrate nitrogen (nitrate-N).
The wet chemistry laboratory is the backbone of a soil
testing program. Unlike research laboratories, service
laboratories must be geared to handle large numbers of
samples, rapidly and accurately. A complete system of
semi-automated apparatus was developed by the ISFEIP for
tropical laboratories in the 1970s and 1980s. This system is
summarized below, abstracted from the first edition of this
book, when ISFEIP was being rapidly scaled up in Latin
America and India. Other models are now in place.
Soil samples are measured volumetrically, eliminating
the time-consuming weighing process, and then placed in
multiple-unit trays, where the extracting and diluting
solutions are added or transferred by specially designed
Fig. 12.5 In-field variability in a maize field in Koraro, Ethiopia.
diluter-dispensers. Other apparatus transfers the aliquots
Photo courtesy of Ray Weil
to spectrophotometers and pH-measuring units

Fig. 12.6 Home garden soils are almost always much more fertile than soils in the larger fields away from home. Mbola, Central
Tanzania. Courtesy of Ray Weil

https://doi.org/10.1017/9781316809785.014 Published online by Cambridge University Press

 318 SOIL FERTILITY PRINCIPLES

automatically. Shaking, stirring and cleaning apparatus are

also semiautomatic and capable of handling 30 units at a
time. A single technician can run 100 soil samples a day,
measuring 10 determinations per sample. The pieces of
apparatus are based on manual operation and use a min-
imum of electrical power. Expensive electronic equipment,
such as atomic absorption spectrophotometers, is used only
during the last stages. An additional advantage is that the
same units can be used for plant tissue as for soil analysis.
The bulk of the work is done with two extractants: the
dilute double acid (Mehlich) or the modified Olsen extrac-
tant for determining available phosphorus, potassium, cal-
cium, magnesium, sodium, iron, manganese, zinc and
copper; and the 1 M KCl extraction for aluminum. Routine
tests for nitrogen, boron, sulfur and molybdenum could not
be adapted to the system.

12.5.4 Soil Test Correlation

A soil test value per se is worthless; it is an empirical number
that may or may not indirectly reflect nutrient availability,
and as mentioned before, what is measured is an artifact of Fig. 12.7 The Cate–Nelson method for determining the
the extraction procedure. Soil test values become useful only critical soil test level. In this case the critical level is 6 ppm
when they are correlated with crop responses. Such correl- available phosphorus by the Bray 1 method, for sugar cane, in
ations are usually conducted at two levels: an exploratory the State of Pernambuco, Brazil. Each dot represents a field plot
one in the greenhouse with a large number of widely diverse (ISFEIP 1967)
soils, and a more definite one in the field with fewer, but
carefully selected, soils. Correlations can be done in many
ways, using fairly complicated formulas, but in my view the 12.5.5 Curvilinear Models
simplest and most practical one is to plot relative yields These classic models are based on the law of diminishing
versus the soil test levels as a scatter diagram. Then simply returns, where curvilinear functions (quadratic, square-root,
use a transparent overlay with a cross, placing it where the logarithmic, Mitscherlich, etc.) are fitted to the yield
number of points in the upper left quadrant and in the lower response data. Statistical techniques determine which func-
right quadrant are at the minimum. This Cate–Nelson tion fits the data best by providing the highest coefficient of
method (Cate and Nelson 1971) gives the soil critical level, determination (R2). The optimum fertilizer rate is deter-
as illustrated in Fig. 12.7. (I am sure a software program has mined at that point where marginal revenue equals the
been developed to do the same.) marginal cost (i.e. the point where the price of the last yield
This concept of a “critical value” for a given soil and increment equals the cost of the last increment of fertilizer).
farming system has important practical implications for The point can be determined mathematically or graphically
efficient phosphorus use. Maintaining the soil at or close by drawing a price:cost ratio line, expressed in the agro-
to the critical value has important benefits to the farmer (in nomic equivalents in the yield response diagram. The opti-
terms of economic return) and to the environment, in terms mum yield is then determined at the point where a tangent
of reducing the risk of phosphorus transfers to surface of the price:cost ratio line intersects the response curve.
waters (Syers et al. 2008). When more than one element is deficient, regression
The Cate–Nelson method also tells us the points with a high models take this into account, as well as the interactions
probability of a phosphorus response (the lower left quadrant), between these elements. The recommended rates are deter-
and those with a low probabiliy of phosphorus response (in the mined by solving simultaneous equations.
upper right quadrant). It does not tell you how much fertilizer Curvilinear models are also used in a different manner to
to add, but it tells you whether you need it or not. This could be produce one function that takes into account soil test levels as
improved with better estimates of uncertainty. variables, as well as other variables related to soil, climate and
If the soil in question is above the critical level, the management properties, in an attempt to account for the
decision is clearly not to apply that particular nutrient elem- uncontrolled variables. A yield equation relating potato yields
ent to that crop. If the soil is below the critical level, then the in Peru needed 27 variables (Ryan and Perrin 1973). Optimum
decision becomes how much to apply. This requires yield fertilizer levels are calculated on the basis of levels of crop
response field trials. There are two main types of models to prices, fertilizer costs, expected rainfall and soil properties.
express yield response curves: the classic curvilinear model Such complex models are effective when there is adequate
and the simpler linear reponse and plateau model. information about the variables involved and when prices are

https://doi.org/10.1017/9781316809785.014 Published online by Cambridge University Press

 12.6 EARLY TWENTY-FIRST-CENTURY PARADIGM SHIFTS 319

stable. They fail in the tropics where there are insufficient data LRP model Quadratic model
to quantify all variables. The use of complex models in tropical 20
regions is limited to after-the-fact analysis in areas with 18
detailed information; they do not serve successfully as predict- 16
ive tools. In addition, Anderson and Nelson (1975) found that

Potato yields (t/ha)

14
quadratic models are biased when there is a marked response
to the first fertilizer increment, followed by little or no 12

response at higher rates. In such cases, optimum fertilizer 10

recommendations became unrealistically high. 8 Price:cost ratio
6
12.5.6 The Linear Response and Plateau Model 4
A series of studies conducted in England by Boyd (1970) and
2
in the United States by Bartholomew (1972) summarized
many fertilizer response functions over the world and con- 0
0 60 120 180 240 0 60 120 180 240
cluded separately that in most instances fertilizer response 75 160
Nitrogen applied (kg N/ha)
curves can be characterized by a sharp linear increase
followed by a flat horizontal line. Bob Cate and Larry Nelson
Fig. 12.8 Comparison between the LRP model and the
in North Carolina then developed the Linear Response and
quadratic model using the same field data from potatoes in Peru.
Plateau model (LRP), which is also based on Liebig’s law of
The recommended rate is much lower using the first model.
the minimum, and is a logical extension of the Cate–Nelson
Adapted from Waugh et al. (1973)
correlation model (Waugh et al. 1975). Fertilizer response
from a field, or a group of fields, is represented in this model
by two straight lines for each nutrient. The first line repre- (Sanchez et al. 1973). By using the graphic technique, plateau
sents the relatively steep response of an added nutrient until yields and recommended rates were computed. The average
it ceases to be a major limiting factor. This is followed by a nitrogen recommendation in this very high-yielding environ-
line showing a flat plateau, where further additions no ment was 224 kg N/ha according to the quadratic model, and
longer increase yields. The fertilizer response curve so con- 170 kg N/ha with the LRP model. The average gross return for
structed consists of two main points. The “threshold yield” fertilization at the recommended rates was not statistically
is the yield at the zero level of the nutrient in question, but significant between the two models. The net return per
not of all nutrients. The “plateau yield” is the yield at the dollar invested in fertilizer was superior with the LRP model.
point where the nutrient ceases to be a limiting factor; it is The LRP model is much simpler, and focuses on the
not the maximum yield because other factors may still limit range that provides the most bang for the buck, the linear
yields. The fertilizer rate needed to reach the plateau yield is portion at the threshold yield point. The quadratic models,
the recommended rate for the particular nutrient. in their search for the optimum value along the flat part of
The comparison between the two approaches is shown in the response curve, are the least cost-effective. While the
Fig. 12.8 for the same data set. The dotted lines indicate how science is correct in their case (fertilizer response curves are
the fertilizer recommendations are arrived at. The LRP indeed curves), the practicalities are not, because there is
model has only one optimum point, independent of cost wide variability in the actual results under field conditions.
and prices. The curvilinear model shows an optimum point As Bob Cate famously said “If you want to go (from North
based on the particular price:cost ratio at the time the Carolina) to Tokyo you go via the Great Circle route, but if
experiments were conducted. you want to go from this building to the next, you go in a
It is very important to note the wide differences in rec- straight line.” So let’s be as simple and practical as we can.
ommended rates between the two methods. The LRP model It is heartening to know that the LRP model is now
recommended a lower fertilizer rate (75 kg N/ha) to reach a widely used in US agriculture (Beegle 2005).
yield plateau of about 19 t/ha of potatoes. This rate in effect
provides nearly maximum yields, while preserving an effi- 12.6 Early Twenty-First-Century
cient return per unit of fertilizer, because it is still along the
increasing slope. The quadratic model in this case more than Paradigm Shifts
doubles the recommendation (160 kg N/ha) in order to obtain
only an additional 1 t/ha of yield. This is in the relatively flat Building on the advances of the last century, particularly the
part of the curve, where variability is quite high. Small last quarter of it, three new tools have changed our
changes in yields result in large changes in recommended approach to soil fertility. These are near-infrared spectros-
rates. In other comparisons, however, the difference in rec- copy, digital soil mapping and bringing the laboratory to the
ommended rates might not be as large as in this example. field in scientifically rigorous ways. Together they represent
Which approach is right? I tried it myself with a set of a paradigm shift, not of the science itself, but in how soil
field trials on the response to sulfur-coated urea by wetland science is applied. I am happy to note that the contribution
rice in the Coast of Peru, where I used the quadratic method of tropical soil scientists is major in all of the three cases.

https://doi.org/10.1017/9781316809785.014 Published online by Cambridge University Press

 320 SOIL FERTILITY PRINCIPLES

12.6.1 Spectrometry spectrometers that can estimate many soil properties, just
The analysis of wavelengths of light is one of the simplest in by passing light though a soil sample and making the neces-
physics, and not until the start of this century has it been sary correlations (Fig. 12.9). The main properties that can be
applied to soil science at scale. Spectrometry has for decades determined at present are: sand, silt, clay, organic carbon,
been used as a standard in various industries, such as exchangeable bases, cation exchange capacity, phosphorus
pharmaceuticals and paints. The near- and mid-infrared retention and some soil minerals. Where spectrometry does
wavelengths can be correlated with many, but not all soil not (yet) work is available phosphorus, potassium and
properties. In the tropics, the seminal papers by Keith Shep- sulfur, very critical soil fertility parameters. Spectrometry
herd and Markus Walsh (Shepherd and Walsh 2002, 2007) is also helpful in determining organic input quality and
showed the value of then near-infrared rudimentary carbon mineralization rate (Shepherd et al. 2005).

Fig. 12.9 Keith Shepherd (top left) using a portable spectrometer in the field in Kenya, and a technician using it in his Nairobi laboratory
(bottom left). The resulting soil spectral signature (top right), and the close correlation between the predicted value by spectroscopy and
the “actual” value of SOC from a wet chemistry laboratory (bottom right).

https://doi.org/10.1017/9781316809785.014 Published online by Cambridge University Press

 12.6 EARLY TWENTY-FIRST-CENTURY PARADIGM SHIFTS 321

12.6.2 Digital Soil Mapping organic carbon, bulk density, pH, effective cation exchange
Traditional soil maps are composed of polygons (mapping capacity, electrical conductivity and bulk density – bulk
units) and their legends, showing mostly soil classification density is needed to express carbon and nutrients on a mass
terms that are of very limited information value to anyone basis (t/ha) for biogeochemical modeling (Sanchez et al.
other than a pedologist. Such maps are static and tell you 2009). These are the first two actions in the process for
nothing about soil properties. Polygon maps assume that developing the digital soil map, as shown in Fig. 12.10. Such
soils are uniform within the mapping unit, which is seldom properties are mapped on a pixel basis with a resolution as
the case. Interpretations made from them in terms of soil small as 30  30 m, which is great for smallholder farms.
properties and agronomy are of course no better than The spatially inferred soil properties are then used to
the map. predict more difficult-to-measure soil functions (action 3,
To address some of these limitations, a digital soil- Fig. 12.10), such as available soil water storage, carbon dens-
mapping working group of the International Union of Soil ity and phosphorus fixation, using pedotransfer functions.
Sciences (IUSS) started work in 2006, and has since produced Rather than calibrating only individual soil properties to
valuable information that can be found in books arising from infrared spectra, multivariate associations are used to iden-
their conferences (Lagacherie et al. 2007, Hartemink et al. tify soil functions. The overall uncertainty of the prediction
2008, Boettinger et al. 2010). In 2008, a group of soil scientists, is assessed by combining the uncertainties of the input data
agronomists and ecologists decided to produce a digital map with the uncertainties of the spatial inference and those of
of soil properties and soil functions of the world (Sanchez the soil functions, using global sensitivity analysis.
et al. 2009, Fisher 2012), and the African portion of that is in The key is to connect all these soil functions with fertil-
progress. izer response or other georeferenced field data from
Soil properties are estimated quantitatively with a statis- well-conducted agronomic experiments (legacy data) as well
tical inference system using spatial (kriging, co-kriging) or as with digital maps of population density, roads, distance to
non-spatial (regressions, neural networks) techniques, or markets, internet access and other socioeconomic digital
both, and expressed as their probability of occurrence with maps (action 4, Fig. 12.10). Analyses of decisions need to be
uncertainty estimates. They are derived using quantitative made, connecting the maps with actual on-the-ground
relationships between the punctual soil measurements and experience. Then the final two actions (management recom-
the spatially continuous soil covariates. This results in maps mendations and reaching the end user) can be achieved.
of soil properties such as the ones selected by the Global Soil Web-based digital soil maps can be made interactive. This
Map Consortium as their minimum data set, i.e. clay, is very much work in progress.

1. Data input 2. Soil properties

Legacy data
Clay
3. Soil functions
Spatial OC
Interfence Pedotransfer AJ toxicity
Covariates models pH functions Fertilizer
response
BD Water
storage

Soil map Erosion

Carbon
management

5. Management
recommendations
aly ion

4. Management
sis
an cis

inputs
De

Fertilizer
Social application 6. End users
covariates Erosion
Policy-makers
control

Crop Extension
Decision selection Training Scientists
analysis Drip Dissemination
Agro-business
Legacy data irrigation Web access
Farmers
Climate
mitigation

Fig. 12.10 Conceptual flow of the soil digital map process. Chart based on text by Sanchez et al. (2009)

https://doi.org/10.1017/9781316809785.014 Published online by Cambridge University Press

 322 SOIL FERTILITY PRINCIPLES

12.6.3 Bringing the Laboratory to the Field as the beer and swimming-pool testing industries. The soil
Although soil nutrient replenishment in sub-Saharan Africa fertility parameters analyzed include soil pH, electrical con-
is widely recognized as a critical biophysical entry point to ductivity (indicative of general fertility levels as well as
an agricultural transformation, the actual application of soil salinity issues), biologically active organic carbon, and 0.01
science is still minimal, and frustratingly so, in large part M calcium chloride (CaCl2)-extractable nitrate-N, sulfate-S,
because of the delay in obtaining the necessary information phosphate-P, and potassium using specific sensors. Add-
from soils sampled in farmers’ fields and sent to a wet- itionally, the kit includes tools to measure soil physical
chemistry, conventional laboratory. The data often take a properties such as surface sealing strength, plow pans and
year to come back (if at all) and the paper-based process is volumetric soil water content. Soil texture is estimated by
prone to errors of recording and transcribing, as well as feel. SoilDoc essentially measures what a conventional wet-
delays in transmission, interpretation and visualization. chemistry laboratory does at a similar level of accuracy
Worse still, the data are usually hard to interpret, largely (Fig. 12.12) plus biologically active SOC and several physical
irrelevant and often of questionable quality. The poor infra- properties (Weil 2011, Sanchez et al. 2013). Like digital soil
structure of Africa and the budget limitations of national mapping, this too is very much work in progress.
institutions are the main reasons used to rationalize this At the time of this writing, we need to use the field-based
useless exercise. This is not a problem in the more advanced wet chemistry of SoilDoc and similar systems in conjunction
tropical countries, but it is also a problem in many areas with near-infrared spectrometers, because spectrometry
outside Africa as well. cannot determine available soil test levels since they are
The way to get around this is to get the data in situ and artifacts of the extractants used. Wavelength data cannot
send it electronically to the central laboratory where senior detect something that does not exist in the soil. To repeat,
people or algorithms make the recommendations and send both approaches are works in progress and their combined
it back electronically to the extension agent or farmer. The use could result in positive synergies.
problem is that, up until recently, field soil test kits have The combination of these approaches, as well as the
proven to be highly inaccurate and not sufficiently precise rapidly developing soil scanning by satellites, shows tremen-
(Vanlauwe 2008). Weil (2011) of the University of Maryland dous potential for faster and more effective ways of evaluat-
and Columbia University has been developing a wet- ing soil fertility.
chemistry “SoilDoc,” which has been proven to work well
and is highly correlated with central wet-chemistry labora-
tories (Fig. 12.11). It enables extension workers to make 12.7 Summary and Conclusions
on-the-spot diagnoses of soil constraints, allowing targeted
recommendations to advise farmers in real time (Sanchez
et al. 2013). • Soil fertility is the capacity of soils to supply essential
This tool uses state-of-the-art battery-powered miniatur- elements to plants. Currently we recognize 17 elements
ized instruments, originally created for other purposes such as essential plant nutrients. Essential means that without
them plants, microbes or animals would not be able to
complete their life cycles.
• There are three main principles of soil fertility: the law of
the minimum, synchrony and nutrient cycling.
• The law of the minimum states that plant growth will be
limited by the essential element that it is most
deficient in.
• The ideal is to synchronize nutrient additions with the
plant’s nutrient requirements as it grows. This is the
“synchrony principle,” which represents the agronomist’s
dream, and is seldom, if ever, realized.
• Nutrient cycling is the third fundamental principle of soil
fertility. Cycles of soil nutrient elements are different,
varying in speed, depending on the types of bonds the
elements have with carbon.
• Organic nitrogen forms covalent bonds with organic
carbon, while organic phosphorus and sulfur form ester
bonds. Most other nutrient elements are either bonded
ionically or are in loose associations with soil organic
Fig. 12.11 SoilDoc, a “lab-in-the-box” field test kit. Ray Weil
carbon (SOC).
is entering the data in a smart phone that transmits it to the
• The covalent bonding of nitrogen with carbon results in a
cloud or to a central location. Arusha, Tanzania,
series of compounds that are chemically recalcitrant or
September 2012.
physically protected in the slow and passive SOC pools.

https://doi.org/10.1017/9781316809785.014 Published online by Cambridge University Press

 12.7 SUMMARY AND CONCLUSIONS 323

7.00
y = 0.901x + 0.627
R2 = 0.942

6.00
rH by field kit

5.00

4.00
4.00 5.00 6.00 7.00
pH in laboratory

80
75 0.7
70
65 kit-N = 0.81* lab-N + 2 0.6
Nitrate-N by kit method, using CaCl2

60 R2= 0.96*
55
0.5
50
SoilDoc P, mg/kg

45
40 0.4
35
30 0.3
25
20
0.2
15
10
0.1
5
0
-5 0.0
-5 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 0 50 100 150 200
Nitrate-N by Cd reduction, KCl extraction method, Mehlich 3 P, mg/kg
in laboratory

Fig. 12.12 Correlations for pH (top), nitrate-N (middle), and available phosphorus (bottom), determinations by the SoilDoc kit and a
wet-chemistry laboratory. Reproduced with permission from Ray Weil, University of Maryland

• Phosphorus and sulfur compounds have a high negative • The demand side of soil fertility is best expressed as the
charge which may prevent them from being entrapped in nutrient uptake at harvest. Different crops exhibit vary-
the slow and passive SOC pools. Organic phosphorus and ing yield levels and nutrient uptake.
sulfur mineralization proceeds rapidly once the extracel- • The high-input agroecosystems generally add similar
lular enzymes break the ester bonds. nutrient inputs to the soils as the natural systems.
• This makes nitrogen mineralization slower and more For example, residues returned to the soil by an annual
costly in terms of energy (fuelled by soil carbon) than maize–maize–soybean rotation in the Peruvian Amazon
the mineralization of phosphorus and sulfur. In terms of provided more nitrogen, phosphorus and potassium and
stoichiometry, plants have C:N ratios of 20 and above, similar levels of calcium and magnesium than litter fall
while soil bacteria have low C:N ratios, about 6, which from a tropical rainforest in similarly acid soils.
means that they require more nitrogen relative to the • As expected, this did not happen in low-input systems, but
energy available, and in fact they are usually nitrogen- when aluminum-tolerant cultivars were used in acid soils,
starved. some systems came surprisingly close to the natural ones.
• It is a priority for soil scientists to accelerate and tighten The trick is to ensure that crop residues are returned to
nutrient cycles in agricultural systems. the soil.

https://doi.org/10.1017/9781316809785.014 Published online by Cambridge University Press

 324 SOIL FERTILITY PRINCIPLES

• Taking a representative soil sample is both the first step spectrometry does not (yet) work is in estimating available
and the largest source of error in soil fertility evaluation phosphorus, potassium, sulfur or nitrogen.
because it normally represents one-billionth (10–9) of the • Digital soil mapping is superior to the traditional poly-
total soil volume for which an analysis is made. gon maps. Digital soil maps, with a resolution of 100-m
• Spatial variability in soil properties is always an issue pixels, can describe soil properties obtained by spectrom-
because soil is far from a homogeneous system and bears etry or wet chemistry. The spatially inferred soil proper-
the footprints of generations of farmers who add to it ties are then used to predict more difficult-to-measure
temporal variability. soil functions, such as available soil water storage,
• Fields close to the household (infields) are usually more carbon density and phosphorus fixation, using pedo-
fertile than those farther away (outfields) because small- transfer functions. Web-based digital soil maps can be
holder households dump a lot of garbage and kitchens made interactive.
ashes, and their domestic animals often urinate and defe- • A new generation of field test kits, such as “SoilDoc,”
cate in these nearby fields. can measure many soil properties using miniaturized
• The most widely used way to assess the status of a nutri- wet-chemistry sensors at a similar level of accuracy
ent in a soil is by an extraction in a wet-chemistry labora- as large laboratories, eliminating the need to send
tory. In modern laboratories, a single technician can run soil samples to the laboratory, a major obstacle in Africa
100 soil samples a day, measuring 10 determinations per and other less-developed countries in the tropics.
sample. The bulk of the work is done with two extrac- The SoilDoc kit is connected electronically to a
tants: the dilute double acid (Mehlich 1) or the modified central laboratory, sending data in a web-based form,
Olsen extractant for determining available phosphorus, and the recommendations, via algorithms or expert
potassium, calcium, magnesium, sodium, iron, manga- opinion, are returned to the extension agents in
nese, zinc and copper; and the 1 M KCl extraction for real time.
aluminum.
• Soil test values become useful only when they are correl-
ated with crop yield responses. Correlations can be done
in many ways, using fairly complicated formulas. In my
References
view, the simplest and most practical one is the Cate–
Nelson method, which plots relative yields versus the soil Anderson RL and LA Nelson. 1975. A family of models intersecting
test levels as a scatter diagram. This method also deter- straight lines and concomitant experimental design useful in evalu-
ating response to fertilizer nutrients. Biometrics 31: 303–318.
mines which soils have a high probability of a nutrient
Barker AV. 2010. Science and Technology of Organic Farming. CRC Press, Boca
response as opposed to those that do not.
Raton, FL.
• There are two main methods to determine the recom- Bartholomew WV. 1972. Soil Nitrogen: Supply Processes and Crop Require-
mended fertilizer rates based on yield response field trials. ments. International Soil Fertility Evaluation and Improvement Program
The first is the traditional curvilinear method, where a Technical Bulletin 6. North Carolina State University, Raleigh, NC.
quadratic or other function is fit and the point where Beegle D. 2005. Assessing soil phosphorus for crop production by soil
marginal revenue equals marginal cost represents the testing. Phosphorus: Agriculture and the Environment, JT Sims and AN
recommended rate. The other is the Linear Response and Sharpley (eds.). American Society of Agronomy, Madison, WI,
Plateau (LRP) model where yields are represented by two pp. 123–143.
straight lines for each nutrient in the model. The first line Boettinger JL, DW Howell, AC Moore, AE Hartemink and S Kienast-Brown
(eds.). 2010. Digital Soil Mapping: Bridging Research, Environmental
represents the relatively steep response of an added nutri-
Application, and Operation. Springer-Verlag, Dordrecht.
ent until it ceases to be a major limiting factor. This is
Bouldin DR, T Greweling and J Lui. 1971. The Effect of Drying Soils From the
followed by a line showing a flat plateau, where further Humid Tropics on Selected Soil Properties. Cornell University Agronomy
additions no longer increase yields. Department, Ithaca, NY.
• The LRP model recommends a lower fertilizer rate, pro- Boyd DA. 1970. Some recent ideas on fertilizer response curves. Proceed-
viding nearly maximum yields, while preserving an effi- ings of the 9th Congress of the International Potash Institute, Antibes,
cient return per unit of fertilizer, because it is still along 1970. International Potash Institute, Berne, pp. 461–473.
the increasing slope. The LRP model is much simpler, and Brady NC and RR Weil. 2008. The Nature and Properties of Soils, 14th edition.
focuses on the range that provides the most bang for the Pearson Prentice Hall, Upper Saddle River, NJ.
buck. It saves on fertilizer at little yield cost and thus Cate RB Jr. and LA Nelson. 1971. A simple statistical procedure for parti-
tioning soil test correlation data into two classes. Soil Science Society
reduces the emission of nitrogen gases and leaching
of America Proceedings 35: 658–659.
losses, providing a superior net return per dollar invested
CEPLAC. 1989. Informe de Pesquisas de 1983. Commissão Executiva do
in fertilizer. Plano da Lavoura Cacaueira, Ilheus.
• The application of near- and mid-infrared spectrometry CIAT. 1985. Tropical Pastures Program. Annual Report for 1984. CIAT, Cali.
can provide accurate estimates of sand, silt, clay, organic DeDatta SK. 1981. Principles and Practices of Rice Production. Wiley, New York,
carbon, exchangeable bases, cation exchange capacity, NY.
phosphorus sorption and some soil minerals. Where Fisher M. 2012. Creating a global digital soil map. CSA News: 57: 4–9.

https://doi.org/10.1017/9781316809785.014 Published online by Cambridge University Press

 12.7 SUMMARY AND CONCLUSIONS 325

Goedert WJ (ed.). 1986. Solos dos Cerrados: Tecnologías e Estrategias de Man- Shepherd KD and MG Walsh. 2002. Development of reflectance spectral
ejo. Nobel, São Paulo. libraries for characterization of soil properties. Soil Science Society of
Hartemink AE, A McBratney and ML Mendonça-Santos (eds.). 2008. Digital America Journal 66: 988–998.
Soil Mapping with Limited Data. Springer-Verlag, Dordrecht. 2007. Infrared spectroscopy: Enabling an evidence-based diagnostic
Henderson JB and EJ Kamprath. 1970. Nutrient and Dry Matter Accumulation surveillance approach to agricultural and environmental manage-
by Soybeans. Technical Bulletin 197. North Carolina Agricultural ment in developing countries. Journal of Near Infrared Spectroscopy 15:
Experiment Station, Raleigh, NC. 1–19.
IRRI. 2011. Crop Manager, Rice Knowledge Bank. International Rice Research Shepherd KD, B Vanlauwe, CN Gachengo and CA Palm. 2005. Decom-
Institute, Los Baños. position and mineralization of organic residues predicted using
ISFEIP. 1967–1974. Annual Reports of the International Fertilizer Evaluation and near infrared spectroscopy. Plant and Soil 277: 315–333.
Improvement Program. North Carolina State University, Raleigh, NC. Swift MJ. 1985. Tropical Soil Biology and Fertility (TSBF): Planning for Research.
Lagacherie P, AB McBratney and M Voltz (eds.). 2007. Digital Soil Mapping: Biology International. Special Issue 9. IUBS, Paris.
An Introductory Perspective. Elsevier, Amsterdam. Syers JK, AE Johnston and D Curtin. 2008. Efficiency of Soil and Fertilizer
Muñiz O. 2008. Los Microelementos en la Agricultura. Instituto de Suelos, Phosphorus Use. Reconciling Changing Concepts of Soil Phosphorus
Ministerio de la Agricultura, La Habana. Behaviour and Agronomic Information. Fertilizer and Plant Nutrition Bul-
Perur NC, CK Subramanian, G Muhr and HE Ray. 1974. Soil Fertility Evalu- letin 18, Food and Agriculture Organization of the United Nations,
ation to Serve Indian Farmers. Mysore Department of Agriculture, Rome.
Bangalore. Szott TT. 1987. Improving the Productivity of Shifting Cultivation in the Amazon
Ryan JC and RK Perrin. 1973. The Estimation and Use of a Generalized Response Basin of Peru through the use of Leguminous Vegetation. PhD Thesis,
Function for Potatoes in the Sierra of Peru. Technical Bulletin 214. North North Carolina State University, Raleigh, NC.
Carolina State Agricultural Experiment Station, Raleigh, NC. TropSoils. 1987. Technical Report for 1985–1986, N. Caudle and CB McCants
Sanchez PA and JR Benites. 1987. Low input cropping for acid soils of the (eds.). North Carolina State University, Raleigh, NC.
humid tropics. Science 238: 1521–1527. Vanlauwe B. 2008. Report on the Evaluation of Commercially Available Test-Kits
Sanchez PA, A Gavidia, GE Ramirez, R Vergara and F Minguillo. 1973. for Soil Fertility Evaluation. Tropical Soil Biology and Fertility Pro-
Performance of sulfur-coated urea under intermittent flooded rice gram, Nairobi.
culture in Peru. Soil Science Society of America Proceedings 37: 789–792. Vitousek PM and RL Sanford. 1986. Nutrient cycling in moist
Sanchez PA, JH Villachica and DE Bandy. 1983. Soil fertility dynamics tropical forests. Annual Review of Ecology and Systematics 17: 137–167.
after clearing a tropical rainforest in Peru. Soil Science Society of Vitousek PM, S Hättenschwiller, L Olander and S Allison. 2002. Nitrogen
America Journal 47: 1171–1178. and nature. Ambio 31: 97–101.
Sanchez PA, CA Palm, LT Szott, E Cuevas and R Lal. 1989. Organic input von Liebig J. 1840. Chemistry in its Application to Agriculture and Physiology.
management in tropical agroecosystems. Dynamics of Soil Organic Taylor & Walton, London.
Matter in Tropical Ecosystems, DC Coleman, JM Oades and G Uehara Waugh DL, RB Cate Jr. and LA Nelson. 1973. Discontinuous Models for Rapid
(eds.). University of Hawaii, Honolulu, HI, pp. 125–152. Correlation, Interpretation and Utilization of Soil Analysis and Fertilizer
Sanchez PA, S Ahamed, F Carré, AE Hartemink, J Hempel, J Huising, P Response Data. International Soil Fertility Evaluation and Improvement
Lagacherie, AB McBratney, NJ McKenzie, Ml Mendonça-Santos, Program Technical Bulletin 7. North Carolina State University,
B Minasny, L Montanarella, P Okoth, CA Palm, JD Sachs, KD Shep- Raleigh, NC.
herd, TG Vågen, B Vanlauwe, MG Walsh, LA Winowiecki and GL Waugh DL, RB Cate Jr., LA Nelson and A Manzano. 1975. New concepts in
Zhang. 2009. Digital soil map of the world. Science 325: 680–681. biological and economical interpretation and fertilizer response.
Sanchez PA, RR Weil and A Rose. 2013. SoilDoc: An innovative soil test kit Soil Management in Tropical America, E Bornemisza and A Alvarado
that provides real-time fertilizer recommendations. IFA Fertilizers (eds.). North Carolina State University, Raleigh, NC, pp. 484–501.
and Agriculture. February issue: 6–7. Weil RR. 2011. Using simple quantitative and qualitative field tests to
Sarmiento G. 1984. The Ecology of Neotropical Species. Harvard University diagnose soil limitations. Abstracts, ASA, CSSA and SSSA Annual
Press, Cambridge, MA. Meetings, October 16–19, San Antonio, TX.

https://doi.org/10.1017/9781316809785.014 Published online by Cambridge University Press