Tải bản đầy đủ - 0trang
III. Higher Production from Existing Arable Soils
POTENTIALLY ARABLE SOlLS O F THE WORLD
families, but returns in the market place may be so low that they have
little incentive to use added inputs for production beyond family needs.
As pointed out in the previous section, several practices must be
combined, in proper relationship to the local kind of soil and to one
another, for rewarding harvests in relation to inputs of labor, skill, and
materials. Because of the great contrasts among the many thousands of
kinds of soil and their associated crops, the “packages” of practices vary
widely from place to place, even between one kind of soil and another
within a small holding. Especially in tropical areas, more local research
is needed to learn how best to combine machines, chemicals, and other
products of the industrial sector of agriculture to have better farming
systems for the many kinds of tropical soils.
Soil surveys are essential for the full use of the results of research and
of experience related to soil management. For orderly use in technical
assistance, these results need to be reported and summarized by specific
kinds of soil. No one has yet found a reasonable alternative for assernbling knowledge about soils for application to the specific land tracts where
1 . Operational Planning
For operational planning and technical assistance the maps in such
soil surveys must be in enough detail to show clearly the kinds of soil
in each important field and holding.
The degree of detail depends on the relative intensity of potential use,
the relative value per hectare of labor and other inputs, the complexity of
the pattern of unlike soils in the landscape, and their degrees of contrast.
Then too, where expensive structural works, such as those for water
control - runoff control, drainage, irrigation, or land shaping-are in
prospect, detailed soil mapping is required for satisfactory results without
waste. Soil surveys are essential for carrying out programs to consolidate
fragmented holdings: programs that are essential to effective water control and field design in millions of farm villages. Thus the scale for adequate soil mapping ranges from about 1 :50,000in areas with simple soil
patterns or only extensive uses to about 1 :8000 for highly intensive farming. (At and around the site of expensive structures, along planned highways, and for housing clusters, field mapping may need to be done at a
large scale, say 1:1000,where the pattern of soils is very complex.)
The soil map must also have plotted on it accurately the roads, streams,
houses, and other features that a user can see and that help him to locate
CHARLES E. KELLOGG A N D ARNOLD C. ORVEDAL
himself precisely on the map. These can be plotted during the soil survey
or taken partly from adjusted aerial photographs or from detailed planimetric surveys.
Usually soil maps are published at a somewhat smaller scale than the
field sheets. For example, field sheets at a scale of about 1:15,000 can
be published at 1:20,000 or 1:24,000 if the symbols and mapping are
properly planned in advance.
Any soil map to be used in operational planning or technical assistance
must be accompanied by a text that describes the soils in terms of the lay
reader and in the technical language of scientists and engineers. If new
practices are developed after the soil survey is published, the scientists
and engineers need the technical descriptions to provide new or revised
Besides the descriptions, the principal alternative uses, management
systems, and expected yields of adapted crops must be given in specific
terms for each kind of soil. If relevant in the area, similar interpretive
ratings are needed for native and sown grasses, trees, and any limitations
of the soils for suggesting houses and other structures.
2 . General Planning
For general planning of the approximate location of projected broad
changes in land use, highways, and other community facilities, a smallscale soil map, generalized from the detailed one or made by reconnaissance, is highly useful. It too must include roads, villages, and other
important features to help the reader locate himself. The map units on
this kind of map are associations of kinds of soil that naturally occur together in the landscape, whether they are similar or highly contrasting.
Each soil association is named from the two or three principal kinds of
soil that occur together. Its description includes the proportion of each
kind of soil that is an important part of the mapping unit. Depending on
the local area, descriptions and interpretations of potential uses can be
given specifically for each kind of soil, but only in broad terms for each
OF SCIENCE A N D TECHNOLOGY
Compared to temperate regions, less research for soil management systems has been done in the Tropics and less experience with modern
systems is available. Yet statements that little is known about tropical
soils are quite untrue. A great deal is known and published. But the number of people well informed about this knowledge is small in relation to the
number of cultivators in both the less developed countries and the ad-
POTENTIALLY ARABLE SOILS OF T H E WORLD
vanced countries. Then too, much useful material is available in many
scientific institutes that has not been summarized by kinds of soil and
published. Considerable research has been done on several kinds of soil
in the Tropics, as in northern Queensland, Hawaii, Puerto Rico, and
especially in the Congo between 1935 and 1958. During the past few
years much important research has been done in other centers in both
East and West Africa, in south Asia, and in parts of South and Central
America, including some that is not yet published. Each year more soil
surveys are being made in tropical areas. Detailed soil maps are available
at a good many locations and small-scale soil maps are available for all
the Tropics and subtropics at scales ranging from about 1:250,000 to
1 : 1,000,000. Unfortunately not many of these vitally needed soil maps
have been published.
Several schematic soil maps at scales of 1:5,000,000 are either available or becoming available, but these soil maps have far too few base
data - villages, cities, roads, streams, and other features. These features
are needed to enable the user to locate himself well enough to note the
kinds of soil on the tracts of land in which he is interested. They are also
needed to enable him to read from the map the kinds of soil on tracts of
land supporting experimental stations or successful farm communities
in other parts of the world. It is from these places that are on similar
kinds of soil that he can obtain valuable data for planning. Even at a scale
of 1:I ,000,000, the smallest practicable individual mapping unit is about
4000 hectares. At a scale of 1:50,000 one can show areas of about 10
hectares. Obviously, this is about the smallest practicable scale of soil
maps for operational planning. Where the individual areas of soil are small
and the kinds of soil are highly contrasting, the scale should be larger for
operational planning of enterprises requiring moderate to large inputs of
labor and capital.
Much more research and study has been done with the reasonably wellwatered tropical soils than with the dry ones, with those of moderate to
high fertility than with those of low fertility, and with those near existing
transport than with those far from existing roads or navigable streams.
Yet much useful material is becoming available within the Tropics and
subtropics for wide use to transfer experience and knowledge between
countries and continents. “The World Food Problem” (President’s
Science Advisory Committee, 1967) emphasizes the urgent need for their
synthesis and publication. For operational planning where the individual
areas are small and the soils are highly contrasting, the scale should be
larger for operational planning of enterprises requiring considerable inputs of labor and capital.
CHARLES E. KELLOGG A N D ARNOLD C. ORVEDAL
The results of research and experience gained on tropical soils are far
more useful for estimating the potentials and best farming systems on
similar tropical soils than results brought from temperate regions.
Despite the urgent need, improved technology is not easily transferred from temperate regions to tropical ones. But the principles developed from scient$c research in the sciences basic to soil use and the
scholarly methods for both basic and adaptive research can be
In technical assistance, either to improve the use of existing arable
soils or to suggest systems for new arable soils, one should take orderly
account of the available experience in the world on comparable soils and
of the skills of the cultivators or of the proposed settlers.
The basic task in technical assistance is for the scientist or a small
group of scientists to connect the basic principles of the sciences relevant
to farming and the other sectors of agriculture to the specific natural,
social, economic, and political environment where the work is to be done
-in other words to invent the most efficient technologies and systems of
soil use for the local kinds of soil that can be carried out by the local
people. Obviously, success requires excellent scholars with deep knowledge of the basic principles and with the skills to work in contrasting
social and natural environments and to communicate with the people.
When we read of “practice” application or that “cultivators are slow to
accept a new practice,” we can be fairly certain that the advisor is too
narrow in his viewpoint. The cultivator is always dealing with a combination of practices.
Here and there good starts have been made and others could be,
especially with local adaptive research for guidance. Yet more basic
research needs to be carried out in the newly developing countries to
reach the best results. New principles from that research are bound to
lead to more adaptive research for the most practicable local systems.
This need is great not only in farming, but also in the service and industrial
sectors of agriculture.
1. Adaptive Research
Even with the best summaries of what has been learned from research
and experience elsewhere, adaptive research in local areas is usually
essential for modern systems.
First of all, one can learn much of importance by studying existing
practices, even primitive ones that cultivators have developed over the
years and passed on to their children. Mere description, however, is
nearly useless unless the soils are named in a standard system or are
POTENTIALLY ARABLE SOILS OF THE WORLD
described well enough so they can be named later. Otherwise there is no
way to apply the results elsewhere.
Second, one can lay out simple field trials as small rectangular plots on
cultivators’ fields located on the different kinds of soil, which are named
or described. These trials can have different combinations of fertilizers,
water control practices, new varieties, and so on. Of course, knowledge
gained from elsewhere on how similar soils respond under good management form the basis for planning such trials. In areas with generally low
yields, precise yield measurement is not necessary. A plot can be treated,
for example, with a moderate application of nitrogen alone, and other
plots with nitrogen in various combinations with phosphorus, potassium,
magnesium, calcium, or other plant nutrients expected to be in short
supply from the appearance of the plants. The yields and condition of the
crops should be recorded during the growing season.
In some areas the appearance of the plants suggests deficiencies of one
or more of the secondary nutrients, such as boron, iron, zinc, or manganese. If such plant symptoms are evident early in the trial, small strips
across the plots can be sprayed with a dilute solution of the individual
trace nutrients to learn whether or not plants respond to them.
Some soil scientists tend to make this kind of adaptive research much
more sophisticated than is necessary. Eventually it is desirable to have
long-term experiments accompanied by soil tests for available nutrients
as well as many detailed soil descriptions, but from many of these test
plots a good deal can be learned that is highly useful in making a start
A farmer in an advanced country, if convinced of a new practice in his
system, will normally adopt it if he expects a 10 or 15 percent increase in
yield. We must recall that this 10 or 15 percent increase is from a high
base yield. The yields of many cultivators are already so low that a 10
or 15 percent increase could hardly be seen, kt alone be enough to compensate him for probable inputs. In many areas of the world, unless the
yield can be at least doubled, a cultivator will not be persuaded to adopt
another combination of practices. Commonly the opportunities are far
greater than for mere doubling of the yield. On some old tropical soils, for
example, a crop like grain sorghum will little more than germinate in the
absence of phosphatic fertilizer (Fig. 3). Yet with it the crop responds to
nitrogen and other inputs to give yield increases of severalfold.
2. Continued Research
Continued research and examination of agricultural systems by
scientists are essential, not only for further progress but also to maintain
CHARLES E. KELLOGG A N D ARNOLD C . ORVEDAL
the progress that has been gained. This important point is often overlooked. Let us say that a new package of soil management practices,
including new varieties, fertilizers, and water control systems, is estab-
FIG.3. An experiment showing the enormous effect of phosphatic fertilizer on a tropical
soil with a sandy surface and clayey B horizon at the Nungua Station near Accra in Ghana.
With nitrogen but no phosphatic fertilizer, the crop in the foreground is worthless. With
fairly high rates of both, good yields of grain sorghum are seen in the background. (The soil
is probably a sandy Paleustult.)
lished and proves to be highly successful. Unless competent agricultural
scientists are available, some new disease or insect could wipe out the
crop. Modern plant breeding, for example, produces plants of high yield
having rather narrow genetic bases. Where cultivators in a broad area
have grown the same new variety and a new disease or insect went unnoted for a short time, the result has been catastrophic. This hazard is
great with many newly introduced varieties. Similarly unnoticed changes
in the soil as a result of a new management system can lead to serious
difficulties unless they are noted promptly and the necessary corrective
measures are taken. In other words, the more efficient and productive the
soil management system, the more important it is that scientists be
available to detect a serious problem before it becomes catastrophic.
Most experienced scientists from the advanced countries are familiar
POTENTIALLY ARABLE SOILS OF THE WORLD
with the kinds of dramatic failures that can happen after the adoption of a
new combination of high-yielding varieties and other practices - failures
resulting from a new disease or insect, waterlogging and salinity, new
gullies from poorly located terrace outlets, and so on. Most are familiar
with weed problems, but some may not visualize the sudden hazard that
weeds can bring about in the absence of modern chemical controls. Under
the old practice of shifting cultivation in tropical areas, the rejuvenating
forest fallow after 2 to 5 years of cropping interrupted the cycles of
insects, diseases, and weeds. Many of the worst weeds are scarce plants
within the native vegetation. Yet the seeds that fall on bare soil grow
promptly. Thus, if open-field cultivation is substituted for shifting cultivation with bush fallow or mixed cultures, weeds must be watched for and
avoided. Weeds can reduce the yields of rice, maize, and many other
crops substantially unless practical control measures are applied promptly.
C . INCENTIVES
Most cultivators in the underdeveloped countries are inclined to be
cautious, even conservative, to new ways of soil management. The
reasons are not hard to find. Few of them have savings or even any easy
way to save. Should the new combination of practices that might be
suggested to them fail, they would suffer greatly. Thus they prefer the
trusty old familiar system.
Actually, most cultivators work hard. Some travelers fail to understand
why they see so many men resting and think them lazy. People on low
diets rest when additional work adds nothing significant to their incomes.
Schultz (1964) explained that, considering their skill, resources, and
markets, most peasant farmers work as near to the margin of economic
return for added inputs as the farmers in advanced countries. Yet Penny’s
studies in Indonesia suggest that some cultivators at a subsistence level
will not change when they have opportunities. He gives several specific
examples (Penny, 1966).
Nevertheless, Gunnar Myrdal (1 968) points out the great importance
of improving the use of labor in the crowded less developed countries and
notes that improved technology, rather than decreasing labor, actually
increases it. Although mechanization of field operations may reduce
some labor requirements in fields, the use of fertilizer, plant protection,
improved storage and processing, adequate market roads, and the like,
increase the opportunities for labor.
In many newly developing countries where additional farm products
are urgently needed for food and for economic growth, the marketing
system is not conducive to increased production. Commonly, losses are
CHARLES E. KELLOGG A N D ARNOLD C. ORVEDAL
so great between the rural village and the primary market that village
prices are low. Few local markets near the cultivators grade their products
according to quality. Special taxes may even further reduce the price.
Many cultivators lack opportunities for education and for learning new
skills. Many lack access to credit at reasonable rates and to such inputs
as improved seeds, fertilizers, and tools at prices they can afford. In
remote areas, until many cultivators adopt improved combinations of
practices, reliable vendors will not have established sales outlets for new
Yet local demonstrations near the homes of cultivators under conditions similar to theirs are convincing to many of them. The advantage
in yield must be clear. Even then some cultivators hesitate to invest. The
risk may seem too great if substantial purchases are required. Much of
this risk can and has been overcome by a qualified government service
that guarantees them at least their traditional yields.
Then too, some advisors fail to inform cultivators of the great advantage
of combinations of practices to have good harvests from their soils. Although occasionally one practice alone may give a large return, advisors
must be cautioned to think of several at the same time, such as fertilizers,
improved seeds, water control, and pest control.
All these difficulties that the cultivator sees must be fairly well overcome before he has the essential incentive to go ahead.
Obviously we are speaking about the bare minima of the several
essential practices for a start, not the ideal. Actually nearly all countries
have the soil and the labor for ample food supplies. But these will not be
forthcoming until cultivators have genuine incentives.
Some system for water control is important on nearly all soils. The
system is so much influenced by field patterns that special emphasis must
be given it in early steps toward the improvement of farming.
For the best results the roots of plants should not lack adequate
supplies of either moisture or oxygen during the growing season. The requirements of both vary among crops. The ideal arable soil is porous
enough to allow the water added by rain or by irrigation to enter the rooting zone. The body of the soil holds enough water to support plants between wettings and allows any excess to pass beneath the rooting zone. If
too much water accumulates in the soil, the roots lack oxygen and crops
do poorly. The excess can be taken away through a drainage system,
either in open ditches or through tile.
POTENTIALLY ARABLE SOILS OF THE WORLD
I . Runof Control
By far the greater part of the crops grown in the world are watered
directly from the rain that falls. The productivity of soils that have a
variable supply of rainfall and are reasonably permeable can be greatly
improved by one of several kinds of terraces. These earth ridges are laid
out in such a way that water moves only slowly down the slope and thus
has more time to penetrate the soil. Enormous areas of arable soil in the
world could benefits from this practice i f i t is well done. If poorly done,
with terraces on old irregular field lines, waterlogging or erosion, or both,
are likely to result in waste.
Many potentially excellent soils in south Asia, for example, fall into
this category. They are moderately permeable, their slopes are long and
gentle, and the rains come unevenly in the monsoon climate. By making
terraces at slight angles to the contours along with protected waterways
for the excess water of heavy rains that cannot soak into the soil, the total
amount of moisture stored for plant roots is greatly increased. Not only
do the roots have more water in the rainy season but also enough water is
commonly stored in the soil to support an additional crop during the drier
season. Then too, wells near the lower parts of slopes have additional
water for irrigating fruit and vegetable gardens.
On many soils this practice for runoff control alone has more than
doubled production. Then, as the water system is supplemented by
fertilizers, improved varieties, and pest control, yields may be higher by
severalfold than those under the current traditional systems (Fig. 4).
These opportunities are widespread in many of the less developed
countries low in capital. A good system of terraces requires little or no
capital from outside the local community. It does require skilled technicians to lay out the system. The earth can be moved in a great variety of
ways. It can be excavated and carried on the head in baskets. Bullocks or
other draft animals can be used with simple equipment. Tractors and
machine graders are commonly used in advanced countries where labor is
scarce, but machines are not necessary. Unhappily some giving technical
assistance have emphasized earth-moving machines rather than precise
designs of terraces.
But we repeat, the job must be done well. In areas with small irregular
parcels a reasonable scheme for consolidating fragmented holdings is
essential to make a start toward modern farming. The terraces must be
made at the proper angles to the contour, which depends on the permeability of the soil, its relief, its water-holding capacity, and the character
of the rainfall. Otherwise water breaks over the terraces to make gullies
CHARLES E. KELLOGG A N D ARNOLD C. ORVEDAL
in some places or water accumulates to waterlog the soil in other places.
On many gently sloping soils under mixed cultures of crops, trees,
shrubs, or palms not requiring tillage, the cover itself may slow down the
FIG.4. Water control for paddy rice in the foreground. The hills in the center background
are under shifting cultivation. Brush and wood are piled in strips up-and-down the slopes,
covered thinly with sods, and burned for crops of maize or potatoes. Some of the slopes
have low terraces. The soils were called Yellow Latosols. (Possibly Haplustults in the new
system.) They have been farmed for along time by Khasi tribal people. Near Shellong,
runoff enough for water to enter the soil. Yet on well-covered moderate
or strong slopes unless the soils are highly permeable, at least some
terraces are necessary to have the optimum amount of water enter the
soil and be held in the rooting zone (Fig. 5).
Irrigation is an old water-control practice and plays a great role in
farming in many crowded areas today. With an effective irrigation system
and a good source of water, responsive soils of the deserts and semideserts can be farmed efficiently. The productivity of many other soils
can be improved by supplemental irrigation during regular dry seasons or
during irregular periods of low rainfall. Such irrigation has been increasing in normally well-watered parts of the United States and Europe
POTENTIALLY ARABLE SOILS OF T H E WORLD
where the soils are permeable. Here too, the matter of design is critical
because in these normally well-watered areas a heavy rain may fall
immediately after the soil has been irrigated. If only a little additional
FIG.5 . Well-made terraces in Algeria for runoff control to promote reforestation on
steeply sloping soil barely moist enough for trees. (The soil is probably in the Xeralf suborder.)
water can soak into the soil, without provisions for runoff control and
drainage, roots can be waterlogged and the crops fail.
For best results with irrigation, soils with irregular surfaces need to be
smoothed for sprinkler irrigation or leveled for border irrigation. Such
smoothing or leveling is not practical with soils that are thin over rock or
have lower layers of hardpans, laterite, caliche, gravel, or other materials
that cannot be made into suitable surface soils for cropping. Thus carefully made detailed soil surveys are a prerequisite for planning irrigation
Costly systems that include large dams and canals to provide irrigation
water are seldom economic unless an excellent job is done with the layout
on cultivators’ fields, including leveling of the surface, terraces or bunds
where needed, and artificial drainage as required. The aim in irrigation
should be a complete water system that avoids deficiencies of either water
or air for pIant roots. If these deficiencies of water and air are overcome,