2.1  Natural Environment

The natural environment of the study area in each country is indicated in Table 1.  It is described in more details below.

Table 1  Summary of Natural Environment of the Study Area in Each Country

(1) Cote d'Ivoire

The valley bottom (bas-fonds in French) along the river in the Djebonoua district about 20km to the south of Bouake, one of the main inland cities, served as the study area.  The area is at an altitude of about 330m.  It belongs to the Kan River system, which is the tributary to the right of the Nji River, one of the three largest tributaries of the Bandama River, and of which one of the most upstream rivers is the Kplaha River.  The width of valley bottom paddy fields around the study area including slopes is 150-400m, and 40-120m at the floor though close to 200m at some places.  The relative height between the plateau and the valley floor is 8-10m.  The inclination of slopes is 4-5% on the left bank and 2-4% on the right bank.

The soils' principal mineral is quartz, its clay kaolinite, and its color generally reddish due to the quartz thinly coated with iron film.  The soil textures belong to sand and sandy loam.  The soils of the plateau are classified as Paleustults and Kandiustalfs, of the slopes Ustropepts and Dystropepts, and of the valley bottoms Fluvaquents.  Since the parent material of these soils is weathered granite, their fertility is generally low.

The soils' low fertility is indicated in the results of an experiment conducted as part of the study (Table 2).  This experiment was for examining the effects of the three major nutrients (N, P and K), and many suggestive findings were obtained.

Table 2  Fertilizer Application Experiment (on Paddy Rice) in Cote d'Ivoire

It is recognizable that crop yields increased on average by the use of the chemical fertilizers.  However, yields varied widely between the experimental fields.  This may be because paddies that were of high potential yields and those that were not existed side by side in the area where the experiment was conducted at the valley bottom.  Since the available phosphate (P₂O₅) content was low at the low yield paddies in the first experiment, a section with a doubled amount of P₂O₅ was added in the second experiment.  However, its effects on crop increase were not so high, for the average yield of the control section was 4.5 ton/ha and that of the doubled P₂O₅ section 5.3 ton/ha.  From this result, the low yields cannot be attributed to phosphate deficiency.  That is, it suggests the possibility that there are other soil factors, which limited the yields besides the three major nutrients.

As for the climate, the area belongs to the Guinea savanna zone.  The dry season and the rainy season are clearly divided, and the rainy season lasts from April through October and the dry season November through March.  Annual average precipitation is about 1,100mm, of which close to 85% concentrates in the rainy season.  Annual average temperature is between 24.3° C and 28.3° C, and it is hottest from February to March at the end of the dry season.  Sunshine hours are 7.7 hours a day on average in January and February of the dry season but are only 2.7 hours on average in August of the rainy season, and are 5.6 hours on average throughout the year.

(2) Tanzania

The study was conducted at the valley bottom (mbuga) in Chekereni Weruweru village about 5km southwest of Moshi City, the capital of Kilimanjaro region, located at the south base of Kilimanjaro volcano.  The area's altitude is about 800m.  Located at the mountain toe of Kililmanjaro volcano, its overall inclination is slightly over 1%.  Several springs are found at this valley bottom, of which the water volume increases toward the end of the rainy season, but the water yield, even totaled, is very small.  On the other hand, the Kiladeda River, the main source of irrigation water for rice paddies of about 20ha spreading across the valley bottom, which runs southward about 200m west of the paddies along the slope of the entire volcano base, often floods during the rainy season, and overflows pour into the valley floor.  From the conditions of the area, it is very likely that the valley was formed by overflows.  The valley bottom paddy fields are about 1km long north and south, and about 300m wide at its widest upstream to the north with a narrow section 20m wide about three quarters down from the top, which becomes wider again about 120m wide downstream.  The highland has undulations and the relative height from the valley floor varies.  There are small hills about several meters high to the north and the west and about 10m high to the east of the narrow section.  Further, there is the Njoro spring that has considerable water volume about 400m to the east of downstream of paddy fields.

The parent material of the soil is aeolian andesite volcanic ash.  Since the area is semiarid with little precipitation and dry and rainy seasons, volcanic ashes are weathered to become halloysitic and the soil so produced is presumed to be fertile Paleustolls.  The soil's high fertility was evident in the results of the fertilizer application experiment conducted as part of the study (Table 3).

Table 3  Fertilizer Application Experiment (on Paddy Rice) in Tanzania

The field without fertilizer application produced unhusked rice yield of 3 ton/ha, which increased up to 7 ton/ha only with the nitrogen fertilizer.  Even compared to the experimental results in the other countries, the soil fertility of this area is overwhelmingly high.  In the neighboring districts under the Kilimanjaro Agricultural Development Project (KADP) implemented in the 1980's with assistance of the Japanese government, high yields exceeding 6 ton/ha have been maintained over 10 years only with nitrogen application.

However, there is sodium accumulation in the subsoil of this area.  If topsoil treatment were done inappropriately during paddy field development, salt damage might ensue, and thus caution should be in order.  This point will be referred to later.

As for the climate, the rainy season is from December to May and the dry season from June to November, and the annual average precipitation is small at 600-900mm.  Having mountain climate characteristics, it tends to rain more the closer to the slopes, and annual precipitation reaches 2,000mm at the middle slopes of Mt. Kilimanjaro.  Rainfall to this mountain turns into springs at its base, putting crop production in the study area and the surrounding valley bottoms at advantage.  However, the sodium content of the spring water may be excessively high depending on the springs and as there have been sporadic reports of salt damage due to the high sodium content, attention should be paid to the quality of water.

The annual average temperature is between 21.8° C and 27.3° C and the range is relatively wide.  Temperature increases during the rainy season and reaches highest in February and March.  It cools down during the dry season, and especially from July to August the lowest temperature may reach as low as about 10° C.

(3) Zambia

The study area is the valley bottom (dambo) formed upstream of the Chambanshi River about 40km north of Kasama, the capital of Northern region, which is about 800km northeast of the nation's capital of Lusaka.  The altitude is about 1,300m.  The Chambanshi River is a tribute of the Chambeshi River about 100km to the east of Kasama City, which runs to the southwest, but its water volume is very small.  The valley bottom spreads long and narrow along the stream of the Chambanshi River, and the right bank is about 200m wide and the left bank about 50m at the study point.  The valley bottom, which is almost totally unutilized, is grassland.  Its highland is woodland, called Miombo woodland, of very simple appearance with legume trees (under 10m high), which gently slopes down toward the river.  If only the grassland is regarded to constitute the valley bottom, the relative height from the river does not exceed 10m.

The principal mineral of the soils of the plateau is mainly quartz sand.  For this quartz sand is accompanied by kaolinitic clay with its quartz thinly coated with iron, the soils' appearance is reddish.  The soils are Kandiustults and Paleustults with a laterite horizon in the substrata as a common characteristic.  For topsoil, the section near streams in the valley bottom has peat and that close to the Miombo woodland at somewhat higher relative height has a mucky humus horizon.  While the principal mineral of these soils is quartz, having almost no iron coating, their color is grayish pink.  Their classification is sandy Fluvaquents, and considered to be Hystic Fluvaquents near the flat valley bottom.  Since the soils were formed over a geochronologically extremely old shield, their fertility is very low.  Various nutrients may be lacking, including not only the three major nutrients but also such nutrients as sulfur, magnesium, boron, and others, and even minor nutrients.[1]  These are suggested by the results of the fertilizer application experiment conducted as part of the study (Table 4) and through observations of the crops grown in the fields and sold at markets.

Table 4  Fertilizer Application Experiment in Zambia

Regardless of treatment, the yields of both paddy rice and maize are extremely low.  This strongly indicates the unusually low natural fertility of the soils.  Also, the effects of sulfur application on crop yields are quite evident for both paddy rice and maize.  Especially for paddy rice, the yield at the field in which sulfur was added reached 2.3 times as high as when only the three nutrients were applied.  It is apparent that the soil of every field is lacking in sulfur.  Moreover, a study on various crops growing around the area and vegetables sold at markets suggests that there are other nutrients such as boron, zinc and magnesium lacking in this area's soils besides sulfur.

The area's climate is clearly divided into the dry season and the rainy season; the rainy season from November to April and the dry season for the rest.  Annual precipitation is about 1,100mm, of which close to 90% concentrates during December through March.  Further, there is practically no substantial rainfall from May to October.  As for temperature, the annual average is about 20° C, and the monthly average is the highest in October at 23.3° C and the lowest in July at 17° C.

(4) Malawi

The study area is the valley bottom (dambo) in Bandawe village of Salima district, about 100km to the east-northeast of the nation's capital of Lilongwe.  This valley bottom, close to Lake Malawi shore, is slightly over 1km long in the major axis and about 400m in the minor axis.  Reportedly, the area used to flood throughout the year but it turned into a marsh as the water level of Lake Malawi dropped.  At its south end, the Nyungwi River runs to the east and flows into Lake Malawi.  During the rainy season, the water level of the Nyungwi River rises and the overflowing floods the valley bottom.  However, during the dry season, the groundwater level drops and the ground dries up to the point of cracking.  The valley bottom is in the shape of a shallow dish and the relative height to the surrounding areas is about 2m.

The soil is Vertisols of sandy loam and its color is close to black.  The soil of the flat valley bottom is fairly fertile, but as it turns more coarse sandy loam toward the upper section of the slope, fertility in the usual sense decreases.  The results of the fertilizer application experiment at the flat valley bottom are presented in Table 5.

Table 5  Fertilizer Application Experiment in Malawi

The effects of the fertilizer applications are evident, but even without the fertilizers the yields are relatively high.  However, in the condition without fertilizer, the timing of heading was delayed for about 20 days, suggesting that growth retardation occurred.  This might have been caused by phosphate deficiency.

The area's climate belongs to tropical savanna, and annual precipitation is slightly less than 1,200mm, of which 90% concentrates in December through March.  Further, there is almost no rainfall from June through October.  Annual precipitation has fluctuated between 600mm and 1,900mm in the past 11 years.  Temperature averages 25° C throughout the year.  It is the hottest in January with the monthly average temperature at 32° C for the highest and 24° C for the lowest, and the coolest in June and July with the monthly average temperature at 27° C for the highest and 17° C for the lowest.  As for the sunshine hours, October has the longest at 10.1 hours per day on average and January the shortest at 5.6 hours.  Annual hours of sunshine are 8.7 hours per day on average.

2.2  Issues on Agricultural Development in Relation to Climate and Soils

(1) Climatic Environment

The division of the rainy season and the dry season is clear in every one of the areas studied.  The rainy season is brought by rainfall caused by the north-south movements and the onset of the intertropical convergence zone (ITCZ).  However, because the movements are irregular, the timing of the beginning and end of rainfall is not necessarily fixed and the amount of rainfall is apt to fluctuate as well.  Further, due to the influence of the ascending current inland, heavy rains occur at times.  These annual fluctuations related to the timing and amount of rainfall and heavy rains work against agricultural production.  As for hours of sunshine, the lack of sunshine during the rainy season could be a negative factor of crop production in Cote d'Ivoire.

Meteorologically, in every study area the monthly average temperature is in the 25-30° C range and even the lowest temperature is generally over 20° C during the rainy season that is the time of crop cultivation.  Therefore, the likelihood of temperature negatively affecting crop cultivation in the rainy season is nil.

On the other hand, in the dry season there is no rainfall and water is a limiting factor of crop production, but hours of sunshine are long and there are many advantageous conditions for crop production.  For example, in the eastern and the southern Africa where altitudes are generally high, the lowest temperature during the dry season drops below 20° C allowing cultivation of temperate zone crops.  In terms of water environment, inland valley bottoms are endowed compared to highlands, for even during the dry season there is surface water and groundwater levels are high.  Thus, it may be argued that crop production potentials at inland valley bottoms are high in that water resources development is easier at relatively low cost.

(2) Soil environment

By and large, the fertility of African soils is low.  This has been recognized anew in the recent years, and the International Workshop on Development of National Strategies for Soil Fertility Recapitalization in Sub-Saharan Africa was held in April of 1997 under cosponsorship of the World Bank and the International Fertilizer Development Center.  At this workshop, it was recognized that soil fertility improvement would play a very significant role in agricultural advancement, rural development and environmental conservation in Sub-Saharan Africa, and it was confirmed that implementation of soil fertility improvement would be reflected in every country's national action plan.  The soil environment of the study areas especially in terms of fertility is explained below.

(i) Low fertility and low productivity of African soils

The reasons for the low fertility of African soils are outlined as follows.  The African continent is geologically very old and has never been subject to diastrophism such as upheavals and subsidences for the past several hundred million years.  Erosion cycles in the geological sense have repeated countless times during the period, turning it into a plateau continent.  As a result of eons of erosion and weathering, the residual soil mineral is quartz that is most resistant to weathering.  Of the study areas in the four countries examined, the soils' principal mineral was quartz (silicon dioxide) except the soil in Tanzania, which was formed by fallen ash from Kilimanjaro volcano to be described later.  An ordinary mineral has aluminum, iron and silicon as main constituent elements, and such minerals are apt to contain metallic elements essential for plant growth as they are attached to iron.  However, in Africa dominant are quartz-based soils with aluminum and even iron leached out, which are less likely to contain essential elements.  Therefore, the effects of fertilizer application generally tend to be marked, as mentioned previously, as observed in the experiments conducted in each country.

Because of the low soil fertility, element deficiency diseases were observed in various crops across the areas examined, but sulfur, boron and zinc deficiencies were especially noticeable.  In particular, the occurrences of symptoms of sulfur deficiency and minor elements deficiencies were widespread in Zambia.  In the country, liberalization policies by IMF and the World Bank were put into effect in 1992, and all the restrictions on import of fertilizers, which had been generally under the government's control, were abolished making it possible to import and sell any kind of fertilizer freely.  As a result, the import of cheaper fertilizers of only NPK compounds increased dramatically.  As of 1997, the final year of the study, according to the only fertilizer trader (House of Kasama) in the central city of Kasama, no NPK fertilizer containing sulfur had ever reached the store for six years since 1992.  It was only 1999 when its import started again.  As seen from the results of the studies conducted by the study team, it is not hard to imagine how grave the farmers' losses must have been in this period.  The unavailability of these fertilizers is apparently beyond the local farmers' control, and also it cannot be denied that the central government's administrative capacity (and the lack thereof) has had very serious effects on the development of individual farmers as well.

Moreover, soil fertility is often discussed with reference to organic matters and humus in soil, and humus normally becomes stable as it combines with metallic nutrients such as aluminum and iron.  However, in case of soil with quartz as the principal mineral, humus decomposes and disappears very rapidly because of its free condition, and thus the accumulated effects of humus through compost and such cannot be expected in the long run.

The aforementioned is generally applicable to soils in not only inland valley bottoms but also Africa as a whole.  The regional distribution of element deficiencies and toxicity in Africa has been reported as shown in Figure 2.

Figure 2  Locations of Micronutrients Deficiencies and
Distribution of Toxic Nutrients
Source: Kiuch & Wakatsuki, 1990.

To improve crop yields, soil fertility improvement is essential including fertilizer application.  However, unfortunately the consumption volume of chemical fertilizers is merely 19kg/ha in the entire Africa.  That is only one fifteenth (1/15) of 287kg/ha in Japan.

(ii) Characteristics of inland valley bottoms

As characteristics of the general shape of an inland valley bottom, first the vastness of its highland plateau may be noted (Figure 3).  Valley bottoms, though reflecting the stability of the African continent as a continent, occupy only a part of the tectonic flat topography.  Also, the inland valley bottom is formed into the topography of an oval, shallow dish, for the rivers flow gently from upstream to downstream.  The gentle slope of the dish-shaped topography is on the fringes of a sloping valley of eroded morphology and thus the substrata of the upland soil are exposed.  Further, in case of the valley bottom farthest upstream where the flow of rivers is extremely gentle, there may remain stagnant water during the dry season for a while at times, and under such conditions peaty soils may form at the bottom of the concave.

Figure 3  Shapes of Inland Valley Bottoms: Models

It should be noted here that at the flat valley bottom, which is more or less level, the fertility of topsoil would most likely differ between the area close to the slope and the middle section.  At the area close to the slope, since the topsoil should be formed by parent material fallen from the slope, the soil texture would be more coarse and its fertility would be low.  On the other hand, the topsoil of the middle section should be of alluvial deposit of parent material carried by river flow.  Considering the near level topography, the speed of river flow would have been slow and the grain size of the carried parent material would be finer, and hence the soil fertility is likely to be higher.  This indicates the possibility that soil fertility may vary depending on the section even within the same valley bottom, and scrutiny on the variance of soil fertility would be called for at the time of development.  Furthermore, if peat soil is present, care must be taken not to let it dry up and decompose, taking into consideration its large water holding capacity and buffer against groundwater runoff.

(iii) Fertile volcanic ash soil and salt damage

In the Great Rift Valley region of the geologically old African continent, soils are generally fertile because there are many volcanoes leading soil rejuvenation to occur with their eruptions, and being in a semiarid zone, volcanic ashes form soils through the halloysitic weathering process.  The southern mountain foot of Kilimanjaro volcano, one of the study areas, is part of the Great Rift Valley.  The area's soil is no exception, very fertile, and crop yields are very high compared to the other areas as shown in Table 3.  Yet, salt damage (sodium damage) was detected in paddy rice as an idiosyncratic problem to this area, and the reason for this problem is discussed below.[2]

The sodium content of the soils in this area becomes higher the deeper into the substrata.  At places where subsoil had been exposed by cutting when paddy fields were developed, the rice growing there showed signs of saline damage.  Also, there are some springs and rivers around the area with water high in sodium, and from irrigating paddies with those as water sources saline-damaged paddy may ensue.  Further, at paddies that were not filled with water due to water shortage, cases were observed, in which rice received salt-related damage.  In these cases, sodium in the soil or the irrigation water became accumulated in the topsoil, accompanied by evaporation of water from the exposed soil.

The high sodium concentration in the soil and rivers/springs in this area was presumed to have resulted from lava runoff from Ol Doinyo Lengai volcano about 100km to the west-northwest of Kilimanjaro volcano.  The major nutrients of the chemical composition of ordinary lava flow and extruded volcanic ash are silicon, aluminum, iron, and others.  However, the lava runoff from the volcano in October 1960 revealed that the lava flow of Ol Doinyo Lengai volcano contained almost none of these nutrients but instead was mainly composed of sodium and carbon dioxide.  In other words, it is speculated that volcanic ashes with sodium carbonate as the main constituent fell to this area in large quantities prior to the ash fall from Kilimanjaro volcano, which forms the present ground surface.  This explains both the salt damage derived from soil as well as the salt damage derived from irrigation water in this area.  The mineral of this type of composition is called carbonatite and there are 330 volcanoes in the world that have been known to produce carbonatite.  Of these about a half are found in Africa, the majority of which in turn are distributed in and around the Great Rift Valley (Figure 4).  Hence, at areas in which carbonatite originates, it is necessary to measure sodium concentration in soils and river/spring water, calculate not only pH or EC (electric conductivity) but also SAR [3] (sodium adsorption ratio) considering the dissolution characteristics of sodium carbonate (i.e., relatively moderate pH increase and moderate EC), and take appropriate measures against salt damage.

Figure 4  Distribution of Carbonatite around the Great Rift Valley