RSS | Register/註冊 | Log in/登入
Site search:
Home>FFTC Document Database>Extension Bulletins>Soil Degradation in the Coastal Lowlands of Southeast Asia
facebook分享
Soil Degradation in the Coastal Lowlands of Southeast Asia
Kazutake Kyuma
Professor Emeritus of Kyoto University, Japan
1-79 Nagatanicho, Iwakura, Sakyoku
Kyoto 606-0026, Japan, 2003-10-01

Abstract

In the coastal lowlands of Southeast Asia, two kinds of soils are vulnerable to degradation under the influence of careless human interventions. One is coastal mangrove areas, and the other is freshwater swamp forests. In recent years, large areas of both kinds of soils are being developed, either for agriculture or for aquaculture. This often results in irreversible degradation. This Bulletin reviews the genesis and general properties of these two soils in coastal lowlands, and discusses the process of their degradation as a result of human activity. Both mangroves and swamp forest are important landscape units in Southeast Asia. They should be preserved by protecting the underlying soils from degradation.

Introduction

In the coastal lowlands of Southeast Asia there are two kinds of soils vulnerable to degradation. One supports mangroves in a brackish water environment, and the other sustains swamp forest in a freshwater environment. In recent years, large areas of both these soils have been reclaimed for human utilization. This often results in irreversible degradation, with total loss of the natural vegetation and soil resources.

Mangrove mud contains sulfidic materials, mainly in the form of pyrite. As the mud is exposed to the air, either by natural land accretion or by artificial drainage, these sulfidic materials undergo oxidation. As a result, sulfuric acid then forms, leading to strong soil acidity. It is not easy to get rid of these harmful oxidation products, whether by washing the sediments or by applying liming materials. As a result, very often the land is abandoned after a short period of utilization, reverting to brackish water swamps with gelam (Melaleuca cajupute) trees. Not only is there strong acidity, but also induced aluminum (Al) and iron (Fe) toxicities. There are also nutrient deficiencies, particularly of phosphate (P). Examples of successful reclamation are rare.

Swamp forest is a landscape unique to insular Southeast Asia. It is found in the tropics, and mostly grows in woody peat which has accumulated in freshwater swamps left after the seaward shift of mangroves. Reclamation of tropical peat has three serious problems: land subsidence, underlying sulfidic sediments, and the extreme oligotrophy of peat, including micronutrient deficiencies

Land subsidence is due to the drying, compaction and decomposition of the peat itself. Of these, it is decomposition which contributes most, after the initial rapid dehydration. Often a shallow layer of peat completely disappears in a few decades, exposing underlying sulfidic sediments. The strong acidity that accompanies oligotrophy, along with copper deficiency, often causes sterility in rice and other crops.

Acid Sulfate Soils Derived from Mangrove Mud

Mangroves are considered to have originated in Southeast Asia (Yamada 1986). They now occur widely in coastal areas of tropical Asia. Mangrove swamps consist of many tree species, all of which can tolerate marine and brackish water. They protect shorelines from wave action, and provide spawning and hatching grounds for fish and prawns, as well as the raw material for making charcoal. Not only do their well-developed root systems create a special kind of landscape, but also their strategic location is important for accelerating the deposition of mud sediments and the seaward build-up of land.

In their natural state, mangroves are a very important part of the coastal ecosystem in tropical and subtropical regions. However, once they undergo a change that exposes mangrove mud permanently to the air, such as drainage and land reclamation, they become profoundly affected by the formation of acid sulfate soils. This change is probably one of the most damaging forms of land degradation.

Chemical Reactions Leading to the Formation of Acid Sulfate Soils

Acid sulfate soils are formed as a result of the oxidation of sulfide-containing sediments. This happens when the sediments are exposed to the atmosphere, in the course of natural land accretion or in the process of artificial land reclamation. The general conditions that lead to the formation of sulfide-containing sediments are:

  • The presence of marine or brackish water as the source of sulfate;
  • Stagnation of water, as typically found in lagoons and bays; and,
  • A supply of decomposable organic matter.

The last condition is also favored by the second. The stagnation of water facilitates the establishment of vegetation such as mangroves. The slow build-up of coastal land by silting may be counted as another condition that favors the formation of sulfide-containing sediments.

The chemical process of the formation of sulfidic compounds in mangrove mud, is shown in Table 1(1125).

It is known that the greater part of these oxidizable sulfide compounds are present in the form of pyrite, (FeS2). This is because pyrite is much more stable than ferrous sulfide, (FeS). When sedimentary pyrite is being formed, there must first be not only sulfate reduction, but also the formation of elemental sulfur. Van Breemen and Pons (1978) have suggested that the partial oxidation of sulfide to elemental sulfur may occur.

This process is mainly anaerobic, alternated with a limited aerobic process. These conditions can best be met in the zone of tidal fluctuation. According to Stumm and Morgan (1970) however, under acid conditions, elemental sulfur is formed as a result of the reduction of SO42- as an intermediate product. Once formed, it may persist as a stable solid phase. Thus, solid elemental sulfur is quite common in recent marine sediments.

Upon exposure to the air, pyrite undergoes oxidation as follows:

FeS2(s) + (7/2)O2 + H2O = Fe2+ + 2SO42- + 2H+ (1)

The ferrous iron is further oxidized to ferric iron. This is precipitated if the environmental pH is higher than about 3.

Fe2+ + (1/4)O2 + H+ = Fe3+ + (1/2)H2O (2)

Fe3+ + 3H2O = Fe(OH)3(s) + 3H+ (3)

Thus, the overall reaction is:

FeS2(s) + (15/4)O2 + (7/2)H2O = Fe(OH)3(s) + 2SO42- + 4H+ (4)

producing four equivalents of acidity from the oxidation of one mole of pyrite.

It is known that the reaction (2) is a slow process if it proceeds in a purely chemical fashion. At a pH of 3, the half life of this reaction is around 1000 days (Stumm and Morgan 1970). In the soil, however, this reaction is mediated by autotrophic iron bacteria, Thiobacillus ferrooxidans and Ferrobacillus ferrooxidans, and proceeds much faster.

Another important reaction is the oxidation of pyrite by Fe3+,

FeS2 + 14Fe3+ + 8H2O = 15Fe2+ + 16H+ + 2SO42-

This produces even more acidity. The reaction runs quite rapidly. The half life of this reaction is from 20 to 1000 minutes. The oxidation of sulfur in this reaction is mediated by another autotrophic bacterium, Thiobacillus thiooxidans.

Stumm and Morgan (1970) provided the following schematic drawing of the overall process of pyrite oxidation ( Fig. 1(1157)) for coal mine waste. They stated that:

  • "To initiate the sequence, pyrite is oxidized directly by oxygen (a) or is dissolved and then oxidized (a'). The ferrous iron formed is oxygenated extremely slowly (b) and the resultant ferric iron is rapidly reduced by pyrite (c), releasing additional acidity and new Fe (II) to enter the cycle via (b). Once this sequence has been started, oxygen is involved only indirectly in the reoxidation of ferrous iron (b), the oxygenation of FeS2 (a) being no longer of significance. Precipitated ferric hydroxide serves as a reservoir for soluble Fe (II) (d). If the regeneration of Fe (II) decreases, it will be replenished by dissolution of solid Fe(OH)3".

In the soil, the reaction (b) is mediated by iron bacteria, and thus it may not be seriously rate limiting. But it has to be noted that to make smooth pyrite oxidation possible, it is necessary to keep the ferric iron activity high. For this to be realized, the pH of medium must be kept sufficiently low. In this relation, Murakami (1965) remarked that it is noteworthy that liming at an early stage in the reclamation of sulfide-containing sediments can retard the oxidation of pyrite.

Jarosite is an intermediate oxidation product of ferrous sulfate in an acid medium:

Fe2+ + SO42- + (1/4)O2 + (3/2)H2O + (1/3)K + = (1/3)KFe3(SO4)2(OH)6 + H+ + (1/3)SO42-

Jarosite appears in the pores and cracks of the soil as pale yellow (or straw yellow) mottles, and is indicative of acid sulfate soils. Jarosite is further hydrolyzed in a less acidic medium, finally to precipitate iron as hydrated ferric oxides, releasing more acidity:

2KFe3(SO4)2(OH)6+6H2O = K2SO4+6Fe(OH)3+3H2SO4

The acidity produced by pyrite oxidation may be neutralized if there is an large amount of lime (CaCO3) contained in the sediment or brought in by water. In some coastal areas, plenty of fine gypsum crystals can be seen in the soil as the result of a neutralization reaction. However, in most of the humid tropics, the carbonate content is low or nil. Acidity remains in the soil to give a pH sometimes as low as 3.

Van Breemen and Wielemaker (1974) indicated that the original clay minerals in the acid sulfate soils in the Bangkok Plain of Thailand _ ferric oxides, jarosite, a basic aluminum sulfate (ALOHSO4), amorphous silica and sometimes gypsum _ maintain a very stable pH of 3.6 to 3.8. They suggested that in sediments with a low content of smectite clays, the equilibrium pH may be close to 3.0-3.5. If we leave these acid sulfate soils with their low pH to the natural submergence and draining cycles, the deacidification mechanism would work finally to raise the soil pH to between 4.5 and 5.

Reclamation and Amelioration of Acid Sulfate Soils for Rice Cultivation

Because of the strong acidity, if acid sulfate soils are to be reclaimed for agricultural production, various human interventions are necessary. As acid sulfate soils are usually found in swampy lowlands, rice is often the first crop to be considered. However, various factors are involved in establishing paddy fields in an acid sulfate soil.

Rice is known to tolerate relatively high acidity. In a solution culture, the pH had to be less than 4 for rice to be adversely affected (Ponnamperuma et al. 1973). Under such a low pH, aluminum toxicity may be more important than the direct effect of hydrogen ion concentrations. In some studies conducted in Japan, rice seedlings started to show growth inhibition when the Al3+ ion concentration exceeded 35-40 mg/kg. Cate and Sukhai (1964) and others quoted an even lower threshold of 25 mg/kg at which rice seedlings started to show aluminum toxicity. These toxic levels of aluminum can occur in acid sulfate soils with a pH of 3.5 or less, according to Van Breemen and Pons (1978).

Ferrous sulfate produced in quantity during the oxidation process of pyrite is also said to affect the growth of rice adversely (Kobayashi 1939). It is known that iron toxicity results from a complex nutrient imbalance, but still the high ferrous ion concentration, coupled with high acidity and/or high salinity, may be an important element of the disorder frequently found in acid sulfate soils.

Of all the nutrient elements, probably phosphorus (P) is the most critical in acid sulfate soils. In addition to the inherent lack of P in these soils, the high activity of aluminum lowers the availability of P to rice. Phosphorus is so acutely deficient in some soils that no response, even to lime and nitrogen (N), can be expected, unless P is applied. Kawaguchi and Kyuma (1969) suggested that one of the benefits of liming and P application lies in the increased supply of N by the stimulation of ammonification. Matsuguchi et al. (1970) found that applying lime and phosphate definitely increased the amount of biologically fixed N. Thus in many acid sulfate soils, P is the limiting factor for the increased growth and yield of rice.

The following measures may be effective in reducing the harmful effects of toxic substances:

  • Preventing the oxidation of pyrites contained in potentially acid sulfate sediments;
  • Leaching harmful substances out of the rooting zone after allowing oxidation to occur; and,
  • Inactivating aluminum by raising the pH of the medium by liming. This also reduces the concentration of ferrous iron in the soil solution.

All of these measures have been recommended by researchers working with acid sulfate soils, to ameliorate them for rice cultivation.

However, the first measure, preventing the oxidation of pyrite, is not a positive solution to the acid sulfate soil problem. It suppresses the natural ripening process of the mud (potential acid sulfate sediments). Because the soil would still have strongly reducing conditions and be very difficult to work, the rice yield would be very low, and yet no positive measures for yield improvement could be taken.

In addition, this `apparently easy' solution is in practice very difficult, because of unpredictabale weather conditions. When unusually long dry spells occur, the soil surface is inevitably oxidized and strongly acidified. This is likely to lead to total failure of the rice crop. Tanaka and Yoshida (1970) reported on a crop failure of this kind in Camarines Sur, Philippines.

The second measure, leaching out harmful substance, can be used only where there is good drainage. One lysimeter experiment was conducted by Murakami (1965) in Japan in a heavy clay soil (50% clay and 42% silt). The plot had tile drainage at a depth of 80 cm. The third crop (one crop a year) gave over 4 mt/ha of paddy rice. In contrast, a plot without drainage gave almost no yield through the first four crops, and only later did the yield increase. In the first three years of percolation treatment, the oxidizable sulfur content decreased from the initial 192 mg to 12 mg/kg soil. This area has around 800 mm of precipitation during the fallow months, which must have contributed greatly to the removal of toxic substances through natural leaching.

We must consider the conditions under which acid sulfate soils occur in Southeast Asia. Most of the areas are low-lying, and natural drainage is severely restricted. During the rainy season, the entire land surface is submerged. During the dry season, however, the top 5-10 centimeters of soil become dry. Surface water during the rainy season is mostly fresh, and washes the top few centimeters of soil. In the dry season, upward movement of water cancels out this effect, bringing the toxic substances back to the surface.

The minimum requirement for the reclamation of acid sulfate soils is to dig open ditches for drainage. The exception is when the first measure, preventing oxidation, is taking place. Often tidal gates have to be installed in coastal areas to prevent the intrusion of salt water during the dry season. This investment would certainly reinforce the washing effect during the rainy season, but it is not effective at all during the dry season unless there is an ample supply of fresh water. This is generally difficult to obtain in a monsoon climate zone.

In areas with a permanently humid climate, as typically seen in Sarawak, East Malaysia, leaching first with sea water or brackish water, followed by leaching with plenty of rainwater, could be an effective way of reclaiming acid sulfate soil. This method of soil amelioration has been adopted in Sierra Leone in West Africa, with some success. Piling up the soil into ridges, or digging a close network of shallow ditches, should also prove effective in such areas for upland crop cultivation, as long as the fields are connected to a major drainage system. The surjan system in Indonesia, with broad ridges (guludan) and shallow furrows (tabukan), is a time-honored local practice for the utilization of acid sulfate soils.

The third measure, liming, is effective in raising the soil pH and thus inactivating toxic aluminum. But if this measure alone is taken, the amount of lime required can easily rise to an uneconomically high level. Van Breemen and Pons (1978) calculated that complete neutralization of the acidity generated in the uppermost 50 cm of soil containing 3% pyrite would require 150 mt/ha of lime, even when half the acidity is removed by leaching. Moreover, it must not be forgotten that acidity is generated only slowly upon exposure of pyritic mud to the air during the dry season. This means that the lime applications have to be repeated year after year for a long time.

Another point to be considered is that the oxidation of pyrite, the dominant form of oxidizable sulfur, is retarded if the pH is higher than 3. Therefore, if lime is applied at the beginning of reclamation, the time required to leach out the toxic products of oxidation would be prolonged.

Thus, liming should be carried out only after oxidation and leaching are quite advanced. In Thailand, liming is recommended in amounts just high enough to inactivate aluminum (Komes 1973). Further liming to produce a higher pH is thought to have the adverse effect of causing sulfate reduction in the rooting zone. As a matter of fact, a significant amount of sulfate reduction begins to take place when the pH is higher than 5 (van Breeman 1975). Murakami's experiment (Murakami 1965), however, indicates that acid sulfate soils usually do not suffer from the harmful effects of hydrogen sulfide because sulfate reduction is rather rare. He argues that in these soils, ferric oxides are liberated and precipitated as a final product of pyrite oxidation, and so the Eh is kept relatively high, high enough to suppress sulfate reduction. (The Eh is a measure of the oxidation-reduction process).

When oxidation and leaching take place, whether or not liming is practiced, soil nutrients are lost. Accordingly, soil fertility is probably reduced. Murakami (1965) however, concluded from his experiments that such effects are not serious. His results revealed that:

  • All of the exchangeable cations showed a marked decrease, but levels of Calcium (Ca), Magnesium (Mg) and potassium (K) in the soil solution were kept at a level high enough to supply rice plants with these elements;
  • The CEC did not decrease appreciably;
  • The P sorption capacity did not increase;
  • The C/N ratio of the soil organic matter fell from about 20, to 10 at the end of three cropping seasons, showing the maturation of humic substances in the soil.

Fertility Characteristics of Acid Sulfate Soils

Some of the studies conducted on the fertility status of acid sulfate soils are introduced below.

Table 1(1125) gives the fertility-related properties of the soils in the Bangkok Plain of Thailand (Attanandana et al. 1981). The samples of acid sulfate soil are at different stages of maturity or amelioration, but their pH is invariably < 5, and even < 4 for Rangsit Very Acid soils. This can be compared with the pH > 5 for non-acid sulfate soils studied as a control. Reflecting the low pH, the level of exchangeable Al is also very high for Rangsit Very Acid soil, higher by two orders than the control. Available P and silica are lower in acid sulfate soils, and again are very low in Rangsit Very Acid soil. In spite of all of these factors, acid sulfate soils still retain a relatively high level of 1.4 nm minerals (mainly high-charge smectites) and thus a high CEC. However, this feature, combined with a very low pH, could lead to high levels of exchangeable aluminum (Al), as seen in Rangsit Very Acid soil. This would cause rice to suffer from Al toxicity. Jarosite sulfur is found only in acid sulfate soils, but total sulfur, as well as water-soluble sulfur, is quite high, even in a control soil. This may be due to the presence of gypsum in the Bangkok soil series.

As shown here, the major fertility problems of acid sulfate soils are a low pH, a high level of exchangeable Al, and low levels of P and silica. However, since the clay mineral composition has not yet been severely degraded, if acidity is carefully amended, rice is expected to respond well to fertilizer applications. Attanandana et al. (1981) conducted a pot experiment using the five soils listed in Table 2(1173). They showed that Sena and Rangsit soils could attain the same yield levels as Bangkok and Ratchaburi soils, if properly limed and fertilized, particularly with respect to P and K. However, rice grown in Rangsit Very Acid soil failed, and showed symptoms of Al toxicity, in spite of a preliminary application of lime. This was probably, due to the new generation of acidity in the rhizosphere from the hydrolysis of basic aluminum, or iron sulfates such as jarosite.

Phosphorus is the crucial nutrient element in acid sulfate soils. In the above pot experiment, the growth of rice in Rangsit very acid soils was almost completely inhibited in the plot with no applied P. Any phosphorus present in the soil would have been caught by the active Al that was abundant in this soil.

However, this may also suggest that phosphate may be effective in suppressing Al toxicity in rice. Attanandana et al. (1982) demonstrated, using an X-ray microanalyzer, that in a high-phosphate plot, Al taken up by rice was concentrated in the epidermis of rice roots along with the P. This suggested that phosphates applied to the vicinity of rice roots could be effective in suppressing Al toxicity. They conducted a field experiment in which they inserted mud balls mixed with phosphate into the root zone of rice seedlings. This gave certain positive results in the yield, as compared with a homogeneous application of the same amount of phosphate.

Taking advantage of the strong acidity of acid sulfate soils, pulverized rock phosphate without any processing may be effectively used as the source of P. Depending on the origin of rock phosphates, the response of rice could vary quite widely, but there will certainly be more use of this method in the future in the tropics.

Organic Soils Sustaining Swamp Forest

Peat occurs throughout the world where the balance between the addition and decomposition of organic matter favors its accumulation. This may be because of excess water, or anaerobic conditions, and/or low temperatures. Globally, peat covers more than 200 million ha of land, of which about 30 million ha are in the tropics.

The proportion of peat in insular Southeast Asia is very high. Southeast Asia has a land area which is only 5% of the world's total tropical land, but it has some 20 million ha or two-third of the total tropical peat area, mostly around the Sunda Shelf (Driessen 1978; Dent 1986). The most widely accepted definition of a peat soil is a soil with an organic layer at least 50 cm deep, which contains more than 65% of organic matter by weight.

Characteristics of Tropical Peat

In temperate and cold regions, it is usual to classify peat into three groups: low-moor peat, intermediate peat, and high-moor peat.

Low-moor peat is formed in a submerged depression. As the depression gradually fills with the plant debris of aquatic vegetation (mainly reeds and sedges that are still rooted directly in the underlying mineral soil), intermediate peat starts to form.

This is made up mainly from the plants that have been rooted in the low-moor peat, and are thus poorer in mineral elements. By this time, free water surface disappears due to the complete burial of the depression by the peat.

High-moor peat then forms on the intermediate peat. It is mainly made up of sphagnum moss. The surface of high-moor peat is raised over the previous water surface, though the sphagnum peat is saturated with water.

High-moor peat is poorest in mineral elements, because it has no footing in underlying mineral soils.

In contrast to the dominantly herbaceous nature of temperate peat, tropical peat is made up mostly of the remains of woody stems and branches of swamp forest vegetation. It is thus called `woody peat'. According to Driessen (1978), tropical peat may be classified into topogenous peat and ombrogenous peat.

Topogenous peat formation is similar to that of low-moor or intermediate peat, starting with the establishment of aquatic vegetation in a permanently submerged depression and ending in the complete burial of the depression with organic debris.

Upon this topogenous peat, ombrogenous peat then forms under conditions of year-round rainfall and excessive humidity, under the closed canopy of swamp vegetation. The peat heaps up gradually, to form a dome. At the same time it extends until it covers the entire basin. Like high-moor peat, ombrogenous peat is poor in mineral elements because it is formed from vegetation that was self-supporting, relying on the limited amount of mineral nutrients released from the underlying peat mass.

Tropical peat, as observed in Sarawak and Sumatra, sometimes has a depth of more than 10 m. Around 75-95% of its pore space is filled with water, so the bulk density ranges from 0.05 _ 0.4 g/cm3. It is very difficult to imagine how this thin slurry of semi-decomposed organic debris can support a swamp forest with 300 mt/ha of above-ground biomass. According to carbon-dating data, the deposition of tropical peat is considered to have begun from 6000 to 6500 years ago, during the Hypsithermal period, with a mean rate of deposition of 2-3 mm per year. This is much faster than the rate of 0.6-1 mm per year measured for herbaceous peat in Japan.

Reclamation of Tropical Peat Soils for Rice Cultivation

When peat soil is being reclaimed, drainage must come first, even if rice is to be planted. Only around 5-7% of the deep, loose peat dome is solid, while the remaining pore spaces are fully saturated with water. In densely packed peat, only 25% of the total volume may be solid. Therefore, unless drainage is provided, peat has almost no capacity to support a standing crop.

To drain peat, one-meter deep, open ditches are usually dug at 20 to 40 m intervals. As drainage proceeds and the peat dries out, considerable subsidence of the land takes place. As the dehydrated peat decomposes, subsidence is further enhanced. Therefore after some years, the drainage system must be adjusted. A rate of subsidence of as much as one meter each year has been recorded in the first year of reclamation of deeply drained ombrogenous woody peat.

According to Driessen (1978), most of the nutrient elements in tropical peat accumulate in the top 25 cm, where a dense root mat is formed ( Fig. 2(1120)). The subsurface layer commonly contains a much lower level of mineral elements, and the layers below 80 cm, which are almost the lower limit of the living root system, contain only tiny amounts. This means that almost all the mineral elements contained in a peat soil are recycled rapidly by the vegetation. Therefore, felling of the vegetation and decomposition of the surface layer of peat inevitably lead to a rapid loss of mineral nutrients from the soil system.

Driessen (1978) gave the example of a deep peat soil under virgin mixed swamp forest. It contained 13,250 kg of mineral constituents per hectare in its upper 80 cm layer, of which some 10,850 kg were involved in cycling, while the rest (2400 kg/ha) were in the structural part of organic materials. In the `padang' forest, which is a poor stand, naturally occurring in the center of a dome peat and often established when the land is used for agriculture and then abandoned, there were only 5630 kg/ha of mineral matter. Of these, 2380 kg/ha were an integral part of the peat structure. This limits the quantity of plant-available elements in the peat to a mere 3250 kg/ha, to a depth of 80 cm.

Table 3(1267) (from Driessen 1978), shows a decrease in the content of total ash, K2O, P2O5, and SiO2, after forest vegetation has been cut. In contrast, the levels of CaO and MgO often show an increase after forest has been cut, as a result of the management practices. What is more impressive is the influence of cropping on subsidence, and the mineral content of peat. Fig. 2(1120) shows that a reclaimed peat subsided by some 90 cm if cropped only three times in sixteen years, while the same peat subsided by as much as 2 m over the same period if it was cropped every year. Fig. 2(1120) and Table 3(1267) both show that peat used for a crop every year shows a much greater loss of nutrients than peat cropped only a few times with long fallow periods.

The N content of tropical woody peat is higher than that of other nutrients. It is often around 2000 to 4000 kg N/ha in the 0-20 cm layer. However, less than 3% of this is readily available to plants (Driessen 1978). Thus, N application is almost always essential for a successful crop.

As stated above, the nutrient status of peat is generally very poor. Applying fertilizer is inevitable if crops are to be cultivated with any success on reclaimed peat. Driessen's general recommendation for micronutrients (1978) is as follows: 15 kg copper sulfate; 15 kg magnesium sulfate; 15 kg zinc sulfate; 7 kg manganese sulfate; 0.5 kg sodium molybdate and 0.5 kg borax per hectare. For major elements, his recommendation for peat soils is 50-130 kg N, 30-70 kg P2O5 and 60-100 kg K2O per hectare per year.

As peat is generally very acid, with a pH between 3 and 5, liming is also essential in almost all cases. Chew (1971) says that 8-10 mt of ground magnesium limestone per hectare, followed by annual applications of 1 mt/ha, would give a good yield for most short-term crops in acid peat soils.

Rice is considered to be a good crop in peat soils, because it requires prolonged submergence of the land. This slows down the rate of the subsidence and decomposition of organic materials. In fact, many attempts have been made to grow rice on reclaimed peat. So far, paddy rice has not been cultivated successfully in acid tropical peat soils.

The sterility of rice grown in deep peat soils is widely known. According to Driessen and Suhardjo (1976), this is probably because of imperfect photosynthesis, or imperfect carbohydrate translocation, or a disturbance of the generative system of the rice plants, or possibly all three disorders. Stagnant groundwater in deep peat areas has a low pH and low base status, and probably contains a special composition of dissolved organic compounds. These organic compounds have a strongly polyphenolic character and can cause an uncoupling of oxidative phosphorylation (Flaig 1968), a process essential for the formation of starch in plants. Copper (Cu) deficiency apparently retards the deactivation of toxic phenols, and also causes male sterility (cf. Graham 1975). Driessen's conclusion is that the sterility of rice is probably due to inhibited phosphorylation, in the presence of certain phenols, in a copper-deficient environment. Copper that is applied direct to paddy fields is rapidly inactivated, and no practical solution to this problem is known.

A case study of temperate peat soils in Hokkaido, Japan

There are about 200,000 ha of peatland in Hokkaido, Japan. Of those, about one-fourth have been reclaimed for agricultural use since the beginning of the 20th century. Three fundamental measures were taken in reclamation:

  • Drainage;
  • Soil dressing; and,
  • Fertilization.

Drainage was carried out by digging open ditches at intervals of 540 m, plus laying closely aligned lines of drainage tiles (at 12.5 m intervals) under the peat. As drainage proceeded, the peat dried up and decomposed. Considerable land subsidence was inevitable during the course of reclamation. The chemical composition of the peat, originally with a high cellulose and high hemicellulose content, changed to a high lignin content in the course of maturation.

Soil dressing is seen as an essential part of peatland reclamation in Japan. About 1000 to 1500 mt of mineral soil are usually dressed per hectare, which gives a surface mineral soil layer of about 10 cm. The frozen ground and deep snow during the winter in Hokkaido would have facilitated the use of sleighs to do the laborious work of soil dressing. The effect of soil dressing, besides amending the rice sterility, may be summarized as follows:

  • Bulk density is increased, thus it becomes possible to prevent the floating of rice plants due to gas evolution during the summer;
  • Mineral elements are supplied; and,
  • The soil temperature rises more rapidly.

It is necessary to apply fertilizer even after soil dressing. Generally speaking, peat paddy soils are poorer in P and K, and much poorer in silica, than alluvial paddy soils. However, the supply of N from herbaceous peat in Japan is quite substantial, so that less N fertilizer is needed. The application of magnesium oxide is often effective on high-moor peat.

After these treatments, peat soils can be made as productive as ordinary mineral paddy soils. There has never been a rice sterility problem in Hokkaido, and by now, the yield level of peat paddy soils is not inferior to that of mineral paddy soils.

A case study of tropical peat soils in Thailand and Malaysia

Kyuma et al. (1992) conducted a study on tropical peat soils which have been reclaimed for agricultural use. Some of the more relevant data are introduced below (cf., Kyuma 1991).

Contribution of peat decomposition to land subsidence

It is known from long-term observations at the two peat experiment stations at Jalan Kebun in Selangor and Pontian in Johore, Peninsular Malaysia, that after many years of agricultural use, the rate of land subsidence becomes almost constant, at 2.5 cm, per year. In order to estimate the contribution of decomposition of peat itself to the rate of land subsidence, soil respiration was measured in the Pontian area, both in a bare field within the station, and in a nearby swamp forest at Ayer Baloi. The swamp forest was no longer intact, being located in an area with open drainage ditches, but it was still supporting a forest with some 290 mt/ha of above-ground biomass, and included trees 40 m high.

Table 4(1101) gives the data of annual soil respiration at both sites. Assuming that 40% of the total respiration at the swamp forest comes from the respiration of living roots (Kira 1976), the figure for the swamp forest was 14 mt/ha of peat dry matter. In contrast, that for the bare field in the station amounted to 42 mt/ha, three times higher than the swamp forest. Using the bulk density values measured at the plots, the contribution of peat decomposition to the yearly land subsidence rate of 2.5 cm was calculated. The results showed that 70% of the total subsidence was due to the decomposition of the peat itself. Clayton (1928), who studied peat in the Everglades, Florida, USA gave a similar value of two-thirds as the contribution of decomposition to land subsidence. It has been shown that tropical peat decomposes quite rapidly after reclamation and agricultural use. The value of subsidence for the swamp forest in Table 4(1101) is an apparent one, since the formation of new peat should compensate for the loss in the intact forest.

Nutrient release from peat decomposition

The oligotrophic nature of tropical peat is well known. Table 5(1162) gives additional data on this point. Using the data in Table 4(1101) and Table 5(1162), the amount of nutrients released by the annual level of peat decomposition was estimated, and is shown in Table 6(1249). The extremely low level of nutrients is evident. A considerable amount of nitrogen may be supplied by the decomposing peat (to be taken into account in fertilizer application), but P and K levels are extremely low, and hardly enough even for short-term crops.

Micronutrients such as copper (Cu), manganese (Mn) and zinc (Zn) would not become available to a crop, even if a sufficient amount were to be released, as they are chelated by the phenolic components of peat. The high values of iron (Fe), Mn and some other elements in the field at the Pontian station are considered to have an artificial source, such as the laterite paving of the road.

Avoidance of Rice Sterility

Even though rice is the crop best adapted to reclaimed peat, the sterility of rice has severely limited the adoption of rice cultivation on peat soils in the tropics. As reported earlier, reclaimed peat in Japan has been used for rice without a sterility problem, and this is thought to be due to the practice of soil dressing at the time of reclamation. In a tropical peat study project carried out in Malaysia and Thailand, Tadano et al. (1992) and Kamarudin et al. (1992) clarified, both in the field and in pot experiments, that copper (Cu) deficiency and polyphenols are certainly inhibitive to rice, but they are not necessarily the decisive factors. Krisornpornsan et al. (1992) conducted a pot experiment with rice using peat from the Bacho swamp, Narathiwat, Thailand. The result of the experiment, as given in Table 7(1133), endorsed, to some extent, the findings of Tadano et al. (1992). Rice grown on peat alone, and on peat with added copper (Cu), did not develop any grains, but when liming was conducted to correct acidity, the peat yielded some grains even without copper (Cu) application. When lime plus copper (Cu) was applied, the rice yielded a fair amount of grain. When the soil was dressed, either applied to the surface or mixed in with the peat, the grain yield increased. Further addition of copper (Cu) had no effect.

Under field conditions, it was much more difficult to grow rice in peat in Narathiwat Province, Thailand, because, in addition to rice sterility, there was a problem of wide fluctuations in the water supply. Flooding occurred during the rainy months, and drought during the drier months. However, Attanandana et al. (1999) finally confirmed the outcomes of the pot experiment in a field trial, which showed that liming and copper application were essential for harvesting a reasonable yield from rice grown in peat soils.

References

  • Attanandana, T., S. Vacharotayan and K. Kyuma. 1981. Chemical characteristics and fertility status of acid sulfate soils of Thailand. Proc. Of the Bangkok Symp. On Acid Sulfate Soils, pp. 137-156.
  • Attanandana, T., S. Vacharotayan, and K. Kyuma. 1982. Fertility problems of acid sulfate soils of Thailand. First Int'l Symp. on Soil, Geology and Landforms-Impact on Land Use Planning in Develop. Countires, Bangkok, pp. C1-1-C1-9.
  • Attanandana, T., B. Chakranon, K. Kyuma and P. Moncharoen. 1999. Improvement of a peat soil for rice cultivation in Thailand. Jpn. Jour. Trop. Agric. 43: 91-96.
  • Cate, R.B. and A.P. Sukhai. 1964. A study of aluminum in rice soils. Soil Sci. 98: 85-93.
  • Chew, W.Y. 1971. Yield and growth of some leguminous and root crops on acid peat to magnesium lime. Malays. Agric. Jour. 48: 142-158.
  • Clayton, B.S. 1928. Subsidence of Florida peat soil. Trans. Int'l. Soc. Soil Sci., Comm. 6B, Zurich, pp. 840-843 (Cited from Coulter 1957).
  • Coulter, J.K. 1957. Developmetn of the peat soils of Malaya. Malay. Agric. Jour., 33: 63-81.
  • Dent, F.J. 1986. Southeast Asian coastal peats and their use-An overview. In: Classification and Utilization of Peat Land: Proc. Of the 2nd Int'l Soil Management Workshop, Thailand/Malaysia, pp. 27-54.
  • Driessen, P.M. 1978. Peat soils. In: Soils and Rice IRRI (Ed.). IRRI, Los Baòos, Philippines, pp. 763-779.
  • Driessen, P.M. and H. Suhardjo. 1976. On the defective grain formation of sawah rice on peat. Soil Res. Inst. Bull., Bogor, Indonesia 31: 56-73.
  • Flaig, W. 1968. Effect of humic substances on plant metabolism. Proc. 2nd Int'l Peat Congr., Leningrad. (Cited from Driessen 1978).
  • Graham, R.D. 1975. Male sterility in wheat plants deficient in copper. Nature 254: 515. (Cited from Driessen 1978).
  • Kamarudin, A., A.B. Zahari and T. Tadano. 1992. The effect of liming and micronutrient application on the growth of crop plants and occurrence of sterility in a Malaysian peat soil. In: Coastal Lowland Ecosystems in Southern Thailand and Malaysia, Kyuma, K., P. Vijarnsorn and A. Zakaria (Eds.). Kyoto University, Kyoto, Japan, pp. 380-397.
  • Kawaguchi, K. and K. Kyuma. 1969. Lowland rice soils in Thailand. Center for Southeast Asian Studies, Kyoto University, Kyoto.
  • Kira, T. 1976. Terrestrial Ecosystems _ An Introduction. Ecology Course 2. Kyoritsu-Shuppan, Tokyo. (In Japanese).
  • Kobayashi, T. 1939. Studies on amelioration of problem soils in polder lands. Spec. Rep. Ibaraki Agric. Expt. Sta. 3: 1-47. (In Japanese).
  • Komes, A. 1973. The Reclamation of Some Problem Soils in Thailand. Soils of the ASPAC Region; Part 5. Thailand. Tech. Bull. No. 14, Food and Fertilizer Technology Center, Taipei, Taiwan ROC.
  • Krisornpornsan, B., T. Attanandana and K. Kyuma. 1992. Effect of soil dressing on rice in tropical peat soils. In: Coastal Lowland Ecosystems in Southern Thailand and Malaysia, Kyuma, K., P. Vijarnsorn and A. Zakaria (Eds.). Kyoto University, Kyoto, Japan, pp. 312-317.
  • Kyuma, K. 1991. Problems related to reclamation and development of swampy lowlands in the tropics. Jpn. J. Soil Physic. Cond. Plant Growth, 63:43-49. (In Japanese).
  • Kyuma, K., P. Vijarnsorn and A. Zakaria (Eds.). 1992. Coastal Lowland Ecosystems in Southern Thailand and Malaysia, Kyoto University, Kyoto, Japan. 416 pp.
  • Matsuguchi, T., B. Tangcham and S. Pakiyuth. 1970. Nitrogen-fixing microflora and its activity in paddy soil of Thailand. Proc., 1st Asean Soil Conf., Bangkok, Thailand.
  • Murakami, H. 1965. Studies on the Characteristics of Acid Sulfate Soils and their Amelioration. Doctoral Dissertation, Kyoto University. (In Japanese).
  • Ponnamperuma, F.N., T. Attanandana, and G. Beye. 1973. Amelioration of three acid sulfate soils for lowland rice. Proc. Int'l Symp. Acid Sulfate Soils, Vol. II: 391-406. ILRI Publication No. 18. Wageningen, Netherlands.
  • Stumm, W. and J.J. Morgan. 1970. Aquatic Chemistry: An Introduction, Emphasizing Chemical Equilibria in Natural Waters. John Wiley & Sons, N.Y.
  • Tadano, T., K. Yonebayashi and N. Saito. 1992. Effect of phenolic acids on the growth and occurrence of sterility in crop plants. In: Coastal Lowland Ecosystems in Southern Thailand and Malaysia, Kyuma, K., P. Vijarnsorn and A. Zakaria (Eds.). Kyoto University, Japan, pp. 358-369.
  • Tanaka, A. and S. Yoshida. 1970. Nutritional Disorders of the Rice Plant in Asia. IRRI, Tech. Bull. No. 10. IRRI, Manila, Philippines.
  • Van Breemen, N. 1975. Acidification and deacidification of coastal plain soils as a result of periodic flooding. Soil Sci. Soc. Amer. Proc. 39: 1153-1157.
  • Van Breemen, N. and L.J. Pons. 1978. Acid sulfate soils and rice. In: Soils and Rice, IRRI (Ed.). IRRI, Los Baòos, Philippines, pp. 739-761
  • Van Breemen, N. and W.G. Wielemaker. 1974. Buffer intensities and equilibrium pH of minerals and soils. II. Theoretical and actual pH of minerals and soils. Soil Sci. Soc. Amer. Proc., 38: 61-66.
  • Yamada, I. 1986. Vegetation in swampy lowlands in Southeast Asia. In: Swampy Lowlands in Southeast Asia, Tropical Agriculture Research Center, (K. Kyuma, (Ed.)), Norin-Tokei-Kyokai, Tokyo, pp. 104-196. (In Japanese).

Index of Images

Table 1 Chemical Process by Which Sulfidic Compounds Are Formed.

Table 1 Chemical Process by Which Sulfidic Compounds Are Formed.

Figure 1 Model for the Oxidation of Pyrite

Figure 1 Model for the Oxidation of Pyrite

Figure 2 Distribution of Mineral Constituents in a Number of Representative Virgin and Reclaimed Ombrogenous Dome Peats from West Kalimantan, Indonesia

Figure 2 Distribution of Mineral Constituents in a Number of Representative Virgin and Reclaimed Ombrogenous Dome Peats from West Kalimantan, Indonesia

Table 2 Fertility Characteristics of Some Representative Acid Sulfate Soils and Non-Acid Sulfate Soils in the Bangkok Plain

Table 2 Fertility Characteristics of Some Representative Acid Sulfate Soils and Non-Acid Sulfate Soils in the Bangkok Plain

Table 3 Total Nutrient Contents of Six Surface Soils from Lowland Peat Areas in Sumatra and Kalimantan

Table 3 Total Nutrient Contents of Six Surface Soils from Lowland Peat Areas in Sumatra and Kalimantan

Table 4 Rates of Soil Respiration, Peat Decomposition and Land Subsidence Due to Peat Decomposition

Table 4 Rates of Soil Respiration, Peat Decomposition and Land Subsidence Due to Peat Decomposition

Table 5 Nutrient Content of Some Tropical Peat Soils

Table 5 Nutrient Content of Some Tropical Peat Soils

Table 6 Rate of Nutrient Release from Peat Decomposition (KG/Ha/Year)

Table 6 Rate of Nutrient Release from Peat Decomposition (KG/Ha/Year)

Table 7 Effect of Soil Dressing, Liming and Copper Application on Rice in Pot Experiment

Table 7 Effect of Soil Dressing, Liming and Copper Application on Rice in Pot Experiment

Download the PDF. of this document(602), 81,137 bytes (79.2 KB).