Literature Review On Soil Erosion

The Latin word erodere, (meaning “to gnaw away”) is the origin of the word erosion (Roose, 1996). Soil Erosion is the physical removal of topsoil by various agents, including falling raindrops, water flowing over the soil profile and gravitational pull (Lal 1990). The Soil Science Society of America defines erosion as ” the wearing away of the land surface by running water, wind, ice or other geological agents, including such processes as gravitational creep” (SCSA, 1982). Physical erosion involves the detachment and transportation of insoluble soil particles (sand, silt and organic matter). Removal of soluble material as dissolved substances is called chemical erosion and this maybe caused by surface runoff or subsurface flow where the water moves from one layer to another within the soil profile (Lal 1990).

According to ASCE, 1975, the physical processes in soil erosion include detachment of soil particles, their transportation and subsequent deposition of soil sediments downslope by raindrop impact and runoff over the soil surface. Rainfall is the most important detaching agent (Morgan and Davidson 1986; Lal, 1990) followed by overland flow in entraining soil particles (Lal 1990).

The process of soil erosion occurs in three main steps, detachment of soil particles, transportation and deposition of soil particles downslope by raindrop impact and runoff over the soil surface (ASCE 1975; Morgan and Davidson, 1986, Lal 1990) followed by overland flow in entraining soil particles (Lal, 1990). Soil erosion reduces soil productivity by physical loss of topsoil, reduction in rooting depth and loss of water. In contrast soil, soil depletion means loss or decline of soil fertility due to crop removal or removal of nutrients by eluviations from water passing through the soil profile (Lal, 1990). Sedimentation however, causes off site effects like degradation of basins, accumulation of silts in water reservoirs and burial of low-lying productive areas and other problems (Lal, 1990). Sediments is the main cause of pollution and eutrophication (Lal, 1990). According to Lal 1990, soil degradation may be caused by accelerated soil erosion, depletion through intensive land use, deterioration in soil structure, changes in soil pH, leaching, salt accumulation, build up of toxic elelments such as aluminum or zinc, excessive inundation leading to reduced soil conditions and poor aeration.

Soil Erosion is the most serious and least reversible form of land degradation (Lal, 1977; El-Swaify, Dangler and Amstrong, 1982). Soil erosion and soil loss , according to Lal (1990) have adverse effects on agriculture because they deplete the soil’s productivity and diminish the resourse base.

2.2 Soil Erosion Process

Geologic erosion can be caused by a number of natural agents including rainfall, flowing water and ice, wind and the the mass movement of soil bodies under the action of gravity which cause the loosened or dissolved earthy and rock materials to be removed from a place and eventually deposited to a new location (Lal,1990; Morgan and Davidson, 1986). The Soil Science Society of America (SCSA, 1982) described geologic erosion as ” the normal or natural erosion caused by geologic processes acting over long periods and resulting in the wearing away of mountains, the building up of flood plains, coastal plains. Etc.” The slow and constructive natural soil erosion process has been significantly accelerated by human activities of poor farming practices, overgrazing, ground clearing for construction, logging and mining (Lo, 1990). Accelerated erosion not only affects the soil but also the environment and is the primary cause of soil degradation (Lal, 1990). Agriculture has been identified as the primary cause of accelerated soil erosion (Pimentel, 1976).

2.3 Soil Characteristics in the Tropics

Extremes of climate and wide variety of parent materials cause great contrast of soil properties in the tropics from soils in other temperate regions. In the tropics soils are highly variable and diverse like the vegetation (Sanchez and Buoi, 1975; Van Wambeke, 1992). The main soil types are alfisols, oxisols, ultisols and inceptisols (El-Swaify, 1990). Tropical soils low in weatherable minerals and basic cations (sodium, calcium, magnesium, and potassium) resulted from continuous weathering of parent materials (Lo, 1990). The ability of these soils to keep plant nutrients is largely dependent on the humus content found in plant biomass and the organic matter (Rose,1993). The inactivity of soil mineral constituents (kaolin and sesquioxides) in these soils, causes deficiency in crop nutrients, lowers the capacity to retain basic cations, limits active relationship with organic matter and excessively immobilizes phosphates and related anions, a condition which are highly toxic to plant roots (Lo, 1990). Crop production in tropical soils are constrained by primarily aluminum- derived soil acidity and infertility but generally their physical properties are favourable (El-Swaify, 1990). Tropic soils have moderate to high permeability under natural conditions, but susceptible to slaking and development of impermeable crust upon action of raindrops and as a result runoff increases with continuous cultivation (Lal, 1982). This crusting cause insignificant reduction of filtration rate, increasing water runoff which leads to acceleration of soil erosion (Falayl and Lal, 1979).

It is important to note however that heavy and intense rains cause severe erosion in the tropics (Morgan, 1974; Wilkinson 1975; Amezquita and Forsythe, 1975; Lal 1976; Aina, Lal and Taylor, 1977; Bois, 1978; Sheng 1982).

2.4 Soil Erosion on Steep Slope

According to Lal 1990, Steeplands refer to lands with a slope gradient greater than 20%. It is important to note however that flat undulating lands have a great potential for crop production and agricultural development. Due to the possibility of soil erosion and the problem of mechanization, the steep areas are considered marginal for agriculture production (Lal, 1990).

The difficult topography in steepland agriculture restricts mechanizations of operations thus, reducing all agricultural activities (land preparation, cultivation and harvesting), limiting the farmer in scale and efficiency. Inputs such as fertilizer and pesticides have to be carried manually by the farmer. As a resulted they are used scarcely. Observably any increase in the use of these agricultural inputs will result in decline in he farmers profits from the generally lower agricultural field (Benvenuti, 1988). For all these reasons steepland farmers tend to concentrate in high value crop production of limited scale (Ahmad, 1987; Ahmad 1990). It is important to note however that farmers prefer steepslopes due to cultural hand cultivation, planting and harvesting can be done in an upright fashion (Williams and Walter, 1988). Futher more subsistence farmers are found on steep slopes because of more favourable environmental conditions such as lower temperatures, reduced diseases and higher reliability of rainfall. (Hurni, 1988).

In the tropics, removal of forest vegetation causes excessive leaching and accelerated soil nutrient loss. Being highly weathered soil types , their contained minerals generally have poor ability to retain sorbed nutrients against leaching. Clay soils with high residualmiron contents are considered superior in resistance to runoff caused soil erosion; thus, soils emanated from basic igneous rocks and red soils developed from calcareous rocks are strongly aggregated due to the cementing property of iron oxides, hence, soil erosion is expected to be less than for most other soils. Also soils developed from fragmentary volcanic materials with andic properties are resistant to soil erosion (Sheng, 1986; Ahmad, 1987; Ahmad, 1990; Lal, 1990). Soils formed from shales, schists, phyillites and sandstones are considered highly erodible. Soils produced from these rocks are high in both sand or silt fraction, and clay minerals and iron oxides are generally insufficient as cementing agents for a stable-structured soil. These parent materials are generally rich in muscovite occurring in all soil particle-size fractions. Micah-rich soils are weak-structured, and thus raindrops can easily dislodged the weak aggregates, while the clay fraction dispersed in water. The resulting mica flakes settling on their flat axes in the water film on the soil surface causes soil crusting. The formation of soil crusts further restricts water entry into the soil (Ahmad and Robin, 1971; Sumner, 1995), resulting to disposal of a much greater volume of runoff water, a condition which leads to further disintegration of soil aggregates and transport of colloidal soil material (Ahmad, 1987; Ahmad 1990). Soil crust restricts gaseous exchange leading to anaerobic soil conditions, denitrification, toxic effects due to ethylene production, and mechanical impedance to seedling emergence (Ahmad 1987; Ahmad, 1990).

Steep slope cultivation can cause certain instability in the ecological system with both onsite and offsite detrimental impacts (El-Swaify, Garnier and Lo, 1987). Soil, climate, land use and farming systems affect the extent and the degree of severity of soil erosion. However, regardless of soil and climatic conditions, intensively used steeplands in densely populated regions experience severe soil erosion problem.

Land use influences the degree of severity of soil erosion on steeplands. Uncontrollable grazing or over grazing, exensive and abusive cultivation, diversified cropping are responsible for severe soil erosion in unprotected arable lands (Roose, 1988; Liao et al 1988). Ahmad (1987;1990) reportd soil loss of approximately 120 t0 180 tonnes per hectare in Tobago Trinidad. In Australia, annual soil loss of 200 t/ha to 328 t/ha has ben reported from sloping sugar cane plantations in central and north Queensland (Sallaway, 1979; Mathews and Makepeace 1981).

There are two types of soil erosion associated with the Caribbean region, land slipping and gullying. Land slipping is a manifestation of mass movement associated with steepland agriculture and the severity being strongly influenced by the parent materials. Land clearing (example deforestation) and crop production can influence land slipping particularly in the early portion of the wet season when the cleared soil wets faster due to saturation of the soil above rock. Serious dislocations, crop loss and destruction of any mechanical anti erosion devices can result from this form of mass movements. Due to drastic changes in hydrological conditions experienced by land naturally prone already to slipping and cleared for agriculture for the first time land slippage would be of common experience (Ahmad 1987; Ahmad 1990).

Gullying is another common form of soil erosion that occurs on steep land bcause of the terrain involved. This is more common on sandy soils, volcanic soils and vertisols, which are all porous materials. Soils easily attain saturated conditions upon the rapid entry of water, consequently breaking the material and ultimately, leading to the formation of gullies. Agricultural activities enables this soil erosion in steeplands by allowing rapid soil wetting upon the start of the wet season. Farming activities though unsuitably oriented field boundaries, foot tracks and the lack of provision for disposal of surface water are some main causes of gullying, even on soils not prone to this tpe of steepland soil erosion (Ahmad 1987;Ahmad 1990).

Since steeplands are traditionally considered marginal for agricultural crop production, most research on soil erosion and soil conservation has been done on either flat land or ‘ rolling land with a maximum slope of about 20%'(Lal, 1988).

2.5 Factors Affecting Soil Erosion

The causes of soil erosion have been intensively discussed during the past 40 years. Soil erosion is a natural process that is enhanced by human activity (Richter, 1998) and occurs in all landscapes and under different land uses. In addition to human activities, soil erosion processes are also caused by morphometric characteristics of the land surface, the erosive forces of rainfall and the erodibility of soils and soil surfaces.

When rainwater reaches the soil surface it will either enter the soil or run off. Runoff occurs when the rainfall intensity exceeds the infiltration capacity of the soil. Water erosion is the result of the dispersion action of rain drops, the transporting power of water and also the vulnerability of the soil to dispersion and movement (Baver and Gardner, 1972). The effects of soil erosion is also classified: definition of gullies and explanation of gully development is given by Morgan (1996), as well as Hudson (1995) who additionally focuses on individual cases of the development of gullies. Toy et al (2002) give detailed definitions of soil erosion features and processes such as sheet erosion and inter-rill erosion, rill erosion, as well as ephemeral and permanent gully erosion.

Rill erodibility depends both directly and indirectly on soil properties such as bulk density, organic carbon and clay content, clay mineralogy, cations in the exchange complex, soil pH and experimental conditions such as moisture content, aging of prewetted soil and quality of eroding water (Rapp,1998). Govers (1990) found that runoff erosion resisitance of a loamy material was extremely sensitive to variation in the initial moisture content and to a lesser extent to changes in bulk density.

The process of water erosion can be separated into two components, rill and interrill erosion (Young and Onstad, 1978). Interrill erosion (sheet erosion) is mainly caused by raindrop impact and removes soil in a thin almost imperceptible layer (Foster, 1989). In interril erosion the flow of water is generally unconfined, except between soil clods and covers much of the soil surface. As the velocity of flow increases the water incises into the soil and rills forms (Evans,1980).

Rill erosion begins when the eroding capacity of the flow at some point exceeds the ability of the soil particles to resistant detachment by flow (Meyer cited by Rapp, 1998). Soil is detached by headcut advance from knickpoints (De Ploey, 1989; Bryan, 1990), rill slide sloughing and hydraulic shear stress (Foster cited by Rapp, 1998) as well as by slumping by undercutting of side walls and scour hole formation (Van Liew and Saxton, 1983). These processes are usually combined into a detachment prediction equation as a function of average shear stress (Foster cited by Rapp, 1998). When the rills develop in the landscape, a three to five fold increase in the soil loss commonly occurs (Moss, Green and Hutka 1982 and Meyer and Harmon 1984).

2.5.1 Vegetative Factors

The effects of vegetation can be classified into three catergories:

The interception of raindrops by the canopy (D’Huyvetter, 1985). Two effects are associated with this. Firstly, part of the intercepted water will evaporate from the leaves and stems and thus reduce runoff. Secondly, when raindrops strike the vegetation, the energy of the drops is dissipated and there is no direct impact on the soil surface. The interception percentage depends on the type of crop, the growth stage and the number of plants per unit area.

A well distributed, close growing surface vegetative cover will slow down the rate at which water flows down the slope and will also reduce concentration of water (D’Huyvetter, 1985). As a result, it will decrease the erosive action of running water.

There is also the effect of roots and biological activity on the formation of stable aggregrates, which results in a stable soil structure and increased infiltration that reduces runoff and decreases erosion (D’Huyvetter, 1985). Increased permeability also reduces erosion as a result of in increased water percolation due to better drainage. Stables aggregrates in the topsoil also counteract crusting.

2.5.2 Rainfall Factors

Raindrop size, shape, duration of a storm and wind speed interactions controls the erosive power of rainfall (D’Huyvetter, 1985). The erosivity of rainfall is expressed in terms of kinetic energy and is affected by various factors.

According to Wischmeier and Smith (1965), the intensity of rainfall is closely related tot e kinetic energy, according to the regression equation

E = 1.213 + 0.890 log I

Where

E = the kinetic energy (kg.m/m2.mm)

I = rainfall intensity (mm/h)

Raindrop size, distribution and shape all influence the energy momentum of a rainstorm. Laws and Parson (1943) reported an increase in median drop size with increase in rain intensity. The relationship between mean drop size (D50) and rainfall is given by:

D50:2.23 I 0.182 (inch per hour).

The median size of rain drops increases with low and medium intensity fall, but declines slightly for high intensity rainfall (Gerrard, 1981). The kinetic energy of an rainfall event is also related to the velocity of the raindrops at the time of impact with the soil (D’Huyvetter, 1985). The distance through which the rain drop must fall to maintain terminal velocity is a function of drop size. The kinetic energy of a rainstorm is related to the terminal velocity according to the equation:

Ek = IV2/2

Where Ek = energy of the rain storm

I = Intensity

V= Velocity of raindrop before impact

Ellison (1945) developed an equation showing that the relationship between the soil detached, terminal velocity, drop diameter and rainfall intensity:

E = KV4.33 d1.07 I0.63

Where E = relative amount of soil detached

K = soil constant

V = velocity of raindrops (ft/sec)

d = diameter of raindrops (mm)

I = rainfall intensity

2.5.2.1 Effect of rainfall intensity on runoff and soil loss

According to Morgan (1995), soil loss is closely related to rainfall partly through the detaching power of raindrops striking the soil surface and the contribution of rain to runoff. If rainfall intensity is less than the infiltration capacity of the soil, no surface runoff occurs and the infiltration rate would equal the rainfall intensity (Horton, 1945) as sited by Morgan (1995). If the rainfall intensity exceeds the infiltration capacity, the infiltration rate equals the infiltration capacity and the excess rainfall forms surface runoff.

According to Morgan (1995), when the soil is unsaturated, the soil matric potential is negative and water is held in the capillaries due to matrics suction. For this reason, under saturated conditions sands may produce runoff very quickly although their infiltration capacity is not exceeded by the rainfall intensity. Intensity partially controls hydraulic conductivity, increasing the rainfall intensity may cause conductivity to rise so that although runoff may have formed rapidly at relatively low rainfall intensity, higher rainfall intensities do not always produce greater runoff (Morgan, 1995). This mechanism explains the reason why infiltration rates sometimes increase with rainfall intensities (Nassif and Wilson, 1975).

2.5.3 Soil Factors

According to Baver et al, (1972), the effect of soil properties on water erosion can be in two ways : Firstly, certain properties determine the rate at which rainfall enters the soil. Secondly, some properties affect the resistance of the soil against dispersion and erosion during rainfall and runoff.

The particle size distribution is an important soil property with regards to erodibility. Generally it is found that erodible soils have a low clay content (D’Huyvetter, 1985). Soils with more than 35% clay are often regarded as being cohesive and having stable aggregates which are resistant to dispersion by raindrops (Evans, 1980). Evans also stated that sands and coarse loamy sands are not easily eroded by water due to its high infiltration rate. In contrast soils with a high silt or fine sand fraction are very erodible.

Erodibility of soil increases with the proportion of aggregates less than 0.5mm (Bryan, 1974). Factors which contribute to aggregate stability include organic matter content, root secretions, mucilaginous gels formed by break down of organic matter, the binding of particles by sesquioxides and the presence of a high Ca concentration on the exchange sites of the colloids instead of a high sodium content (D’Huyvetter, 1985).

The depth of erosion is determined by the soil profile (Evans, 1980). According to Evans soil horizons below the A horizon or plough layer are often more compact and less erodible. The texture and chemical composition of the sub surface horizon can also have an adverse effect. Normally deep gullies can be cut if the parent material is unconsolidated. If resistant bedrock is near the surface only rills will develop. Soil rich in surface stones are less susceptible to erosion (Lamb, 1950 and Evans, 1980). Stones protect the soil against erosion and also increase the infiltration of the flowing water into the soil.

The antecedent soil moisture and the surface roughness are both regarded by Evans (1980) as important soil factors affecting erosion. The ability of a soil to accept rainfall depends on the moisture content at the time of the rainfall event.

2.5.3.1 Factors affecting aggregate stability

Soil structure is determined by the shape and size distribution of aggregates. Aggregrate size and strengthe determine the physical properties of a soil and its susceptibility to breakdown due to water forces. Their stability will have a decisive effect on soil physical properties (Lynch and Bragg, 1985). The main binding materials giving stable aggregates in air dry state are the glueing agents in organic matter (Chaney and Swift, 1984; Tisdale and Oades, 1982) and sesquioxides (Goldberg and Glaubic, 1987).

2.5.3.1.1 Aluminium and Iron Oxides

The soil used by Kemper and Koch (1966) contained relatively little free iron, although it did contribute to aggregrate stability. Their data show a sharp increase of free iron from 1 to 3%. Goldberg and Glaubic (1987) concluded that Al-oxides were more effective than Fe-oxides in stabilizing soil structure. Al-oxides have a greater proportion of sub-micrometer size particles in a sheet form as opposed to the spherical form of Fe-particles.

Shainberg, Singer and Janitzky (1987) compared the effect of aluminium and iron oxides on the hydraulic conductivity of a sandy soil.

2.5.3.1.2 Organic Matter

Organic matter can bind soil particles together into stable soil aggregates. The stabilizing effect of organic matter is well documented. Little detailed information is available on the organic matter content required to sufficiently strengthen aggregates with ESP values greater than 5 or 7, and containing illite or montmorrillionite, so as to prevent their dispersion in water (Smith, 1990). High humus content makes the soil less susceptible to the unfavourable influence of sodium (Van den Berg, De Boer, Van der Malen, Verhoeven, Westerhof and Zuur, 1953). Kemper and Koch (1966) also found that aggregate stability increased with an increase in the organic matter content of soils. A maximum increase of aggregate stability was found with up to 2% organic matter, after which aggregate stability increased very little with further increases in organic matter content.

2.5.3 Slope Factors

Slope characteristics are important in determining the amount of runoff and erosion ( D’Huyvetter, 1985). As slope gradient increases, runoff and erosion usually increases (Stern, 1990). At low slopes due to the low overland flow velocities, detachment of soil particles from the soil surface into the water layer is due to detachment alone (Stern, 1990). Additionally, at low slope gradients, particles are splashed into the air in random directions unlike the case with steeply sloping land where down slope splash occurs (Watson and Laflen, 1985).

As slope gradient increases, the ability for surface runoff to entrain and transport sediments increases rapidly until the entrainment by the surface runoff becomes dominant contributing to sediment transport (Stern, 1990). Foster , Meyer and Onstad (1976) presented a conceptual model that showed that at lower slopes, interill transport determined erosion, while at steeper slopes, raindrop detachment determined it. Th uniform bed characteristics of sheet flow transport tend to be replaced by channels because of instability and turbulent flow effects (Moss, Green and Hutka, 1982).

There are many empirical relationships relating soil transport by surface wash to slope length and slope gradient. Zingg (1940) showed that erosion varied according to the equation:

S = X1.6 tanB1.4

Where S = soil transport cm/yr

X = slope length (m)

B = slope gradient (%)

Studies conducted by Gerrard (1981), showed that plane and convex slopes did not differ significantly in the amount of soil lost by surface runoff, but concave slopes were less eroded.

Some researchers such as Zingg (1940) and Mc Cool et al (1987) indicated that soil erosion increases exponentially with increase in slope gradient. The relationship is indicated after Zing (1940) by: E = aSb where E is the soil erosion, S is the slope gradient (%) and a and b are empirical constants. The value of b ranges from 1.35 to 2.0. The other relationship between erosion and slope gradient for inter-rill erosion is given by Mc Cool et al (1987)

E = a sin b Q+C

Q is the slope angle in degrees

A,b and C are empirical constants.

However, even if the effect of slope gradient on erosion is well recognized, several studies indicate that the power relationship between slope gradient and soil loss over predicts interrill erosion rate by as much as two or more times (Torri, 1996;Fox and Bryan, 1999), and the relationship is better described as linear.

2.8 Soil Erosion Impacts

2.8.1 Soil Physical Properties

Progressive soil erosion increases the magnitude of soil related constraints for crop production. These constraints can be physical, chemical and biological. The important physical constraints caused by erosion are reduced rooting depth, loss of soil water storing capacity (Schertz et al 1984; Sertsu, 2000), crusting and soil compaction and hardening of plinthite (Lal, 1988). Erosion also results in the loss of clay colloids due to preferential removal of fine particles from the soil surface (Fullen and Brandsma, 1995). The loss of clay influences soil tilth and consistency. Exposed subsoil is often of massive structure and harder consistency than the aggregated surface soil (Lal, 1988).

Development of rills and gullies may change the micro-relief that may make use of farming machinery difficult. Another effect of erosion is that the manangement and timing of farm operations.

2.8.2 Soil Chemical Properties

Soil erosion reduces the fertility status of soils (Morgan, 1986; Williams et al., 1990). Soil chemical constraints and nutritional problems related to soil erosion include low CEC, low plant nutrients (NPK) and trace elements (Lal, 1988; Fullen and Brandsma, 1995). Massy et al (1953) reported an average loss of 192 kg of organic matter, 10.6 kg of N and 1.8kg per ha on a Winsconsin soils with 11% slope. Sharpley and Smith (1990) reported that the mean annual loss of total P in runoff from P fertilized watersheds is equivalent to an average of 15%, 12% and 32% of the annual fertilizer P applied to wheat, mixed crop grass and peanut – sorghum rotation practices respectively. Researchers (Massy et al 1953; Lal, 1975) have also reported extensive loss of N in eroded sediments.

2.8.3 Productivity

Quantifying the effects on crop yields is a difficult task. It involves the evaluation of interactions between soil properties, crop characteristics and climate. The effects are also cumulative and not observed until long after accelerated erosion begins. The degree of soil erosion’s effects on crop yield depends on soil profile characteristics and management systems. It is difficult to establish a direct relationship between rates of soil erosion and erosion induced soil degradation on the one hand and crop yield on the other (Lal, 1988).

It is well known that soil erosion can reduce crop yields through loss of nutrients, structural degradation and reduce of depth and water holding capacity (Timilin et al, 1986; Lal,1988). Loss of production in eroded soil further degrades its productivity which in turn accelerates soil erosion. The cumulative effect observed over a long period of time may lead to irreversible loss of productivity in shallow soils with hardened plinthite or in soils that respond to expensive management and additional inputs (Lal,1988).

2.8.4 Off Site Effects of Soil Erosion.

Effects of erosion include siltation of rivers, crop failure at low lying areas due to flooding, pollution of waterbodies due to the various chemicals brought by the runoff from different areas. Several studies reported the significance of the off site effects of soil erosion on land degradation (eg. Wall and ven Den,1987; Lo, 1990; Robertson and Colletti, 1994; Petkovic et al, 1999)

Rainwater washes away materials that originate from fertilizers and various biocides (fungicides, insecticides, herbicides and pesticides) which are applied in large concentrations. They reappear in greatr quantities in the hydrosphere polluting and contaminating the water environment (Zachar,1982;Withers, and Lord, 2002; Verstraeten and Poesen, 2002). Chemical pollution of water mainly by organic matter from farm fields causes rapid eutrophication in waterways (Zachar, 1982;Zakova et al, 1993; Lijklema, 1995).

2.8.5 Soil Erosion Models

Modelling soil erosion is the process of mathematically describing soil particle detachment, transport and deposition on land surfaces (Nearing et al, 1994). Erosion models are used as predictive tools for assessing soil loss and project planning. They can also be used for understanding erosion processes and their impacts (Nearing et al 1994). There are three main types of models, empirical or statistical models, conceptual models and physically based models (Morgan 1995, Nearing et al 1994, Merritt et al 2003). It is important to note however that there is no sharp difference among them.

2.8.5.1 Physically Based Models

These models are based on solving fundamental physical equations describing stream flow and sediment and associated nutrient generations in a specific catchment (Merritt et al ., 2003). They are developed to predict the spatial distribution of runoff and sediment over land surfaces during individual storms in addition to total runoff and soil loss (Morgan, 1995). Physically based models are also called process based models (Morgan, 1995) as they rely on empirical equations to determine erosion processes. These models use a particular differential equation known as the continuity equation which is a statement of conservation of matter as it moves through space over time. The common physically based models used in water quality studies and erosion include : The Areal Non-Point Source Watershed Environment Response Simulation (ANSWERS) (Beasley et al., 1980), Chemical Runoff and Erosion from Agricultural Management Systems (CREAMS) (Knisel, 1980), Griffith University Erosion System Template (GUEST) (Misra and Rose, 1996), European Soil Erosion Model (EUROSEM) (Morgan, 1998), Productivity, Erosion and Runoff, Functions to Evaluate Conservation Techniques (PERFECT) (Littleboy et al., 1992) and Water Erosion Prediction Project (WEPP) (Laflen et al., 1991).

2.8.5.2 Empirical Models

These models are based primarily on observations and are usually statistical in nature. They are based on inductive logic, and generally are applicable only to those conditions for which the parameters have been calibrated (Nearing et al., 1994, Merritt et al., 2003). The main focus of these models have been in predicting average soil loss although some extensions to sediment yield have been developed (Williams, 1975 as quoted by Nearing et al.,1994). Empirical models are generally based on the assumption that the underlying conditions remain unchanged for the duration of the study period. They are not event responsive and ignore the process of rainfall – runoff in the areas being modeled. Empirical models are frequently used in preference to the more complex models and are particularly useful as first step in identifying sources of sediment and nutrient generations (Merritt et al.,2003). Among the commonly used models are: The Universal Soil Loss Equation (USLE) (Wischmeier and Smith, 1978), Revised Universal Soil Loss Equation (RUSLE) (Renard et al., 1994) and the Soil Loss Estimation Model for Southern Africa (SLEMSA)(Etwell, 1978).

2.8.5.3 Conceptual Models

These models are based on on spatially lumped forms of water and sediment continuity equations (Lane et al., 1988 in Nearing et al., 1994). They intend to include a general description of catchment processes, without including the specific details of process interactions which would require detailed catchment information (Merritt et al., 2003). These models can provide an indication of the qualitative and quantitative effects of land use changes, without requiring large amounts of spatially and temporally distributed data. The main feature that distinguishes these conceptual models from empirical models is that the conceptual model, whilst they tend to be aggregated, they still reflect the hypothesis about the processes governing the system behaviours (Merritt et al.,2003). The Agricultural Non-Point Source Model (AGNPS) (Young et al., 1989), Agricultural Catchment Research Unit (ACRU) (Schulze, 1995), Hydrologic Simulation Program Fortran (HSPF) (Walton and Hunter, 1996), and Simuator for Water Resources in Rural Basins (SWRRB) (Arnold et al., 1990) are among the models (Merritt et al., 2003) used in erosion and water quality studies.

2.7 Soil Erosion in the Caribbean

Soil Erosion in the caribbean in mainly affected by two types of factors, climatic factors and topographic factors. It can be seen that the soils of the islands of the west indies should be subject to a great deal of erosion by water. The amount of soil erosion occurring in the Caribbean has not been quantitively determined. The severity of the erosion depends on topography, rainfall, natural vegetation, erodibility of the soils, land use and soil management.

According to Breckner 1971, topographic effects are as extreme as the climatic effects. Several Caribbean islands are characterized by steep slopes with a high percentage (58%) of the land area having slopes greater than 30 degrees. Many slopes are greater than 45 degrees and farming is practiced on these slopes (Gumbs 1997).

2.7.1 Trinidad

Erosion studies on a range of soil types in many tropical countries have shown that soil losses can be extensive (Suarez De Castro 1951, 1952; Smith and Abruna 1985; Sheng and Michaelsen 1973; Lal 1976). The earliest comprehensive report on soil erosion in Trinidad was done by Hardy (1942). He reported considerable gulling and sheet erosion in the foothills of the northern range particularly in the western part where intensive cultivation is carried out. Burning to clear the land for wet season and landslips are common in some places where the soil is within six inches of parent rock. Also according to Hardy sheet erosion is an important in the Caroni plain. He suggested that the soil on the sides of the highly cambered beds of the sugar cane areas is directly exposed to the rains and is thus just as prone to erosion as exposed soil on steep hills of the northern range.

In Las Lomas the sandy soils are described as being very erodible. Since much of the land is a forest reserve, erosion is not a severe problem. In the central range land creep is a major problem and the sothern slopes show many land slip scars. Chenery (1952) mentioned that Brasso clay, the most wide spread soil of the central range is very eroded due to prolonged cultivation. Both Hardy and Chenery commented on the severe erosion of the marl soild and the associated red soils of the Naparima district in southern Trinidad, with caps of exposed white marl being a common feature of the hills of the region.

Alleyne and Percy (1966) measured the soil loss from the major soil type (Maracas clay loam – orhoxic tropudult) in the northern range under pineapple (Ananas comosus) with 50% of the area terraced and pangola grass (Digitaria decumbens). Under both types of vegetation the runoff was less than 10% of rainfall and the soil losses were both very slight ( < 0.4 and 0.05 tonnes per hectare during the wet season). Lindsay and Gumbs (1982) have shown that this soil type is only slightly erosible but the large amounts of soil can be lost from the bare soil (Gumbs and Lindsay 1982).

Report on losses of nitrogen by erosion either in runoff or eroded sediments are limited. Neal (1944) considered that the majority of nitrogen loss by erosion occurs in the organic fraction of the soil as water easily loosens and floats away organic matter. Lal (1976) found there was a tendency for greater losses of inorganic nitrogen in the erosed sediments than in the runoff water.

2.7.2 Tobago:

Information on the erosion situation in Tobago are Brown et al (1965), Hardy (1942) and Breckner (1971). All the mountainous areas of Tobago is highly prone to erosion with soils formed from such highly erosive parent materials as diorite and schists predominating. Volacanic tuffs and breccias make up the third major type of prent rock. In the south of the island erosion becomes a more serious problem. Hardy has described heavy erosion in the Castara – Parlat area on the leeward side of the island.

On the winward side of the island, large patches of land on the volcanic soils are still being cleared, mainly by burning and planting with arable crops such as corn, ground provisions and tomatoes. The worst eroded area is in Mason Hall. – Les Coteaux district. The sandy clay loam soils of this area formed from dioritein very irregular, steeply sloping topography appear to be highly erodible. This is due to the main area experiencing peasant farming which involves intensive cultivation.

It is important to note however that brown et al estimated that between 1956 and 1965 1500 acres of land have been treated by soil conservation measures under the subsidies scheme. The authour saw no evidence of treated land. It was observed that soil conservation measures were practicised.

Limited soil conservation is practiced mainly in the form of intercropping and occasionally trash mulching (Gumbs 1997).

2.7.3 Antigua:

The low rainfall and level topography of much of both Antigua and Barbuda has meant that not as much erosion has occurred as in some other parts of the West Indies (Hill 1964; Vernon and Lang 1964). However much accelerated erosion has occurred in the hilly areas of Antigua. Cotton was grown extensively and requires a long fallow period under poor poor management . Monoculture of sugar cane in the past has also added to the loss of much soil in hilly areas. Harsh and Torrential rains occur often after long periods of droughts when vegetative cover is sparse. This has contributed to the erosion problem.

In the hills of the central region, Indian Creek loam and Liberta clay loam are both described as being very eroded with parent material being exposed in some places. In the south west mountains, frys clay loam and springhill loam, little soil is left at all on the steeper slopes.

2.7.4 Barbados:

According to Veron and Carroll (1966) about 25% of Barbados occurs occupies relatively flat coral soils on which erosion is not considered to be severe. They suggested that erosion control measures, maybe necessary on the soils of upland plateau of St. john’s Valley.

In the hilly Scotland District erosion is very severe. Cumberbatch (1985) reported that it was estimated that 70% of the area was threatened by erosion and that 11% of it had reached a very severe state of degradation. L andslides and gullying are common.

2.7.5 Dominica:

The soils of Dominica are highly permeable except soils formed on igneous rocks, the shoal soils and other soils become less permeable during pedologic development. As a result not as much erosion occurs. Dominica is charaterised by steep slopes where 86% of the land area has slopes greater than 20 degrees and only 2% has slopes between 0 to 5 degrees. Slopes of over 60% with natural vegetation and cultivated slopes over 50 degrees are reported by Lang (1967), indicating that the soil is of unusual stability. It is important to note however that erosion is limited because much of the land is still under forest.

On the Leeward side of the island, most of the shoal soils and other soils of low permeability occur. Poor soil and crop management has been the major contributor to the soil erosion problem. The slow regeneration of soil suitable for cropping in he dry areas (as in St. Lucia also) increases he intensity of soil erosion.

According to the environment profile of Dominica prepared under the advice of the Caribbean Conservation Association in 1991, Dominica has great potential for agricultural development without damaging or removal of the forest lands. Forestry and forestry development are important. Timber extraction is undertaken but according to Russell (1974) it damages only a relatively small area and the erosion caused is not significant.

2.7.6 Grenada and Carriacou:

According to Vernon et al (1958), although some severe erosion can be seen in the hills where shifting cultivation, fragmentation of land, poor land distribution and poor cultivation practices exist, Grenada has suffered less from erosion and fertility exhaustion than many of the Caribbean islands. There are two main reasons for this;

The island’s agriculture is mainly on tree crops, cocoa (Theobroma cacao) and nut meg (Myristica fragrans), banana an food crops are frequently interplante with cocoa; and major forest fires are not prevalent.

The soil parent material is very base rich an the soil has been further enriched by additions of volcanic ash from eruptions in nearly islands in recent times. Even if some surface soil is lost, the underlying exposed material is almost as fertile and promotes raid vegetative growth.

Caribbean has suffered very severe erosion over nearly the whole island and in many cases only the parent rock remains.This has been as a result of the erodible nature of the soil, unsustainable cultivation methods an overgrazing. The soils of cariacou are skeletal soils over ash and agglomerate soils formed from other igneous rocks and those formed from limestone. They are all susceptible to erosion and deep gullies.

2.7.7 Monsteratt:

Lang (1976) describes the whole island as suffering from severe soil erosion and the greater portion having lost its top soil. Unlike the soils of Dominica, they are generally unstable and many soils are highly erodible if cultivated on slopes greater than ten degrees.

2.7.8 St. Vincent:

According to Watson et al (1958) soil and land use survey of St. Vincent, he mentioned that because of certain types of crops grown, soil erosion is a serious problem. The three crops which contribute to much of the erosion are cotton (Gossypium sp.), arrowroot (Maranta arundinacea) and ground nuts (Arachis hypogaea). Cultivation of these crops disturb the soil and leave the soil bare for long periods. Erosion can be seen in all but the flattest areas. It is important to note however that the farmers of St. Vincent are conscious of the erosion problem and soil conservation methods are of high standards.

2.7.9 St Lucia:

Soil erosion is a major problem in St. Lucia. Many of the soils interior are susceptible to heavy rains and soil can be seen being washed away by even the smallest rivulets (Stark et al 1966). In areas of allophonic clay soils, slumping is a problem and gullying also is common throughout the island. Common practices such as clean cultivation of very steep land and of clearing steeply sloping forested land which would be left under its original vegetation have accentuated the situation. Probably the most severe affecs of erosion are on the so called ” shoal soils” which contain a layer of indurated material in the substratum.

2.7.9.1 St. Kitts and Nevis:

Information on soil erosion on these two islands is taken from the soil survey report by Lang and Carrol (1967) Erosion in the mountainous areas of the islands has much been reduced by conserving the area in forest. Where the forest has been cleared and planted to provision crops, some accelerated erosion is seen. The subsoil of most of the island’s soils is free draining and contour cultivation is common the worst effects of erosion are usually avoided. On the island of nevis in contrast, much erosion can be seen. In much of the principal agriculture areas the surface soil has been lost completely.

2.7.9.2 Jamaica:

The soil survey reports of Jamaica (Barker, 1963 and 1970; Finch 1959 and 1961; Morgan and Baker, 1963; Price 1959a and 1959b; Stark 1963, 1964a and 1964b; Vernon, 1959 and 1960) indicate that soil erosion is rampant in many parts and on many soils of the island. One of the major factors is population pressure which results in increasing frequency and intensity of land use or the use of land beyond its capability. The parishes of Westmorland, Portland, Hanover, Clarendon and St. Elizabeth are especially noted as suffering from severe accelerated erosion. Soils where erosion is a serious problem are limestone soils, shale deriveds and the soils formed on grano-diorite. Also the shale soil soils are noted for their shallowness, poor infiltration and permeability, factors which contribute to serious