You are on page 1of 4

relationship with mean annual rainfall.

I should be stressed that extrapolating such relationships beyond the data from which they have been derived in order to apply them elsewhere is dangerous. For example, applying the equation developed of malaysia to mean annual rainfall totals below 900 mm yields estimated of erosivity that are not obviously nonsense. I should also be note that result of different researchers are not always comparable because of assumptions made when calculating the R value. The relationship developed between mean annual R and mean annual precipitation by Bollinne et al. (1979) for Belgium is based on rainstorms greater than 1,27 mm, whereas that proposed by Rogler and Schwertmann (1981) for baravia, Germany, considers storms only with rainfall greater than 10 mm and I30 values greater than 10 mmh-1. Erosion risk in Great Britain was assesed using the KE > 10 indeks. The annual values are rather low (Morgan 1980b; Fig 4.7), rising above 1400 jm-1 onlyn in part of the Pennines, the Welsh mountain, Exmoor and Dartmoor. They are less than 900 Jm-1 along most or the west coast and below 700 in the Outer Hebrides, Orkneys, Shetlands and on the nort coast f Scotland. Values over much of eastern and southern England are around 1100-1300. Since these are main areas devoted to arable to arable farming, it is here that the greatets risk of agricultural soil erosion occurs.

4.1.2 Rainfall aggressiveness The most commonly used index of rainfall aggressiveness, shown to be significantly correlated with the sediment yield in rivers (Fournier 1960), is the ratio p2/P, where p is the highest mean monthly precipitation and P is the mean annual precipitation. It is strictly an index of concentration of precipitation into a single month and thereby gives a crude measure of intensity of rainfall and, in so far as high value denotes a strongly seasonal climatic regime with a dry season during which the plant cover decays, of erosion protection by vegetation. The index was used by Low (1967) to investigate regional variation in erosion risk in Peru and by Morgan (1976) in Peninsular Malaysia (Fig 4.8). Using data from 680 rainfall stations, a low but significant correlation was obtained in Malaysia between p2/P and drainage texture, defined as the number of first-order streams per unit area (r=0,38; n=39; Morgan 1976). Since drainage texture is analogous to gully density, p2/P may be regarded as an idicator of the risk of gully erosion. In contrast, mean annual erosivity values reflect the risk of erosion by rainsplash, overland flow and rills. By superimposting the maps of p2/P and erosivity, a composite picture of erosion risk is obtained (Fig. 4.9)

As expected from above, there is often a poor relationship between p2/P and mean annual R. The emphasis given in p2/P to the month with the highest rainfall underplays the contribution of the rainfall in the rest of the year to erosion. If the mean annual rainfall increase but the highest monthly total remains the same, the p2/P actually falls in value whereas the potential for erosion should increase, since a proportion rains is likely to be erosive. Arnoldus (1980) proposed a way of overcoming this defect by considering the rainfall of all month and developing a modified Fournier Index (MFI): Where p is the mean monthly precipitation and P is the mean annual rainfall. Based on significant correlation beetween MFI and mean annual R for different climatic regions, mean annual erosivity maps have been produced for the Middle East and Africa north of the equator (Arnoldus 1980) and for 16 countries of the European Union (Gabriels 2002; Fig. 4.10).

4.1.2 Rainfall aggressiveness A simple scoring system for rating erosion risk was devised by Stocking and Elwell (1973b) for Zimbabwe. Taking a 1:1000,000,000 base map, the country was devided on a grid system in to units of 184 km2. Each unit was rated on a scale from 1 (low risk) to 5 (high risk) in respect of erosivity, erodibility, slope, ground cover and human occupation, the latter taking account of densityof the population and the type of settlement. The five factor score were summed to give total score, which was then compased with an arbirarily chosen classification system to categorize areas of low, moderate and high erosion risk. The scores were mapped and areas of similiar risk delineated (Fig. 4.11). Several problems are associated with technique. First, the classification may be sensitive to different scoring systems. For example, the use of different slope groups may yield different assessments of degree of erosion risk. Second, each factor is treated independently, whereas there is often interaction between the factors. Slope steepness may be much more important in areas of high than in areas of low erosivity. Third, the factors are combined by addition. There is no reason why this should be a more appropiate method of combining them than multiplication, although multiplication often results in the score for one factor dominating and the total score and, for that reason is dificult to the use with zero values in the scoring system. Fourth, each factor is given equal weight. Despite these difficulties, the

technique is easy to use and has the advantage that factors which cannot be easily quantified in any other way can be readily included. When use carefully, factorial scoring can provide a general appreciation of erosion risk and indicate vulnerable areas where more detailed assessments should be made. A system based on the susceptibility of soil to crusting (four clases), the shear strength of the soil (three classes), land cover (nine classes) and rainfall erosivity (four classes) gives good correlations for the cultivated areas of France between erosion risk and the spatial frequency of muddy floods (Le Bissonnais et al. 2002).

Keterangan: Erosivity values Mean annual erosivity (KE>10) in Great Britain (after Morgan 1980b)

Erosion risk Moderate High Values of p2/P in Peninsular Malaysia

Severe High (gullying) High (overland flow) Moderate Reconnaissance survey of soil erosion risk in Peninsular Malaysia

Mean annual erosivity for 16 countries of the European Union based on the modified Fournier index for rainfall stations with ten years or more of monthly rainfall record (after Gabriels 2002). Values under 100 are considered low and values over 300 high. No data were avaliable for much of France and central England

Major groups Low Very low Below average Average Above average

High Very high

Factor score

Subgroups according to dominant factors Erosivity Cover Slope Erodibility Human occupation No dominant factor

Categories

Erosivity (J mm m h )
-2 -1

Cover (mm of rainfall and basal cover est. (%) above 1000 7-10 800-1000 5-8 600-800 3-6 400-600 1-4

Slope (degrees)

Erodibility

Human occupation*

Low

below 5000

0-2

orthoferralitic regosoils

Extensive large scale commercial ranching National Parks or Unreserved Large scale commercial farms Low density CLs (below 5 p.p.km2) and SCCF Moderately settled CLs (5-30 p.p.km2)

Below average II

5000-7000

2-4

Paraferralitic

Average

III

7000-9000

4-6

Fersiallitic siallitic

Above average IV

9000-110000

6-8

vertisoils lithosoils Non-

High

Above 110000

below 400 0-2

above 8

calcichydromorphic sodic

Desenly settled CLs (above 30 p.p.km2)

(Notes: cover, Erodibility and Human occupation are only tentative and cannot as yet be expressed on a firm quantitative basis) * p.p.km2 = persons per square kilometre CL = Communal Lands SCCF = Small Scale Commercial Farms Erosion survey of Zimbabwe (after Stocking & Elwell 1973b)

You might also like