Use of 137Cs technique for soil erosion study in the agricultural region of Casablanca in Morocco

Journal of Environmental Radioactivity 68 (2003) 11–26

www.elsevier.com/locate/jenvrad

Use of ¹³⁷ Cs technique for soil erosion study in the agricultural region of Casablanca in

Morocco

A. Nouira

, E.H. Sayouty

, M. Benmansour

aDe´partement de Physique, Groupe Me´thodes Nucle´aires d’Analyse, Faculte´ des Sciences Aı¨n Chock, Universite´ Hassan II, Casablanca, Morocco

bCentre National de l’Energie des Sciences et des Techniques Nucle´aires, Rabat, Morocco Received 20 February 2002; received in revised form 9 December 2002; accepted 12 December 2002

Abstract

Accelerated erosion and soil degradation currently cause serious problems to the Oued El Maleh basin (Morocco). Furthermore, there is still only limited information on rates of soil loss for optimising strategies for soil conservation.

In the present study we have used the

Cs technique to assess the soil erosion rates on an agricultural land in Oued el Maleh basin near Casablanca (Morocco). A small representative agricultural field was selected to investigate the soil degradation required by soil managers in this region. The transect approach was applied for sampling to identify the spatial redistribution of

Cs. The spatial variability of

Cs inventory has provided evidence of the importance of tillage process and the human effects on the redistribution of

Cs. The mean

Cs inven- tory was found about 842 Bq m

, this value corresponds to an erosion rate of 82 tha

yr

by applying simplified mass balance model in a preliminary estimation. When data on site characteristics were available, the refined mass balance model was applied to highlight the contribution of tillage effect in soil redistribution. The erosion rate was estimated about 50 tha

yr

. The aspects related to the sampling procedures and the models for calculation of erosion rates are discussed.

 2003 Elsevier Science Ltd. All rights reserved.

Keywords:¹³⁷Cs; Erosion; Agricultural field;¹³⁷Cs calibration models; Morocco

∗ Corresponding author. Fax:+212-22-23-06-74.

E-mail address: [email protected] (E.H. Sayouty).

0265-931X/03/$ - see front matter2003 Elsevier Science Ltd. All rights reserved.

doi:10.1016/S0265-931X(03)00025-0

1. Introduction

Current concern for the global environment is focused on soil degradation and erosion. There are especially major constraints to a sustainable development of agri- cultural productivity in Morocco. The main impact of erosion is the reduction in the natural soil fertility, which leads to the progressive disappearance of arable lands, and the sedimentation in water reservoirs. It is estimated that out of 22.7 million hectares, potentially exploitable in the septentrional part of Morocco, 77% are exposed to very high erosion risks (Belkheri, 1988). Thus, it is important to collect reliable information on the rates of soil loss in Morocco in order to quantify the severity of erosion and to design effective strategies for soil conservation.

Using classical erosion techniques for monitoring soil loss has many limitations:

they are difficult to handle, time consuming, and expensive (Laflen et al., 1991). A number of researchers have explored the potential for using fallout of radio-nuclides (He and Walling, 1997; Walling and Quine, 1995), and particularly

Cs, to obtain estimates of the rates of deposition and soil erosion in agricultural lands. The

Cs technique provides retrospective information on long-term rates of soil loss, on the spatial pattern of erosion and on deposition based on a single site visit. It was applied in a wide range of environments, and the basis of the technique is well established and documented (Bernard and Lavadiere, 1990; Elliot et al., 1990; Quine and Wall- ing, 1991; Ritchie and McHenry, 1990; Ritchie et al., 1974; Rogowski and Tamura, 1970; Walling and Quine, 1993).

However, the use of the technique in Morocco has been very limited (Benmansour et al., 2000; Bouhlassa et al., 1994; Lahlou, 1997). This article deals with erosion assessment in the region of Aı¨n Harouda near Casablanca. This region is of great interest for the soil managers because it suffers from erosion and there are no quanti- tative data on erosion available to provide objective assessments of the magnitude of the problem in order to optimise the required management.

The aims of the study are: (i) to evaluate the reliability of the

Cs technique for the quantification of erosion in the Aı¨n Harouda region; (ii) to quantify the spatial soil redistribution using a transect approach; and (iii) to estimate erosion or depo- sition rates on a representative cultivated field.

2. The basis of caesium-137 technique

137

Cs is an artificial radionuclide with a half-life of 30.2 yr. Its distribution in the

environment began with atmospheric testing of thermonuclear weapons, primarily

during the period from 1954 to mid-1970s. The

Cs was released into the atmos-

phere, distributed globally and deposited as fallout. The amount of deposition

depends on the radionuclide concentration in the atmosphere and rainfall. Its use as

a sediment tracer lies in its rapid and strong adsorption by fine particles (Bachhuber

et al., 1982; Livens and Baxter, 1988) so that in most agricultural environments its

subsequent redistribution is a direct reflection of the erosion, transport and deposition

of soil particles occurring over a period of ca. 35 yr (Walling and Quine, 1995).

13 A. Nouira et al. / J. Environ. Radioactivity 68 (2003) 11–26

Assessment of

Cs redistribution, and thereby soil redistribution, is commonly based upon comparison of measured inventories of

Cs from the sample sites com- bined with an equivalent estimate of cumulative atmospheric deposition for the stud- ied area. Generally, the cumulative input of the reference inventory is established by sampling nearby, undisturbed locations, which are known to have suffered no erosion or deposition disturbance during the period since the onset of

Cs depo- sition. Therefore, it is essential to determine the reference inventory with a great care.

The use of the

Cs technique has the following advantages (Walling and Quine, 1993).

앫

The technique permits retrospective assessment of medium-term rates of erosion.

앫

The application of the technique requires only a single site visit, and the results can be provided within a relatively short time.

앫

The rates estimated are the consequence of all erosive process. Both rates and patterns of soil redistribution may be quantitatively estimated.

앫

Estimates of erosion rate may be obtained, including mean rates of erosion and deposition and net rates of soil exported from the field.

앫

The spatial resolution of the data obtained is defined by the sampling strategy.

3. Characteristics of the site and sampling 3.1. Geographic localisation

The representative cultivated land is located in the Oued El Maleh basin in the region of Aı¨n Harouda at 20 km north of Casablanca. The selected field situated at 7°25 ⬘ N and 33°37 ⬘ W was chosen to be representative of the surrounding area with a surface of about 0.3 ha. The field altitude ranges from 82 to 87 m above sea level with an average slope of 5° (Fig. 1).

The climate of the region is classified as Mediterranean, characterised by mean temperatures varying from 13 °C during the coldest months (December–February) to 23 °C in summer (June–September). Rainfall regime is irregular with a dry period in summer and a wet period in autumn and spring. The mean annual precipitation is of ca. 400 mm. The main soil type is composed of a great percentage of loam- clay ranges approximately from 50 to 70% and sand mixture (Beaudet, 1969). The land studied has been cultivated without the use of any special soil conservation tech- niques.

The choice of a suitable reference site is important. The reference site should

have received the same annual precipitation and have the same geomorphological

parameters as the studied field. There were no uncultivated native grassland sites in

the general area and it was difficult to find sites that had clearly been undisturbed

since 1950. Thus, an undisturbed reference site was selected in the forest (named

Cascade) with a flat topography for this study. It was not affected by soil erosion

or forest deposition, and it is located 2.5 km away from the investigated field with

an altitude of about 66 m above sea level.

Fig. 1. Map of Mohammedia–Casablanca (Morocco) 1:50000.

3.2. Sampling methodology

The sampling procedure shown in Fig. 2 is based on the transect approach; this consists of a sequence of samples collected along the axis of the greatest slope from the upslope to downslope boundary. A grid framework was established consisting of five parallel transects across the field, approximately 6 m apart. Along each tran- sect, six soil cores were collected. The distance between two adjacent sampling points was 10 m. A motorised cylindrical tube (ca. diameter: 67 cm

and length: 1 m) was inserted to a depth of 32 cm to ensure that all

Cs is retained. To obtain information concerning the depth distribution of

Cs concentration, the soil core was divided into 1–2 cm incremental samples.

Seven reference samples were collected inside a circle 20 m in diameter in the

undisturbed site. The positions of all sampling points were determined by using a

GPS system.

15 A. Nouira et al. / J. Environ. Radioactivity 68 (2003) 11–26

Fig. 2. The sampling procedure using the transect approach.

4. Analysis of samples

The soil samples were dried at 103 °C for 24 h, disaggregated, passed through a 2 mm sieve, mixed and weighed. The obtained whole samples were put in Marinelli beakers of a volume 500 ml whereas the incremental samples were prepared in cylin- drical containers of a volume 200 ml.

Measurements of

Cs concentrations were conducted on a sub-sample of the finer fraction (⬍2 mm) of each sample by gamma-ray spectrometric analysis, using two high purity coaxial germanium detectors (苲30% efficiency,

coupled to a multichannel analyser system. The

Cs activities were determined by calculating the net area under the Gamma (γ) peak on the spectrum at 662 keV. The use of certified multi-gamma liquid solution and IAEA reference materials (IAEA 375, soil 6) allowed us to calibrate the detection system efficiency for each counting geometry. The radioactivity of the samples was counted for 12–24 h with a precision lower than ca. 10% at the 95% level of certainty.

The

Cs inventory A (in Bq m

) may be expressed as:

A ⫽ CM

S (1)

where C is the

Cs activity of the sample (in Bq kg

), M the total mass of the

collected soil core (in kg), and S is the cross-section of the sampling tube (in m

).

5. Results and discussion

5.1. Depth distribution of

Cs activity

The distribution of

Cs activity as a function of the soil depth in the reference and agricultural sites was examined in order to provide background information on the behaviour of

Cs fallout within the soil.

The

Cs activity concentration profile for the identified reference site shows a sharp decrease of

Cs activity concentration with increasing depth (Fig. 3), which can be fitted by an exponential function. This distribution is typical of an undisturbed site (Bachhuber et al., 1982; Frissel and Pennders, 1983). Most of

Cs is contained within the top 20 cm from surface, with retention of 80% of the

Cs in the upper 14 cm and sharp drop in

Cs activity below that depth. This provides further con- firmation of the validity of using data from this site to establish the reference inven- tory of the field studied.

The depth distribution of

Cs for the cultivated site is shown in Fig. 4. The

Cs found, was quite equally distributed throughout the plough layer due to cultivation practices. The plough layer is about 30 cm. The

Cs activity concentration value at the surface is of 3.6 Bq kg

. This value is close to the results obtained earlier in a study related to the superficial distribution of

Cs within the Moroccan soil (Ait haddou et al., 1996).

Fig. 3. Depth distribution of¹³⁷Cs for the local reference site.

17 A. Nouira et al. / J. Environ. Radioactivity 68 (2003) 11–26

Fig. 4. Depth distribution of¹³⁷Cs for the cultivated site.

These results confirm that

Cs behaviour is in agreement with the expectations and the assumptions of the technique used in this studied area.

5.2.

Cs inventories

The

Cs inventories for the seven cores collected from the reference site are given in Table 1. The values are quite close except the R

and R

sampling points, which represent a percentage deviation of about 20% from the mean reference value.

The main cause of this variation results probably from the spatial variability of

Cs inventory due to soil heterogeneity such as soil microreliefs, argillaceous level, veg- etable density and biological activity as well as by splash or runoff phenomena. Such phenomena have also been observed in other studies (Owens and Walling, 1996;

Sogon, 1999). The mean

Cs activity obtained for the reference site is about 1587

Table 1

137Cs Measurement inventories in reference site

Sampling R11 R12 R13 R14 R15 R16 R17 Mean value (Bq m^⫺2) point

Activity (Bq 1803 1930 1268 1638 1407 1353 1712 1587±32 m^⫺2)

Bq m

with a relative standard deviation of 16% (i.e. relative standard deviation is the standard deviation divided by the mean value).

Concerning the area studied in this work, all the

Cs inventories established from the sampling points collected at each transect are distinctly lower than the reference inventory, and this could suggest that some erosion or soil removal occurred at this location. The

Cs inventory values ranged between 383 and 1358 Bq m

with standard deviations of the transects varying from 105 to 300 Bq m

. The lower percentage of

Cs observed for nearly all the samples (Tables 1 and 2) is an indi- cation of the important soil loss produced in this land.

Since all sampling points present a

Cs activity lower than the reference activity, no deposition points are observed downslope the field.

The variation in

Cs activity with the distance from upslope to downslope along the transects is presented in Fig. 5. Usually the longitudinal variation in

Cs activity, related to the soil loss, depends on the variations of the topography and the position of the sampling point from hilltop.

We can see from transect 1 that

Cs inventory decreases rapidly with distance until 20 cm starting from the top. The maximum loss reaches 76% of the reference inventory showing hence an evidence of even serious soil loss. Then

Cs activity loss becomes lower. This can be interpreted as erosion or export of soil are not sig- nificant.

On the other hand the transects 2, 3 and 5 show an irregular and oscillating behav- iour of

Cs activity loss suggesting a variation of soil loss along the transects. The mean

Cs activity values of transects 2, 3 and 5, respectively are of 35, 54 and 40% loss of

Cs input.

Transect 4 presents a great decrease of

Cs inventory loss downslope. The

Cs inventory loss is about 71% of the reference value upslope and reaches about 18%

downslope. This highlights a significant erosion in the top of the transect, whereas downslope, the lower

Cs inventory loss seems to correspond to soil redeposition.

Table 2

137Cs measurement inventories in agricultural site

Transect Distance (m)

0 10 20 30 40 50

T1 1009 563 383 714 856 584

T2 868 1358 1014 1127 956 817

T3 769 676 899 581 717 729

T4 465 751 650 756 1048 1305

T5 898 1268 770 896 1046 778

Lateral mean erosion 802 923 743 815 926 843

rate (Bq m^⫺2)

Standard deviation 207 363 243 208 139 273

(Bq m^⫺2)

Relative standard 26 39 33 25 15 32

deviation (%)

19 A. Nouira et al. / J. Environ. Radioactivity 68 (2003) 11–26

Fig. 5. The¹³⁷Cs activity redistribution along all the transects.

Variations of

Cs inventory loss are observed to be significant along the transects.

The behaviour of

Cs redistribution within the studied field can be explained as follows: (i) The influence of the shape of each transect of the field. Indeed the slope angle was not uniform along the transects, reflecting a complex topography of the studied land. The

Cs inventory loss increases as the slope increases, then progress- ively decreases with gradual increase of the slope. (ii) The heterogeneity of the soil is due to the tillage redistribution and human effects which may cause important soil movement.

The lateral variability in the

Cs activity may be also examined for each level (Table 3). Analysis of the data shows the lateral mean

Cs inventory for each level

Table 3

The erosion rates Y (tha^⫺1yr^⫺1), estimated in the studied site using simplified mass balance model

Transects Distance (m) Mean erosion rate (tha^⫺1yr^⫺1)

0 10 20 30 40 50

T1 55 125 171 97 75 121 107±5.4

T2 73 19 55 42 62 81 55±2.8

T3 88 103 69 121 96 94 95±4.8

T4 148 91 108 90 51 24 85±4.3

T5 69 27 88 69 51 86 65±3.3

ranges from 743 to 926 Bq m

with standard deviations varying from 139 to 363 Bq m

. The highest deviation between the transects was observed at level 10 m from the top with a relative standard deviation of about 39%. The deviation can be due to the influence of the topography on the lateral

Cs activity redistribution, hence showing the necessity to undertake sampling along several transects. The mean

137

Cs inventory obtained for the five transects is about 842 Bq m

with relative standard deviation of about 17%.

5.3. Estimation of soil erosion rates

Although the comparison of measured inventories with the local reference value provide useful qualitative information on soil erosion and deposition, the quantitative estimates of soil depend upon the existence of reliable relationship between the meas- ured

Cs inventory at specific sampling points and the rate of erosion and deposition at this point.

To derive quantitative estimates of the rates of soil erosion and deposition from

137

Cs radioactivity measurements, it is necessary to establish a relationship between the magnitude of the deviation from the reference inventory and the extent of soil loss (or gain).

5.3.1. Use of simple models

At the beginning when data on site characteristics were not available, two simple models were tested: the proportional model and the simplified mass balance model.

The proportional model has been widely used to estimate soil erosion rates from

137

Cs measurements on cultivated soils particularly for low soil erosion rates ( ⬍ 30 tha

yr

) (Mitchell et al., 1980). It is based on the assumption that

Cs fallout inputs are completely mixed within the plough or cultivation layer and that soil loss is directly proportional to the reduction in the

Cs content in the soil profile. The mean annual soil loss rate Y (tha

yr

) can be written as follows:

Y ⫽ 10 BdX

100T (2)

where X is the percentage reduction in total

Cs inventory (defined as (A

⫺ A) / A

× 100), d the depth of the plough or cultivation layer (m), B the bulk density of the soil (kg m

), T the time elapsed since initiation of

Cs accumulation (yr), A

the local

Cs reference inventory (Bq m

), and A is the total measured inven- tory at the sampling point (Bq m

).

The proportional model does not take into account the dilution of

Cs concen- trations in the plough layer after surface lowering by erosion. The results obtained may underestimate the actual rates of soil loss.

The simplified mass balance (Kachanoski, 1993; Quine et al., 1996; Zhang et al.,

1990) attempts to overcome the limitation assuming that the total input of

Cs

fallout occurred in 1963 instead of over the period from 1954 to 1970 and that the

depth distribution of

Cs in the soil profile is not time dependent.

21 A. Nouira et al. / J. Environ. Radioactivity 68 (2003) 11–26

According to this model, the mean annual soil loss rate (tha

yr

), Y is expressed as follows:

Y ⫽ 10dB 冋 ^{1 ⫺} 冉 ^{1 ⫺ 100 X} 冊

册 ⁽³⁾

where t is the time since 1963.

This model is considered by the above mentioned authors to be more realistic than the proportional model. The difference between the two models becomes important for the high soil erosion rates (⬎30 tha

yr

). The proportional model underestimates the soil erosion rates. An example of this comparison can be observed in Fig. 6, which illustrates the erosion rate redistribution along one of the transects (cf. transect II in which the bulk density B is 1469 kg m

and the plough layer is about 0.3 m). The soil loss rates derived from the proportional model and the simpli- fied mass balance model demonstrate a good agreement in terms of behaviour whereas there is a clear difference in the magnitude of the rates between the two models for all sampling points of the transect (Fig. 6). The erosion rates derived from the proportional model remain always lower than those predicted with simpli- fied mass balance model. This clearly indicates the dilution effect of soil within the plough layer by soil containing

Cs below the original plough depth. Therefore, a preliminary estimation of the soil loss in this field, when data on site characteristics were not available, is given by the simplified mass balance model.

Table 3 reproduces the erosion rate estimates for the different transects using the simplified mass balance model. It shows a spatial pattern of soil redistribution occur- ring within the studied land. The assessed erosion rate ranges from 55 to 107 tha

yr

with a mean erosion of about 82 ± 4.1 tha

yr

. This spatial soil distribution is due to a number of interactions between the topographic and the anthropic factors which influence the removal and transport of soil particles.

Fig. 6. The erosion rates estimated using simplified mass balance model (MB) and proportional model (PP) in transect II.

Assessed erosion rates which correspond to the mean erosion rate over a long period of about 35 yr, constitute the contribution of the tillage effect to soil and water erosion. It is well known that tillage erosion results from the mechanical dis- placement of the soil-surface material. Previous studies (Govers et al., 1994, 1996;

Lindstrom et al., 1992) carried out in agricultural regions of Canada and Europe to examine the relative importance of water and tillage erosions, have strongly sup- ported the idea that tillage contributes significantly to the soil redistribution. Further- more, the studied agricultural field was ploughed in the same direction as the slope.

5.3.2. Use of refined mass balance model incorporating tillage effects

Once the data on site characteristics are available, we have used the refined mass balance model in order to distinguish contribution of tillage effect to soil redistri- bution. The model constitutes the third version of mass balance approach (Govers et al., 1996; Walling and He, 1997). It incorporates the influence of tillage in soil redistribution and only applies to slope transects parallel to the flow directions.

The variation of total

Cs inventory A(t) (Bq m

) for a point experiencing ero- sion, is expressed as follows:

dA(t)

dt ⫽ (1 ⫺⌫ )I(t) ⫹ R

C

(t) ⫺ R

C

(t) ⫺ R

C

(t) ⫺ lA(t) (4) where C

, C

and C

(Bq kg

) are the

Cs concentrations of the sediment associated with tillage input, tillage output and water output, respectively, R

(kg m

yr

) the water erosion rate and ⌫ is the percentage of the freshly deposited

137

Cs fallout removed by erosion before being mixed into the plough layer. ⌫ can be expressed as ⌫ = Pg(1 ⫺ e

), where g is the proportion of the annual

Cs input susceptible to removal by erosion.

The net erosion rate is given by

R ⫽ R

⫺ R

⫹ R

(5)

For a point experiencing deposition, the expression of relation (4) will be:

dA(t)

dt ⫽ I(t) ⫹ R

C

(t) ⫺ R

C

(t) ⫹ R ⬘

C

(t) ⫺ lA(t) (6) where R

is the rate of water induced deposition.

The net erosion rate R becomes as follows:

R ⫽ R

⫺ R

⫺ R ⬘

(7)

The contribution of tillage effect is estimated from the following equation

R

⫺ R

⫽ ⌽ (sin b

⫺ sin b

) / L (8)

where b

and b

are the slope angle degrees before and after the sampled point, L the slope length and ⌽ (kg m

yr

) is a constant.

The

Cs concentration of the soil within the plough layer C

(t⬘) (Bq kg

) can

be expressed as

23 A. Nouira et al. / J. Environ. Radioactivity 68 (2003) 11–26

C

(t⬘) ⫽ A(t ⬘ )

d for a net erosion site C

(t ⬘ ) ⫽ 1

d 冋 ^{A(t ⬘ ) ⫺ |R| d} 冕

A(t ⬙ ) e

dt ⬙ 册

for a net deposition site

(9)

where |R|(R ⬍ 0) is the net deposition rate. The relationships between C

,C

and C

are as follows:

C

(t ⬘ ) ⫽ C

(t ⬘ ) ⫽ C

(t ⬘ )C

(t ⬘ ) ⫽ PC

(t ⬘ ) ⫹ I(t ⬘ )

R

Pg(1 ⫺ e

) (10) while the

Cs concentration of water-derived deposited sediment C

(t ⬘ ) (Bq kg

) can be expressed as:

C

(t ⬘ ) ⫽ 1

冕

^RdS 冕

^{P ⬘ C}

^{(t ⬘ )RdS (11)}

The refined mass balance model represents an important improvement over the simplified mass balance model and proportional model. Its application requires infor- mation on the annual atmospheric

Cs deposition flux I(t) (Bq m

yr

). In our case, according to the geographic situation of Morocco and the absence of such information, the local temporal patterns of

Cs deposition flux has been assumed to be similar to those of the total

Cs deposition to the northern hemisphere esti- mated from the data recorded at various stations in the northern hemisphere (cf.

Cambray et al., 1989). The proportion factor g and relaxation mass depth H for the initial distribution of

Cs fallout in surface soils must also be supplied. The value of g depends on the temporal distribution of local rainfall and the timing of culti- vation. Since there is only one annual cultivation operation in study zone and accord- ing to the local rainfall regime, the value of g is found to be of 0.63. For the relaxation mass depth H, because of the difficulty to determine this parameter in the studied land, we have assumed H to be 4.0 kg m

(Bouhlassa et al., 2000; He and Wall- ing, 1997).

Table 4 summarises the average erosion rates estimated from each transect by using the refined mass balance model (Y RMB), and distinctly the contribution of tillage redistribution (Y RMB(t)) and water erosion (Y RMB(w)). The mean erosion constitutes the algebraic sum of water erosion and tillage redistribution, it is esti- mated to be about 50 tha

yr

.

It is clearly observed from Table 4 that there is no great variation in term of

magnitude of erosion induced by tillage effect between the transects, only transect

4 exhibits a value of 23 tha

yr

, which is few times larger than the others. This

can be explained by the microtopography of the site. The mean tillage erosion esti-

mated of 14 tha

yr

remains a high value which is leading to an important impact

of tillage contribution to soil redistribution of about 28% of the total erosion.

Table 4

The mean erosion rates Y (tha^⫺1yr^⫺1), estimated in the studied land by applying the refined mass balance model and distinctly the contribution of tillage redistribution (Y RMB(t)) and water erosion (Y RMB(w)) of each transect

Transects Erosion rate (tha^⫺1yr^⫺1)

Y RMB Y RMB (t) Y RMB (w)

T1 66 11 55

T2 34 12 22

T3 52 13 39

T4 58 23 35

T5 39 11 28

Mean erosion (tha^⫺1 50 14 36

yr^⫺1)

Although the present results have proved an important effect of tillage contribution in soil movement, the major influence is from water erosion, which corresponds to a soil loss of approximately 36 tha

yr

. The estimated water erosion rate seems to be relatively high, taking into account the mean annual rainfall regime and the field topography. A previous study made in the Zitouna basin in north Morocco where the selected fields are characterised by slopes varying from 7 to 20% and rainfall regime of approximately 400 mm, has provided by applying

Cs technique an erosion rate ranging from 11.9 to 22.3 tha

yr

(Benmansour et al., 2000).

According to this study, the mean erosion rate obtained of 50 tha

yr

in studied area can be considered as relatively high. This can be explained by the following factors: (i) the surface encrusting which involves more runoff, (ii) the floods due to disturbance occurring in the hydrological regime, and (iii) the geomorphological parameters of this land. The results from this model are likely to be closer to reality for the studied field than those derived from the simplified mass balance model and the proportional model presented in Section 5.

6. Conclusion

The study of the Aı¨n Harouda region (Morocco) has demonstrated the potential of the

Cs technique to provide erosion rate data. The detailed study and the sam- pling method using several transects have allowed us to obtain on one hand the spatial distribution of

Cs in the soil and to highlight the impact of the agricultural practices and on the other hand to contribute to the assessment of long term ero- sion rates.

The mean erosion rate was estimated to be about 82 tha

yr

by applying the

simplified mass balance model whereas by using the refined mass balance model it

was found to be about 50 tha

yr

. The approach of the refined mass balance

model has allowed us to show the contribution of tillage effect, which corresponds

25 A. Nouira et al. / J. Environ. Radioactivity 68 (2003) 11–26

to an erosion rate of 14 tha

yr

in studied area. The obtained high erosion rate of about 50 tha

yr

confirms the significance of soil erosion problem for this study cultivated field.

Due to the severity of soil erosion problem for the basins in Morocco, the use of the

Cs technique appears to be a suitable method to estimate the soil loss.

References

Ait haddou, A., Ibn majah, M., Benmansour, M., Embarech, K., Badraoui, M., 1996. Evaluation of¹³⁷Cs pollution in Morocco. In: October 9th–11th, Mohammedia, Proceeding of the International Symposium on Environmental and Impact Assessment, 64.

Bachhuber, H., Bunzl, K., Schimmack, W., Gans, I., 1982. The migration of¹³⁷Cs and⁹⁰Sr in multilayered soil: results from batch, column and fallout investigations. Nuclear Technology 59, 291–301.

Beaudet, G., 1969. Le plateau central Marocain et ses bordures, e´tude ge´omorphologique. Ouvrage publie´

avec l’aide du M.E.S du Maroc et M.F.E.N et des A.Rabat, 275 pp.

Belkheri, A., 1988. Conse´quences de la de´gradation des bassins versants sur les retenues de barrages.

Rapport du Se´minaire National sur l’ame´nagement des bassins versants “Diagnostic de la situation actuelle”. MARA., T.P., PNUD/FAO. 18–28 Janvier, Rabat, 23 pp.

Benmansour, M., Ibn majah, M., Marah, H., Marfak, T., Walling, D.E., 2000. Use of the¹³⁷Cs technique in soil erosion investigation in Morocco-case study of the Zitouna basin in the north. In: October 16th–20th, Proceeding of an International Symposium on Nuclear Techniques in Integrated Plant Nutrient, Water and Soil Management. AIEA/FAO, Vienna, pp. 308–315.

Bernard, C., Lavadiere, M.R., 1990. Variabilite´ spatiale de l’activite´ en Cs-137 et ses re´percussions sur l’estimation de l’e´rosion hydrique. Pe´dologie 40, 299–310.

Bouhlassa, S., Azenfar, A., Machrouh, A., 1994.¹³⁷Cs fallout as tracer of erosion and sedimentation in big catchment. Applied Radiation and Isotopes 46 (6/7), 659.

Bouhlassa, S., Moukhchane, M., Aiachi, A., 2000. Estimates of soil erosion and deposition of cultivated soil of Nakhla watershed, Morocco, using¹³⁷Cs technique and calibration models. Acta Geologica Hispanica 35 (3/4), 239–249.

Cambray, R.S., Playford, K., Carpenter, R.C., 1988. Radioactive fallout in air and rain: results to the end of 1988. In: UK Atomic Energy Authority Report. HMSO, London, UK (AERE-R 10155).

Elliot, G.L., Campbell, B.L., Lougran, R.J., 1990. Correlation measurements and soil¹³⁷Cs contents.

Applied Radiation Isotopes 4 (8), 713–717.

Frissel, M.J., Pennders, R., 1983. Models for the accumulation and migration of⁹⁰Sr,¹³⁷Cs^239,240Pu and

241Am in the upper layers of soils. In: Coughtrey, P.J. (Ed.), Ecological Aspects of Radionuclide Release. Special Publication of British Ecological Society, No. 3, pp. 63–72.

Govers, G., Quine, T.A., Desmet, P.J.J., Walling, D.E., 1996. The relative contribution of soil tillage and overland flow erosion to soil redistribution on agricultural land. Earth Surface Processes and Land- forms 21, 929–946.

Govers, G., Vandaele, K., Desmet, P.J.J., Poesen, J., Bunte, K., 1994. The role tillage in soil redistribution on hillslopes. European Journal of Soil Science 45, 469–478.

He, Q., Walling, D.E., 1997. The distribution of fallout¹³⁷Cs and²¹⁰Pb in undisturbed and cultivated soils. Applied Radiation and Isotopes 48, 677–690.

Kachanoski, R.G., 1993. Estimating soil loss from changes in soil caesium-137. Canadian Journal of Soil Science 73, 629–632.

Laflen, J.M., Lane, L.J., Foster, G.R., 1991. WEEP a new generation of erosion prediction technology.

Journal of Soil and Water Conservation 46.

Lahlou, Y., 1997. Erosion dans le bassin versant du barrage Sidi Driss: me´thodologie d’approche et quantification (cas du sous bassin versant Mhasser). The`se de 3e´me cycle Universite´ Mohamed V Faculte´ des Sciences de Rabat, Maroc.

Lindstrom, M.J., Nelson, W.W., Schumacher, T.E., 1992. Quantifying tillage erosion rates due to mold- board plowing. Soil and Tillage Research 24, 243–255.

Livens, F.R., Baxter, M.S., 1988. Chemical association of artificial radionuclides in Cumbrian soils.

Environmental Radioactivity 7, 75–86.

Mitchell, J.K., Budenzer, G.D., McHenry, J.R., Ritchie, J.C., 1980. Soil loss estimation from ¹³⁷Cs measurements. In: DeBoodt, M., Gabriels, D. (Eds.), Assessment of Erosion. Wiley, Chichester, UK, pp. 393–401.

Owens, P.N., Walling, D.E., 1996. Spatial variability of¹³⁷Cs inventories at reference sites: an example from two contrasting sites in England and Zimbabwe. Applied Radiation and Isotopes 47 (7), 699–707.

Quine, T.A., Walling, D.E., 1991. Rates of soil erosion on arable fields in Britain: quantitative data from caesium-137 measurements. Soil Use and Management 7 (4), 169–176.

Quine, T.A., Walling, D.E., Govers, G., 1996. Simulation of radiocaesium redistribution on cultivated hillslopes using a mass-balance model: an aid to process interpretation and erosion rates estimation.

In: Anderson, M.G., Brooks, S.M. (Eds.), Advances in Hillslope Process. Wiley, Chichister, UK, pp.

561–588.

Ritchie, J.C., McHenry, J.R., 1990. Application of radioactive fallout¹³⁷Cs measuring soil erosion and sediment accumulation rates and patterns: a review. Journal of Environmental Quality 19, 215–233.

Ritchie, J.C., Spraberry, J.A., McHenry, J.R., 1974. Estimating soil erosion from the distribution of fallout

137Cs. Soil Science Society of America Proceedings 38, 137–139.

Rogowski, A.S., Tamura, T., 1970. Erosional behavior of¹³⁷Cs. Health Physics 18, 467–477.

Sogon, S., 1999. Erosion des sols cultive´s et transport des matie`res en suspension dans un bassin versant de Brie. The`se de doctorat, Universite´ Paris 1, France.

Walling, D.E., He, Q., 1997. Models for converting¹³⁷Cs measurements to estimates of soil redistribution rates on cultivated and uncultivated soils (including software for model implementation). In: A Contri- bution to the IAEA. (Co-ordinated research programmes on soil erosion (D1.50.05) and sedimen- tation (r3.10.01)).

Walling, D.E., Quine, T.A., 1993. Use of ¹³⁷Cs as tracer of tracer of erosion and sedimentation. In:

Handbook for the Application of the ¹³⁷Cs Technique. UK Overseas Development Administration Research Scheme, R4579, Department of Geography, University of Exeter, UK.

Walling, D.E., Quine, T.A., 1995. Use of fallout radionuclide measurements in soil erosion investigations.

In: Proceedings of an International Symposium on Nuclear and Related Techniques in Soil–Plant Studies on Sustainable Agriculture and Environment Preservation. IAEA, Vienna, pp. 597–619.

Zhang, X.B., Higgitt, D.L., Walling, D.E., 1990. A preliminary assessment of the potential for Using

137Cs to estimate rates of erosion in the Loess Plateau of China. Hydrological Sciences Journal 35.