Quantification of Soil Sensitivity to Water Erosion by the RUSLE Model in the Oued Amter Watershed, Northwestern Morocco

Abstract


Introduction
The Water erosion is a process that expresses itself through soil degradation.It threatens a large surface on a global scale and describes the phenomena due to man and/or climatic aggressiveness that reduces the production potential of soils and the quality of natural resources.It is the first manifestation of the phenomenon of desertification and is mainly focused in regions with a semi-arid climate and in the Mediterranean areas.Water erosion is considered an important phenomenon affecting the environment.It has negative consequences on the socio-economic framework at the local, regional and national level, in a general way, because of the demographic growth and also the climatic changes.It constitutes a danger on the infrastructures, on the agricultural productions and also on the quality of the waters.
A study on soil degradation in Morocco, carried out by the Food and Agriculture Organization of the United Nations (FAO), showed the extent of this phenomenon with a value of 12.6 million hectares of crops and rangelands that were considered to be under threat (Zouagui et al., 2018).Another more recent FAO study, in 1990, found that a percentage of 40% of the land in the whole territory was affected by water erosion (Lufafa et al., 2003;Benzougagh et al., 2022).The average specific degradation exceeds 50 T/ha/ha in several areas, particularly in the Rif, due to the aggressive climate, the rugged terrain, the great sensitivity of the lithological layers and the impact of the anthropic factor.This results in a significant and worrying degradation of soil fertility and productivity.But also by the reduction of the water storage capacity by the effect of the silting of dams.
The present work aims to assess the extent of erosion risk in the watershed of Oued Amter and the realization of the spatial model of soil loss using remote sensing (RS) and geographic information system (GIS).In addition, there are several models for quantifying water erosion and the choice is made according to the available data.The most widely used method at the national scale is the Revised Universal Soil Loss Equation (RUSLE) by (Wischmeier and Smith, 1978).It's used to figure out how much soil erodibility, topography, plant cover, rainfall erosivity, and erosion management measures affect the environment's ability to retain water.

Study Area
The Oued Amter watershed covers an area of 300 Km² and a perimeter of almost 100Km.It is part of the province of Chefchaouen which is located in the north-western part of Morocco (Fig. 1), it is located southeast of the city of Tetouan (between Oued Laou and Jebha) in the region of Tangier-Tetouan-Al Houceima (Fig. 1).The Mediterranean Sea, at an outlet X = 555115.06m Y = 516611.278m, and a mountain range, to the south, limit its northern extent (Bab Berred).It is located between latitudes 35° and 35°15' N and longitudes 4°40' and 4°50'W.It has a mountainous character with altitudes that vary between 0 m and 2100 m. (Jebel Tizirane).It shows a diversity of reliefs, with structural forms, depressions, ravines ... etc.

Precipitation
This basin is characterized by a semi-arid Mediterranean climate with strong seasonal contrasts and irregularities in rainfall.This basin is marked by highly variable rainfall from one month to another with an average annual rainfall of about 600 mm/year.The rainy season in the study area is spread out between the months of November and April with marked maxima in December and January.The driest months are July and August (Fig. 2).

The Geological Framework
The watershed of Oued Amter extends along the internal Rifain domain.The geology of the watershed plays a major role in the circulation of surface water.When a watershed present very permeable formations support a vegetation cover, then the density of drainage will be low which explain by an important infiltration.Conversely, if the geological formations of the watershed are impermeable with medium vegetation, this will facilitate the flow of water to the surface and then erosion and by effect the drainage density will be high, so the runoff from the surface is important than infiltration.It is generally formed by 3 morphostructural domains, from downstream to upstream we distinguish: the internal domain, Flysch nappes and the intrarificial zone (Fig. 3).

The Relief
The analysis of the distribution of the altitude ranges shows that the average altitude of this watershed is 700m.Before starting the part concerning the quantification of soil losses, we summarize the different Physiographic parameters analyzed in the following (Table 1).

Annual soil loss rate
The soil loss rate (SLR) is calculated using Wischmeier and Smith's empirical formula (1978).It is a mathematical model integrated with GIS tools that is frequently used to measure soil loss around the world.This equation is based on numerous experiments on agricultural plots in the USA.It is based on 5 explanatory factors of water erosion (Fig. 5).
After a downpour, the erosivity index.kinetic energy and the peak 30-minute rainfall intensity in millimeters per hour, both expressed as a percentage.In terms of soil erosion and sediment distribution, rainfall is one of the most important factors.Thirteen (13) stations are used to collect this information.The use of historical rainfall data and the application of multiple different formulas is required to determine the erosion factor in a given location.In places where there is a lack of reliable climate data, the R-factor is difficult to estimate.After doing many regression studies, Renard and Freimund (1994) came up with a novel model: (2) (3) Where: P is the annual rainfall (mm) and R is the rainfall erosivity (MJ mm ha).

The Soil Erodibility Factor (K)
Soil variables' cumulative influence is quantified by calculating the K-factor.However, it shows that a certain soil type is generally resistant to erosion.It is a measure of how quickly rainwater and runoff separate and transport soil particles.K values were calculated from FAO data, (World Digital Soil Map, 2019) http://www.fao.org/geonetwork/srv/en/metadata),due to the lack of soil quality data in the Oued Amter basin.The k-factor (soil erodibility factor) determines the vulnerability of the soil to erosion as well as the erosion rate.William et al. method is used to calculate the K factor (Sharpley and Williams, 1990).

2.4.4.The Topographic Factor (LS)
The slope L (meters) and the slope S (%) are the only two variables that have any meaning ( % ).Geomorphology's impact on water erosion is exemplified by this sculpture.The formula of Moore and Wilson, 1992, was used to estimate this factor: (5) Where: FA: flow accumulation map (The total flow accumulated on a frame basis in each cell, weighted for all cells flowing into each downstream cell).Cellsize: the grid cell size derived from the 30 m resolution of the DTM.Slope: slope map expressed in degrees.

2.4.5.The vegetation cover factor (C)
The C-factor (dimensionless) is based on the density and height of the vegetation cover of the ground surface.It is based on vegetation height, cover index, percent grass cover, and vegetation residue or litter.The C value for a completely covered field ranges from 0.001 to 1 for a fallow field (Wischmeier and Smith, 1978).The vegetation cover for this research was derived by supervised classification of a LANDSAT-8 satellite image extract recorded on July 18, 2019.NDVI For the normalized difference vegetation index, the R and PIR channels are combined.The normalized vegetation index vividly illustrates the distinction between the red and near infrared bands observed in the image. (6) The vegetation often has NDVI values between 0.1 and 0.7.The greater the value, the denser the cover must be.NDVI values range from -1 to 1, with higher values indicating greener vegetation and lower values indicating no vegetation, barren terrestrial and aquatic sources show negative NDVI values ( Karaburun, 2010).For the calculation of C , we used the following equation.(7) Where: α=2 and a β=1 value seem to give reasonable results

2.4.6.The anti-erosion practices factor (P)
The factor P represents the effect of cultural practices that allow for a reduction in the volume and speed of water, hence reducing ruissellement and, as a result, erosion.Because there are no data on antierosion measures available, the factor P was given a value of 1 over the whole surface of the bassin versant.

Results
The application of the Wischmeir and Smith (1978) equation in the Oued Amter watershed allowed us to estimate soil losses and map potential erosion.This evaluation is carried out by integrating all of the RUSLE model's variables (R, K, LS, C, and P) utilizing GIS spatialization tools and remote sensing.This allowed us to develop a spatialized soil loss map of the Oued Amter watershed.

The Erosivity Factor (R)
The rainfall study of the thirteen climatic stations showed an irregularity of precipitation in time and space (Fig. 6), with average annual precipitation varying between 1458 mm for the Jbel Outka station and 316 mm at the Targuist station (Table 2).IDW (inverse distance weighting) was used to create the rainfall-runoff erosivity map, we may forecast the erosivity of cells on the map that are unknown.This approach has the ability to restrict the impact of distant spots.2. Average annual precipitation in mm and factor R Fig. 7 shows the rainfall/stream erosivity map of the watershed as a result of using this method to determine the R-factor.Annual precipitation was chosen in this study because it is readily available, simple to calculate, and consistent across regions.When performing the R-factor analysis, we choose to use the yearly mean precipitation.Between 1637.09 and 1937.61MJ mm/ha/hr/yr is the range of the Rprojected factor's value.

The Soil Erodibility Factor (K)
The values of the soil erodibility factor K (Fig. 8) vary from 0.139 to 0.140 t ha h/ha/MJ/mm.Lower K-factor values were associated with high permeability soils, low antecedent moisture content and low water content soils.

The Topographic Factor (LS)
The LS factor is a topographic index that represents the morphology of the terrain.It is calculated from the Digital Terrain Model (DTM) by superimposing maps of the length of slopes L (m) and their degree of inclination S (%).The slope has an important influence on the process of water erosion.It aggravates the effect of rainwater runoff.Results for LS range from 0 to 309.The slope gets steeper and steeper as the flow builds up.Fig. 9 depicts that while the majority of basins have LS values in the range of 15 to 309, only around 10% of watersheds have such high LS values.Most of the high LS factors are found at the watershed network passage level in the easternmost reaches.

The vegetation Cover Factor (C)
The C-factor numerically expresses the results of the available vegetation cover for all land areas in the RUSLE model, whether it is agriculture or soil erosion mitigation techniques.The Normalized Difference Vegetation Index is a popular metric of vegetation growth in remote sensing (NDVI).This satellite image index was built using QGis and Equation 6.The C factor is determined by the kind and extent of vegetation cover.To put it another way, there is a wide range of values for the NDVI, ranging from 0 to 1. Negative NDVI values represent water surfaces, while positive NDVI values represent bare soil.For Oued Amter watershed, the RUSLE,Map C Factor (A) and NDVI Map (B) are shown in (Fig. 10).Can say that from the point of view of vegetation cover, the watershed of Oued Amter is not well protected, 28.66% of its area of the basin is constituted of Arboriculture.The surface that remains protected by natural vegetation is modest (Forest) and represents only 12.65% of the surface of the basin (Table 3).With a variable value of C from 0.07 to 0.48, therefore little protection against the impact of rain and showers.

The Factor of Erosion Control Practices (P)
Using soil protection and anti-erosion methods, the "P" factor reduces water erosion by reducing the speed of runoff.It depends on how far the project has progressed.When it comes to using any of the following techniques, the Wadi Amter basin receives a rating of 1.Based on the practice and slope, the P factor has different values.The research region does not have these safeguards in place.Because P is a constant (P = 1) ( Benzougagh et al., 2020).

Evaluation of Soil Losses (A)
The map of soil losses elaborated by multiplying the maps corresponding to the five (5) factors described above under GIS shows that the Oued Amter watershed presents a very important variability in terms of soil erosion, with values ranging from less than 10 tons per hectare to more than 2500 tons per hectare (Fig. 11).The upstream areas are the most prone to erosion, due in part to the uneven terrain and lack of vegetation cover.The class of soils with losses below the limit of 7 t/ha/year, considered as a tolerance threshold (Renard et al., 1997) occupy only 11% of the watershed territory.
In the Oued Amter watershed, losses are highly concentrated in the 10-300 t/ha/yr range, i.e. 93.08% of the territory.While the low losses below the tolerance threshold concern 6.92% of the basin area.A total of 31.36 percent of the basin's land area is affected by losses ranging from 0 to 30 t/ha/yr, those with 30 to 100 t/ha/yr occupy 28.14%, and those ranging from 100 to 300 t/ha/yr occupy 46.93%, and those exceeding 300 t/ha/yr occupy only about 0.19% of the territory (Table 4).
Thus, considering the tolerance threshold of 7 t/ha/year, we can see that the majority of the watershed is subject to significant erosion and to varying degrees 99.79% of the watershed.The losses thus generated represent a significant threat in the medium and long term of sedimentation at the level of the Oued Amter at the outlet of the watershed.These results are similar to those obtained for the Nakhla watershed, with an area of about 110 km 2 , this watershed is located in the Western Rif in northern Morocco (Northern Rif).This study is based on the use of the revised universal soil loss equation.The results obtained show that the rate of erosion is high in areas with steep slopes ranging from 12 to 40 degrees.Indeed, these areas with predominantly calcimagnesic soils suffer losses of about 95.5 t/ha/year.

Conclusions
The present work is carried out with the aim of quantifying the soil losses in the Oued Amter watershed using the Revised Universal Soil Loss Equation (RUSLE) and the GIS tool.The studied watershed extends along the internal Rifain domain, located in the region of Tangier-Tetouan-Al Houceima, Chefchaouen province in northwestern Morocco.It has a mountainous character with altitudes that vary between 0 m and 2100 m. (Jebel Tizirane) from North to South, it shows a diversity of reliefs, with structural forms, depressions, ravines ... etc.It is placed in a transitional position, it is essentially located between the internal domains and the flysch nappes its two domains are generally characterized by one another from the lithological point of view (nature of the substratum).Then we can synthesize that the nature of the substratum does not react in the same way to the presence of water (water erosion), from that we can distinguish that the flysch areas in general soft impermeable and semi permeable materials (marl and clay) which form the upstream part of our basin are affected by erosion and also present low reliefs (moderate and weak slopes).On the contrary, the zones of internal domain are of hard and resistant material, the existing reliefs are strong (extreme slopes).The effect of erosion in these areas is concentrated in gullies and talwegs.
Slope that varies from 0 to 58.58%.It is characterized by significant erosion, with variable 99.79% of the area of the basin exceed the tolerance threshold and only 0.19% of losses are less than 7t/ha/year.This is related to the different values of the factors involved in soil water erosion.The values of the erosivity factor R range from 1637.09 to 1937.61 with an average of 1787.35Mj.mm/ha.h. an (Millijoule.Millimeters/hectare. Hour.Year).Likewise, the erodibility factor K presents two values from 0.139 to 0.140 and it showed that the soils little evolved and the Vertisols are the most vulnerable to erosion.The LS factor related to the topography varies between 0 and 309.06 of which the upstream is more uneven and more sensitive.It can be said that from the point of view of vegetation cover, the watershed of Wadi Amter is not well protected, 28.66% of its area of the basin consists of Arboriculture.The surface that remains protected by natural vegetation is modest (Forest) and represents only 12.65% of the surface of the basin.With a value of C variable from 0.07 to 0.48, thus little protection against the impact of rainfall and showers, in general, the watershed is exposed to a great risk of water erosion.
The results obtained give an important idea for decision makers to know the areas at risk and propose adequate interventions for the fight against erosion from the in-depth description of each of the factors related to the erosion processes.There are some recommendations to be followed to control the risk of erosion such as adding to the universal soil loss equation of Wischmeier other models to calculate the amount of sedimentation produced at the level of the studied watershed as well as the rate of phosphorus and organic matter, and taking into account the relative importance of agricultural and non-agricultural pressures and the fragility of the environment and program interventions to stabilize the soil and improve water infiltration and decrease the topographic impact.

Fig. 1 .
Fig.1.Map of the geographical situation, administrative division, outlet point, hydrographic network and elevation in meters, of the Oued Amter watershed

Fig. 3 .
Fig.3.Map of the geological facies of the Oued Amter watershed.

Fig. 10 .
Fig.10.Map of the vegetation index by normalized difference (a), Map of the factor C at the level (b) of the Oued Amter watershed.

Table 1 .
Morphological characteristics of the Oued Amter watershed

Table 3 .
Types of vegetation cover in the Oued Amter watershed

Table 4 .
Area and percentage of soil loss classes in t/ha/yr.