Application of the Erosion Potential Method in Vithkuqi Watersheds (Southeastern Albania)

Soil erosion is one of the most important phenomena affecting land composition and settlement. Among all natural causes of soil erosion such as rainfall intensity, temperature and wind, the human activity; massive deforestation and intensive agriculture, including the latest climate changes are considered as very important factors, especially nowadays. Thus, calculating the soil erosion coefficient appears very important in order to prevent the phenomena. Many methods are used to calculate such coefficient but in the presented research, the Erosion Potential Method was chosen. In this study, eight watersheds in southeastern Albania were evaluated. Results show that erosion is present in all considered watersheds. In one case (Panariti watershed) the erosion coefficient was very high; excessive, while in others it varies from heavy to very slight erosion. In conclusion, it can be stated that the Erosion Potential Method can be applied in the Albanian contest, same as in other neighbor countries. The results from Panariti, Roshani and Gianci should be further investigated due to the high quantity of soil eroded.


INTRODUCTION
Quantitative displacement and transportation of different portions of the land resulting in soil degradation are commonly considered as erosion [Joy et al. 2002]. Intensity of soil erosion phenomena is mainly dependent on natural factors and human influence. Progress of such phenomena can be defined as slow, high or even very high in the cases where the factors causing it are very consistent. Landscape properties which support erosion, such as rainfall, wind and temperature changes, can be listed as natural factors. However, nowadays climate changes are thought to be one of the main natural factors affecting soil erosion [Borelli et al. 2020; Nearing et al. 2004].
On the other hand, the intensive agricultural use of the land, large deforestation process, the increasing number of the population and other human activities represent the main factors in . Furthermore, sediment yields may cause major damages, such as influencing the flow rate when they settle at one final point. The complexity of the land degradation causes has made it difficult to predict the exact impacts on soil erosion. Considering all these factors, generation of land erosion and sediment yield maps would be an important step to oppose the process. The methodology for evaluation of sediment yield and erosion has been described by many authors and authorities. The [Gavrilovic 1988].
The EPM method was initially proposed by Gavrilovic for former Yugoslavia conditions and then applied in many similar situations [Gavrilovic 1988]. The method aims to estimate the amount of sediment production and transportation by indicating the areas potentially threatened by the erosion phenomena. The methodology proposed by Gavrilovic represents a semi quantitative analysis that can be applied in arid and semi-arid areas to estimate erosion. This approach has been widely used to estimate the erosion in many countries [Haghizadeh et al. 2009 In Albania, the erosion phenomena are consistently present, especially in specific regions where natural conditions and are most affected by human activity [Marko et al. 2011]. It is presumed that application of such methodology in specific areas of our country can help in identifying the erosion scale and in preventing the phenomena to prosper. Since in Albania and in the specific area proposed the most significant issue is the erosion caused by the water, this phenomenon will be focused on in this study.

Model description
The Erosion Potential Method (E.P.M), was designed by Gavrilovic [Gavrilovic 1988] and used for the estimation of sediment production and transportation, as well as the erosion coefficient and its classification. This methodology was firstly applied to the erosion problems in Serbia, then in most of the Balkan countries, and nowadays worldwide, e.g. Italy, Switzerland, Croatia, The annual volume of detached soil W a (m 3 /year) has been determined using the following equation: where: S -is the watershed area (km 2 ); T -is the temperature coefficient (-); h -is the mean monthly precipitation (mm); Z -is the erosion coefficient (-).
The temperature coefficient T (-) depends on the mean annual temperature t (°C), and has been calculated using the following equation: The erosion coefficient Z (-) has been estimated using the following equation and the detailed information of Table 1.
where: x (-) indicates the protective nature of the land cover and is a function of land use; y (-) describes soil erodibility and is a function of geological characteristics; φ (-) shows the observed active erosion processes; i m (%) is the mean slope of the studied area.
The total volume of sediments produced does not fully reach the outlet. A portion of it is redeposited in streams or other areas of the basin; therefore, it is important to calculate the real sediment production G (m 3 /year) by the following equation: where: W a (m 3 /year) is the annual volume of detached soil, and Dr (-) is the sediment delivery ratio, which represent the quantity of sediments that reach the downstream.
The equation for Dr, is as follows: where: H -is the mean height distance of the basin (or sub unit), (km); P -is the perimeter of the basin (or sub unit), (km); L -is the length of the basin (km)

Study area
The Vithkuqi area is located in the Korca district, at the South-East of Albania ( Figure 1a). The study area is one of the most typical areas of Albania in terms of erosion development phenomenon due to its slope, especially in the upper part of the watersheds, and degradation of forest vegetation. The study area with a total surface of 22392 ha or 223.92 km 2 is divided into eight watersheds: Panariti, Shera, Rungaja, Katundi, Vithkuqi, Lubonja, Gjanci, and Roshanj, as it is shown in (Figure 1b). The data regarding the surface and perimeter of each watershed is given in Table 2.
The application of the Erosion Potential Method in this area were based on the data gathered from different field surveys and satellite sources.
Geological maps of the Albanian Geological Service were used to determine the geological structure and the coefficient of soil erodibility (y) of the studied area, as it is shown in Table 2.
The information about land use and cover type of each watershed was obtained using CO-RINE Land Cover maps. For the evaluation of the land cover coefficient x (data on Table 2), the coverage level map was used (Figure 1d).
Another important parameter for the determination of the erosion coefficient is the mean slope of the watershed i m (data on Table 2), which was derived from the slope maps of each watershed generated by a Digital Elevation Model (DEM).
The mean elevation of the watersheds, varying from 1140 to 1700 m a.s.l., was derived from the elevation map of the studied area, as it is shown in Figure 1f.
The spatial distributions of precipitation and temperature, expressed as monthly rainfall (h, mm) and monthly temperature (t, °C), for the time period 1990-2018, were obtained from the meteorological stations of the area, situated at 1250 m a.s.l. and 1600 m a.s.l., respectively, as it is shown in Table 3.

RESULTS AND DISCUSSION
In the current work, a very large dataset comprising the surface and perimeter, the coefficient of soil erodibility, land cover, mean slope and mean elevation of the watersheds was collected using sources from a variety of academic, governmental and non-governmental institutions.
Detailed information about the above mentioned dataset, used for the application of Erosion Potential Method in Vithkuqi area is given in Table 2.
The study area, almost all of its surface, is crossed by the Osumi River, in the West-East direction. This river comprises the water inflows of all watersheds of the area, which are also its main source suppliers. Some of the suppliers are the Lisec stream, Qafe Kumbulla stream, Shen Thanas stream, etc. The map of the hydrographic network of the studied area is shown in Figure 1b and its density varies from 1.97 km/km 2 to 5.77 km/km 2 .
The geological structure of Vithkuqi area consists mainly on magmatic rocks of the Jurassic On the basis of the information gathered from the maps of the coverage level, forest covers the largest area (42%), followed by the pasture (26%), bare soil (15%), cultivated land (12%),  bare rocks (4%), and urban area covers only 1% of the Vithkuqi area. The data collected in the field showed that the main tree species covering the land are Pinusnigra, Abies alba, Fagus sylatica, Quercussp etc. According to the climatic indicators of the area, the Vithkuqi area is classified in the mountain Mediterranean climate zone, South-East sub-zone, characterized by relatively high temperatures during summer, and quite low during winter (Table 3), as well as low rainfall during the summer period, and higher during the months of November -February (Table 3), in most cases as snowfall. Average annual temperatures and precipitation of the area are respectively: 13.2 °C and 1174.5 mm for the meteorological station at 1250 m a.s.l.; and 11.1 °C and 1443.4 mm for the meteorological station at 1600 m a.s.l.
As mentioned previously, all the parameters of Tables 2 and 3 were used for the application of the Erosion Potential Method (explained in section 2.1), obtaining the results for the erosion coefficient, the amount of eroded sediment, the sediment delivery ratio, and the sediment yield of the watersheds. Table 4 present the results obtained from the calculations performed according the equations 1, 2, 3, 4, and 5 of the Erosion Potential Method. Moreover, the results about the specific eroded sediment per each watershed, calculated as report of eroded material and the surface of the watershed expressed in ha, are shown in Table 4.
As it is shown in Table 4 On the basis of other studies, the parameter which mostly affects the value of the eroded sediment W, is the coefficient of soil erodibility y, where the Panariti watershed with the largest amount of eroded material has the highest value, equal to 1.5 as can be seen in Table 2. According to the sensitive analyses performed  In this study, the specifi c eroded sediment values E were also calculated. The Panariti watershed has the largest amount of eroded sediment per hectare per year, followed by the Roshanji, Gjanci, Katundi, Rungaja, Lubonja, Vithkuqi, and Shera watersheds. It was already mentioned that the results for this parameter (E) are calculated as a report of the amount of eroded material and the surface of each watershed. The results for the specifi c eroded sediment (E) are not in the same order as those for the eroded sediment (W), due to the fact that even the surface of the watersheds does not follow that order. Considering the results obtained by the calculations of specifi c eroded sediment (E), it is reconfi rmed once again that the Panariti watershed is the area with the highest risk of erosion.
Since these watersheds have diff erent sediment delivery ratio values, as it is shown in Table 4, their sediment yield is 240800.  Figure 2 for all watersheds of Vithkuqi area.
As it is shown in Figure 2 and in the absolute values of Table 4, there are differences between watersheds in terms of sediment yield and eroded material. The connection that exists between Dr and the transported material G is clearly noticed, since the watershed which has the largest amount of deposited material is exactly Shera (98.96% of eroded sediment) which has the highest value of sediment delivery ratio Dr, followed by Rungaja (97.68%), Vithkuqi (93.37%), Lubonja (91.47%), Gjanci (88.45%), Panariti 88.36%), Katundi (87.77%) and finally Roshanji (64.74%). Dragicevic et al. [25] conducted a sensitivity analysis of the EPM method and classified the sediment delivery ratio Dr as a very high sensitive parameter that affects only the sediment yield G. It is also worth mentioning the contribution of another parameter affecting the sediment yield of the Shera watershed, which -having the smallest amount of eroded material, has transported almost all of its quantity as it has the lowest land cover coefficient Xa. As it can be seen from the data used for the EPM evaluation, the first three watersheds (Shera, Rungaja and Vithkuqi) that have transported almost the entire amount of eroded material, are precisely the ones with the highest altitudes above sea level and with the most pronounced slopes. As mentioned above, the watershed that has transported the smallest amount of eroded material is Roshanji which has the smallest Dr, as it has the lowest value of both mean elevation (1140 m a.s.l.) and the perimeter of the watershed, only 3.2 km.

CONCLUSIONS
In this work, the application of the Erosion Potential Method in an Albanian watershed was proposed for the first time. The EPM method provides and estimates the amount of sediment production, specific eroded sediment, the sediment yield, erosion coefficient and the erosion intensity and risk. The EPM method was applied to 8 watersheds of Vithkuqi area, in southeastern Albania. The results for the study area show that the overall sediment production is 606578.2 m 3 / yr; the overall specific eroded sediment is 210.12 m 3 /ha/yr, and the overall real sediment production 540994.4 m 3 /yr. The major contributor in all the three values obtained from the applications of the EPM method is the Panariti watershed. The erosion coefficient Z of the watersheds was calculated and it varies between 0.11 (Shera watershed) and 1.56 (Panariti watershed). According to the Gavrilovic classifications, the study area shows excessive erosion (I erosion category) for the Panariti watershed (Z = 1.56); heavy erosion (II erosion category) for the Roshanji (Z = 0.95) and Gjanci (Z = 0.78) watershed; medium erosion (III erosion category) for the Katundi (Z = 0.66), Lubonja (Z = 0.58), Rungaja (Z = 0.57), and Vithkuqi watershed (Z = 0.53); and very slight erosion (V erosion category) for the Shera watershed (Z = 0.11).
In conclusion, it can be said that the application of the Erosion Potential Method is feasible for the specific study area. Moreover, this method can be applied in other country areas presenting the same situation contest with erosion problems. Results from this paper can be used by the policy makers in order to take measures for the prevention of the erosion in the study area, and not only. However, further studies focusing on other areas of the country with high erosion risk should be performed in order to better determine the efficacy and applicability of the proposed method.