**2. Case study: The Ramla watershed in Gozo**

The Ramla watershed in Gozo is roughly a 6 km<sup>2</sup> catchment area extending from its watershed divide down to the Ramla beach (**Figure 1**). The socioeconomic and environmental importance of this sensitive watershed is based on intensive farming activities [4], the presence of perched aquifers along its slopes (MT015 & MT016. [5]), unique biodiversity [6], and the presence of a NATURA 2000 site [4, 7].

According to Ref. [8], the Ramla watershed has formed through the fragmentation and dislodgment of the upper coralline plateau along the edges of the headlands overlooking, which is being sustained by the erosion of the underlying blue clay strata. The detachment of rock mass of various sizes then continues to fragment and weather into small pieces. Its geomorphology is characterized by somewhat steep relief, with elevations ranging between 120 and 28 m above sea level and an average slope of 18.1%. Its climate features mild, humid winters and hot, dry summers an annual mean air temperature of 18.6°C, and a mean precipitation of 574 mm [9]. The rainfall regime is intermittent with baseflow from October to January.

The Ramla watershed has been much affected by human activities. Agriculture has evidently shaped the watershed into a system of terraced fields that are delineated by rubble walls to define and protect the land parcels (**Figure 2b**). However, the area still continues to show natural depositional features including gullies and rills, as well as its principal watercourse and associated tributaries. The slopes are affected directly by rainfall resulting in surface runoff in the form of stream flow, which at higher slope elevations of the Ramla catchment removes the weathered material of the clay and makes the surface of the slope smoother. Splash erosion can happen when clay particles are moved about by raindrops. This becomes significant over the barren parts of slopes.

### **2.1 Methodology**

#### *2.1.1 Morphometric analysis*

National 1 1 m DEM derived from LiDAR and 15 cm pixel resolution aerial orthophotographs [10] were used for morphometric analysis in a GIS environment. The terrain analysis approach of SAGA GIS V8.5.1 [11] was used to derive morphometry characteristics of the watershed. A number of specific software tools were used: channels [12], hydrology, and morphometry [12]. These tools extracted the presence of watersheds and sub-watersheds in an automated manner in the form of GIS vector layers. For a proper determination of flow direction and flow accumulation, sinks were identified from the DEM and filled. After the drainage networks were extracted and sub-watersheds were delineated, four sets of morphometric parameters were calculated using the mathematical formulations as described in **Table 1**.

The stream order is the primary step in quantitative interpretations of a drainage network, and the stream number is the number of streams in each order. These two

#### **Figure 2.**

*(a) Topographic relief of the island of Gozo showing the area of interest (inset), (b) onsite agricultural land cover of the watershed taken from south-eastern watershed divide, and (c) analytical Hillshading derived from LiDARderived elevation data. (Source: Authors).*


#### **Table 1.**

*Parameters derived from the morphometric analysis.*

*Soil Erosion Risk Analysis of a Small Watershed DOI: http://dx.doi.org/10.5772/intechopen.111424*

parameters produce a geometric relationship that resulted in the calculation of the stream length of the drainage that was based on the surface runoff characteristics of the basin. The total length of individual stream segments of each order was, thus, derived, while the extracted Mean stream length provides information of the drainage size and its contributing basin surface. The terrain ruggedness index (TRI) was also calculated by calculating the sum change in elevation between a grid cell and its eight neighbor grid cells, where xij is the elevation of each neighbor cell to cell. The watershed basin characteristics, the channel network, and sink route were subsequently derived from the DEM to produce a basin shapefile, along with other shapefiles [12].

#### *2.1.2 Soil erosion*

Soil erodibility within the Ramla watershed was estimated using the RUSLE [16]. The soil erodibility was estimated using the following equation:

$$\mathbf{A} = \mathbf{R}^\* \text{ K}^\* \text{ L} \mathbf{S}^\* \text{ C}^\* \text{ P.}$$

where A is the estimated soil loss in tons per hectare per year, R is the rainfall erosivity factor, K factor is a measure of soil erodibility, LS is the slope length and steepness factor, C is the cover and management factor, and P is the support practice factor.

### *2.1.3 Estimation of CHLA and TSM*

The derivation of the climatological profile of CHLA and TSM was based on the analysis of 10 years' worth of monthly 300 m MERIS L1B sensor data derived by ENVISAT-1. The observation period ranged between May 17, 2002 until April 8, 2012, and is equivalent to 3614 days for the area of interest (**Figure 3**). The algorithm used to derive CHLA (in mg.m�<sup>3</sup> ), and TSM (in g.m�<sup>3</sup> ) was based on Case 2 water monitoring [17] with a spatial resolution of 0.3 km/pixel data products. Average concentrations of CHLA were derived using the OC4 algorithm and TSM [18]. The monthly climatological processing of the CHLA and TSM was conducted using MATLAB coding.

For this case study, near real-time, high-resolution COPERNICUS Sentinel-2A and 2B (L1C) imagery overpasses were used to derive CHLA and TSM quantitative maps that coincided with the rainstorm events observed between the January 2 and 12, 2021. Prior to processing, images were resampled using Band-2 due to its high spatial resolution (10 m). Salinity and temperature parameters needed for algorithm processing were set to reflect the values observed in Maltese coastal waters based on their monthly averages.
