**4. Discussion**

This study presents a practical and multidisciplinary approach that is essential for risk management of watersheds because it helps identify the areas and factors that are most vulnerable to erosion and coastal sedimentation. Similar problems remain practically unexplored in Malta.

Soil erosion is a significant environmental issue for Malta and is having serious consequences for the sustainability of local watersheds and the agriculture sector that depends on them. Among other things, it leads to increased sedimentation along the drainage pattern identified by this study, and ultimately to increased risks of flooding and landslides within the watershed basin itself. In addition, unless properly managed, changes in the watershed dynamics will negatively affect the sensitive ecosystem dynamics observed at the Ramla valley mouth that are critical to its normal functioning.

From a risk management plan of this highly sensitive ecological watershed, it is being recommended that the impact of soil erosion hotspots identified by this study (both within the watershed itself and resulting impact on coastal waters next to the valley mouth), are to be further monitored, and if possible, analyzed, and mitigated. Sustainable management of watersheds necessitates the use of appropriate hydrological, ecological, and socioeconomic management tools that unfortunately tend to operate, if at all, in mutual isolation. Such a thorough and holistic approach would support the much-needed management of local watersheds, which is particularly concerning and urgent in view of the need for Malta to adhere to a number of important European directives, such as the Water Framework and the Marine Strategy Framework Directive.

Future studies related to coastal water sedimentation can consider using other Earth observations platforms, such as OLCI and LANDSAT 9, to analyze more rainstorms of varying severities. Ideally, such studies would include improved algorithms to detect CHLA and TSM in coastal waters associated with watersheds. Such an integrated monitoring approach, involving both *in situ* and remote sensing analysis must also be able to give value to the validation of the entire soil erosion and sedimentation process at a highly localized scale.

It is hoped that the analysis and quantification of this phenomenon will contribute to an understanding of the dynamics of an important watershed that so far has never been studied from a soil erosion risk management point of view. The results could be used in future multi- and inter-disciplinary studies to better understand the impact of unsustainable land use practices both within watersheds and related littoral sediment budgets and to manage and develop future risk management plans aimed at minimizing soil erosion under a changing climate over the Maltese islands where evapotranspiration, rainfall, and temperatures regimes will change significantly over the next 50 years [19].

The study has a number of methodological and logistical limitations that stem from its multidisciplinary nature. In the case of the estimation of the degree of soil erosion, scientific literature on this topic is filled with the application of RUSLE for relatively large areas and therefore, the present estimates should be interpreted with care. The biggest challenge, however, concerns the *in situ* validation of the degree of soil erosion mapping estimations as identified by the 3D GIS visualization. Unfortunately, no data on actual soil losses exist at the local scale against which the present estimates can be compared.

The sedimentation processes identified in the coastal waters just outside the valley mouth can be affected by inaccuracies due to differences in resolutions used between the near-real-time Sentinel-2 datasets (having a ground resolution of 10 m) against the MERIS imagery (with a ground resolution of 300 m) used to derive climatological baselines. The use of Sentinel data at the finest resolution constituted the best freely-available data available that can be used for our soil erosion and sedimentation study at the Ramla valley mouth.

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

The processing of Earth observation data also had its own limitations. Firstly, restricted satellite visiting times have led to limited availability of data over the Ramla watershed, especially when studying limited, high-intensity, low-duration rainfall events. A further limitation on the methodology was posed by the presence of clouds in the satellite scenes, which tend to contaminate the sensor information acquired over any area of interest. Naturally, clouds are present during rainfall and so optical remote sensing of coastal waters can be quite challenging during, and exactly following rainstorms.
