**4. Results**

In this section, the results of the SWOT analysis were summarized, considering the two parts of RS potential and potential assessment.

## *4.1. Remote Sensing Techniques*

Strengths: Recently, ESA designed, developed, and launched a new family of satellites called Sentinel (includes S1 to S6) as a part of the Copernicus Program. Sentinel-1 (S1, SAR) is able to perform very detailed analysis in this area and would like to thank the different polarization modes: single polarization (vertical-vertical (VV) or horizontal-horizontal (HH) and dual-polarization (VV + VH or HH + HV) [65]. The VV polarization is very useful for detecting wind speed in ocean and also for understanding the di fferent kind of ocean and sea activities such as fisheries, ship routing, coastal surveillance, o ffshore installations and exploration. The main reason to select the VV polarization is because of its success in detecting wind speed, since this kind of polarization is sensitive to sea and ocean surface roughness (sea surface water). Images obtained at VV polarization by the SAR satellite are highly sensitive to wind speed variations by means of RMSE which is lower for VV polarization than HH polarization for Sentinel-1B [66]. For C band SAR images analysis, such as Sentinel 1, the C-band model CMOD (C geophysical model function) family (such as CMOD 4, CMOD 5, CMOD 5.n, CMOD 7) and a new model function called C-SARMOD2 can be used. Measuring surface roughness caused by wind is an important feature of the SAR images. SAR satellite capability depends on Doppler information to achieve good resolution in the along-track direction [31].

The Italian Space Agency, along with the Italian Ministry of Defense have developed the COSMO-SkyMed system which is the unique constellation of four radar satellites for earth observation. Those four radar satellites of COSMO-SkyMed system have advanced technology and uses high-resolution radar sensors to observe the earth's day and night, regardless of weather conditions with varying revisit time. Main theme areas are emergency prevention, strategy, scientific and commercial purposes, providing data on a global scale to support a variety of applications among which forest & environment protection, risk management, natural resources exploration, land management, maritime surveillance, defense and security, food and agriculture management. The COSMO-SkyMed satellites have main payloads of X-band, multiresolution and multi-polarization imaging radar, with various resolutions (from 1 to 100 m) over a large access region. Since, it is equipped with a fixed antenna, having electronic steering capabilities that could manage many operative modes for the image acquisition and for internal calibrations. The nominal incidence angles varied between 20◦ and 59◦.

SAR images have grea<sup>t</sup> potential for the observation, monitoring and detection of marine sources. It is important to have a wind parameters long-term reference analysis that can cover a large geographical scale. This is more important when many of the observations made from ocean tool measurements are scattered [67]. One of the most important reason to use S1 satellite data and software is because of its free access, supported by unlimited policy with just a sign-up before trying to download the images. However, the data are systematically provided by delivering within an hour of reception for near-real-time (NRT) emergency response, within three hours for NRT priority areas and within 24 h for systematically archived data. Images taken from satellites at di fferent frequencies are used to analyze and map wind parameters at of the seawaters. Recently, the OW field retrieval method has been developed based on satellite data sources and image processing techniques [68]. In many of the Mediterranean coastal, nearshore, and o ffshore areas, there are no observations of the tall tower for validation. Hence, maritime validation relies on reports of ships and buoys.

Not only SAR satellite imagery can provide OW field data with a long time-series of large and small zones, but also satellite imagery is playing a significant role in o ffshore observation and research. Even though sea-level wind measurements can be carried out using buoys in di fferent periods with very high resolution, these buoys are usually installed at a distance of 10 to 100 km from the shoreline indicating the possibility of the lack of access at greater distances [69]. Ocean winds recorded from scattering and radiometers have a higher temporal resolution [70]. This higher temporal resolution can be improved by using the cooperation of several satellite data. Due to the increase in the number of observations and time resolution, the accuracy of OW sources estimation can be gradually improved [71].

Furthermore, the ERA-interim reanalysis dataset is a reanalysis project dataset designed and developed by the ECMWF [37]. The ECMWF uses predicted models and data capture techniques, including 4D analysis with a 12 h analysis to describe the atmospheric parameters of the land and oceans such as wind speed, evaporation, surface pressure, surface roughness and surface net solar radiation [72]. In ECMWF dataset, the wind speed at 10 m height, 10 m U wind component, 10 m

wind speed and 10 m V wind component are available since 1979. The intended spatial resolution is characterized by monthly, yearly and four hourly intervals for each day [73].

Even though the vastness and infinity of wind energy in many parts of the world are having good wind power [74], the problems related to the determination of wind measurement accurately making it out the scheme more difficult. The availability of meteorological and floating data information especially in coastal areas, making the situation difficult by lack of access to the complete data set for a study area [75]. On the other hand, good data are available to identify suitable locations through meteorological models and satellite observations. It should be noted that the reanalysis data collects a complete form of data on terrestrial existence, for example, meteorological stations, buoys and cub anemometers, ships and satellite data, which can provide a more accurate display of wind resources on a scale. Such data are regularly monitored with high quality without delay (unlike floating devices: buoys and cub anemometers). In this case, reanalysis data showed the lowest overall error compared with buoys and ship data [76].

Due to the dramatic increase in the use of renewable energy around the world, the need to identify suitable areas for installation and the size of wind farms has increased. In this case, the reanalysis data can be used with a grea<sup>t</sup> ability to identify these areas. For example, at different altitudes, it can measure the wind for industrial use of the area, so we called industrial wind. Which can show itself in the view of an accurate wind atlas. Wind atlas contains different wind parameters in different regions, for example, maps, wind speed and direction, time series and frequency distributions. A wind atlas covers the average of important wind parameters at different altitudes for long periods that can come to governments, companies, and academic projects, for example (https://globalwindatlas.info/). An atlas map of offshore wind energy potential that considers all wind parameters can identify potential areas for the nearshore and offshore wind farms installation for later stages [77].

Opportunities: RS data and techniques guarantee a high level of reproducibility because S1 data (SAR) is free and has global coverage. To be more precise and accurate, SAR images with longer time intervals should be used. SAR satellites have several advantages, such as the high spatial resolution, the coverage of large areas, obtained in all-weather conditions, day and night (24 h) [78]. The ability to identify hot spots or focus on ROI makes it an interesting tool for preliminary analysis with different goals and by different target groups. On the other hand, the researcher can obtain data over a long period of time. Many researchers used the reanalysis dataset, which is long term time series of data on wind speed analysis, for network integration studies on wind potential areas [79]. The most important advantage of reanalysis dataset is that they are generally free. This type of information received from the global observation system are made up by different observations tools such as satellites, meteorological stations, and ships to cover a large area [80].

Weaknesses: There are several sensor options for measuring wind at sea and ocean waters using satellite RS, but those satellite and sensors like many other measurement tools, has limitations. For this reason, many researchers have used synthetic winds derived from multimeter scattering, radiometry, and reanalysis data [81,82]. In addition, mostly the assessments were confined to a single or dual scatterometer or using different data sources like reanalysis model, a single scatterometer data is limited to a specific period of time [81]. On one hand, SAR satellite images relative with ocean and sea water surface usually manifests expressions of atmospheric phenomena occurring in the boundary layer can be attributed to the phenomenon [83], such as boundary layer rolls, atmospheric convective cells, atmospheric internal gravity waves, tropical rain cells [84]. Furthermore, SAR satellites have a limit on scanning different areas, which can include one or two scans per week and/or day and are not universal. Much of these satellite data can be accessible with some restrictions of user need (users need a proposal), such as TerraSAR-X, COSMO SkyMed and Radarsat-2. These restrictions also include time constraints with the launching time of satellites. The accuracy of offshore wind assessment using SAR satellite imageries may be affected by different reasons, such as land contamination and lower water depth in the coastline areas [56].

Other factors such as image quality, in-orbit time of the satellite and also hard targets such as oil spills, land, islands, ships, WTGs and WECs are limiting ability to measure wind at the ocean and sea areas from the satellite's imageries. Especially while researchers using the SNAP software, all these hard targets should be masked at the first stage of the processing. For example, in SAR image processing for wind field estimation, the studied areas should be checked out for oil spill and removed from the image because of decreasing the water roughness of the sea surface [85]. Another important weaknesses, the wind direction (SAR images can be analyzed in SNAP software to achieve wind direction with 180 degree ambiguity) obtained from the SAR satellites cannot be verified because the available data will not really show the wind field in the coastal and open seas. This is the main reason for selecting local data (in situ data) in many studies. Another limitation of wind speed estimation based on satellite data are that they measure wind speed data at 10 m above sea surface water. By using this type of data to obtain information about the wind sources at a height of 100 hub meters, it is necessary to implement theoretical models using surface roughness coe fficient [76].

Threats: Interdisciplinary knowledge is needed to work with RS methods as this involves obtaining and analyzing satellite data. Then, researchers need to be familiar with various software applications to properly analyze satellite data. It is known that satellites are highly capable of observing the Earth, since they have applications in many di fferent fields. Hence, studying the specific application mainly depends on the software knowledge of researchers.
