Abstract
The present study analyzes the wind energy potential of Qatar, by generating a wind atlas and a Wind Power Density map for the entire country based on ERA-5 data with over 41 years of measurements. Moreover, the wind speeds’ frequency and direction are analyzed using wind recurrence, Weibull, and wind rose plots. Furthermore, the best location to install a wind farm is selected. The results indicate that, at 100 m height, the mean wind speed fluctuates between 5.6054 and 6.5257 m/s. Similarly, the Wind Power Density results reflect values between 149.46 and 335.06 W/m2. Furthermore, a wind farm located in the selected location can generate about 59.7437, 90.4414, and 113.5075 GWh/y electricity by employing Gamesa G97/2000, GE Energy 2.75-120, and Senvion 3.4M140 wind turbines, respectively. Also, these wind farms can save approximately 22,110.80, 17,617.63, and 11,637.84 tons of CO2 emissions annually.
Introduction
The growth of interest toward sustainability has increased countries’ and researchers’ interest to impulse renewable energy forms. This approach’s overall goal is to reduce greenhouse gas (GHG) emissions generated daily worldwide. A common and ever-growing renewable form of energy is wind technology. This method used to harness the wind’s kinetic energy has improved its efficiency throughout the years while lowering its cost; this increase has been translated into a market growth from 91,632 up to 513,547 MW of installed capacity between 2007 and 2017 (Key Statistics, n.d.; IRENA International Renewable Energy Agency, 2018).
Unlike other renewable technologies like geothermal or tidal, wind energy is not highly limited to specific locations worldwide. Its potential may be higher or lower depending on the site; with this in mind, feasibility studies have been carried out worldwide. Countries like Austria, Algeria, Egypt, Greece, India, Jordan, Nigeria, USA, Thailand, and Turkey, which present a diverse range of geografic locations and wind profiles, have done studies related to implementation of wind turbines. However, the outcomes were not always the same, offering possibilities to install small/medium scale farms (in Turkey) to large installations in Greece (Adekoya and Adewale, 1992; Ammari et al., 2015; Boudia et al., 2016; Draxl and Mayr, 2011; Fyrippis et al., 2010; Gökçek et al., 2007; Hamed et al., 2016; Janjai et al., 2014; Khan et al., 2019; Quan and Leephakpreeda, 2015; Ramachandra and Shruthi, 2003). Similarly, studies in the Gulf Cooperation Council (GCC) region have also been performed. The previous findings highlighted that the Arabian Gulf central region (Bahrain, Qatar, and Saudi Arabia) presents adequate wind potential to install wind farms up to a large scale depending on its location (Al-Salem et al., 2018; Elgabiri et al., 2021).
Few studies focused on Qatar’s potential to produce electricity by implementing wind energy. These analyses concluded that there is a potential to harness the wind’s kinetic energy and generate green electricity. This conclusion was made even though the analyzed data was taken between 10 and 25 m above the ground (Alnaser and Almohanadi, 1990; Moghbelli et al., 2011). This potential can increase with altitude since wind speed increases with elevation. Moreover, other studies highlighted the economic relevance of using this technology in the country, indicating that a wind farm’s energy production could be 15% lower than a traditional power plant (Marafia and Ashour, 2003). Also, since this type of energy could substitute a part of gas power plant electricity production, a decent environmental and economic benefit can be obtained by implementing wind technology, saving up to 3.32 million US$ by implementing a 5 × 3.4 MW wind farm (Méndez and Bicer, 2019).
Based on the studies mentioned above, the present research focuses on the necessity to generate a site-specific analysis for Qatar, including the generation of a wind atlas and a Wind Power Density (WPD) map highlighting the best locations to harness the wind’s kinetic energy. WindSim 10 software is employed to create these maps. To carry out this task, 40 measurement points including hourly wind speed, direction, temperature, and atmospheric pressure data between January 1979 and August 2020, obtained from ERA-5, are used. Furthermore, it is relevant to indicate that the outcome is based on wind speed at 100 m above ground level, indicating a more realistic potential based on commercial turbines’ hub height. Figure 1 indicates the steps taken to generate the wind map using WindSim software.
Wind maps generation flowchart.
Finally, to complement the wind atlas and WPD maps, three wind turbine types are used to model three scenarios of a case study with ten wind turbines as a result, the overall electricity production of these wind farms is obtained, and the best location option is presented.
Terrain
As shown in Figure 2(a) and (b), Qatar’s terrain is relatively flat. It does not present significant changes in its behavior, with a maximum elevation of 63.767 m in the north and 136.99 m in the south sector. Furthermore, the height difference is gradual, and there no abrupt presence of any mountain in any location. This element is relevant to the analysis because it indicates that there will not be a lot of turbulence generation in the area, meaning that there will be a minor influence on the wind speed based on terrain distribution. Furthermore, it also enables the possibility of locating a wind farm without significant installation difficulties associated with rugged terrains.
Qatar’s terrain elevation: (a) north and (b) south.
Wind potential
To create the wind maps, the measurements from the 38 points indicated in Table 1 are used.
Environmental measurement locations.
Wind speed
From the measurements taken at the locations indicated in Table 1, Figure 3 is generated. In such an illustration, two statistical methods are implemented: wind speed recurrence and Weibull distribution. The latter is implemented to determine the wind regime’s characteristics. This method is focused on two fundamental elements are calculated, these are the shape (k) and scale (A) parameters. To determine these parameters, diverse methods could be implemented, such as wind variability, standard deviation, power density, Moroccan method, Weibull probability density function, and so on (El Khchine et al., 2019). In the present study, these parameters are calculated by implementing the WAsP method; this technique employs an iterative approach to calculate k via equation (1) followed by the use of such result in equation (2) to obtain A (Bingöl, 2020).
where:
Up is the proportion of the values above the mean.
U is the mean wind speed (m/s).
k is the shape parameter.
A is the scale parameter.
After implementing both statistical methods, the resemblance between these results indicates that wind speeds between 3 and 9 m/s present the highest probability of occurrence throughout all the measurement points.
Wind frequency and Weibull distribution at 100 m per measurement location.
It is imperative to indicate that the results reflected in Figure 3 are presented at the same height as the measurements (100 m above ground), meaning that no data modification is required. However, it is relevant to indicate that wind speed is greatly influenced by height and terrain roughness. Meaning that for calculations regarding wind turbines with hub heights other than 100 m, the following expression was used to determine the wind speed at the desired altitude (Hernández and Ortega, 2014):
where:
V is the wind speed at height H (wind turbine hub height).
V 0 is the wind speed at height H0 (100 m in this case).
α is the friction coefficient or Hellmann Index.
The friction coefficient is everchanging, and it depends on the studied terrain, as indicated in Table 2. For Qatar, this value has a minor shift in its behavior due to the small elevation modifications throughout the country; however, most of the surface presents values under 0.15.
Friction coefficient (Fernández, 2002).
Wind direction
Since wind energy is always fluctuating and the primary purpose is to harness this natural resource in the best way possible, turbine distribution and orientation are also important. The airflow path is determined by implementing a vane measurement instrument, enabling the creation of a wind rose plot for each point. This plot indicates the direction in which the air is flowing in the form of bar extensions originated from the circumference center. Each element’s length indicated the recurrence in which a certain wind speed moves in a particular direction, meaning that a bar with an alignment to the south shows that the wind will flow toward the north (Ahrens, 1998; Méndez and González, 2009).
Figure 4 highlights that the wind is mainly originated from the North-West (NW), moving toward the South-East (SE) of the country. It is also visible that, in southern Qatar, there is a broader range of dispersion; however, the SE is also the most predominant direction.
Wind rose plot per measurement location.
Wind atlas
Once analyzed the terrain and wind parameters, these are merged with other parameters such as temperature and atmospheric pressure creating the wind map reflected in Figure 5(a) and (b).
Qatar’s average wind speed profile on land at 100 m: (a) north and (b) south.
Figure 5 reflects that minor wind speed changes exist throughout the country, with values between 5.2662 and 6.2329 m/s. It is also visible that northern Qatar has a broader presence of the wind with velocities above 6 m/s, meaning that it has more potential than the south side of the country. Furthermore, as a reference, the measurement points used to generate the map are also visible in Figure 5(a) and (b). To associate the wind profile generated with the actual surface and observe if the locations where the wind speed has its highest values are available, Figure 6(a) and (b) illustrates the wind map over Qatar.
Qatar’s average wind speed (on land) map at 100 m: (a) north and (b) south.
Since Qatar’s borders include an area of the Arabian Gulf, the offshore wind potential is also considered. Figure 7 indicates how Qatar presents its highest values in the northern zone (Figure 7(a)), including an extensive area of approximately 224,612.37 hectares with values between 6.2 and 6.3 m/s. Moreover, by comparing Figure 7(a) with Figure 7(b), a pattern is noticeable, wind speed tends to present lower values the further it gets from shore, specifically toward the South-East region.
Qatar’s average wind speed profile offshore at 100 m: (a) north and (b) south.
Similarly, as done for the onshore wind profile, Figure 8(a) and (b) reflects the wind map over Qatar’s offshore area. This figure corroborates that the wind speed gets lower toward the South-East.
Qatar’s average wind speed (offshore) map at 100 m: (a) north and (b) south.
Wind power density
Wind speed is an excellent way to observe the potential that a particular area has to produce energy. However, the WPD showcases the capacity of energy production per swept area (W/m2). Figure 9(a) and (b) indicates that Qatar presents values between 149.46 and 267.08 W/m2. Moreover, similarly to the wind profile, Qatar’s northern side offers a higher WPD potential, including a large sector with values above 240 W/m2. Furthermore, Figure 10(a) and (b) presents a similar representation as to the onshore wind map (Figure 6), confirming that northern Qatar offers a higher potential than the south.
Qatar’s average wind power density (W/m2) profile on land at 100 m: (a) north and (b) south.
Qatar’s average wind power density (W/m2) (on land) map at 100 m with some GIS data: (a) north and (b) south.
Regarding the offshore area, Figure 11(a) and (b) indicates that Qatar presents wind power density values between 210.02 and 335.06 W/m2 offshore, with the north as its most appealing sector. Furthermore, Figure 12(a) and (b) reflects how the WPD potential decreases toward the South-East area of Qatar’s border in the Arabian Gulf, in parallel with the wind speed behavior in the sector.
Qatar’s average wind power density (W/m2) profile offshore at 100 m: (a) north and (b) south.
Qatar’s average wind power density (W/m2) (offshore) map at 100 m: (a) north and (b) south.
Site selection
Based on the information above, Qatar’s north area presents better potential than the south, both on and offshore. With this perspective, Figure 13 shows the zoomed-in version of the best onshore location, while Figure 14 highlights the site with more potential offshore. Both areas are large and seem to have most of their extent free of any construction, meaning that these locations are initially available to be used for the construction of a wind farm. The onshore option remains uncertain until the owner of such land is located and a formal agreement is made to use it. It is relevant to indicate that the area in which a wind farm is constructed may be used for other purposes, implying that only the area in which the wind turbines, underground wiring, and substation are located is indisposed. Hence, other activities like farming or the construction of parks, roads, and even a sustainability museum may be carried out without significant issues.
Location with the highest average wind speed (m/s) in northern Qatar (on land).
Location with the highest average wind speed (m/s) in north eastern Qatar (offshore).
The offshore location does not depend on a potential single owner. However, the area is located near a port, meaning that extensive vessel transit may be expected, and by building a wind farm, this route may be affected. Furthermore, this area is located above a large natural gas field (Méndez and Bicer, 2019), implying that it is needed to locate such natural gas wells to extract such resources in the future. However, this location should not be discarded entirely and future studies analyzing vessel transit, and potential well locations must be integrated to determine the viability of installing an offshore wind farm.
Regarding the wind recurrence and direction, Figure 15 indicates how both locations present a high repetition between 2 and 9 m/s. However, Figure 15(a) shows that wind speeds above 2 m/s offer a higher frequency of occurrence on land than offshore, which presents higher frequencies for values below 2 m/s (see Figure 15(b)). Moreover, offshore delivers an average wind speed of 6.25 m/s, which is just 0.17 m/s (or 2.72%) higher than the onshore location, which has a mean of 6.08 m/s. This difference is generated by a higher presence of sporadic (with a frequency under 1%) wind speeds above 14 m/s.
Wind frequency and Weibull distributions at 100 m in the best locations: (a) onshore and (b) offshore.
Similarly, the wind direction in both locations has a resemblance between them, as indicated by Figure 16(a) and (b). Despite existing a wide dispersion of wind orientation, there is a clear tendency for the wind to flow toward the SE, in which the onshore location presents a slight increase in frequency toward the south (Figure 16(a)). This lower southern inclination is translated into an even more dominant SE tendency for the offshore location. However, wind turbines can rotate on their yaw axis to catch the prevailing wind direction, enabling the possibility of harnessing this resource despite a momentary 45° shift in wind direction. For these reasons, for both locations, it is advised that the wind farm should face NW.
Wind rose plots in the best locations: (a) onshore and (b) offshore.
Based on the analysis made above, there is less than 3% and 12% differential between both locations’ average wind speeds and WPD values, respectively. Moreover, the higher cost that offshore wind farm represents, the land-based option is initially chosen to carry out the study of Qatar’s implementation of a possible wind farm.
Energy generation
There are several ways to calculate the energy production from a wind turbine, all of which involve its power curve and the turbine’s characteristics. A simple approach is to match the desire wind speed with its value in the power curve; however, this approach depends on the density conditions under which this curve was generated. Another form to obtain the wind turbine’s output is to calculate the power generation based on the power coefficient (Cp). To do so, equation (4) is applied (Sedaghat et al., 2016).
where:
ρ is the air density (kg/m3),
AR is the turbine’s rotor swept area (m2),
V is the wind speed at the hub height (m/s),
Cp is the turbine’s power coefficient.
In the present study, each wind turbine’s output is determined by its power curve; however, WindSim software creates an adjustment based on the selected area’s air density. On this note, it is relevant to indicate that the chosen turbines have their power curves based on a value of ρ = 1.225 kg/m3. Moreover, the software makes an adjustment based on ρ = 1.161 kg/m3, which is the average air density in Qatar (Méndez and Bicer, 2019, 2020).
Wind turbine selection
To select an appropriate wind turbine, several elements are considered. First, the cut-in wind speed must be low (around 3 m/s); additionally, it must present a hub height over 100 m to use as much as possible the available wind potential. Finally, the turbine must take advantage of the most recurrent wind velocities, which (as indicated above) are between 2 and 9 m/s. Based on the criteria mentioned above, the wind turbines indicated in Table 3 are chosen.
Selected wind turbine models (Welcome to wind-turbine-models.com, 2021).
Figure 17 highlights both the Cp and power curves of the three turbine models. This plot reflects that all the models have a cut-in wind speed of 3 m/s; however, the Senvion option can generate energy up to 22 m/s. Furthermore, the Gamesa turbine requires 14 m/s to produce its nominal capacity, while the GE Energy and Senvion need 12 and 11 m/s, respectively, to do so. Additionally, option three can generate from 77.60 kW up to 2868 kW with wind speeds between 3 and 9 m/s, representing a value approximately 32% higher than the closest alternative.
Selected wind turbines power curves (Welcome to wind-turbine-models.com, n.d.).
Based on the power curves, the Senvion 3.4M140 is the most viable option; however, to confirm such a selection, all three options’ performance are analyzed. For this reason, a wind farm conformed of 10 turbines is simulated, and three scenarios (one per turbine model) are compared. Considering the results obtained in previous sections, all the turbines face NW, and they are distributed as indicated in Table 4.
Wind turbines distribution at the best-selected location.
Since the designated area is wide enough, all 10 turbines may be placed in a single line, lowering the wake effect’s influence. Using the Senvion model as a reference, Figure 18 highlights how the wind speed reduction due to the wake effect does not significantly influence the wind velocity that any turbine is facing.
Wake effect influence in the wind farm (wind speed variation).
Regarding energy generation, Table 5 reflects the Annual Energy Production (AEP) of the three scenarios. The results confirm that the Senvion wind farm is the best option since its AEP is approximately 25.50% and 89.99% higher than the GE and Gamesa farm, respectively. Furthermore, this outcome implies that scenario three can generate an average of 12.96 MWh of energy, while the GE and Gamesa options yield a mean of 10.32 and 6.82 MWh, respectively.
Annual energy production per each turbine type at the selected location.
Environmental assessment
GHGs are mainly originated from energy (either in terms of electricity or heating/cooling) production worldwide. Qatar is not the exception, since natural gas power plants and diesel generators are the leading electricity generation forms. With this in mind, it is imperative to point out that the use of these technologies translates into estimated emissions of about 0.181 kg CO2/kWh for natural gas and 0.250 kg CO2/kWh for diesel (Frequently Asked Questions (FAQs) – U.S. Energy Information Administration (EIA), n.d.).
Considering the CO2 emissions per kWh generated and assuming that 80% of the electricity generation is originated by natural gas power plants in Qatar, while diesel generators produce 20%; Table 6 indicates the annual CO2 emission savings per wind farm scenario.
Annual savings of CO2 emissions compared to diesel and natural gas power plants.
Table 6 highlights that the more energy a wind farm produces, the less CO2 emissions are generated countrywide. Hence, since the Senvion scenario generates more clean electricity, it also saves the most annual CO2 emissions (22,110.80 tons of CO2). In comparison, the GE Energy and Gamesa options prevent about 17,617.63 and 11,637.84 tons of CO2 annually, respectively.
Conclusions
The present study analyses Qatar’s wind potential by generating a wind atlas and a wind power density map using WindSim software by over 41 years of data. Furthermore, the best location is determined, and a case study of three wind farms is carried out. From this assessment, the following main results are obtained:
Qatar’s height difference is gradual, which means that the country’s surface is relatively flat.
Throughout the measurement points, at 100 m of elevation, wind speeds between 3 and 9 m/s present the highest probability of occurrence. Moreover, values below 3 m/s offer the lowest likelihood of incidence with a possibility of less than 8% and 3% for velocities of 2, 1 m/s, respectively.
Across Qatar, the mean wind speed at 100 m fluctuates between 5.6054 and 6.2329 m/s on land. Similarly, the offshore average velocity varies between 5.8875 and 6.5257 m/s. Furthermore, northern Qatar presents the highest mean values, both on land and offshore.
Throughout Qatar, the wind power density means at 100 m varies between 149.46 and 267.08 W/m2 onshore, while offshore, it ranges from 210.02 to 335.06 W/m2.
The northern onshore location is the best option to place a wind farm, considering that the selected areas present just a 2.72% difference between their average wind speeds at 100 m (6.25 m/s offshore and 6.08 m/s onshore). Also, the offshore area with the most potential is located above a natural gas reservoir, which may interfere in exploiting such natural resource.
The wind direction is scattered in every direction (especially near the southern border); however, the predominant direction originates from North-West, moving toward the country’s South-East region.
Annual electricity production values of 59.7437, 90.4414, and 113.5075 GWh are obtained by implementing a wind farm conformed by ten Gamesa G97/2000, GE Energy 2.75-120, Senvion 3.4M140 turbines, respectively.
The Senvion wind farm scenario outperforms the other options by 25.50% and 89.99%, generating 113.5075 GWh of electricity annually.
By implementing the Senvion wind farm, the emissions of about 22,110.80 tons of CO2 is prevented annually. Such reduction is created by replacing natural gas and diesel-based electricity generation with clean wind-based electricity generation.
Footnotes
Appendix
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The authors recognize the support provided by the Hamad Bin Khalifa University, Qatar Foundation (210008390). Also, the authors would like to appreciate the assistance provided by Pablo Duran during the simulations. The contents herein are solely the responsibility of the authors.
