Evaluation of the Wind Energy Potential in the Coastal Environment of Two Enclosed Seas

Abstract

The work presents a comprehensive picture of the wind energy potential in the coastal environment of the Black and the Caspian Seas. 10-year of data coming from the US National Centers for Environmental Prediction was considered as the main source. This dataset was subsequently compared with both in situ and remotely sensed measurements. The results show that the western side of the Black Sea has an enhanced wind power potential, especially in the vicinity of the Crimean Peninsula. As regards the Caspian Sea, the northeastern sector can be considered more energetic. A direct comparison of various wind parameters corresponding to the locations with higher potential in the two target areas considered was also carried out, in order to notice the similarities and the key features that could be taken into account in the development of an offshore wind project. Finally, it can be concluded that the coastal environments of the Black and the Caspian Seas can become in the near future promising locations for the wind energy extraction, as well as for the hybrid wind-wave energy farms that could play an important role also in the coastal protection.

Full text

Research Article
Evaluation of the Wind Energy Potential in the Coastal
Environment of Two Enclosed Seas
Florin Onea, Alina Raileanu, and Eugen Rusu
Department of Mechanical Engineering, “Dunarea de Jos” University of Galati, 6200 Galati, Romania
Correspondence should be addressed to Eugen Rusu; erusu@ugal.ro
Received  April ; Revised  May ; Accepted  May 
Academic Editor: Julio Diaz
Copyright ©  Florin Onea et al. is is an open access article distributed under the Creative Commons Attribution License,
which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
e work presents a comprehensive picture of the wind energy potential in the coastal environment of the Black and the Caspian
Seas. -year of data coming from the US National Centers for Environmental Prediction was considered as the main source. is
dataset was subsequently compared with both in situ and remotely sensed measurements. e results show that the western side of
the Black Sea has an enhanced wind power potential, especially in the vicinity of the Crimean Peninsula. As regards the Caspian
Sea, the northeastern sector can be considered more energetic. A direct comparison of various wind parameters corresponding to
the locations with higher potential in the two target areas considered was also carried out, in order to notice the similarities and
the key features that could be taken into account in the development of an oshore wind project. Finally, it can be concluded that
the coastal environments of the Black and the Caspian Seas can become in the near future promising locations for the wind energy
extraction, as well as for the hybrid wind-wave energy farms that could play an important role also in the coastal protection.
1. Introduction
e present work is focused on two enclosed seas, the Black
andtheCaspianSeas,andithasasmainobjectivetoassessthe
wind energy potential in these marine areas. e Black Sea is
located between the Anatolian Peninsula and the southeast-
ernpartofEuropeanditisconnectedtotheMediterranean
Sea throughout the Marmara and Aegean seas, respectively.
is basin can be divided into two main zones (west and
east) with particular features, the coastlines of the sea being
distributed between Bulgaria and Romania (west), Russia
and Ukraine (north), and Georgia (east) and Turkey (south).
Regarding the geographical characteristics of this sea, an
average depth of m, a total area of  km2,anda
water volume of  km3can be mentioned (Rotaru [],
Rusu et al. []). In the short term, the regional climate is
under the inuence of the NAO (North Atlantic Oscillations)
mechanism, according to which during the winter time the
storm events signicantly increase under the inuence of
the cold air arriving from the northern regions. On a local
scale, as in the case of the Novorossiysk region (Russia), the
Bora events can restrict the maritime trac since the wind
conditions can easily exceed  m/s (Alpers et al. []).
Regarding the Caspian Sea, it can be mentioned that this
is in fact the largest enclosed water body in the world (%
of the inland waters) being surrounded in about  km
by Russia and Kazakhstan (north), Turkmenistan (east), Iran
(south), and Azerbaijan (west). It is located between Europe
and Asia, and it has a surface of  km2and a volume
of  km3, while a particularity of this basin is that it is
characterized by important oil elds (Rusu and Onea []).
is basin has in the northern part a continental climate (hot
summers and cold winters), while the southern part is dened
by a subtropical regime (Stolberg et al. []). e weather
conditions in this region are under the inuence of the arctic
air ow, including also the dry continental air masses from
Kazakhstan and the air masses generated over the Atlantic
Ocean (Mamaev []).
At this moment, the Black and Caspian Seas can be
considered important sources of energy, but this regards
mainly the fossil fuel reserves, while the possible benets
from the renewable energy resources are not yet well taken
into consideration. From this perspective, the novelty of
the present work consists in the fact that the wind energy
potential in the vicinity of the coastlines of the two inland
seas is discussed from a meteorological perspective.
Hindawi Publishing Corporation
Advances in Meteorology
Volume 2015, Article ID 808617, 14 pages
http://dx.doi.org/10.1155/2015/808617

Advances in Meteorology
47
45
43
41
27 33 38 43
B1
B2
B3 B4 B5 B6
B7
B8
B9
B10
B11
B12
AB
C
D
Longitude (∘)
Latitude (∘)
(a)
P1
P2
P3
P4
P5
Meteorological stations
(b)
C1
C2
48
42
39
35 46 51 54
C3
C4
C5
AB
C
C6
C7
C8
C9
C10
C11
C12
D
Longitude (∘)
Latitude (∘)
(c)
F : e geographical locations of the reference points
considered in the coastal environments of (a) the Black Sea, (b)
meteorological stations, and (c) the Caspian Sea. e in situ stations
are located in the Black Sea area; most of them are in sector A.
Figures are processed from Google Earth ().
2. Methods and Materials
2.1. e Target Areas. Figure  illustrates the basins of the
Black and of the Caspian Seas, which are targeted in the
present study. Each of the two basins was divided into four
distinct rectangular areas denoted by A, B, C, and D, and
twelve reference points were considered in each basin for
a detailed evaluation. e B group and C group points are
placed in the vicinity of the coastlines in water depths which
do not exceed  m, especially in the case of the Caspian
Sea, which is in the northern part characterized by large
shallow water areas. For the Black Sea area, the reference
points are located in the vicinity of Romania (B), Ukraine
(B), Russia (B–B), Georgia (B), Turkey (B–B), and
Bulgaria (B), respectively. In order to assess the accuracy
and the limitations of the NCEP model, ve meteorological
stationslocatedinthesectorAweretakenintoaccountfor
comparisons, as follows: Gloria (denoted by P), Primorskoe
(P), Chernomorskoe (P), Khersonesskiy-Mayak (P), and
Zavetnoe (P). e rst point is placed in the Romanian
sector, more precisely in the oshore area at a water depth
of about  m, while the rest of the points are located near the
Ukrainian coast. All the in situ measurements are related to
a  m height above the sea level and correspond to the time
interval –, except for the point P, which covers only
thetimeperiod–.Somemoredetailsaboutthese
datacanbefoundinOneaandRusu[].
Regarding the Caspian basin, the C points are divid-
ed between Russia (C–C), Kazakhstan (C–C), Turk-
menistan (C-C), Iran (C-C), and Azerbaijan (C and
C), respectively. For these target areas, it can be mentioned
that Russia has coastlines in both seas, a fact which can be
considered somehow an advantage, since the wind projects
can be focused on one or another area according to the
most favorable wind regime. e coastlines of Turkey can be
also considered as representing another important area, since
Turkey has the largest opening to the Black Sea (km),
compared, for example, with Romania ( km).
2.2. e NCEP-CFSR Dataset. In general, a gap in the assess-
ment of the conditions in the marine environment is repre-
sentedbythelimitedamountoftheinsitumeasurements.
Nevertheless, during the recent years, this aspect was over-
reached by the development of the numerical models, which
can produce extended reanalysis databases in both space and
time.istypeofdatawasalsousedinsomepreviousstudies
to assess the global renewable energy resources in the marine
areas (Rusu [], Rusu and Onea []). A well-known reanalysis
database is the one produced by the National Centers for
Environmental Prediction (NCEP) from the United States,
which provides information for a -year period (–).
e full name of this project is NCEP-CFSR (Climate Forecast
System Reanalysis) and it will be denoted in the present work
by NCEP.
is is based on a system which uses a -day average and
-hour forecast, being capable to simulate various parameters
on a global scale, such as precipitation, temperature, pressure
at the surface, or ice thickness. e wind conditions are
reported at  m above the sea level (a.s.l.) in terms of the
𝑈and 𝑉components (in m/s) being computed based on 
pressure layers, which are structured between the surface and
. hPa ( levels) and from the surface to a  m altitude
(rest of the levels). More details regarding the bias correction,
the assimilation processes, and the data quality can be found
in the specication of the project (Kalnay et al. [], Saha et
al. []).
As a rst step of the present work, the available NetCDF
les were processed for the -year time interval –,
obtaining in this way daily values of the wind conditions with

Advances in Meteorology 
0 5 10 15 20 25
0
5
10
15
20
25
Measurements
Numerical model
U10 (m/s)
(a)
0 2 4 6 8 10 12 14 16 18 20
0
2
4
6
8
10
12
14
16
18
20
Measurements
Numerical model
U10 (m/s)
(b)
05 10 15 20
0
5
10
15
20
Measurements
Numerical model
U10 (m/s)
(c)
0 2 4 6 8 10 12 14 16 18 20
0
2
4
6
8
10
12
14
16
18
20
Measurements
Numerical model
U10 (m/s)
(d)
F : Scatter plots of the 𝑈10 parameter based on the in situ and NCEP model data considering the entire time interval – and
the reference points: (a) P, (b) P, (c) P, and (d) P. e point P is reported only to the time period –.
a step of  hours (––– UTC). In order to investigate
dierent wind patterns, the initial data were particularly
selected for the winter time period (October–March) and
also for the diurnal (– UTC) and nocturnal (– UTC)
intervals.
2.3. Methods. In Figure ,adirectcomparisonbetweenthe
in situ measurements and the NCEP data is presented
corresponding to the reference points P, P, P, and P,
considering the entire time interval –. In this
connection, a second processing of the NCEP data was
performed in order to t the hours corresponding to the in
situ measurements. us, the NCEP model is characterized
by a temporal resolution of  hour ( data points per day)
and for each particular point only the hours at which the
measurements were reported were considered, as follows:
P (01–07–– UTC), P (–21 UTC), P (03–––
21 UTC), and P (–21 UTC). e hours considered to be
part of the diurnal interval are represented with bolds. e
scatter diagram corresponding to the point P is presented
in Figure (a),whereitcanbeobservedthattheNCEP
model tends to underestimate the wind speeds with higher
T  : 𝑈10 statistics, in situ measurements at the meteorological
stations(pointsP–P)againsttheNCEPdataforthesamegeo-
graphical locations. e results are available for the time interval
–, being related to the total time.
Point 𝑋med
(m/s)
𝑌me d
(m/s)
Bias
(m/s) RMSE SI 𝑅
P . . . . . .
P . . −. . . .
P . . −. . . .
P . . −. . . .
P . . −. . . .
values. e best correlation can be observed for the point
P (Figure (b)),whileforthepointsP(Figure (c))andP
(Figure (d)) the numerical model seems to overestimate the
local wind conditions.
e statistical parameters associated with the scatter
diagrams are presented in Tab l e  .𝑋med represents the mean
valueofthemeasurementsandhasamaximumvalueof

Advances in Meteorology
2.5
5.5
9
P1
P2
P3
P4
P5
U10 (m/s)
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
In situ
Month
(a)
P1
P2
P3
P4
P5
3
5.5
8.5
U10 (m/s)
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
NCEP
Month
(b)
P1
P2
P3
P4
P5
−0.6
0
0.6
1.4
U10 (m/s)
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov De c
Month
(c)
P1
P2
P3
P4
P5
−0.6
0
0.4
1
U10 (m/s)
Jan Feb Mar Apr May Jun Jul Aug Sep O ct Nov Dec
Month
(d)
F : Evaluation of the wind conditions reported to the meteorological station sites for the time interval –. e analyses are
based on the in situ measurements and the NCEP data and correspond to the mean values of (a) monthly wind speeds registered at the
meteorological stations, (b) monthly wind speeds provided by the NCEP model, (d) and (c) diurnal-nocturnal variations of the wind speeds
according to the in situ measurements and the NCEP data, respectively.
. m/s for the point P located in the oshore area, compared
to the points P–P (which are in the nearshore), where
values in the range .–. m/s are encountered. 𝑌med
is associated with the NCEP model and has signicantly
higher values, especially for the points P–P, while a good
approximation is provided for the point P (. m/s). e
bias is negative for all the points, except for P where a value
of . m/s results. e Root Mean Square Error (RMSE)
is frequently considered to assess the dierences between
thevaluesprovidedbythenumericalmodelsandtheones
coming from the measurements, where the zero value is
considered as a reference (perfect score). According to this
index, the best results are accounted by the point P (.),
while the point P has a value of .. Regarding the point P,
the higher values of the RMSE index (.) can be considered
to be normal if we take into account that the errors tend to
increase for higher wind speeds. e scatter index (SI), which
links the RMSE to the 𝑋med,presentsvaluesintherange.–
., while for the Pearson correlation index (R)alsothepoint
P presents the best results (.) compared to P (.) and
P (.).
3. Results
3.1. Evaluation of the Wind Conditions. Figure  illustrates
the distribution of the wind conditions corresponding to
the parameter 𝑈10 for the P group points. e monthly
distribution of the in situ measurements (mean values) is
indicated in Figure (a),fromwhichitcanbenoticedthatthe
point P (located oshore) stands out with higher wind speed
values and also indicating clearly the dierences between the
summer and winter intervals. It has to be mentioned that
in the present work the winter time is considered the -
month period between October and March. is point (P)

Advances in Meteorology 
1
4
8
Tot al ti me
Winte r time
B1B2B3B4B5B6B7B8B9B10 B11 B12
U10 (m/s)
(a)
0
220
450
Tot al ti me
Winter time
B1B2B3B4B5B6B7B8B9B10 B11 B12
Power density (W/m2)
(b)
8.5
5.5
3
B2
B5
B7
B10
Jan Feb Mar Ap r May Jun
Month
Jul Aug Sep Oct Nov Dec
U10 (m/s)
(c)
2
5
8.5
B2-diurnal
B5-diurnal
B7-diurnal
B10-diurnal
B2-nocturnal
B5-nocturnal
B7-nocturnal
B10-nocturnal
Jan Feb Mar Apr May Jun
Month
Jul Aug S ep Oct Nov De c
U10 (m/s)
(d)
F : Assessment of the wind conditions in the Black Sea reported to a height of  m. e results are based on the NCEP data (–
) for the mean values of (a) wind speeds—total and winter time, (b) power density—total and winter time, (c) monthly wind speed (B,
B, B, and B), and (d) monthly wind speed (B, B, B, and B)—diurnal and nocturnal.
presents during the winter values in the range .–. m/s,
while a minimum of . m/s is encountered during June and
August, respectively. As regards the points P–P, it can
be mentioned that during the winter time more important
valuesareobservedinMarchwithm/sinP,.m/sin
P, . m/s in P, and . m/s in P. Compared to this time
interval, which is typically considered to be more energetic,
some summer months may report similar values, as in the
case of May: . m/s in P, . m/s in P, . m/s in P, and
. m/s in P. On the opposite side, the NCEP model presents
a smooth distribution of the monthly values (Figure (b)),
clearly highlighting the winter season and indicating the
point P as being the less energetic one. In this case, the point
P is considered to be more energetic with values located
in the range .–. m/s (winter) and .–. m/s (summer),
closelyfollowedbythegrouppointsP–Pwith.–.m/s
(winter) and .–. m/s (summer), respectively.
e dierences between the diurnal and nocturnal inter-
vals, as reected by the in situ measurements, can be
observed in Figure (c). e point P presents in general
considerably higher values of the wind speed corresponding
to the nocturnal conditions, indicating a maximum dierence
of.m/sinDecember.Atthesametime,thepoint
P presents higher nocturnal values, the dierences being
locatedintherange–.m/s.Moresignicantdiurnalwind
conditions can be encountered near the points P–P, higher
variations being reported by P, especially in June (. m/s)
or by P, with a dierence of .m/s in May. Regarding
the NCEP model (Figure (d)), it can be observed that this
indicates more important diurnal conditions for the points
P and P, respectively. According to this dataset, the point P
presents more important diurnal conditions during October,
when a dierence of . m/s is encountered.
Figure  illustrates the distribution of the parameters 𝑈10
in the Black Sea area, structured in total and winter time
intervals, respectively. According to the NCEP data, it can be
noticed that the wind speed tends to decrease from the point
B to B which are aligned along the northern, northeastern,
and southern coasts of the sea, respectively. e winter season
presents higher wind speeds and, according to Figure (a),

Advances in Meteorology
2
5
8
Tot al ti me
Winter time
C1C2C3C4C5C6C7C8C9C10 C11 C12
U10 (m/s)
(a)
0
200
450
Tot al ti me
Winter time
C1C2C3C4C5C6C7C8C9C10 C11 C12
Power density (W/m2)
(b)
3
5.5
8.5
C3
C6
C7
C11
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Month
U10 (m/s)
(c)
2
5
8.5
Month
C3-diurnal
C6-diurnal
C7-diurnal
C11-diurnal
C3-nocturnal
C6-nocturnal
C7-nocturnal
C11-nocturnal
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
U10 (m/s)
(d)
F : Assessment of the wind conditions in the Caspian Sea reported to a height of  m. e results are based on the NCEP data (–
) for the mean values of (a) wind speeds—total and winter time, (b) power density—total and winter time, (c) monthly wind speed (C,
C, C, and C), and (d) monthly wind speed (C, C C, and C)—diurnal and nocturnal.
thefollowingpointsstandoutfromeachreferencesector:
B—. m/s (sector A), B— m/s (sector B), B—. m/s
(sector C), and B—. m/s (sector D).
Besides the wind speed, another parameter that is used
frequently to express the wind energy potential is the power
density (in W/m2), which can be dened as (Fiedler and
Adams []):
𝑃=1
2𝜌𝑈103,()
where 𝜌refers to the air density (. kg/m3)and𝑈10
represents the wind speed reported to a m height above the
sea level. Regarding the power density parameter, presented
in Figure (b), more important energetic peaks are noticed
in B ( W/m2), which is followed by the points B and B
with values in the range – W/m2. For the group points
B–B, the values do not exceed  W/m2.
e monthly distribution of the parameter 𝑈10 (mean
values) for the points B, B, B, and B is presented
in Figure (c). From this gure, it can be observed that
the values reect the total/winter distributions, where the
point B is more energetic and the point B is the least
energetic one. More signicant values are noticed during
the winter time, in particular during January and February,
when values of B– m/s, B–. m/s, B–.m/s, and B–
. m/s are encountered in these points. Figure (d) illustrates
the diurnal/nocturnal distributions of the wind conditions
from which it can be mentioned that the point B presents
the smallest variations. Regarding the point B, this presents
higher nocturnal values during the interval May–October,
when a maximum dierence of .m/s is noticed in Septem-
ber. As regards the point B, during the summer time (May–
August), the diurnal values are considerably higher, while a
reversetrendisobservedinthewinter,whenthediurnalwind
speed has a maximum value of  m/s, the dierences being
about . m/s. Regarding the point B, it can be noticed
that the diurnal conditions are more consistent throughout
the entire year, especially for the interval April–October.
For the Caspian Sea area, a similar analysis is presented
in Figure , considering the same time interval (–).
Regarding the distribution of the wind speed (Figure (a))it

Advances in Meteorology 
T : Statistical analysis of the NCEP data, corresponding to the total time (TT) and winter time (WT), respectively. e results cover the
ten-year time interval –.
Point
Results Time interval Black Sea Caspian Sea
B B B B C C C C
𝑈10 (m/s) TT . . . . . . . .
WT . . . . . . . .
% (m/s) TT . . . . . . . .
WT . . . . . . . .
Extreme (m/s) TT . . . . . . . .
WT . . . . . . . .
Power density (W/m)TT . . . . .  . .
WT . . .  . . . .
can be noticed that higher values were reported by the group
points C–C, where the point C presents a maximum of
. m/s (total time) and . m/s (in winter). On the opposite
side, the point C has values which do not exceed . m/s,
with the mention that the values encountered during the total
time are more signicant. In relationship to each reference
sector, it can be mentioned that the most important values
are noticed in C—.m/s (sector A), C—. m/s (sector
B), C—. m/s (sector C), and C—. m/s (sector D).
e evolution of the power density is presented in
Figure (b), where the points C and C present higher
values, especially during the winter time, while the point
C and the group of points C–C do not exceed the limit
of  W/m2. Similar to the wind speed, from the sector
considered, the following points present maximum values:
C— W/m2,C—W/m
2,C—W/m
2, and C—
 W/m2. Since some reference points already stand out in
terms of their energy, a more detailed evaluation of the
monthly distribution is presented in Figure (c). As expected,
the point C presents lower wind speed values (<. m/s)
while the point C seems to be more energetic during the
time interval November–March, when the mean wind speed
values can reach a maximum of . m/s. e point C presents
lower wind speed values during the interval April–May, when
a minimum of  m/s was registered.
e diurnal/nocturnal distributions of the wind condi-
tions are given in Figure (d), where various patterns can be
noticed. us, for the point C, a higher dierence is noticed
during the summer time, when the nocturnal conditions
are also signicantly higher with dierences of .m/s in
September and October. For the point C, the nocturnal
values can have dierences of . m/s (February), . m/s
(March), and almost  m/s (August and September). e
diurnal wind conditions are higher in the point C, especially
in the interval April–September, when a dierence of . m/s
is noticed in July, while a similar pattern is characteristic to
the point C, which for the interval March–October presents
a maximum dierence of .m/s (in June) and a minimum of
. m/s (in October).
Table  presents a statistical analysis of the main wind
parameters from the Black and the Caspian Seas, where
besides the average values of the wind parameters 𝑈10 and
the power density, there were also included the  percentile
(denoted by %) and the extreme values. e  percentile
presents for the Caspian Sea values in the range .–. m/s
(total time) and .–. m/s (winter season), while in the
Black Sea these values are in the interval .–. m/s (total
time) and .–. m/s (in winter).
Anotherimportantparameterintheprocessofevaluating
a particular location is the direction from which the wind
is blowing. Figure  illustrates the wind roses, which are
represented corresponding to the meteorological points P–
Pbasedonthetimeperiod–,theresultsbeing
structured in the diurnal and nocturnal intervals. At a
rst analysis, it can be observed that the NCEP data have
smoother wave roses. Moreover, they are grouped around a
particular direction, compared to the in situ measurements,
which present a more irregular distribution. For the point
P, the in situ data indicate the northern sector as being
dominant, with the mention that during the diurnal time it is
possible to encounter wind conditions from the northeastern
direction. e NCEP model suggests that during the diurnal
period the southern sector is the dominant one, while during
the nighttime the northeastern-southwestern line appears to
be the most relevant. As regards the point P, it can be
mentioned that there is a small agreement between the two
datasets, in the sense that both indicate that the southern
sector is dominant during the nocturnal interval, with the
mention that the in situ values are located more in the
southeastern sector. Maybe the best agreement is reported by
the point P, which indicates that the southeastern sector is
one of the most important (diurnal and nocturnal), while the
in situ values also indicate the northern sector.
Based on the NCEP dataset, Figure  illustrates a similar
analysis, but this time only for the total time data (–
) and by considering only some reference points from
the two target areas. In general, it can be noticed that
the wind direction from the two seas does not follow a
regular pattern, each region being dened by some particular
features. Regarding the Black Sea (Figure (a)), it can be
mentioned that the points from the northern part of the
basin present a grouped distribution of the wind directions,
with no clear peaks, except for the point B, which indicates
the northeastern sector, an aspect that could be associated

Advances in Meteorology
5%
10%
15%
20%
WE
S
N
5%
10%
15%
20%
WE
S
N
5%
10%
15%
20%
WE
S
N
5%
10%
15%
20%
WE
S
N
P1
Diurnal
Nocturnal
NCEP
03
6≥9
U10 (m/s)
In situ
(a)
5%
10%
15%
20%
WE
S
N
5%
10%
15%
20%
WE
S
N
03
6≥9
U10 (m/s)
5%
10%
15%
20%
WE
S
N
5%
10%
15%
20%
WE
S
N
P2
NCEP
In situ
(b)
5%
10%
15%
20%
WE
S
N
5%
10%
15%
20%
WE
S
N
5%
10%
15%
20%
WE
S
N
5%
10%
15%
20%
WE
S
N
P3
NCEP
03
6≥9
U10 (m/s)
In situ
(c)
5%
10%
15%
20%
WE
S
N
5%
10%
15%
20%
WE
S
N
5%
10%
15%
20%
WE
S
N
5%
10%
15%
20%
WE
S
N
P4
Diurnal
Nocturnal
NCEP
03
6≥9
U10 (m/s)
In situ
(d)
F : Wind roses reported to the meteorological stations, during the time period –. e results are structured in the diurnal
and nocturnal intervals, corresponding to the points (a) P, (b) P, (c) P, and (d) P.

Advances in Meteorology 
10%
20%
15%
47
45
43
41
27 33 38 43
B3
B5
5%
10%
15%
20%
WE
S
N
5%
10%
15%
20%
WE
S
N
5%
10%
15%
20%
WE
S
N
5%
10%
15%
20%
WE
S
N
5%
10%
15%
20%
WE
S
N
5%
10%
15%
20%
WE
S
N
B7
B9
B11
B1
Longitude (∘)
Latitude (∘)
03
6≥9
U10 (m/s)
B3-wind roses
5%
(a)
46 51 54
5%
10%
15%
20%
WE
S
N
5%
10%
15%
20%
WE
S
N
5%
10%
15%
20%
WE
S
N
5%
10%
15%
20%
WE
S
N
5%
10%
15%
20%
WE
S
N
036≥9
U10 (m/s)
C2
C4
C6
C8
C10
C12
Longitude (∘)
48
42
39
35
Latitude (∘)
5%
10%
15%
20%
WE
S
N
(b)
F : Wind roses corresponding to some relevant points from (a) the Black Sea and (b) the Caspian Sea. e NCEP data correspond to
the -year time interval (–).

 Advances in Meteorology
0
250
550
B2
C6
Power density (W/m2)
Jan Feb Mar Apr May Jun Ju l Aug S ep Oct Nov De c
Month
(a)
5%
10%
15%
20%
WE
S
N
5%
10%
15%
20%
WE
S
N
03 6≥9
U10 (m/s)
B2 diurnal
B2 nocturnal
(b)
5%
10%
15%
20%
WE
S
N
C6 diurnal
C6 nocturnal
5%
10%
15%
20%
WE
S
N
03 6≥9
U10 (m/s)
(c)
F : Comparison of the wind conditions corresponding to the points B (the Black Sea) and C (the Caspian Sea), where (a) monthly
mean power density, (b) diurnal and nocturnal wind roses for the reference point B, and (c) diurnal and nocturnal wind roses for the
reference point C are reported.
with the inuence of the Bora events from the Novorossiysk
region.MoreobviouspeaksareobservedinthepointsB
(from the east), B (from the southwest), and B (from the
northeast), which instead present lower wind speeds than the
points located close to the northern coast. In the Caspian
Sea (Figure (b)),itcanbenoticedthatthepointsfrom
the northern part (C, C, and C) present a signicant
percentage of the wind conditions higher than  m/s. In terms
of the wind direction the point C indicates the northeastern
sector as dominant, while C is under the inuence of the
northwestern and northeastern winds. For the point C,
the general occurrence of the wind can be associated with
the onshore area while in about % of the cases the wind
conditions can occur from the northern sector (along the
coastline).
A closer look at the wind conditions of the two target
areasispresentedinFigure ,consideringonlythemost
representative locations. As it was previously noticed, based
ontheNCEPdata,itisshownthatthepointBseemsto
present the best wind resources from the entire sea, while
the point C, located in the vicinity of Kazakhstan, was
considered since it presents a signicant wind potential, as
itcanbenoticedfromFigure (a) (mean values).
Figure (a) illustrates the monthly evolution of the
power density. First of all, the dierences between the win-
ter time and the rest can be noticed and also that the

Advances in Meteorology 
point B does not present higher values than the point C.
Regarding the point B, it can be observed that a max-
imum of  W/m2is reported in January, while com-
pared to C the following dierences can be noticed: Jan-
uary— W/m2,March—W/m
2,May—.W/m
2,July—
 W/m2, September— W/m2, and November— W/m2,
respectively.
e most severe variation can be associated with the
diurnal/nocturnal distribution of the wind direction, which
is illustrated for the point B in Figure (b).Duringthe
diurnal interval, a signicant percentage of the wind is
coming from the southwestern sector (sea region) compared
to the night interval, when the northern sector (onshore
region) is dominant, while regarding the wind speeds there
are no visible variations. For the point C (Figure (c)), the
northwestern sector seems to be more representative, while
a reverse trend is observed for the nocturnal period when
the wind conditions from the southeastern sector occur more
frequent.
4. Discussions of the Results
Although at this moment in Europe, the ocean boundaries
present more interest, since the wind conditions seem to
be in general more energetic there, possible benets can be
obtained also from the inland basins such as the Mediter-
ranean and Black seas (Ahmed Shata and Hanitsch [], Onea
and Rusu []) or from the Caspian Sea (Kerimov et al.
[], Rahmanov et al. []), which is relatively close to this
geographic region.
Since the NCEP dataset can be considered a blended
source of data, from the comparisons with the in situ
measurements (wind speed and direction), it was noticed that
this type of data tends to average the wind conditions both in
the oshore or the nearshore locations. In order to provide a
better perspective on the results obtained for the two basins,
in this section, also some satellite measurements will be ana-
lyzed, which were processed corresponding to the locations
of the same reference points, respectively: B–B (Black Sea)
and C–C (Caspian Sea). ese measurements are coming
from the AVISO (Archiving, Validation and Interpretation
of Satellite Oceanographic Data) program [], which is a
multimission project where multiple satellite missions are
combined and calibrated in order to obtain an accurate
dataset available on a global scale. In the present work, the
wind measurements were processed for the time interval
– ( years), each time series being characterized by
one value per day.
Figure  illustrates the evolution of the parameter 𝑈10
for the reference points in the Black Sea. e seasonal
distribution is presented in Figure (a) for the total and the
winter time periods, where it can be observed that the points
located in the western part of the sea present more energetic
features. According to these measurements, the point B
presents more important values with . m/s during total
time and . m/s in winter, closely followed by B with  m/s
(winter), B—. m/s (winter), B—. m/s (winter), and
B–B with values located close to . m/s (winter). On the
opposite side, the points B and B can be considered, where
the wind speeds do not exceed . m/s during the total time or
. m/s in winter. e monthly distribution of the wind speed
for the most relevant points is presented in Figure (b),from
which the inuence of the winter season can be observed.
Firstofall,itcanbenoticedthatthedierencesbetweenthe
points are very small, the point B presenting slightly higher
values during January–March (. m/s) or for September-
October interval (. m/s). During the summer time, in the
reference points, wind speeds in the interval .–. m/s
can be noticed, with some energetic peaks during June
and September. Regarding the structure of the wind, in
Figure (c), it is represented the histogram corresponding to
the point B. As it can be observed, the wind occurrences
from the interval – m/s are the most important, while the
values higher than  m/s are almost close to zero. During the
winter time, the wind conditions located under m/s seem to
be insignicant reported to the total wind budget.
A similar analysis is performed in Figure  in the
Caspian Sea area. e seasonal mean values (Figure (a))
indicate that the point C seems to have the most important
wind resources indicating a maximum of . m/s during
winter, this point being also indicated by the NCEP dataset.
Other important points can be considered: C with . m/s
and . m/s (total-winter time), C →. m/s–. m/s and
C →. m/s–. m/s. e point C appears to be the
less energetic one, since the wind conditions do not exceed
. m/s, even during the winter, this aspect being also indi-
cated by the NCEP data. Regarding the monthlydistribution,
theseasonalpatternofthewindconditionscanbeobserved
when a maximum of . m/s appears during February, while
the point C has considerably lower values for the interval
October–December. During the summer time, the selected
points have wind speed values between  and m/s, lower
valuesbeingregisteredinMay.ewindhistogramis
indicated in Figure (c) for the point C, where during the
total time a similar distribution can be observed as in the case
of the point B, while during the winter season an energetic
peak for the interval – m/s is reported.
Table  presents a statistical analysis of the wind condi-
tions considering the same reference points as in the case
presented in Tab l e  (NCEP data). In general, it can be
observed that the values indicated by the satellite measure-
ments are signicantly lower than those given by the model,
both during the total and the winter time. From the analysis
ofthe%itcanbeobservedthattheBlackSeapointspresent
values in the range .–. m/s (total time) and .–
. m/s (in winter) compared to the Caspian Sea, where
during the winter time the values are close to  m/s. For the
extreme values, a maximum of . m/s is indicated by the
point C, being followed by B with .m/s, while on the
opposite side the point B presents a minimum of  m/s.
During the winter time, the power density takes values in
the range – W/m2for the B points, compared to –
 W/m2in the Caspian basin.
5. Conclusions
In the present work a comprehensive picture of the wind
energy potential in the Black and the Caspian Seas is provided

 Advances in Meteorology
2.5
4
5.5
Tot al ti me
Winter time
B1B2B3B4B5
Points
B6B7B8B9B10 B11 B12
U10 (m/s)
Black sea
(a)
2
4
6
B1
B2
B3
B4
Jan Feb Mar Apr May Jun
Month
Jul Aug Sep Oct Nov Dec
U10 (m/s)
(b)
0
600
1200
0–33–66–99–12 12–15 ≥15
U10 (m/s)
N
Tot al ti me
Winter time
B2
(c)
F : Evaluation of the wind conditions in the Black Sea based on satellite measurements. e results are reported to the time interval
–, for (a) mean wind speed for the total and winter time, respectively, (b) the monthly evolution of the 𝑈10 parameter in the points
B, B, B, and B, and (c) wind speed histogram corresponding to the point B.
T : Statistical analysis of the satellite measurements, corresponding to the total time (TT) and winter time (WT), respectively. e results
cover the ve-year time interval –.
Point
Results Time interval Black Sea Caspian Sea
B B B B C C C C
𝑈10 (m/s) TT . . . . . . . .
WT . . . . . . . .
% (m/s) TT . . . . . . . .
WT . . . . . . . .
Extreme (m/s) TT . . . . . . . .
WT . . . . . . . .
Power density (W/m)TT  . . . . . . .
WT . . . . . . . .
from a meteorological point of view. e analysis is based
both on the reanalysis data coming from the National Centers
for Environmental Prediction (NCEP), which cover a -year
interval (–), and also throughout some measured
data (both in situ and remotely sensed).
By dividing each target area into four rectangular
domains, it was possible to identify in each basin the most
promising locations from the point of view of the wind energy
potential. Regarding the Black Sea region, it was found that
the windiest locations seem to be situated in the northwestern
part of the sea, especially in the vicinity of the point B, which
is located close to Ukraine. For the Caspian region, the points
C and C (Kazakhstan) stand out with signicant values,
especially during the winter time. e less energetic areas
were found to be in the southern part of the Black Sea, while
intheCaspianSealowervaluesofthewindspeedandpower

Advances in Meteorology 
3.5
5
7
Tot al ti me
Points
Winter time
C1C2C3C4C5C6C7C8C9C10 C11 C12
U10 (m/s)
Caspian sea
(a)
7.5
5.5
3
C1
C2
C6
C12
Jan Feb Mar Apr May Jun
Month
Jul Aug S ep Oct Nov Dec
U10 (m/s)
(b)
0
300
600
N
0–33–66–99–12 12–15 ≥15
U10 (m/s)
Tot al ti me
Winter time
C6
(c)
F : Evaluation of the wind conditions in the Caspian Sea based on satellite measurements. e results are reported to the time interval
–, for (a) mean wind speed for the total and winter time, respectively, (b) monthly evolution of the 𝑈10 parameter in the points C,
C, C, and C, and (c) wind speed histogram corresponding to the point C.
density were reported in the points located in the southern
andsouthwesternpartsofthebasin.
ese results are also conrmed by the satellite mea-
surements, which indicate, with a good accuracy, the most
or the less energetic points from the two areas targeted,
with the mention that the values are usually below the
onesprovidedbytheNCEPmodel.Ifwetakeintoaccount
that the most energetic area presents also relevant wave
energy resources, it can be expected in the near future to
bedevelopedhybridwind-wavesfarms,whichmightalso
play an important role in the coastal protection (see, e.g.,
Zanopol et al. [,]). is especially takes into account the
fact that many of the coastal areas targeted are subjected to
intense processes of coastal erosion (Diaconu and Rusu []).
Regarding the diurnal/nocturnal distributions of the wind
conditions, it was found that these conditions are dierent
for each geographical region, while on a local scale the most
important variations occur in the directional patterns of the
wind.
BasedbothontheNCEPdataandonthesatellite
measurements, it is shown that the two enclosed seas targeted
in the present work can provide suitable conditions for the
implementation of the oshore wind parks. On the other
hand,ifwetakeastepbacklookingatthebigpicture,it
canbeassumedthatsincetheCaspianSeaisanarearichin
oil, there are few chances at this moment to see a renewable
project in this area. As regards the Black Sea, although the
strongest wind conditions were noticed near the coastlines
of Ukraine, considering the current geopolitical climate, it
might be rather unrealistic to make plans there for an oshore
farm. In this context, it can be concluded that one of the most
interesting regions is located in the vicinity of the Romanian
coast, nearshore that has a good wind potential as also the
capacity to develop a renewable project in the oshore areas.
Conflict of Interests
e authors declare that there is no conict of interests
regarding the publication of this paper.
Acknowledgments
is work was supported by a grant of the Romanian
Ministry of National Education, CNCS-UEFISCDI PN-II-
ID-PCE--- (project DAMWAVE). e work of the
rst author has been funded by the Sectoral Operational
Programme Human Resources Development – of
the Ministry of European Funds through the Financial

 Advances in Meteorology
Agreement POSDRU//./S/. e wind dataset cor-
responding to the Ukrainian coastal environment of the
Black Sea was kindly provided by the Ukrainian Research
Hydrometeorological Institute. e altimeter products were
produced by Ssalto/Duacs and distributed by Aviso with
support from Cnes.
References
[] A. Rotaru, “Some geo-aspects of the Black Sea basin,” in Pro-
ceedings of the 3rd International Conference on Environmental
and Geological Science and Engineering, pp. –, .
[] E. Rusu, F. Onea, and R. Toderascu, “Dynamics of the envi-
ronmental matrix in the Black Sea as reected by recent
measurements and simulations with numerical models,” in e
Black Sea: Dynamics, Ecology and Conservation, Nova Science
Publishers, New York, NY, USA, .
[] W. Alpers, A. Ivanov, and J. Horstmann, “Observations of Bora
events over the Adriatic Sea and Black Sea by spaceborne
synthetic aperture radar,” Monthly Weather Review,vol.,no.
, pp. –, .
[] E. Rusu and F. Onea, “Evaluation of the wind and wave energy
along the Caspian Sea,” Energy,vol.,no.,pp.–,.
[] F.Stolberg,O.Borysova,I.Mitrofanov,V.Barannik,andP.Eght-
esadi, Caspian Sea, GIWA Regional Assessment 23, University of
Kalmar,Kalmar,Sweden,.
[] V. Mamaev, Seas Around Europe. Europe’s Biodiversity—Bioge-
ographical Regions and Seas, European Environment Agency,
.
[] F. Onea and E. Rusu, “Wind energy assessments along the Black
Sea basin,” Meteorological Applications,vol.,no.,pp.–
, .
[] E. Rusu, “Assessment of the wave energy conversion patterns in
various coastal environments,” Energies,vol.,no.,pp.–
, , Proceedings of the st International e-Conference on
Energies.
[] L. Rusu and F. Onea, “Assessment of the performances of
variouswaveenergyconvertersalongtheEuropeancontinental
coasts,” Energy,vol.,pp.–,.
[] E. Kalnay, M. Kanamitsu, R. Kistler et al., “e NCEP/NCAR
-year reanalysis project,” Bulletin of the American Meteorolog-
ical Society,vol.,no.,pp.–,.
[] S. Saha, S. Moorthi, H.-L. Pan et al., “e NCEP climate forecast
system reanalysis,” Bulletin of the American Meteorological
Society,vol.,no.,pp.–,.
[] B.H.FiedlerandA.S.Adams,“Asubgridparameterizationfor
wind turbines in weather prediction models with an application
to wind resource limits,” Advances in Meteorology,vol.,
Article ID ,  pages, .
[] A. S. Ahmed Shata and R. Hanitsch, “Evaluation of wind
energy potential and electricity generation on the coast of
Mediterranean Sea in Egypt,” Renewable Energy,vol.,no.,
pp.–,.
[] F. Onea and E. Rusu, “An evaluation of the wind energy in the
North-West of the Black Sea,” International Journal of Green
Energy,vol.,no.,pp.–,.
[] R.Kerimov,Z.Ismailova,andN.R.Rahmanov,“Modelingof
wind power producing in Caspian Sea conditions,” International
Journal on Technical and Physical Problems of Engineering,vol.
,no.,pp.–,.
[] N. Rahmanov, R. Kerimov, E. Gurbanov et al., “Assessing the
wind potential of Caspian Sea region for covering demand
in neighboring countries and reducing of carbon emission,”
in Proceedings of the 2nd International Symposium on Energy
Challenges & Mechanics, Aberdeen, UK, August .
[] AVISO, , http://ww w.aviso.altimetry.fr/en/home.html.
[] A. Zanopol, F. Onea, and E. Rusu, “Coastal impact assessment
of a generic wave farm operating in the Romanian nearshore,”
Energy,vol.,pp.–,.
[] A. Zanopol, F. Onea, and E. Rusu, “Evaluation of the coastal
inuence of a generic wave farm operating in the Romanian
nearshore,” Journal of Environmental Protection and Ecology,
vol. , no. , pp. –, .
[] S. Diaconu and E. Rusu, “e environmental impact of a wave
dragon array operating in the Black Sea,” e Scientic World
Journal,vol.,ArticleID,pages,.