Industry

A five-yrs study of coastal recirculation and its effect on air pollutants at the East Mediterranean region

Description
A five-yrs study of coastal recirculation and its effect on air pollutants at the East Mediterranean region
Categories
Published
of 14
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
  A five-year study of coastal recirculation and its effect on airpollutants over the East Mediterranean region Ilan Levy, 1 Uri Dayan, 1 and Yitzhak Mahrer  2 Received 23 October 2007; revised 24 January 2008; accepted 22 May 2008; published 30 August 2008. [ 1 ] Many studies have shown that air pollutants concentrations in coastal cities may begravely affected by coastal recirculation. In this study an attempt is made to examinethe properties of coastal recirculation over a long period (5 yrs) at multiple sites along theEast Mediterranean Sea (EMS). For this purpose, a single station quantitative measureof horizontal recirculation is used based on wind field measurements over periods of 1–96 hrs. The horizontal recirculation is examined with respect to the integration time period, synoptic flow, seasonality, coastline variations, elevation, and air pollutantsconcentrations. The interaction between synoptic and mesoscales is shown to be agoverning factor by allowing or overruling the land sea breeze winds. Favorite conditionsfor coastal recirculation are shown to be light or variable winds such as under a Cole or aHigh-Pressure system. The monthly distribution of the recirculation potential has a bimodal behavior with two peaks during the transitional seasons and October in particular.This is as a result of the annual cycle of night-time land-sea temperature difference drivingthe land breeze and the more frequent passage of synoptic scale flows with an easterlywind component at the EMS. Two factors leading to variations along the coastline arethe urban heat island, weakening the breeze winds and reducing recirculation potential,and the concaved shape of the southern shoreline that causes a convergence andstrengthening of the land breeze, thus supporting recirculation. The primary pollutants NO x and SO 2 have the highest concentrations during weak daily mean wind speeds. O 3 levels depict an almost opposite image of NO x , with higher values for both high and lowrecirculation, possibly resulting from either long range transport or coastal recirculation. Citation: Levy, I., U. Dayan, and Y. Mahrer (2008), A five-year study of coastal recirculation and its effect on air pollutants over theEast Mediterranean region, J. Geophys. Res. , 113 , D16121, doi:10.1029/2007JD009529. 1. Introduction [ 2 ] The effect of land and sea breeze (LSB) on coastalmeteorology in general and the interaction between LSBand air pollutants in particular plays an important role indetermining many aspects of coastal environments aroundthe world [  NRC  , 1992, chapter 7]. In these regions that holda large part of the population along with pollution emissionsources, air quality is an outcome of the combined effect of several governing factors, such as long range transport,short-range transport, local emissions, photochemical activ-ity and the meteorological conditions for transport anddiffusion [  NRC  , 1992; Banta et al. , 2005].[ 3 ] A meteorological phenomenon that is often associatedwith the LSB is air mass recirculation in coastal regions[  Hsu , 1988; Miller et al. , 2003]. Coastal recirculation occursin two mechanisms. Vertical recirculation refers to theascent of sea breeze air at the breeze front, a return flowaloft few hundreds of meters above the surface breeze layer and a descent back into the sea breeze layer near the surface[  Hsu , 1988]. Horizontal recirculation refers to the clockwise(or anticlockwise) rotation of the wind direction due to thediurnal cycle of the LSB, causing an air mass to return to itssource region the next day.[ 4 ] Apart from the main driving force of the land-seatemperature difference (LSTD), the LSB winds are knownto interact with other factor, such as prevailing synoptic pattern [e.g., Ma and Lyons , 2003; Oh et al. , 2006]; urbanheat island (UHI) [e.g., Yoshikado , 1992; Ohashi and Kida ,2002; Martilli , 2003; Freitas et al. , 2007]; vicinity to amountain ridge [e.g., Lalas et al. , 1983; Lu and Turco ,1994; Perez-Landa et al. , 2007; Porson et al. , 2007];curvature of the shoreline [  Alpert and Getenio , 1988] andother topographic settings of the coastal region, like landcover, land use and vegetation type.[ 5 ] Many studies have shown that air pollutants concen-trations in coastal cities may be gravely affected by coastalrecirculation [e.g., Lyons and Cole , 1976; Ma and Lyons ,2003; Baumgardner et al. , 2006; Ainslie and Steyn , 2007].In the Mediterranean basin, that is mostly surrounded bymountains and heavily populated, multiple evidence exist asto the combined effect of coastal recirculation, topograph- JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 113, D16121, doi:10.1029/2007JD009529, 2008 ClickHere for Full Article 1 Department of Geography, Hebrew University of Jerusalem, Jerusalem,Israel. 2 The Seagram Center for Soil and Water Sciences, Hebrew Universityof Jerusalem, Jerusalem, Israel.Copyright 2008 by the American Geophysical Union.0148-0227/08/2007JD009529$09.00 D16121 1 of 14  ical settings, the UHI and large-scale synoptic flow on air quality. Studies performed in the Athenes air basin [e.g.,  Lalas et al. , 1983; Kallos et al. , 1993; Asimakopoulos et al. ,1999] have shown that during the summer season thecombined effect of the sea breeze, local topography andsynoptic flow often result in elevated levels of both primaryand secondary pollutants. Other studies over the Iberian peninsula [e.g., Millan et al. , 1984, 1987; Massons et al. ,1997; Salvadoret al. , 1999; Millan et al. , 2000; Perez-Landaet al. , 2007], the Adriatic coast [e.g., Fortezza et al. , 1993],southern France [e.g., Lasry et al. , 2005], Tunisia [e.g.,  Bouchlaghem et al. , 2007] and Israel [ Segal et al. , 1982;  Peleg et al. , 1994; Robinsohn et al. , 1992; Alper-Siman Tov et al. , 1997] have all shown similar results.[ 6 ] The above mentioned studies have all focused on theanalysis of breeze and recirculation in a single case analysis, based on field measurements and\or using meteorologicaland dispersion modeling. However, such an approach failsto provide a complete picture of the recirculation phenom-enon. As far as the authors are aware, no studies have been performed that neither quantify coastal recirculation, nor compare different sites/regions, or examine its frequencyover long periods of time over the Mediterranean basin. Thestudy presented here aims at giving a quantitative estimateof the horizontal recirculation based on a method presented by Allwine and Whiteman [1994, hereafter AW], withrespect to the integration time period, prevailing synopticconditions and seasonality. In addition, the effect of the UHIis shown as well as the concaved shape of the shoreline.Finally, the resulting impact on air pollutants concentrationsis examined. 2. Methodology and Data Sets 2.1. Quantitative Measure of Recirculation [ 7 ] An objective quantitative measure of air mass stag-nation, recirculation and ventilation was proposed by AW, based on a single station integral measure of wind data.Given a series of N discrete observations of wind speed anddirection in a measuring site: V  i ¼ u i þ v  i ; i ¼ 1 ; 2 ; . . . ;  N  ð 1 Þ with an averaging interval of T hours (e.g., 5-yrs data seriesof 30-min averaged observations). Based on the wind data,the authors define the discrete integral quantities of ‘‘resultant transport distance’’ (L) that is the net vector displacement, ‘‘wind run’’ (S) that is the wind scalar sumand the ‘‘recirculation factor’’ (R) based on the ratio between L and S:  L i t  ¼ T  X i À t  þ 1  j  ¼ 1 V   j   ¼ T  X i À t  þ 1  j  ¼ i u  j  ! 2 þ X i À t  þ 1  j  ¼ i v   j  ! 2 2435 12 ð 2 Þ S  i t  ¼ T  X i À t  þ 1  j  ¼ i V   j   ¼ T  X i À t  þ 1  j  ¼ i u 2  j  þ v  2  j   12 ð 3 Þ  R i t   1 À  L i t  S  i t  ð 4 Þ where t  is the wind run time for integration (e.g., 24 hrs).The calculated values of L, S and R presented in this studydiffer from AW’s values only in that they are looking backward in time rather then its future progress, in order torelate current measurements of air pollution with the recent history of an air mass. L i , S i , and R  i are calculated at everyending time step t  i to the start time t  i À t  +1 . The indexnotation R  24 = R  ( t  =24) will be used hereafter. Figure 1illustrates S and L for the cases of high and low recirculation.[ 8 ] The resultant transport distance L is a measure of thenet distance an air parcel had dislocated in the previous t  hours. S is a measure of the accumulated distance the parceltraveled in that time and when divided by t  is the scalar average wind speed. Hereafter the notation S 0 will be usedfor the average wind speed (S/  t  ) expressed in m s À 1 . Low S 0 values are an indication of stagnant wind conditions. Thecombination of stagnant conditions and high recirculation isa measure of poor ventilation, whereas nonstagnant con-ditions and low recirculation factor are a measure of goodventilation. Here we use AW’s criteria for ventilation basedon ground measurements only, as opposed to the ventilationcoefficient that is the product of the mixing height and theaveraged wind speed from the ground to that height.[ 9 ] It is important to note that the quantities of L, S and R are an exact measure of the parcel’s travel only for an idealhomogeneous wind field, i.e., when the wind observed at the measuring site is uniform throughout the region. Sincethis study applies these quantities for a LSB environment,that is not the case. Therefore L, S and R should only be Figure 1. Schematic illustration of the net vector dis- placement (L, dashed line) and progressive vector diagram(S, solid line) for (top) high and (bottom) low recirculation.The dot marks the monitoring site. D16121 LEVY ET AL.: 5-YR COASTAL RECIRCULATION AND POLLUTION2 of 14 D16121  regarded as a measure of the wind field in the vicinity of themeasuring site. Nevertheless, when examining the climato-logical aspects of regional air flow and pollutants behavior,they allow an important insight. 2.2. Data Sets [ 10 ] The study area is the Israeli coastal plain at thesoutheastern part of the Mediterranean, a relatively flat terrain bounded from the east by a mountain ridge withheights up to 800 m above sea level (ASL) at a distance of 40–60 km from the coast (Figure 2). This study focuses onthree airsheds along the coastal plain: Hadera, Tel Aviv andAshdod, each having local emission sources from transpor-tation, industry and energy production activities. TheHadera airshed holds the medium-sized city of Hadera[population of 75,000; CBS  , 2006] and a few small towns,located 45 km north of Tel Aviv (Figure 2) and distancedsome 20 km from the mountain ridge (400 m ASL). It alsoholds Israel’s largest coal-fired power plant operated by theIsrael Electric Corporation (IEC), IEC Orot Rabin (Table 1).The Tel Aviv airshed has the largest metropolitan area inIsrael with a population of over 1 million people [ CBS  ,2006], with its associated mobile emission sources, and asmall power plant (IEC Riding). The third airshed of Ashdod includes the medium-sized city of Ashdod[200,000 inhabitants; CBS  , 2006], located 30 km south of Tel Aviv, at a distance of 30 km from the mountain ridge.The Ashdod airshed holds the medium-sized power plant of IEC Eshkol, as well as the Ashdod Oil Refinery. The secondlargest power plant of IEC Rotenberg is also located some20 km south of Ashdod (Figure 2 and Table 1).[ 11 ] 30-min averages of meteorological parameters (windspeed, wind direction, air temperature and relative humid-ity) as well as air pollutants concentrations (NO x , O 3 andSO 2 ) were retrieved for a period of 5 yrs at 29 sites alongthe Israeli coastline, most of them located at the threeairsheds described above, and the others further inland(Figure 2). The resultant transport distance L, wind run Sand recirculation parameter R, were derived for each site at every time step, calculated backward in time for  t  = 1, 2, ..,96 hours.[ 12 ] Vertical profile of the horizontal winds were obtained by a 1290-MHz radar wind profiler located 3.4 km onshoreand 15 m ASL, near Hadera. The data used are in 30 mintime resolution and 60 m vertical resolution (a total of 25 vertical layers: 144–1532 m above ground level, AGL)[  Dayan et al. , 2002]. Since available data for the period isnot continuous, recirculation values where calculated onlyfor three uninterrupted periods: between 15/07/2002-6/2/ 2003 (207 days), 30/5/2003-19/10/2003 (143 days) and 26/ 7/2004-4/11/2004 (102 days). Daily horizontal recirculationfactors ( t  = 24) were derived for each of the 25 height levels provided by the wind profiler, based on the 30 min averagedwind observations. A strict criterion was used for more then85% of the observations to be available during the 24 hrs period of calculation, meaning that no more then 7 obser-vations could be missing.[ 13 ] In addition, hourly averages of sea surface tempera-ture (SST) were available from two buoys located about 2 km offshore from Haifa and Ashdod (Figure 2), for thesame 5-yrs period. Since each buoy’s data set had large gapsof missing data, and since the difference in measurements between the two was relatively small ( $ 0.1 ° ), the two datasets were merged into a single uninterrupted record of hourly SST measurements by averaging simultaneousobservations. Daily mean zonal and meridional componentsof wind speeds and air temperature at the 925 hPa levelwere extracted from the NCEP/NCAR Reanalysis Project [  Kalnay et al. , 1996; Kistler et al. , 2001] for the entire 5-yrs period. The 925 hPa level was chosen to represent ambient  Figure 2. Map of the study region, indicating monitoringstations at three airsheds and downwind, buoys measuringsea surface temperature, major emission sources, maincities, and urban areas. Contours indicate topographicelevation above sea level (m). Table 1. Major Emission Sources at the Coastal Plain [ Weinroth ,2004; E. Weinroth, personal communication, 2007] SourceStack Height (m) TypeSO 2 (Mol hr  À 1 ) NO(Mol hr  À 1 )IEC a  Haifa 80 PP  b 28,500 17,300Oil Refinery Haifa OR  c 17,200 13,400IEC Orot Rabin, Hadera 250–300 PP 321,000 409,000MBPH d , Hadera 40 Ind e 3,250 1,960IEC Riding, Tel Aviv 150 PP 34,900 21,200Oil Refinery Ashdod OR 1,550 2,020IEC Eshkol, Ashdod 150 PP 67,100 40,700IEC Rotenberg, Ashkelon 250 PP 118,000 150,000 a  IEC - Israel Electric Corporation.  b PP - Power Plant. c OR - Oil Refinery. d MBPH - Mondi Business Paper Hadera. e Ind - Industry. D16121 LEVY ET AL.: 5-YR COASTAL RECIRCULATION AND POLLUTION3 of 14 D16121  conditions being within the mixing layer most of the year [  Dayan and Rodnizki , 1999], yet above the breeze layer.Moreover, these data are coarse enough (2.5 ° Â 2.5 ° ) not to be directly affected by the land-sea interactions.[ 14 ] Finally, a subjective manual classification of thedaily synoptic circulation patterns in the EM region wasused, along the 12:00 UTC sea level pressure chartsextracted from NCEP/NCAR Reanalysis Project [  Kalnayet al. , 1996; Kistler et al. , 2001]. The classification is basedon the 3-yrs (2000–2002) classification described in moredetails by Dayan and Levy [2005], and was extended for two more years (2003–2004) to include a total of 1827days, classified into 23 synoptic flow patterns over the EM.The different flow patterns are listed in Table 2 along withtheir associated ground level wind direction, 925 hPa meanair temperature and mean zonal and meridional windcomponents, 5-yrs frequencies and monthly distribution. 3. Results and Discussion 3.1. Varying Run Time for Integration [ 15 ] Examining the recirculation factor for varying trans- port times ( t  = 1, .., 96) and at different starting times(between 00:00–23:30 every 30 min), provides some in-sight on the governing scales of motion in the lower atmosphere. Figure 3 shows the 5-yrs average of R( t  ) for different starting times (thin lines) and the mean R for every t  (bold line), for the prevailing summertime synoptic flow pattern of ‘‘Persian Trough’’ (PT), with its three modes:Weak (WPT), Modal (MPT) and Deep (DPT) (patterns 4, 5and 6 respectively in Table 2 and Figure 4). The PT is a low pressure system srcinating from the Persian Gulf andconfined to the shallow atmospheric levels up to about 1000 m MSL, advecting cool and moist air onshore the EM,with some variations according to its three different modes[  Dayan et al. , 2002], shown in Figure 4.[ 16 ] A striking feature of R( t  ) in Figure 3 is the conver-gence to a single value of R every 24 hrs, regardless of thestarting time. To explain this behavior, let us consider a puresinusoidal wind oscillating at a period of P = 24 hrs and anamplitude of 1 m s À 1 in the north-south and east-west components. The hourly R( t  ) plot (Figure 5a) will beindependent of the starting time and will reach perfect recirculation (R = 1) every t  = N P (N = 1, 2, . . . ). If wenow add a constant ambient wind component with anamplitude of 3 m s À 1 to the sinusoidal breeze, the resultant R( t  ) plot would still have maximum values at  t  = NP, but R would be smaller then 1 at all times (Figure 5b), with thestronger the ambient wind, the lower R. Nevertheless, R  t  =NP would still be independent of the starting time since theoscillation is of period P. Finally, changing the ambient windfrom a constant to a periodic wind with period P 0 (P 0 = MP,M = 2, 3, . . . ) will result in R = 1 whenever  t  = NP 0 andlocal peaks at  t  = NP. For example, Figure 5c presents R( t  )for P = 24 hrs and P 0 = 96 hrs. In this case, R  t  = NP will nolonger be constant for all starting times since it is also afunction of the P 0  period. For a more detailed discussion onthe analytical aspects of the recirculation parameter, see  Levy et al. [2008]. Looking back at Figure 3 it is nowevident that the constant asymptotic value of the mean R reached after less then 24 hrs and the constant value of R for every starting time at R  t  = N Á 24 for the WPT (Figure 3a), area manifestation of the semi permanent nature associatedwith this synoptic flow pattern that has the highest 5-yrsfrequency (20%) among all synoptic patterns, concentratedduring the summer months (Table 2). The same behavior isshown to a lesser extent for the MPT, as opposed to the DPTthat is much less frequent and hence has a more transient nature (Figures 3b and 3c, respectively).[ 17 ] During the WPT, the weaker synoptic scale forcingallows the LSB to be more dominant over the wind field,resulting in higher recirculation values for the shorter  periods (in the order of 24 hrs), that reaches an asymptoticvalue after about 24 hrs with R  24 = 0.37. As the barometrictrough intensifies to the Modal and Deep modes, the effect of the low periods is weakening, resulting in lower recir-culation factors of R  24 = 0.32 and 0.21 for MPT and DPT,respectively. The highest recirculation values during the first 24 hours for all 3 modes of the PT (R = 0.57, 0.48 and 0.30for the WPT, MPT and DPT, respectively) are obtained inthe morning hours (07:00–08:00 LST) for  t  = 16 hrs, i.e., arecirculating air mass that started at 15:00–16:00 LST the previous day. Such an air mass had traveled onshore withthe sea breeze during the afternoon, veering clockwise untilchanging direction with the land breeze during the late night and early morning, and returned to the shore in the morning.R  t  =24 will be hereafter used as default because of its relativeindependence of the starting time. 3.2. Synoptic Scale Control [ 18 ] The effect of different synoptic scale forcing on thespatial distribution of R is further exemplified in Figure 6for two distinct flow patterns. During days belonging to afrontal low passage over Israel (patterns 11–16), lowrecirculation values (0–0.30) are measured at all sites.During days belonging to a barometric high-pressure systemover Israel (pattern 10), ambient winds are much weaker,allowing the LSB winds to come into effect, resulting ina much higher recirculation factor at all monitoring sites,and particularly in the coastal plain airsheds, with valuesexceeding 0.44.[ 19 ] Figure 7 gives the mean R (top) and mean S 0 (bottom) at each airshed for days belonging to synoptic patterns that are prone to weak synoptic scale forcing near the surface (patterns 3, 10 and 20) versus synoptic patternsthat are prone to a stronger forcing (patterns 11 and 14). Thedifferent forcing of each synoptic pattern is manifested bythe daily mean wind speed, S 0 . For the group of strongsynoptic forcing and higher wind speeds, the resulting R islower (0.2–0.3) compared to the group of weak synopticforcing (0.3–0.5). 3.3. Local Scale Variations Along the Coast [ 20 ] Because of the dominant westerlies at the lower atmosphere in the EM region ( $ 30 °  N) on one hand, andsea breeze being normally stronger than the land breeze onthe other, ideal conditions for recirculation would be astrong land breeze that could counter the sea breeze. Twofactors are examined here that have a major effect on theintensity of the LSB winds: the LSTD and the topographicsettings of the concaved southern shoreline. Figure 8 givesthe difference between SST and daily minimal temperatureover land for each month as well as the mean SST, dailyminimum, maximum and mean temperatures over land for  D16121 LEVY ET AL.: 5-YR COASTAL RECIRCULATION AND POLLUTION4 of 14 D16121        T    a      b      l    e      2  .     F    i   v   e  -    Y   e   a   r    D    i   s    t   r    i    b   u    t    i   o   n   o    f    S   y   n   o   p    t    i   c    F    l   o   w    P   a    t    t   e   r   n   s    O   v   e   r    t    h   e    E    M   a   n    d    T    h   e    i   r    A   s   s   o   c    i   a    t   e    d    G   r   o   u   n    d  -    L   e   v   e    l    W    i   n    d   s ,    9    2    5    h    P   a    W    i   n    d   s   a   n    d    T   e   m   p   e   r   a    t   u   r   e   a   n    d    M   o   n    t    h    l   y    D    i   s    t   r    i    b   u    t    i   o   n     N   o .    N   a   m   e    A    b    b   r   e   v    i   a    t    i   o   n    A   s   s   o   c    i   a    t   e    d    G   r   o   u   n    d  -    L   e   v   e    l    W    i   n    d   s    9    2    5    h    P   a   w    i   n    d   s   a   n    d    t   e   m   p   e   r   a    t   u   r   e    F    i   v   e  -    Y   r    C   o   u   n    t    F   r   e   q .    (    %    )    M   o   n    t    h    l   y    D    i   s    t   r    i    b   u    t    i   o   n    (    d   a   y   s    )   u    (   m   s   À             1     )   v    (   m   s   À             1     )    T    (    k    )    J   a   n    F   e    b    M   a   r    A   p   r    M   a   y    J   u   n    J   u    l    A   u   g    S   e   p    O   c    t    N   o   v    D   e   c    0    U   n    d   e    f    i   n   e    d    Á   Á   Á   Á   Á   Á     1 .    5    À     1 .    0    2    9    0 .    2    1    1    1    6    5    8    1    5    1    3    1    1    5    3    3    1    2    1    5    1    1    1    0    1    R   e    d    S   e   a    T   r   o   u   g    h   w    i    t    h    E   a   s    t   e   r   n   a   x    i   s    R    S    T    E    N    E  -    N    W    À     0 .    3    À     2 .    5    2    8    8 .    6    2    1    0    1    1    1    4    1    2    3    7    2    2    2    3    0    0    0    1    0    4    7    2    3    2    2    2    R   e    d    S   e   a    T   r   o   u   g    h   w    i    t    h    W   e   s    t   e   r   n   a   x    i   s    R    S    T    W    S    E    À     1 .    5    2 .    0    2    8    5 .    5    2    3    1    3    3    2    2    0    0    0    0    0    0    6    7    3    R   e    d    S   e   a    T   r   o   u   g    h   w    i    t    h    C   e   n    t   r   a    l   a   x    i   s    R    S    T    C    E    À     3 .    0    À     0 .    5    2    8    9 .    0    1    3    9    8    1    5    1    9    1    2    4    7    3    0    0    3    1    8    3    9    1    9    4    W   e   a    k    P   e   r   s    i   a   n    T   r   o   u   g    h    W    P    T    N    W    3 .    0    À     3 .    1    2    9    6 .    9    3    7    2    2    0    0    0    0    0    1    4    7    8    9    4    1    1    0    6    1    1    5    0    0    5    M   o    d   a    l    P   e   r   s    i   a   n    T   r   o   u   g    h    M    P    T    W    4 .    4    À     3 .    1    2    9    7 .    2    1    1    2    6    0    0    0    0    1    2    8    3    9    3    3    1    1    0    0    0    6    D   e   e   p    P   e   r   s    i   a   n    T   r   o   u   g    h    D    P    T    W  –    S    W    5 .    8    À     1 .    9    2    9    6 .    3    1    7    1    0    0    0    0    0    3    6    4    4    0    0    0    7    H    i   g    h    t   o    t    h   e    E   a   s    t    H    E    S    E    À     1 .    7    2 .    9    2    8    7 .    5    7    7    4    1    5    1    5    2    7    1    0    0    0    0    1    1    2    2    4    8    H    i   g    h    t   o    t    h   e    W   e   s    t    H    W    N    W    4 .    0    À     2 .    3    2    8    9 .    2    2    3    9    1    3    1    6    1    4    2    9    3    6    3    8    2    3    1    3    3    2    4    2    5    1    1    7    9    H    i   g    h    t   o    t    h   e    N   o   r    t    h    H    N    N  –    N    E    À     0 .    6    À     3 .    1    2    8    7 .    1    1    1    8    6    1    9    1    1    9    1    1    1    7    4    0    1    9    8    1    3    1    6    1    0    H    i   g    h   o   v   e   r    I   s   r   a   e    l    H    I    V    R    B    1 .    2    À     1 .    9    2    8    8 .    2    1    0    4    6    4    1    3    1    1    2    4    1    3    2    0    1    1    2    1    1    5    8    1    1    D   e   e   p    L   o   w    t   o    t    h   e    E   a   s    t    D    L    E    N    W    9 .    6    À     0 .    3    2    8    1 .    1    2    2    1    4    5    5    1    0    0    0    0    0    0    5    2    1    2    D   e   e   p    C   y   p   r   u   s    L   o   w    t   o    t    h   e    S   o   u    t    h    D    C    L    S    S    E    À     4 .    5    6 .    6    2    8    4 .    4    1    0    0    1    0    0    0    0    0    0    0    0    0    0    1    3    S    h   a    l    l   o   w    C   y   p   r   u   s    L   o   w    t   o    t    h   e    S   o   u    t    h    S    C    L    S    S    E    2 .    0    4 .    0    2    8    4 .    1    8    0    2    2    0    0    1    0    0    0    0    0    2    1    1    4    D   e   e   p    C   y   p   r   u   s    L   o   w    t   o    t    h   e    N   o   r    t    h    D    C    L    N    W    9 .    2    6 .    0    2    8    4 .    0    1    1    1    3    2    2    0    0    0    0    0    0    1    1    2    1    5    S    h   a    l    l   o   w    C   y   p   r   u   s    L   o   w    t   o    t    h   e    N   o   r    t    h    t    h    S    C    L    N    W    7 .    3    3 .    3    2    8    5 .    9    4    8    3    1    0    1    1    3    5    4    0    0    0    0    2    5    8    1    6    C   o    l    d    l   o   w    t   o    t    h   e    W   e   s    t    C    L    W    S    W    3 .    1    6 .    8    2    8    5 .    3    1    3    1    3    0    3    0    0    0    0    0    0    0    1    6    1    7    S    h   a    l    l   o   w    L   o   w    t   o    t    h   e    E   a   s    t    S    L    E    N  –    N    N    W    4 .    3    À     2 .    0    2    8    5 .    9    8    8    5    1    9    1    8    7    1    3    1    3    4    0    0    4    4    0    6    1    8    S    h   a   r   a   v    L   o   w    t   o    t    h   e    W   e   s    t    S    L    W    S    E    1 .    1    2 .    2    2    9    1 .    3    1    7    1    0    1    6    7    3    0    0    0    0    0    0    0    1    9    S    h   a   r   a   v    L   o   w   o   v   e   r    I   s   r   a   e    l    S    L    I    S    E  –    S    W    0 .    2    2 .    0    2    9    3 .    0    1    1    1    0    0    4    3    4    0    0    0    0    0    0    0    2    0    C   o    l   e  –    V    R    B    0 .    8    1 .    5    2    8    6 .    2    4    2    2    1    5    3    5    1    0    0    0    0    0    2    1    1    5    2    1    S    h   a    l    l   o   w    S   y   r    i   a   n    L   o   w   w    i    t    h    H    i   g    h    t   o    t    h   e    N   o   r    t    h    S    S    L    N    W    5 .    5    À     1 .    1    2    8    6 .    0    2    4    1    3    1    1    0    5    0    0    0    0    6    4    4    2    2    S    h   a    l    l   o   w    C   o    l    d    L   o   w    t   o    t    h   e    W   e   s    t    S    C    L    W   w   e   a    k    S    W    3 .    6    4 .    2    2    8    3 .    9    2    0    1    5    3    2    1    0    0    0    0    0    0    1    8 D16121 LEVY ET AL.: 5-YR COASTAL RECIRCULATION AND POLLUTION5 of 14 D16121
Search
Tags
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks